earth sciences research journal earth sci. res. j. vol. 20, no. 2 (june, 2016) letter from the editor dear readers. the earth sciences research journal has been recently flooded with dozens of good papers expecting to be published. thanks to you, readers, reviewers, authors for your trust. we are continuously working to improve our editorial processes. the trend topics of those papers go from gas and petroleum exploitation, land use, geological hazards, to sustainable development, environmental protection and optimization of operation techniques. without forgetting the basic research. all those subjects are welcome in earth sciences research journal. proudly, we remark our journal, your journal, as the natural stage for the debate between economic development and sustainable development. there is no a better referee for this discussion than education entities. if your manuscript put lights on this important subjects, do not hesitate to submit it. is this not a minor topic! nowadays, every political decision, every policy designed, every industrial and human activity are seen through the lens of the environmental awareness. we can not quit on this responsibility. as an example of how important those topics are, during the coming days, us president barack obama is going to enact a law expanding the protection areas in hawaii in close four times the current size. it means longlines fishing operators will not be permitted in the 582581 square miles; neither the seabed mining in a region with rich deposits of manganese, nickel, zinc, cobalt, and titanium. similar decisions have been taken in chile and costa rica, this last one with a 26% of its land territory protected as a natural park. just talking about some cases in this side of the world, where we are more familiar with. the next meeting of geological society of america, taking place in denver, on september, will hand on this same topics. are the extracting and exploitation of resource prohibitions the solution to environmentalism? welcome all your papers cracking this nut. carlos alberto vargas jiménez. editor in chief, earth sciences research journal l e t t e r f r o m t h e e d it o r earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016) letter from the editor as we are sure you’ve noticed, the earth sciences research journal has undergone some substantial changes this year. the hard work of the last few years has resulted in a some major steps forward for this journal, and now we want to announce our four volumes to be published since 2016 every year. our goal is to become a premier journal, and it is to this end that we work day in and day out. recently several conferences have turned to us for special issues, and we expect this to be an ongoing relationship where we are able to provide readers with novel articles. by supplying conferences, we are able to broaden the pool of scientific papers by giving those who might otherwise have been denied it the opportunity of publishing to a worldwide audience. it is not just earth sciences research journal that is changing, so too is the global petroleum market. the fluctuating price of oil and technological advancements have played a role in upturning the previous status quo. many previously unknown actors are now operating on the global stage in this new emerging situation. management of resources is in the political spotlight; energy and water use, as well as environmental concerns, have been thrust to the forefront of discussion. fracking and shale extraction, as well as well as the stretching of exploration into more and more delicate environments have caused substantial damage, and pose the risk of causing immeasurable destruction in the future. this new reality has brought a whole new focus to a segment of the scientific community, and our journal is poised to accept these new articles. we want to encourage everyone involved in the scientific research on earth sciences to submit their papers. we have substantially modernized our processing, revising all submitted works, and otherwise improving the experience for the authors involved. the review process has undergone marked update, reducing wait times while improving quality. in the future, we intend to collaborate with other organizations in the earth sciences field. in us, you will find a friendly partner with whom to collaborate and bring our field into a new age. for the volume 20, we are presenting 17 selected papers in different areas of study; original contents we hope will be of interest and use. comments, reviews, and suggestions are always welcomed in earth sciences research journal through the mail account earthjour_fcbog@unal.edu.co. thanks for reading and submitting your valuable scientific papers! carlos alberto vargas jiménez. editor in chief, earth sciences research journal l e t t e r f r o m t h e e d it o r it is vital for analyzing groundwater chemical characteristics to develop, manage and protect groundwater resources. hydrogeochemistry simulation, graphic method, and other methods are applied to analyze 22 pieces of groundwater chemical data in the urban area of zhumadian city. the results show that shallow groundwater is mainly of hco3-ca type. besides, from northeast to southwest, salinity and hardness increase gradually, the hydrodynamic condition becomes worse, and groundwater is significantly affected by human activities. this study is of important significance to scientifically using and managing groundwater resources in the urban area of zhumadian city. el análisis de las características químicas de las aguas freáticas o subterráneas es determinante para desarrollar, administrar y proteger los recursos hídricos. este trabajo se basó en la simulación hidrogeoquímica, el método gráfico y otros procedimientos para analizar la información química de 22 muestras de aguas freáticas en el área urbana de zhumadian, en el este de china. los resultados muestran que las aguas freáticas poco profundas son principalmente del tipo hco3-ca. además, del noreste al suroeste, la salinidad y la rigidez aumentan gradualmente, las condiciones hidrodinámicas empeoran y las aguas freáticas son afectadas por las actividades humanas. este estudio es de una importancia significativa para el manejo científico y la administración de los recursos de aguas subterráneas en el área urbana de zhumadian. earth sciences research journal earth sci. res. j. vol. 20, no. 3 (september, 2016) : g1 g5 abstract resumen keywords: urban area of zhumadian city; groundwater; hydrochemical characteristics palabras clave: área urbana de zhumadian, aguas freáticas, características hidroquímicas. chemical characteristics of groundwater and its spatial distribution in urban area of zhumadian city record manuscript received: 16/12/2015 accepted for publication: 21/11/2016 how to cite item zhongpei, l., jianhua, p., & yuping, h. (2016). chemical characteristics of groundwater and its spatial distribution in urban area of zhumadian city. earth sciences research journal, 20(3), g1-g5. doi:http://dx.doi.org/10.15446/esrj.v20n2.54710 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.54710 liu zhongpei1,2, ping jianhua3, han yuping1* 1 department of water resources, north china university of water resources and electric power, zhengzhou-450045, china 2 china institute of water resources and hydropower research, beijing-100038, china 3 school of environment and water conservancy engineering, zhengzhou university, zhengzhou 450001, china *for correspondence: hanyp@ncwu.edu.cn g e o c h e m is t r y características químicas y distribución espacial de las aguas freáticas en el área urbana de zhumadian (china) liu zhongpei, ping jianhua, han yupingg2 analysis of groundwater chemical characteristics is an essential content of hydrogeology subject. it will be of important significance to both use and management of region groundwater resources and construction and protection of ecological environment (zhang, deng, & he, 2006). chemical characteristics of the groundwater are determined by sedimentary environment and sedimentary association characteristics of provenance. it is also subject to the interaction of natural and human factors such as flood, saline intrusion, ecologic water conveyance and agricultural irrigation (an, zhao, ye, liu & ding, 2012). through study on groundwater chemical characteristics, the scientific basis will be provided for interaction mechanism of groundwater and environment, description of groundwater recharge, runoff and discharge conditions, as well as groundwater resources management. in recent twenty years, for rapid economic and social development and climate changes in zhumadian city, the conflict between water resources supply and demand in the urban area of zhumadian city increase prominently. the urban area of zhumadian city is located at the slight area of the mountain front, where surface water resources are deficient and limited surface water resources are not enough to meet the increasing water resources demand, and groundwater plays a vital role in development and use of water resources (liu, zhang, & ren, 2002). study on chemical characteristics of shallow groundwater in the urban area of zhumadian city is conductive to understand further of typical features of current shallow groundwater in the area. moreover, it can provide scientific bases and management concept for groundwater use in this area, and it is good to safeguard sustainable development of water resources. also, it is of great theoretical guidance significance and practical value of reasonable use and protection of groundwater resources in the urban area of chinese cities. 1.2 geology and hydrogeology shallow groundwater in the urban area of zhumadian city mainly lies in silty, clayey soil porosity of quaternary pleistocene series. aquifer group structure is single-layer clayed soil structure. water storage capacity of shallow groundwater is different in different parts of the study area. quaternary pleistocene series lithology in this area is thick-layer brownish yellow, auburn yellow, yellowish-brown and ginger silty clay with muddy medium-coarse sand, muddy medium-fine sand, fine muddy silt and muddy gravel layer. silty clay porosity and fracture from earth surface to the layer 30m deep are well developed and connected and constitute a storage space of shallow groundwater. a supply source of shallow groundwater in the urban area of zhumadian city is mainly from atmospheric precipitation, following lateral groundwater flow and well irrigation area infiltration. shallow groundwater flows from west to east, without obvious depression cone. the drainage mainly includes industrial and agricultural exploitation, evaporation and lateral flow. 2 data and methods 2.1 data twenty-two sampling wells were arranged in the study area, as shown in figure 1, with a radius of 0.25-0.50m and depth of 15-60m to reveal the chemical characteristics and spatial distribution characteristics of shallow groundwater in the urban area of zhumadian city. the analysis items of groundwater quality mainly include bicarbonate ion, carbonate ion, free carbon dioxide, calcium ion, magnesium ion, potassium-sodium ion, chloride, sulfate, ammonia nitrogen, total hardness, total dissolved solids, ph value, microelement, metal ion, conductivity, total acidity and total alkalinity. 2.2 methods this paper studies water chemistry of groundwater and its spatial distribution characteristics in the urban area of zhumadian city using graphical method (zhang et al., 2006; an et al., 2012; lu, zhou, wang, yang, & lu, 2010), hydrogeochemistry simulation (tang, chen & qian, 2012; jiang, li, & hu, 2009; li, hou, zhou, & liu, 2009) and other methods. 1.1 physical geography the study area is located zhumadian city, henan province, with geographical coordinates of east longitude 113°56′02″～114°08′52″ and northern latitude 32°54′29″～33°04′31″. the total area is 360 km2, and the number of population is 570400. the study area is a relatively gentle terrain and ground slope gradient of 1～10 ‰. bounded by guanwang temple, shun river, and li tower, its western part is alluvial-proluvial plain, and its eastern part is alluvial-lacustrine plain, as shown in figure 1. the earth surface is composed of hard and dense silty clay of the middle pleistocene and slight yellow muddy clay limnetic facies of upper pleistocene. it is in the transition zone from northern subtropics to warm temperate zone and belongs to a mainland monsoon type climate, with primary weather features of prominent monsoon climate, four distinctive seasons, moderate climate, the same season of rain and hot, multi-year mean evaporation capacity of 935.5mm and multi-year mean precipitation of 966.0mm. the water system is not well developed in the study area, in addition to a few of streams, only a relatively large lianjiang river passes the area from the south boundary. lianjiang river is a branch of ru river and flows into suya lake reservoir from west to east through the urban area. figure 1. sampling point location chemical characteristics of groundwater and its spatial distribution in urban area of zhumadian city g3 3 results and discussion 3.1 spatial distribution characteristics of water chemistry composition 3.1.1 spatial distribution characteristics of tds and hardness under the usual state of many water chemical properties, tds can reflect the overall condition of chemical composition field, and total hardness is an important indicator of water quality and significantly affects industrial water (zhang & shen, 2000). based on the results of chemical analysis of water samples in the study area, contour map of tds of shallow groundwater in the urban area of zhumadian city (fig. 2) and contour map of total hardness (by caco3) (fig.3) are drawn as below. it can be learned from the analysis as below. (1) according to tds level, groundwater can be classified into fresh water (tds<1g /l), brackish water (1g /l≤tds≤3g /l) and salt water (tds>3g /l). shallow groundwater in the urban area of zhumadian city is mainly fresh water, with brackish water in southwestern part and no salt water. tds becomes lower from southwest to northeast, and it changes rapidly in the southwest part, but changes slowly in other regions. based on water quality monitoring data, tds of well 18 and 19 is relatively high, contents of cland so4 2ions at these two points are far superior to that of other monitoring points and may be affected by pollution. (2) according to hardness classification standard (by caco3), groundwater can also be classified into soft water (<150mg/l), little hard water (50-300mg/l), hard water (300-450mg/l) and very hard water (>450mg/l). hard water and very hard water mainly distribute around the urban area, and groundwater is of relatively small hardness in the center area. based on features of the city, groundwater around urban area of zhumadian city is relatively high and may be polluted by urban domestic water, industrial water, and other wastewater. (3) based on figure 2 and figure 3, tds and salinity of the southwest part of the urban area of zhumadian city are both relatively high, which less likely result from natural causes. by access to information, it is found that there is a quarrying transport factory at xiangshan town at the southwest part of the urban area of zhumadian city, and there are power plants and coal mines in gucheng town, so it is speculated that such human activities result in pollution of local groundwater. 3.1.2 spatial distribution characteristics of hydrochemical type piper trilinear chart of shallow groundwater (figure 4) is drawn, and graphical method is applied to analyze and study hydrochemical characteristics and its spatial distribution rule of shallow groundwater in the study area. figure 4. piper trilinear chart of shallow groundwater. from the figure, it can be drawn as below. (1) water sampling points in the study area are mainly distributed at the left bottom of the delta-shaped region, indicating groundwater in the study area is primarily of hco3-ca type, or hco3-ca·mg, so4-ca·mg and so4-ca type. (2) hydrochemical types of the study area from southwest to northeast are transiting from hco3-ca to hco3-ca·mg, so4-ca·mg and so4-ca· and spatial variation degree of the anion is higher than that of positive ion. (3) carbonate hardness of water sample point is greater than 50%, indicating that chemical composition of water is mainly of carbonate lixiviation. besides, groundwater is dominated by hco3 with little mineralization degree and also reflecting that shallow groundwater in the urban area of zhumadian city is mainly from atmospheric precipitation. 3.2 ion proportionality coefficient characteristics in chemical composition of groundwater, content proportionality coefficient among all compositions is often used to study some figure 2. contour map of tds figure 3. contour map of total hardness (by caco3) liu zhongpei, ping jianhua, han yupingg4 hydrogeochemical problems and judge the groundwater causes, groundwater composition source and groundwater classification li, sun, & liu, 2000). coefficient ƴna/ƴcl and ƴcl/ƴca are respectively used to reflect the causes and hydrodynamics of shallow groundwater in the study area jiang et al., 2009. figure 5. [ na+ ] vs [ cl] figure 6. [cl/ca2+] vs tds (1) ƴna/ƴcl coefficient is called as groundwater cause coefficient and is a hydrogeochemical parameter of sodium ion concentration degree of natural groundwater. the mean value of ƴna/ƴcl coefficient of standard saline water is 0.85; low salinity water has relatively high ƴna/ƴcl coefficient (ƴna/ƴcl>0.85) and high salinity water has relatively small ƴna/ƴcl coefficient (ƴna/ƴcl<0.85). from figure 5, it can be drawn that analysis point of shallow groundwater sample in the urban area of zhumudian city is at or close to line 1:1. it indicates that groundwater will continuously dissolve rock minerals by the hydrolysis and acidification action in the runoff process and release na+ from feldspar so as to make na+ concentration larger than clconcentration. it also indicates that groundwater has a rock-salt lixiviation action in the runoff process. sampling well 18 is abnormally large due to clmilliequivalent and ƴna/ƴcl coefficient is relatively small, which reflects southwestern part of the study area may be polluted. (2) ƴcl/ƴca coefficient can be used as a parameter describing hydrodynamics. from figure 6, it can be drawn that when tds<1000mg/l, ƴcl/ƴca coefficient does not change significantly with tds. however, when tds>1000mg/l, ƴcl/ƴca coefficient increases sharply. it shows that northeast part of the study area is a relatively good hydrodynamic condition and southeast part is a relatively weak hydrodynamic condition. at the same time, there are two significant deviations sampling points, well 18 and 19, which also reflects that they are most likely polluted. 3.3 saturation index analysis saturation index refers to the parameter which is used to determine the status of water and minerals through getting water-rock interaction information, represented by “si.” when si>0, minerals in water are saturated and will subside; when si<0, minerals in water are unsaturated and will be dissolved; when si=0, minerals in water are in an equilibrium state of dissolution and sediment, that is, the minerals will not subside and also not be dissolved (li et al, 2009). phreeqc software is used to simulate and calculate saturation index of mineral facies of shallow groundwater in the study area to quantitatively describe evolution state of the chemical field of shallow groundwater. because shallow groundwater is mainly of hco3-ca type, calcite saturation index is selected to draw a contour map of mineral saturation index to analyze hydrochemical characteristics of groundwater in the study area. figure 7. calcite si contour map from calcite si contour map (fig.7), it can be drawn as below. (1) calcite mineral si in the study area mostly is less than 0, and the fields with an si larger than 0 are mainly at southwest, southeast and northeast part of the study area, reflecting the study area is dominated by calcite dissolution. however, at the southwest, southeast and northeast part of the study area, calcite sediment is separated out. (2) mineral saturation index may be affected by groundwater circulation and exchange degree. calcite saturation indexes at the southwest, southeast and northeast part of the study area are larger than that of other regions, and that of the southwest part is the largest. it reflects runoff condition at such sections is relatively worse than that of other areas, water circulation and exchange are relatively slow, and mineral saturation index is relatively high, especially water circulation and exchange in southwestern part are the weakest. (3) parts with relatively high mineral saturation index are around the urban area, similar to hardness distribution rule. based on the study area including large urban, living and industrial activity strength and shallow groundwater, groundwater chemistry is more easily affected by human endeavor, reflecting that shallow groundwater in the urban area of zhumadian city is significantly affected by human activity. chemical characteristics of groundwater and its spatial distribution in urban area of zhumadian city g5 4. conclusion based on spatial distribution characteristics, ion proportionality coefficient and saturation index analysis results of chemical composition of water in the study area, conclusion can be drawn as below: (1) shallow groundwater in the urban area of zhumadian city is mainly of hco3-ca type, salinity, and hardness decrease from southwest to northeast and hydrodynamic conditions become worse from northeast to southwest. (2) intensive human activities occur in the study area, which may significantly affect groundwater, and the groundwater at the southwest part of the study area is likely polluted. acknowledegment this study was supported by the national natural science foundation of china (grant no. 51209090 and 71271086), the “948” program of the ministry of water resources in china (grant no. 201328). references an, l. s., zhao, q. s., ye, s. y., liu, g. q., ding, x. g. (2012). hydrochemical characteristics and formation mechanism of shallow groundwater in the yellow river delta. chinese journal of environmental science, 33(2), 370-378. jiang, l., li, p., & hu, a. (2009). the groundwater chemical characteristics in the yaoba oasis of alxa area, inner mongolia. journal of arid land resources and environment, 23(11), 105-110. li, x., sun, z., & liu, j. (2000). hydrogeochemical. atomic energy press, 3rd. edition. li, x. q., hou, x. w., zhou, z. c., & liu, l. x. (2009). hydrogeochemical characteristics and formation evolutional mechanism of the groundwater systems in the taiyuan basin. geoscience-journal of graduate school, china university of geosciences 23(1), 1-8. liu, g., zhang, j., & ren, h. (2002). assessment of shallow groundwater environmental in zhumadian urban. henan water conservancy, 4, 75-80. lu, x., zhou, a., wang, m., yang, l., lu, h. (2010). characteristic analysis of phreatic water equality evolution by piper diagram in huaihe river drainage area, jiangsu province. geotechnical investigation & surveying, 2, 42-47. tang, h., chen, j., & qian, h. (2012). application of saturation index to research of water-rock interaction and its sensitivity analysis. journal of water resources and water engineering , 6, 180-183. zhang, g. x., deng, w., & he, y. (2006). hydrochemical characteristics and evolution laws of groundwater in songnen plain, northeast china (in chinese). advances in water science, 17(1), 20-28. zhang, z., & shen, z. (2000). evolution of groundwater environment in north china plain. geological publishing house. keywords: regularization; iterative continuation; anomaly separation; physical parameters. palabras clave: regularización; continuación iterativa; separación de anomalías; parámetros físicos; inversión. how to cite item wang, y., li, j., wang, x., & jian, x. (2019). inversion technique of physical parameters based on regularization extension depth constraint. earth sciences research journal, 23(4), 331-338. doi: https://doi.org/10.15446/ esrj.v23n4.84340 through the regularization downward continuation of gravity and magnetic anomalies, the depth of the field source can be solved. however, due to the gibbs effect, the horizontal resolving power of the field source is poor. in view of this, based on the depth of field source established by regularization downward continuation, this paper proposes a physical property parameter inversion method based on iterative continuation and anomaly separation, which can effectively improve the inversion accuracy of superimposed anomaly physical parameters, and provide a new idea for solving the physical parameters of superposition gravity and magnetic anomalies. abstract inversion technique of physical parameters based on regularization extension depth constraint técnica de inversión de parámetros físicos basada en la regularización de restricción de extensión de profundidad issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.84340 a través de la regularización de la gravedad y las anomalías magnéticas, se puede resolver la profundidad de la fuente del campo. sin embargo, debido al efecto gibbs, el poder de resolución horizontal de la fuente de campo es poco. en vista de esto, basado en la profundidad de la fuente de campo establecida por la continuación de la regularización hacia abajo, este artículo propone un método de inversión de parámetros de propiedades físicas basado en la continuación iterativa y la separación de anomalías, que puede mejorar efectivamente la precisión de inversión de los parámetros físicos de anomalías superpuestas, y proporcionar una nueva idea para resolver los parámetros físicos de la superposición de gravedad y anomalías magnéticas. resumen record manuscript received: 01/04/2019 accepted for publication: 09/08/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 331-338 yang wang1, 2, jun li1, 3*, xuben wang1, 3, xingxiang jian1, 3 1college of geophysics, chengdu university of technology, chengdu 610059, china 2public class teaching department, sichuan vocational and technical college of communications, chengdu 611130, china 3ministry of education key laboratory of earth exploration and information technology, chengdu university of technology, chengdu 610059, china * corresponding author: lijun3@cdut.edu.cn g eo ph y si c s 332 yang wang, jun li, xuben wang, xingxiang jian introduction the downward continuation of the potential field can provide abundant information on the spatial distribution of the potential field and is an important part of the interpretation of gravity and magnetic anomalies (zhu, cao, & lu, 2014; deng, liu, song, & chen, 2014). luan wengui adopted another regularization functional and deduced another response. according to the same regularization functional, another response formula was derived by luan wengui, which is suitable for understanding the distribution characteristics of the source potential field in the near field (zhang, dai, & liu, 2014). c. w. groetsch, p mauriello, d patella, xu deshu, zeng hualin, wang xuben, gao yongcai and other scholars have also studied the extension method. through research and comparison, it shows that the regularization method has the best downward extension effect. but due to the influence of the gibbs effect, on both sides of the main anomaly zone, there is often a “positive and negative alternating “ ” shaped weak anomaly zone with the main source distribution area as the center, which may be a separate anomaly or composite anomalies formed by the superposition with other anomalies, thus increasing the difficulty to explain the field source. to solve this problem, based on the gravity and magnetic data regularization downward extension, this paper puts forward downward gravity and magnetic property parameters inversion method of downward continuation and anomaly separation based on the iterative method, providing a new way of thinking for the comprehensive interpretation of the field source from the material properties and geometric parameters (al farajat et al., 2015; li & zhang, 2017; mazgaj, szular, & szczurek, 2017). basic principle regularization extension technique according to the laplace equation and the boundary condition of the potential field, there is: ∂ /∂ +∂ /∂ φ 2 2 2 2 0 0 u x z x u x z z u x z u x z f x z x l z x , , , , , ( ) ( ) = ( ) = ( ) = ( )  → = ( )       (1) u x z,( ) is the potential field at the measuring point x z,( ), and f x z,( ) is the known field value on the measuring point;  x( ) is the height of the measuring point. by using the method of separating variables and the boundary condition, the general formula of the series decomposition of the field potential can be obtained (boehm & ulbrich, 2015). f x z a x ei i z i n , cos ,( ) = ( ) = ∑   0 (2) set m as the number of points at the measuring point, and x is the distance between the measuring points, there are x i x= ( ) ⋅1 ∆ λ π / −i i m x= ⋅ ( ) ⋅1 ∆ this result is the series decomposition method of the traditional downward continuation (raknes & arntsen, 2014). in order to overcome the oscillatory effect, the series is still taken as the basic form of the field value solution. in solving the continuation function, a decomposition function with a relatively smooth variation is selected and the error of fitting it to the original field is minimized on the original observation profile, so that the analytical function of regularized downward continuation series after correction can be obtained (zhang et al., 2015; fontchastagner, lubin, mezani, & takorabet, 2018). f x z a a x e vi i z i i n i, cos( ) = ( ) = ∑12 0 1 + λ λ (3) vi is called the regularization factor, which is 1 1 2 2/ +αλ λi ze i( ), a function related to the frequency i , continuation depth z and coefficient . it corrects the frequency of the series term of the continuation field, and it is proved by practice that as long as the regularization function is properly selected, it cannot only guarantee the non-singularity of the potential field in the process of continuation, so as to obtain the complete field value distribution in the lower half space, but also the gibbs effect cannot be very significant. iterative continuation technique it is often necessary to extend downwards to the top of different depth layers to separate the anomalies produced by different depth layer sources on the ground. the stability and depth of continuation directly affect the result of continuation. to deal with the instability of potential field downward continuation and the limited depth of downward continuation by the classical fft method, xu shizhe put forward a downward continuation method in spatial domain, which is implemented based on stable fft upward continuation. the principle is simple and the downward continuation is stable. the basic idea of the downward continuation of the iterative method is to assume that there are two planes, a and b, representing the ground surface and the plane of a certain depth under the ground, respectively. between the two, there is no field source distribution, that is, there is a passive space, so that the downward continuation iterative solution of the known depth based on upward continuation can be realized according to the following diagram (chen, 2015). figure 1. implementation diagram of downward extension through the iterative method inversion of physical parameter of gravity and magnetic anomaly assuming the thickness of a certain depth of the underground layer as h, the layer is divided into mn mn m n= ×( ) small vertical prisms of the same size with a finite extension, and the coordinate of the center point of the small prisms is x yj j0 0,( ) and residual density is  j j mn=( )1 2 3, , , , ; its side length is a and b respectively; the top depth of the prism is h and the height is dh. thus, the formula for the frequency spectrum g u vj ,( ) of the gravity anomaly g x yj ,( ) generated by the ground x y,( ) of each prism in the frequency domain is as follows: g u v g uv au bv e r ej j hr dhr, sin sin( ) = ⋅         ⋅ ⋅ ( )2 4 2 2 1π ρ − − − ⋅⋅ ( )e i ux vyj oj− +0 (4) 333inversion technique of physical parameters based on regularization extension depth constraint thus, by fourier inverse transformation, the residual density values of each point x y, �( ) on the top surface of the depth layer can be obtained: ρ π − − − x y f g u v g uv au bv e r dhr , , sin sin ( ) = ( ) ⋅         ⋅  1 2 4 2 2 1               (9) formula 9 is the inversion formula for calculating the relative physical parameters of the gravity anomaly of the known field source. the apparent density corresponding to the gravity anomaly can be calculated by the formula. for the magnetic source anomaly, the gravity anomaly of the magnetic source can be obtained first, and the magnetic source density or magnetization intensity can be inversed by the same method. in view of the space, its calculation formula is not deduced in this paper. gravity anomaly regularized continuation depth constraint inversion regularized continuation depth constrained inversion of single model gravity and magnetic anomalies in this paper, a 101 × 101 mesh is designed, the distance between points and that between lines is all 1 (unit: meter), that is, the measuring area is 100 × 100m2. the specific parameters of the sphere model are shown in table 1. table 1. specific parameters of a single sphere model spherical center coordinate (m) sphere radius (m) residual density (g/cm3) magnetization intensity (a/m) magnetization inclination magnetization deviation angle (50,50,20) 5 0.5 100 45° 0° the corresponding gravity anomalies (δ g) and magnetic anomalies (δ t) are derived from the forward calculation of the above model, as shown in figure 2: figure 3 is a section of the regularization result of the above magnetic anomalies, which shows that the center of the anomaly is consistent with the depth of the given model. however, due to the influence of the gibbs effect, in the upper formula, u, v represent the wave number of x, yrespectively; g is the gravitational constant; and r u v= +2 2 . the frequency spectrum g u v,( ) is made fourier inversion to obtain the gravity anomaly g x y,( ) in the space domain. according to the superposition property, the frequency spectrum g u v,( ) of gravity anomaly g x y,( ) produced by the whole depth layer at the ground x y, �( ) is as follows: g u v g uv au bv e r e dhr dhr j , sin sin( ) = ⋅         ⋅ ⋅ ( )2 4 2 2 1π − ρ − − ⋅⋅ ( ) = ∑ e i ux vy j mn oj− +0 1 (5) formula 5 is the forward formula of field source gravity anomaly with known physical parameters. if the gravity anomaly g x y,( ) generated by the depth layer in the top surface is known, h = 0 (if h ≠ 0, that is, the buried depth is not zero, and it can be extended to the top interface first, and then: ρ ρ − + u v ej i ux vy j mn oj,( ) = ⋅ ( ) = ∑ 0 1 (6) thus:  g u v g uv au bv e r u v dhr , sin sin ,( ) = ⋅         ⋅ ( ) ⋅ ( )2 4 2 2 1 π − ρ − (7) the following formula can be deduced from the above:   ρ π − − u v g u v g uv au bv e r dhr , , sin sin ( ) = ( ) ⋅         ⋅2 4 2 2 1 (8) figure 2. gravity anomaly (left) and magnetic anomaly (right) of the sphere model 334 yang wang, jun li, xuben wang, xingxiang jian on both sides of the main anomaly zone, there is often a “positive and negative alternating “øè” shaped weak anomaly zone with the main source distribution area as the center, which may be a separate anomaly or form composite anomalies by the superposition with other anomalies, thus increasing the difficulty to explain the field source. figure 3. magnetic source gravity anomaly of sphere model figure 4. the principal curve of magnetic source gravity anomaly and its regularization downward extension fracture surface regularization continuation depth constrained inversion of superimposed gravity and magnetic anomalies in order to carry out theoretical calculation and verification, a 201 × 201 mesh is designed in this paper. the distance between points and that between lines is all 5 (unit: meter, that is, the range of measuring area is 1000 × 1000m2. the model consists of three vertical prisms, in which the column centers of model 1 and model 3 are on the same vertical line, and the specific parameters of each column model are shown in table 2: according to the relevant parameters of the above model, the gravity anomaly and magnetic anomaly (unit: nt) of the model can be obtained by forward modeling (unit: mgal). the result is shown in figure 6. in order to extend this method to the processing of magnetic anomalies effectively, the magnetic anomalies’ magnetic source gravity is solved first, and the magnetic anomalies are also solved according to the gravity anomaly process. figure 7 shows the magnetic source gravity anomaly of the magnetic source model. the anomaly of the main section of the model is selected for regularized downward continuation, so that the depth of the field source is obtained respectively. it can be seen from the diagram that the depth of the regularization extension has a high accuracy. the regularization is used to establish the depth of the field source for iterative continuation. considering that the basic requirement of iterative continuation is that the two depth interfaces are passive field spaces, the idea of this paper is to carry out the continuation of shallow source depth first; then, the shallow-source anomaly is separated by cutting method, and the cutting radius is established according to the anomaly range of shallow-source anomaly. (the idea of this paper is to use laplace operator to establish abnormal boundary and then the radius is taken as (l+d)/2) according to the anomaly boundary.) 335inversion technique of physical parameters based on regularization extension depth constraint figure 5. magnetic source gravity anomaly (left) and apparent density inversion results (right) table 2. specific parameters of the combined model parameters cylinder cylindrical center coordinate (m) cylinder length (m) cylinder width (m) cylinder height (m) density (g/cm3) magnetization intensity (a/m) declination magnetic dip 1 (650,650,500) 300 300 100 0.4 80 0° 45° 2 (200,200,50) 50 50 50 0.5 100 0° 45° 3 (650,650,50) 50 50 50 0.5 100 0° 45° (a) spatial position of the model (b) model plane position and parameters figure 6. model and its forward modelling anomaly 336 yang wang, jun li, xuben wang, xingxiang jian (c) gravity anomaly g of model forward modelling (d) magentic anomaly ta of model forward modelling figure 6. model and its forward modelling anomaly figure 7. magnetic source gravity of the model figure 8. anomaly curve of diagonal section of magnetic source gravity anomaly and its regularized extension 337inversion technique of physical parameters based on regularization extension depth constraint figure 9. magnetic source gravity’s anomaly value after the downward extension of 50 m figure 10. shallow anomaly (left) and deep anomaly (right) obtained by cutting 50 m depth anomaly figure 11. the density value obtained by the shallow anomaly inversion 338 yang wang, jun li, xuben wang, xingxiang jian conclusion based on the potential field theory and regularization theory, a method for inversion of gravity and magnetic physical property parameters of downward continuation and anomaly separation based on iterative method is proposed. the following points should be pointed out: 1) the inversion effect based on single model depth constraint is better; 2) the lateral superposition model can obtain higher lateral resolution anomalies in the process of iterative downward continuation. therefore, the inversion of physical parameters has a higher accuracy, while the vertical superposition anomaly and physical properties are related to the separation accuracy of anomalies. 3) the study of high accuracy anomaly separation is the key to the inversion of depth constrained physical parameters of complex geological models. acknowledgement we acknowledge the support sullied by the national key research and development program (granted by 2017yfc0601504) and the valuable review comments proposed by the reviewers. references al farajat, m., schaefers, b., al hassanat, h., al atteyat, n., al jahed, n., & khataibeh, j. (2015). using gis and geophysics in selecting suitable basins with freshwater aquifers for an efficient exploration strategy a case study from petra-region, jordan. earth sciences research journal, 19(1): 39-50. boehm, c. & ulbrich, m. (2015). a semismooth newton-cg method for constrained parameter identification in seismic tomography. siam journal on scientific computing, 37(5), s334-s364. chen, y. (2015). iterative deblending with multiple constraints based on shaping regularization. ieee geoscience & remote sensing letters, 12(11), 2247-2251. deng, c. l., liu, h. j., song, b. h., & chen, j. (2014). density parameters inversion for structures based on regularization method. journal of guilin university of technology. fontchastagner, j., lubin, t., mezani, s., & takorabet, n. (2018). design optimization of an axial-field eddy-current magnetic coupling based on magneto-thermal analytical model. open physics, 16(1), 21-26. li, c. & zhang, f. (2017). ava inversion based on the l1-norm-based likelihood function and the total variation regularization constraint. geophysics, 82(3), 1-54. mazgaj, w., szular, z., & szczurek, p. (2017). calculations of magnetic field in dynamo sheets taking into account their texture. open physics, 15(1), 1034-1038. raknes, e. b. & arntsen, b. (2014). time-lapse full-waveform inversion of limited-offset seismic data using a local migration regularization. geophysics, 79(3), wa117-wa128. zhang, f., dai, r., & liu, h. (2014). seismic inversion based on l1-norm misfit function and total variation regularization. journal of applied geophysics, 109, 111-118. zhang, y,. yan, j., li, f., chen, c., mei, b., jin, s., & dohm, j. h. (2015). a new bound constraints method for 3-d potential field data inversion using lagrangian multipliers. geophysical journal international, 201(1), 267-275. zhu, z. q., cao, s. j., lu, g. y. (2014). 3d gravity inversion with bound constraint based on hyper-parameter regularization. the chinese journal of nonferrous metals, 24(10), 2601-2608. figure 12. abnormal value of the extension of 450 m of depth anomaly (left) and its inversion density (right) articulo 3.indd earth sciences research journal hydrogeology earth sci. res. sj. vol. 15, no. 2 (december, 2011): 101 108 hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india k. brindha and l. elango department of geology, anna university, chennai600 025, india. corresponding author. e-mail: elango34@hotmail.com, elango@annauniv.edu abstract hydrochemical study was carried out in madhuranthakam located near chennai in tamil nadu, india with an objective of understanding the suitability of local groundwater quality for domestic and irrigation purposes. twenty groundwater samples were collected in february 2002 and analysed for physical and chemical parameters. groundwater in this area was found to be within the desirable bureau of indian standards and world health organisation limits for drinking water. ca-hco 3 was the dominant groundwater type. groundwater in this area was assessed for irrigation purposes on the basis of sodium percentage (na%), magnesium hazard (mh), residual sodium carbonate (rsc), sodium absorption ratio (sar), permeability index (pi) and united states department of agriculture (usda) classification. most of the groundwater samples were suitable for irrigation, except in a few locations (15%) based on mh. overall the groundwater quality was suitable for drinking and domestic purposes and permissible for irrigation activities. resumen un estudio hidroquímico fue llevado en la localidad madhuranthakam cerca chennai en tamil nadu, india, con el propósito de evaluar la calidad de las aguas subterráneas locales para usos domésticos y fines de riego. veinte muestras de agua subterránea fueron recogidas y analizadas en términos de parámetros físicos y químicos en febrero 2002. las aguas subterráneas en esta área fueron encontradas aptas como potables según los limites sugeridos por el bureau estándares de india y los límites de la organización mundial de la salud para el agua potable. ca-hco 3 es dominante en las aguas subterráneas. las aguas subterráneas en esta área fueron asignada para propósitos de riego sobre la base del porcentaje de sodio (na%), magnesio peligroso (mh), carbonato sodio residual (rsc), proporción de absorción de sodio (sar), índice permeabilidad (pi) y la clasificación del departamento de agricultura de los estados unidos (usda). la mayoría de las muestras de las aguas subterráneas son aptas para el riego, excepto en unas pocas ubicaciones (15%), basadas sobre mh. en general, la calidad del agua subterránea fue apta para el consumo, uso doméstico y permisible para las actividades de riego. palabras claves: agua subterraneas, madhuranthakam, calidad de agua potable, calidad de agua de riego, tamil nadu, india. . keywords: groundwater, madhuranthakam, drinking water quality, irrigation water quality, tamil nadu, india. record manuscript received: 10/03/2011 accepted for publications: 24/11/2011 introduction groundwater is a precious source of fresh water, being the most distributed form on the earth, excluding the polar icecaps and glaciers. groundwater studies are gaining more importance in the present day as it is used for almost all purposes such as domestic, industrial and agricultural activities in most parts of the world. improper management of this replenishable resource may lead to groundwater contamination and scarcity. when some ions and minerals are present beyond the permissible limit, they become unsuitable for drinking and irrigation purposes, which may be due to natural and also anthropogenic causes. groundwater quality has been given lot of importance and studied worldwide (lahermo and backman, 1999; omo-irabor et al., 2008; baalousha, 2010). several regions in india have encountered degradation in groundwater quality too, due to rapid urbanisation and an exponential increase in population (ramesh and elango, 2005; brindha and elango, 2010; brindha et al., 2011). the present study was carried out with the aim of understanding groundwater quality and its suitability for drinking and irrigation purposes in madhuranthakam, located near chennai in tamil nadu, india. groundwater is the major source for domestic and irrigation practices in this area. there has been an increase in the demand for groundwater due to the growth of the local population. groundwater is usually put to direct use in rural areas without proper monitoring and treatment. groundwater may also become contaminated by the agrochemical products used for irrigation. the groundwater quality in the nearby regions, namely chennai, kancheepuram and chengalpet, k. brindha and l. elango102 has been studied earlier (elango and manickam, 1987; ramesh, 1999; rajmohan et al., 2000; elango et al., 2003a; elango et al., 2003b; kumaresan and riyazuddin, 2006). however, no studies have been carried out in the madhuranthakam region pertaining to groundwater quality. the suitability of groundwater for domestic and irrigation purposes thus had to be determined based on the presence of major ions in the groundwater of this region. the present study, which was carried out in 2002, will serve as baseline data for comparing future groundwater quality. study area the study area is situated 76 km south of chennai city and forms a part of madhuranthakam taluk, kancheepuram district, tamil nadu, india (figure 1). most of the annual rainfall of 1,206 mm is received from october to december and the rest during the southwest monsoon season from june to september. climatic condition in the study area varies from 39oc to 40oc during summer (april to june) and 20oc to 24oc during winter (october to december). the study area is surrounded by a number of tanks which are used for drinking and agricultural purposes. this area is intensively cultivated by pumping groundwater from dug wells as well as surface water resources. geologically this area is characterised by charnockites of precambrian era. outcrops of these rocks are found in many parts of the study area. sandy clay and clayey sand overlay the charnockites with thickness ranging from 1.5 to 5m. groundwater occurs under unconfined condition mainly in the weathered and fractured part of this charnockites in this area. figure 1. location of study area and sampling wells. parameter unit minimum maximum mean ph 7.2 8.2 7.69 ec µs/cm 200 1900 807.6 calcium mg/l 11.84 93.9 52.21 magnesium mg/l 3.65 132.46 24.37 sodium mg/l 27.59 82.76 50.35 potassium mg/l 0 2.74 1.17 carbonate mg/l 6 54.01 28.5 bicarbonate mg/l 30.51 256.27 136.9 chloride mg/l 20.35 32.41 21.91 sulphate mg/l 30.48 44.89 35.64 table 1. statistical summary of physicochemical parameters. hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india 103 sampling and analysis a well inventory survey was carried out in february 2002 to obtain background information about the well population, lithology, well use etc. forty-seven wells were investigated and 20 wells (figure 1) were chosen as representative wells for groundwater sampling based on electrical conductivity (ec). the groundwater samples were collected in clean 500 ml polyethylene bottles. the sampling bottles were soaked in 1:1 diluted hcl solution for 24 hours and washed twice with distilled water before sampling. they were washed again in the field with groundwater sample filtrates. the groundwater samples were collected from bore wells after pumping the water for about 10 minutes. field parameters such as ec and ph were measured in the field using portable digital meters. samples were analysed for major ions in the laboratory using the standard recommended methods (apha, 1998). sulphate concentration in the groundwater samples was analysed using a uv visible spectrophotometer. sodium and potassium content was determined by using a flame photometer and calcium, magnesium, chloride, carbonate and bicarbonate by titration technique. total ions measurement precision was checked by calculating the ion balance error (ibe). the ibe was within ±10%. total dissolved solids (tds) were calculated by using the formula: tds (mg/l) = ec (µs/cm) x 0.64; total hardness (th) was calculated by using: th = 2.497ca + 4.115mg in mg/l. results and discussion table 1 shows the maximum, minimum and mean concentrations of various parameters. the order of cation dominance was ca2+>na+>mg2+>k+ and hco3 ->so4 2->co3 2->clfor anions. the general chemical nature of groundwater can be understood by plotting major cation and anion concentrations on a piper trilinear diagram (piper, 1944). ca-hco3 was the major water type dominant in this area (figure 2). the next dominant water type was mixed ca-na-hco3. drinking water quality groundwater used for domestic purposes, such as drinking and cooking, should be free from toxic chemicals and pathogens. domestic water quality indicates that a particular parameter at a given concentration may be suitable for the human body beyond which it is unsuitable. the concentration of various ions in the groundwater samples was compared with bureau of indian standards (bis, 2003) and world health organisation (who, 1993) standards, which are given in table 2 wherein all the groundwater samples were found to be within the suitable limits. the groundwater in this area was thus seen to be fit for domestic consumption, based on the major ion analysis carried out in this study. the groundwater samples were classified regarding tds and th (tables 3, 4 and 5). most of the groundwater was fresh and suitable for drinking purposes based on tds (freeze and cherry, 1979; davis and deweist, 1966). groundwater in this study area varies from soft to very hard (sawyer and mccarty, 1967). irrigation water quality good quality irrigation water is essential for achieving maximum crop productivity. groundwater suitability for irrigation purpose in this study area was assessed using sodium percentage (na%), magnesium hazard (mh), residual sodium carbonate (rsc), sodium absorption ratio (sar), permeability index (pi) and united states department of agriculture (usda) classification. groundwater in most of the study area was found to be suitable for irrigation. irrigation water having high ec content will affect root area and water flow. groundwater in this area was grouped according to the guidelines figure 2. groundwater type. k. brindha and l. elango104 parameter bis (2003) who(1993) percentage of samples above maximum allowable limit highest desirable limit maximum permissible limit highest desirable limit maximum permissible limit ph 6.5– 8.5 6.5 – 9.2 6.5 – 8.5 6.5 – 9.2 nil th (mg/l) 300 600 100 w500 nil calcium (mg/l) 75 200 75 200 nil magnesium (mg/l) 30 100 50 150 nil chloride (mg/l) 250 1000 200 600 nil sulphate (mg/l) 150 400 200 400 nil sodium (mg/l) 200 nil table 2. comparison of groundwater samples with bis and who standards. table 3. freeze and cherry (1979) classification of groundwater based on tds (mg/l). table 4. david and deweist (1966) classification of groundwater based on tds (mg/l). table 5. sawyer and mccarty (1967) classification of groundwater based on th (mg/l). table 6. suitability for irrigation based on usda classification. tds (mg/l) water type number of samples percentage <1,000 fresh 19 95 1,000 10,000 brackish 1 5 10,000 1,00,000 saline nil nil >1,00,000 brine nil nil tds (mg/l) classification number of samples percentage <500 desirable for drinking 11 55 500 1,000 permissible for drinking 8 40 1,000 3,000 useful for irrigation 1 5 >3,000 unfit for drinking and irrigation nil nil total hardness (mg/l) type of water number of samples percentage <75 soft 2 10 75 150 moderately high 7 35 150 300 hard 7 35 >300 very hard 4 20 ec (µs/cm) salinity class percentage of samples remark on quality <250 c1 5 excellent or low 250-750 c2 45 good or medium 750-2250 c3 50 permissible or high 2250-5000 c4 nil unsuitable or very high hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india 105 established by the united states salinity laboratory (freeze and cherry, 1979) based on ec (table 6). this showed that groundwater in this area was permissible for irrigation. sodium is an important parameter for irrigation water and is denoted as na% which was calculated from the formula given in equation 1 (wilcox, 1955) and all concentrations were expressed in meq/l. the na% in this area is given in table 7. groundwater was suitable for irrigation in 60% of the samples while 40% were permissible to doubtful. ec and na% are plotted in figure 3 which showed that most of the groundwater samples were good for agriculture. the concentration of bicarbonate and carbonate higher than calcium and magnesium will influence the suitability of water for irrigation purposes. the rsc value was computed using the following formula (equation 2) where ions were expressed in meq/l. rsc = (co 3 2+ hco3 -) – (ca2+ + mg2+) however, regarding rsc, all samples fall within the safe category for irrigation. sar is another important parameter for determining the desirability of irrigation water. the sar values were calculated by using equation 3 (richards, 1954), where all the concentrations were given in meq/l. all the groundwater samples collected from this area were excellent on the basis of sar. water used for irrigation can be classified into four typesc1, c2, c3 and c4 based on salinity hazard and s1, s2, s3 and s4 based on sodium hazard. figure 4 shows the plot of groundwater samples grouped on the above basis. most groundwater samples fall under c1s1 and c2s1 which are suitable for irrigation. comparatively few samples were grouped under c3s1 (which is permissible for irrigation). thus the groundwater in this area was seen to be suitable for irrigation based on salinity hazard and sodium hazard. mh for irrigating water was calculated using the formula mh = mg/ (ca + mg) x 100, where concentrations were given in meq/l (szabolcs and darab, 1964). mh above 50 is considered to be unsuitable for irrigation. groundwater had an mh above 50 in 15% of the samples, thus not being fit for irrigation. table 7. suitability for irrigation based on sodium percent. % na suitability for irrigation number of samples percentage <20 excellent 5 25 20-40 good 7 35 40-60 permissible 6 30 60-80 doubtful 2 10 >80 unsuitable nil nil (1) (2) (3) )( 100)( % 22 ++++ ++ +++ + = knamgca xkna na 2 22 ++ + + = mgca na sar figure 3. suitability of irrigation water, based on ec and sodium percent. k. brindha and l. elango106 the suitability of groundwater for irrigation based on pi was calculated using equation 4, where concentrations were in meq/l. class i and class ii waters are considered to be good and suitable for irrigation while class iii water is unsuitable for irrigation (doneen, 1964). figure 5 shows that one groundwater sample was not suitable for irrigation based on pi whereas the rest of the samples were good. in general, the groundwater fall within the permissible category for irrigational use, except for a few locations where it was unsuitable based on mh. conclusion groundwater quality of an area must be studied to understand its suitability for domestic and irrigation purposes. all the groundwater samples collected from the madhuranthakam area, tamil nadu, india, showed that the major ions fall within the permissible range. the dominant groundwater type was ca-hco 3. based on tds, 95% of the groundwater was fresh and permissible for drinking. the groundwater varied from soft to very hard on the basis of th. the suitability of groundwater for irrigation was assessed from na%, ec, rsc, sar, mh, pi and usda classification. the groundwater in this area was seen to be good and suitable for drinking and domestic purposes. however, the groundwater was unsuitfigure 4. irrigation water salinity and alkalinity hazard. figure 5. classification of irrigation water based on permeability index. )( 1003 +++ + ++ + = namgca xhcona pi (4) hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india 107 able for irrigation in a few places, based on mh. overall, groundwater in the madhuranthakam area remains usable. the present groundwater quality status must be maintained by taking precautionary measures such as rainwater harvesting, less use of chemical fertilisers and the ongoing monitoring of groundwater quality in this region. acknowledgements the authors would like to thank the department of science and technology-funds for improvement in science and technology (dst-fist) (grant no. sr/fst/esi-106/2010) and the university grants commission-special assistance programme (ugc-sap) (grant no. ugc drs ii f.550/10/drs/2007(sap-1)) for providing financial support which helped in creating facilities to carry out this work. references apha (1998). (1998). standard methods for the examination of water and wastewater, 20th edn, american public health association, washington dc, usa. baalousha, h. (2010). assessment of a groundwater quality monitoring network using vulnerability mapping and geostatistics: a case study from heretaunga plains, new zealand, agricultural water management. 97, 240–246. bis (2003). bureau of indian standards specification for drinking water, is: 10500:91. revised 2003, bureau of indian standards, new delhi. brindha, k. and elango l. (2010). study on bromide in groundwater in parts of nalgonda district, andhra pradesh, earth science india, 3, no. 1, 73-80. brindha, k., rajesh, r., murugan, r. and elango, l. (2011). fluoride contamination in groundwater in parts of nalgonda district, andhra pradesh, india, environmental monitoring and assessment, 172, 481-492. davis, s. n. and dewiest, r. j. (1966). hydrogeology, wiley, new york. doneen, l. d. (1964). water quality for agriculture, department of irrigation, university of calfornia, davis, 48 pp. elango, l. and manickam, s. (1987). hydrogeochemistry of the madras aquifer, india spatial and temporal variation in chemical quality of groundwater, geological society of hong kong bulletin, no. 3, 525-534. elango, l., kannan, r. and senthilkumar, m. (2003a). major ion chemistry and identification of hydrogeochemical processes of groundwater in a part of kancheepuram district, tamil nadu, india, environmental geosciences, 10, no. 4, 1-10.  elango, l., kumar, s. s. and rajmohan, n. (2003b). hydrochemical studies of groundwater in chengalpet region, indian journal of environmental protection, 23, no. 6, 624-632. freeze, r. a. and cherry, j. a. (1979). groundwater, prentice hall inc. englewood cliffs, new jersey, 604 pp. kumaresan, m. and riyazuddin, p. (2006). major ion chemistry of environmental samples around sub-urban of chennai city, current science, 91, no. 12, 1668-1677. lahermo, p. and backman, b. (1999). nitrates in groundwater in finland: the most endangering quality problem, hydrogeology and land use management, 329–333. omo-irabor, o.o.,  olobaniyi, s.b., oduyemi k. and akunna, j. (2008). surface and groundwater water quality assessment using multivariate analytical methods: a case study of the western niger delta, nigeria, physics and chemistry of the earth, parts a/b/c, 33, no. 8-13, 666-673. piper, a. m. (1944). a graphical procedure in the geochemical interpretation of water analysis, transactionsamerican geophysical union, 25, 914-928. rajmohan, n., elango, l., ramachandran, s. and natrajan, m. (2000). major ion correlation in groundwater of kancheepuram region, south india, indian journal of environmental protection, 20, no.3, 188-193. ramesh, k. and elango, l. (2005). groundwater quality assessment in tondiar basin, indian journal of environmental protection, 26, no. 6, 497 –504. ramesh, r. (1999). groundwater quality management: pollution perspectives impacts of urban growth on surface water and groundwater quality, proceedings of iugg 99 symposium hs5, birmingham, july, iahs publ. no. 259, 47-55. richards, l. a. (1954). diagnosis and improvement of saline and alkali soils, us department of agriculture, hand book 60. sawyer, g. n. and mccarty, d. l. (1967). chemistry of sanitary engineers, 2nd edn, mcgraw hill, new york, 518 pp. szabolcs, i. and darab, c. (1964). the influence of irrigation water of high sodium carbonate content of soils, proceedings of 8th isss, trans vol. ii , 802-812. who (1993). guidelines for drinking water quality, vol. 1, 2nd edn, recommendations, who, geneva, 130 pp. wilcox, l. v. (1955). classification and use of irrigation waters, usda, circular 969, washington dc, usa. keywords: three-dimensional modelling; key algorithms; geological bodies; virtual display. palabras clave: modelado tridimensional; algoritmos clave; cuerpos geológicos; pantalla virtual. how to cite item feng, w. (2019). virtual display system of geological body based on key algorithms of three-dimensional modelling. earth sciences research journal, 23(3), 185-189. doi: https://doi.org/10.15446/esrj.v23n3.81809 to study the virtual display system of geological body, the key algorithm of three-dimensional modelling is used to analyze it, and the virtual display scheme of geological body is realized. three-dimensional modelling of geological bodies is a frontier field of current geoscience research, and also a hot issue in computer science. its research can complement and improve the theoretical basis of virtual reality technology and expand its application fields. geological body three-dimensional modelling software can meet the needs of geoscientists to study geological problems more effectively from three-dimensional space, and has considerable practical value. in the network environment, users' information needs are diversified and individualized, and special attention is paid to the breadth, novelty and timeliness of information. faced with massive information resources, the information users need is a drop in the ocean in the huge and disorderly network information space. users need to spend a lot of time and energy to find valuable information in all kinds of uniform systems. traditional system information services can no longer help users effectively get rid of information overload and information lost. therefore, the concept of personalized service system has become a hot spot of people' s attention as soon as it is put forward. in recent years, personalized systems have emerged in endlessly. in the early stage of formulation and implementation of personalized service strategy, the existing personalized service generally lacks the attention to context information such as user' s environment. there are weak links in personalization and intellectualization of service, which damages user experience. therefore, shepard algorithm and kriging algorithm of three-dimensional modelling are used to study the virtual display system of geological bodies. the results show that these two key algorithms play a very important role in the research process of geological virtual display system, and also provide a new idea for further research on the popularization of geological knowledge. abstract virtual display system of geological body based on key algorithms of three-dimensional modelling sistema de visualización virtual del cuerpo geológico basado en algoritmos clave de modelado tridimensional issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n3.81809 para estudiar el sistema de visualización virtual de un cuerpo geológico se utiliza el algoritmo clave del modelado tridimensional, y se realiza el esquema de visualización virtual del cuerpo geológico. el modelado tridimensional de cuerpos geológicos es un campo fronterizo de la investigación actual en geociencias, y también un tema importante actualmente en la informática. su investigación puede complementar y mejorar las bases teóricas de la tecnología de realidad virtual y ampliar sus campos de aplicación. el software de modelado tridimensional del cuerpo geológico puede satisfacer las necesidades de los geocientíficos para estudiar problemas geológicos de manera más efectiva desde el espacio tridimensional, y tiene un valor práctico considerable. en el entorno de red, las necesidades de información de los usuarios están diversificadas e individualizadas, y se presta especial atención a la amplitud, novedad y puntualidad de la información. frente a los recursos de información masivos, la información que los usuarios necesitan es una gota en el océano, en el enorme y desordenado espacio de información de la red. los usuarios necesitan gastar mucho tiempo y energía para encontrar información valiosa en todo tipo de sistemas uniformes. los servicios tradicionales de información del sistema ya no pueden ayudar a los usuarios a deshacerse efectivamente de la sobrecarga de información y la información perdida. por lo tanto, el concepto de sistema de servicio personalizado se ha convertido en un punto de atención de las personas tan pronto como se presenta. en los últimos años, los sistemas personalizados han surgido en forma interminable. en la etapa inicial de formulación e implementación de la estrategia de servicio personalizado, el existente generalmente no presta atención a la información de contexto, como el entorno del usuario. existen vínculos débiles en la personalización e intelectualización del servicio, que dañan la experiencia del usuario. por lo tanto, el algoritmo de shepard y el algoritmo de kriging del modelado tridimensional se utilizan para estudiar el sistema de visualización virtual de cuerpos geológicos. los resultados muestran que estos dos algoritmos clave desempeñan un papel muy importante en el proceso de investigación del sistema de visualización virtual geológico, y también proporcionan una nueva idea para futuras investigaciones sobre la divulgación del conocimiento geológico. resumen record manuscript received: 20/05/2019 accepted for publication: 15/07/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 3 (september, 2019): 185-189 wei feng institute of mud logging technology and engineering, yangtze university, hubei jingzhou, 434023, china g eo lo g y 186 wei feng introduction china is a country with large geological resources. generally speaking, china has a well-developed stratum, diverse sedimentary types, complex geological structures, frequent coexistence of magmatic activities in active zones and stable areas, and diverse types of long-term geohistory metamorphism with different metamorphic degrees (sinitski, thompson, godsell, honey, & besemann, 2018). china is one of the regions with important geological characteristics in the world, and also one of the important areas to study the evolution of continental geological structures, especially the mesozoic and cenozoic crustal structures (naqvi, ali hassan, pervaiz, & ni, 2018). therefore, in order to better and more comprehensively understand the current situation of geological resources in china, many researchers use three-dimensional modelling algorithm and virtual display technology to study geology, which can not only help people to have more knowledge of geography, but also help researchers to study geological strata (handong, tuo, & changhong, 2006). this is better than simple data for researchers to have a deeper understanding and experience of geological formations. visualization of scientific computing emerged in the late 1980s (li, cai, & lu, 2018). it uses computer technology to convert the data of scientific or engineering calculation process and calculation results into graphics or images, and display these graphics or images on the screen for interactive processing theory, methods and techniques (wu, zhu, lin, zhuang, & ma, 2018). three-dimensional geological modelling technology is a branch of scientific computing visualization technology. it is the development and application of scientific computing visualization technology in the field of geosciences. therefore, in order to exploit and utilize the underground resources according to the specific national conditions of china, it is necessary to clarify the shape, attributes and spatial relationship of the underground geological structure, understand the basic structural and material characteristics of the geology, so as not to blindly exploit and utilize the underground resources and avoid unreasonable development and loss (liu et al., 2018). however, due to the late start of three-dimensional geological modelling technology in china, the lack of geological data, and the complexity, fuzziness and uncertainty of geological spatial distribution and structure, it is difficult to study threedimensional geological modelling and visualization (zhao, wang, & martinez, 2018). however, with the development of science and technology, some new methods and techniques have emerged in the exploration and acquisition of underground resources. people can obtain more detailed, comprehensive and abundant geological data through high-tech, which also provides material and data basis for the establishment of three-dimensional geological visualization model (buttussi & chittaro, 2018). on this basis, an intuitive, clear and complete three-dimensional geological visualization model is established, and the structure and characteristics of geological bodies are presented concisely and concisely by means of virtual images (zhu, lu, & rätsch, 2018). this model can not only describe the geological structure and characteristics comprehensively, but also relieve the pressure of staff members. at the same time, it can also provide certain theoretical basis for relevant staff to make reasonable exploration and development decisions. on the other hand, it is also conducive to the popularization of geological knowledge. methodology key algorithms for 3d geological modelling geological body is a term often used in geological work. it usually refers to the product of geological processes that occupy certain space in the earth’ s crust, especially the inherent components and can be distinguished from the surrounding materials. no matter how large or small the geological body is, whether it reaches a rock mass or an inclusion, it can be called a geological body. in geophysical prospecting, it refers to the object to be looked for when applying geophysical methods to study geological problems. therefore, geological body is often the geological factor causing anomalies. sometimes, it is common with anomalies. geological bodies occupy a certain space in the crust and have certain composition, which is a natural entity different from its surroundings. the study of geological bodies is of great significance in searching for various useful deposits and seismotectonic active belts. the data format and data structure of geological data play an important role in the three-dimensional geological body modelling. its organization mode determines the methods used in the three-dimensional geological modelling. therefore, understanding the data organization commonly used in the threedimensional geological modelling is of great value to the three-dimensional geological modelling. the geological data which are studied more at present is chosen as the research object, that is, borehole data. it provides data support for the algorithm research of three-dimensional geological modelling. diamond data is the most common type of data in three-dimensional geological modelling. borehole data is obtained by engineering drilling method. it can directly obtain detailed rock and soil distribution by drilling core. it is necessary to obtain borehole data, distribution data of boreholes and strata, and the description of the global bottom, which lays a theoretical foundation for the follow-up study of geological virtual display system. borehole data is very important in three-dimensional geological modelling because of its intuitive, accurate and detailed characteristics. the flow chart of three-dimensional geological modelling of borehole data is shown in figure 1. stratigraphic division is to determine how many strata there are in a geological body and the interface data of each stratum. assuming that there are five drilling points in a drilling area and obtain data from them, each drilling point has coordinates of x and y directions. geological information obtained at different heights may be different. stratigraphic division is based on the numerical information of each drilling hole in different z coordinates. in this study, shepard processing algorithm and kriging algorithm are used to process it. the corresponding interpolation models are as follows: ξ µ µ x y z r x y r x y z r ii n i i i i n , , , ( ) = ( )  ( )  =       = = ∑ ∑ 1 1 1 0       = ri 0 (1) r x y x x y yi i i,( ) = ( ) ( )− + −2 2 represents the distance from the interpolated point (x, y) to the sampling point (xi, yi ). the value of  is taken as 2. the influence of the function value on the weight function is inversely proportional to the distance from the interpolated point to the sampling point. the larger the distance is, the smaller the impact is. the smaller the distance is, the greater the impact is. the two points are close to each other and have similar relationship in geological data, that is, the properties of the adjacent points are similar. kriging method is a spatial local interpolation method. it evaluates regionalized variables in the interpolation region by using covariance function, variation function and other models. the formula is as follows: z x z xi i i n * 0 1 ( ) = ( ) = ∑ (2) where z x* 0( ) represents the value of the point to be interpolated, z xi( ) represents the value of the first sampling point, and i is the effect of the i-th sampling point on the interpolation point. in kriging algorithm, the ordinary kriging method has been widely studied in the field of geology and has shown good results. therefore, due to the uncertainty of drilling location and geological structure, some variables in the address data show random and structural characteristics and regionalized variables in spatial distribution. the difference between regionalized random variables can be expressed by spatial covariance. cov z x z x h e z x z x h e z x e z x h( ) ( )  = ( ) ( )  ( )  ( ) , + + − + (3) the variation function of z x( ) at points x and x+h is denoted as  x h,( ), and its expression is as follows: 187virtual display system of geological body based on key algorithms of three-dimensional modelling environment through three-dimensional equipment. moreover, people can freely control the three-dimensional graphics environment to achieve specific purposes. virtual reality technology has the characteristics of immersion, interaction, multi-perception and sense of autonomy. its greatest feature is that people have natural functions and perception, and can interact with computergenerated virtual systems in real time. virtual reality system provides us with a convenient and realistic way to communicate with computers. it is a powerful tool for us to understand and transform the world. it has attracted the attention and research of many wellknown scholars all over the world. through the efforts of various researchers, it has made some brilliant achievements. it can be foreseen that with the rapid development of computer hardware and software technology, virtual reality technology will further develop and change our lives. as a result, the national geological information library successfully launches the virtual exhibition hall. on the one hand, it can give full play to the role of scientific popularization of geological data and documents, and popularize geological knowledge. on the other hand, it also provides scientific and technological means for researchers to study geological conditions and disaster situations more comprehensively. in the specific design of the display system, many perceptrons, reasoners, ontologies and service providers are involved in the structural model. next, these key points are elaborated in detail. by constructing a series of generic ontologies, a bridge for context-aware intelligent mobile services can be built to communicate and express with each other. in this three-dimensional virtual display platform, the ontology structure is divided into several sub-ontologies, which are object ontology, service ontology, computing ontology, perception ontology and storage ontology. object ontology includes user ontology and product ontology. service ontolo gy includes services. when the system decides that users need some information, γ − + − + x h var z x z x h e z x z x h,( ) = ( ) ( )  = ( ) ( )  1 2 1 2 2 − − + 1 2 2 e z x z x h( ) ( ) { } (4) on this basis, unbiased estimates are made, and the estimated variance is as follows: σ − − λe i i i n e z x z x e z x z x2 2 1 2 = ( ) ( )   = ( ) ( )         = ∑* (5) σ − λ + µe i i i n c x x c x x2 1 = ( ) ( ) = ∑, , (6) by analyzing the variance of kriging estimation, whether the estimated value is the best one can be judged. virtual display system as one of the emerging technologies, spatial information has become the main source of information for human beings to understand the real world around them and explore the unknown world. establishing realistic largescale virtual scene can help us understand spatial information. therefore, in recent years, researchers pay more attention to virtual simulation technology. moreover, it can realize human-computer interaction. computer generates three-dimensional graphics and makes people feel three-dimensional start select drilling hole and extract drilling data standardized processing of drilling data underlying partition algorithm generate low level data ground surface reconstruction algorithm texture mapping algorithm for true feeling of plane end 3 d geological modeling algorithm for zone faults fast slicing algorithm construct 3d geological body model satisfied ？ three dimensional analysis of the model y n figure. 1 flow chart of 3-d geological modelling based on borehole data 188 wei feng it feeds back the information to users and service system to recommend products to users in time. computing ontology contains three reasoners, which is the core part of the system. there are four sensors in perceptual ontology, and storage ontology is fused in the above ontology in practical application. to deal with the huge and strict data information, it is necessary to use data storage and computing tools in colleges and universities, as shown in figure 2. results and discussion this study collects borehole data of a certain area, as shown in table 1. table 1 borehole data of a certain area geological condition well drilling p[1] well drilling p[2] well drilling p[3] well drilling p[4] below depth (1) 1 1 2 1 below depth (2) 3 2 3 2 below depth (3) 5 3 5 4 below depth (4) 6 5 6 5 below depth (5) 7 7 7 6 stratigraphic division of geological conditions is carried out by the algorithm, and the results are obtained, as shown in figure 3. the three-dimensional geological model thus established is shown in figure 4. the result of virtual display is shown in figure 5. in addition, through the analysis of the calculation results of the algorithm, it can be seen that the parallel algorithm of shepard method and kigring method is more efficient than that of the single one in the case of two cores. moreover, in the case of four cores, the efficiency of the parallel algorithm is much higher than that of the two cores, which shows that the parallel algorithm can reduce the events of program execution and improve the efficiency of modelling. it is also very helpful to the operation of virtual display system. in summary, on the basis of collecting relevant data of geological body, using the key algorithm of three-dimensional modelling not only improves the operation efficiency, but also provides a good data support for the establishment of virtual display system, so that the final image can display the specific effect well, which is conducive to people’ s further understanding of geological conditions and provides a good model for researchers to study. user ontology product ontology reasoning machine i reasoning machine ii reasoning machine iii sensor i sensor ii sensor iii sensor iv storage of ontology service i service ii figure. 2 five ontology structures and their relationships figure. 3 stratigraphic division result map figure. 4 main interface of 3d geological simulation system 189virtual display system of geological body based on key algorithms of three-dimensional modelling conclusion with the continuous development of virtual display technology, researchers have a better understanding of the relevant information of geological bodies, and use the three-dimensional modelling algorithm to analyze, and ultimately complete the design of the display system. based on the three-dimensional geological modelling process of borehole data, the three-dimensional geological modelling modules are divided. the algorithm of stratum division and surface reconstruction are studied and realized. in the algorithm of surface reconstruction, shepard and kriging are used to implement the two methods respectively, and the two algorithms are analyzed and compared in detail. the design and research of the two algorithms in the geological virtual display system are realized. however, in the design of virtual display system, less consideration is given to the data that may affect the model, so more relevant data will be studied in the future. this study also provides ideas for other aspects of three-dimensional modelling, and has a reference significance for the wider application of virtual display technology. references buttussi, f., &chittaro, l. (2018). effects of different types of virtual reality display on presence and learning in a safety training scenario. ieee transactions on visualization and computer graphics 24.2,1063-1076. handong, t., tuo, t., & changhong, l. (2006). the parallel 3d magnetotelluric forward modelling algorithm. applied geophysics 3.4,197-202. liu, m., zhu, c., cui, t., zhang, h., zheng, w., & you, s an alternative algorithm of tunnel piston effect by replacing three-dimensional model with two-dimensional model. building and environment 128,55-67. naqvi, s. a. r., ali hassan, s., pervaiz, h., & ni, q. (2018). drone-aided communication as a key enabler for 5g and resilient public safety networks. ieee communications magazine 56.1,36-42. sinitski, e. h.., thompson, a. a., godsell, p. c., honey, j., & besemann, m. (2018). postural stability and simulator sickness after walking on a treadmill in a virtual environment with a curved display. displays 52,17. li, w. d., cai, y. l., & lu, w. f. (2007). a 3d simplification algorithm for distributed visualization. computers in industry 58.3,211-226. wu, w., zhu, h., lin, j., zhuang, x., & ma, g. (2018). tunnel stability assessment by 3d dda-key block analysis. tunnelling and underground space technology 71,210-214. zhao, r., wang, y., & martinez, a. (2018). a simple, fast and highlyaccurate algorithm to recover 3d shape from 2d landmarks on a single image. ieee transactions on pattern analysis and machine intelligence 40.12,3059-3066. zhu, x., lu, h., & rätsch, m. (2018). an interactive clothing design and personalized virtual display system. multimedia tools and applications 77.20,27163-27179. figure. 5 virtual display result image after the late 1990s, gnss/gps network rtk technology has become the preferred technique in the world, today widely used in various applications such as monitoring, early warning systems, and mapping and engineering applications. by the end of 2008, the cors-tr network was being used for engineering applications to determine 3d positions across the whole of turkey. on the other hand, single base rtk stations are being operated in turkey. the yldz is one of these types of stations. it was established at the yıldız technical university by the department of geomatic engineering in 2012. in this paper, the performance of the yldz station is examined by use of measurements in 5 istanbul gps triangulation network (igna) benchmark points, with five repeatability measurements up to 50 km from the station. the cors-tr network rtk solution is used at the same points, and the results are compared. the measured coordinates produced by the two rtk solutions are compared with the igna point coordinates. satellite geometry, initialisation time and the standard deviation of repeatability are also determined. the differences between measured and known coordinates are calculated as 1-2 cm horizontally and close to 3 cm vertically using the yldz single base rtk method. después de la década de los 90, el sistema global de navegación por satélite/sistema de posicionamiento global (gnss/gps, en inglés), basado en tecnología de navegación cinética satelital en tiempo real (rtk, del inglés real time kinematic) se convirtió en la técnica preferida en el mundo y que todavía se utiliza en varias aplicaciones como el monitoreo, sistemas de alerta temprana y mapeo e ingeniería de aplicativos. para finales de 2008, los aplicativos de ingeniería utilizan la red de estaciones de referencia de funcionamiento continuo (cors-tr, un proyecto establecido en turquía) para determinar posiciones en tres dimensiones a lo largo del territorio turco. por otro lado, las estaciones rtk de una sola base han funcionado en turquía. la yldz es una estación de este tipo que fue establecida en la universidad técnica de yildiz por el departamento de ingeniería geomática en 2012. en este trabajo se examina el desempeño de la estación yldz a través de la medición de cinco puntos de referencia del sistema de triangulación gps de estambul (igna, en inglés) con cinco mediciones repetitivas a más de 50 km de la estación. similares medidas rtk se hicieron con la red cors-tr y se compararon los resultados. las dos mediciones producidas por las soluciones rtk se compararon con los puntos de coordenadas igna. también se determinó la geometría satelital, el tiempo de inicialización y la desviación estándar de la repetibilidad. las diferencias entre las coordenadas medidas y las coordenadas conocidas se calcularon en un rango de 1-2 centímetros horizontalmente y cerca de 3 centímetros verticalmente para la estación yldz con el método rtk. abstract resumen keywords: rtk gnss, cors, single base rtk palabras clave: navegación cinética satelital en tiempo real, sistema global de navegación por satélite, estaciones de referencia de funcionamiento continuo, base única rtk. record manuscript received: 12/06/2015 accepted for publication: 28/07/2015 how to cite item aykut, o.a., gülal e., and akpinar b. (2015). performance of single base rtk gnss method versus network rtk. earth sciences research journal, 19(2) 135 139. http://dx.doi.org/10.15446/esrj.v19n2.51218 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n2.51218 desempeño de un método de navegación cinética satelital en tiempo real (rtk) de una sola base en comparación de un sistema rtk earth sciences research journal earth sci. res. j. vol. 19, no. 2 (december, 2015): 135 139 g e o d e sy performance of single base rtk gnss method versus network rtk nedı̇m onur aykut*1, engı̇n gülal1, burak akpinar1 1yıldız technical univesity, department of geomatics engineering, esenler, istanbul, turkey 136 nedi̇m onur aykut, engi̇n gülal and burak akpinar http://www.ggntr.com/ link. on the other hand, as of august 2014, the yldz station is a euref proposed station with a 20815m001 number. the coordinates of the yldz station are determined by processing the gnss data by gamit/globk software in irtf2008 datum, using 38 reference stations in the igs network. the velocities of these points are millimetre sensitive for calculating orbit information and earth rotation parameters spread over a 1000 km diameter of the yldz point. in this data process, the igs final ephemeris data was used to calculate precise coordinates and velocities. the coordinates and standard deviation of yldz station were then transformed to itfr96 datum for national applications (table 1). table 1. cartesian coordinates and standard deviation of yldz station introduction global navigation satellite systems (gnss) provides various kinematic and post processing methods of positioning solutions (feng and wang 2008). different real-time kinematic (rtk) gnss methods have been used for determining the position of any point on the globe. the differential gnss method has been used since 1980, although rather for navigation applications, using pseudo-range data with the position accuracy of submetres. carrier phase data has been used in real time kinematic applications since 1990 to increase the position accuracy at a centimetre level (langley 1998). both of these methods require at least two receivers. one receiver is the reference, the other is the rover. the correction data is transmitted by radio wave or satellite communication. a typical rtk gps accuracy may be given as 8 mm + 1 ppm horizontally and 15 mm + 1 ppm vertically (spectra precision, manual). this accuracy varies depending upon the distance between the reference and moving receiver. the limitation of rtk is due to distance-dependent biases, such as orbit bias, ionosphere bias and troposphere bias (rizos and han 2003). this restricts the distance to 10 kilometres or less. the network rtk (nrtk) or continuously operating reference stations (cors) has been developed to overcome the distance dependency by using a network of gnss reference stations spread over a wide area (rizos 2003). in turkey, the cors tr network has been used in the operation of 150 reference stations since 2008. single base rtk gnss at least two receivers are required for classical or single base rtk survey. the first receiver is the base receiver, which is fixed to a point with known coordinates. the other receiver is the rover, used for measuring the unknown coordinates of the points in real time. there is a communication link between both receivers to send corrections from base to the rover receiver by radio waves or gsm links. the real time kinematic gnss employs a method of carrier phase differential gnss positioning, whereby users can obtain centimetre level position accuracy in real time (langley 1998). the number of full waves between the receiver and satellite cannot be accounted for at the initial signal acquisition (xu 2007). the ambiguous number of these waves in carrier phase measurement is resolved by on-the-fly (otf) techniques in real time (hu et al. 2003). the pseudo-range phase is measured at the base station a at epoch t0 may be represented by the equation given below (hoffmann et al. 2008): (1) is the geometric range, is the satellite-receiver dependent bias, is satellite dependent only, is receiver dependent only. is the phase ambiguity. (2) described in equation 2 are the results for corrected phase ranges where and is the difference of phase ambiguities. the point positioning at the rover site b is performed with the corrected pseudorange phase (hoffmann et al. 2008). the yldz single base rtk station is operated by the department of geomatic engineering, located in yildiz technical university in istanbul, turkey. the aerat1675_120 spke gnss antenna is located on the roof of the civil faculty. gnss data is collected by the ashtech proflex pf800 receiver located in the geodetic metrology laboratory of the department. the proflex web server is used for the ntrip caster and for monitoring the base station. the corrections are sent for real time applications via internet communication. the yldz single base rtk system ip is 193.255.212.232 and a 2101 port. users can access the web page of the station using the network rtk the limitation of the single base rtk technique is the distance between the base receiver and the rover receiver due to distancedependent biases; namely, orbit bias, ionosphere bias and troposphere bias. techniques have been developed to overcome this distance dependence using a network of gnss reference stations spread over a wide geographic area (hu et al. 2003; rizos and han 2003). the aim of network rtk (nrtk) is to compute network corrections from the observations of several reference stations and send them to the user (wegener and wanninger,2005). nrtk requires an ambiguity resolution ‘engine’ which will fix the integer ambiguities between the static multiple reference receivers, located at stations of known position, and so make up the network (rizos and han 2003). the reference station separation is usually restricted to less than 100 km to allow for quick and reliable ambiguity resolution. as the number of stations increases, better corrections can be attained (ahmed el-mowafy 2012). there are several correction computing solutions, such as the vrs (virtual reference stations), mac (master auxiliary concept), prs (pseudo reference stations) and fkp (flächenkorrektur parameter) methods. the vrs technique is currently the most popular and efficient method of transmitting corrections through a data link to the network users for rtk positioning (hu et al., 2003). the basic theory of the vrs method is to transform measurements made at real reference stations to the location of the vrs and, therefore, to a different location at the same epoch. the pseudo-range phase is described by equation 3 after considering a substitution of f s=c/λs. on the left side, is the measured carried phase which is modeled by , the geometric distance between receiver and satellite, integer ambiguity and , the combined receiver and satellite clock bias (hoffmann et al. 2008). (3) the terms and are changed in accordance with varying locations in this equation. the receiver r is once located at the real reference station a, represented by the coordinate factor xa, and once at the vrs, represented by xv. the results of these two locations from (3) are the two following equations (hoffmann et al. 2008). (4) (5) 137performance of single base rtk gnss method versus network rtk satellite orbit errors, ionospheric and tropospheric refraction errors are included in term, which is modelled or solved by all baselines of the reference network. calculating the differences of these equations gives the following result. (6) turkey has had all of the covered network, named cors-tr or tusaga active, which currently has 150 reference stations (4 of them are located in north cyprus) providing 24/7 rtk positioning applications, geodetic/cadastral surveys, modelling the atmosphere researches, monitoring plate tectonics, earthquake prediction early warning systems and gis applications (gülal et al., 2013) in itrf96 datum and 2005.0 epoch. the positioning accuracy of real-time measurements is under 3 cm horizontally and 5 cm vertically (eren et al. 2009). the master and auxiliary control centres broadcast the coordinate correction using vrs cmr+, vrs rtcm 3.1, sapos fkp 2.3, rtcm3net (mac) and dgps techniques (yildirim et al. 2011). the cors-tr network was established at the end of 2008 with the support of the tubitak (the scientific and technological research council of turkey) project. the istanbul kültür university has taken over the responsibility of making the network fully operational under the supervision of the general command of mapping and general directorate of registration and cadastre. this network was built with the benefit of some tutga (turkish national fundamental gps network) points which serve as a reference to geographic information, established with the cooperation of the general command of mapping and general directorate of land registry and cadaster (kurt and cingoz 2013). performance test of the yldz single base rtk station the performance tests of the yldz station were carried on in two stages. in the first stage, the coordinates of yldz station were determined in the cors-tr network by real-time measurements using the vrs method. as a result of this measurement, the precision of coordinates and standard deviations of the yldz point was calculated in the cors-tr network system datum. the ashtech proflex 800 gnss receiver and aerat1675_120 spke geodetic antenna were used. in total, 6500 epoch by epoch coordinate data with an interval of 10 seconds was collected in approximately 18 hours. the average value of the measured coordinates and standard deviations was calculated. the differences in these coordinates are given in fig. 1, while the differences percentage of all the surveyed epochs are given in table 2. -2 -1 0 1 2 -2 -1 0 1 2 3 -4 -2 0 2 4 6 north (cm)east (cm) up (c m ) figure 1. the deviation of rtk coordinates from average value table 2. differences of the rtk coordinates the average value of yldz coordinates of 6500 epochs is determined by the cors-tr network, while the differences between north east up coordinates are calculated from static and real-time observations. the post processed coordinates of yldz station were entered into the ashtech web server software to be used in real-time measurements in order to determine the coordinates of benchmark points in the next stage. on the other hand, the yldz point was determined by cors-tr network in real time to calculate the coordinate differences between these two systems for precision of the yldz single base rtk solution versus the cors-tr solution. the differences between the north, east and up coordinates are close to 2 cm. in the second stage, test surveys were carried out in 5 pillar-constructed benchmarks with a 5 repeatability up to 50 kilometres from yldz single base station, in distances of 10, 15, 25, 40 and 50 kilometres respectively. real-time surveys were carried out with the spectra precision sp80 gnss receiver, which has an 8 mm + 1 ppm horizontal, 15 mm + 1 ppm vertical accuracy. at each benchmark point, the coordinates were determined by the yldz single base rtk method, the cors-tr network and the vrs method. an average of 5 epochs of north, east, up coordinates, initiation time, measuring time, pdop, observed satellite numbers, horizontal rms and vertical rms values of coordinates were recorded at all 5 benchmark points with 5 repeatability in order to determine the precise coordinates and standard deviation between the sessions. benchmark points are istanbul gps triangulation network (igna) points which are in the itrf96 datum 2005.0 epoch. the igna network was constituted in 2005 by processing gnss data through the fixing of 11 tutga points close to istanbul city. rtk measurements were carried out in the itrf96 and 2005.0 beginning epoch at the benchmark points with 5 repeatability in order that the standard deviation values between sessions be calculated. the coordinate differences between known and measured coordinates are examined in fig. 2 for two rtk solutions. figure 2 (a). coordinate differences (cors-tr network vrs solution) 138 nedi̇m onur aykut, engi̇n gülal and burak akpinar figure 2 (b). coordinate differences (yldz single base rtk solution) in figure 2 (a), the deviation between known and measured coordinates by the cors-tr network was under 4 cm both horizontally and vertically. the cors-tr solution is not due to the distance from yldz single base station but the cors-tr network points are far away from the measured points. the distance from reference points to measured benchmarks is about 50 km. in particular, the last of two test results have high coordinate differences from the known values, perhaps due to the distances from the reference points of cors-tr network. in fig.2 (b), the results of the yldz single base solution are given. the deviation of the coordinates is under 2-3 cm horizontally and vertically up to 50 km from the yldz station. during the test survey, the repeatability of the rtk solutions was also tested. five sets of rtk measurements were attempted. at each of the benchmark points, the gnss receiver was initiated to the cors-tr and yldz casters one by one. when the gnss receiver was initiated to cors-tr, measurements were taken by calculating the mean value of 5 surveyed epochs. this constituted a first set of rtk measurements. the gnss receiver was then initiated to the yldz caster and the first set of the measurements was completed. this operation was continued until 5 sets of rtk measurements were completed. we now had 5 by 5 north, east and up coordinates of each benchmark points solved by the cors-tr and yldz solutions. the standard deviation between these measurements was calculated to describe the measurement differences in respect to their average value. if the deviation of the sets were at a high value, it showed us that the measurements were far from the average value. if the standard deviation value was then small, the measurements were close to the mean value and so the measurements were coherent in relation to each other. figure 3 (a). standard deviation of 5 rtk measurement sets (cors-tr network solution) figure 3 (b). standard deviation of 5 rtk measurement sets (yldz single base rtk solution) looking at figure 3, it can be said that the standard deviation of the coordinates is small and close to zero value. hence, the measurement quality was good. only in the third benchmark point were the values higher in contrast to others, but still under 4 cm for both cors-tr and yldz solution. the reason might be poor satellite geometry. figure 4 shows the satellite numbers which were observed by both the cors-tr and yldz solutions during benchmark tests with 5 repeatability. figure 4. satellite numbers at the benchmark points initialisation time is another important criterion in network rtk solutions (eren 2009). fig. 5 shows the average of the initialisation times of 5 rtk sets for all the benchmarks. the yldz rtk solution initialisation time was sometimes close to the cors-tr solution. figure 5. initialisation time of rtk solutions 139performance of single base rtk gnss method versus network rtk conclusion network rtk solutions are useful across the world for real time positioning. in turkey, the cors-tr rtk network has operated since 2008 for real time applications in itrf96 datum. as a result of the measurements carried with the cors-tr network at the benchmark points, the differences between the known and measured coordinates are under 3-4 cm horizontally and vertically (fig. 2a). the results of the measurements with 5 repeatability are close to each other at the same benchmark points. the standard deviations of these are near/under 1 cm horizontally and 3-4 cm vertically (fig. 3a). the coordinate differences of 4 and 5 points are higher than the others. at the 4th point, the observed satellites are 13-14 numbers. the measurements made at the 4th point are the same because the standard deviation of these measurements is 0.5 cm horizontally and almost 2 cm vertically (fig.3a). at the 5th point, the observed satellite numbers are 10 (fig. 4). besides, the initialisation time of this solution is approximately 40 seconds; therefore higher than the other values. satellite geometry may be a reason for dropping positioning accuracy. in addition to these, at the 3rd point, the differences of north, east and up coordinates are under 2 cm but the standard deviation of the up coordinates is 4 cm (fig. 3a). the observed satellites are 12. cors-tr network positioning accuracy, which was tested at 5 benchmark points, was calculated as 3-4 cm horizontally and vertically (as given in the literature). this value depends on satellite geometry and therefore on the initialisation time and the distances to reference points. when examining the yldz single base rtk solution, the coordinate differences of known and measured are both near and under 1-2 cm horizontally and near 3 cm vertically (fig. 2b). the standard deviations of 5 repeatability of the measurements are an average of 1 cm horizontally and 3 cm vertically. the higher value is 4 cm with the 3rd test point at the up coordinate, just as with the cors-tr network solution. however, at the benchmark points 4 and 5, the coordinate differences calculated by yldz solution are under 1 cm and 2 cm respectively (fig. 2b). the observed satellites in these points 14 and 15 (fig. 4). the average initialisation time of yldz single base solution is under 20 seconds. the differences of known and measured coordinates change from 0.5 cm to 2 cm horizontally and from 0.5 to 3 cm vertically. the distance between base and rover receiver does not change perceptibly for the coordinate differences up to 50 km. however, at the 5th point, the differences of north coordinates are about 2 cm and higher than the other points (fig. 2b). judging from this result, it may be said that the distance between the receivers is important for coordinate precision. in addition to this claim, the initialisation time (fig. 5) is changed with the distance between the base and the rover receiver. the coordinate differences determined by the two rtk solutions were under 1.5 cm with 99% of measured epochs (table 2). going by these measurements, it can be said that the yldz single base rtk solution is a choice for users up to 50 km away from base station. the distance among base and rover receiver may therefore be efficient for coordinate accuracy. in addition to this conclusion, the satellite geometry has been shown to be important for rtk measurements when determining precise coordinates. being that the initialisation time is dependent on the satellite geometry, this is important for the accuracy of positioning. acknowledgments this research has been supported by the yildiz technical university scientific research projects coordination department (project number: 2012-05-03-kap01). the authors wish to thank geomatics ankara-turkey for gnss equipment and technical support. references ahmed el-mowafy 2012. global navigation satellite systems – signal, theory 162 and applications, part 7, precise real-time positioning using network rtk, pages 161-164. eren, k., uzel, t., gülal, e., yıldırım, ö, cingöz, a. 2009. results from a comprehensive global navigation satellite system test in the cors-tr network: case study, journal of surveying engineering © asce / february 2009, pages 10-18. feng, y. and wang, j. 2008. gps rtk performance characteristics and analysis, journal of global positioning systems (2008) vol. 7, no. 1, pages 1-8. gülal, e., erdoğan, h., tiryakioğlu, i̇. 2013. research on the stability analysis of gnss reference stations network by time series analysis, digital signal processing 23 (2013) pages 1945–1957. hofmann-wellenhof, b., hoffmann, et al,, h., wasle, e. 2008. gnss – global navigation satellite systems, gps, glonass, galileo, and more, springerwiennewyork, isbn 978-3-211-73012-6. hu, g.r., khoo, h.s., goh, p.c., law, c.l. 2003. development and assessment of gps virtual reference stations for rtk positioning, journal of geodesy (2003) 77: pages 292–302. kurt, a.i̇, cingöz, a. 2013. birlikte çalışabilirlik kapsamında türkiye ulusal sabit gnss i̇stasyonları ağı (tusaga-akti̇f), tmmob coğrafi̇ bi̇lgi̇ si̇stemleri̇ kongresi̇, ankara. landau, h., chen, x., kipka, a., vollarth, u. 2007. network rtk, getting ready for gnss modernization, gps world, april 2007, pages 50-55. langley 1998. rtk gps, gps world; sep 1998; 9, 9; proquest science journals, pages 70-76. rizos c. and han, s..2003. reference station network based rtk systemsconcepts and progress, wuhan university journal of natural sciences vol. 8 no. 2b, 2003, pages 566-574 rizos, h. 2003. network rtk research and implementation – a geodetic perspective. spectra precision sp-80 gnss receiver manual. volker wegener and lambert wanninger 2005. communication options for network rtk / sapos® realization, proceedings of the 2nd workshop on positioning, navigation and communication (wpnc’05) & 1st ultra-wideband expert talk (uet’05), pages 115-118. yildirim, ö., salgin ö., bakici, s. 2011. the turkish cors network (tusaga-aktif), fig working week 2011, bridging the gap between cultures, marrakech, morocco. xu, g. 2007. gps, theory, algorithms and applications, 2nd edition, springer, isbn:978-3-540-72714-9. this paper aims at the problem that the migration quality is poor in deep water. it analyzed the influence that velocity model accuracy had on migration and studied the impact that variable seawater velocity makes on movement effect. at first, variable seawater velocity affected by temperature, pressure, and salinity is defined to replace the actual seawater velocity. then variable seawater velocity’s influence on interface migration location, layer sickness and movement energy focusing degree are analyzed in theory. moreover, finally, a deep water layered medium model containing variable seawater velocity, a syncline wedge shape model, and a complex seafloor velocity model are constructed. by changing the seawater velocity of each model and comparing migration results of constant seawater velocity model and variable seawater-velocity model, it was found the conclusion. under the condition of deep water, variable seawater velocity’s impact on the quality of seismic migration is significant, which not only can change the location of geologic body migration result but also can influence the resolution of the geologic interface in the movement section and maybe can cause migration illusion. este artículo se enfoca en el problema de la baja calidad de la migración en aguas profundas. se analiza la influencia que tiene el modelo de precisión de velocidad en la migración y se estudia el impacto que la variación de velocidad del agua marina tiene en el efecto de movimiento. en primera instancia, se define la variación de la velocidad del agua marina afectada por la temperatura, la presión y la salinidad para reemplazar la velocidad del agua marina actual. luego se analiza la teoría de la influencia de la velocidad del agua marina sobre la interfaz de la ubicación de migración, el grosor de la capa y la energía de movimiento de acuerdo con la inclinación. además, finalmente, se construyó un modelo medio por capas de aguas profundas que contiene las variaciones de velocidad del agua marina, un modelo de forma de cuña sinclinal, y un modelo de complejo de velocidades del lecho marino. al cambiar la velocidad del agua marina en cada modelo y comparar los resultados de migración del modelo de velocidad de agua marina constante con los del modelo modelo de velocidad variable se encontró la conclusión. bajo las condiciones de agua profunda, el impacto de la variación de la velocidad del agua es significativo en la migración sísmica, lo que no solo puede cambiar el resultado de migración en la ubicación del cuerpo geológico sino que además puede influir en la resolución de la interconexión geológica en la sección de movimiento y podría causar la ilusión de migración. earth sciences research journal earth sci. res. j. vol. 20, no. 3 ( september, 2016 ) : h1-h7 abstract resumen keywords:seawater velocity, deep-water geology, migration imaging, migration velocity palabras clave: velocidad del agua marina, geología de aguas profundas, imágenes de migración, velocidad de migración. research into the effects of seawater velocity variation on migration imaging in deep-water geology record manuscript received: 22/03/2016 accepted for publication: 23/11/2016 how to cite item sun, h., sun, j., han, f., sun, z., liu, z., liu, m., & huang, x. (2016). research into the effects of seawater velocity variation on migration imaging in deep-water geology. earth sciences research journal, 20(3), j1-j6. doi:http://dx.doi.org/10.15446/esrj.v20n3.56382 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n3.56382 hui sun1, jianguo sun2, fuxing han1, zhangqing sun1, zhiqiang liu1, mingchen liu1, xingguo huang1 1. college for geoexploration science technology, jilin university, changchun, 130026 correspondence should be addressed to jianguo sun, e-mail: sun_jg@jlu.edu.cn s e is m ic investigación de los efectos de la variación en la velocidad del agua marina sobre las imágenes de migración en la geología de aguas profundas hui sun, jianguo sun, fuxing han, zhangqing sun, zhiqiang liu, mingchen liu, xingguo huangh2 introduction as more and more offshore oil and gas fields have been discovered recently, the exploration difficulty for new ones increases sharply. thus, growing number of human and financial resources has been invested into deep-sea oil and gas exploration worldwide (barely, 1999). the south china sea is rich in oil resource while 70% of that located in the deep sea (li, 2006). during latest ten years the south china sea oil and gas exploration experienced both more and more attention from the government and also a series of problems of which one exists in migration processing on marine seismic data. the problem can be described as the movement energy not high enough in deep water so that the real structure and stratum could not be recognized clearly and exactly. the paper aims at solving it, thus analyzes the affection of variable seawater-velocity on migration in deep water from the perspective of velocity model’s accuracy. physical oceanography study reveals that the main influencing factors on the acoustic velocity of seawater are temperature, pressure, and salinity (feng, 1999; jones, 1999) which are also correlated with depth. therefore, seawater acoustic velocity is a function of depth as well. it is well known that pre-stack depth migration is so sensitive to velocity model that even a tiny error could change the final result (versteeg, 1993; herron, 2000). the seawater velocity is always processed as constant speed during marine seismic data processing. such processing has little affection on seismic data of shallow seawater but a big one on that of deep water which could not be neglected. the relationship between seawater velocity and wave field propagation has been studied and concluded by many researchers as follows. dyk and swainson (1953) drew the conclusion from the perspective of kinematics that the seawater acoustic velocity which varies as depth has a great effect on ray path. barley (1999) pointed out in his paper that seawater velocity variation is one of the important reasons that reduce the efficiency of marine seismic data processing. mackey and fried (2003) revealed further that it not only affects the event continuity of marine seismic data but also influences stack section showing inaccuracy amplitude information and wrong submarine tectonics. the similar question is put forward by lacombe et al. (2006), that seawater velocity variation can cause seismic event dislocation. xu and pham (2003) and jones (2010) studied seismic imaging and achieved the conclusion that variable seawater velocity sets back seabed imaging. papenberg et al. (2010) and song et al. (2010), who used seismic data to achieve the temperature and salinity variation curve form with the depth of seawater, provide the theory basis of variable seawater velocity employed in this paper. the influence of seawater velocity on seismic travel time, ray path, and amplitude is revealed based on empirical formulas of temperature, depth, and salinity (han et al., 2012). the paper firstly studies the practical method of seawater acoustic velocity and draws the variation curve of acoustic velocity with depth. then based on that method, the effect of the speed is researched theoretically on interface deviation, layer thickness, and focusing of migration energy. finally, a layered model, a syncline model, and a complex model are built to contrast movement results between variable seawater-velocity models and constant seawater-velocity models and reveal the effect of variable velocity on migration imaging in deepwater geology structure. 2. methodology 2.1 construction of seawater acoustic velocity model seawater velocity is a function of temperature, salinity, and pressure and always described in empirical formula. the one it was applied in this paper is the wilson empirical formula simplified by frye and pughfrom (feng, 1999). (1) in the above formulas c represents seawater velocity, t expresses time， s means salinity and p describes pressure. variable seawater-velocity model intuitively reflects the position relationship between seawater velocity and depth. thus temperature, pressure and salinity’s relationship with depth are needed. the changed law of pressure with depth is quite simple in linear form as is shown in figure 1a. temperature and salinity variation with depth are rather complicated because of sun radiation, ocean current, water masses and so on. the temperature and salinity are assumed to change evenly with depth in different extreme sections. then approximate variation curve of temperature and salinity with depth is formed in consideration of physical oceanography knowledge and temperature and salinity data (jones, 1999; chen and wen, 2010; zhao, 1985) collected in a region of south china sea in summer as is shown in figure 1b and figure 1c. the variable seawater velocity curve with depth is obtained after the equal pressure, temperature and salinity are calculated in equation 1. figure 1. the changing curve of pressure a, temperature b, salinity c and seawater velocity d in depth 2.2 effect of seawater velocity error on pre-stack depth migration the accuracy of velocity has a direct influence on migration and even stronger impact on pre-stack depth migration. in this section, the effect of seawater velocity on seawater movement is analyzed theoretically. in migration, imaging mv is assumed to be constant seawater movement velocity and ( )v z is the function of actual seawater velocity with depth. in seawater velocity model including horizontally stratified medium, each interface can still be recognized in migration section even when there is a significant difference between mv and true seawater velocity. however, the position of each interface changes as well as the layer thickness. the following part will discuss the influence of seawater velocity on interface position and layer thickness quantitatively. figure 2. the sketch map of layers the figure 2 shows the seawater velocity model including the horizontally stratified medium in which 1h describes seawater layer’s thickness while 1nh + represents the thn horizontal stratified medium’s thickness. zero offset time of geophone through single pass from interface one can be described as follows. research into the effects of seawater velocity variation on migration imaging in deep-water geology h3 (2) seawatert and vaverage respectively represents zero offset time of single pass and average seawater velocity. the whole layer position will climb when migration velocity mv is lower than vaverage . meanwhile, each layer is assumed to be thick enough to acquire the exact position and climbing height of each one as is shown below. (3) in the function, ' nh means the migration depth of the thn interface and nh� demonstrates the difference between migration depth and real depth. as is shown in equation 3 migration position error of the thn interface correlates with seawater migration velocity mv and velocity in the overlying layer nv , which decreases with mv and increases with nv . because of different speed in overlaying layers for a different interface, thus the climbing height of each layer is different in migration section. therefore, movement thickness is different from the true layer thickness. (4) in the above function, , 1n nh � means the thickness of the horizontal layer between thn and 1thn � interface. the average geological structure enjoys bigger velocity in the downside strata. when mv is lower than vaverage , the thickness of horizontal stratified medium decreases. furthermore, the more significant the overlying stratum velocity is, the less the migration thickness of the layer is. when mv is higher than vaverage , migration location of each interface will fall. (5) it can be seen from the functions that migration position error is directly proportional to movement velocity in seawater. different from the condition when movement velocity in seawater is lower than average speed, it is velocity in the strata below the interface that affects migration location error as well as the thickness of each horizontal layers. (6) the equation 6 shows that when migration velocity is higher in seawater, the movement thickness of the horizontal layer will increase. figure 3. the sketch map of simple layers in the figure 3, the horizontal strata below the seawater have the velocity of 2500m/s and the thickness of 1000m while the second horizontal stratified medium enjoys the velocity of 3500m/s. velocity in seawater varies with depth and velocities at different depth can be matched in figure 1d. table 1 shows the location error of interface one and thickness error of layer one in corresponding migration velocity at a different depth. table 1: the location error of interface one (m) and thickness error of layer one (m) remarks: + means the falling of migration location and the increase of migration thickness of the 1th layer. otherwise, − shows the climbing of migration area and the decrease of migration thickness of 1st layer the data given in table 1 are consistent with the analysis conclude before. the effect of variable seawater velocity on interface location and layer thickness has been taken into consideration. however, its influence on migration energy focusing can not be neglected as well. supposing v represents seawater acoustic velocity and mv is the applied migration velocity, the function of migration depth is obtained of common imaging point gathers by zhu (1998). (7) in equation 7, mz acts as migration depth and /mv vb = exists. when migration velocity is smaller, b is less than 1, otherwise b is more than 1. after then /mv vd = � and mz z z= + � are calculated in equation 7, the equation 8 is acquired. (8) in marine data processing, is less than z and velocity variation is much lower than true velocity. the equation 9 can be obtained after second order term of and are omitted. (9) based on functions above, the cigs of an impulse response was drawn in figure 4 to illustrate velocity error’s impact on cigs. figure 4 shows that when seawater migration velocity is quite low, the orbit of pulse imaging points is the one above the true orbit. when the seawater movement speed is higher, the orbit of pulse imaging points is the one below the true orbit. what’s more, depth error of the migration will increase with the growth of horizontal distance to object point. the velocity will affect the focusing of the movement energy and lower the effectiveness of migration imaging of marine geology structure. figure 4. the sketch map of velocity error’s impact on cigs hui sun, jianguo sun, fuxing han, zhangqing sun, zhiqiang liu, mingchen liu, xingguo huangh4 3. examples and results a layered model, a syncline model, and a complex model were built and then it was simulated to acquiring the seismic record. the effect of variable velocity on geology structure migration in deep water is achieved after the contrast between true velocity model and different constant seawater velocity model. seawater velocity in the three models all takes temperature, pressure, and salinity into consideration. the figure 1d describes variable seawater velocity curve with depth. 3.1 influence of variable seawater velocity on migration of horizontally stratified medium the figure 5 demonstrates the horizontally stratified medium model in deep water with the size of a 1501 x 1601, grid spacing of 4m x 4m and seawater depth of 3500m at which the data is collected in a region of south china sea. the horizontally stratified medium can be divided into three layers whose velocity are respectively 3250m/s , 4000m/s , and 4750m/s from up to down. figure 5 layered model in deep water firstly, the migration of true velocity model is obtained and then the seawater velocity is substituted with different constant seawater velocities, 1450m/s, 1470m/s, 1500m/s, 1520m/s, 1570m/s, to get similar constant velocity model and migration section of each one as is shown in figure 6. the contrast of movement results in different models shows that when seawater velocity is 1450m/s the lower the layer is, the higher the horizontal interface climbs. though three interfaces can be demonstrated in migration section, movement energy can not be entirely focused on each interface but both on the interface and the area above it. the more shallow the interface location is, the less focused the energy is in the interface. as the increase of migration velocity of seawater, the upward movement of each interface decreases and migration energy focus strengthens. when seawater velocity is 1500m/s, the movement result which still has some difference with the one achieved at variable velocity is most similar to the original model. when seawater velocity is 1500m/s, the migration result which still has some difference with the one achieved at variable velocity is most similar to the original model. as the increase of migration velocity of seawater, the downward movement of each interface increases and is directly proportional to the depth of interface. movement energy mainly distributes on the interface and in the region below it. figure 6. migration results of layered model in deep water, the corresponding seawater velocity of a, b, c, d, e, f is true seawater velocity 1450m/s, 1470m/s, 1500m/s, 1520m/s, 1570m/s research into the effects of seawater velocity variation on migration imaging in deep-water geology h5 figure 8. migration results of syncline model in deep water, the corresponding seawater velocity of a, b, c, d, e, f is true seawater velocity, 1450m/s, 1470m/s, 1500m/s, 1520m/s, 1570m/s 3.2 influence of variable seawater velocity on migration of syncline model the figure 7 shows the syncline model in deep water with the size of 1501 1601� , grid spacing off 4m 4m� and seawater depth of 3500m. the horizontally stratified medium can be divided into three layers whose velocity are respectively 2500m/s , 3000m/s , and 3500m/s . the velocity of wedge geologic body is 5500m/s . figure 7. syncline model in deep water the figure 8 demonstrates the variable seawater velocity model of syncline and migration corresponding to constant seawater velocity model at the speed of 1450m/s, 1470m/s, 1500m/s, 1520m/s and 1570m/s. the contrast of migration results in different models reveals that when seawater velocity is 1450m/s all movement location are higher than true interface position. the distance between two horizontal layers decreases as well as the angle of wedging geology body. migration interface of syncline is steeper than original interface. movement energy is not well focused on interfaces but also distributed in the region above migration interface. as the increase of seawater velocity, the location of movement section and radian of syncline interface decreases. the distance between two layers increases and the focus of migration energy is improved. when seawater velocity is 1520m/s, the movement result is most similar to the one applying real seawater velocity. when seawater velocity is as high as 1570m/s, radian of syncline interface and distance between two horizontal layers decreases and migration section moves downside. meanwhile, energy is scattered both above and under movement interface. therefore, the shape of movement interface is a quite different from that of the true interface. hui sun, jianguo sun, fuxing han, zhangqing sun, zhiqiang liu, mingchen liu, xingguo huangh6 3.3 effect of variable seawater velocity on complex seabed geology structure model in deep water the figure 9 displays the complex seabed geology structure model in deep water with the size of 1501 1601� and grid spacing of 4m 4m� . velocities of layer in the model are respectively 2500m/s, 2700m/s, 2900m/s, 3000m/s, 3100m/s, 3300m/s and 3500m/s from up to down. the velocity of embed body on the right side is 3300m/s. figure 9. complex model in deep water figure 10 illustrates migration results of complex model in deep water at constant seawater velocity of 1450m/s, 1470m/s, 1500m/s, 1520m/s, 1570m/s. it can be concluded from the contrast of these migration sections that when seawater velocity is 1450m/s all interfaces move upward in the migration section. at different depth, different horizontal position in migration section moves upwards in different distances which increase with deeper depth of the sea. furthermore, movement energy of interfaces cannot be well focused mainly on interface two and interface three whose interfaces can not be recognized clearly in migration section. meanwhile, the up and down interfaces are also unable to be connected to embed layer. when seawater velocity is 1470m/s or 1500m/s, the problem is partially improved in contrast with the condition when the velocity is 1450m/s but still having a severe influence on migration results. then when the seawater velocity is as high as 1520m/s, movement achieves the best result among ones of constant seawater velocity. however, migration energy still can not be well focused. afterward, when the velocity is 1570m/s, the up and down interfaces can not be connected to embed layer. interfaces can be recognized reluctantly but in low resolution. in comparison with a true condition, there is a marked difference in space position and geometry. figure 10. migration results of complex model in deep water, the corresponding seawater velocity of a, b, c, d, e, f is true seawater velocity 1450m/s, 1470m/s, 1500m/s, 1520m/s, 1570m/s. research into the effects of seawater velocity variation on migration imaging in deep-water geology h7 4. conclusions acoustic velocity of seawater is mainly influenced by temperature, pressure, and salinity and is also correlated with depth. the seawater acoustic velocity is always processed as constant velocity during marine seismic data processing. such processing has a severe effect on seismic data of deep seawater which can not be neglected. meanwhile, the accuracy of velocity model can influence accuracy and results of migration. this paper begins with improving the weaker movement field and worse migration results in deep water region of south china sea and then analyzes effects of seawater velocity on migration results. it can be concluded from layer model and syncline including wedge model that variable velocity has an adverse influence on interface location, the shape of geology body and focusing on migration energy. the analysis of complex seabed geology model reveals that variable seawater velocity can make interfaces of geology body hard to be distinguished in movement section. furthermore, variable speed even leads movement section to show wrong geology structure. therefore, seawater velocity should not be only taken as constant in deep-water seismic data processing. 5. acknowledgments this study was supported financially by the national natural science foundation of china (42174120, 41404085, 41504084). we are grateful to danian huang of jilin university, changchun, for his kind computer supporting. references barley, b. (1999). deepwater problems around the word. the leading edge, 18(4), 488-493. doi: 10.1190/1.1438319 chen, j., & wen, n. (2010). the geophysical atlas of the south china sea. science press, beijing. dyk, k., & swainson, o. w. (1953). the velocity and ray paths of sound waves in deep sea water. geophysics, 18(1), 75-103. doi: 10.1190/1.1437867 feng, s. z. (1999). marine science introduction. higher education press, beijing. han, f. x., sun, j. g., & wang, k. (2012). the influence of seawater velocity variation on seismic traveltimes, raypaths, and amplitude. applied geophysics, 9(3), 319-325. herron, d. a. (2000). pitfalls in seismic interpretation: depth migration artifacts. the leading edge, 19(9), 1016-1017. doi: 10.1190/1.1438756 jones, e. j. w. (1999). marine geophysics. john wiley & sons, england. jones, s. m., sutton, c., hardy, r. j. j. & hardy, d. (2010). seismic imaging of variable water layer sound speed in rockall trough, ne atlantic and implications for seismic surveying in deep water. geological society, london, petroleum geology conference series 7, 549-558. doi: 10.1144/0070549 lacombe, c., schultzen, j., butt, s., & lecerf, d. (2006). correction for water velocity variations and tidal statics. 68th eage conference & exhibition. doi: 10.3997/2214-4609.201402385 li, q. p. (2006). the situation and challenges for deepwater oil and gas exploration and exploitation in china. china offshore oil and gas, 18(2), 130-133. mackay, s., fried, j., & carvill, c. (2003). the impact of water-velocity variations on deepwater seismic data. the leading edge, 22(4), 344-350 papenberg, c., klaeschen, d., & krahmann, g. (2010). ocean temperature and salinity inverted form combined hydrographic and seismic data. geophysical research letters, 37(4). doi: 10.1029/2009gl042115 song, y., song, h. b., chen, l. (2010). seawater thermohaline structure inversion from seismic data. chinese journal of geophysics, 53(11), 2696-2702. doi: 10.1002/cjg2.1569 versteeg, r. j. (1993). sensitivity of prestack depth migration to the velocity model. geophysics, 58(6), 873-882. doi: 10.1190/1.1443471 xu, s., & pham, d. (2003). global solution to water column statics: a new approach to an old problem. 73rd ann. internat. mtg.: soc. expl. geophys. zhao, x. y. (1985). the comprehensive research report of the south china sea. science press, beijing. zhu, j., lines, l., & gray, s. (1998). smiles and frowns in migration/ velocity analysis. geophysics, 63(4), 1200-1209. keywords: loess; triaxial test; seismic subsidence changes; depth of soil layer; subsidence extent. palabras clave: suelos poco compactos; reistencia triaxial; cambios de subsidencia sísmica; profundidad de capas de suelo; subsidencia extendida. how to cite item yan, x., wang, s., & wang, n. (2020). analysis and evaluation methods of seismic subsidence characteristics of loess and field seismic subsidence. earth sciences research journal, 24(4), 485-490. doi: https://doi.org/10.15446/ esrj.v24n4.91593 the objective of this research is to analyze the dynamic degeneration of loess and the evaluation method of field seismic subsidence. in this study, q3 loess is taken as the research object, and the dynamic properties of loess with 10%, 20%, 30% and 35% moisture content are tested by triaxial experiment. in addition, seismic subsidence characteristics of loess with dry densities of 1.4g/cm3, 1.6g/cm3, and 1.8g/cm3 and consolidation stress ratios of 1.0, 1.2, 1.4, and 1.6 are analyzed. then the simplified seismic subsidence estimation method is used to calculate the relationship between seismic subsidence coefficients at different soil depth in one dimensional field, cycle times, and subsidence depth. the results show that the higher the water content of loess is, the greater the change of seismic subsidence appears. the larger the dry density of loess is, the smaller the change degree of seismic subsidence appears. the larger the consolidation stress ratio is, the greater the change of seismic subsidence occurs in loess. when the depth of soil reaches 9.5m, the maximum seismic subsidence coefficient can reach 0.8%. when the depth of soil layer is 10m, the degree of seismic subsidence is the largest. when the depth of soil layer is 12~16m, the settlement depth caused by earthquake subsidence is small. while the depth of soil layer is 8~12m, the settlement degree is large. abstract analysis and evaluation methods of seismic subsidence characteristics of loess and field seismic subsidence análisis y métodos de evaluación de las características de subsidencia sísmica en suelos poco compactos y subsidencia sísmica de campos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.91593 el objetivo de esta investigación es analizar la dinámica de degeneración de suelos poco compactos y el método de evaluación de subsidencia del campo sísmico. en este estudio, suelos poco compactos tipo q3 fueron objeto de investigación, y las propiedades dinámicas de estos suelos con 10, 20, 30 y 35 por ciento de contenido de húmedad fueron evaluadas en experimentos triaxiales. además, se analizaron las características de subsidencia sísmica de los suelos poco compactos con densidades secas de 1.4g/cm3, 1.6g/cm3, y 1.8g/cm3 y un índice de consolidación de esfuerzo de 1.0, 1.2, 1.4 y 1.6. luego, el método de estimación de la subsidencia sísmica simplificada se utilizó para calcular la relación entre los coeficientes de subsidencia sísmica en diferentes profundidades en un campo dimensional, los tiempos del ciclo y la profundidad de subsidencia. los resultados muestran que entre más alto sea el contenido de agua en los suelos poco compactos es mayor el cambio de subsidencia sismica. y que entre mayor sea la densidad seca de estos suelos, más pequeño es el grado de cambio de la subsidencia sísmica. y, finalmente, cuando el índice de consolidación de esfuerzo es mayor también será mayor el cambio de subsidencia sísmica que ocurre en los suelos poco compactos. cuando la profundidad del suelo alcanza los 9.5 metros, el coeficiente de subsidencia sísimica puede alcanzar el 0.8 por ciento. cuando la profundidad de la capa del suelo es de 10 metros, el grado de subsidencia sísmica es mayor. cuando la profundidad de la capa del suelo es entre 12 y 16 metros, el asentamiento causado por la subsidencia del movimiento de tierra es menor. mientras la profundidad de la capa del suelo se encuentre entre 8 y 12 metros, el grado de asentamiento es mayor. resumen record manuscript received: 12/03/2019 accepted for publication: 31/01/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 485-490 xiaotong yan1, shun wang2*, ningwei wang1 1institute of engineering mechanics. cea, heilongjiang, china 2shenyang jianzhu university, liaoning, china *corresponding author: xiatong.yan@126.com se is m o l o g y https://doi.org/10.15446/esrj.v24n4.91593 https://doi.org/10.15446/esrj.v24n4.91593 1.4g/cm 1.6g/cm 1.8g/cm https://doi.org/10.15446/esrj.v24n4.91593 1.4g/cm 1.6g/cm 1.8g/cm 486 xiaotong yan, shun wang, ningwei wang introduction loess is a kind of soil with weak cementation, large pore size, and easy denaturation when encountering water (qiu et al., 2018). however, under the action of water and dynamic load, serious geological disasters such as collapsibility and seismic subsidence would occur (hao et al., 2018). the judgement of seismic subsidence of loess is an important means of seismic safety evaluation. the subsidence of soil under the action of earthquake is seismic subsidence. due to its unique dynamic characteristics, loess is usually damaged when strong earthquakes occur (cheng et al., 2018). at present, some experts and scholars have comprehensively analyzed the dynamic characteristics of loess by the triaxial test system. the specific physical properties of loess, such as dry density and water content, have a great impact on the dynamic characteristics of loess and play an important role in the study of the characteristics of loess (liu et al, 2017). when earthquake subsidence occurs, the soil would subside under the action of dynamic load, and some soil with high water content would collapse due to vibration (chen et al., 2017). dry density is also an important internal factor that reflects the seismic subsidence degeneration of loess. the seismic subsidence performance can be calculated by obtaining the volume strain variable through the dynamic single shear experiment. or, on the basis of the volume strain variable data obtained from the dynamic single shear experiment, the incremental calculation equation is established, and the calculation equation of seismic subsidence is established according to the analysis results of field earthquakes (wang et al., 2017; drzewiecki & piernikarczyk, 2017). it is mainly used to analyze the equivalent linearized site and determine the shear strain time history of different soil layers, and then convert the random vibration waves into harmonics to obtain the equivalent shear strain amplitude and vibration times. under the same vibration condition, the longitudinal strain value of vibration is obtained through the single shear test of the soil samples in different soil layers, and finally the settlement amount of seismic subsidence under different soil layers is analyzed and calculated (gao et al., 2017). the simplified seismic subsidence estimation method can obtain the equivalent shear strain of the soil layer through shear model, maximum ground acceleration, overburden load of the upper soil, and stress attenuation coefficient. according to the relationship between the circular cycle and the volume strain, the volume strain value of equivalent shear strain is obtained. after the loading cycle, the volume strain value of the magnitude is calculated (araujo & castro, 2017; sarhosis et al., 2018). in this study, by analyzing the seismic subsidence characteristics of loess and the influencing factors of seismic subsidence, the seismic subsidence changes of loess with different moisture content, dry density, and consolidation stress ratio are analyzed. the seismic subsidence characteristics of onedimensional loess field and the variation of seismic subsidence in different soil layers are analyzed by the simplified seismic subsidence estimation method. this study aims to provide theoretical basis for the subsequent analysis of loess seismic subsidence characteristics and field earthquakes. methodology characteristics and influencing factors of loess seismic subsidence the main mechanism, disaster development mode, and disaster types of loess seismic subsidence are influenced by the material index of soil, topographic characteristics of loess field, and earthquake. when loess earthquake subsidence occurs, soil quality can significantly affect the intensity of earthquake and the severity of earthquake subsidence. the degree of loess earthquake subsidence is obviously different in different areas. different forces act on the primary structure of loess during the earthquake. when the external force exceeds the connection strength between the soil particles, the pores in the primary structure of loess would be destroyed and the soil particles would be rearranged. at this time, the surface would show sudden settlement. seismic subsidence in loess is the result of the interaction of various factors, and the difference in soil structure has a significant impact on the occurrence of seismic subsidence, such as the water content and pore ratio of soil particles. loess is composed of grain size, and the content of clay between soil grains has a great influence on the occurrence of seismic subsidence. the loess with different water content would also exhibit different kinetic characteristics, and the seismic subsidence of loess would increase with the increase of water content of soil. when the pore ratio is greater than 0.75, the seismic subsidence would increase with the increase of the pore ratio of soil particles. seismic subsidence test of loess under different conditions generally speaking, residual strain is used for the representation of loess earthquake subsidence, and the electro-hydraulic servo dynamic triaxial testing machine controlled by microcomputer (xi’an lichuang, china) is adopted for the test. during the test, the ratio of fixed junction stress of loess samples is not equal to 1, and the center of dynamic strain amplitude would move in the direction of compression with the cyclic action of dynamic stress. accumulated residual strain that can’t be completely recovered would occur in the loess after the cycle stops. the residual strain can be expressed as equation 1: µr n h h h ( ) = − (1) among them, εr(n) is the residual strain, that is, the seismic subsidence coefficient, h is the height of the loess sample before the dynamic load, and h is the height of the loess sample after dynamic loading. figure 1a shows the relationship between the time course of loess dynamic strain and the number of vibration with the passage of cycle number n. it can be observed from figure 1b that after n cycles stop, the accumulated residual strain value of loess is the corresponding strain value when the dynamic stress under this cycle number is equal to 0. the q3 loess, which is 3.5m in the third terrace of the xxx river, is collected from xxx county of xxx city. the loess structure is loose. after testing, it was found that the selected loess has a natural water content of about 11.98%, a dry density of about 1.42g/cm3, and a plasticity index of 7.55. the dynamic properties of loess with 10%, 20%, 30%, and 35% moisture content are tested, and seismic subsidence characteristics of loess with dry densities of 1.4g/cm3, 1.6g/cm3, and 1.8g/cm3 and consolidation stress ratios of 1.0, 1.2, 1.4, and 1.6 are analyzed evaluation of seismic subsidence of one-dimensional loess the one-dimensional loess site is shown in figure 2. when an earthquake occurs in a one-dimensional field, the loess layer would extend horizontally in all directions in the same direction and be propagated by vertical and upward shear waves. at this time, the loess volume would only generate horizontal vibrations conducted by bedrock. when seismic subsidence occurs and the loess is of different densities and of the same type, it can be considered that the seismic subsidence deformation of the same type of loess follows the ratio of the same seismic subsidence table 1. the seismic subsidence model of loess and its condition destruction mode vibrating type liquefied type seismic landslide soil large pores and weak cementation structure; less clay content; low water content high content of powder and sand; high water content large pores and weak cementation structure; less clay content; low water content terrain plateau, beam or loess hills loess basin tableland edge or loess basin strength level 8 and above level 7 and above level 8 and above 1.42g/cm 1.4g/cm 1.6g/cm 1.8g/cm 487analysis and evaluation methods of seismic subsidence characteristics of loess and field seismic subsidence the calculation equation of seismic load cycle is as follows: n n a aeq a i i m i =     = ∑2 37 1 2 . max (4) among them, ai is the acceleration amplitude of level i, nai is the cycle number of ai, amax is the maximum acceleration value. with the change of depth h, the relationship between the stress attenuation coefficient rd of the loess field is shown in figure 3. therefore, the stress attenuation coefficient of loess can be expressed by the equation 5. r h h h h h h d ( ) = ≤ < − ≤ < − ≤ < 1 0 3 34 43 31 43 3 14 82 82 105 88 14 32 0 48 , . . , . . , . ,332 ≤          h (5) here, the equivalent shear strain of the soil layer at any depth is as follows. (6) among them, δ0 is the pressure covering the soil layer, f is the conversion coefficient of frequency addition, and g is the acceleration of gravity. coefficient increment to the dynamic strain amplitude. when the moisture content, dry density, and consolidation pressure of the loess in the same soil field are the same, the seismic depression coefficient of the loess with different soil depth can be calculated by the equation 2. (2) among them, △εp is the increment of the seismic subsidence coefficient, γd is equivalent shear strain amplitude of loess body during earthquake, εp is the seismic subsidence coefficient of the cycle that has occurred, ρ0 is the dry density of loess, ρref is the referenced dry density, and a and b are the reference values of earthiness, as shown in equation 3. (3) 0.0 0.8 1.6 0 1 2 2ε n,d ε σσ σ n n i n ε ε n,r σ σ d ε ε n,r 2ε n,d o a b figure 1. definition of accumulated residual strain of loess (a is the process of dynamic stress and accumulated residual strain changing with time; b is the dynamic stress lag loop of loess) earth's surface z=0 1 n-1 bedrock z=h 1 n-1 h1 hn-1 d1 d2 d3 d4 d5 dn-2 dn-1 dn z1 … … i    figure 2. hierarchical and discrete structure of one-dimensional loess field 488 xiaotong yan, shun wang, ningwei wang result and discussion the relationship of pore ratio of soil grains, dry density and moisture content to seismic subsidence coefficient the density of loess can determine the maximum deformable quantity of earthquake subsidence, and it is reflected by dry density and void ratio. it can be observed from figure 4a that the earthquake subsidence coefficient would increase with the increase of the pore ratio of soil particles, and there is a linear growth relationship between the two. as can be observed from figure 4b, the dispersion between the seismic subsidence coefficient of loess and the dry density of soil particles (g/cm3) is relatively large, but when the dry density of soil particles is large, a large seismic subsidence also occurs. it can be observed from figure 5 that when the pore ratio of soil particles is relatively low, the loess with high water content may not be seismic subsidence. when the void ratio is relatively high and the water content of loess is low, seismic subsidence may also occur. this indicates that the influence of water content limit should be considered for loess with low porosity. however, the boundary of water content does not need to be considered in loess with higher porosity. seismic subsidence variation rule of loess with different moisture content when the consolidation stress and ratio are equal, the influence of different water contents of loess on seismic subsidence is shown in figure 6. when the consolidation stress ratio is 1.0, the consolidation stress is 200kpa and the dry density of the loess is 1.4g/cm3 and the pressure is applied on the loess samples, the loess with different moisture contents all have seismic subsidence changes to some extent. it can be observed from figure 4a that when the moisture content of the loess is 10%, different dynamic stresses have little influence on the residual strain of the loess. however, when the moisture content of the loess is greater than or equal to 20% and the dynamic stress exceeds 100kpa, the residual strain of the loess increases significantly, and the soil also shows obvious seismic subsidence changes. it can be observed that the seismic subsidence of loess is mainly caused by the destruction of the pore structure of soil particles under the action of dynamic load. external pressure would reduce the pore ratio of loess soil particles, and the soil mass would become more compact and deformed (anbazhagan et al., 2017). however, when the water content of loess increases, the cementation between the soil particles would be strengthened, thus reducing the friction strength between the soil particles. therefore, under the action of the same load and consolidation stress, the pore ratio of the loess with high water cut is smaller, the seismic subsidence is more likely to occur, and the deformation of seismic subsidence is more serious (xu & yang, 2017). seismic subsidence variation rule of loess with different dry density when the consolidation stress and ratio are equal, the influence of different dry densities of loess on seismic subsidence is shown in figure 7. it can be observed from figure 7a that when the dry density of loess is 1.4g/cm3, its influence on dynamic stress is lower than that of dry densities of 1.6g/cm3 and 1.8g/cm3, indicating that under the action of the same dynamic stress, the dry density of loess is inversely proportional to the deformation generated by earthquake subsidence. this may be because the higher the dry density of the loess is, the higher the compactness between the soil particles is, and the higher the dynamic stress is needed to make the loess volume change. as can be observed from figure 7b, the residual strain generated by the loess volume decreases with the increase of the dry density of the soil, and when the dry density of the loess is larger than 1.6g/cm3, the influence of the residual strain on the volume of the loess also decreases gradually. 0 10 20 30 40 50 60 70 0.2 0.4 0.6 0.8 1.0 zk1 zk2 zk3 zk4 zk5 zk6 zk7 zk8 zk9 zk10 zk11 zk12 zk13 zk14 zk15 zk16 r d h/m figure 3. relationship between stress attenuation coefficient and soil depth of loess field 0.8 0.9 1.0 1.1 1.2 1.3 0.00 0.02 0.04 0.06 0.08 0.10 se is m ic c oe ff ic ie nt pore ratio 1.2 1.4 1.6 1.8 0.00 0.02 0.04 0.06 0.08 0.10 se is m ic c oe ff ic ie nt dry density a b figure 4. relationship between seismic subsidence coefficient and void ratio and dry density of loess (a is the void ratio; b is the dry density) 0.7 0.8 0.9 1.0 1.1 1.2 1.3 0 5 10 15 20 25 30 seismic sink none-seismic sink w at er c on te nt (% ) pore ratio figure 5. relationship between water content, seismic subsidence coefficient, and pore ratio of loess 1.4g/cm 1.4g/cm 1.6g/cm 1.8g/cm 1.6g/cm 489analysis and evaluation methods of seismic subsidence characteristics of loess and field seismic subsidence the variation rule of seismic subsidence of loess under different consolidation stress ratio figure 8a shows that when the consolidation stress is 100kpa, the dry density of the loess is 1.4g/cm3 and the moisture content is 20%, the consolidation stress ratio is proportional to the residual stress generated by the volume of the loess after the pressure is applied to the sample. it is possible that due to the increase of deviator stress on the loess, the pore structure among the soil particles is gradually destroyed, which leads to the increase of residual strain of the loess volume. it can be observed from figure 8b that the residual stress generated by loess would increase with the increase of consolidation stress ratio, showing a linear growth relationship, which is consistent with the research results of song et al., 2017. analysis of seismic subsidence characteristics of one-dimensional field the variation of seismic subsidence coefficient and cycle times of onedimensional field at different soil depth is shown in figure 9. it can be observed that the seismic subsidence coefficient produced by loess soil is proportional to a b 50 100 150 200 250 0 1 2 3 r es id ua l s tr ai n /% dynamic stress 10% 20% 30% 35% 1 2 3 4 0.0 0.5 1.0 1.5 2.0 2.5 35%30%20% r es id ua l s tr ai n /% water content 105 kpa 140 kpa 175 kpa 210 kpa 10% figure 6. relationship between residual strain and dynamic stress of loess with different moisture content (a is the relationship between residual strain and dynamic stress; b is the relationship between water content and residual strain) a b 50 100 150 200 250 0 1 2 3 r es id ua l s tr ai n /% dynamic stress 1.4g/cm3 3 1.6g/cm3 1.8g/cm3 0.0 0.5 1.0 1.5 2.0 2.5 1.81.6 r es id ua l s tr ai n /% dry density g/cm 105 kpa 140 kpa 175 kpa 210 kpa 1.4 figure 7. relationship between residual strain and dynamic stress of loess with different dry density (a shows the relationship between residual strain and dynamic stress; b shows the relationship between dry density and residual strain) a b 1.0 1.2 1.4 1.6 0 1 2 3 4 r es id ua l s tr ai n /% consolidation stress ratio 105 kpa 140 kpa 175 kpa 210 kpa 50 100 150 200 0 1 2 3 4 r es id ua l s tr ai n /% dynamic stress 1.0 1.2 1.4 1.6 figure 8. relationship between residual strain and dynamic stress of loess with different consolidation stress ratios (a shows the relationship between residual strain and dynamic stress; b shows the relationship between consolidation stress ratio and residual strain) 1.4g/cm 490 xiaotong yan, shun wang, ningwei wang the cycle times, but the seismic subsidence amplitude is inversely proportional to the development of seismic subsidence. it can be observed from figure 9 that when the depth of the soil layer reaches 9.5 m, the earthquake subsidence coefficient is the largest, the generated earthquake subsidence is the strongest, and the maximum earthquake subsidence coefficient of this deep soil layer can reach 0.8%. the results of the relationship between seismic subsidence coefficients at different soil depths and seismic subsidence depths are shown in figure 10. it can be observed from figure 10a that the seismic subsidence coefficient first increases and then decreases with the increase of soil depth. and it can be observed from 10b that when the depth of soil layer is 12~16m, the settlement depth caused by earthquake subsidence is small, while the settlement degree in the soil depth of 8~12m is larger. conclusion based on the previous research results, the seismic subsidence characteristics of the original loess with different moisture content, density, and consolidation stress are analyzed by triaxial experiment. it is found that the seismic subsidence is stronger when the loess has higher water content, lower density, and higher consolidation stress. after the calculation of seismic subsidence coefficient of one-dimensional loess field, it is found that when the depth of soil layer is 10m, the seismic subsidence coefficient is the largest and the deformation of seismic subsidence is the largest. in this study, the mechanical properties of loess are not considered comprehensively, and the calculation amount of models used for seismic subsidence analysis is too small. however, the results of this study can provide certain theoretical basis for the subsequent study on seismic subsidence and evaluation of loess. references anbazhagan, p., uday, a., moustafa, s. s. r., & alarifi, n. (2017). soil void ratio correlation with shear wave velocities and spt n values for indo-gangetic basin. journal of the geological society of india, 89(4), 398-406. araújo, m., & castro, j. m. (2017). simplified procedure for the estimation of local inelastic deformation demands for seismic performance assessment of buildings. earthquake engineering & structural dynamics, 46(3), 491-514. doi: https://doi.org/10.1002/eqe.2825 chen, t., ma, w., & wang, j. (2017). numerical analysis of ground motion effects in the loess regions of western china. shock & vibration, 3, 1-9. doi: https://doi.org/10.1155/2017/1484015 cheng, x. s., ma, l., yu, d. p., fan, j., & li, d. (2018). seismic stability of loess tunnels under the effects of rain seepage and a train load. science china technological sciences, 61(5), 735-747. drzewiecki, j., & piernikarczyk, a. (2017). the forecast of mining-induced seismicity and the consequent risk of damage to the excavation in the area of seismic event. journal of sustainable mining, 16(1), 1-7. gao, g., nie, c., & shi, c. (2017). seismic subsidence of sand ground subject to multidirectional earthquake load. journal of harbin engineering university, 38(7), 1100-1106. hao, l., meng, j., zhang, y., & yang, l. (2018). pliocene seismic stratigraphy and deep-water sedimentation in the qiongdongnan basin, south china sea: source-to-sink systems and hydrocarbon accumulation significance. geological journal, 54(1), 392-408. doi: https://doi. org/10.1002/gj.3188 liu, p., lian, p., zhang, m., he, b., & dou, l. (2017). dynamic characteristics of a building with seismic joints based on ambient vibration. journal of hunan university, 44(1), 95-101. doi: 10.16339/j.cnki. hdxbzkb.2017.01.012 qiu, j., wang, x., lai, j., & wang, j. (2018). response characteristics and preventions for seismic subsidence of loess in northwest china. natural hazards, 92(7), 1-27. sarhosis, v., milani, g., formisano, a., & fabbrocino, f. (2018). evaluation of different approaches for the estimation of the seismic vulnerability of masonry towers. bulletin of earthquake engineering, 3–4, 1-35. song, b., tsinaris, a., & anastasiadis, a. (2017). small-strain stiffness and damping of lanzhou loess. soil dynamics & earthquake engineering, 95, 96-105. wang, y., gang, t., & yu, l. (2017). the effect of explosion parameters on seismic source wavelet calculation and its characteristics. journal of applied geophysics, 145. xu, j. s., & yang, x. l. (2017). effects of seismic force and pore water pressure on three dimensional slope stability in nonhomogeneous and anisotropic soil. ksce journal of civil engineering, 2, 1-10. 0 2 4 6 8 10 12 14 16 0.00 0.02 0.04 0.06 0.08 se is m ic s ub si de nc e co ef fi ci en depth/ m a b 0 2 4 6 8 10 12 14 16 0.0 0.2 0.4 0.6 se is m ic s ub si de nc e depth/ m 1 figure 10. relationship between seismic subsidence coefficients at different soil depths and seismic subsidence depths 0 2 4 6 8 10 12 0.00 0.02 0.04 0.06 0.08 0.10 number of cycles se is m ic s ub si de nc e co ef fi ci en 0.5m 3.5m 6.5m 9.5m 12.5m 15.5m figure 9. relationship between seismic subsidence coefficient and cycle times at different soil depths https://doi.org/10.1002/eqe.2825 https://doi.org/10.1155/2017/1484015 https://doi.org/10.1002/gj.3188 https://doi.org/10.1002/gj.3188 10.16339/j.cnki.hdxbzkb 10.16339/j.cnki.hdxbzkb keywords: flac; neural network; landslide; deformable body; jm tunnel. palabras clave: flac; red neuronal; deslizamiento de tierra; cuerpo deformable; túnel jm. how to cite item zhou, h., wang, y., shen, t., & feng, q. (2019). the comprehensive treatment of the ancient landslide deformable body in the giant deep rock bedding based on the flac analysis model. earth sciences research journal, 23(4), 303-308. doi: https://doi.org/10.15446/esrj. v23n4.84008 in previous studies, most of the studies are based on the failure mechanism of the landslide deformable body on tunnel and the safety monitoring and early warning after the completion of the tunnel, while there is less research on tunnel construction. in order to study the comprehensive treatment of the deformable body of the ancient landslide in the giant deep rock bedding better, the treatment of jm tunnel landslide deformable body was selected to carry on the empirical study. after a detailed analysis of the geological characteristics of jm tunnel engineering landslide, the stability analysis of the landslide and the deformation body and the calculation of the thrust value based on the flac model were carried out on the basis of the monitoring data. after the construction of the model, the intensity parameters of the slip mass deformation and the sliding zone were obtained from the back analysis. on the basis of the above, the comprehensive treatment scheme for the rear side of the old landslide deformation body was determined, including earthwork cleanup, grading supporting, and the back pressure by abandoning the cleaned earthwork at the leading edge. the study is of great significance to the design of the construction of undercrossing tunnel of landslide under the complicated geological conditions. abstract the comprehensive treatment of the ancient landslide deformable body in the giant deep rock bedding based on the flac analysis model tratamiento integral del deslizamiento de un cuerpo deformable antiguo en un gigante lecho rocoso profundo con base en el modelo flac issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.84008 estudios anteriores suelen basarse en el mecanismo de falla de un cuerpo deformable por deslizamiento de tierra en un túnel y en el monitoreo de seguridad y alerta temprana después de la finalización del túnel, mientras que hay menos investigación sobre la construcción del túnel. para estudiar mejor el tratamiento integral del cuerpo deformable de un antiguo deslizamiento de tierra en un lecho de roca profunda gigante, se seleccionó el tratamiento del cuerpo deformable de deslizamiento de túnel del túnel jm para llevar a cabo el estudio empírico. después de un análisis detallado de las consideraciones de deslizamiento por las características geológicas del túnel jm, se realizaron el análisis de estabilidad del deslizamiento de tierra y el cuerpo de deformación y el cálculo del valor de empuje basado en el modelo flac y en el monitoreo de información. después de la construcción del modelo, los parámetros de intensidad de la deformación de la masa de deslizamiento y la zona de deslizamiento se obtuvieron del análisis posterior. después de esto, se determinó el esquema de tratamiento integral para la parte posterior del antiguo cuerpo de deformación por deslizamiento de tierra, incluida la limpieza del movimiento de tierras, el soporte de nivelación y la contrapresión al abandonar el movimiento de tierra limpio en el borde delantero. el estudio es de gran importancia para el diseño de la construcción del túnel de deslizamiento de tierra bajo condiciones geológicas complicadas. resumen record manuscript received: 27/04/2019 accepted for publication: 25/09/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (january, 2020): 303-308 hongsheng zhou1,2,*, yunsheng wang1,2, tong shen1,2, qianqian feng1,2 1state key laboratory of geohazard prevention and geoenvironment protection, chengdu university of technology, chengdu, 610000, china; 2college of environment and civil engineering, chengdu university of technology, chengdu, 610000, china * corresponding author: geophysics_chow@sina.cn g eo lo g ic a l en g in ee r in g 304 hongsheng zhou, yunsheng wang, tong shen, qianqian feng introduction the excavation of the tunnel may lead to the deformation of the original mountain, coupled with the original landslide geological conditions, the comprehensive consideration of all kinds of complex factors is necessary in the corresponding reinforcement construction (chen, xiao, sun, & fang, 2014). there are many studies on the investigation and support design of slope landslide in china, while there are few studies on the existing landslide with complicated construction conditions (liu & qian, 2014). the traditional analysis methods of slope stability are mainly qualitative analysis and quantitative analysis of the two, in which, the qualitative analysis includes the geological analysis method, the engineering analogy method and the graphic method, while the quantitative analysis method includes the limit equilibrium method, the numerical analysis method and so on (shuan-cheng, wang, wang, & he, 2015). these methods can be used to analyze and predict the process of slope landslide, but the parameters needed are many, and the calculation is large (ming-zhu, kong, & zheng 2011). to this end, the empirical study on the comprehensive treatment of the jm tunnel landslide deformable body was carries out, the relevant parameters of the landslide deformable body were calculated on the basis of the construction of the flac two-dimensional numerical analysis model and the uniform design test method (sun, wong, shang, shen, & lü, 2010). the study has certain reference significance to the construction of the modern expressway in china and the reinforcement supporting engineering of slope under the complicated geological conditions, which is helpful to promote the application of neural network model in engineering design. description of the problem jm tunnel from xy to gl expressway is divided into two tunnels, on the left and right; the tunnel passes through the landslide two times, the left hole body is 103m, the entrance section is 347m and the right hole section is 204m. the jm landslide belongs to the old landslide of the giant bedding rock, therefore, it has the characteristics of multi-sliding, which is influenced by tunnel excavation; the large sliding deformation of landslide deformable body causes significant surface cracks and tunnel cracks, as shown in figure 1. the main treatment difficulties include: (1) the geological condition is very complex, the jm landslide belongs to the multi-level sliding deformation, the landslide areas are many, and the stress change is complex; (2) the duration of construction is tight, the entire expressway from xy to gl must be opened to traffic in 2016; (3) the project cost is needed to control. therefore, it is necessary to make a comprehensive analysis of the jm expressway landslide deformable body, so as to provide the basis for the selection of the treatment plan. for the supporting scheme of landslide deformation, it is necessary to analyze the sliding surface and the sliding bed characteristics of the landslide and deformable body, and calculate the stability and the thrust value of the landslide and the deformation body, so as to provide data basis for the choice of treatment plan. china is vast in territory, and the geological conditions are complicated (xiao & huang, 2013). affected by natural factors, the main threat to road safety in mountainous areas is the landslide (zhou, zeng, & zhu, 2012). in the engineering construction of several decades, affected by the landslide, there have been many traffic disruption and engineering accidents caused by landslide disaster in china, which has caused serious economic losses to our country; this is not conducive to the construction of infrastructure in our country (zhenyu, chen, & wan, 2014). but whether it is expressway construction or railway construction will face the situation through the landslide mountain areas (song-yue, cao, hong, & zhou, 2009). in the process of slope excavation and tunnel construction, there are serious security problems because of the landslide (deng, 2014). especially in the southwest, heavy rain may further lead to instability of the slope in the rainy season, and exacerbate landslide deformation (jiang, 2010). the use of tunnel construction is to avoid the disturbance caused by the slope excavation; however, affected by natural factors and human impact of construction, it is difficult to carry on the supporting work of the slope (lin, 2011). figure 1. deformation and cracks of tunnel landslide in jimei 305the comprehensive treatment of the ancient landslide deformable body in the giant deep rock bedding based on the flac analysis model furthermore, assuming that the velocity component of node was ui, then the constant strain triangular element was obtained on the basis of the gauss’s theorem: ∂υ ∂x a u n s i i i s = ∑1 ∆ (2) the nodal velocity was used to calculate the strain increment of a time step unit: ∆ ∆e u x u x tij i j j i = 1 2 ∂ ∂ + ∂ ∂ (3) the total stress was calculated by the constitutive relation function. assuming that the stress tensor was ij, so there was: ∆ ∆ij ijf e= ( )……, (4) assuming that the unbalanced component of node in i direction was f ti( ) at t time, so that the motion equation of node was expressed as:   u t f t m i i= ( ) (5) based on the principle of virtual work and the formula (5), formula (6) was obtained: u u f t mi t t i t t i t+ − + ∆ ∆ ∆2 2         ( )= ∑ (6) firstly, the inverse analysis of parameters of slip soil was carried out under the natural working condition, the 6-6’ section plane in the zoneⅰ was used as the main sliding profile, and the flac two-bit numerical analysis model was established by means of the generalized engineering geological profile. at the same time, the homogeneous design test method was used to establish the computational scheme table of numerical analysis experiment homogeneous design of sliding deformation parameters and sliding zone intensive parameters (yan, guan, & xie, 2006). according to the homogeneous design test plan, the experimental parameters were substituted into the flac model, thus, the displacement values of the same position of the model and field monitoring were calculated (wang, kang, & wang, 2011). the coefficient reduction was used to obtain the parameters of the new rock and soil to carry out the cycle computing, so as to make the stability judgment. assuming that the reduction coefficient was f, the shear strength indexes were ct t, after the reduction, so there was: c c ft = / (7)  t f= ( )arctan tan / (8) the stability limit of the slope was determined by the mohr kulun yield criterion. the ratio of the maximum shear strength of slope and the actual shear stress of slope was expressed as: f c dt dll t n t t = ( )∫ ∫ +σ ϕ τ tan (9) combined with the strength factor reduction method, there was: 1 =     = ( )∫ ∫ ∫ ∫ c f f dll c dll dll n t n t+σ σ τ +σ ϕ τ tan tan (10) based on the analysis of the geological conditions of each partition, it can be seen that the whole jm landslide initially slid from the soft stratum of bedrock, and gradually disintegrated under the influence of external stress after a long period of time, which was characterized by multi-zone, multi-layer and multi-period sliding (shi, 2012). among them, there was an extrusion affected zone at the junction between the southeast of ⅰ3 and the zoneⅱ, the bedrock occurrence of the landslide was 14-55°∠33-15°, while the bedrock occurrence of slide bed in zoneⅰ3 was 14-31°∠33-35°, and the angle between the deformation direction of deformable body was 19-36°. the sliding surface slope of bedding landslide and the direction of the maximum sliding force were bedding tendency, namely 14-31°. in which, the direction of the maximum slip force was the northwest direction. the traction, extrusion and rub occurred in the east side of zoneⅰbecause of the influence of sliding body, thus forming a lateral impact area, as shown in figure 2. as a result, the surface topography in the affected area was messy, and the secondary landslide was developed. figure 2. deformation force and deformation direction of deformable body methodology in order to put forward the feasible measures for the treatment of jm tunnel landslide, it is necessary to analyze and calculate the displacement deformation value of the deformable body and tunnel, so as to ensure the safety of the whole project, and provide data support for the optimization of the scheme (xueyong, cai, zhao, & zhu, 2013). taking into account the actual situation of the project, the calculation results of special numerical analysis which are widely used in the analysis and calculation of slope and landslide of hydropower system at present were used as the main basis, at the same time, the monitoring data of the deformable body and tunnel deformation was adjusted and perfected to realize informatization construction and dynamic design (mei, peng, & xiao-hong, 2008). based on this, the flac model was used to carry on the specific analysis and research: flac finite difference mesh was used to carry on the calculation, which was divided into quadrilateral elements, each quadrilateral was made up of 2 triangular elements with constant strain. the weighted average values of all variables in a triangular element were taken as the element variable value of the quadrilateral. assuming that the integrals of the closed surface boundary and surface area of were s , a. the unit normal vector of the surface s is ni, and the increment ar length was ds. the lagrange’s method and the gauss’s theorem were used for the calculation:   f x da fn dsi sa = ∫∫ (1) 306 hongsheng zhou, yunsheng wang, tong shen, qianqian feng then the deformation displacement was used as the input variable, and the sliding deformation parameters and sliding zone intensive parameters were taken as output variables, so that the neural network rbf model was established (shuanping, 2010). the calculation model of 6-6’ profile is shown in figure 3. the homogeneous design test analytical data sample was used to train the grid, so that a neural network model was established to accurately reflect the nonlinear relationship among the deformation parameters of the sliding body, the sliding zone intensive parameters and the displacement value of the monitoring points, then the intensive parameters of slip deformation and slip zone were obtained by substituting the measured displacement values into the model. figure 3. calculation model of section 6-6 ‘section the anti-slide pile was used in the treatment scheme, and the pile element was used to simulate the anti-slide pile in the latter part of the calculation, so that the solid element was used to simulate the soil of sliding zone (liu, 2012). the shear force was generated between the constitutional unit of anti-slide pile and the surrounding rock mesh because of the slippage. assuming that the axial displacement of the pile was up, and the axial displacement of the rock and soil surface was um, and the length of the unit was l , so there was: f l cs u us sstiff p m= ( ) (11) the maximum shear force of unit length pile was: f l cs cs perimeters scoh c t sfric max tan= × ( ) ×+σ (12) the calculated results of the back analysis of parameters are shown in table 1. table 1. back analysis of parameters elastic modulus of sliding body e/ (pa) poisson’s ratio μ elastic model of slip zone e/(pa) poisson’s ratio μ slip band cohesion c/(kpa) internal friction angle of sliding belt phi(°) 7.00e+07 0.33 4.00e+07 0.35 33.2 18.6 secondly, the numerical analysis of the deformable body and tunnel was carried out, so that the rationality of the subsequent program was verified to guide the later supporting design. in this paper, the 6-6’ profile was selected to carry out the stress-deformation analysis of typical profile, and four kinds of working conditions were fully taken into account in the calculation: (1) natural + tunnel excavation + earthwork non-cleanup; (2) natural + tunnel excavation + earthwork cleanup; (3) rainstorm + tunnel excavation + earthwork cleanup; (4) rainstorm + tunnel excavation + earthwork cleanup + supporting structure. among them, for the third working condition, the strength of slip soil was reduced by 1.15 under the condition of rainstorm; for the fourth working condition, a row of anti-slide piles was arranged at the right side of the left and right tunnel hole of 10m, respectively. the calculated parameters of the model are shown in table 2: table 2. model parameter table lithology elastic modulus e/pa poisson ratio μ cohesive force c/ pa internal friction angle /° bulk density kn/m3 natural saturated slip zone 4.00e+07 0.35 30 19 2350 2400 sliding body 7.00e+07 0.33 36 28 2350 2400 bedrock 5.00e+09 0.26 6000 42 2640 primary support 9.00e+08 0.24 500 38 2500 two support 9.00e+09 0.2 6000 40 2500 results analysis through the analysis of the different working conditions of the 6-6’ profile after the tunnel excavation, the deformation of the landslide was understood, and the equivalent nephogram of landslide deformation was obtained, as shown in figure 4. as can be seen from the figure, under the condition of the natural state of tunnel excavation, the maximum deformation of slope was 57cm, the deformation mainly was concentrated in the upper part of the deformable body, the lower part would not happen (yang, peng, gong, & wu, 2012). under the condition of rainstorm + earthwork cleanup, the deformation of the slope was only about 2.3cm, and the deformation around the tunnel was very small. figure 4. equivalent nephogram of landslide deformation the maximum shear stress equivalent nephogram of the landslide is shown in figure 5. as can be seen from the figure, the maximum shear stress was mainly concentrated around the tunnel; the maximum shear stress value 307the comprehensive treatment of the ancient landslide deformable body in the giant deep rock bedding based on the flac analysis model was about 5.4mpa under the working condition of the natural + earthwork noncleanup, while the maximum shear stress value was about 2.5mpa under the working condition of rainstorm + earthwork cleanup. figure 5. the maximum shear stress equivalent nephogram of landslide through further analysis, the plastic zone distribution map of the deformation was obtained, as shown in figure 6. as can be seen from the figure, there was a shear stress strain increment zone in the upper part of landslide under natural conditions, but there was no in the lower part. it can be found that the whole slope was stable and there might be local sliding. under the working condition of rainstorm + earthwork cleanup, the strength of slip soil was reduced by 1.15 times, and the shear stress strain increment zone mainly appeared in the lower part of the landslide. thirdly, as can be seen from the analysis of the deformation and stress distribution law of tunnel, under the condition of rainstorm, the maximum shear stress around the tunnel was about 2.45mpa after the earthwork cleanup of deformable body, and the displacement value was about 5.7mm. when the anti-slide pile was applied to the periphery, the stress and displacement value of the tunnel were not changed greatly. in the previous calculations, under the condition of rainstorm, soil cohesion was reduced by 26.09kpa according to the natural condition, and the friction angle was also reduced by 16.6°. according to the previous calculation method, the 6-6’ profile was calculated. under the condition of rainstorm + earthwork cleanup, the deformation range of the landslide increased with the decrease of the cohesion of the sliding body, the corresponding displacement values of the sliding body also increased, the deformation of the left line tunnel was obvious, while the deformation of the right line tunnel was relatively small. when the soil cohesion values of sliding zone were 26.09kpa, 22.0kpa and 18.0kpa, respectively, the maximum displacement values of the corresponding slope were 2.13cm, 2.55cm and 5.38cm. the maximum shear stress increment of the landslide concentration gradually increased with the decrease of sliding soil cohesion, and enhanced with smooth consistency. it can be concluded that under the working condition of rainstorm + earthwork cleanup, the slope safety coefficient of different sliding soil cohesion of the landslide was taken, as shown in table 3. as can be seen from the table, the slope safety coefficient decreased with the decrease of sliding soil cohesion, but which was still greater than the requirements of the safety factor. table 3. the slope safety coefficient of different sliding soil cohesion of the landslide sliding soil cohesion (kpa) 26.09 22 18 safety factor 1.38 1.35 1.3 the results were calculated by using the neural network rbf model, and the inverse analysis was used to obtain the sliding soil cohesion value and the internal friction angle value under the natural working condition. according to the actual situation of the landslide, the slip soil cohesion value of zoneⅰwas 30.0kpa under the natural working condition, and the internal friction angle was 19°. under the working condition of natural + tunnel excavation, each section of the landslide body underwent significant deformation; and the deformation was concentrated in the middle and upper part of the deformable body, while the deformation of the lower part was not obvious. on the whole, jm landslide was in a relatively stable state. based on the comparison and analysis of the scheme of earthwork cleanup, the measures of earthwork cleanup for the rear part of the deformed body were determined; at the same time, taking into account the rainstorm condition, the sliding soil strength was reduced by 1.15 times, so that the overall sliding surface was not formed, which shows that the deformation induced by jm landslide and tunnel excavation was in a stable state. a row of anti-slide piles was arranged at the right side of the left and right tunnel hole of 10m, which had no effect on the deformation of the tunnel, therefore, the implementation of the anti-slide pile supporting engineering of two holes was not considered. through the comprehensive analysis, the in-situ treatment was determined, and the anti-slide pile was given priority. because the protective range of anti-slide piles on the tunnel and the top was greater than 5m, the pile top and above sliding body could be transferred to the leading edge slide to offset the sliding thrust, so that the sliding thrust of the tunnel range was considered. the residual sliding thrust values of the right side of left and right hole were calculated as 13674kn/m, 11653kn/m. therefore, multi rows of anti-slide pile needed to be set up in the upper deformation. figure 6. distribution of plastic zone of landslide body 308 hongsheng zhou, yunsheng wang, tong shen, qianqian feng conclusions because the excavation construction of jm tunnel may cause the old landslide which originally belongs to the giant bedding rock have further large sliding deformation, the cracks appeared in the tunnel and the ground surface need to be treated. in order to put forward the feasible measures for the treatment of jm landslide, it is necessary to analyze and calculate the displacement deformation value of the deformable body and tunnel, so as to ensure the security implementation of the whole treatment project and provide data support for the optimization of the scheme. the 6-6’ profile was taken as the research object, and the flac two-dimensional numerical analysis model was constructed; combined with the analytic results of uniform design test and actual monitoring displacement value, the intensity parameters of the slip mass deformation and the sliding zone were obtained by forward and reverse analysis. at the same time, under four different working conditions, the maximum displacement value of the slope obtained from different sliding soil cohesion was considered, it can be seen that the safety coefficient of slope decreased with the decrease of sliding soil cohesion, but which was still greater than the requirements of the safety coefficient. on the basis of the above analysis, the comprehensive treatment scheme of the jm tunnel landslide deformation body was proposed, including earthwork cleanup, grading supporting, and the back pressure by abandoning the cleaned earthwork at the leading edge. this study is based on the reinforcement supporting of landslide deformation as the main research object, although the mechanism of the tunnel and landslide deformation has been studied, the reinforcement supporting of the tunnel has not been studied in depth. acknowledgments this study has been supported by the funds for creative research groups of china (grant no. 41521002), the national key research and development program of china (grant no. 2017yfc1501000), the national natural science foundation of china (grant no. 41877235), and the independent fund of state key laboratory of geohazard prevention and geoenvironment protection (grant no. sklgp2015z001). the authors would like to express their gratitude for these financial supports. references chen, c., xiao, b., sun, x., & fang, g. (2014). causes analysis and treatment of the landslide of a highway tunnel entrance. construction technology, s2, 165-168. deng, x. p. (2014). analysis and comprehensive treatment of landslide for wenling tunnel. highway, 5, 119-122. jiang, b. (2010). diversion tunnel landslide treatment of bulungkol-gongur hydropower station. guangxi water resources & hydropower engineering, 2010-01. lin, x. (2011). long deep highway landslide disaster action mountain and treatment project countermeasures. fujian architecture & construction, 2, 67-69. liu, h. (2012). geological problems and treatment measures for highway tunnel through ancient landslide. journal of engineering geology, 20(4), 540-546. liu, j. & qian, y. (2014). cause analysis of qingfeng landslide and its treatment measures of gu-zhu highway. transportation science & technology, 2014-3. mei, l., peng, j. l., & xiao-hong, h. u. (2008). research on present situation of highway tunnel landslide and its treating method. storage transportation & preservation of commodities, 30(4), 41-42. ming-zhu, h. u., kong, d. g., & zheng, j. (2011). stability evaluation and treatment measures of miyi landslide along xichang-panzhihua highway. subgrade engineering, 6, 169-173. shi, x. (2012). landslide treatment and safety monitoring research of lowerimbedding unsymmetrical-pressure qing-ping tunnel. highway engineering, 4, 157-160. shuan-cheng, g. u., wang, b. q., wang, j., & he, w. g. (2015). comprehensive treatment technology and evaluation for tangjiayuan tunnel through the landslide section. journal of railway engineering society, 2015-01. shuanping, l. i. (2010). analysis and treatment for arch landslide on emergency lay-by on dawangding tunnel. technology of highway & transport, 2010-06. song-yue, y. e., cao, r. h., hong, h. t., & zhou, z. q. (2008). analysis of reviviscence characteristic and optimal selection of treatment projects of ancient landslide body located on the exit of zhangping tunnel. communications standardization, 7. sun, h. y., wong, l. n. y., shang, y. q., shen, y. j., & lü, q. (2010). evaluation of drainage tunnel effectiveness in landslide control. landslides, 7(4), 445-454. wang, k., kang, h. g., & wang, h. t. (2011). recovery execution in landslide at the portal of highway tunnel. advanced materials research, 250253, 1823-1826. xiao, z. w. & huang, l. (2013). analysis and treatment of landslide at the tunnel portal in nanjing road. applied mechanics & materials, 353356, 686-691. xueyong, y. u., cai, q., zhao, d., & zhu, d. (2013). analysis on genesis of landslide at the exit of wusha tunnel and the treatment design. subgrade engineering, 4, 167-171. yang, x. g., peng, j. g., gong, x. w., & wu, s. (2012). comprehensive treatment on landslide at the entrance of shejiapo tunnel on xuyong-gulin railway. subgrade engineering, 2012-05. yan, l. i., guan, z. h., xie, z. w. (2006). analysis on the causes of landslide in beihai road tunnel and the plans of emergency repair. west-china exploration engineering, 18, 164-166. zhenyu, y. i., chen, h., & wan, p. (2014). discussion of treatment techhique on great collapse of hanlinjie tunnel. highway engineering, 1, 177179. zhou, k., zeng, y., & zhu, w. (2012). slope treatment of highway tunnel with water-softening surrounding rock under complex geological conditions. journal of water resources & water engineering, 23(6). keywords: probability integral, subsidence predict, thick loose layer, main influence radius. palabras clave: probabilidad integral, predicción de subsidencia, capa suelta gruesa, radio de influencia principal. how to cite item li, j., yu, x., & liang, y. (2020). a prediction model of mining subsidence in thick loose layer based on probability integral model. earth sciences research journal, 24(3), 367-372. doi: https://doi.org/10.15446/esrj.v24n3.90111 the probability integral method is the most commonly used mining subsidence prediction model, but it is only applicable to ordinary geological mining conditions. when the loose layer in the geological mining conditions where the mining face is located is too thick, many inaccurate phenomena will occur when the movement deformation value is predicted by the probability integral method. the most obvious one is the problem that the predicted value converges too fast compared with the measured value in the edge of the sinking basin. in 2012, wang and deng proposed a modified model of probability integral method for the marginal errors in the model of probability integral method and verified the feasibility of the method through examples. in this paper, the method is applied to the prediction of surface movement under thick and loose layers after modified. through practical application, it is found that due to the angle between the working face and the horizontal direction, the average mining depth in the strike direction is different from the average mining depth in the inclined direction, and the main influence radius of the two main sections are often. therefore, based on this problem, this paper divides the main influence radius into trend and tendency and adjusts the parameters in the model to find the rules of the parameters. the original method uses a dynamic scale factor to adjust the predicted shape of the graph by adjusting the sinking coefficient. this study is aimed to set the scale factor to 0.5 and fix the value of the sinking factor, and propose to adjust the integral range and then adjust the shape of the graph to make it more in line with the actual measurement situation. abstract a prediction model of mining subsidence in thick loose layer based on probability integral model modelo de predicción de subsidencia en minas para una capa gruesa e inestable con base en el modelo integral de probabilidad issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n3.90111 el método integral de probabilidad es el modelo de predicción de subsidencia en minas más utilizado, pero solo es aplicable a las condiciones geológicas ordinarias de la minería. cuando una capa suelta en las condiciones de extracción geológica donde se encuentra la cara de extracción es demasiado gruesa, el método integral de probabilidad no es exacto al predecir el valor de deformación del movimiento. el problema más obvio es que el valor predicho converge demasiado rápido en comparación con el valor medido en el borde de la cuenca de hundimiento. en 2012, wang zhengshuai propuso un modelo modificado del método integral de probabilidad para los errores marginales en el modelo del método integral de probabilidad y verificó la viabilidad del método a través de ejemplos. en este documento, el método se aplica a la predicción del movimiento de la superficie bajo capas gruesas y sueltas después de la modificación. a través de la aplicación práctica, se encuentra que debido al ángulo entre la cara de trabajo y la dirección horizontal, la profundidad promedio de extracción en la dirección de impacto es diferente a la profundidad promedio de extracción en la dirección inclinada, y frecuentemente al radio de influencia principal de las dos secciones principales. por lo tanto, con base en este problema, este documento divide el radio de influencia principal en tendencia y ajusta los parámetros en el modelo para encontrar las reglas. el método original usa un factor de escala dinámico para ajustar la forma pronosticada del gráfico ajustando el coeficiente de hundimiento. este estudio busca establecer el factor de escala en 0.5, fijar el valor del factor de hundimiento, ajustar el rango integral y luego ajustar la forma del gráfico para que esté más en línea con la situación de medición real. resumen record manuscript received: 01/06/2019 accepted for publication: 07/05/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 3 (september, 2020): 367-372 jingxian li1,2, xuexiang yu2,*, ya liang 1school of earth and environment, anhui university of science and technology, huainan 232001, china 2school of geomatics, anhui university of science and technology, huainan 232001, china 3school of mathematics and big data, anhui university of science and technology, huainan, 232001, china *corresponding author: xxy16321@163.com m in in g e n g in e e r in g https://doi.org/10.15446/esrj.v24n3.90111 https://doi.org/10.15446/esrj.v24n3.90111 mailto:xxy16321@163.com 368 jingxian li, xuexiang yu, ya liang introduction as one of the main energy sources for human life and economic development, coal occupies a very important position in the world energy consumption structure. although with the continuous promotion of global clean energy, the proportion of coal in primary energy continues to decline. anyway, coal is still the main component of the world’s primary energy consumption structure (29.2%) (“geo-informatics,” 2019). as the world’s largest energy producer and consumer, coal has always dominated the primary energy structure in china. it is expected that by 2030, the proportion of coal will remain above 50%. coal can drive the country’s economic growth, but the problem of irreversible impact on the surface during mining and even for a long time after mining cannot be ignored. because the underground coal seam is artificially destroyed, the original stress balance of the rock layer is destroyed, then the phenomenon of surface subsidence and movement occurs when the impact is propagated to the surface, resulting in geological disasters such as damage to landmark buildings and water accumulation. so it is particularly important to predict the surface movement and deformation that it may cause before mining. the subsidence prediction methods established by mining subsidence workers in china mainly include probability integral method, negative exponential function method, typical curve method, integral grid method, weibull distribution method, spline function method, hyperbolic function method, pearson function method (zhang & lei, 2012), mountain surface deformation prediction method, three-dimensional layered medium theory prediction method, and strip mining prediction method based on pallet theory. among these prediction methods, other methods have not been widely used due to their limitations or scope of use. probability integral has become the main method for predicting mining subsidence in china due to its solid theoretical foundation, easy computer implementation, and good application effect. the probability integral method was developed by scholars such as liu baochen and liao guohua of china in the 1960s on the basis of the stochastic medium theory proposed by the polish scholar litwiniszyn in 1957 (wang et al., 2013). it is called the probability integral method because of the calculation of the probability integral part of its calculation process. the probability integral method is mainly applicable to the case of near-horizontal coal seams or gently inclined coal seams (fan et al., 2014). the probability integral method is currently the most commonly used model in the area of mining subsidence prediction in china. its basic idea is to treat the surface and downslide movements as random events and decompose the mining area into infinitely many tiny units to calculate the sum of the surface influences and surface subsidence. this method is based on the stochastic medium model theory. the method is expected to have good results when the geological factors of the mining have little effect on the surface movement, but if the geological conditions of the working face are special (such as loose layers thicker), which has a greater impact on surface subsidence, the probability integral method will not reflect these changes well (wang et al., 2019). so it needs to be adjusted on the basis of the original model to make it more suitable for the prediction under different geological mining conditions. aiming at the problem of converging too fast at the edge of the subsidence basin when predicting the surface subsidence using the probability integral method, wang and deng (2012) proposed that treat surface subsidence as a weighted sum of the effects of mining from two working faces with different main influence radius (hereinafter referred to as the original model) (wang et al., 2018). this to a certain extent corrects the problem of converging too fast at the edge of the basin when using the probability integral model to predict surface subsidence (zhou et al., 2016). however, the original model uses the same influence radius along the strike and tendency direction. in actual application, it is found that the main influence radius values of the strike and tendency are often different. so the main influence radius in this model is divided into a strike influence radius and a tendency influence radius and integrates parameters are chosen to be the changeable parameters instead of subsidence coefficient in the original model in this article. characteristics of mining subsidence under thick loose layer the loose layer is the quaternary and neogene strata and is composed of soil, sand, gravel, and pebble layers. after the hard rock has undergone geological processes such as erosion, transportation, and deposition, evacuated sediments formed due to the unconsolidated hardened diagenesis are called loose layers. since loose layers are formed in different environments, times, and places, they will have different properties. in terms of how thick the loose layer can be called “thick”, it is generally believed that the loose layer is called a thick loose layer when the thickness exceeds 50m and is called a huge thick layer when the thickness exceeds 100m. it is of great theoretical and practical significance to deeply study the particularity of surface movement and deformation and the parameter variation law of thick loose layer mining area for accurately predicting the impact of underground mining on various protective objects on the surface (“geoinformatics,” 2019). at present, the law of surface movement and deformation of underground mining in thick and loose layer mining areas has been studied in china and other countries (ding et al., 2018), and the particularity of surface movement and deformation of thick and loose layer mining areas has been obtained: the subsidence coefficient is too large, many of which are around 1.1 or greater; the vertical deformation above the coal pillars in the thick loose layer mining area is abnormal compared to general geological mining conditions; the surface movement range is large, and the horizontal movement range is often greater than the vertical movement range; the upper mountain boundary angle, the moving angle is less than or close to the lower mountain boundary angle and moving angle, etc. although these studies have obtained some macroscopic laws, due to the differences in the characteristics of geological mining conditions in various mining areas, there has not yet been obtained a comprehensive study on the relationship between the characteristics of surface subsidence in thick loose layer mining areas and geological mining factors, and the influence laws. as a result, it is difficult to select the parameters when predicting the surface movement and deformation of coal seam mining in thick and loose layer mining areas (ma & hu, 2013). compared with the actual observation values, the prediction error is large. therefore, it is of great theoretical and practical value to deeply study the subsidence mechanism and deformation law of coal seam mining in a thick and loose layer mining area, to accurately predict the movement and deformation of the ground surface, to reasonably keep various types of protective coal pillars, and to protect various protective objects on the surface (hou et al., 2018). figure 1. comparison of surface subsidence predicted by probability integral method and real subsidence value comparison of two model original model assuming that a unit volume of coal seam b (s, t) is mined in a gently inclined coal seam, the subsidence value of the surface point a (x, y) caused by the coal seam mining in this unit can be calculated by equation 1 in the original model. w x y r es x s y t h c r , tan ( ) = ∗ ( ) + + ∗( )       1 2 2 2 2 −π − − θ (1) where r is the main influence radius, h is the mining depth, and θ is the maximum sinking angle. assuming that the working face during actual mining is a rectangular working face, the calculation formula for the sinking value of any point a (x, y) of the near-horizontal working face with the strike length of d3 and the inclined length of d1 according to the probability integral method is: 369a prediction model of mining subsidence in thick loose layer based on probability integral model where s3 is the offset of the inflection point in the downhill direction, l = d3-s3-s4, which is the length of the theoretical strike, d3 is the length of the real strike, l d s s= ∗ ( )1 1 2− − θ+α θsin / sin , which is the theoretical tendency length, d1 is the real tendency length, α is the coal seam inclination, θ is the maximum sinking angle, r is the main influence radius of the tendency, r is the main influence radius of the strike direction. thus modified model sinking calculation formula is: w x y w x y w x ys r r s r r, , . , .( ) = ( ) ∗ + ( ) ∗1 1 2 20 5 0 5 (7) where w x ys r r1 1 ,( ) and w x ysr r2 2 ,( ) are the calculation results brought into equation 6 while r = r1, r = r1 and r = r2, r = r2 . w x y r r e x s r y t h c r s co , * * tan ( ) = ( ) + + ∗( )        ∗ 1 1 1 1 2 2 2 2 2 2 1 −π − − θ ss l s x s r y t dtds r r e α −π − − ∫∫ + ( ) + 3 2 2 2 1 1 1 0 5 1 1 * . * * ++ ∗( )        ∗∫∫ h c r s cos l s dtds tan * . θ α 2 2 2 13 1 0 5 (8) parameter analysis main influence radius-r affected by underground coal mining, the subsidence value of the surface approaches or reaches the maximum subsidence value w0 at the position of x> r, and other surface movement and deformation mainly occur within the range of x = -r ~ + r which is shown in figure 2. so we call r the main influence radius. the main influence radius r is defined as the distance from the inflection point of the subsidence curve to the maximum subsidence point or the distance from the inflection point to the boundary point of the moving basin, which is a parameter that characterizes the concentration of surface movement. they are mainly related to the nature of overburden and mining depth. figure 2. schematic diagram of main influence radius in order to study the effect of r2 and r2 values on the predicted subsidence shape along strike direction and tendency direction. the control variable method is used to compare them in this paper as follows: (1) the effect of r2 on the shape of the main section subsidence predict the results calculated by the improved model with a different value of r2 while other parameters keep unchanged are shown in figures 3 and 4. figure 3. the impact of r2 on the predicted main section w x y r es x s y t h c r si cos l s , tan ( ) = ∗ ( ) + + ∗( )        ∗∫ 1 2 1 2 2 2 3 −π − − θ α∫∫ dtds (2) where s3 is the offset of the inflection point in the downhill direction, l = d3-s3-s4, which is the theoretical strike length, d3 is the real strike length, l d s s= ∗ ( )1 1 2− − θ+α θsin / sin , which is the theoretical tendency length, d1 is the real tendency length, α is the coal seam inclination, θ is the maximum sinking angle. since one of the main problems of the probability integration method is the phenomenon of convergence too fast at the edge of the subsidence basin compared with the actual subsidence, a modified probability integration method model is proposed by wang and deng (2012): w x y w x y r p w x y r ps s, , ; , ;( ) = ( ) ∗ + ( ) ∗ ( )1 2 1 (3) where w x y rs , ; 1( ) and w x y rs , ; 2( ) are the calculation results brought into equation 2 while r = r1 and r = r2 and p is the proportionality coefficient. w x y r e si cos l s x s y t h c r, tan ( ) = ∗ ∗ ( ) + + ∗( )     ∫∫ 1 1 2 1 3 2 2 1 2 α −π − − θ   ∗ ( ) + + ∗ + ∗∫∫ dtds p r e si cos l s x s y t 1 2 2 1 3 2 α −π − − hh c r dtds p ∗( )        ∗ ( ) tan θ − 2 21 2 1 (4) it can be seen from equation 4 that the correction method regards the ground subsidence caused by coal mining as the combination of the coal seam with the main influence radius r1 and the coal seam with the main influence radius r2 according to a certain proportional coefficient. it is verified by an example that this method has a certain effect to improve the phenomenon that the probability integral method converges too fast at the edge of the subsidence basin compared with the actual subsidence. however, in practical application, it was found that when analyzing the ground subsidence caused by mining in a working face, the main influence radius along the strike direction and the main influence radius along the inclined direction are different, which reduced the accuracy of the model. therefore, this paper proposes to divide the main influence radius into the influence radius in the striking direction and the influence radius in the tendency direction on the basis of this original method, and changed the parameter adjustment method in the original model, verified the accuracy of this method in predicting the deformation value of the ground movement through examples. improved model according to the above description, the main influence radius in the formula is divided into the influence radius along the strike direction and the influence radius along with the trend. the calculation formula of the sinking value of the surface point a (x, y) caused by unit mining is: w x y r r es x s r y t h c r , tan ( ) = ∗ ∗ ( ) + + ∗( )       1 1 2 2 2 2 2 −π − − θ (5) where r is the main influence radius in the strike direction, and r is the main influence radius in the tendency. h is the coal mining depth, and θ is the maximum angle of influence. assuming that the working face during actual mining is a rectangular working face, the calculation formula for the sinking value of any point a (x, y) of the near-horizontal working face with the strike length of d3 and the inclined length of d1 according to the probability integral method is: w x y r r e si cos l s x s r y t h c r, tan ( ) = ∗ ∗ ∗ ( ) + + ∗( )   ∫∫ 1 1 3 2 2 2 2 21 α −π − − θ      dtds (6) 370 jingxian li, xuexiang yu, ya liang figure 4. the effect of r2 on the prediction of the prone main section (1) the effect of r2 on the shape of the main section subsidence predict the results calculated by the improved model with a different value of r2 while other parameters keep unchanged are shown in figures 5 and 6. figure 5. the impact of r2 on the predicted main section figure 6. the effect of r2 on the prediction of the prone main section after the statistical analysis of the measured data, the main effecting angle (tan β) estimated by the new model for the subsidence predicted value of the near-level coal seam mining are: tan1=3.177-1.232*h/h tan2=0.804+0.127*h/h tan1=2.223+1.505*h/h tan2=0.915+0.202*h/h it can be seen that the main influence of tangent is proportional to the thickness of the loose layer. taking the main influence tangent of the strike direction as an example, tanβ1 is larger than tanβ0, and tanβ2 is smaller than tanβ0. assuming that the main influence radius in the original probability integral method model is r0, r1 is smaller than r0, and r1 mainly controls the position of the inflection point, r2 mainly controls the degree of boundary convergence. it also can be seen that when the strike and the tendency direction are both fully mined, the r2 value mainly controls the convergence of the edge subsidence value in the inclined direction, while the r2 value mainly controls the convergence of the subsidence value along the strike line of the subsidence basin. it can be seen from the statistics of the examples that the main influence radius of the trend and the tendency are more or less different, so this method is helpful to improve the prediction accuracy. points range in addition, according to the definition of the probability integral method, the value of the sinking coefficient q is mainly related to the roof management method. but the definition of the inflection point offset is based on the principle of equivalent coal seam due to the cantilever effect of the coal goaf, the predicted subsidence calculated by the original coal seam directly according to the probability integration method does not match the actual situation, so the definition of the inflection point offsets are proposed to build an ideal coal seam to make predicted subsidence value more match the actual situation. so the integral range of the trend and tendency during the prediction is the range of the imaginary working face (the trend is s3-1, and the tendency direction is s1*cosα). the principle of the new model is that two-unit coal seams with different influence radium will jointly affect the surface subsidence with the same ratio, which changes the integration range. therefore, this article chooses to change the integral range on the basis of the original inflection point offset. after the statistical analysis of the measured data, the integration range estimated by the new model for the subsidence value of the near-level coal seam mining are: ax=1.375+0.409*s3 bx=l ay=36.801+0.097*s1 by=271.014+0.002*l example verification the situation of the observatory in the mining area xieqiao mine 21113 working face is located in huainan city, anhui province, and is the first mining face of xieqiao mine. the upper boundary depth of the working face is 714m, the lower boundary depth is 764m, the average mining depth is 739m. the seam thickness is 5.2m, and the inclination angle is 12 °, which belongs to the class of gently inclined coal seam. the coal mining method is the long-arm method, and the roof management method is the full collapse method. the real strike length of this work face is 1968.7m, the real tendency length is 234.2m. one observation line is set along the strike direction and two half observation lines are set along the tendency direction. the setting diagram of the 21113 working face observation station is shown in figure 7. figure 7. schematic diagram of 21113 working face observation station calculation results the comparison of the results of solving the parameters of the working face by using the mold loss method combined with two different models is shown in table 1. the predicted results of the model corresponding to the parameters calculated by the original model and improved model proposed in this paper are compared with the measured values of the trend and tendency direction as shown in figures 8 and 9. the contour map of expected sinking corresponding to the probability integral method is shown in figure 10, and the contour map of expected sinking corresponding to the improved model is shown in figure 11. figure 8. comparison chart of expected value on the main section 371a prediction model of mining subsidence in thick loose layer based on probability integral model expected, and both the original model and the improved model have a significant effect in boundary correction. conclusion from the above analysis, it can be seen that the original model regards the surface movement deformation as a weighted superposition of the effects of two mining units with different main influence radius, which has an optimization effect on the problem that the probability integral method converges too fast at the edge of the subsidence basin compared with the actual subsidence. on the basis of the original model, the improved algorithm proposes a method in which the main influence radium is divided into two directions-strike main influence radium and tendency direction main influence radium, which improves the accuracy while optimizing the boundary problem. compared with the original model, different parameters were selected for an adjustment instead of the subsidence coefficient in the original model, and it is assumed that the impact of the two mining units is equal, in other words, the scale factor in the new model is 0.5. but the integral parameters can no longer be set according to the boundary range of the ideal coal seam, so they were set as new adjustment parameters. the actual application results prove that the new model is more in line with the actual coal mining subsidence. however, if it is used in steeply inclined coal seams, the main influence radius of the uphill direction of the inclined direction is different from the main influence radius of the downhill direction. acknowledgment the work is supported by the research on key technology of highprecision and rapid monitoring of mining subsidence based on gps / bds combination (no. 41474026). references ding, r., liu, h., & dong, j. (2018). gisbased subsidence prediction of yuzhou goaf area at the middle route of the south-to-north water diversion project. geotechnical and geological engineering. fan, h. d., gu, w., qin, y., xue, j. q., & chen, b. q. (2014). a model for extracting large deformation mining subsidence using dinsar technique and probability integral method. transactions of nonferrous metals society of china. hou, d., li, d., xu, g., & zhang, y. (2018). superposition model for analyzing the dynamic ground subsidence in mining area of thick loose layer. international journal of mining science and technology. ma, w. j., & hu, h. f. (2013). the strata movement regularity and parameter simulation analysis in multiple seams repeated mining. applied mechanics and materials. wang, l., deng, k., fan, h., & zhou, f. (2018). monitoring of large-scale deformation in mining areas using sub-band insar and the probability integral fusion method. international journal of remote sensing. wang, l., deng, k., & zheng, m. (2020). research on ground deformation monitoring method in mining areas using the probability integral model fusion d-insar, sub-band insar and offset-tracking. international journal of applied earth observation and geoinformation. wang, l., jiang, c., wei, t., li, n., chi, s., zha, j., & fang, s. (2019). robust estimation of angular parameters of the surface moving basin boundary induced by coal mining: a case of huainan mining area. ksce journal of civil engineering. wang, y. h., zhou, j. w., & wen, b. (2013). subsidence prediction under thick alluvium based on probability integration method. applied mechanics and materials. wang, z., & deng, k. (2012). edge correction model of subsidence prediction by probability integral method. journal of xi’an university of science and technology, 32(04):495-499. table 1. parameters fitted by the three models according to the measured values probability integral method improved model revised model subsidence coefficient 0.94 1.14 0.94 main influence radius(m) 339 r1=249.66 r1=339 r1=209 r2=731.68 r2=939 r2=739 mining impact propagation angle(°) 85.48 85.48 85.48 points range s3=66 s3=66 ax=-34 s4=45 s4=45 bx=1969 s2=-6 s2=-6 ay=49.02 s1=-21 s1=-21 by=309.79 figure 9. comparison chart of expected value on the main section figure 10. surface subsidence contour predicted by probability integral model figure 11. surface subsidence contour predicted by an improved model from the comparison of the calculation results of probability integral method and the improved model in figures 7 to 10, it can be seen that the accuracy of the improved model has been improved when the full face is 372 jingxian li, xuexiang yu, ya liang xie, y., zhang, a., liu, h., & feng, l. (2019). geo-informatics in sustainable ecosystem and society. springer science and business media llc. zha, j. f., feng, w. k., zhu, x. j., & mi, l. q. (2011). analysis accuracy and robustness of parameters inversion in probability integral method by genetic algorithm. seventh international conference on natural computation. zhang, z., & lei, t. (2012). one design method of pipeline in mined-out area, volume 4 pipelining in northern and offshore environments strain-based design risk and reliability standards and regulations. zhou, d., wu, k., li, l., diao, x., & kong, x. (2016). a new methodology for studying the spreading process of mining subsidence in rock mass and alluvial soil: an example from the huainan coal mine, china. bulletin of engineering geology and the environment. keywords: geological ecology; high resolution remote sensing; sampling point; bing algorithm; selective search algorithm. palabras clave: geología ambiental; teledetección de alta resolución; punto de muestreo; algoritmo bing; algoritmo de búsqueda selectiva. how to cite item bi, t. (2020). optimal allocation algorithm of geological and ecological high-resolution remote sensing monitoring sampling points. earth sciences research journal, 24(1), 105-110. doi: https://doi.org/10.15446/esrj. v24n1.85531 the purpose of this study is to solve the problem of an unsatisfactory image representation of monitoring sampling points in high-resolution remote sensing due to the complexity of geological ecology. firstly, three algorithms used in remote sensing technology were introduced, that is, extraction algorithm of monitoring sampling point (selective search algorithm), discriminant algorithm (support vector machine), and bing algorithm. then, the bing algorithm was improved. finally, the superiority of the improved bing algorithm was verified through the experimental data set. the results showed that the selective search algorithm could generate more candidate windows in remote sensing images and had better adaptability. the improved algorithm had a higher quality of candidate windows extracted from remote sensing images. although the ibing algorithm could significantly improve the extraction speed of remote sensing, the detection time of each image became larger. such testing times were still acceptable. therefore, in this research, the allocation algorithm of geological and ecological high-resolution remote sensing monitoring sampling points was optimized, which had a good guiding significance for the application of remote sensing technology in environmental and geological research. abstract optimal allocation algorithm of geological and ecological high-resolution remote sensing monitoring sampling points algoritmo de asignación óptima de puntos de muestreo de monitoreo de detección remota geológica y ecológica de alta resolución issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.85531 el propósito de este estudio es resolver el problema de la representación de imagen insatisfactoria de los puntos de muestreo de monitoreo en la teledetección de alta resolución debido a la complejidad de la ecología geológica. en primer lugar, se introdujeron tres algoritmos utilizados en la tecnología de detección remota, a saber, el algoritmo de extracción del punto de muestreo de monitoreo (algoritmo de búsqueda selectiva), el algoritmo discriminante (máquina de vectores de soporte) y el algoritmo bing. luego de esto el algoritmo bing fue mejorado. finalmente, la superioridad del algoritmo bing mejorado se verificó con base en un conjunto de datos experimentales. los resultados mostraron que el algoritmo de búsqueda selectiva podía generar un mayor número de ventanas candidatas en la imagen de teledetección y tenía una mejor adaptabilidad. el algoritmo mejorado tenía mayor calidad de ventanas candidatas extraídas de imágenes de teledetección. aunque el algoritmo ibing podría mejorar en gran medida la velocidad de extracción de la teledetección, el tiempo de detección de cada imagen se hizo mayor. estos tiempos de prueba aún eran aceptables. por lo tanto, en esta investigación se optimizó el algoritmo de asignación de puntos de muestreo de monitoreo de detección remota geológica y ecológica de alta resolución, que tenía una alta importancia orientadora para la aplicación de la tecnología de detección remota en la investigación geológica y ecológica. resumen record manuscript received: 06/01/2019 accepted for publication: 26/07/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 105-110 taifu bi 1key laboratory of regional environment and eco-restoration, shenyang university, shenyang liaoning 110044, china * corresponding author: co m pu te r sc ie nc e a pp li ed to g eo lo gy 106 taifu bi introduction the existing remote sensing technology has many advantages, such as wide coverage, fast updating speed and large amount of information. the above advantages enable remote sensing technology to acquire large range of data in a relatively short time, so remote sensing technology is widely used. at present, it is most widely used in environmental and geological fields, which can monitor soil erosion, environmental protection, vegetation cover and land use (gorji et al., 217). remote sensing technology can be divided into broad sense and narrow sense. in the broad sense, it refers to the use of remote non-contact means to capture the information of monitoring research objects. in a narrow sense, the specific meaning refers to the technology that uses existing space technologies (satellites or aircraft) as a carrier platform and carries relevant sensors to perceive or observe some characteristics of the research target and analyze its characteristics (gao et al., 2017). in remote sensing image technology, target detection sampling point is one of the difficulties in remote sensing monitoring for geological ecological monitoring system. with the development of science and technology, the operating time and spatial resolution of remote sensing technology have been continuously improved, and the information of research objects collected by remote sensing images has become richer (lengyel et al., 2017). at present, the application of target detection technology based on remote sensing image is more and more extensive. high-resolution optical remote sensing images can obtain more shape and texture information of ground objects, have stronger discrimination ability to ground objects, and can more intuitively display the surface conditions. however, due to the complexity of geology and ecology, it is difficult to extract images and express effects when monitoring sampling points. therefore, these problems of target detection based on high resolution optical remote sensing image need to be solved. most of the target detection technologies of remote sensing images are infrared remote sensing, synthetic aperture radar or medium-low resolution remote sensing images. these traditional detection technologies can’t effectively adapt to the complexity of high-resolution optical remote sensing images and the diversity of targets (silva et al., 2017). and as the spatial resolution of remote sensing gets higher and higher, the image gets closer and closer to the natural image. however, the target detection technology of natural image (such as artificial intelligence algorithm) has developed rapidly in recent years. the combination of artificial intelligence algorithm and image target detection will further improve remote sensing technology. therefore, it is of practical significance to explore more effective target detection algorithms based on high-resolution optical remote sensing images. to sum up, in this research, the configuration algorithm of high-resolution remote sensing detection technology was optimized by studying highresolution remote sensing target detection technology. it is hoped that the effect of high-resolution remote sensing target detection can be improved. methods monitoring sampling point extraction algorithm selective search algorithm firstly divides the pixels of the same object with similar features (such as color, texture and scale) in the image into super pixels, and obtains the initial small area set. then, the similarity definition and merge strategy definition of pixels are carried out to merge super pixels and form a hierarchical structure, which can generate candidate windows. in order to obtain better robustness, selective search algorithm adopts multiple color spaces and multiple similarity calculation strategies for better robustness (sürme et al., 2017). bing algorithm is simple and efficient in the extraction of candidate windows, which is very suitable for the extraction of candidate windows. bing algorithm is a classifier with two stages. first, the first stage is to scan the images waiting to be detected according to different scales of sliding. then the subgraph of the candidate window is unified to 8*8, and finally the linear scoring model needs to be used for scoring. the linear scoring model can be expressed as equations (1), (2), and (3). w r 64 (1) s w gl l= ( ), (2) l i x y= ( ), , (3) among them, l represents the window position; i represents the window scale; (x, y) indicates the coordinates of the upper left corner; sl represents filter score; and gl represents the window feature vector. in the bing algorithm, the w linear model is trained by svm in the first stage, the normal pattern gradient features of the real target window are taken as positive samples, and the paradigm gradient feature of the randomly sampled background window is taken as a negative sample. the second stage is to solve the problem that different scales contain different possibilities of targets. it needs to train a scoring calibrator for each scale window. the candidate window score extracted in the first stage is corrected, as shown in equation (4). l i l iv s t= • + (4) among them, vi and sl represent the calibration coefficients of i learning at different scales. in order to speed up the process of image extraction and scoring, the bing algorithm uses a binary approximation method. the two-valued basis vector is used to approximate the linear model, as shown in equation (5). w j j j nw ≈ β α =∑ 1 (5) among them, nw represents the number of base vectors; j represents a base vector; and j indicates the coefficient. j can be represented by another binary vector and its complement form, as shown in equation (6). α α −αj j j= + + (6) among them, α ∈j + { }0 1 64, . equation (6) can perform the fast bitwise and bit count operation of the score of the binarized feature b, which can be expressed by equation (7). w b b bj j j nw , ,( ) ( )( )+=∑≈ β α −21 (7) bing algorithm expresses ng eigenvalue through equation (8): g bl b k k l k ng ≈ −2 1 , =∑ (8) through the above changes, the score of whether the final image window contains the target can be approximated as equation 9: s w g b bl l j b k k n j k l j n gw = ( ) ( )( )= += ∑∑, , ,≈ β α −−2 221 (9) among them, 2 2b k j k lb b − α −+( )( ), , can perform cpu atomic operations, which improves the speed of bing algorithm. discriminant algorithm support vector machine (svm) is adopted in the general discriminant algorithms of remote sensing image technology. the basic idea of support vector machine is shown in figure 1. the algorithm derivation process of svr is as follows (ellis & coppins, 2017). first, the training set sample pairs were given:t x y i li i= ( ) = …{ }, , , , ,1 2 , x ri d , y ri  . in the high-dimensional eigenspace, the linear regression function is established. 107optimal allocation algorithm of geological and ecological high-resolution remote sensing monitoring sampling points f x a a x x b a a k x x b i i i l i j i i i j i l (16) the principle diagram of support vector regression machine is shown in figure 2. k(x1,x) k(x2,x) k(xn,x) ∑ the hidden layer the input layer x=[x1,x2,…,xn] y bias x2 x1 xn figure 2. schematic graph of svr improved bing algorithm because the bing algorithm was easy to miss the fuzzy monitoring samples, in this research, the candidate region extraction algorithm was improved. firstly, different weak feature channels were extracted from each training sample, and a scoring model was developed for each weak feature channel. then the score of a scoring model was trained by each weak feature channel as the score of candidate window. because the integration part adopted a simple linear weighting method, the accuracy of each model determined the weight of the model. thus, the corresponding scoring model for each window was obtained, as shown in equation (17) and (18). s w gkl k kl= ( ), (17) l i x y= ( ), , (18) among them, wk is the parameter of scoring model k; skl represents the score of window l by the scoring model; and gkl indicates the feature value k at window 1. therefore, the final window score in the first stage is expressed by equation (19). g sk kl n = ∑  (19) among them, k is the weight of the eigenvalue of the scoring model. after correction according to the scoring model of the second stage, the window score is shown in equation (20). o a s v tl k kl i i n = ∗ +∑ (20) in bing algorithm, ng feature graph is calculated for test sample points, and then binary approximation window ng feature and scoring model f x w x b( ) = ( ) + (10) among them, w is weight vector; (x) represents the nonlinear mapping function; and representative bias. the c-linear insensitive loss function is defined as equation (11). l f x y y f x y f x y f x ( )( ) = ( ) ( ) ( )     , , , , ε − ≤ε − −ε − >ε 0 (11) among them, f (x) is the predicted value, y is the corresponding real value, and  is the insensitive coefficient to control the fitting accuracy. secondly, relaxation variables, that is, i and j, are introduced. according to srm criterion, the minimization objective function is constructed, as shown in equation 12. min * 1 2 2w c i i i l l + +( ) = ∑   (12) s t y w x b y w x b i l i i i i i i i . . , , , , , * − φ − ≤ε ξ − φ ≤ε ξ ξ ≥ ξ ≥ ( ) + + ( ) + + = …   1 2 0 0     (13) among them, c > 0 is the penalty factor, and represents the punishment intensity for samples with an error greater than 1 during training. lagrange equation is introduced into equation (13) to obtain dual optimization problem, as shown in equation (14) and (15): min * * * * , 1 2 111 a a k a a a a a a yi i j j i i i i i i l i l i j − − − − ε −( ) ( ) ( ) + ( ) === ∑∑ ll ∑           (14) s t a a a a c i i i l i i . . , , *− ∈ ( ) =         = ∑ 0 0 1 (15) among them, k x x x xi j i j,( ) = ( ) ( )  is a kernel function. the commonly used kernel functions are polynomial kernel function, radial basis kernel function and sigmoid kernel function. and the radial basis kernel function was selected in this research. fourthly, ai and ai * can be solved from equation (15), and then w* and b * can be obtained, the regression function is shown in equation (16). f(x) support vector optimal classification line * * * * * * * * * * * · · ·· · · · · · figure 1. the basic idea of svr 108 taifu bi parameters are used. such a process is to only use cpu atomic operation. ng feature graph calculates the gradient amplitude of each pixel by collecting the edge intensity information in the original graph. however, because the logic of the ng feature map is relatively simple, the influence of noise on it is particularly prominent. based on the shortcomings of bing algorithm mentioned above, sobel feature graph with better edge information capture effect was adopted in this research to replace ng feature graph with weak edge information. moreover, the local binary pattern (lbp) feature graph and difference of gaussian (dog) feature graph were introduced into the sobel diagram. the meaning expressions of sobel feature graphs are shown in equations (21)~(23): g ax = + + +           ⊗    1 0 1 2 0 2 1 0 1 (21) g ay = + + + +           ⊗ 1 2 1 0 0 0 1 2 1    (22) g g gx y= + 2 2 (23) among them, a represents the original graph matrix; gx is the gradient of the horizontal direction of the image; gy is the gradient for the vertical direction of the image; and g is the gradient magnitude of the feature graph matrix. as can be concluded from equation (21), sobel feature diagram adopts the grayscale diagram of the original image of 3*3 sobel operator convolution. the approximate gradients of the horizontal and vertical directions are calculated respectively, and then the sum of the squares of the approximate gradients of the horizontal and vertical directions is carried out to obtain the corresponding gradient amplitude of each pixel. lbp feature map is a texture feature proposed by yavad and congalton (2017) to describe images. with the development of technology, it is widely used in human classification and pedestrian detection. the calculation process of lbp code is shown in figure 3. dog feature map is obtained by subtracting two fuzzy images of different degrees from the original image. the two-dimensional gaussian kernel function is defined as follows: g x y x y σ πσ − σ1 1 2 21 2 2 2 1 2 , exp( ) = +       (24) g x y g x y f x y1 1, , ,( ) = ( ) ∗ ( ) (25) g x y g x y f x y2 2, , ,( ) = ( ) ∗ ( ) (26) the dog feature map can be obtained by subtracting the two images, as shown in equation (27). g x y g x y g x y f x y g x y f x y1 2 1 2, , , , , ,( ) ( ) = ( ) ∗ ( ) ( ) ∗ ( )− −σ σ = ( ) ∗ ( )g g f x yσ σ−1 2 , = ∗ ( )dog f x y, (27) that is, dog can be expressed by equation (28): dog g g e e x y x y ≈ − π σ − σ σ σ − σ − σ 1 2 2 2 1 2 2 2 2 21 2 1 1 1 2 2 2 =       +( ) +( )/ / (28) among them,  1 and 2 represent gaussian kernel parameters. when dog images are applied in different scenarios, the proportions of two gaussian kernel parameters are also different. for example, when dog is used for image enhancement,  1 2 1 4: := . in this study, the purpose of dog was image enhancement, so 1 0 5= . , and 2 2 0= . were adopted. results and discussion analysis of the results of the selective search algorithm in this research, the applicability of selective search algorithm in remote sensing image to target extraction of monitoring sampling points and the superiority of improved bing algorithm were mainly verified. the data sample set was divided into training set and test set, and the ratio of training set to test set was 2:8, this experiment was tested in fast mode, and only two different color spaces and two similarity functions were used. in remote sensing images, the size of target collection points was determined by the spatial resolution of images. therefore, in this research, unreasonable candidate windows were filtered when selective search algorithm results were obtained, as shown in figure 4. according to figure 4, with the increase of pixel size of remote sensing image, the number of candidate windows generated by selective search algorithm was also increasing. in addition, when there was no constraint, the average number of candidate windows generated by selective search algorithm under different pixel sizes was greater than the average number of candi date windows generated by selective search algorithm after filtering unreasonable windows. moreover, when there was no constraint, the maximum number of candidate windows generated by selective search algorithm under different pixel sizes was larger than that generated by selective search algorithm after filtering unreasonable windows. studies showed that selective search algorithm could generate about 2000 candidate windows on average in natural images (fan et al., 2017). in this experiment, the adopted remote sensing images could generate 3000 candidate windows on average. by comparison, selective search algorithm could generate more candidate windows in remote sensing image. table 1 shows the running time of selective search algorithm in figure 3. calculation process of lbp code 109optimal allocation algorithm of geological and ecological high-resolution remote sensing monitoring sampling points remote sensing images of different sizes. as can be observed from table 1, with the increase of pixel of remote sensing image, the running time of the algorithm had been increasing. when the pixel size was 1500*1500, it took 30s to get a candidate window of image in remote sensing image, which indicated that this algorithm did not occupy an advantage in remote sensing pixel size. figure 4. the average number of candidates extracted by selective algorithm for remote sensing images of different scales table 1. the average running time of selective algorithm for candidate extraction of remote sensing images in different scales remote sensing image scale / pixel 500*500 800*800 1000*1000 1500*1500 average processing time / s 2.7 7.6 14.3 29.8 comparison and analysis of the effects of bing algorithm and ibing algorithm figure 5 is the candidate effect diagram of extraction on high-resolution remote sensing images by bing algorithm and ibing algorithm. as can be concluded from figure 5, when the number of candidate windows was the same, the dr/mabo value of ibing algorithm was 97.37%, while the dr/ mabo value of bing algorithm was 95.13%. therefore, the extraction effect of ibing algorithm was better than that of bing algorithm. similarly, the mabo value of ibing algorithm was 65.36%, while the dr/mabo value of bing algorithm was 63.21%, indicating that the quality of candidate boxes extracted from remote sensing images by the improved algorithm was higher. figure 6 shows the extraction effect diagram when the number of candidate areas is 1000~10000. it can be observed that as the number of candidate regions increased, both the dr and mabo indicators tended to rise and then stabilize. when the number of candidate areas was less than 5000, the index was rising, and when the number was greater than 5000, it did not change. this indicated that the number of targets in large-format remote sensing images was relatively large, and the range of possible positions of targets increased significantly, so more candidate windows were needed to achieve a certain detection rate. comparison and analysis of average detection time between bing algorithm and ibing algorithm figure 7 shows the average detection time of bing algorithm and ibng algorithm on different data sets. as can be observed from the figure, under the bing algorithm, the detection time increased with the increase of size, but the maximum time did not exceed 1s. however, the increase of ibing algorithm was obvious, which increased from 0.18s to 0.64s. this showed that although ibing algorithm could greatly improve the extraction speed of remote sensing, the detection time of each image was increased, and such detection time was acceptable. figure 7. average detection time of bing algorithm and ibng algorithm on different data sets figure 5. candidate renderings extracted from high-resolution remote sensing images by bing algorithm and ibing algorithm figure 6. extraction effect in 1000~10000 candidate areas 110 taifu bi conclusion in this research, the algorithm was optimized for high-resolution remote sensing detection sampling points under geological ecology. the three algorithms involved in traditional remote sensing technology are selective search algorithm, svm algorithm and bing algorithm respectively. in this paper, the bing algorithm was improved to obtain the ibing algorithm. according to different data sets, the selective search algorithm in remote sensing image was verified for the applicability of target extraction of monitoring sampling points, and the improved bing algorithm was also verified. the results showed that selective search algorithm could generate more candidate windows in remote sensing image and had better adaptability. the improved algorithm had higher quality of candidate boxes extracted from remote sensing images. although the ibing algorithm could greatly improve the extraction speed of remote sensing, the detection time for each image became larger. and such detection time was also acceptable. the algorithm optimization of high-resolution remote sensing monitoring sampling points in geological ecology was carried out, and the superiority of the algorithm was verified through experiments, which promoted the application of high-resolution remote sensing technology in geological ecology. reference ellis, c. j., & coppins, b. j. (2017). taxonomic survey compared to ecological sampling: are the results consistent for woodland epiphytes. lichenologist, 49(2), 141-155. fan, l. m., ma, x., li, y., li, c., yao, c., xiang, m., wu, b., & peng, j. (2017). geological disasters and control technology in high intensity mining area of western china. journal of china coal society, 42(2), 276-285. gao, b., lu, a., pan, y., huo, l., yunbing, g., li, x., li, s., & chen, z. (2017). additional sampling layout optimization method for environmental quality grade classifications of farmland soil. ieee journal of selected topics in applied earth observations & remote sensing, 7(99), 1-9. gorji, t., sertel, e., & tanik, a. (2017). monitoring soil salinity via remote sensing technology under data scarce conditions: a case study from turkey. ecological indicators, 74:384-391. lengyel, s., kosztyi, b., schmeller, d. s., henry, p. y., kotarac, m., lin, y. p., & henle, k. evaluating and benchmarking biodiversity monitoring: metadata-based indicators for sampling design, sampling effort and data analysis. ecological indicators, 85, 624-633. silva, a. m., abessa, d. p., zaitune, p. a., bohrer-morel, m. b. (2017). ecological risk assessment of a subtropical river influenced by discharges of residues from water and sewage treatment plants. management of environmental quality an international journal, 28(2), 156-174. sürme, y., bişgin, a. t., uçan, m., & narin, i. (2018). cloud point extraction and flame atomic absorption spectrometric determination of cd(ii) in industrial and environmental samples. journal of analytical chemistry, 73(2), 140-144. yavad, k., & congalton, r. (2017). issues with large area thematic accuracy assessment for mapping cropland extent: a tale of three continents. remote sensing, 10(2), 53-57. palabras clave: grava; forma de gránulo; módulo de corte; relación de amortiguación. how to cite item cui, k., & sheng, h. (2019). effects of granular shape on shear modulus and damping ratio of gravel. earth sciences research journal, 23(1), 87-91. doi: https://doi.org/10.15446/esrj. v23n1.78727 the effects of the consolidation ratio, effective confining pressure, gravel content, and granule breakage on the shear modulus and damping ratio of gravel have been extensively researched in recent years. however, studies on the effect of the granular shape are rare. thus, under different confining pressures, dynamic triaxial tests were performed on gravel specimens to investigate the effect of granular shape on the shear modulus and damping ratio of gravel specimens by using a multifunctional triaxial testing instrument. the samples consisted of two kinds of gravel with the same grain composition and relative density of 45%. the test results indicate that, when the confining pressure and shear strain amplitude exceed 300 kpa and 7×10-4, respectively, gravel with a round granular shape has a higher shear modulus compared to an angular shape. conversely, when the shear strain amplitude exceeds 2×10-4, the damping ratio of angular gravel exceeds that of round granules. abstract keywords: gravel; granule shape; shear modulus; damping ratio effects of granular shape on shear modulus and damping ratio of gravel efectos de la forma granular sobre el módulo de corte y la relación de amortiguamiento de la grava issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n1.78727 los efectos de la relación de consolidación, la presión de confinamiento efectiva, el contenido de grava y la rotura de gránulos en el módulo de corte y la relación de amortiguación de la grava se han investigado ampliamente en los últimos años. sin embargo, los estudios sobre el efecto de la forma granular son escasos. por lo tanto, bajo diferentes presiones de confinamiento, se realizaron pruebas triaxiales dinámicas en muestras de grava para investigar el efecto de la forma granular en el módulo de corte y la relación de amortiguamiento de las muestras de grava utilizando un instrumento de prueba triaxial multifuncional. las muestras consistieron en dos tipos de grava con la misma composición de grano y una densidad relativa del 45%. los resultados de la prueba indican que, cuando la presión de confinamiento y la amplitud de tensión de corte superan los 300 kpa y 7×10-4, respectivamente, la grava con una forma granular redonda tiene un módulo de corte más alto en comparación con una forma angular. a la inversa, cuando la amplitud de la tensión de cizallamiento excede de 2×10-4, la relación de amortiguamiento de la grava angular excede la de los gránulos redondos. resumen record manuscript received: 24/07/2018 accepted for publication: 01/12/2018 earth sciences research journal earth sci. res. j. vol. 23, no. 1 (march, 2019): 87-91 kai cui1*, hang sheng2 1key laboratory of high-speed railway engineering of the ministry of education, southwest jiaotong university, chengdu, 610031, china. * corresponding author: cuikai@swjtu.cn g eo te c h n ic a l en g in ee r in g 88 kai cui, hang sheng introduction the shear modulus and damping ratio, which can describe the dynamic characteristics of gravel, must be used in the equivalent linearisation method in seismic response analysis, the evaluation of interaction of soil-structure systems, and the stability analysis of sites. therefore, there is an urgent need to study the important parameters of the dynamic shear modulus and damping ratio of the gravel in order to analyse the seismic response of gravel. many factors can affect these two parameters in all types of soil. the main factors are the soil type (rollins, evans & diegl, 1998), confining pressure (wang, geng & jin, 2005; ling, fu & cai, 2009; wang, jia & deng, 2006; wan & yang, 2013), loading time (sas, gabryś & szymański, 2015), consolidation ratio (liu, yang & chen, 2010; sun, gong & tao, 2013; sun, yuan & gong, 2011), and initial stress state (wang, luo & guo, 2011; wang, luo & guo, 2012), among other factors (ling, zhang, li, an & wang, 2015; delfosse-ribay, djeran-maigre, cabrillac & guovenot, 2004). some results on the effect of effective confining pressure, consolidation ratio, gravel content, and granule breakage, among other factors, on the dynamic shear modulus and damping ratio of gravel have been reported. wang et al. (2005), ling et al. (2009), and wang et al. (2006) studied the influences of confining pressure on these two parameters of gravel. liu et al. (2010) studied, analysed, and proposed the relationship of the dynamic shear modulus and damping ratios with various factors (consolidation ratio, porosity, test grading, etc.) and a comprehensive formula which takes into consideration the consolidation stress ratio, void ratio, and test grading curve, and can be used to calculate the maximum dynamic shear modulus. rollins et al. (1998) found that, with the increase of gravel content from 0% to 60%, the value of gmax can be as high as 38%. hardin and kalinski (2005) found that the gmax value of gravel samples clearly increases through large-scale resonant column tests (specimen height of 300 mm or 150 mm) on different gradations of sand and sand-gravel mixtures. based on the resonant column tests of soil samples with different grain-size distributions, wichtmann and triantafyllidis (2009) pointed out that, for a constant void ratio, the values of gmax do not change with the variation in the mean grain size d50. sahaphol and miura (2005) used bender element tests to study the influence of the particle breakage characteristic of volcanic soils on the maximum shear modulus. however, studies on the influence of the granular shape on the macro-physical and dynamic properties are rare. the effect of the granular shape on the shear modulus and damping ratio of gravel soil cannot be ignored, because the granular shapes and occlusal area are different among the soil particles. qian, gray and woods (1993) found that, under the same conditions (with an effective confining pressure less than 100 kpa), the maximum shear modulus of unsaturated angular sand is 5%−25% greater than that of round sand. meidani, shafiee and habibagahi (2008) found that, under the same conditions, the shear modulus and damping ratio of round gravel are greater than those of angular gravel. the granule shape has a close relationship with the parent soil composition, weathering, transportation, and deposition of the environmental conditions. tong and wang tong, wang and yu-hsing (2014) showed that the shear modulus and damping ratio can also be increased and decreased, respectively, by increasing the particle aspect ratio ar (i.e., by decreasing sphericity). however, the quantitative relation could be applied to a sand but not sure about gravel soils. some scholars attempted to use numerical techniques to simulate the effect of granule shape on the mechanical properties (liu, xie, ma & chang, 2011). these studies are also limited to a general description of the granule shape, such as “granules with a much more regular round shape” and “granules with an irregular shape.” a model or formula that calculates the dynamic properties (such as shear modulus and damping ratio) while considering the granule shapes has not been reported to the authors’ knowledge. the reason for the current situation is that a standard and simple means to effectively quantify the granule shape is lacking. many scholars noted the importance of granule shape and performed a series of studies and proposed many types of quantitative methods, such as the fourier description method and method of fractal theory. however, the quantitative description of granule shape is still not mature because many aspects continue to present difficulties: (1) the descriptions of individual granules; (2) the characteristic descriptions of the distributions of group shapes; (3) the descriptions of group shapes. this research focuses on examining the differences between the properties conferred on the mechanical characteristics of soil by angular gravel and round gravel. further, a conventional optical microscope is used to obtain digital images of the granules, which enable us to analyse the test results and discuss the effect of shear characteristics (shear modulus and damping ratio), compression characteristics, and other mechanical indices of the granule shape by performing conventional mechanical tests. dynamic triaxial test test apparatus the dynamic triaxial test was conducted in the laboratory with dyntts soil multifunctional triaxial testing instruments manufactured by gds company in the uk. as shown in figure 1, the instrument is mainly composed of the host, the axial-force measuring system, the axial-deformation measuring system, the axial control system, the confining-pressure control system, the back-pressure control system, the pore-water-pressure measuring system, and the computer system. the systems for measuring the axial-force and axialdeformation, together with the axial control system, are installed in the control cabinet; the host consists of the triaxial cell and the axial loading system; the pressure control system and the back-pressure control system are connected to two volume controllers, respectively. the greatest advantage of this instrument is the loading and pressure control system. by adjusting the pressure of the cap, the soil can be pressed. this system configuration enables all procedures to be controlled. while collecting and processing the test data, test results can be simultaneously displayed on the monitor. for testing dynamic loads, a load frequency can be selected in the range of 0.01–5 hz with a sine wave, triangular wave, rectangular wave, or any user-defined waveform. test principle under cyclic loading, the stress-strain relationship of the soil has distinctly nonlinear characteristics and shows hysteresis. the hyperbolic model, in which the loading and unloading curves are obtained from the masing rule, is widely used to describe the nonlinear characteristics of soil. in the equivalent linear dynamic viscoelastic model, the equivalent dynamic shear modulus and equivalent damping ratio can describe the nonlinear dynamic characteristics of the soil. in the dynamic stress-strain curve shown in figure 2, the shear modulus is defined as the slope of a secant line that connects the extreme points on a hysteresis loop. the damping ratio can be computed using the equation d a l a t= ( ) ( ) / 4  (1) where a(l) is the area enclosed by the hysteresis loop and a(t) is the area of the shaded triangle shown in figure 2 tested materials and sample preparation in the present paper, angular gravel and round gravel are used as the research subjects. they have low dry density and low degree of saturation. their figure 1. dynamic triaxial apparatus 89effects of granular shape on shear modulus and damping ratio of gravel uniformity and curvature coefficients are 2.09 and 1.04, respectively. gravel samples are prepared using angular gravel and round gravel, respectively, according to figure 3, which shows the grain-size distribution curve. the cylindrical reconstituted specimens are sand samples 39.1 mm in diameter and 80.0 mm in height. first, the samples are subjected to the combined saturation method using the hydraulic and back-pressure, where the saturation of samples can be greater than 98%. second, after the consolidation of the samples has been completed, cyclic loading should be applied to the samples. the loading waveform is a sine wave, and the loading frequency is 0.2 hz. the unidirectional vibration test is carried out under different amplitudes of stress. in order to avoid the influence of vibration effects on the test results, the vibration times should be minimal, and vibration times should generally be 3-5 cycles. testing program in order to study the effect of the granular shape on the dynamic shear modulus and damping ratio of gravel, the grain-size distribution and consolidation condition of angular and round samples are maintained at constant levels throughout the test. by testing the dry density of the angular and round samples under different relative densities, it is shown that, when the relative density is 45%, the difference of dry density between them is less than 3%. this implies that we can ignore the effect of dry density on the test results. thus, we maintain the relative density of all samples as 45% during reconstitution. p5 is defined as the content of gravel having a particle diameter greater than 5 mm. the specific testing program is listed in table 1. table 1. testing program gravel shape p5/% effective confining pressure/kpa dry density/g•cm3 angular sample 36 100,400,600,800,900 1.84 round sample 36 100,400,600,800,900 1.83 angular sample 44 100,400,500 1.89 angular sample 61 100,400,500 1.83 round sample 46 100,400,500 1.87 round sample 61 100,400,500 1.84 digital images of granules the digital images of the granules were acquired using a conventional optical microscope and are shown in figure 4 (grains approximately 1 mm in diameter). the ground shapes of the grains are shown in figure 4(a), and the angular shapes of the grains are shown in figure 4(b). because the angular shapes of the grains differ greatly, they are shown in figure 5 (angular grains approximately 2 and 5 mm in diameter). figure 4. digital images of grains of approximately 1 mm in diameter: (a) round gravel and (b) angular gravel figure 5. digital images of angular grains with an approximate diameter of: (a) 2 mm and (b) 5 mm results and discussions test results the loading equipment controlled by a servo motor may lose testing data at the moments of initial and final loading. when processing the test results, we need to select the 2nd, 3rd, and 4th circular data of each loading stage for the calculation in order to avoid this problem. the test results can be obtained by calculating the average dynamic shear modulus or damping ratio. in accordance with the difference of p5 values, some results can be obtained, such as the dynamic shear modulus gd and the damping ratio λ corresponding to different dynamic shear stress amplitudes γd. when the effective confining pressure is not greater than 300 kpa, under the same conditions and the shear stress amplitude value γd > 7×10 -4, the round sample has a greater dynamic shear modulus. while with the shear stress figure 2. stress−strain curve figure 3. grain-size distribution curve 90 kai cui, hang sheng amplitude value of γd ≤ 7×10 -4, an angular sample has a greater dynamic shear modulus. when the effective confining pressure is greater than 300 kpa, under the same conditions, the round sample has a higher dynamic shear modulus than the angular one. this is mainly because, under larger confining pressure and shear stress amplitude, the local broken angularity of the angular sample causes the occlusal force to decrease among the granules of the gravel, or local shear dilation phenomenon causes the dynamic shear modulus to decrease. when the shear stress amplitude value γd ≤ 2×10 -4, under the same conditions, the damping ratio of the round sample is approximately equal to that of the angular one. when the shear stress amplitude value γd > 2×10 -4, under the same conditions, the round sample has a higher damping ratio than the angular one. the reason is that, when the external profile of granules is more irregular, the occlusal points among the granules are more in number, whereas the contact area of a single occlusal point is less. an increase in the shear stress amplitude would be expected to adversely affect relative sliding or rolling among the granules of the angular sample. therefore, more energy is required to overcome the occlusal force and finishes the reconstitution of its structure. thus, under the same conditions, the damping ratio of the angular sample is higher than that of the round one. the effect of granule shape on the dynamic shear modulus in order to investigate the effect of granule shape on the dynamic shear modulus, the effective confining pressure is normalised as follows: g g pd d o a= ( )/ / . 0 5 (2) where gd is the normalised dynamic shear modulus, d is the dynamic shear modulus, σ’0 is the initial effective consolidation confining pressure, and pa = 100 kpa is the atmospheric pressure. the calculation results of the normalised shear modulus corresponding to p5 and the shear stress amplitude value 1×10 -4, 1×10-3, and 1×10-2 are listed in table 2. this table indicates that, when the shear stress amplitude is 1×10-4, the angular sample has a greater dynamic shear modulus. the increase in the shear stress amplitude results in a large amount of relative rotation among the granules of the round sample, and much more obvious breakage is observed among the granules of the angular sample, which can lead to a significant decrease of the dynamic shear modulus of the angular sample. when the shear stress amplitude reaches 1×10-2, the large deformation will play a leading role in determining the soil properties but not the granule size and shape. in addition, the difference between the dynamic shear modulus of the round sample and the angular sample is small, but with the increase of gravel content, their dynamic shear modulus will increase. by using the hyperbolic curve model, the relationship between the dynamic shear modulus gd and the shear stress amplitude γd can be expressed as follows: g m nd d= +( )1/  (3) when γd0, gdmax=1/m. thus, the maximum dynamic shear modulus gdmax of all samples can be calculated. the increase in the effective confining pressure led to an increase in the maximum dynamic shear modulus of the gravel. however, when the shape and p5 of the gravel are different, the growth rate is also different. when p5 = 37% or 45%, with the increase in the effective confining pressure, the growth rate of the maximum dynamic shear modulus of the round sample is greater than that of the angular one. when the effective confining pressure is greater than 300 kpa, under the same effective confining pressure, the maximum dynamic shear modulus of the round sample is 10% to 15% greater than that of the angular one; when p5 is 60% and the effective confining pressure is not greater than 300 kpa, the growth rate of the maximum dynamic shear modulus of the angular sample is faster, and its value is 3% to 5% greater than that for the round one. critical gravel content p5cr of the maximum dynamic shear modulus with the increase in gravel content, the maximum dynamic shear modulus of breccia gravel and round gravel all increase first and then decrease. this indicates that the effect of p5 on the maximum dynamic shear modulus may have a limit. when p5 reaches the critical value, the maximum dynamic shear modulus of the gravel also reaches its critical value. in order to verify this estimate and determine the critical value, the relationship between the maximum dynamic shear modulus and the confining pressure is investigated. when the effective consolidation pressures are 100, 300, and 500 kpa, respectively, the maximum dynamic shear modulus of the round sample and angular sample showed a significant “parabolic” trend with varying p5. this indicates that a critical value exists under different values of p5, and the value is called p5cr, which is the critical gravel content of the maximum dynamic shear modulus. when p5 is less than the critical value, the maximum dynamic shear modulus of the gravel will increase with increasing p5. when p5 is greater than the critical value, the maximum shear modulus of gravel soil will decrease with increasing p5. in order to investigate the effect of the granule shape on p5cr, the effective confining pressure is normalized as follows: g g pd d o amax max . / /=       ⋅  0 5 (4) where dmax is the normalised maximum dynamic shear modulus, gdmax is the maximum dynamic shear modulus, σ’0 is the initial effective consolidation confining pressure, and pa = 100 kpa is the atmospheric pressure. the relationships between the normalised maximum dynamic shear modulus and p5 have been derived. according to the test results, the parabolic fitting effect is appropriate. when the relative density is 45% and the effective confining pressure is not greater than 50kpa, the value of p5cr of the angular and round samples is approximately 51.2% and 48.5%, respectively. conclusions dynamic triaxial tests on round and angular samples of gravel were conducted to study the effects of the granular shape on the dynamic shear modulus and damping ratio at a relative density of 45%. the main conclusions are as follows: the effects of the granular shape on the dynamic shear modulus are related to the effective confining pressure and shear stress amplitude. when the effective confining pressure is greater than 300 kpa or γd > 7×10 -4, the round sample has a higher dynamic shear modulus than the angular sample. however, when the effective confining pressure is less than or equal to 300 kpa or γd ≤ 7×10-4, the angular sample has a higher dynamic shear modulus than the round one. a critical value of p5 exists for which the dynamic shear modulus of the gravel is maximised. under the same conditions, the p5cr of the angular sample is greater than that of the round sample. when γd > 2×10 -4, under the same conditions, the angular sample has a greater damping ratio than the round one. acknowledgements this work was supported by the national natural science foundation of china (grant no. 41572245), sichuan science and technology program (grant no. 2019jdjq0018) and supported by the fundamental research funds for the central universities (grant no. 2682018gj02). table 2. relationship between p5 and normalised shear modulus under specific shear stress amplitudes (gd /σ ’ 0 /pa) 0.5 /mpa p5=37% p5=45% p5=60% angular sample γd=1×10 -4 61 71 63 round sample γd=1×10 -4 60 64 59 angular sample γd=1×10 -3 40 45 41 round sample γd=1×10 -3 42 51 44 angular sample γd=1×10 -2 9 12 17 round sample γd=1×10 -2 10 18 16 91effects of granular shape on shear modulus and damping ratio of gravel references delfosse-ribay, e., djeran-maigre, i., cabrillac, r., & gouvenot, d. (2004). shear modulus and damping ratio of grouted sand. soil dynamics and earthquake engineering, 24, 461-471. hardin, b.o., & kalinski, m.e. (2005). estimating the shear modulus of gravelly soils. journal of geotechnical and geoenvironmental engineering, 131, 867-875. ling, h., fu, h., & cai, z.y. (2009). experimental study on dynamic deformation behaviors of dam materials. chinese journal of geotechnical engineering, 31, 1920-1924. ling, x. z., zhang, f., li, q. l., an, l. s., & wang, j. h. (2015). dynamic shear modulus and damping ratio of frozen compacted sand subjected to freeze–thaw cycle under multi-stage cyclic loading. soil dynamics and earthquake engineering, 76, 111-121. liu, d., xie, t. t., ma, g., & chang, x. l. (2011). numerical simulation of true triaxial test for behavior of rockfill based on grain shape. water resources and power, 29, 68-72. liu, h. l., yang, g., chen, y. m. (2010). experimental study of factors influencing dynamic shear modulus and damping ratio of dam inverted filler. rock and soil mechanics, 31, 2030-2034. meidani, m., shafiee, a., & habibagahi, g. (2008). granule shape effect on the shear modulus and damping ratio of mixed gravel and clay. iranian journal of science and technology, transaction b: engineering, 32 (b5), 501-518. qian, x., gray, d., & woods, r. (1993). voids and granulometry effects on shear modulus of unsaturated sands. journal of geotechnical and geoenvironmental engineering, 119, 295-314. rollins, k. m., evans, m. d., & diehl, n. b. (1998). shear modulus and damping relationships for gravels. journal of geotechnical and geoenvironmental engineering, 124, 396-405. sahaphol, t., & miura, s. (2005). shear moduli of volcanic soils. soil dynamics and earthquake engineering, 25, 157-165. sas, w., gabryś, k., & szymański, a. (2015). effect of time on dynamic shear modulus of selected cohesive soil of one section of express way no. s2 in warsaw. acta geophysica, 63, 398-413. sun, j., gong, m. s., & tao, x. x. (2013). dynamic shear modulus of undisturbed soil under different consolidation ratios and its effects on surface ground motion. earthquake engineering and engineering vibration, 12, 561-568. sun. j., yuan, x. m., & gong, m. s. (2011). the effect of nonlinear dynamic shear modulus under anisotropic consolidation on response spectrum of soil layer. advanced materials research, 243, 2250-2253. tong, l., & wang, y. hsing. (2015). dem simulations of shear modulus and damping ratio of sand with emphasis on the effects of particle number, particle shape, and aging. acta geotechnica, 10, 117-130. wang, a. p., geng, y., & jin, j. (2005). study on dynamic characteristics of rock filling materials of zhangfeng reservoir. journal of north china institute of water conservancy and hydroelectric power, 26, 16-18. wang, r. h., jia, b., & deng, a. f. (2006). dynamic triaxial testing study on dynamic characteristics of sandy pebble soil. chinese journal of rock mechanics and engineering, 25(supp.2), 4059-4064. wang, z. j., luo, y. s., guo, h., & tian, h. (2012). effects of initial deviatoric stress ratios on dynamic shear modulus and damping ratio of undisturbed loess in china. engineering geology, 143, 43-50. wang, z. j., luo, y. s., & guo, h. (2011). effects of complex initial stress state parameters on dynamic shear modulus of loess. advanced materials research, 243, 2601-2606. wan, x. l., & yang, d. l. experimental study on dynamic shear modulus and damping ratio of soft soils in binhu new district of hefei. applied mechanics and materials, 405, 1957-1960. wichtmann, t., & triantafyllidis, t. (2009). influence of the grain-size distribution curve of quartz sand on the small strain shear modulus gmax. journal of geotechnical and geoenvironmental engineering, 135, 1404-1418. keywords: 226ra, rainbow trout, pylon ab6a, lr-115, mining. palabras clave: radio 226; trucha arcoíris; pylon ab6-a; lr-115; minería. how to cite item viloria avila, t. j., pesantez, a., vazquez silva, e., delgado, e. m., duque, p., & sajo-bohus, l. (2020). levels of radium-226 in the rainbow trout, macroinvertebrates-substrates and water adjacent to the mining concession loma larga, azuayecuador. earth sciences research journal, 24(1), 29-34. doi: https://doi.org/10.15446/esrj. v24n1.79728 mineral processing has been identified among the twelve main industrial processes that must be under control due to the expose to natural radioactive sources. the modification and, generally, the increase in the concentration of naturally occurring radioactive materials (norm) in the earth's crust cause an imbalance in the ecosystem. imbalance that affects the fauna, which are able to bio accumulate these radioisotopes and introduce them into the trophic chain. the main purpose of this work was to establish a radio-ecological baseline. for this purpose, the concentration levels of 226ra in rainbow trout, in macro invertebrate matrices-substrates and water from the irquis and portete rivers were evaluated. a study area adjacent to the loma larga mining concession was selected. the measurements were made using the lucas cells coupled to a pylon ab6a counter and the "can technique", using lr-115 passive detectors. the results indicated that rainbow trout bio accumulated in bones and organsa mean of 94.7% of the total 226ra detected, which values varied from 4.69 to 2.46 bq/kg, while macro invertebrate-substrate matrices showed a concentration that ranged between 0.7 and 1.3 bq/kg in the irquis river, while in the portete river was between 1.1 and 19.3 bq/kg. the water samples, in the sampling points of the irquis river, showed concentrations that ranged between 37.3 and 119.1 mbq/l and in the portete river between 43.5 to 146.4 mbq/l. abstract levels of radium-226 in the rainbow trout, macroinvertebrates-substrates and water adjacent to the mining concession loma larga, azuay-ecuador niveles de radio 226 en la trucha arcoíris, macroinvertebrados-sustratos y agua adyacente a la concesión minera loma larga, azuay, ecuador issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.79728 el procesamiento de minerales se ha identificado entre los doce principales procesos industriales que deben estar bajo un control a la exposición de fuentes radiactivas naturales. la modificación y, por lo general, el aumento de la concentración de norm en la corteza terrestre causa un desequilibrio en el ecosistema. desequilibrio que afecta a la fauna, la cual es capaz de bioacumular estos radioisótopos e introducirlos en la cadena trófica. la finalidad principal de este trabajo fue establecer una línea base radio-ecológica; para ello se evaluaron los niveles de concentración de 226ra en la trucha arcoíris, en matrices macroinvertebrados-sustratos y el agua de los ríos irquis y portete, ubicados en la provincia del azuay, ecuador. se seleccionó una zona de estudio adyacente a la concesión minera loma larga. las mediciones se realizaron haciendo uso de las celdas lucas acopladas a un contador pylon ab6-a y la “can technique”, utilizando detectores pasivos lr-115. los resultados indicaron que la trucha arcoíris bioacumuló el 95.2 % del total del 226ramdetectado, en los huesos y órganos, las matrices macroinvertebrados-sustrato presentaron una concentración que osciló entre 0.7 y 1.3 bq/kg en el río irquis, mientras que en el río portete resultó entre 1.1 y 19.3 bq/kg. las muestras de agua en los puntos de muestreo del río irquis, reflejaron una concentración variable entre 37.3 y 119.1 mbq/l y en el río portete entre 43.5 a 146.4 mbq/l. resumen record manuscript received: 17/05/2019 accepted for publication: 04/10/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 29-34 tony jesús viloria ávila1, adriana pesantez1, efrén vázquez silva1, ernesto manuel delgado1, paola duque1, laszlo sajo-bohus2 1universidad politécnica salesiana, sede cuenca, ecuador 2universidad simón bolívar, caracas, venezuela * corresponding author: tviloria@ups.edu.ec e n v ir o n m e n ta l g e o l o g y 30 tony jesús viloria ávila1, adriana pesantez, efrén vázquez silva, ernesto manuel delgado, paola duque, laszlo sajo-bohus introduction the us environmental protection agency (us-epa) acknowledged in 1987 that the problems related to mining waste can be described as surpassed only by global warming and depletion of stratospheric ozone in terms of ecological risk. the release to the environment of mining waste can cause a profound, and generally irreversible, destruction of ecosystems (durand, 2012). according to the international atomic energy agency (iaea, 2006), any mining operation or activity in which a mineral is involved has the potential to increase the effective dose that a person can receive from natural sources as a result of exposure to naturally occurring radioactive materials (norm) contained in industrial waste. the aquatic systems among others contain norm related to the human activity, and therefore are responsible for a fraction of the dose that the human being receives from the environment. the global average internal dose due to radionuclides other than radon and its decay products is 0.29 msv/y of which 0.17 msv/y comes from 40k, 0.12 msv/y comes from the uranium and thorium series, and 12 μsv/y come from 14c (united nations scientific committee on the effects, 2008). the natural radioactivity of water comes essentially from the 40k and the disintegration series of the 238u and 232th. for groundwater, the main radioactive isotopes found are those mentioned above, and their daughters; within the hazardous elements are found 226ra, 228ra and 222rn, also their progeny. these elements when inside the body move to the bone tissue and concentrate (cox, 1993). the prolonged exposure can cause bone cancers and other bone irregularity; their concentrations vary depending on the geochemical history of the water and the industrial activities of the environment. the bioaccumulation of natural or artificial radionuclides in aquatic organisms is the focus of radio-ecological studies, because they act as an additional source to the radiation dose; of particular importance for the aquatic flora and fauna that form food chains used by the population (stewart et al., 2008). environmental problems associated with norm and solid mineral mines occur during the process of drilling, leaching, handling, storage, transportation of minerals and related equipment (innocent et al., 2013). in ecuador the mining is very active, because the regions have several minerals, such is the case from the southern region of ecuador, known as austro, where we found an important variety of minerals such as gold, silver, copper, lead and zinc (rea-toapanta, 2017). therefore, it is important to carry out baseline studies regarding radioactive elements, such as 226ra, in different media near the mines, in order to be able to calculate the radiological impact, considering that the exploitation mining activity in most cases significantly alters the natural levels of these. as expected, they also increase the concentration levels of the norm in the biota of the rivers. in this study, the concentration levels of 226ra were evaluated in three matrices, the rainbow trout (oncorhynchus mykiss), the macroinvertebrate and the water of the irquis and portete rivers. the rainbow trout is an ichthyic species of the salmonidae family, present in south america, where its habitat is generally located in cold waters of rivers and lakes, tolerating a temperature range between 0 to 30 °c, are a very resistant species, capable of displacing endemic species. the macroinvertebrate are bio indicators of the water quality with biological indexes, giving a significant importance to these matrices. materials and methods the technique used in this study for the quantification of the content of 226ra in the organic matrix (fish) was the can technique making use of solid-state nuclear track detectors (ssntds), specifically the lr-115 detector (mohamed abd-elzaher, 2012). selection of sampling areas this investigation was carried out in the mining project called loma larga which covered an area of 7,960 ha, where three concessions operate, known as cerro casco, cristal and río falso. it is located in the south of ecuador 15 km northwest from the town called girón and 30 km southeast of cuenca city, at an altitude between 3 500 and 3 900 mamsl (jarrín-jurado & morán-reascos, 2017). the area of study was río falso concession with an area of 3 168 ha, located in the parishes baños, victoria del portete, san gerardo, chumblín, san fernando and girón, belonging to the province of azuay (fig. 1) (inv minerales ecuador sa, 2016). the physico-chemical properties of the site analyzed in this study were characterized by presenting strata of andesitic porphytic lavas of very fresh acicular pagioclase and amphibole, which are volcanic or igneous extrusive rocks containing between 52% and 63% silica (sio2), in addition 5% iron (feo) and 7% calcium (cao), on average. in relation to hydrology, the drainage network is a radial dendritic type constituted by a series of streams. drainage to the southeast occurs through two main rivers, the irquis and the portete. these effluents, being also transport vectors of matter of the mining exploitation, have been analyzed in this study to determine the 226ra levels. figure 1. location map of the loma larga mining project, in the province of cuenca, ecuador. coordinates-psad-1956-utm, zone17s, (december 2018). in this work, sampling points for the rainbow trout, macroinvertebrate matrices and water, were selected as established by the world health organization (who, 2006). it considers that the number of sampling points depend on the characteristics of the water quality under analysis. for components whose concentrations tend to vary during distribution, the number of strategic points should be three, distributed along the region of studies (see table 1). in this work these were situated in the headwater of the river, in the middle course of the region of interest (adjacent to the mining activity) and at the exit of the concession (see fig. 2). water samples collection of surface water samples obtaining water samples was carried out by the immersion method, following the protocol established for the pylon ab6 detector (pylon electronics inc., 2015). to take the liquid sample, a 250 ml glass container was used, ensuring the absence of bubbles, which was hermetically sealed and each of the containers identified with the necessary data. 31levels of radium-226 in the rainbow trout, macroinvertebrates-substrates and water adjacent to the mining concession loma larga, azuay-ecuador conc ncpm nae cf eff v a eff g bq lra c s d = ( ) * * * * * * / 2 (1) where, ncpm is the number of counts per minutes, nae is the number of alpha emissions (for 222rn is 3), cf is the conversion factor (for this calculation is 60 dpm to dps), effc is the cell efficiency (0,745 cpm/dpm), vs is the sampling volume (0,19 l), effd is the 222rn extraction efficiency (for the lucas cell 600a is 90 ± 5%) and g is the growth correction factor. the growth correction factor is calculated by the following equation, g e tdelta= − −( )1 1 (2) where, t t t mindelta a1 1 0= − ( ) and t0a is the time in which the sample was initially degassed, t1 is the time in which the sample was degassed for measurement. after each measurement the lucas cells were flushed with nitrogen, at a flow rate of 5 l/min for 3 minutes, in order to prevent contamination and reduce background count buildup that could affect future measurements. the cells were stored for a minimum of 24 hours, before being reused again. biological samples. collection of samples. the macro invertebrates and the substrate in which they live, such as sediments, algae and macrophytes, were collected with surber sampler nets, located at the three selected points of each river. the rainbow trout were collected at the midpoint of the irquis river, adjacent to the mine mouth of the concession denominated río falso by gillnets. preparation of the trout samples each of the individuals was dissected, preparing with them two matrices, one of muscle tissue, and another of bones and organs. organic matrices of both measurement of 226ra activity concentrations in water. to measure the 226ra activity concentrations in the water sample, the 222rn was evacuated from the latter, flowing nitrogen through it for a period of approximately 15 minutes; the samples were stored for a period of 3.8 days, after this time, they were degassed, transferring the gas to a lucas cell model 600-a, according to the pylon ab6 protocol (pylon electronics inc., 2015). the formula used to calculate the ra activity concentrations in water was: figure 2. map of the location of the sampling points in the irquis and portete rivers. coordinates-psad-1956-utm, zone17s (december 2018). kilometers 0 0.45 0.9 1.8 2.7 3.6 table 1. coordinates of the sampling points in the study area where the investigation was conducted. code coordinates psad-1956-utm-zone 17s sampling area river 1i 698141 9664023 irquisheadwater 2i 699137 9663840 quebrada high quinuahuaycu irquisadjacent to entrance of the mine 3i 700018 9662981 quebrada quinuahuaycu irquis-exit of the concession 1p 698448 9662815 porteteheadwater 2p 699249 9661368 quebrada kalluancay portete middlecourse 3p 699916 9660990 quebrada kalluancay portete exit of the concession 32 tony jesús viloria ávila1, adriana pesantez, efrén vázquez silva, ernesto manuel delgado, paola duque, laszlo sajo-bohus macro invertebrates and rainbow trout were dehydrated in an oven at 110 °c for 24 hours. and then crushed to obtain a finer granular consistency for favoring the exhalation rate of 222rn; after this the weighing was carried out. taking into account that the radon exhalation of 222rnfrom the matrices is conditioned by the physical properties of it, among which the porosity is one of the most important, and considering that the porosity is not controlled by grain size, as the volume of between-grain space is related only to the method of grain packing, but the emanation of radon depends on the grain size and the shape, because they determine in part how much radium is near enough to the surface of the matrix grain to allow the newly formed radon to escape into a pore space, we considered appropriated to make a fine matrix to reduce the probability of 222rn to remain embedded in the grain. content measurement of 226 ra in macro invertebrates-substrate matrices. the samples of macro invertebrates and rainbow trout previously processed were placed in pvc cylindrical containers, of 10 cm high and 7 cm diameter in which a passive detector was positioned. in this case, 1.5 cm x 2 cm lr-115 type ii (kodak-pathe, france) detectors were used, located in the internal upper part of the container, in such a way that the sensitive part of the detector is oriented towards the sample, at approximately 4 cm from the sample surface (khan et al., 1992). in order to avoid the registration of alpha particles emitted in the decay of the isotopes of radon, prior the 226ra and the 222rn reached the secular equilibrium within the cylindrical containers, the detectors were covered for four weeks with a metal foil that was held in place with the help of a small magnet on the outside of the chamber. after that time, the metal sheet was released, exposing the detector for the following 60 days (sonkawade et al., 2008). after this time, the detectors were removed and etched, placing each detector in a 2.5 n naoh solution for 90 minutes, at a temperature of 60 °c in a laboratory water bath (zubair et al., 2012). the nuclear tracks density was quantified with the help of an optical microscope with a magnification of 400x. the content measurement of 226ra was calculated by the following equation. c h a k t m bq kgra rn eff = ( )* * * * / (3) where, ρis the radon track density, h is the distance between the surface of the sample and the detector in meters, a is the cross-sectional area of the container in m-2, krn is the calibration factor of lr-115 plastic track detector (0.033 track cm2 d-1/ bq m-3), teff is the effective exposure time, and m is the mass of the sample in kg. the effective exposure time can be calculated using the formula t t eeff t= − −( )−1 1   (4) where, t is the actual exposure time and λ is the decay constant for 222rn. results the 226ra concentrations in the matrices of macro invertebrates-substrate of the irquis river ranged from 0.90 ± 0.07 bq/kg to 1.30 ± 0.08 bq/kg, while in the portete river the minimum concentration was 1.10 ± 0.08 bq/kg and the maximum of 19.30 ± 0.08 bq/kg (fig. 3). because of the difficulty to fish in others sampling points along the rivers, only two sampling were done at only one sampling point, the midpoint of the irquis river analysis region (adjacent to the mine mouth); in the first sampling the average 226ra levels in the muscle tissue of the rainbow trout was 0.17 ± 0.04 bq/kg, and the corresponding to the bone-organs matrix was 4.69 ± 0.51 bq/kg. while in the second sampling the muscle tissue concentration was 0.19 ± 0.03 bq/kg, and the bone-organs sample showed a 226ra concentration of 2.46 ± 0.31 bq/kg. all values correspond to dry weight (see fig. 4). it is worth to noting that, the difference between the 226ra concentration in the boneorgans matrices in the first and second sampling was 47.74 %. the original purpose of the two samplings was to obtain a better statistic, but that noticeably greater difference between the values of the 226ra concentrations in the bone-organs matrices, makes a comment on it worthwhile, even though it is not part of the main objective of this work. the bioaccumulation factor depends on different factors, such as the duration of exposure, the rate of metabolism, the age and the diet of the species, among others. 0 1 2 3 4 5 bones-organs muscle tissue total c on ce nt ra tio n of r ad iu m -2 26 type of sample concentration of ra-226 (bq/kg) first sampling concentration of ra-226 (bq/kg) second sampling figure 4. concentration of 226ra in the rainbow trout of the irquis river in the area adjacent to the mine entrance. the variation of the radium levels in both rivers, have a similar behavior, they increase as the measurements are done downstream. the levels of radium in the headwater of the irquis river and the portete river were 39.3 and 57.5 mbq/l respectively, while in the lowest points downstream were 119.1 and 146.39 mbq/l for the irquis and portete river respectively (see fig. 5). discussion the iaea (2014) reports that in uncontaminated surface waters the concentration range of radioisotope 226ra in fish (whole body) ranges between 0.68 and 1.12 bq /kg (fresh weight). swanson s. m. (1983) reports concentrations of 226ra in fish, among them trout, in uncontaminated surface water, with levels between 0.296 and 3.400 bq/kg in dry weight, values similar to those obtained in this investigation, in which was obtained an average concentration for rainbow trout of 3.75 bq/kg (dry weight). the concentration levels of 226ra in the bone-organ matrices were higher than those in the muscle tissue matrices, this is due to the fact that both calcium and radium belong to the same group in the periodic table, so both elements have very similar chemical properties, and that is why in human and animal organisms, radium is assimilated as calcium and deposited in the bones. clulowet al. (1998), in their study in lakes adjacent to non-industrialized zones, found levels of 226ra in the bones of lake trout (salvelinus namaycush), 0 5 10 15 20 1 2 3 mean c on ce nt ra tio n of 2 26 r a 1n b q/ kg sampling points radium activity concentration (cra) (bq/kg) irquis river radium activity concentration (cra) (bq/kg) portete river figure 3. concentration of 226ra in macro invertebrates of the irquis and portete rivers (1 corresponds to the source of the two rivers, 2 represents the midpoint of both the irquis and portete rivers, and 3 are the sampling points at the exit of the concession) 33levels of radium-226 in the rainbow trout, macroinvertebrates-substrates and water adjacent to the mining concession loma larga, azuay-ecuador with average values less than 20 bq/kg, as well as skipperud, et al. (2013) in their work claim that the concentration of radionuclides bio accumulates in large quantities in bones rather than in gills, more than in kidneys, and more than in the liver, while in smaller amounts it is in the muscle, agreeing satisfactorily with the result of this study, which was that of the total 226ra measured, 95.21% was accumulated in bones and organs, and only 4.79% in muscle tissue. the results reported by siddeeg, et al., (2014), and sethy, et al., (2011), on 226ra concentrations in sediments of rivers that are considered affected by mining activity, are in a range between 15. 0 and 179.0 bq/kg, while in the works of sethy, et al., (2011) and rowan, et al., (2013) in rivers not affected by mining activity, concentrations of 226ra in the sediments are between 1.99 bq/kg and 4.50 bq/kg; these last values are comparable with those obtained in this work, which were between 0.90 ± 0.07 bq/kg and 1.30 ± 0.08 bq/kg for the irquis river, and 1.10 ± 0.08 bq/kg and 19.30 ± 0.08 bq/kg for the portete river. the levels of dissolved 226ra activity in river waters are related to the composition of the soil and the physical and chemical properties of the waters. siddeeg (2013) asserts that the physical properties of the bedrock along the river bed play an essential role in the contribution of the concentration of 238u, 232th and 226ra in natural waters. the geochemical processes of 238u, 232th and 226ra, such as the adsorption by clay minerals, insoluble oxides such as iron oxide, play a key role in the understanding of the behavior of these radioisotopes and in the analysis of their concentration levels present in both water and in the soil. on the other hand, porcelli, et al., (2001), affirm that these levels of concentrations vary, when “the streams drain limestone, phosphates and rocks rich in 238u”, factor that influences so that the concentration of this radionuclide is greater. when comparing the results obtained by hameed et al.,(1997) (0.07-314 mbq /l) and de lama & osores (2002) (2.1-175.4 mbq/l) whose works were carried out in areas with lithological similarities to the present work, where predominate the andesitic strata from volcanic (igneous) rocks, which contain relative high percentage of silica, iron oxides and calcium oxides, we found that the values obtained are within the ranges mentioned above; in the portete river the activities of the dissolved 226ra ranged between 43.5mbq/l and 146 mbq/l, while in the irquis river the values were between 37 mbq/l and 119 mbq/l. some authors, such as de oliveira, et al., (1996) and the international atomic energy organization (2014), state that 226ra at low or high ph is more soluble than at medium ph. szabo et al. (2012) analyzed the correlation of the physical-chemical parameters of water with the 226ra content and affirm that this alkaline earth metal has a significant correlation with ph, dissolved oxygen (do) and total dissolved solids (sdt), as well as with some metals. high concentrations of 226ra occur mainly in anoxic waters (low do) with high content of iron, manganese, calcium, magnesium, barium, strontium and eventually with total dissolved solids, potassium, sulfates (so4) and carbonates (hco3), as well as when the water is acidic (low ph) and contains high concentration of nitrates (no3) and aluminum. both the waters of the irquis and portete rivers, which were analyzed in this study, differ completely from the aforementioned characteristics. conclusions the values obtained in this work, concerning to the amount of 226ra measured in trout bone-organ matrices, the trout muscle-tissue matrices, and water of the rivers irquis and poetete, were very similar to values obtained by other authors who measured the same magnitudes in uncontaminated waters. it should be noted that the values obtained in this work are the first of their kind and could serve as a reference database for future works whose main objective will be to measure the radiological impact due to the mining activity in loma larga. it is suggested to expand the sampling area around the loma larga concession, the number of samples, and the sampling period and calculate the bioaccumalation factor (baf). references clulow, f., davé, n., lim, t. & avadhanula, r. (1998). radium-226 in water, sediments, and fish from lakes near the city of elliot lake, ontario, canada. environmental pollution, 99(1), 13-28. cox, j. r. (1993). naturally occurring radioactive materials in the oilfield: changing the norm. tulane law review. de lama, g. & osores, j. (2002). determinación radioquímica de radio-226, radio-228 y plomo-210 en aguas del río rimac. http://dspace.ipen.gob. pe/bitstream/ipen/179/1/pag.%2047-50-%20ict%201998-2001.pdf (last accessed january 2019) de oliveira, j., mazzilli, b., de oliveira-sampa, m. h. & silva, b. (1996). la variación sazonal de 226ra y 222rn en fuentes de agua mineral en aguas de prata-brasil. proteccion radiológica en américa latina y el caribe, proyecto arcal xvii/oiea, 1, pp. 33-37. durand, j. (2012). el impacto de la extracción de oro en el witwatersrand en los ríos y el sistema kárstico de gauteng y la provincia del noroeste, sudáfrica. journal of african earth sciences, 68, 24-43. hameed, p. h., shaheed, k., somasundaram, s. s. n. & riyengar, m. a. (1997). radium-226 levels in the cauvery river ecosystem, india. journal of biosciences, 22(2), 225-231. khan, a. j., prasad, r. &tyagi, r. k. (1992). measurement of radon exhalation rate from some building materials. nuclear track radiation measurement, 20(4), 609-610. (https://doi. org/10.1016/1359-0189(92)90013-l) innocent, a. j., onimisi, m. y. & jonah, s. a. (2013). evaluation of naturally occurring radionuclide materials in soil samples collected from some mining sites in zamfara state, nigeria. british journal of applied science and technology, 3(4), 684-692. international atomic energy agency (iaea). (1990). the environmental behavior of radium. 1 (310), vienna, austria. international atomic energy agency (iaea). (2014). the environmental behaviour of radium: revised edition. (476), vienna, austria. international atomic energy agency (iaea). (2006). assessing the need for radiaton protection measures in work involving minerals and raw materials. viena, austria. inv minerales ecuador s.a. (2016). auditoría ambiental de cumplimiento. concesiones mineras: cerro casco y río falso (proyecto loma larga) periodo: oct. 2014oct. 2016. jarrín-jurado, j. r. & morán-reascos, d. v. (2017). análisis de continuidad de extensión de la mineralización alrededor del sistema de alta sulfuración loma larga. tesis de grado, universidad central del ecuador, quito, ecuador. (http://www.dspace.uce.edu.ec/handle/25000/10566) (last accessed december 2018) mohamed abd-elzaher. (2012). an overview on studying 222rn exhalation rates using passive technique solid-state nuclear track detectors. american journal of applied sciences, 9(10), 1653-1659. organización mundial de la salud (oms) (world health organization, who) (2006). guías para la calidad del agua potable. primer apéndice a la tercera edición. y = 34,306x r² = 0,7569 y = 51,448x 11,473 r² = 0,9861 0 20 40 60 80 100 120 140 160 1 2 3 c on ce nt ra tio n of r ad iu m -2 26 (m b q/ l) sampling points irquis river portete river lineal (irquis river) lineal (portete river) figure 5. average concentrations of 226ra in the irquis and portete rivers in the three strategic points (1 corresponds to the point of the source of the rivers, 2 is the adjacent point of the mine entrance in the irquis river and the middle point of the portete river, and 3 is the point at the exit of the concession in each river respectively) https://en.wikipedia.org/wiki/tulane_law_review http://dspace.ipen.gob.pe/bitstream/ipen/179/1/pag. 47-50ict 1998-2001.pdf http://dspace.ipen.gob.pe/bitstream/ipen/179/1/pag. 47-50ict 1998-2001.pdf https://doi.org/10.1016/1359-0189(92)90013-l https://doi.org/10.1016/1359-0189(92)90013-l http://www.dspace.uce.edu.ec/handle/25000/10566 34 tony jesús viloria ávila1, adriana pesantez, efrén vázquez silva, ernesto manuel delgado, paola duque, laszlo sajo-bohus pylon electronics inc. (2015). http://radoncorp.com/testing/pylon-monitors. php (last accessed december 2018) porcelli, d., andersson, p. s., wasserburg, g. j. & baskaran, m. (2001). transport of uand th-series nuclides in a baltic shield watershed and the baltic sea. geochimica et cosmochimica acta, 65(15), 24392459. rea-toapanta, a. r. (2017). mining policy and environmental sustainability in ecuador. revista digital figempa: investigación y desarrollo, 1(2), 41-52. rowan, d., silke, r., & carr, j. (2013). biota-sediment accumulation factors (bsaf) for radionuclides and sediment associated biota of the ottawa river. international atomic energy agency (iaea). (https://pubs. cnl.ca/doi/abs/10.12943/anr.2013.00013) (last accessed december 2018) salman, d. j. & al-khalifa, i. j. (2013). the determination of radon activity concentration in local and imported fish of basra governorate/ (iraq) by using ssntds technique. journal of basrah researches (sciences), 39(1), 135-142. sethy, n. k., jha, v., shukla, a., sahoo, s., tripathi, r. & puranik, v. (2011). natural radionuclide (u and 226-ra) in water, sediment, fish and plant species in the aquatic environment around uranium mining and ore processing site at jaduguda, india. journal of ecosystem & ecography, 1, 103. siddeeg, s. m. (2013). geochemistry of natural radionuclides in uraniumenriched river catchments. ph.d. thesis, school of chemistry, faculty of engineering and physical science, university of manchester, england. siddeeg, s. m., bryan, n. d. & livens, r. f. (2014). dispersion of u-series radionuclides in stream sediments from edala valley, uk. journal of environmental science: processes & impacts, (5), 991-1000. doi: 10.1039/c3em00609c. skipperud, l., stromman, g., yunusov, m., stegnar, p., uralbekov, b., tillovoev, h., zjazjev, g., heier, l. s., rosseland, b. o. & salbu, b. (2013). environmental impact assessment of radionuclide and metal contamination at the former u sites taboshar and digmai, tajikistan. journal of environmental radioactivity, 123, 50-62. doi: https://doi. org/10.1016/j.jenvrad.2012.05.007 somogyi, g. (1990). methods for measuring radium isotopes: track detection. in iaea. the environmental behavior of radium, 1, 229-256. sonkawade, r. g., kant, k., muralithar, s., kumar, r. & ramola, r. c. (2008). natural radioactivity in common building construction and radiation shielding materials. atmospheric environment, 42, 2254-2259. stewart, g., fowler, s. & fisher, n. (2008). the bioaccumulation of uand th-series radionuclides in marine organisms. radiactivity in the enviroment, 13, 269-305. swanson, s. (1983). level of ra-226, pb-210 and u-total in fish near a saskatchewan uranium mine and mill. health physics, 45(1), 67-80. szabo, z., kraemer, t., de paul, v. & jacobsen, e. (2012). occurrence and geochemistry of radium in water from principal drinking-water aquifer systems of the united states. applied geochemistry, 27(3), 729-752. doi: https://doi.org/10.1016/j.apgeochem.2011.11.002 united nations scientific committee on the effects (unscear) (2008). sources and effects of ionizing radiation. volume ii. https://www. unscear.org/unscear/en/publications.html) (last accessed december 2018) zubair, m., shakir khan, m. & verma, d. (2012). measurement of radium concentration and radon exhalation rates of soil samples collected from some areas of bulandshahr district, uttar pradesh, india using plastic track detectors. iranian journal of radiation research, 10(2), 83-87. http://radoncorp.com/testing/pylon-monitors.php http://radoncorp.com/testing/pylon-monitors.php https://pubs.cnl.ca/doi/abs/10.12943/anr.2013.00013 https://pubs.cnl.ca/doi/abs/10.12943/anr.2013.00013 https://doi.org/10.1016/j.jenvrad.2012.05.007 https://doi.org/10.1016/j.jenvrad.2012.05.007 https://www.sciencedirect.com/science/article/pii/s1569486007000083 https://www.sciencedirect.com/science/article/pii/s1569486007000083 https://doi.org/10.1016/j.apgeochem.2011.11.002 https://www.unscear.org/unscear/en/publications.html https://www.unscear.org/unscear/en/publications.html 97 earth sciences research journal earth sci. res. sj. vol. 17, no. 2 (december, 2013): 97 103 a multivariate statistical approaches on physicochemical characteristics of ground water in and around nagapattinam district, cauvery deltaic region of tamil nadu, india s. venkatramanan1, 2, s. y. chung1*, t. ramkumar2, g. gnanachandrasamy2, s. vasudevan2 1 institute of environmental geosciences, department of earth & environmental sciences, pukyong national university, 599-1 daeyeon-dong nam-gu, busan 608-737, south korea 2department of earth sciences, annamalai university, annamalai nagar 608-502, tamil nadu, india *corresponding author email: chungsy@pknu.ac.kr, venkatramanansenapathi@gmail.com hydrogeology ground water samples collected at different locations in and around the nagapattinam district were analyzed for their physicochemical characteristics. the ground water samples were collected from fifty two dug and deep wells during the monsoon and summer seasons in june and december, 2011. the present investigation is focused on the determination of physico-chemical parameters such as ph, ec, tds, ca, mg, na, k, hco3, so4 and cl. ground water suitability for drinking, domestic and agricultural purposes was examined by using who standards. correlation, factor and cluster analyses were applied to classify the ground water qualities and to categorize the geochemical processes controlling ground water geochemistry. factor analysis indicates that seawater intrusion and agriculture runoff are dominant factors controlling the hydrogeochemistry of ground water in the study area. cluster analysis was helpful for the classification on the basis of contamination characteristics of ground water quality. this study also elucidates that multivariate statistical analyses can be used to improve the understanding of ground water status and assessment of ground water quality. en este estudio se analizan las características fisicoquímicas de muestras de aguas subterráneas tomadas en diferentes locaciones en y alrededor del distrito de nagapattinam. las muestras se recolectaron en 52 pozos cavados y perforaciones profundas durante las subestaciones del monzón y el verano, en los meses de junio y diciembre (2011). la presente investigación está enfocada en la determinación de parametros fisicoquímicos como ph, ec, tds, ca, mg, na, k, hco3, so4 y cl. se examinó la pertinencia de estas aguas para consumo y para irrigación a la luz de los estándares de la organización mundial de la salud. se aplicaron análisis de correlación, factores y cúmulos para clasificar las muestras y categorizar los procesos geoquímicos que controlan las aguas subterráneas. el análisis de factores indica que la intrusión del agua marina y los vertidos de la agricultura son los principales agentes que intervienen la hidroquímica de las aguas subterráneas en el área de estudio. el análisis de cúmulos ayudó a la clasificación de las bases de contaminación características de las aguas subterráneas. este estudio, además, dilucida que los estudios estadísticos multivariantes pueden ser utilizados para mejorar el conocimiento y la evaluación de la calidad de las aguas subterráneas. palabras clave: calidad de aguas subterráneas, análisis estadístico, incursión de agua marina, nagapattinam, costa este de la india. record manuscript received: 03/12/2012 accepted for publication: 01/11/2013 abstract resumen key words: ground water quality, multivariate statistical analyses, ground water suitability, correlation, factor and cluster analyses. introduction one of the most severe problems concerning ground water quality is the rapid salinization of water resources. the increase in ground water salinity, particularly in coastal areas, may be due to the inflow of natural saline water, such as sea water intrusion, dissolution of soluble salts in the unsaturated zone, and saline water naturally derived from adjacent aquifers (vengosh and rosenthal, 1994, vengosh et al., 1999, maslia and prowell, 1990, jones et al., 1999, rosenthal et al., 1992, glynn and plummer 2005, cloutier et al., 2008, brinda and elango, 2011). anthropogenic contamination is another major cause of salinization and water-quality degradation. ground water irrigation causes salts recycling and accumulation in aquifers. the salts are accumulated in the soil and flushed through the unsaturated zone to the aquifer. moreover, irrigation with wastewater, which is generally more saline than regional ground water, increases the salinization rate of shallow ground water. this problem is more conspicuous 98 in arid and semi-arid zones, where potable water is replaced by wastewater for irrigation in order to save the depleting water resources. ground water salinization and degradation is usually observed through the sampling of pumping wells. in these wells, the screens are located in the deeper parts of the aquifer, far below the water table. in the light of this fact, one of the management strategies for ground water resources especially in the coastal belt is to ensure that ground water extraction does not lead to the intrusion of saline seawater into the aquifers. however, due to the relatively low elevations in the area and the lack of strict regulation, ground water salinization continues to be a problem in some of the coastal areas. on the other hand, this resource is the only source of water for agricultural activities in view of the erratic nature of rainfalls in recent times. the high salinity of ground water and the heavy reliance on ground water for irrigation necessitate the overall assessment of ground water quality for irrigation purposes. (hussein, 2004, ramkumar et al. 2011, anithamary et al. 2012, venkatramanan et al. 2012, gnanachandrasamy et al. 2012). thus, the suitability of ground water for irrigation and drinking purposes thus had to determined based on the presence of major ions in the ground water of this coastal region. this study proposes using multivariate statistical analysis to identify a correlation between fresh and saline water incursion. description of the study area the study area is located in east coast of india, and its stretches between coleroon in the north and kodikarai in the south (fig. 1). the district has a coastline stretching of about 190 km. the nagapattinam district water quality parameters in mg/l who standard (1993) monsoon season summer season ph 6.5-8.5 7.4-8.2 7.2-7.8 electrical conductivity (μs/cm) (ec) 846.8-2476.3 987.6-2576.6 total dissolved solidsc(tds) 500-1500 912-1875 846-1973 calcium (ca) 200 32-312 28-322 magnesium (mg) 150 35.4-318 38.5-321 sodium (na) 200 109.3-358.6 118.8-375.1 potassium (k) 14.3-312.4 18.9-324.9 bicarbonate (hco3) 97.3-326.2 107.5-324.9 sulphate(so4) 250 52.7-378.2 58.6-398.6 chloride(cl) 250-600 354.8-821.9 355.4-845.2 table. 1. physico-chemical parameters of ground water in the study area table. 2. correlation coefficients of ground water chemical components in monsoon and summer seasons monsoon ph ec tds ca mg na k hco 3 so 4 cl ph 1.00 ec -0.01 1.00 tds -0.05 0.84 1.00 ca -0.13 0.18 0.26 1.00 mg 0.13 -0.20 -0.28 0.06 1.00 na -0.03 0.74 0.68 0.14 0.17 1.00 k -0.14 0.00 -0.04 -0.10 0.24 -0.15 1.00 hco 3 0.10 -0.15 -0.07 0.13 0.10 0.70 -0.15 1.00 so 4 0.36 0.54 0.64 -0.12 0.45 0.70 -0.27 -0.03 1.00 cl -0.15 0.76 0.84 -0.37 0.16 0.76 0.12 0.11 0.34 1.00 summer ph 1.00 ec -0.12 1.00 tds -0.13 0.94 1.00 ca 0.18 0.17 0.24 1.00 mg 0.13 -0.21 -0.18 0.07 1.00 na 0.29 -0.07 0.84 0.11 0.18 1.00 k -0.27 -0.02 -0.03 -0.11 0.25 -0.13 1.00 hco 3 0.24 -0.14 -0.14 0.13 0.07 0.70 -0.16 1.00 so 4 0.31 0.65 0.72 -0.11 0.45 0.78 -0.27 -0.03 1.00 cl -0.11 0.88 0.95 -0.37 0.12 0.82 0.14 0.06 0.68 1.00 s. venkatramanan, s. y. chung, t. ramkumar, g. gnanachandrasamy, s. vasudevan 99 physico-chemical parameters monsoon season summer season factor 1 factor-2 communality factor 1 factor-2 communality ph 0.193 0.096 0.343 0.140 0.629 0.284 electrical conductivity (μs/cm) (ec) -0.772 -0.083 0.759 -0.854 -0.090 0.900 total dissolved solids (tds) -0.807 -0.018 0.776 -0.840 -0.074 0.902 calcium (ca) -0.479 0.345 0.289 -0.438 0.441 0.293 magnesium (mg) 0.514 0.200 0.525 0.450 0.152 0.464 sodium (na) 0.040 0.879 0.540 0.062 0.780 0.558 potassium (k) 0.105 -0.344 0.406 0.113 -0.441 0.388 bicarbonate (hco3) 0.062 0.878 0.552 0.113 0.771 0.517 sulphate (so4) 0.532 0.092 0.652 0.469 0.140 0.574 chlordie (cl) 0.536 0.054 0.469 0.480 -0.163 0.386 eigenvalues 2.41 1.801 2.417 24.179 total % of variance 24.152 18.011 1.982 19.826 cumulative % of variance 24.152 42.16 24.17 44.00 table. 3. varimax rotated factor loadings in monsoon and summer seasons mainly comprises of the quaternary sediments, which increases towards the southern part of coleroon river. these sediments have been delineated as alluvial plain deposits of the cauvery river and its tributaries, narrow fluvio-marine deltaic plain deposits and marine coastal plain deposits (east coast formation). the fluvial deposits comprise of flood plain, flood basin, point bar, channel bar and palaeo channels with admixtures of sand, silt and clay. the deltaic plain includes palaeo tidal flats with clays, and sands and sand ridges or gray brown sand. the marine coastal plains include beach, tidal flats, salt marsh, mangrove swamps, and deposits of sand and clay. the cretaceous formations of the coastal track of the cauvery basin consist of faunal, rich marine sedimentary rocks, such as limestones, sandstones, clays and sandy beds, etc. the mouth of the river comprises of alluvial deposits including clays and silts. ground water occurs in these formations and is extracted by filter point wells, tube wells, shallow boreholes and infiltration wells, especially from the sandy aquifers. the pliocene and quaternary shallow aquifers consists of sand, gravel and clay. the aquifer is more clayey towards the east and south eastern part of the district, except the coastal stretch of beach sands. the depth to shallow aquifers varies between 3 m and 35 m, whereas the depth to deep aquifers ranges from 80 m to 100 m. the average annual rainfall in the monsoon season is about 451.48 mm, whereas during that of summer, rainfall averages is about 43.12 mm. method of study fifty two ground water samples were collected from di fferent water sources such as dug and bore wells in june and december, 2011, using high-density polythe ne containers. the water samples were analyzed for major physico-chemical parameters adopting standard procedures (apha 1995). ph was measured by a ph pen (eco tester ph) and electrical conductivity (ec), total dissolved solids (tds), ca, na, mg, k, hco3, cl, and so4 were measured using a multi-parameter ion analyzer (hanna, hi83099). all concentrations are expressed in milligrams per litre (mg/l), except ph and ec. multivariate statistical analyses were performed to find out determine the relationship between ground water parameters using statistica software (ver.8). the results were evaluated in accordance with the norms prescribed under ‘who’ (world health organization, 1993) (table 2). results and discussion summary of ground water analysis the water quality analysis of different ground water samples were carried out to determine different physicochemical parameters such as for ph, ec, tds, ca, mg, na, k, hco3, so4, and cl; and the results are presented in table 1 and figure 2. ph value of ground water samples between 7.4-8.2 and 7.2-7.8 during the monsoon and summer seasons, respectively. ground water is a little alkaline, but the quality is within the limiting value of the drinking water standard. the general increase of ph in a sedimentary terrain is related to weathering of plagioclase feldspar in sediments. this is aided by dissolved atmospheric carbon dioxide resulting in the release of sodium and calcium, which progressively increases ph and alkalinity of the ground water. ec varied to 846.8-2476.3 and 987.62576.6 μs/cm, and tds changed to 912-1875 mg/l and 846-1973 mg/l during monsoon and summer seasons, respectively. ec and tds exceed who (1993) permissible limits. the summer season recorded high levels of ec and tds compared to the monsoon season. the reasons are the intrusion of residual solids into the aquifer, the movement of water through sediments containing higher soluble mineral matter, and the influx of industrial and municipal wastes with incursion of seawater. the calcium and magnesium concentrations are generally derived from the leaching of limestone, dolomites, gypsum and anhydrites (garrels 1976, ramkumar et al. 2010, anithamary et al. 2012). calcium and magnesium ranged between 32-312 mg/l and 28-322 mg/l, respectively during the monsoon sub season, and 35.4-318 mg/l and 38.5-321 mg/l, respectively during the sum mer season. most of the ground water samples exceed who (1993) permissible limits in ca and mg. the increased concentration in ca and mg in ground water samples is due to the leaching of limestone, dolomites, gypsum and anhydrites (aiuppa et al., 2003, banaszuk et al., 2005, samsudin et al., 2008). the ionic concentration ranged between 109.3358.6 mg/l and 118.8-375.1 mg/l for so dium, and 354.8-821.9 mg/l and between 355.4-845.2 mg/l for chloride, respectively. thus, sodium might have been derived from the dissolution of evaporated minerals and its subsequent mixing with ground water. the higher na and cl concentration ground water in the coastal region is attributed to the influence of seawater on the coastal aquifer, which was highly visible during the summer due to a decline in water table (einsiedl 2012, kouzana et al. 2010, yidana, and yidana, 2009). ionic concentrations were 14.3-312.4 mg/l for potassium, and 18.9-324.9 mg/l for bicarbonate, and 97.3-326.2 mg/l for sulphate during the monsoon season; and 107.5-324.9 mg/l for potassium, 52.7378.2 mg/l for bicarbonate, and 58.6-398.6 mg/l for sulphate during the summer season, respectively. the large portion of potassium is due to the intrusion of saline water as well as the fertilizers of agricultural activity. a multivariate statistical approaches on physicochemical characteristics of ground water in and around nagapattinam district, cauvery deltaic region of tamil nadu, india 100 in this study, it was observed that na, k, ca, and mg were higher in summer than in monsoon season. na and mg were lower, but k and ca were higher during monsoon. it indicates that the seasonal variation of cation concentrations was very conspicuous in this environment. the relative abundance of sodium and magnesium during the summer revealed the influence of monsoon rainwater on the ground water. it led to an increase in these elements from irrigation runoff, thereby increasing their concentration in the ground water during summer season. the weathering of plagioclase feldspar and the nature of dissolved atmospheric carbon dioxide releases sodium and calcium, which progressively increases the sodium and calcium concentrations. chloride and sulfate were dominated during summer season, but chloride showed more variations than sulfate during two seasons. it is resulted from agricultural activity and seawater incursion in the shallow aquifer system. correlation between variables the correlations between major cations and anions were carried out using pearson’s correlation. a correlation analysis is a bivariate method applied to describe the degree of relation between hydrochemical parameters. associated the variable representing with correlation coefficient is r, and the multiple correlations, which is the percentage of variance in fig.1. location map of the study area s. venkatramanan, s. y. chung, t. ramkumar, g. gnanachandrasamy, s. vasudevan 101 the dependent variable, is explained collectively by all of the independent variables. the results of the correlation analysis are considered in the subsequent interpretation section. a high correlation coefficient (near 1 or -1) means implies a good strong relationship between the two variables, and a correlation coefficient around of approximately zero means implies that there is no relationship between them two variables, at a significance level of, 0.05. more precisely, it can be said that parameters showing r > 0.7 are considered to be strongly correlated, whereas when r has a value between 0.5 and 0.7, a moderate correlation is shown is said to exist (kim et al. 2002; helena et al. 2000, shammas and jacks, 2007). positive values of r indicate a positive relationship, while negative values indicate an inverse relationship. the correlation coefficients of the studied parameters are shown in table 2. in the present study, ec and tds are strongly correlated with na, cl and so4 during the monsoon, and similar type of result was observed during summer. they thus indicate seawater incursion and leaching of secondary salts. the lower concentration of calcium compared with sodium and chloride is a result of the cation exchange process that occurs naturally when seawater is introduced into the freshwater aquifer system (appelo and postma 2005, zeng and rasmussen, 2005, venkatramana et al., 2009). furthermore, during the summer season, sodium is positively correlated with cl, ec, and tds, indicating seawater incursion due to continuous pumping activities in the study area. the ca shows significant correlation with ec and tds and good correlation was noted in hco3. the na was strongly correlated with cl, clearly indicating that the seawater incursion plays a vital role in this present study (ramkumar et al., 2011 and anithamary et al 2012). varimax rotation factor analysis factor analysis is a frequent common statistical technique employed in hydrochemical studies (subbarao et al., 1996, adami et al., 1997, davis, 1986). the main goals of factor analysis are: (1) to reduce the number of ‘variables’and (2) to detect structure in the relationships between ‘variables’(analytical parameters) and/or ‘cases’(sampling points in our study). the mathematical base of factor analysis and a review of the chemical aspects can be found in mermet et al, (1998). currently, the calculations of factor analysis calculations and their graphical results performed using by davis (1986). consequently during the summer season, the first factor accounts for 24.17% whereas the second factor accounts for 44.00% (table 3). factors 1 and 2 show a good strong relationship with ec, tds, hco3, na, cl and mg, which is indicative of seawater incursion and agricultural runoff incursions in the study area. the concentration of na and cl can be ascribed to the incursion of seawater into the aquifer systems. furthermore, during the monsoon season, the first factor accounts for 24.15% and second factor accounts for 42.16%. the factor analyses performed in this study has clearly indicated that seawater incursion into the coastal aquifer as is the primary source of ground water pollution, the similar observations was made by venkatramanan et al. (2012). cluster analysis cluster analysis is used for grouping the cases based on the similarity of the responses to several variables. on the basis of the connecting distances between parameters and their positions on the dendrogram, distinctive clusters of the variables were defined for each of three locations along the basin. though this procedure is subjective, the distinction between clusters in this analysis is quite clear from the dendrogram. cluster analysis is a useful method for combining ground water wells into homogenous groups according to their water quality. in this study, three types of clustering nature was observed during the monsoon and summer seasons. the first clusters are ph, k, ca, na, hco3 and so4, whereas the second cluster is cl accompanied by ph, k, ca, na, hco3 and so4. the third cluster is accompanied by ec and tds. these variables would be found in natural mineralization and anthropogenic processes. the non-homogenous nature of the aquifer led to different hydrogeochemical facies for ground water such as na and cl. a close relationship between na and cl, may be originated from sea water incursion into coastal aquifers (figs. 3a, b). fig.2. distributions of physico-chemical parameters of ground water in the study area fig.3a. dendrogram of 10 chemical parameters for 10 cases (monsoon) fig.3b. dendrogram of 10 chemical parameters for 10 cases (summer) a multivariate statistical approaches on physicochemical characteristics of ground water in and around nagapattinam district, cauvery deltaic region of tamil nadu, india 102 conclusions the problem of ground water in nagapattinan district coastal region is rather complicated, and is derived from excessive exploitation of the aquifer systems. the study area depends totally on the water derived from the aquifer for its irrigation use. the overpumping of fresh water caused the deterioration of the water quality by the upwelling of deep saline water. fifty two ground water sam ples were collected near the coastal village of the nagapattinam district during monsoon and summer seasons, and physico-chemical characteristics of the samples were studied to determine the quality and the suitability of the ground water for various purposes as well as the characteristics of ground water quality. the maximum concentrations were observed near the coastal area because of sea water incursion. in addition, irrigation return flow also plays a major role in controlling ground water quality in the nagapattinam district. a dendrogram of the 10 cases and 10 variables are plotted and grouped into three main clusters. the values of correlation coefficients and their significance levels contributed to the selection of proper treatments to minimize the contamination of the ground water in the study area. the factor analysis reveals that the ground water from nagapattinam area has been greatly influenced by lithologic and environmental events in the area. a continuous monitoring program of the water quality will help to avoid further deterioration of the ground water quality in this coastal region. acknowledgements the authors are grateful to two anonymous referees for their constructive comments and suggestions which led to significant improvements to the manuscript. we also thank for the valuable suggestions given by carlos alberto vargas jimenez, editor, which greatly helped in the final presentation of the paper. this research was supported by a grant (code 13awmpb066761-01) from awmp program funded by the ministry of land, infrastructure and transport of korean government and department of science and technology (project no: dst/war-w/wsi/02/2010), new delhi, india. references adami, g., barbieri, p., picciotto, a. and reisenhofer, e. (1997). principal factor analysis as applied in environmental chemistry the study of eutrophication in a shallow lake, toxicological of environmental chemistry, 61, 99–108. anitha mary, i., ramkumar, t. and venkatramanan, s. (2012). application of statistical analysis for the hydrogeochemistry of saline groundwater in kodiakarai, tamilnadu, india, journal of coastal research, 28, 89–98. aiuppa, a., bellomo, s., brusca, l., alessandro, w. and federico, c. (2003). natural and anthropogenic factors affecting groundwater quality of an active volcano (mt. etna, italy) applied geochemistry, 18, 863–882. apha. (1995). standard methods for the examination of water and waste water, 19th edn. newyork, usa. appelo, c. a. j. and postma, d. (2005). geochemistry, groundwater and pollution, 2nd edition. rotterdam: balkema. banaszuk, p., wysocka-czubaszek. and a., kondratiuk. (2005). spatial and temporal patterns of groundwater chemistry in the river riparian zone agriculture, ecosystems and environment, 107, 167–179. brindha, r. and elango, l. (2011). hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india, earth sciences research journal, 15, 101-108. cloutier, v., lefebvre, r., therrien, r. and savard, m. m. (2008). multivariate statistical analysis of geochemical data as indicative of the hydrogeochemical evolution of groundwater in a sedimentary rock aquifer system. journal of hydrology, 353, 294–313. desimone, l. a., howes, b. l. and barlow, p. m. (1997). mass-balance analysis of reactive transport and cation exchange in a plume of wastewater-contaminated groundwater, journal of hydrology, 203, 228–249. einsied, f. (2012). sea-water/groundwater interactions along a small catchment of the european atlantic coast, applied geochemistry, 27, 73–80. engosh, a. and keren, r. (1996). chemical modifications of groundwater contaminated by recharge of sewage effluent, journal of contamination hydrology, 23, 347–360. garrels, r. m. (1976). a survey of low temperature water mineral relations in interpretation of environmental isotope and hydrogeochemical data in groundwater hydrology: vienna, international atomic energy agency, 22, 65–84. glen, e. p., cohen, m. j., morrison, j. i., valdes-casillas, c. and fitzsimmons, k. (1999). science and policy dilemmas in the management of agricultural waste waters: the case of the salton sea, ca, usa, environmental science policy, 2, 413–423. glynn, p. d. and plummer, l. n. (2005). geochemistry and the understanding of groundwater systems, hydrogeology journal, 13, 263–287. gnanachandrasamy, g., ramkumar, t., venkatramanan, s., anithamary, i., vasudevan, s. (2012) gis based hydrogeochemical characteristics of groundwater quality in nagapattinam district, tamil nadu, india. carpathian journal of earth and environmental sciences, 7, 205–210. hao, x. and change, c. (2002). does long-term heavy cattle manure application increase salinity of a clay loam soil in semi-arid southern alberta? agriculture ecosystem environment, 18, 1–16. hern, j. and feltz, h. r. (1998). effects of irrigation on the environment of selected areas of the western united states and implications to world population growth and food production, journal of environmental management, 52, 353–360. helena, b., pardo, r., vega, m., barrado, e., fernandez, j. m. and fernandez, l. (2000). temporal evolution of groundwater composition in alluvial aquifer (pisuerga river, spain) by principal component analysis, water research, 49, 359-372. hussein, m. t. (2004). hydrochemical evaluation of groundwater in the blue nile basin, eastern sudan, using conventional and multivariate techniques, hydrogeology journal, 12, 144–158. jones, b. f., vengosh, a., rosenthal, e. and yechieli, y. (1999). geochemical investigations, in: bear, j., cheng, a. h. d., sorek, s., ouazar, d., herrera, i. (eds.), seawater intrusion in coastal aquifers-concepts, methods and practices. kluwer, dordrecht, 51–72. kanfi, y., ronen, d. and magaritz, m. (1983). nitrate trends in the coastal plain aquifer of israel, journal of hydrology, 66, 331–341. kouzana, l., benassi, r., mammoua, a. b. and mennoubi, s. (2010). geophysical and hydrochemical study of the seawater intrusion in mediterranean semi arid zones. case of the korba coastal aquifer (cap-bon, tunisia), journal of african earth sciences, 58, 242–254. kim, r. h., yum, b. w. and chang, h. w. (2002). hydrogeochemical and isotopic characteristics for salinization of a shallow groundwater in coastal area, youngkwang, korea, proc. 17th salt water intrusion meeting, delft, the netherlands, 227–237. maslia, m. l. and prowell, d. c. (1990). effect of faults on fluid flow and chloride contamination in a carbonate aquifer system, journal of hydrology, 115, 1–49. mermet, j. m., otto, m. and widmer, h. m. (1998). analytical chemistry, wiley, new york. mitchel, j. p., shennan, c., singer, m. j., peters, d. w., miller, r. o., prichard, t., grattan, s. r., rhoades, j. d., may, d. m. and munk, d. s. (2000). impacts of gypsum and winter cover crops on soil physis. venkatramanan, s. y. chung, t. ramkumar, g. gnanachandrasamy, s. vasudevan 103 cal properties and crop productivity when irrigated with saline water, agriculture water management, 45, 55–71. ramkumar, t., venkatramanan, s., anithamary, i. and syed ibrahim, s. (2011). evaluation of hydrogeochemical parameters and quality assessment of the groundwater in kottur blocks, district, tamilnadu, india, arabian journal of geosciences, doi:10.1007/s12517-0110327-2. ramkumar, t., venkatramanan, s., anitha mary, i., tamilselvi, m. and ramesh, g. (2010). hydrogeochemical quality of groundwater in vedaraniyam town, tamilnadu, india, research journal of environmental and earth sciences, 2, 44-48. ronen, d., kanfi, y. and magaritz, m. (1983). sources of nitrates in groundwater of the coastal plain of israel—evolution of ideas, water research, 17, 1499–1503. ronen, d. and magaritz, m. (1985). high concentration of solutes at the upper part of the saturated zone (water table) of a deep aquifer under sewage irrigation land, journal of hydrology, 80, 311–323. rosenthal, e., vinokurov, a., ronen, d., magaritz, m. and moshkovitz, s. (1992). anthropogenically induced salinization of groundwater: a case study from the coastal plain aquifer of israel, journal of contamination hydrology, 11, 149–171. samsudin, a. r., haryono, a., hamzah, u. and rafek, a. g. (2008). salinity mapping of coastal groundwater aquifers using hydrogeochemical methods: a case study from north kelantan, malaysia, environmental geology, 55, 1737–1743. sreekanth, j. and datta, b. (2010). multi-objective management of saltwater intrusion in coastal aquifers using genetic programming and modular neural network based surrogate models, journal of hydrology, 393, 245–256. shammas, m. i. and jacks, g. (2007). seawater intrusion in the salalah plain aquifer. oman, environmental geology, 53, 575– 587. subbarao, c., subbarao, n. v. and chandu, s. n. (1996). characterization of groundwater contamination using factor analyses, environmental geology, 28, 175–180. vengosh, e. and rosenthal, e. (1994). saline groundwater in israel; its bearing on the water crisis in the country, journal of hydrology, 156, 389–430. vengosh, a., spivack, a. j., artzi, y. and ayalon, a. (1999). geochemical and boron, strontium and oxygen isotopic constraints on the origin of the salinity in groundwater from the mediterranean coast of israel, water resource research, 35, 1877–1877. venkatramanan, s., ramkumar, t., vasudevan, s., ramesh, g., anitha mary, i. and vijayakumar, v. (2009). assessment of hydrogeochemistry of groundwater in muthupet coastal region, tamilnadu, india, international journal of applied environmental sciences, 4, 169–176. venkatramanan, s., ramkumar, t. and anithamary, i. (2012). a statistical approach on hydrogeochemistry of groundwater in muthupet coastal region, tamilnadu, india, carpathian journal of earth and environmental sciences, 7, 47 – 54. yidana, s. m. and yidana, a. (2009). assessing groundwater quality using water quality index and multivariate statistical analysis-the voltaian basin, ghana, environmental earth science, 42, 132– 145. zeng, x. and rasmussen, c. d. (2005). multivariate statistical characterization of water quality in lake lanier, georgia, usa, journal of environmental quality, 34, 1980–1991. who. (1993). fluorine and fluorides, environmental health criteria, 36. world health organization, geneva. a multivariate statistical approaches on physicochemical characteristics of ground water in and around nagapattinam district, cauvery deltaic region of tamil nadu, india archivo03.indd 25 manuscript received june 12, 2006. earth sciences research journal ground surface temperature histories inferred from 15 boreholes temperature profiles: comparison of two approaches l.v. eppelbaum1 , i.m. kutasov1 and g. barak1 1dept. of geophysics and planetary sciences, raymond and beverly sackler faculty of exact sciences, tel aviv university, ramat aviv 69978, tel aviv, israel corresponding author: l.v. eppelbaum, e-mail: levap@post.tau.ac.il abstract understanding the climate change processes requires application of special methodologies for revealing a ground surface temperature history (gsth). it was proved by different authors that the gsth may be determined on the basis of analysis of the temperature field observed in short boreholes. in this paper, the authors analyze four mathematical models describing the gsth: (1) sudden change, (2) linear increase, (3) square root of time increase and (4) exponential increase. fifteen borehole temperature profiles from europe, asia and north america were selected in three groups based on their geographical proximity. after careful analysis of temperature-depth profiles in these boreholes it was found out that two models (linear increase and square root of time increase) provide the best fit with field data. the calculated warming rates in the 20th century were compared with those obtained by a few parameters estimation (fpe) technique. key words: temperature, borehole, climate modeling. resumen para entender los procesos de cambio climático se requiere la aplicación de metodologías especiales que revelen una historia de la temperatura de la superficie del suelo (gsth = por su abreviación en inglés). se ha probado por diferentes autores que gsth puede ser determinado con base en el análisis del de campo temperaturas observado en pozos cortos. en este artículo, los autores analizan cuatro modelos matemáticos describiendo el gsth: (1) cambio súbito, (2) incremento linear, (3) raíz cuadrada del incremento del tiempo e (4) incremento exponencial. quince perfiles de temperatura de pozos de europa, asia y norte américa fueron seleccionados en tres grupos con base en su proximidad geográfica. luego de un cuidadoso análisis de perfiles temperatura-profundidad en estos pozos se encontró que dos modelos (incremento linear y raíz cuadrada del incremento del tiempo) proporcionan los mejores ajustes a los datos de campo. las tasas de calentamiento en el siglo xx fueron comparadas con aquellas obtenidas con la técnica denominada estimación de unos cuantos parámetros (fpe). palabras clave: temperatura, pozo, modelamiento climático. earth sci. res. j. vol. 10, no. 1 (jun. 2006): 25-34 26 l.v. eppelbaum , i.m. kutasov and g. barak introduction at present many efforts are made to determine the trends in ground surface temperature history (gsth) from geothermal surveys (e.g., lachenbruch and marshall, 1986; baker and ruschy, 1993; pollack et al., 2000; majorowicz and safanda, 2005). in this case accurate subsurface temperature measurements are needed to solve this inverse problem − namely the estimation of the unknown time dependent ground surface temperature (gst). the variations of the gst during the long term climate changes resulted in disturbance (anomalies) of the temperature field of formations. thus, the gsth can be evaluated by analyzing the present precise temperature-depth profiles. the effect of surface temperature variations in the past on the temperature field of formations is widely discussed in the literature. three approaches are used in deriving climate information from borehole temperature profiles. in the first case the ground surface temperature history (gsth) is reconstructed as an arbitrary function of time (e.g., cermak, 1971; lachenbruch and marshall, 1986; beltrami et al., 1992; shen and beck, 1992; baker and ruschy, 1993; clauser and mareschal, 1995; harris and chapman, 1995; pollack et al., 2000; jain and pulwarty, 2006). huang et al. (1996) called such an approach an arbitrary function reconstruction (afr). the second approach for inversion of temperature profiles – a few parameter estimation (fpe) technique was suggested by huang et al. (1996). as it was demonstrated by the authors, the fpe technique allows comparison of the inversion results, both spatially and temporally. the third approach is the generalized inverse method named the functional space inversion (fsi) technique (shen and beck, 1991; shen et al., 1995). the fsi method allows for uncertainties in temperature-depth data, thermal properties of formations and heat flow density to be incorporated into the model. in this paper we will compare results of our gsth calculations with those obtained by the fpe technique. for this reason we will consider some of the main features of the fpe method. the main considerations to utilize the fpe technique include (huang et al., 1996): (a) the resolving power of a temperature profile for gsth reconstruction. due to the amplitude attenuation of thermal diffusion and the presence of observational and representational noise in borehole data, vigorous estimations can often be made for only a few parameters such as the trend, duration, and the overall amplitude of the ground surface temperature change in the past. (b) the need to simplify and standardize the procedures for reconstruction of gsth. the problem of inverting borehole temperatures to yield a ground surface temperature (gst) is an ill-posed problem, and some constraints are required for a stable solution. to allow a more consistent comparison of temperature inversion results, a standardization of surface temperature reconstruction is needed. the standardization is a difficult task in an afr because of the high degrees of freedom involved in representing a gsth. (c) convenience in comparing results. an afr techniques attempts to reconstruct a gsth at various time scales and degrees of details. however, in a regional or in a continent-wide study only a comparison of general features is needed instead of the details of gsth’s obtained from different areas. the forward calculation approach (afr) was used in the analysis and interpretation of borehole temperatures in terms of a gsth. fifteen borehole temperature profiles from europe, asia, and north america, were selected (huang and pollack, 1998; www.geo.lsa.umich.edu/~climate). three groups based on geographical proximity were formed (table 1). four mathematical models to describe the gsth (sudden change, linear increase, square root of time, and exponential increase) were used to approximate the temperature-depth profiles of the boreholes. the objective of this study is to calculate the warming rates (r) during the 20th century by the afr method and to compare them with those obtained by the few parameter estimation (fpe) technique. it is also reasonable to assume that for close spaced boreholes the values of r should vary with narrow limits. methodology mathematical models and assumptions let us assume that t x years ago from now the ground surface temperature started to increase (warming) or reduce (cooling). prior to this moment the subsurface temperature was (figure 1): 27 ground surface temperature histories inferred from 15 boreholes temperature profiles: comparison of two approaches table 1. input data for 15 boreholes and results of temperature inversions. (huang and pollack (1998); www.geo.lsa.umich.edu/~climate). w el l # well code longitude latitude h a h b , m h t h c , m t o , oc r, rate after 1900 k/100a north america 1 2 3 4 5 6 ca-9901 ca-9906 ca-9907 ca-9804 ca-9806 ca-9807 -101.50 -101.84 -100.56 -100.76 -101.57 -101.57 54.72 54.77 54.93 55.16 54.79 54.79 49.81-119.58 49.97-149.78 48.00-113.78 48.30-106.25 42.49-84.52 47.06-100.80 189.47-596.47 199.38-599.14 196.03-523.34 154.41-307.76 167.22-498.78 175.81-446.92 2.7 2.1 1.1 2.1 2.8 2.5 2.487 1.219 1.384 0.378 0.803 0.419 europe 7 8 9 10 11 cz-127127 cz-hu-7 cz-hu-9 cz-mj-5 cz-mj-8 14.87 12.81 12.81 14.86 14.58 50.73 50.11 50.11 50.57 50.36 80.80-140.00 50.00-120.00 50.00-100.00 60.00-180.00 50.00-120.00 200.00-440.00 180.00-350.00 200.00-460.00 230.00-290.00 180.00-310.00 6.3 5.6 8.1 5.5 0.855 1.533 3.751 0.212 1.787 asia 12 13 14 15 cn-fj-ql17 cn-gd-c3901 cn-jxck46-25 cn-jxzk59-38 116.94 113.18 116.33 116.33 26.33 25.42 27.97 27.97 60.00-100.00 50.00-120.00 60.00-140.00 50.00-110.0 140.00-430.00 160.00-260.00 200.00-300.00 200.00-380.00 18.5 18.4 19.9 13.7 2.532 1.870 1.181 0.300 figure 1. temperature profiles: a principal scheme. ab – cooling, ac – warming. and γ is the geothermal gradient. is also assumed that the formation is a homogeneous medium with constant thermal properties. now the current (t = t x ) subsurface temperature is (in case of warming): )(),( zftttzt ocxc +== (2) where t oc is the current (at the time (date) of temperature logging) mean ground surface temperature; and f(z) is a function of depth that could be obtained from the field data. in some cases the value of t oc can be obtained by extrapolation of the function t c to z = 0. however, in most cases, the value t oc can be estimated by trial and error method: assuming an interval of values for t oc , calculating for each t oc value of the temperature profiles t c (using a computer program) for various models of change in the ground surface temperature (gst) with time and, finally, finding a best match between calculated and field measured t c profiles. in our study we found that a quadratic regression program performed for the section h a – h b (figure 1) can be utilized to estimate the value of t oc = a o (table 2), zttzt oaa γ+== )0,( (1) where t oa is the mean ground surface temperature at the moment of time t = 0 years; z is the vertical depth 28 l.v. eppelbaum , i.m. kutasov and g. barak table 2. reduced temperatures and results of temperature inversions for the borehole ca-9901. t xc = 52.9 yrs, t xl = 137.5 yrs, t xs = 94.0 yrs, t xe = 282.8 yrs, ∆t rc = 0.092 oc, ∆t rl = 0.043 oc, ∆t rs =0.039 oc, ∆t re = 0.275 oc, a l =0.0245 oc/yr α s =0.3474 oc/yr1/2 a e =0.002971 oc/yr h, m t rc , oc t rl , oc t rs , oc t re , oc t ro , oc 29.89 39.85 49.81 59.77 69.72 79.66 89.62 99.61 109.60 119.58 2.10 1.73 1.39 1.10 0.85 0.64 0.48 0.34 0.24 0.17 2.02 1.68 1.39 1.15 0.94 0.76 0.61 0.49 0.39 0.31 2.04 1.69 1.39 1.13 0.91 0.73 0.58 0.45 0.34 0.26 1.72 1.55 1.39 1.25 1.12 1.01 0.90 0.80 0.71 0.63 2.15 1.71 1.39 1.13 0.92 0.75 0.60 0.48 0.37 0.29 2 21),( zazaattzt oxc ++== (3) where a o , a 1 , and a 2 are the coefficients. the temperature-depth values for all wells were taken from the database (huang and pollack, 1998; www.geo.lsa.umich.edu/~climate). four models of changing gst values with time were considered. the corresponding mathematical solutions are presented in the literature (e.g., carslaw and jaeger, 1959; cermak, 1971; lachenbruch and marshall, 1986; lachenbruch et al., 1988; powell et al., 1988). in the first model (model c) we assumed that t xc years ago the gst value suddenly changed from t o to t oc . the current temperature anomaly (the reduced temperature) is ztzftzt oocr γ−−+= )()( (4) and the solution is xcorrc tt at z xttt =      φ∆== 2 )(* (5) ooco ttt −=∆ (6) where a is the thermal diffusivity of formation and φ*(x) is the complementary error function. in the second model (model l) we assumed that t xl years ago the gst value started gradually to change from t o to t oc . we assumed that gst is a linear function of time and xllooc ttt α+= (7) where a l is some coefficient. the solution is * expt t t at z at z at z at z t t1 2 2 4 rl r l xl 2 2 = = + =a r ub d bl n l( 2 (8) in the third model (model s) we also assumed that t xs years ago the gst value started gradually to change from t o to t oc . we assumed that gst is a square root function of time and xssooc ttt α+= (9) where a s is a coefficient. the solution is *expt t t at z at z at z t t 4 2 2 rs r s xs 2 = = =a r ub dl n) 3 (10) and, finally, in the fourth model (model e) we assumed that the gst value exponentially increases with time and )exp( xeeooc ttt α= (11) where a e is a coefficient. the solution is (t re =t r ) * * exp exp exp t t z a at z t z a at z t t t 2 1 2 2 re e e e e e xe = + + = a a a a a u u ] b d b d g l n l n < f (12) 29 ground surface temperature histories inferred from 15 boreholes temperature profiles: comparison of two approaches computer programs were used for processing field data and calculating values of t x , t oc , t o , γ, f(z), a l , a s , and a e . to demonstrate the calculation technique we present below a field example. example of calculations for the interval 49.81-119.58 m of the borehole ca-9901 (table 1 and figure 2) the temperature profile can be approximated by a quadratic equation (standard regression program was applied) 20001927.003626.092.5),( zzttzt xc +−== (13) therefore, the present mean ground surface temperature (gst) is 5.92oc (figure 1). from the observed t-z data it follows that in the 189.47-596.47 m section of the well; the temperature gradient is practically constant. it was assumed that the temperature gradient in this section did not change. for this interval (here a standard regression program was applied) ztztc 01224.056.2)0,( +== (14) therefore, in the past (t x years ago) the mean ground surface temperature t o = 2.56oc and warming occur. figure 2. observed and calculated temperature profiles, borehole ca-9901, north america (see table 1). in our case the observed reduced temperatures (for z < 119.6 m) are: ztzt obsor 01224.056.2)( −−= (15) where t obs values are the measured temperatures (figure 2). for the calculations below we used the depth z = h 1 = 49.81 m, where t r =1.39 oc. the temperature change (anomaly) of the gst is 5.92oc – 2.56oc = 3.36oc. a computer program calculates the reduced temperatures versus depth for the four models of change in gst and compares them with the observed anomalies. for the first model (model c) the reduced temperature is t rc (equation 5). in our case the gst changes from 2.56oc to 5.92oc. in the second model (model l) the reduced temperature is t rl (equation 8) and the ground surface temperature linearly changes from 2.56oc to 5.92oc, a change that can be modeled using the following equation xll tα+= 56.292.5 in the third model (model s) the gst is a square root function of time. the reduced temperature is t rs (equation 10) and xss tα+= 56.292.5 in the fourth model (model e) the gst is an exponential function of time. the reduced temperature is t re (equation 12) and )exp(56.292.5 xee tα= for all models the values of t x and a were estimated by a computer program. the value of thermal diffusivity a = 0.004 m2/hr was used in the temperature inversions. the results of calculations are presented in tables 2 and 3, and in figures 2 and 3. analyzing the data from table 2 and figure 2 we can conclude that the best fit (minimum values of average squared deviations ∆t rl = 0.043 oc and ∆t rs = 0.039 oc) is achieved when the models l or s are applied. 30 l.v. eppelbaum , i.m. kutasov and g. barak table 3. coefficients in equation 1 (b o = t oa , b 2 = γ) and equation 3. t 1,r is the reduced temperature at z = h 1 . w el l # well code h 1 m t 1 , r oc well section h a h b , m well section h t h c , m a o , oc a 1 ·102, oc /m a 2 ·103, oc /m2 b o , oc b 1 ·102, oc /m north america 1 2 3 4 5 6 ca-9901 ca-9906 ca-9907 ca-9804 ca-9806 ca-9807 49.81 49.97 48.00 48.30 42.49 47.06 1.39 1.08 0.76 0.20 0.48 0.52 5.92 4.06 2.51 2.76 3.97 3.70 -3.626 -0.954 -0.133 -0.241 -1.289 -0.210 0.1927 0.0588 0.0317 0.0464 0.1354 0.0515 2.56 2.09 1.19 2.17 2.69 2.65 1.224 1.125 1.211 0.756 1.134 1.092 europe 7 8 9 10 11 cz-127127 cz-hu-7 cz-hu-9 cz-mj-5 cz-mj-8 70.00 50.00 50.00 60.00 50.00 0.91 1.49 0.80 1.31 2.39 8.68 9.73 10.38 9.23 10.46 -0.003 -1.825 -0.781 0.854 -1.042 0.0328 0.2084 0.1606 0.0503 0.1673 6.85 6.24 8.07 6.65 5.86 1.562 3.134 3.056 3.211 4.199 asia 12 13 14 15 cn-fj-ql17 cn-gd-c3901 cn-jxck46-25 cn-jxzk59-38 60.00 50.00 40.00 50.00 0.44 1.08 1.54 2.22 19.77 21.32 24.07 18.64 -0.959 -1.140 -1.327 -1.161 0.1132 0.1023 0.1027 0.1250 18.18 19.02 20.77 14.77 1.723 1.846 2.154 2.813 table 4. results of temperature inversions for two models. ∆t l and ∆t s are the average squared temperature deviations, r l and r s are the warming rates. well codes are presented in tables 1 and 3. w el l # model l model s r l , k/100a r s at t = t xs k/100atxl, yr ∆tl, k al, k/yr txs yrs ∆ts, k as, k/yr 0.5 north america 1 2 3 4 5 6 137.5 278.6 319.1 92.7 81.9 192.1 0.043 0.055 0.079 0.011 0.027 0.034 0.02450 0.00786 0.00411 0.00636 0.01558 0.00545 94.0 185.4 210.8 64.3 56.5 128.9 0.039 0.032 0.068 0.012 0.033 0.040 0.3474 0.1450 0.0904 0.0734 0.1697 0.0922 2.450 0.786 0.411 0.636 1.558 0.545 1.792 0.532 0.311 0.458 1.129 0.406 europe 7 8 9 10 11 415.2 148.8 98.6 328.1 241.9 0.078 0.092 0.097 0.174 0.082 0.00441 0.02341 0.02343 0.00785 0.01901 279.0 101.4 68.5 219.8 161.7 0.064 0.118 0.079 0.144 0.037 0.10971 0.34591 0.27927 0.17361 0.36149 0.441 2.341 2.343 0.785 1.901 0.328 1.718 1.690 2.482 1.421 asia 12 13 14 15 100.3 183.5 260.7 331.7 0.084 0.082 0.060 0.060 0.01590 0.01251 0.01268 0.01165 70.9 124.0 176.2 219.5 0.078 0.062 0.061 0.041 0.18942 0.20623 0.24911 0.26089 1.590 1.252 1.268 1.165 1.125 0.926 0.939 0.907 31 ground surface temperature histories inferred from 15 boreholes temperature profiles: comparison of two approaches close spaced boreholes (figures 7 and 8). analysis of figure 7 indicates that the duration of the warming period was five centuries and the warming rate is minimal in the last century. however, it is widely accepted that most of the warming occurs in the last century. at the same time the gsth reconstructed from borehole no. 12 temperature-depth data (figure 8) indicates that the cooling period lasted for three centuries and a very high warming rate for the 20th century was calculated by the fpe technique. inversion results in most cases the model l (linear increase) and model s (square root time increase) provide the best fit to the field data (table 4, figure 4). in several cases the model c (sudden change) allows to approximate the measured temperature-depth data with good accuracy (figures 5 and 6). for boreholes no. 16 (north america, table 4, model l) the current warming rates vary in the 0.4112.450 k/100a range. the wide range for the warming rate of 0.328-2.482 k/100a was also determined for five boreholes in europe (table 4, model l). interesting results we obtained for four boreholes in china (table 4). in this case the warming rate (r l ) varies with relatively narrow limits (1.1651.590 k/100a.) it was found that the duration of warming periods (table 4, model l) is: 82-279 years for wells no. 1-6; 99-415 years for wells no. 7-11; and 100-332 years for wells no. 12-15. similar results (r s and t xs , table 4) were obtained for the model s. discussion and conclusions as we mentioned in the introduction section the main objective of our study is to calculate the warming rates (r) during the 20th century by the afr method and to compare with those obtained by the few parameter estimation (fpe) technique (table 1). the warming rate estimated by the fpe technique varied in wide ranges: 0.378-2.487 k/100a (north america); 0.212-3.751 k/100a (europe); and 0.3002.532 k/100a (asia). in our case the fpe method allowed to determine temperature trends (warming or cooling rates) over the past five centuries (huang et al., 2000). in the inversion we employ a priory null hypothesis for the gsth that there has been no climate change. from tables 1 and 4 follows that for the boreholes no. 1-11 (north america and europe) both approaches (afr and fpe) provide practically the same ranges of warming rates. for boreholes in asia (no. 12-15) the afp technique gives a more consistent (narrow) range of warming rates (1.1651.590 k/100a). let us compare the gsth of two figure 3. the possible scenarios of the ground surface temperature history, borehole ca-9901, north america (see table 1). 32 l.v. eppelbaum , i.m. kutasov and g. barak the results of temperature inversion by both techniques show that probably some non-climatic effects (well shut-in periods, vertical and horizontal water flows, sedimentation, uplift, erosion, steep topography, lakes, vertical variation in heat flow, lateral thermal conductivity contrasts, thermal conductivity anisotropy, forest fires, farming, etc.) may have perturbed the borehole temperature profiles (e.g., carslaw and jaeger, 1959; lachenbruch, 1965; kappelmeyer and haenel, 1974; powell et al., 1988, majorowicz and skinner, 1997; guillou-frottier et al., 1998; kutasov and eppelbaum, 2003; gruber et al., 2004; bodri and cermak, 2005; kutasov and eppelbaum, 2005; majorowicz and safanda, 2005; mottaghy et al., 2005). this study shows that extreme caution should be used in the selection of temperature-depth profiles for inferring the ground surface temperature histories. a good example in this regard was demonstrated in the study conducted by guillou-frottier et al. (1998), where only 10 out of 57 temperature profiles were figure 5. the possible scenarios of the ground surface temperature history, borehole cz-127127, europe (see table 1). figure 4. observed and calculated temperature profiles, borehole ca-9807, north america (see table 1). selected for inversion of the past ground surface temperatures. we can conclude that only the calculated warming rates for wells in asia (1.2-1.6 k/100a for model l or 0.9-1.1 k/100a for model s, table 4) can be used for forecasting of short term warming trends. 33 ground surface temperature histories inferred from 15 boreholes temperature profiles: comparison of two approaches references • baker, d. g., and d. l. ruschy (1993). the recent warning in eastern minnesota shown by ground temperatures, geophysical research letters, 20, 371-374. • beltrami, h., a. m. jessop, and j.-c. mareschal (1992). ground temperature histories in eastern and central canada from geothermal measurements: evidence of climate change, palaeogeography, paleoclimatology & palaeoecology (global planetary change section), 98, 167-183. • bodri, l., and v. cermak (2005). borehole temperatures, climate change and the pre-observational surface air temperature mean: allowance for hydraulic conditions, global planetary change, 45, 265-276. • cermak, v. (1971). underground temperature and inferred climatic temperature of the past millennium, palaeogeography, palaeoclimatology & palaeoecology, 10, 1-19. • clauser, c., and j. c. mareschal (1995). ground temperature history in central europe from borehole temperature data, geophysical journal international, 121, 805-817. • gruber, s., l. king, t. kohl, t. herz, w. haeberli, and m. hoelzle (2004). interpretation of geothermal profiles perturbed by topography: the alpine permafrost boreholes at stockohorn plateau, switzerland, permafrost and periglacial processes, 15, 349-357. • guillou-frottier, l., j. c. mareschal, and j. musset (1998). ground surface temperature history in central canada inferred from 10 selected borehole temperature profiles, journal of geophysical research, 103 (b4), 7385-7397. • harris, r. n., and s. d. chapman (1995). climate change on the colorado plateau of eastern utah inferred from borehole temperatures, journal of geophysical research, 100, 63676381. figure 6. the possible scenarios of the ground surface temperature history, borehole ca-9806, north america (see table 1). figure 7. the ground surface temperature history, borehole cn-jxzk59-38, asia (see table 1). acknowledgements the authors thank dr. raisa dorofeeva (institute of the earth’s crust, russian academy of sciences (siberian branch), and irkutsk, russia), dr. arkady pilchin (universal geoscience & environmental consulting co., ontario, canada) and dr. john sánchez, editorin-chief of the earth sciences research journal, for their useful comments and suggestions. figure 8. the ground surface temperature history, borehole cn-fj-q117, asia (see table 1). 34 l.v. eppelbaum , i.m. kutasov and g. barak • huang, s., p. y. shen, and h. n. pollack (1996). deriving century – long trends of surface temperature change from borehole temperatures, geophysical research letters, 23, 257-260. • huang, s., and h. n. pollack (1998). global borehole temperature database for climate reconstruction, igbp pages/world data center-a for paleoclimatology data contribution series #1998-044, noaa/ngdc paleoclimatology program, boulder co, usa, (see also http://www.geo.lsa.umich.edu/~climate, last accessed july 05, 2006.) • huang, s., h. n. pollack, and p. y. shen (2000). temperature trends over past five centuries reconstructed from borehole temperatures, nature, 403, 756-758. • jain, s., and r. s. pulwarty (2006). environmental and water decision-making in a changing climate, eos, 87, no. 14, p.139. • kappelmeyer, o., and r. haenel (1974). geothermic with special reference to application, gebruder borntrager, berlin. • kutasov, i. m., and l. v. eppelbaum (2003). prediction of formation temperatures in permafrost regions from temperature logs in deep wells – field cases, permafrost and periglacial processes, 14, no. 3, 247-258. • kutasov, i. m., and l. v. eppelbaum (2005). determination of formation temperature from bottom-hole temperature logs – a generalized horner method, journal of geophysics and engineering, 2, 90-96. • lachenbruch, a. h., and b. v. marshall (1986). changing climate: geothermal evidence from permafrost in the alaskan arctic, science, 234, 689-696. • lachenbruch, a. h. (1965). rapid estimation of the topographic disturbance to superficial thermal gradients, review of geophysics, 6, 365-400. • lachenbruch, a. h., t. t. cladouhos, and r. w. saltus (1988). permafrost temperature and the changing climate, 5th international conference on permafrost, 3, tapir publishers, trondheim, norway, 9-17. • majorowicz, j. a., and w. p. skinner (1997). potential causes of differences between ground and surface air temperature warming across different ecozones in alberta, canada, global planetary change, 15, 79-91. • majorowicz, j. a., and j. safanda (2005). measured versus simulated transients of temperature logs – a test of borehole climatology, journal of geophysics and engineering, 2, 1-8. • mottaghy, d., r. schellschmidt, y. a. popov, c. clauser, i. t. kukkonen, g. nover, s. milanovsky, and r. a. romushkevich (2005). new heat flow data from immediate vicinity of the kola super-deep borehole: vertical variation in heat flow confirmed and attributed to advection, tectonophysics, 401, 119-142. • pollack, h. n., h. shauopeng, and p.y. shen (2000). climate change record in subsurface temperatures: a global perspective, science, 282, 279-281. • powell, w. g., d. s. chapman, n. balling, and a. e. beck (1988). continental heat-flow density, r. haenel, l. rybach, and l. stegena, (editors.), handbook of terrestrial heat-flow density determination, kluwer, 167-222. • shen, p. y., and a. e. beck (1991). least squares inversion of borehole temperature measurements in functional space, journal of geophysical research, 96, 19,965-19,979. • shen, p. y., and a. e. beck (1992). paleoclimate change and heat flow density inferred from temperature data in the superior province of the canadian shield, palaeogeography, palaeoclimatology & palaeoecology (global planetary change section), 98, 143-165. • shen, p. y., h. n. pollack, s. huang, and k. wang (1995). effects of subsurface heterogeneity on the inference of climate change from borehole temperature data: model studies and field examples from canada, journal of geophysical research, 100, 6383-6396. keywords: geomenbrane; stress; clay interface. palabras clave: geomembrana; esfuerzo; interfaz de arcilla. how to cite item zhang, h. (2020). study on the internal force of geomembrane of landfill in heavy metal contaminated area. earth sciences research journal, 24(1), 111-118. doi: https://doi. org/10.15446/esrj.v24n1.85231 in order to analyze the tension stress of geomembrane in the seepage control system of a landfill in a heavy metal contaminated area under the action of the overlying landfill weight, the internal force of geomembrane in a landfill is studied in this paper. firstly, according to the internal force analysis of geomembrane model, the basic equation of geomembrane is obtained, and the equation is analyzed by finite difference numerical solution, so as to analyze the internal force of geomembrane in a landfill. the method can analyze three stress states of geomembrane-clay interface at different positions: elastic state, softening state and residual state, which makes the stress analysis of geomembrane on landfill slope more reasonable. parametric analysis shows that when the strength loss and the difference between residual displacement and peak displacement are small in the softening stage, the peak strength of geomembrane-clay interface has little influence on the maximum tensile stress of geomembrane and its residual strength is the main controlling factor, and when the softening characteristics of geomembrane-clay interface are obvious, the influence of peak strength and residual displacement is more obvious. abstract study on the internal force of geomembrane of landfill in heavy metal contaminated area estudio sobre la fuerza interna de la geomembrana del relleno sanitario en un área contaminada con metales pesados issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.85231 con el fin de analizar el esfuerzo de tensión de la geomembrana (una lamina geosintética que busca evitar la migración de contaminantes al suelo) en el sistema de control de infiltración de vertederos en áreas contaminadas con metales pesados bajo la acción del peso del vertedero suprayacente, en este documento se estudia la fuerza interna de la geomembrana en los vertederos. en primer lugar, de acuerdo con el análisis de fuerza interna del modelo de la geomembrana se obtiene la ecuación básica de la lámina, y la ecuación se analiza por solución numérica de diferencia finita para medir su fuerza interna en el relleno sanitario. el método puede analizar tres estados de tensión de la interfaz de geomembrana-arcilla en diferentes posiciones: estado elástico, estado de reblandecimiento y estado residual, lo que hace que el análisis de tensión de la geomembrana en la pendiente del vertedero sea más razonable. el análisis paramétrico muestra que cuando la pérdida de resistencia y la diferencia entre el desplazamiento residual y el desplazamiento máximo son pequeñas en la etapa de reblandecimiento, la resistencia máxima de la interfaz geomembrana-arcilla tiene poca influencia en el esfuerzo de tensión máximo de la geomembrana, y su resistencia residual es el principal factor de control; y cuando las características de ablandamiento de la interfaz geomembranaarcilla son obvias, la influencia de la resistencia máxima y el desplazamiento residual es más obvia. resumen record manuscript received: 07/07/2019 accepted for publication: 13/11/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 111-118 hongzhou zhang institute of architectural civil engineering, langfang normal university, langfang, 065000, china * corresponding author: zhanghongzhou@lfnu.edu.cn so il s 112 hongzhou zhang introduction geomembrane is one of the most important components of bottom liner system and top cover system in modern sanitary landfill. it plays an important role in preventing waste leachate from polluting groundwater and protecting the surrounding environment. with the increase of landfill height, the geomembrane in heavy metal contaminated area will produce tensile stress under the action of overlying heap weight. in order to increase the interface strength, reduce the tension stress in geomembrane and improve the stability of landfill interface, rough geomembrane is more and more used in landfill’s slope liner system. according to the indoor shear test results, the interface of geosynthetics-geosynthetics and geosynthetics-clay presents obvious strain softening characteristics. therefore, the key to calculate the tensile stress of geomembrane is to select the strength parameters. at present, the conventional limit equilibrium method cannot solve the tension stress of geomembrane in the state of softening interface. (1) the failure process of polyurea composite impervious layer in shrinkage joint is analyzed by using v-zone model (czm). based on the loaddisplacement-deformation data obtained from the debonding test of impervious layer, the interfacial stiffness and viscosity coefficient of the bonding element of the bonding layer are calibrated. in the numerical example, the critical load of coating debonding is determined, and the influence of the impervious layer’s thickness of polyurea composite on the critical load of debonding is analyzed. however, when water pressure is applied to the open shrinkage joint, the thickness of the coating, the elastic modulus and the shear stress of the bonding parts will affect the debonding length of the coating, which will cause water seepage in the pit and depression areas. (2) a generalized version of the model is created by regression fitting all size categories of the analysis unit at a given research site. the root mean square error of the corrected pic estimates ranges from 1.5% to 10.7%; these values are much better than the rmse values of the uncorrected pic estimates. for calibrated and uncalibrated pics, the accuracy of estimation increases with the increase of the size of the analysis unit. however, the unit size, geographic location or land’s cover data set coefficients of landfills are similar, which also leads to a significant decline in the performance of the model when applied outside the development area. (3) based on the saturated-unsaturated seepage finite element theory, the three-dimensional seepage field of a typical earth-rock dam with random distribution of impervious geomembrane defects is simulated. considering the number, type, location, size and distribution of imperfections in geomembrane, the seepage quantity of imperfections in geomembrane and the variation law of infiltration surface behind the membrane are mainly analyzed, so as to analyze the internal force of geomembrane according to the law of analysis. however, the interface strength model ignores the transition stage of softening, which is inconsistent with the actual situation. therefore, it is necessary to adopt a more reasonable analysis model for the interface softening of geomembrane to analyze the tension stress of geomembrane on landfill slope. in view of the above problems, this paper studies the internal force of geomembrane on the basis of clay interface, taking landfill as the core. while describing in detail, the model of geomembrane is constructed and its internal force is analyzed. the influence of the height, slope gradient and anchorage length of geomembrane on the tensile stress of geomembrane is analyzed by the cable fractional solution. geomembrane model as shown in figure 1, geomembrane is an important part of landfill liner system (zhang, weng & shao, 2017). in order to reduce the tension stress of geomembrane, an anchorage ditch is designed at intervals on the slope. the unbalanced shear stress on the upper and lower surfaces of geomembrane results in large tensile deformation of geomembrane (ru et al., 2018). the force acting on the geomembrane element is shown in figure 2. generally speaking, the shear strength of the interface between geomembrane and bottom clay is the weakest in the cushion system (lin et al., 2017), and the accumulation of shear stress will form a larger tensile stress in the anchorage ditch. this is the focus of this study. some variables used in the analysis are shown in figure 3. the internal force of geomembrane in figure 3 is characterized by steep slope, and the overall stability of 1:3.3 rockfill dam slope is sufficient. however, the anti-sliding stability of the geomembrane and the cement figure 1. typical garbage liner system figure 2. stress analysis of geomembrane element figure 3. variable diagram 113study on the internal force of geomembrane of landfill in heavy metal contaminated area hyperbola: b m= 1 /  , m is the ultimate shear stress of the contact surface of the geomembrane., and τ σ ∅m n= tan 1. generally speaking, the foundation medium under geomembrane is clay cushion (wüthrich et al., 2018). assuming that its shear modulus is g and its thickness is d, the clay cushion can be used to estimate ks: k g ds = (2) when formula (2) is substituted for formula (1) and there is no dimensionalization, it can be obtained that: d u dx ru u a bx a b x 2 2 1 11  + = + + (3) where u h h l a k h r m k u u u x x r l k e x m s m s t = = = = = = ( ) τ − − ∅ β = ; ; ; * *; sin 1 2 1 2 1 2 −− − ∅− ∅ β ∅ ∅ ∅ ∅ 1 2 1 2 1 2 2 1 k r a k h b x x l ( )( ) = +( ) sin cos tan sin cos sin tan* == +( )( )                kx lcos sin sin cos tan tan 2 2∅ ∅ ∅− ∅ β ∅ (4) in the case of small slope angle, assuming that the vertical and horizontal directions are the principal stress directions of refuse dump (jin et al., 2017), the normal stress and active shear stress at the interface are respectively as: σ θ θ γ τ σ − σ − γ θ n x u v n x k h k h = +( ) = ( ) ( ) cos sin sin sin 2 2 1 2 2 1 2 1 2 (5) because the tensile stiffness of geotextile is one order smaller than that of geomembrane, it can be assumed that the shear stress at the geotextilegeomembrane interfaceis in the limit state, and the shear stress is as follows: τ σ ϕu t n tc= + tan (6) where, ct and 1 are the interface strength parameters of geotextilegeomembrane (choi et al., 2017). the active shear stress formula (6) and the interfacial shear stress formula (4) of geomembrane-clay are the smaller values of the overlying shear stress of geomembrane. most interface tests show that the strength of geomembrane-clay interface presents obvious softening characteristics (tang et al., 2017). in this paper, the tensile stress of geomembrane on landfill slope is analyzed by using the geomembrane-clay interface softening model as shown in figure 5. when the relative displacement of the interface reaches the ultimate displacement up, the maximum shear strength p is excited, and the shear strength will continue to decrease with the increase of the relative displacement. when the relative displacement reaches the residual displacement ur, the shear strength will stabilize at the residual strength r (if up = ur, the ideal elastic-plastic solution of the softening section cannot be considered). the shear strength and relative displacement of the plane can be divided into three stages: elastic stage, softening stage and residual stage. the interfacial shear strength conforms to the mohr-coulomb failure criterion (mcclung & ibáñez, 2017) shown in figure 6, in which cp, p and cr, r are the peak strength parameters and residual strength parameters of the geomembrane-clay interface, respectively. for the three stages of relative displacement of the above interface, formula (7) can be expressed as follows: mortar surface at its bottom depends on the bonding force between the membrane and the mortar (huang et al., 2017). the bonding force between the membrane and the mortar is greater than 0.1 mpa if the geomembrane is laid immediately before the cement mortar hardens and the sand bag is pressed on the membrane, so that the safety of the anti-sliding stability is sufficient. if only the friction force is used, the safety is insufficient. the anti-sliding stability between reinforced concrete slope protection and geomembrane depends on the cohesive force of asphalt (xu et al., 2017), which can reach 0.5 mpa, providing a high safety factor. usually in the design and construction, the composite geomembrane has not been produced yet, so the polyvinyl chloride film is used. some engineers attribute the landfill site to concrete face rockfill dam, which is very inappropriate, because the reinforced concrete mesh is not divided into blocks and without water stop sheet, it will inevitably lead to crack leakage. polyvinyl chloride (pvc) is an impervious body, so it is appropriate to belong to geomembrane rockfill dam. if it is attributed to the mixed dam type of geomembrane and concrete face slab, this dam type saves the expansion joint water stop, reduces the construction difficulty greatly, saves funds, and avoids the common problem of expansion joint leakage. analysis of internal force of geomembrane figure 4 is a schematic diagram of the yard structure. in this paper, all additional loads on geomembrane are regarded as a whole and expressed as a single material. assuming that there is no slip on the boundary surface of the upper layer of geomembrane, i.e. u > 1 (u, 1) is the friction angle on the contact surface of the upper layer and the lower layer of geomembrane respectively. the interaction between the interior of garbage and its intermediate contact layer such as the overlaying soil, is ignored, so that the estimated results tend to be safe and the analysis process will be obviously simplified. figure 4. diagram of landfill structure fundamental equation of the model the shear stress on the contact surface of the geomembrane is dependent on the shear-displacement relationship between the geomembrane and the soil (tang et al., 2017). it can be observed from drawing and shear tests that the shear stress  on the contact surface increases nonlinearly with the increase of shear displacement u. in this paper, a simplified hyperbolic relationship is adopted. 1 = + u a bu (1) in the formula, 1 is the shear stress of the contact surface of geomembrane’s lower layer ; a ks= 1 / , ks is the initial shear stiffness of 114 hongzhou zhang d u dx te ku u u d u dx te k u u u u u u p p p u p 2 2 2 2 1 1 = ( ) ≤ = ′ ( ) +  ≤ − τ − τ − τ < , ) , rr d u dx r u rte u u 2 2 1 = ( )            τ − τ >, (7) the solutions of formula (7) are respectively u ae a e u u u b bx b bx k u k u u u u ax ax p p u p p r = ≥ = + + ′ + ′ ⋅ ≤ 1 2 1 2 , , cos sin − τ − τ < == ( ) + +          τ − τ > r u rte x c x c u u 2 2 1 2 , (8) in the formula, the constants a1, a2, b1, b2, c1 and c2 depend on the boundary conditions; a k te= / and b k te= ′ / are stiffness ratio constants. the boundary conditions at both ends of geomembrane are respectively u du dxx x l  = = = = 0 0 0; (9) because the shear strength and relative displacement of geomembraneclay interface can be expressed in three stages (li, wu & choi, 2017), the geomembrane interface can also be divided into three stages from anchoring end to free end: elastic stage u up≤( ), softening stage u u up r ≤( ) and residual stage u ur( ). in this paper, the elastic stage length is defined as lp and the softening stage length is lt. when x = lp, u = up; when x = lp+lt , u = ur. according to the boundary conditions of formula (9), formula (8) can be rewritten as u e e e e k e e k p u u al ax u p u al ax al al u p p p p = +( ) + ( ) ( ) + τ − τ τ τ − τ − τ − τ − − − ,, 0 ≤ x lp< (10) u bl k bl b x l k u a p u t p p u = ( ) ( ) ′ ( ) + ′ + τ − − − − 1 cos cos sin τ − − < p p u p p p t k b x l k l x l l ′ + ( ) ( ) ′ ≤ +            cos (11) u te x l l te x l lr u p r r u p t= ( ) +( )    ( ) ⋅ +( )  τ − τ − − τ − τ − 2 2 2 + + ≤u l l x lr p r, < (12) it is concluded that the tensile stress of geomembrane is as follows. t te du dx a p u u al ax u p u p = = +( )τ − τ − − −− aal ax al al p p p pk e e x l ( ) ( ) ≤            − − − <0 (13) t te du dx b bl k bl b x l b r u p u t t p p u = = ( ) ( ) ′ ( ) + ( τ − τ − τ − τ − − τ − τ cos cos cos )) ( ) ′ ≤ +             sinb x l k l x l l p p p t − < (14) t te du dx te x l l l x l r u p t = = ( ) ( ) + ≤      τ − τ − < , (15) in the formula, lp and lt can be obtained by the internal force coordination conditions of elastic and transitional stages and the internal force coordina tion conditions of transitional (fils et al., 2018) and plastic stages, respectively. a e e k e e b bl p u al al u al al r u p u p p p p τ − τ τ − τ − τ − τ − τ − − ( ) +( ) + ( ) ( ) ( ) 2 cos tt t pk bl x l ′ =         sin , ´ (16) figure 5. softening model of geomembrane-clay interface figure 6. mohr-coulomb criterion for geomembrane-clay interface 115study on the internal force of geomembrane of landfill in heavy metal contaminated area b bl k bl te l l l x l l r u t p u t r u p t p t τ − τ − τ − τ τ − τ − ( ) ( ) ′ = ( ) +( ) = + cos sin ,      (17) finally, the values of lp and lt obtained by formula (16) and formula (17) are substituted for formula (11)-(13) to obtain the tensile stresses at all points on geomembrane. the tensile stress at x = 0 is the maximum tensile stress of geomembrane (zhang et al., 2018), and its expression is as follows: t ate e e k e e p u u al al al al p p p p max = ( ) + +( ) ( ) 2 τ − τ τ − − −´ (18) numerical solution of finite difference because formula (18) is a non-linear differential equation, the finite difference method is used to solve it numerically. dimensionless geomembrane length is divided into n elements, then the number of nodes is n i n+( ) = …1 0 1 2, , , , , , and ∆x n= 1 / , and a virtual node (n + 1) is added to the free end of geomembrane . for the second order differential, the difference scheme is used: ∆ ∆ ∆ 2 2 1 1 2 2u x u u u x i i i= + ( ) = + (19) for first order differential, the difference scheme is at node 1 = 0, ∆ ∆ ∆ u x u u u x = + 2 1 04 3 2 (20) at node i n= …1 2, , , , ∆ ∆ ∆ u x u u x i i= +1 1 2   (21) it can be obtained by finite difference method and boundary conditions at both ends of geomembrane. t n l r u u i n l r u u i n i m m i i = ( ) = ( ) = …       + τ − τ − 2 4 0 2 1 2 1 2 1 1 , , , , , (22) based on the above method, the tension and displacement of geomembrane can be obtained by numerical iteration (su et al., 2019), and the influence of various parameters of landfill slope on the displacement and tension of geomembrane can be analyzed by numerical method. in this paper, the main parameters affecting the displacement and tension of geomembrane, inclination q, length l, elastic modulus e and height h, are studied. the results are shown in figure 7-10. from figure 7, it can be seen that the displacement and tension of geomembrane are non-linear along its length. when other conditions are the same, both displacement and tension increase with the increase of inclination angle (zhang, dai, liu, et al., 2017), and when q is small, the change of q has a greater impact on the tension and displacement. from figure 7 to 10, it can be seen that the influence of the length, modulus of elasticity and height of geomembrane on the displacement and tension of geomembrane is similar to that of dip angle q. the peak strength τ σ ϕp p n pc= + tan of the interface between geomembrane and clay is the maximum strength that the interface can exert when the relative displacement of the interface is up, and the residual strength is formed when the relative displacement of the interface is greater than the residual displacement ur due to the rearrangement of clay particles on the interface or the polishing of the surface of geosynthetics. the figure 7. the influence of inclination angle q on displacement x and tension t of geomembrane figure 8. the influence of length l on displacement x and tension t of geomembrane 116 hongzhou zhang peak strength p and residual strength r of rough geomembrane-clay interface are usually quite different. therefore, the selection of interface strength is one of the key factors that affect the safety of geomembrane. when the geomembraneclay interface enters the residual state (figure 11), the shear stress of the geomembrane-clay interface from anchorage to free end is elastic, softening and residual respectively. figure 11 shows that the smaller the residual displacement ur is, the easier the residual strength of the geomembrane-clay interface will be. therefore, the influence of residual displacement ur and peak shear strength p on the tensile stress of geomembrane will be smaller (figure 12). at this time, the exertion of surface shear stress under geomembrane depends on the residual strength of the interface, when ur is larger, the length of the interface between geomembrane and clay in the softening zone is longer, and the influence of ur and p on the tensile stress of geomembrane is more obvious.in this case, if the softening of the interface strength between geomembrane and clay is not considered (luo et al., 2017), the maximum tensile stress of geomembrane obtained by the ideal elastic-plastic method according to the peak strength or residual strength of the interface will be inconsistent with the actual situation. as shown in figure 13, it is not safe to calculate the maximum tensile stress of geomembrane by the ideal elastic-plastic method (erhart & hirche, 2017). because the difference between residual displacement ur and peak displacement up is small, the influence of peak shear strength p of geomembrane-clay interface on the maximum tensile stress of geomembrane in figure 13 is also small. figure 14 and figure 15 show the tensile displacements and stresses at different positions of geomembrane under different residual strengths. it can be seen that the residual strength seriously affects the tensile displacements and stresses of geomembrane. figure 11. curve of shear stress variation with distance under overburden figure 12. effect of residual displacement on the maximum tensile stress of geomembrane figure 9. the influence of elastic modulus e on displacement x and tension t of geomembrane figure 10. the influence of height h on displacement x and tension t of geomembrane 117study on the internal force of geomembrane of landfill in heavy metal contaminated area figure 13. effect of peak shear strength on the maximum tensile stress of geomembrane figure 14. curve of relative displacement variation with distance figure 15. curve of tensile stress in geomembrane with distance simulation experiment the dam body is raised to 40m elevation at the original refuse dam and continued to be landfilled to 120m elevation. a horizontal anti-seepage system is added to the reservoir area in combination with the closure system of the first phase of the project and a leachate drainage system is added to the reservoir area of the first phase of the project. some of the original sewage treatment tanks are used, and a leachate regulating tank is expanded downstream of the original sewage interception dam. horizontal anti-seepage system consists of protective layer, leachate collection layer, main anti-seepage layer of geomembrane, secondary anti-seepage layer (gcl geosynthetical bentonite cushion) or clay liner (ccl), strengthening layer, clay protective layer, strengthening layer, gas conducting layer and isolation layer from top to bottom. the calculation parameters are as follows: the geotechnical composite drainage meshgeomembrane strength parameters are c = 0kpa  = 13, the interface parameters of geomembrane-geosynthetical bentonite cushion are cp = 2.2kpa, p = 21.6, cr = 1.0kpa, r = 12.8, the geomembrane-clay interface parameters are cp = 0kpa, p = 11, cr = 0kpa, r = 6.5 respectively, the landfill weight is maw = 10.2kn/ m3 and the slope gradients are  = 14, h = 12 ~ 40m, l = 10m, up = 0.012m, ur = 0.016m, t = 0.0015m, e = 180mpa kx = 0.56. the maximum tensile stress of geomembrane varies with the height of landfill as shown in figure 16. when the geomembrane is gcl, there is no tensile stress in the geomembrane because the shear strength of the surface under the geomembrane is greater than that of the surface above. when the geomembrane is lined with clay, the lower strength of the geomembrane-clay interface makes the geomembrane have great tensile stress. figure 16. the influence of landfill height on the maximum tensile stress of geomembrane conclusion based on the softening model of interface strength between geomembrane and clay, the equilibrium equation of geomembrane force is established, and the analytical solution of tension stress and displacement of geomembrane is obtained. through further parameter study, the following conclusions are drawn: (1) in the strength softening model of geomembrane-clay interface, the longer the softening section is, the greater the influence of the peak strength and residual displacement of the interface on the tensile stress of geomembrane is. on the contrary, the residual strength of the interface is the main controlling factor of the maximum tensile stress of the geomembrane when the softening stage is short. (2) when the strength of geotextile-geomembrane interface is close to or less than the residual strength of geomembrane-clay interface, the tensile stress of geomembrane is very small or even not. (3) slope gradient, anchorage length and height of overlying pile of geomembrane have great influence on the maximum tensile stress of geomembrane. (4) the maximum tensile stress of geomembrane can be effectively reduced by using rough hdpe geomembrane with high interfacial strength, choosing appropriate geotextile, reducing the anchorage length of geomembrane and limiting the height of pile on geomembrane. acknowledgments: langfang science and technology research and development program: landfill slope geomembrane internal force research in langfang area (2016013105) 118 hongzhou zhang references cen, w., du, x., geng, l., & he, h. (2018). seepage properties of geomembrane faced earth-rock dams under random multiple defects. advances in science and technology of water resources, 38, 60-65. chand, s., crémière, a., lepland, a., thorsnes, t., brunstad, h., & stoddart, d. (2017). long-term fluid expulsion revealed by carbonate crusts and pockmarks connected to subsurface gas anomalies and palaeochannels in the central north sea. geo-marine letters, 37, 215-227. choi, j., maniquiz-redillas m. c., hong, j., & kim, l. (2017). selection of costeffective green stormwater infrastructure (gsi) applicable in highly impervious urban catchments. ksce journal of civil engineering, 22, 1-7. coronado, g. d., rivelli, j. s., fuoco, m. j., vollmer, w. m., petrik, a. f., keast, e., narker, s., topalanchik, e., & jimenez, r. (2018). effect of reminding patients to complete fecal immunochemical testing: a comparative effectiveness study of automated and live approaches. journal of general internal medicine, 33, 72-78. erhart, s., & hirche, s. (2017). internal force analysis and load distribution for cooperative multi-robot manipulation. ieee transactions on robotics, 31, 1238-1243. fils, s. c. n., mimba, m. e., dzana, j. g., etouna, j., mounoumeck, p. v., & hakdaoui, m. (2018). tm/etm+/ldcm images for studying land surface temperature (lst) interplay with impervious surfaces changes over time within the douala metropolis, cameroon. journal of the indian society of remote sensing, 46, 131-143. gholamreza, s., mousa, a., & meguid, m. a. (2017). plausible failure mechanisms of wildlife-damaged earth levees: insights from centrifuge modeling and numerical analysis. canadian geotechnical journal, 54, 1496–1508. huang, w. c., ali, f., zhao, j., rhee, k., mou, c., & bettinger, c. j. (2017). ultrasound-mediated self-healing hydrogels based on tunable metal-organic bonding. biomacromolecules, 18, 1162-1171. jin, w., wu, z., wu, c., cao, z., fan, w., & tarolli, p. (2017). improving impervious surface estimation: an integrated method of classification and regression trees (cart) and linear spectral mixture analysis (lsma) based on error analysis. giscience & remote sensing, 55, 1-21. kaiser, p., schmoelz, w., schoettle, p., zwierzina, m., heinrichs, c., & attal, r. (2017). increased internal femoral torsion can be regarded as a risk factor for patellar instability a biomechanical study. clinical biomechanics, 47, 103-109. lin, b., yang, x. z., cao, x. w., zhang, t. z., wang, f. j., & zhao, j. (2017). a novel trichosanthin fusion protein with increased cytotoxicity to tumor cells. biotechnology letters, 39, 1-8. li, b., zhang, y., liu, x., li, z. (2017). study on the polyurea-coat debonding failure of impervious structure in contraction joints. journal of hydraulic engineering, 48, 70-77. li, w., wu, c., & choi, w. (2017). predicting future urban impervious surface distribution using cellular automata and regression analysis. earth science informatics, 11, 1-11. luo, y., nie, m., & xiao, m. (2017). flume-scale experiments on suffusion at bottom of cutoff wall in sandy gravel alluvium. canadian geotechnical journal, 54:1716-1727. mcclung, t., & ibáñez, i. (2017). quantifying the synergistic effects of impervious surface and drought on radial tree growth. urban ecosystems, 21, 1-9. ostfeldt, c., beguin, j. s., pedersen, f. t., polzik, e. s., muller, j. h., & appel, j. (2017). dipole force free optical control and cooling of nanofiber trapped atoms. optics letters, 2017, 42:4315-4318. parent, j. r., & qian, l. (2018). estimating percent impervious cover from landsat-based land cover with a simple and transferable regression model. international journal of remote sensing, 39, 3839-3851. ru, x., zhang, h., & hui, l. (2018). annual dynamics of impervious surfaces at city level of pearl river delta metropolitan. international journal of remote sensing, 39, 3537-3555. su, h., cui, s., wen, z., & xie, w. (2019). experimental study on distributed optical fiber heated-based seepage behavior identification in hydraulic engineering. heat and mass transfer, 55:421-432. tang, j., di, l., xiao, j., lu, d., & zhou, y. (2017). impacts of land use and socioeconomic patterns on urban heat island. international journal of remote sensing, 38:3445-3465. wüthrich, d., pfister, m., nistor, i., & schleiss, a. j. (2018). experimental study on the hydrodynamic impact of tsunami-like waves against impervious free-standing buildings. coastal engineering journal, 60, 180-199. xu, z., mountrakis, g., & quackenbush, l. j. (2017). impervious surface extraction in imbalanced datasets: integrating partial results and multitemporal information in an iterative one-class classifier. international journal of remote sensing, 38, 43-63. zhang, l., shan, b., zhao, y., & tang, h. (2018). comprehensive seepage simulation of fluid flow in multi-scaled shale gas reservoirs. transport in porous media, 121:263-288. zhang, l., weng, q., & shao, z. (2017). an evaluation of monthly impervious surface dynamics by fusing landsat and modis time series in the pearl river delta, china, from 2000 to 2015. remote sensing of environment, 201:99–114. zhang, w., dai, b., liu, z., & zhou, c. (2017). unconfined seepage analysis using moving kriging mesh-free method with monte carlo integration. transport in porous media, 116, 163-180. hugomonsalve.indd 85 earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 85-98 seismological observatory at quindio university (o.s.q.): a firm long-term step toward resolving regional seismicity monsalve jaramillo hugo1,2, correa montaño isabel cristina1,2, and valencia mina william1 1 ceifi, quimbaya research group quindío university 2 seismological observatory at quindío university osq. e-mail: hugom@uniquindio.edu.co resumen el eje cafetero (zona central de colombia, cultivada con café), y en particular, el departamento del quindío es constantemente golpeada por sismos de diferentes magnitudes. los sismos son el resultado de la compleja tectónica regional debido a la interacción entre las placa nazca, caribe y sur américa. la convergencia de estas placas es responsable de la formación de los andes. por estas razones, el osq (observatorio sismológico del quindío) e ingeominas, como las entidades estatales encargadas de monitorear la actividad sísmica, han de contribuir, a si como los centros regionales de investigación, al desarrollo de los estudios sismológicos y el análisis de la actividad tectónica local, para así obtener nuevos conocimientos a cerca de los desastres naturales que amenazan a las comunidades en la región y entender mejor la estructura interna de la corteza en la región del eje cafetero. la combinación de diferentes modelos de velocidad para la región del eje cafetero ha permitido la definición de elementos clave para tener una mejor percepción de la estructura de la corteza. en consecuencia, la determinación de un modelo de velocidades mejorado para la región cafetera ha ayudado a localizar mejor los sismos y a determinar con mayor precisión los espesores de la corteza y la estructura del manto superior entre los terrenos adyacentes que fueron generados en diferentes períodos geológicos. la magnitud ml fue determinada para los campos cercanos e intermedios y usando el modelo presentado en este articulo, el problema del rango dinámico de saturación de los sismógrafos durante sismos relevantes en campo cercano es solucionado. la magnitud local de algunos sismos superficiales fue calculada usando sismos con distancias epicentrales entre 10km y 300km asi como las magnitudes de algunos sismos profundos con distancias hipocentrales entre 100 km y 600 km. la evaluación del potencial sísmico en la región del eje cafetero empieza con el registro de la sismicidad reciente en chocó y en la parte norte del departamento del valle, y está basada en el análisis comparativo entre esta sismicidad y la registrada antes de la ocurrencia de los sismos mas importantes de los últimos 42 años. finalmente, considerando las dos fuentes de subducción, un sismo cuya magnitud de momento sísmico entre 6.7 y 7.3 podría esperarse en la próxima década, con una probabilidad de 87%. palabras clave: quindío, sismicidad, instrumentación. (manuscript received may 2005, paper accepted october 2005) 86 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina abstract the coffee belt (the central coffee-growing zone in colombia), and more particularly, the state of quindio is constantly hit by earthquakes of different magnitudes. earthquakes result from the complex regional tectonics, brought about by interaction between the nazca, caribbean, and south american plates. the convergence of these plates is responsible for the formation of the andes. for these reasons, the osq (sesimological observatory of quindio, from its acronym in spanish) and ingeominas, as government offices in charge of monitoring earthquake activity, have to contribute, along with research centers of the region, to the development of seismological studies and analysis of local tectonic activity in order to obtain new knowledge regarding natural disasters which are threat to the communities in region and be able to better understand the internal crustal structure in the coffee belt region. the combination of different velocity models for the coffee belt region has allowed the definition of key elements to have better insight about the structure of the crust. consequently, determining an improved new velocity model for the colombian coffee-growing region has helped to a better location of earthquakes and to a more accurate determination of differences in crustal thickness and the structure of the upper mantle between adjacent terrains that were built during different geological periods. the ml magnitude was determined for the near and intermediate fields, and, using the model presented in this article, the problem of dynamic range saturation of seismographs during relevant earthquakes in near field is overcome. the local magnitude of some shallow earthquakes was calculated using shocks with epicentral distances ranging between 10km and 300km as well as the magnitudes of some deep earthquakes with hypocentral distances ranging between 100km and 600km. seismic potential assessment in the coffee belt region begins with recent seismicity recorded in chocó and the northern part of valle (a state in colombia) and is based on comparative analysis between this seismicity and the one recorded before the occurrence of the most important earthquakes in the last 42 years. finally, taking into consideration the two subduction sources, an earthquake with seismic moment magnitude ranging between 6.7 and 7.3 could be expected within the next decade, having a probability of 87%. key words: quindío, sismicity, instrumentation. © 2005 ersj unibiblos introduction as a result of the emergency caused by the january 25, 1999, mw 6.2 quindio earthquake, quindio university and ingeominas worked jointly in the installation and operation of a temporary seismograph network, assigning this responsibility to the studies and research center of the engineering faculty (ceifi) which became, since february 1999, the seismological observatory of quindio university, using telemetry instruments, seismographs, and accelerographs borrowed from ingeominas in order to monitor the aftershocks. currently, there are eight accelerographs in operation and six telemetered seismograph stations (table 1), five of which are located in armenia and one in manizales (rodeo, which belongs to manizales volcanological and seismological observatory ovsm). since may 1999, when the observatory began operations, and up to now, the o.s.q. has been in charge of carrying out seismic monitoring activities in quindio associated to active faults in the area, however, other local and relevant worldwide seismic activity is also detected. in terms of knowing regional seismicity, the information collected by the local network has been of vital importance in understanding the local geodynamic processes related to urban planning and risk management activities in the region, identifying seismic sources, activity that ceifi has been conducting. in addition, the information that is analyzed and provided by ceifi is, and has been, of 87 seismological observatory at quindío university (o.s.q.): a firm long-term step toward resolving regional seismicity great support to different government officials involved in civil protection and disaster prevention activities in order to carry out programs of information diffusion and disaster management. table 1. osq station network. type statión departament city/town latitude longitude accelerographic (quindío university) uniquindío quindío armenia 4.506° -75.624° accelerographic (accelerographic (agreement with colombian national accelerographs network– rnac)) calarcá quindío calarcá 4.549° -75.618° poscosecha quindío armenia 4.506° -75.624° génova quindío génova 4.219° -75.775° dagua valle dagua 3.662° -76.708° calima valle calima-darien 3.884° -76.580° trujillo valle trujillo 4.228° -76.334° toche valle toche 3.650° -76.090° seismological guayaquil quindío planadas 4.460° -75.625° la mina * quindío pijao 4.377° -75.727° rodeo tolima machin volcano 4.490° -75.392° caicedonia valle caicedonia 4.320° -75.814° riber * valle ribera alta 4.519° -75.889° modin * valle cartago 4.670° -75.875° (* not operating permanently) a seismograph network is a useful tool to know the inner structure of the earth, earthquake waves propagation patterns, earthquake-source mechanisms, as well as urban seismic microzoning and hazard assessment of a specific region. for this purpose, it is necessary to have qualified staff that is trained in processing and analyzing data, capable of improving their knowledge of seismic phenomena and generating new hypothesis regarding seismic behavior of the studied region. the installation of an observatory in a region enables gathering of valuable information to conduct studies and increase the knowledge of local seismicity as well as studies about the regional crustal structure. the observatory becomes thus a frim long-term step to help resolve regional seismicity osq goals given the fact that quindio department and the coffee belt region as well as other zones in the country are usually threatened by earthquakes, the seismological observatory has as top priority goal of offering logistic support to the research center for the analysis of tectonic activity in the state of quindio for which constant monitoring, recording, and broadcasting activities are being carried out 365 days a year. in order to generate new knowledge about the natural hazards to which the region is constantly exposed, research projects are developed. these projects allow knowing, assessing, and improving social development conditions and subsoil conditions to provide early advise on disaster prevention and the determination of potential areas for future urban development of engineering works. regional tectonic setting the state of quindio has geological units of oceanic and continental origin which date back to paleozoic and mesozoic ages (arquia complex, quebradagrande complex, ultrabasic rock fragments, and cordoba igneous complex) which are intruded into by igneous units from the late mesozoic and overlaid by tertiary sedimentary sequences. these units, which comprise the regional geological basement, are separated by big regional faulting and are covered by a big fluviovocanic unit of quaternary age, quindio and pereira fans, and by local quaternary units. the general geological and tectonic schemes are presented in figure 1 (espinosa a. 2005). 88 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina from the tectonic point of view, the state of quindio sits on the romeral fault system. the three main faults of the system, cauca–almaguer, silvia–pijao, and san jerónimo, go across the region with strike component n15e. the importance of those faults through the geological history of the zone has been unquestionable, but its activity during the quaternary has been inconsistent (espinosa a. 2005). seismic activity in this zone locates between isodepths of 80-160km within the wadati-benioff zone of nazca plate, which subducts along the colombian northwest (monsalve, 1998), and also along numerous crustal faults. out of these faults, the most noticeable are cordoba, navarco, silvia-pijao, buenavista, el salado, cauca-almaguer, armenia, salento, montenegro, palestina, and cauca-patía faults which are active faults with probable magnitudes ranging between 6.1 and 6.9 mw (guzmán, 1998). taking into consideration all this tectonic complexity, the seismological observatory at quindio university (osq) contributes to the development of projects that generate new knowledge about the hazards that threaten the region, in order to improve plans and processes regarding both disaster mitigation and disaster prevention. figure 1. geological scheme of quindío department. (espinosa , 2005). 89 seismological observatory at quindío university (o.s.q.): a firm long-term step toward resolving regional seismicity seismicity of quindío departament 1999 2005 colombia is located in one of the most active seismic regions of the world dubbed “the pacific ring of fire”. colombia’s tectonics is complex because of the convergence of three plates: nazca, south american, and caribbean plates. these plates interact with one another originating compression, tension, and shear stresses within the region, which in time cause accumulation of energy. this interaction has originated three mountain ranges: the western, the central, and the eastern ranges, being the central one the oldest of all, and the eastern one the youngest. the most active zones in terms of seismic activity are located throughout the andean region of the country, along which the mountain ranges are settled; all this forms thus a complex morphological setting that is currently object of research. quindío department is located in the andean region of colombia, more exactly, to the western flank of the central mountain range. its boundaries are defined by risaralda department to the north, tolima department to the east, and valle del cauca to the southwest. for the geological complexities previously mentioned and the critical location, it is obvious that the coffee belt region is prone to significant shaking from earthquakes. quindío department is one of the most seismically active areas in colombia. this was confirmed by the occurrence of the january 25th quindio earthquake (1999), which took place at 18:19:16 ut (13:19:16 hl), with a magnitude mw=6.2 and a depth of 18 km. a mw 5.5 aftershock occurred at 22:40:12 ut (17:40:12 hl) the same day. the main shock ruptured along a left strike slip with normal component fault surface. this faulting located at shallow depth in the continental crust. its rupture length was 11.7 km and 10.6 km width; this yields a total rupture area of 124 km2. 391 well-identified and well-located aftershocks were used to constrain the rupture area. the aftershocks were recorded between depths of 9 and 20 km (monsalve y vargas 2002). this seismic event was the largest shallow earthquake instrumentally-recorded ever in the coffee belt region. its epicenter was located near cordoba (a municipality of quindio department) 17km from armenia. the death toll was 1,185; about 8,523 were injured, and more than 100.000 constructions were affected in 28 municipalities of three states (quindío, risaralda, caldas, tolima y valle) (monsalve et al, 2002). between february 1999, and august 2005, the osq has recorded 13.607 events (table 2). not all of them have been located due to the little coverage of the network. the events (figure 4), which have been located with the velocity model mentioned above, have yielded magnitudes ml ranging between 0.04 and 6.2, all of them having shallow depths. (for further information about the earthquakes location, see the location catalog on the osq web page: http:// www.uniquindio.edu.co menú investigaciones, ceifi, grupos, o.s.q ) between may 1st, 1999 and march 31st, 2000, 4,256 aftershocks were recorded. this produced a maximum daily rate of 58 events (october 21st) and an average of 15 events/day. we point out that during june 1999, a maximum number of 56 events took place and the minimum were registered during december 1999 march 2000, yielding an average of 14 and 13 earthquakes respectively (figure 3). in general, the occurrence of aftershocks continues with a fast exponential decrease rate as a function of time. the seismicity during 1999 remained at high levels most of the year as a result of the numerous aftershocks generated by january 25th earthquake (1999). during the year 2000 seismicity was low, except during april 10th 11th where 108 and 115 earthquakes/day were registered. because of their magnitude, the following events were remarkable during the year 2000 catalog: february 4th m l = 3.8 (córdoba fault) february 17th m l = 3.6 (armenia fault) february 19th m l = 3.6 (córdoba fault) march 11th m l = 4.0 (armenia fault) april 10th m l = 3.7 (córdoba fault) october 31st m l = 3.5 (armenia fault) 90 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina table 2. 1999 – 2005 earthquake catalog osq. 1999 seismicity 2000 seismicity 2001 seismicity 2002 seismicity month number of events month number of events month number of events month number of events january 2338 january 199 january 39 january 45 february february 237 february 236 february 30 march march 188 march 65 march 45 april april 547 april 186 april 100 may 478 may 139 may 378 may 128 june 637 june 53 june 302 june 17 july 645 july 281 july 217 july 24 august 541 august 250 august 159 august 49 september 362 september 218 september 181 september 51 october 470 october 234 october 106 october 28 november 260 november 205 november 123 november 26 december 227 december 15 december 150 december 32 2003seismicity 2004seismicity 2005seismicity month number of events month number of events month number of events january 34 january 12 january 19 february 30 february 26 february 123 march 24 march 21 march 107 april 25 april 77 april 67 may 29 may 29 may 49 june 38 june 24 june 331 july 17 july 46 july 225 august 25 august 12 august 221 september 15 september 14 october 27 october 101 november 43 november 255 december 9 december 12 91 seismological observatory at quindío university (o.s.q.): a firm long-term step toward resolving regional seismicity figure 2. number of local earthquakes registered by the osq, from 1999 to august 2005. figure 3. number of daily events between may 1st, 1999 and march 31st, 2000. maximum and minimum seismicity levels can be observed. seismicity during 2001 had its highest level during the period february may. the events occurred on january 16th 09:56 pm (local time), and march 2nd 06:46 and 06:47 (local time), were felt in pijao and felt in the southern part of armenia, respectively. the first of them had a ml magnitude of 3.1 and was located 8.3km southwest of armenia. during 2002 and 2003 seismic activity remained very low, with mild increase from may to june 2002, and june and november 2003. the three first months of 2004, seismicity levels were very low with a subtle increase in april with increases in activity on the 19th and on the 20th during which 76 local earthquakes whose magnitudes 92 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina ranged 0.6 and 2.9 ml. two of which were felt in armenia and calarcá. the first, april 19th at 04:04 local time (09:04ut), had a m l magnitude of 2.9. the second, same day (19:19 ut), reached a magnitude 2.3 m l. the following days seismic activity remained normal until september, where a significant increase in activity arose, with an important event on november 15th at 4:06 am local time with magnitude 6.7 in richter scale and epicenter located in the pacific ocean, 50.7 km to the southwest of bajo baudó (choco). until 2005, seismicity continues to be stable and it is worth mentioning the occurrence of an earthquake (june 22nd, 2005) at 03:07am local time (08:07 ut), with magnitude of 4.5 m l , 12.92km deep. this event was strongly felt in the whole quindío and in some municipalities of risaralda state. there were no casualties or material losses. more than 336 aftershocks were reported until june 30th, all of them having magnitudes noticeably inferior to the main shock. the aftershock occurred the same day at 09:25 am local time, with a magnitude m l of 2.9 which was felt in some towns near calarcá. this seismicity is associated to the tectonically active area of quindio department around cordoba municipality and is directly related to the activity that originated the january 25th earthquake in armenia in 1999. figure 4. events located by the osq since 1999 to 2005. six events with magnitudes m l equal to or greater than 4.0, from 1999 to 2005 are shown in figure 5. the osq recorded four of these six local earthquakes; the other two are relate to the seismic activity occurred on january 25th, 1999 and were located by the colombian national seismograph network (rsnc, from its acronym is spanish) in view of the fact that then, quindio university had not fully set up the seismological observatory and started its activities. 93 seismological observatory at quindío university (o.s.q.): a firm long-term step toward resolving regional seismicity figure 5. location of earthquakes with magnitudes m l greater than or equal to 4.0. table 3. localización sismos con magnitudes mayores o iguales a 4.0. year month day local time latitude longitude depth magnitude located by 1999 01 25 13:19 4.432 on -75.703 ow shallow (less than 30 km.) mw= 6.2 rsnc 1999 01 25 17:40 4.408 on -75.732 ow shallow (less than 30 km.) mw= 5.6 rsnc 1999 02 21 00:27 4.482 on -75.689 ow shallow 12 km. m l = 4.07 osq 1999 02 25 10:39 4.381 on -75.664 ow shallow 15.95 km. mw= 4.35 osq 1999 10 13 08:10 4.324 on -75.680 ow shallow 6.67 km. m l = 4.12 osq 2005 06 22 03:07 4,433 on -75,684 ow shallow 12,92 m l = 4.5 osq model for the location of seismic events in order to locate the earthquakes registered by the osq, the phase information recorded by different stations of ovsm and osso was required, and the model developed by (espinosa and monsalve, 2000) was used. this model is a one-dimensional velocity model derived for the coffee belt region using seismic tomography and well-located earthquakes between 4ºn and 6ºn within the coffee belt region with the software hypo71 and the non-linear ray inversion method velest. these earthquakes were later relocated using the joint hypocentral determination (jhd), (dewey, 1971) in order to verify the quality of the model generated for this zone (espinosa y monsalve 2002). the data used for velocity structure inversion included 566 earthquakes suitable for a good model (out of 1,337). the quality of the located events was 94 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina high: rms= 0.057 s, 12km deep, average magnitude of 1.25 and a gap of 158º. from a seismological point of view, a good velocity structure model is important to obtain location of events, as well as reliable source parameters to understand ground shaking during strong ground motion the results obtained with the one-dimensional model show a uniform distribution of velocities. the final result of the inversion using an initial model shows a velocity model of flat layers with a good resolution of superficial layers (0-100km depth). however, for deeper layers the resolution is not really good, because most earthquakes located at shallow levels and there are not many seismic rays that cross those regions (espinosa and monsalve 2002). considering the results of different inversions, we model the continental crust for the coffee belt region as consisting of six flat layers where the moho is situated, from 30 to 35km deep. the wave velocities up to 100km depth, are: 6.0km/s at depths of 6km, 7.0km/s at depths of 24km, 7.4km/s for a 22km-thick layer, 7.8km/s for a 14km-thick layer, and finally, a zone with velocities of 8km/s (figure6). table 4. depths, layer, initial and final velocity ranges and final model. the velocity range is based on four output models. the final model is defined as the lowest average rms (espinosa and monsalve 2002). layer depth range of initial velocities range of final velocities final velocity model 1 5.1 4.0-5.50 4.68-5.53 5.2 2 -6 5.5-6.10 5.51-6.78 6.1 3 -30 6.1-7.00 6.41-7.40 7.0 4 -52 6.5-7.50 7.40 7.4 5 -66 7.8-8.00 7.82 7.8 6 -100 8.0-9.00 8.00 8.0 this study allowed to establish a velocity model for the crust below the zone as well as to understand better the tectonic features of the region, which is important to have more accurate knowledge of the crustal structure in the colombian coffee belt and therefore to improve local earthquake hazard assessment studies. figure 6. final velocity models (blue line: average model (final)). (espinosa and monsalve 2002). calculation of local magnitude (m l ) using strong motion accelerograms besides earthquake location, it is really important to obtain the local magnitude for strong earthquakes using the accelerograms. for this purpose, the method proposed by barragán (2003) is used. this method was developed processing the most important time history records of shallow events (39 events) and events of intermediate depth, (40 events) that have occurred in the central region, los santanderes region, and the western region of the country which were provided by the colombian national accelerograph network (rnac), from 1994 to 2001. the accelerograph records are used as an acceleration input to the equation of motion of the wood-anderson torsion seismograph to produce synthetic seismograms which is then read in the standard manner thus permitting an easy and dependable calculation of the local magnitude ml of important events (barragán 2003). the colombian national accelerographs network (rnac) reported a magnitude m l 6.2 for quindio 95 seismological observatory at quindío university (o.s.q.): a firm long-term step toward resolving regional seismicity earthquake (january 25th, 1999), and with this method, a magnitude m l 6.22 was calculated using the complete procedure and records from 23 accelerographic stations; and a magnitude m l of 6.10 using a preliminary procedure with records of 26 stations (barragán l. f. 2003) which confirms the reliability and simplicity of the method. to have a summary of the process to calculate ml with the method proposed by barragán (2003), let us start by the data that is collected during normal processing activities; it includes instrumental and baseline correction of the records as well as their separation into three orthogonal components: vertical, east-west, and north-south. using the software degtra for windows version 1.4.7, maximum registered amplitudes in cm/s2 are obtained from the horizontal components (eastwest and north-south) and the formula (barragán, 2003) used for shallow or intermediate earthquakes is applied as follows: • for shallow earthquakes an empirical expression is proposed in order to directly determine local magnitude m l using horizontal ground accelerations m l = log 10 a ace + 1,091ln δ 0,1443 where a ace is the maximum amplitude in cm/s2 (from zero to maximum) registered by an accelerograph set on the bedrock for a particular earthquake at a specified distance and δ is the epicentral distance. • for intermediate depth earthquakes between 70 and 300km deep, another empirical expression for the direct calculation of local magnitude m l using horizontal ground acceleration m l = log 10 a ace + 1,059ln δ 0,1136 where a ace is the maximum amplitude in cm/s2 (from zero to maximum) registered by an accelerograph set on the bedrock for a particular earthquake at a specified distance and δ is the hypocentral distance in km. these equations are only applicable to horizontal strong motion components of accelerograph records. in this way, the local magnitude of earthquakes in the near field is calculated, using strong motion accelerograms and applying this to the regional geotectonic setting. earthquake potential in the coffe belt region earthquake potential was assessed based on the seismicity registered in northern valle and chocó. comparative analysis were done between this seismicity and the one that took place before the occurrence of the most important historical earthquakes that have affected the region during the last 42 years. all this information about the seismicity, registered from 1961 to 2003, was compiled based on the minimum magnitude threshold of ms> 4.4. this refers to seismicity that has been experienced in the region (figure 7) which, because of the lack of a national seismological network at the time, it was necessary to resort to the usgs-neic catalog. with this catalog, the comparative analysis was performed. this analysis comprised the seismicity that took place before the occurrence of the most important earthquakes that have struck the coffee belt region during the last 42 years. through this analysis, it was possible to find a time span (in years) of events recurrence that will repeat in the different sources any time within the next decade. knowing this, and the type of faulting mechanisms associated to the different zones, it is possible to obtain better insight of the earthquake potential in the region (monsalve y ospina 2004). according to the observation of these recurrence patterns an earthquake of magnitude (m w ) ranging from 6.5 to 7.2 could be expected at the subduction zone h=120km between filandia and circasia which would occur the next decade with a probability of 57% and intensities up to vii (ems-92 intensity scale). similarly, an earthquake of magnitude (m w ) ranging from 6.7 to 7.3 in the subducton zone h=60 km, within risaralda and chocó states would occur the next decade with a probability of 51% and intensities up to viii or greater in ems-92 scale. finally, considering the two subduction zones altogether, an earthquake of magnitude (m w ) from 6.7 to 7.3 could be expected the next decade with a probability of 87%. 96 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina figure 7. magnitude distribution for seismicity occurred between january 1960 and november 2003. important earthquakes dates during the last 42 years in the coffee belt region are in blue (monsalve and ospina 2004). in armenia the panorama of the last hundred years is quite clear. it can be observed that earthquakes with intensities equal to or greater than vii have a recurrence of about 20 years; there is a return period between 12 and 13 years for events with intensities equal to or greater than vi. it can also be observed that with recurrence of at least 5 years, earthquakes that may cause important damage have struck and will probably strike the region (monsalve y ospina 2004). figure 8. recurrence of historical events with i o (msk) > vi for armenia. modified by espinosa (1996). 97 seismological observatory at quindío university (o.s.q.): a firm long-term step toward resolving regional seismicity figure 9. earthquakes with seismic moment magnitudes 5.0< mw < 7.2 in the colombian northwestern corner (monsalve 1998). historically, there are several earthquakes occurring at deep and intermediate subduction sources (60220km deep). among them, we draw attention to the earthquakes that are remarkable because of their destructive potential. first, the one occurred on december 20th 1961 (mb=6.3, 176 km. deep), july 30th, 1.962 (mw=6.7, 108 km. deep), november 23rd, 1979 (mb=6.4, 108 km. deep), and february 8th, 1995 (mw=6.6, 100 km. deep); which have directly affected quindío and risaralda department (monsalve, ospina 2004). in figure 9, earthquakes with seismic moment magnitude 5.0< mw < 7.2 in northwestern colombian are shown (monsalve 1998). conclusions in earthquake-prone zones, such as the coffee belt region, it is of paramount importance to know and understand the response of structures to the high seismic activity and resulting hazard that this region faces. this can be achieved by specialized and detailed analyses which can lead to the improvement of earthquake hazard assessment studies in the region. all the studies carried out in the coffee belt region have provided people with knowledge and understanding geological and tectonic characteristics in the zone, thus making it possible to take a step further in the knowledge of crustal structure underlying this region. the answer to many seismological questions may be furnished by the installation of this local network. through the development of different projects that have been conducted since the osq started operations, new knowledge regarding the natural hazards that the coffee belt region is exposed to has been generated. these projects have allowed us to evaluate, and improve the social development conditions, and determine regional subsoil conditions to provide early control in terms of disaster prevention and the determination of potential areas for future urban development of engineering works. 98 hugo monsalve jaramillo, isabel cristina correa montaño and william valencia mina references • barragán, l.f. (2003). determinación de la magnitud local m l a partir de acelerogramas de movimiento fuerte. bachelor thesis, civil engineering. quindío university. • dewey, j.w. (1971), seismicity studies with the method of join hypocenter determination: ph. d thesis, university california berkeley. • espinosa, a., 1996. sismicidad histórica del área pereira-dos quebradas-santa rosa de cabal. informe final sismicidad histórica, seismic risk mitigation project pereirados quebradas and santa rosa de cabal. carder, pereira, 40 p., inédito. • espinosa, a., (2005). un modelo del cuaternario del quindío y risaralda. journal of the colombian academy of physical, natural, and exact sciences, bogotá. • espinosa-ospina, l. and monsalve-jaramillo, h. (2002). determinación de un modelo unidimensional de velocidades para el eje cafetero colombiano utilizando tomografía sismica. bachelor thesis, civil engineering. quindío university. • ingeominas (1999). terremoto del quindío (enero 25 de 1999). technical-scientific report. bogotá. • mc court, w.j., aspden, j.a., brook, m., (1984a). new geological and geochronological data from the colombian andes: continental growth by multiple accretion. j. geol. soc. london, 141 (5), 831 – 845. • monsalve, h., 1998. geometría de la subducción de la placa nazca en el noroeste de colombia: implicaciones tectónicas y sísmicas, master thesis, instituto de geofísica, universidad nacional autónoma de méxico, 106 p. • monsalve-jaramillo, h. and ospina-ostios, l.m. (2004). potencial sísmico de la zona de subducción en el quindio y risaralda (colombian coffee belt region). • monsalve-jaramillo, h. and vargas-jiménez, c.a. (2002). el sismo de armenia (25-01-99, mw=6.2): análisis telesísmico y características de las ondas de cuerpo. journal ipgh, no 57, 21-56. • monsalve, h., castaño, p., díaz, j., fernández, a., pérez, a., calero, l. and pinzón, g (2002). la reconstrucción del quindío lecturas desde la academia. quindío university. armenia – colombia. • parís, g., and marulanda, n., (1975). mapa geológico de la parte occidental de la plancha 387 – bolívar, scale 1:100.000. ingeominas – popayán, inédito. • trenkamp, r., mora, h. and salcedo, e. (2004). possible rapid intrablock strain accumulation rates near cali, colombia determined from gps measurements (19962003). first latin american congress of seismology. armenia (colombia). • vargas, c.a. and carvajal, c.a. (1998). estudio de refracción sísmica de armenia-quindio. journal red sismológica regional del eje cafetero viejo caldas y tolima. 3:16-23. issn:0123-9074. keywords: west liaoning province; zhangwuheishan area; mesozoic volcanic rock; isotopic age; geological significance. palabras clave: provincia liaoning; área zhangwuheishan; rocas volcánicas del mesozoico; edad isotópica; significancia geológica. how to cite item hou, c., cai, h., & pei, s. (2020). the shrimp u-pb isotope dating of mesozoic volcanic from zhangwu-heishan area, west liaoning province, china. earth sciences research journal, 24(4), 409-417. doi: https://doi. org/10.15446/esrj.v24n4.91114 the zhangwu-heishan area is located in the east of the fuxin-yixian basin. besides the quaternary soil, the study area is mostly covered with volcanic rock. the horizon and age of volcanic rock play an essential role in understanding fossil beds, structures, and sedimentary evolution of west liaoning province and coal seeking. during this work, 11 volcanic rock samples were measured by shrimp u-pb isotope analysis. based on the reported data on the age of the mesozoic volcanic rock in west liaoning province, in combination with new measurement data, cretaceous volcanic activities in west liaoning province can be divided into five stages, namely 132±1 ma, 126±1 ma, 122±2 ma, 115±2 ma, and 100±5 ma. based on statistical results, this paper concluded that the thinning time of the crust in northeast china is from 132±1 ma to 115±2 ma. abstract the shrimp u-pb isotope dating of mesozoic volcanic from zhangwu-heishan area, west liaoning province, china datación isotópica u-pb a través de la microsonda de iones de alta-resolución en rocas volcánicas del mesozoico para el área zhangwu-heishan, en la provincia de liaoning, en el occidente de china issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.91114 el área zhangwu-heishan se ubica en el este de la cuenca fuxin-yixian. el suelo del cuaternario en el área de estudio está cubierto por rocas volcánicas. el horizonte y la edad de la roca volcánica juegan un rol esencial en la comprensión de las capas fósiles, las estructuras y la evolución sedimentaria de la provincia liaoning, en el occidente de china, y en la búsqueda de carbón. en este trabajo se estudiaron 11 muestras de roca volcánica a través del análisis isotópico shrimp u-pb (datación u-pb a través de la microsonda de iones de alta-resolución). de acuerdo con la información reportada en las rocas volcánicas de edad del mesozoico al oriente de la provincia liaoning, en combinación con información de nuevas mediciones, las actividades volcánicas del cretácico en la zona se pueden dividir en cinco etapas: 132±1 ma, 126±1 ma, 122±2 ma, 115±2 ma, y 100±5 ma. con base en los resultados estadísticos, este estudio concluyó que el tiempo de adelgazamiento de la corteza en el noroeste de china está entre 132±1 ma y 115±2 ma. resumen record manuscript received: 18/11/2019 accepted for publication: 17/08/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 409-417 chaoyong hou*, houan cai, senlong pei china non-ferrous metals resource geological survey, beijing 100012, china *corresponding author: 45139139@qq.com g e o l o g y https://doi.org/10.15446/esrj.v24n4.91114 https://doi.org/10.15446/esrj.v24n4.91114 https://doi.org/10.15446/esrj.v24n4.91114 mailto:45139139@qq.com 410 chaoyong hou, houan cai, senlong pei introduction due to the discovery of precious fossils, west liaoning province has been drawing extensive interests in recent years (chen et al., 1998; sun, 1998; swisher, 1999; xu, 2001; zhou, 2003; zhou, 2003). meanwhile, the researches on the stratigraphy, paleo-geography, paleoclimate, and paleomagnetism of the mesozoic volcano-sedimentary basin in west liaoning province are gradually performed (pan et al., 2001; zhang et al., 2002; zhu et al., 2001 & 2002; gong et al., 2007; li et al., 2002; ren, 1997). as an essential terrane in north china where meso-cenozoic volcanic activities were developed, outstanding achievements have also been made in the researches on the chronology (figure 1) and geochemistry of volcanic rock in west liaoning province (jia et al., 1999; zhou, 2001; zhang & zheng, 2003; yang, 2007; wu et al., 2003). whereas shrimp zircon u-pb isotopic dating of volcanic rock in west liaoning province mainly concentrates on yixian basin, there are only sporadic isotopic reports dating of volcanic rock in zhangwu-heishan area (chen et al., 1997; huang, 2006 & 2007; xiao, 2008). west liaoning province is located in the east section of the yanshan platformal fold belt in the north of sino korean paraplatform and belongs to the east extension of yanshan orogenic belt (figure 1). yanshan orogenic belt lies inside the north china platform and has experienced intense tectonic deformation, magmatic and volcanic activities, and syn-orogenic sedimentation jurassic-cretaceous yanshan movement, and finally becomes a typical intraplate orogenic belt (zhang ch, 1999) far away from any contemporary plate boundary. the subduction of the pacific plate has resulted in ne-trending thrust nappe structure of late jurassic epoch and extensional structure of the early cretaceous epoch. before the mesozoic period, the study area is a large tectonic uplift area. at the first stage of the yanshan movement, the fault-folding formed lower-middle jurassic series in the nne-trending depression area, including coal-bearing strata beipiao formation. at the yanshan movement’s medium setting, a large-scale fracture occurred and was accompanied by a magmatic eruption, and then a mesozoic tectonic framework was formed. denomination and age of volcanic rocks although numerous detailed stratigraphic researches have been carried out in west liaoning province, the knowledge of stratigraphic sequence, the horizon of volcanic rock, and volcanic eruption periods are unified yet. for instance, whether volcanic rock in yixian formation is of late jurassic epoch (wang, 1987; ren et al., 1997; wang, 1998; luo et al., 1999), late jurassic to early cretaceous epoch (zhang et al., 1992; chen et al., 1998; ji et al., 1998; sun & zheng, 2000; jiang et al., 2000; wang et al., 2002) or early cretaceous epoch (li et al., 2001; wang et al., 1999; wang et al., 2001; xu et al., 1999; ji et al., 2002; wang et al., 2005; zhang et al., 2006) is an open question. for more details of the stratal division in this area and volcanic rock’s denomination, see the references wang et al., 2004; cai, 2009). it should be noted that the scientific research institute of liaoning bureau of coalfield geological exploration (1970) advocates that extensive volcanic rock in this area belongs to dalinghe formation and is of cretaceous epoch and concluded that there might be coal seams of fuxin formation under this suite of volcanic rock. likewise, the 107th coalfield geological brigade of liaoning province held in 1974 suggested that coal-bearing series of badaohao formation could not be correlated with shahai formation or fuxin formation in fuxin basin, and gangtaishan volcanic rock overlies badaohao formation and also drew the conclusion that the coal possibly could be found below those volcanic rocks. isotopic dating and eruption period of volcanic rock isotopic dating of volcanic rock in west liaoning province starts in the 1980s (diao, 1983; wang, 1984 & 1985). some fundamental findings have been made since the 1990s (for the data on isotopic dating, see figure 2). by far, over 120 pieces of data derived from various dating methods have been published. these data are systematically collected, and the distribution map (see figure 2) was plotted in this work. from the figure, the age of volcanic rock figure 1. geological map of zhangwu-heishan area and sampling point locations 411the shrimp u-pb isotope dating of mesozoic volcanic from zhangwu-heishan area, west liaoning province, china (zhang, 2003; zhang, 2005). the time when yixian formation was formed is also the critical period when the earth’s crust (lithosphere) in northeast china was thinned, so the age of top and bottom boundaries of volcanic rock in yixian formation is the critical period when the earth’s crust (lithosphere) in west liaoning province, even northeast china, was thinned. conclusions the horizon and age of volcanic rock play an essential role in understanding fossil beds, structures, and sedimentary evolution of west liaoning province and coal seeking. during this work, 11 volcanic rock samples were measured by shrimp u-pb isotope analysis. based on the reported data on the age of the mesozoic volcanic rock in west liaoning province, in combination with new measurement data, cretaceous volcanic activities in west liaoning province can be divided into five stages, namely 132±1 ma, 126±1 ma, 122±2 ma, 115±2 ma, and 100±5 ma. based on the statistical result, this paper concluded that the thinning time of the crust in northeast china is from 132±1 ma to 115±2 ma. which crops out in west liaoning province varies between 56 ma and 147 ma. yixian formation of early cretaceous epoch in the yixian basin contains four suites of volcanic rock and sedimentary rock (ji, 2004), which represent four volcanic-sedimentary activities. as shown in table 1, the periods of volcanic activities are 132±1 ma, 126±1 ma, 122±2 ma, and 115±2 ma. also, a suite of alkaline basalt was formed in the period of 90-105 ma. this is identical to the east part of north china (zhai et al., 2003). there is also a paleogene olivine-inclusion-bearing basalt suit with the age of about 56ma in the zhangwu-heishan area. in 2008 the international commission on stratigraphy adjusted the jurassic-cretaceous boundary from originally 135 ma (140 ma) to 145 ma±4 ma and the age of lower cretaceous–upper cretaceous limit from 95 ma to 99.6 ma. therefore, previously delineated volcanic rock of late jurassic epoch is now of early cretaceous epoch, and the peak period of volcanic activities is the medium stage (130 ma-122 ma or so) of early cretaceous epoch, namely barremian to aptian stage. it is an indisputable fact that the lithospheric mantle in north china craton was thinned on a large scale in the mesozoic and cenozoic periods (gao et al., 2004; liu et al., 2004). the earth’s crust in west liaoning province was also thinned when volcanic rock in the yixian formation was formed figure 2. summarizing of mesozoic and paleogene volcanic rock dating in western liaoning note: data source includes: diao, n. c.; wang, d. f.; guo, h.z.; zhang, z.c.; smith; li, p. x.; wang, s. s.; luo, z. k.; wang, d. y.; zhu, r. x.; zhu, q. j.; zhang, x. h.; peng, y. d.; zhang, h. f.; zhen, j. p.; wang, w. l.; ji, q.; yang, w.; zhang, h.; li, d. m.; huang, h.; shao, j. a.; xiao, g. q.; wang, q. l.; meng, f. x., and this paper. 412 chaoyong hou, houan cai, senlong pei table 1. rock samples and dating results of volcanic in fuxin and heishan-zhangwu area no sample location lithology test method age/ma data source 1 zhuanchengzi in yixian andesite, basalt k-ar 139.2±6.1 diao et al. 2 zhuanchengzi in yixian andesite, rhyolite k-ar 117.1±5 diao et al. 3 zhuanchengzi in yixian andesite rb-sr 125.9±5.3 wang et al. 4 zhuanchengzi in yixian basalt-basaltic andesite k-ar 136.9 wang et al. 5 zhuanchengzi in yixian rhyolite rb-sr 142.5±4 wang et al. 6 wendi countryside rhyolite rb-sr 126±5 wang et al. 7 rhyolite, andesite rb-sr 127±3 guo et al. 8 dakangbo in yixian andesite, dacite rb-sr 127.0±3 zhang et al. 9 jingangshan in yixian dolerite ar-ar 120.8-121 smith et al 10 jiashangou in beipiao basalt ar-ar 121.2±0.3 smith et al 11 huanghuashan in yixian volcanic breccia ar-ar 121.5±0.9 smith et al 12 zoujiagou in yixian basalt ar-ar 121.6±0.4 smith et al 13 central yixian formation zircon u-pb 136.7 luo et al. 14 yixian andesite ar-ar 129.2 chen et al. 15 zhuanchengzi in yixian phosphorite u-pb 136±7 chen et al. rb-sr 130±8 chen et al. 16 hejiagou in yixian andesite ar-ar 119.0±10.0 chen et al. 17 zhujiagou in yixian andesite ar-ar 124.5±4.9 chen et al. 18 dehuiyingzi in zhangwu trachyandensite, dacite ar-ar 123.7±4.4 chen et al. 19 dingjiafang in zhangwu andesite ar-ar 125.9±2.0 chen et al. 20 yemaotai in zhangwu dacite ar-ar 112.0±0.9 chen et al. 21 yemaotai in zhangwu andesite ar-ar 115.0±0.5 chen et al. 22 paozi in fuxin county hyalinite andesite u-pb 116 chen et al. 23 sihetun, jiashangou tuff ar-ar 145.3±4.4 luo et al. 24 sihetun, jiashangou tuff ar-ar 144.8±1.5 luo et al. 25 sihetun, jiashangou tuff ar-ar 147.1±1.8 luo et al. 26 fuxin basalt k-ar 84.76±1.6 zheng et al. 27 sihetun tuff permeates feldspar ar-ar 124.6±0.3 swisher et al 28 sihetun tuff permeates feldspar ar-ar 125.0±0.18 swisher et al 29 hengdaozi sanidine ar-ar 125.0±0.19 swisher et al 30 dijiagou trachyte ar-ar 147.3±0.6 li et al. 31 dijiagou trachyandensite ar-ar 140.4±0.4 li et al. 32 xijianshan section andesite ar-ar 121.9±0.6 li et al. 33 xijianshan section dolerite k-ar 116.7±2.3 li et al. 34 jingangshan in yixian andesite k-ar 120.1±2.0 li et al. 35 xiaoyantun in yixian basalt k-ar 91.81±2.13 li et al. 36 xiaoyantun in yixian andesite k-ar 125.6±0.4 li et al. (continued) 413the shrimp u-pb isotope dating of mesozoic volcanic from zhangwu-heishan area, west liaoning province, china table 1. rock samples and dating results of volcanic in fuxin and heishan-zhangwu area no sample location lithology test method age/ma data source 37 dijiagou—zhuanchengzi basalt k-ar 131.2±2. 6 wang et al. 38 sihetun tuff u-pb 125.2±0.9 wang et al. 39 sihetun basaltic andesite ar-ar 128.2±0.8 wang et al. 40 sihetun basaltic andesite ar-ar 128.4±1.4 wang et al. 41 sihetun basaltic andesite ar-ar 128.4±0.2 wang et al. 42 sihetun dolerite ar-ar 122.3±0.5 wang et al. 43 sihetun dolerite ar-ar 122.1±1.2 wang et al. 44 sihetun dolerite ar-ar 121.8±1.3 wang et al. 45 paishanlou in fuxin dioritic porphyrite u-pb 126±2 luo et al. 46 paishanlou in fuxin granite porphyry u-pb 124±1 luo et al. 47 paishanlou in fuxin dioritic porphyrite u-pb 125±1 luo et al. 48 dashitougou in fuxin biotite granite u-pb 124±1 luo et al. 49 jianguo in fuxin basalt k-ar 92.12±2.08 wang et al. 50 sihetun basalt k-ar 124.16-133.59 zhu et al. 51 zhuangchengzi andesite k-ar 120.42-122.31 zhu et al. 52 zhuanchengzi in yixian andesite k-ar 122.31-120.42 zhu et al 53 jianguo in fuxin basalt ar-ar 105.5±0.5 zhu et al 54 jianguo in fuxin andesite ar-ar 100.4±1.6 zhu et al 55 songjia andesite rb-sr 140±6 zhu et al. 56 yiwulvshan miliolite ar-ar 219±4 zhang et al. 57 yiwulvshan miliolite ar-ar 116±2 zhang et al. 58 yiwulvshan miliolite ar-ar 127±3 zhang et al. 59 sihetun olivine basalt ar-ar 132.9±1.5 peng et al. 60 huangbanjigou liparitic tuff ar-ar 127.4±1.3 peng et al. 61 sihetun ellipsoidal lava ar-ar 126.1±1.7 peng et al. 62 sihetun olivine basalt k-ar 133.3±2.6 peng et al. 63 sihetun olivine basalt k-ar 133.6±2.6 peng et al. 64 sihetun olivine basalt k-ar 124.16±2.4 peng et al. 65 sihetun olivine basalt k-ar 124.42±2.4 peng et al. 66 sihetun olivine basalt k-ar 124.91±2.4 peng et al. 67 houyanzigou olivine basalt ar-ar 130.6±0.5 peng et al. 68 sihetun olivine basalt ar-ar 127.7±0.2 peng et al. 69 jianguo basalt k-ar 107.3 ± 1.6 zhang et al. 70 jianguo ring tephrite k-ar 94.2 ± 1.4 zhang et al. 71 jianguo basalt k-ar 97.1±1.4 zheng et al. 72 jianguo basalt k-ar 98.0±1.4 zheng et al. 73 zaocishan dacite porphyry ar-ar 122.09±0.29 wang et al. (continued) 414 chaoyong hou, houan cai, senlong pei table 1. rock samples and dating results of volcanic in fuxin and heishan-zhangwu area no sample location lithology test method age/ma data source 74 zaocishan dacite porphyry ar-ar 122.36 wang et al. 75 zaocishan dacite porphyry ar-ar 121.80±1.36 wang et al. 76 sihetun basalt u-pb 112±3 ji et al. 77 zhuanchengzi in yixian basalt u-pb 120±2 ji et al. 78 jingangshan in yixian dacite porphyry u-pb 126±0.5 ji et al. 79 zhuanchengzi in yixian andesite u-pb 132±1 ji et al. 80 sihetun tuff u-pb 124.4±1.6 yang et al. 81 huangbanjigou tuff u-pb 123.1±2.1 yang et al. 82 huangbanjigou tuff u-pb 122.1±1.4 yang et al. 83 jianguo basalttrachyandensite ar-ar 116.8±3.0 zhu et al. 84 fanjiagou in yixian rhyolite u-pb 118.9±1.4 zhang et al. 85 fanjiagou in yixian rhyolite u-pb 119.8±1.9 zhang et al. 86 huanghuashan breccia dacite porphyry ar-ar 122.1±0.3 zhang et al. 87 huanghuashan breccia dacite porphyry ar-ar 121.8±1.4 zhang et al. 88 sihetun basaltic andesite ar-ar 126.1±0.4 zhang et al. 89 sihetun basalt k-ar 124.2±2.4 li et al. 90 sihetun basalt k-ar 124.4±2.4 li et al. 91 sihetun basalt k-ar 125.0±2.4 li et al. 92 sihetun basalt k-ar 124.6±2.5 li et al. 93 sihetun basalt k-ar 124.8±2.3 li et al. 94 sihetun basalt k-ar 125.1±2.4 li et al. 95 sihetun basalt k-ar 124.9±2.4 li et al. 96 sihetun basalt k-ar 126.1±2.6 li et al. 97 sihetun basalt k-ar 129.3±2.5 li et al. 98 sihetun basalt k-ar 129.2±2.6 li et al. 99 sihetun basalt k-ar 133.3±2.6 li et al. 100 sihetun basalt k-ar 133.6±2.6 li et al. 101 sihetun basalt k-ar 133.1±2.6 li et al. 102 maozishan in lingyuan tuff u-pb 125.4±0.9 zhang et al. 103 maozishan in lingyuan rhyolite u-pb 125.3±0.9 zhang et al. 104 maozishan in lingyuan rhyolite u-pb 124.4±1.1 zhang et al. 105 guojiagou in lingyuan tuff u-pb 122.3±2.1 zhang et al. 106 dawangzhangzi in lingyuan rhyolite u-pb 123±1.7 zhang et al. 107 zhangwu andesite u-pb 126±2 huang et al. 108 lihenggou in yixian rhyolite ar-ar 113.2±1.6 shao et al. 109 daxingzhuang in linghai amphibole andesite u-pb 114.9±2.5 shao et al. 110 dasijiazi in zhangwu rhyolite u-pb 122.4 ±0.4 ma xiao et al. (continued) 415the shrimp u-pb isotope dating of mesozoic volcanic from zhangwu-heishan area, west liaoning province, china table 1. rock samples and dating results of volcanic in fuxin and heishan-zhangwu area no sample location lithology test method age/ma data source 111 kazuo butte andesite k-ar 125.7 ±4.3 wang et al. 112 kazuo butte andesite u-pb 126.0±1.5 wang et al. 113 lingyuan rhyolite u-pb 124.4±1.4 meng et al. 114 badaohao in heishan trachyandensite u-pb 125.7±1.7 this paper 115 zhangluotun in heishan rhyolite u-pb 124.9±2.9 this paper 116 yaocangtu in heishan coloradoite u-pb 115.5±1.5 this paper 117 nalishan in yaocangtu basalt u-pb 56±2.9 this paper 118 liangshan in heishan trachyandensite u-pb 129.7±1.6 this paper 119 beizhuanchengzi andesite u-pb 132.3±2.3 this paper 120 huanghuashan in yixian dacite u-pb 126.3±1.7 this paper 121 juliangtun in yixian trachyandensite u-pb 125.8±1.9 this paper 122 lijiatai andesite u-pb 127.0±5.1 this paper 123 xinglonggou in yixian andesite u-pb 122.3±4.5 this paper 124 songbahu in yixian andesite u-pb 129.0±2.4 this paper references bing, z. b., liu, w. h., huang, z. a. (2003). stratigraphic sequence and age of the upper cretaceous daxingzhuang formation in western liaoning province (in chinese with english abstract). geological bulletin of china, 2003, 22(5), 351-355. cai, h. a. (2009). research on the sequence stratigraphy of badaohao coal basin, volcanics horizon in west liaoning province and the coal-searching prospect. phd dissertation of china university of geosciences (in beijing), 1-115 (in chinese with english abstract). chang, z. l., wang, y. q, & wang, x. l. (1999). the late mesozoic strata and discussion on related questions. proceedings of the seventh annual symposium of the chinese society of vertebrate paleontology. beijing: geological publishing house, 13-26 (in chinese with english abstract). chen, p. j., dong, z. m., & zhen, s. n. (1998). an exceptionally well-preserved theropod dinosaur from the yixian formation of china. nature, 391, 147-152. chen, w., zhang, y., & ji, q. (2005). age of the valuable fossiliferous beds of the shihetun village-new evidence of ar/ar and shrimp u-pb dating (in chinese with english abstract). journal of stratigraphy, 29, 582588. chen, y. x., & chen, w. j. (1997). mesozoic volcanic rocks in western liaoning and adjacent area. beijing, seismological press. 297 p. (in chinese). diao, n. c., li, x. z. (1983). the k-ar isotopic dating of mesozoic volcanic rocks in western liaoning (in chinese with english abstract). journal of geological society liaoning province, 1, 71-83. gong, e. p., liang, j. h., & li, x. d. (2007). comprehensive research on the sedimentary environment and palaeogeography of early cretaceous jehol bita in western liaoning, china (in chinese with english abstract). acta geologica sinica, 81(1), 1-8. guo, h. z., & zhang, z. c. (1992). essential characteristics of mesozoic volcanic rocks in western liaoning (in chinese with english abstract). acta petrologica et mineralogica, 11(3), 193-204. huang, h., gao, s., & hu, z. c. (2007). geochemistry of the high-mg andesites at zhangwu, western liaoning: implication for delamination of newly formed lower crust (in chinese). science in china (series d), 37(10), 1287-1300. ji, q., currie, p. j., norell, m. a. (1998). two feathered dinosaurs from northeastern china. nature, 393, 753-761. ji, q. (2004). mesozoic jehol biota of western liaoning, china. beijing: geological publishing house, 95-159 (in chinese). jia, b. w., zhou, a. c., & gu, d. q. (1999). geochemistry and provenance analysis of late paleozoic volcanic event deposits in the west liaoning (in chinese with english abstract). acta sedimentologica sinica, 17(3), 473-477. jiang, f. x., chen, j. h., cao, m. z. (2000). a discussion on the age of the feathered dinosaurs bearing beds of liaoning, china (in chinese with english abstract). acta palaeontologica sinica, 39(2), 307-311. li, d. m., chen, c. l., & li, q. (2006). constrain the time of the paleocoenosis fossil stratum by k-ar age of sihetun volcanic rock sequence (in chinese with english abstract). acta petrologica sinica, 22(6), 16271632. li, p. x., chen, z. w., & pang, q. q. (2001). the horizon and age of the confuciusornis in beipiao, western liaoning (in chinese with english abstract). acta geologica sinica, 75(1), 1-13. li, w. p., lu, f. x., & li, x. h. (2001). the origin of early cretaceous volcanic rocks of yixian formation and crust-mantle interaction in west liaoning province, eastern china (in chinese with english abstract). journal of mineralogy and petrology, 21(4), 1-6. liu, y. s., gao, s., & yuan, h. l. (2004). u-pb zircon ages and nd, sr, and pb isotopes of lower crustal xenoliths from north china craton: insights on evolution of lower continental crust. chemical geology, 211(1-2), 87-109. luo, q. h., chen, p. j., & zhou, d. y. (1999). age of sinosauropterys and confuciusornis 40ar/39ar laser single-grain and k-ar dating of the yixian formation, ne china (in chinese with english abstract). geochimica, 28(4), 405-409 luo, y., ma, h. f., & wang, m. t. (2004). petrogeochemical characteristics and genesis of the mesozoic volcanic rocks in the yanshan-liaoning region (in chinese with english abstract). bulletin of mineralogy, petrology and geochemistry, 24(3), 112-117. luo, z. k., miao, l. c., & guan, k. (2001). shrimp u-pb zircon age of magmatic rock in paishanlou gold mine district, fuxin, liaoning province, china (in chinese with english abstract). geochimica, 30(5), 483-490. 416 chaoyong hou, houan cai, senlong pei meng, f. x., gao, s., & liu, x. m. (2008). u-pb zircon geochronology and geochemistry of volcanic rocks of the yixian formation in the lingyuan area, western liaoning, china (in chinese with english abstract). geological bulletin of china, 27(3), 364-373. pan, y. x., zhu, r. x., & john, s. (2001). magnetic polarity ages of the fossilbearing strata at the sihetun section, west liaoning: a preliminary result (in chinese). chinese science bulletin, 46(8), 680-683. peng, y. d., zhang, l. d., & zhang, c. j. (2003). kar and 40ar/39ar ages determination of volcanic rocks of yixian formation, western liaoning (in chinese with english abstract). geochimica, 32(5), 427435. ren, d., guo, z. g., & lu, l. w. (2001). a further contribution to the knowledge of the upper jurassic yixian formation in western liaoning (in chinese with english abstract). geological review, 43(5), 449-459. shao, j. a., zhang, h. f., & liu, x. m. (2007). the chronology record of early masonic volcanic underplating on north margin of north china: u-pb dating evidence of masonic andesite from western liaoning, china (in chinese). progress in natural science, 17(5), 609-613. smith, p. e., york, d., & chang, m. m. (1995). dates and rates in ancient lakes: 40ar/39ar evidence for an early cretaceous age for the jehol group, northeast china. canadian journal of earth sciences, 32, 1426-1431. sun, g., dilcher, d. l., & zheng, s. l. (1998). in search of the first flower: a jurassic angiosperm, archaefructus, from northeast china. science, 282(5394), 1692-1695. swisher, c. c., wang, y. q., & wang, x. l. (1999). cretaceous age of the feathered dinosaurs of liaoning, china. nature, 400(1), 58-61. swisher, c. c., wang, x. l., & zhou, z. h. (2001). new evidence of isotope age of yixian formation and 40ar/39ar age dating of tuchengzi formation (in chinese). chinese science bulletin, 46(23), 2009-1012. tan, x. c. (1926). beipiao coalfield of rehechaoyang county (in chinese). geology report, 8(45-57). wang, d. f., zhou, s. g., & sun, c. l. (1985). ree determination and ree distribution patterns of meso-cenozoic volcanic rocks in centralwestern liaoning province and their geotectonic implications (in chinese with english abstract). geochimica, 2, 134-141. wang, d. f. (1983). variations in isotopic ages of cretaceous volcanic rocks in eastern china and isotopic dating of the lower boundary of the cretaceous system (in chinese with english abstract). geochimica, 4, 100-104. wang, d. y., xu, w. l., & feng, h. (2002). nature of late mesozonic lithospheric mantle in western liaoning province: evidences from basalt and the mantle-derived xenoliths. journal of jilin university (earth science edition), 32(4), 319-324. wang, g. h., zhang, c. h., wang, g. s. (2001). tectonic framework of western liaoning province and its evolution during mesozoic (in chinese with english abstract). geoscience, 15(1), 1-7. wang, q. l., chen, w., & zhang, y. (2009). laser 40ar/39ar and shrimp u-pb age of the andesite in the yixian formation, gushan area (in chinese with english abstract). geological review, 54(1), 125-133. wang, s. s., hu, h. g., & li, p. x. (2001). further discussion on the geologic age of sihetun vertebrate assemblage in western liaoning, china: evidence from ar-ar dating (in chinese with english abstract). acta petrologica, 17(4), 663-668. wang, w. l., zhang, h., & zhang l. j. (2004). standard section of tuchengzi stage and yixian stage and their stratigraphy, palaeontology and tectonic-volcanic actions. beijing: geological publishing house, 343434 (in chinese with english abstract). wang, x. r., gao, s., & liu, x. m. (2005). geochemistry of the high-mg andesites at sihetun, western liaoning: implication for delamination of lower crust and sr/ychange. science in china (series d), 35(8), 700-709 (in chinese). wu, g., zhang, h., & zhang, l. j. (2003). geochemical characterstics and genesis of the early mesozoic volcanic rocks in west liaoning area (in chinese with english abstract). journal of mineralogy and petrology, 23(3), 44-50. xiao, g. q., gao, s., & huang, h. (2008). zircon u-pb geochronology and geochemistry of mesozoic volcanic rocks from dasijiazi area at zhangwu, west liaoning province (in chinese with english abstract). earth science (journal of china university of geosciences), 33(2), 151-164. xu, w. l., wang, d. y., & wang, q. h. (2004). 40ar/39ar dating of hornblende and biotite in mesozoic in intrusive complex from the north china block: constraints on the time of lithospheric thinning (in chinese with english abstract). geochimaca, 33(3), 221-231. xu, w. l., zheng, c. q., & wang, d. y. (1999). the discovery of mantle-and lower crust-derived xenoliths in mesozoic trachybasalts from western liaoning ant their geological implications (in chinese). geological review, 45(supp.), 444-449. xu, x., zhou, z. h., & prum, r. (2001). branch edintegumental structures in sinornithosaurus and the origin of feathers. nature, 410, 200-204. yang, w. (2007). geochronology and geochemistry of the mesozoic volcanic rocks in western liaoning: constraints on mechanism for the lithospheric thinning in the north china craton. phd dissertation of university of science and technology of china, hefei, anhui province, china, 1-121 (in chinese with english abstract). zhai, m. g., zhu, r. x., & liu, j. m. (2003). the key time of tectonic transform in mesozoic in eastern north-china (in chinese with english abstract). science in china (series d), 33(10), 913-920. zhang, h., liu, x. m., & li, z. t. (2005). crust-thinning in large scale in fuxinyixian basin and adjacent in western liaoning in early cretaceous as well as its forming reason (in chinese with english abstract). geological review, 51(4), 360-372. zhang, h., liu, x. m., & yuan, h. l. (2006). u-pb isotopic age of the lower yixian formation in lingyuan of western liaoning and its significance (in chinese with english abstract). geological review, 52(1), 63-71. zhang, h. f., sun, m., & zhou, x. h. (2003). secular evolution of the lithosphere beneath the eastern north china craton: evidence from mesozoic basalt and high-mg andesites. geochimical et cosmochimica acta, 67(22), 4373-4387. zhang, h. f., & zheng, j. p. (2003). the geochemical characteristics of mesozoic basalt in north china and the cause of the rock’s formation: take example for the rock in fuxin liaoning province (in chinese with english abstract). chinese science bulletin, 48(6), 603-609. zhang, l. d., guo, s. z., & zhang, c. j. (2002). the discovery of pillow lavas in yixian formation, western liaoning, and its significance (in chinese with english abstract). acta geosicientia sinica, 23(6), 491494. zhang, x. h., li, t. s., & pu, z. p. (2002). 40ar-39ar thermochronology of two ductile shear zones from yiwulvshan, west liaoning region: age constraints on the mesozoic tectonic events (in chinese). chinese science bulletin, 47(9), 697-701. zhang, z. c., li, z. d., & wang, f. b. (1994). essential characteristics of volcanic rocks and discussion on their genesis in yixian basin, western liaoning (in chinese with english abstract). geoscience-journal of graduate school, china university of geosciences, 8(4), 441-451. zheng, c. q., xu, w. l., & wang, d. y. (1999). the petrology and mineral chemistry of the deep-seated xenoliths in mesozoic basalt in fuxin district from western liaoning (in chinese with english abstract). acta petrologica sinica, 15(4), 616622. zheng, j. p., zhang, r. s., & yu, c. m. (2004). zircon u-pb dating, hf isotope and trace elements of monzonitic enclave in basalt of eastern hebeiwestern liaoning, and their tracing on the 169 ma and 107 ma thermal event in yanshan-liaoxi belt (in chinese). science in china (series d), 34(supp.), 32-44. zheng, p. j., griffin, w. l., & o’reilly, s. y. (2007). mechanism and timing of lithospheric modification and replacement beneath the eastern north china craton: peridotitic xenoliths from the 100 ma fuxin basalts and a regional synthesis. geochimica et cosmochimica acta, 71, 52035225. http://dlib3.edu.cnki.net/kns50/navi/bridge.aspx?linktype=baselink&dbcode=cjfd&tablename=cjfdbaseinfo&field=baseid&value=dqxb&navilink=%e5%9c%b0%e7%90%83%e5%ad%a6%e6%8a%a5 417the shrimp u-pb isotope dating of mesozoic volcanic from zhangwu-heishan area, west liaoning province, china zhou, x. h., zhang, g. h., & yang, j. h. (2001). sr-nd-pb isotope mapping of late mesozoic volcanic rocks across northern margin of north china craton and implications to geodynamic processes (in chinese with english abstract). geochimica, 30(1), 10-23. zhou, z., barrett, p. m., & hilton, j. (2003). an exceptionally preserved lower cretaceous ecosystem. nature, 421, 807-814. zhou, z. y., & zheng, s. l. (2003). the missing link in ginkgo evolution. nature, 423, 821-822. zhu, q. j., chen, j., & yue, w. z. (2002). the application of isotopic analysis to the tectonic evolution of songjia depression in liaohe oilfield (in chinese with english abstract). journal of precious metallic geology, 11(4), 228-232. zhu, r. x., pan, y. x. & shi, r. p. (2002). new cretaceous palaeointensity data and the constraints on geodynamics. science in china (series d), 45(10), 931-938. zhu, r. x., pan, y. x., & shaw, j. (2001). geomagnetic palaeointensity just prior to the cretaceous normal superchron. physics of the earth and planetary interiors, 128(1-4), 207-222. keywords: hydrological wetland; soft rock soil; foundation; consolidation; deformation characteristics; settlement. palabras clave: humedal hidrológico; suelo de roca blanda; fundación; consolidación; características de deformación; asentamiento. how to cite item liu, b., & jiang, x. (2020). consolidation and deformation characteristics of soft rock foundation in a hydrological wetland environment. earth sciences research journal, 24(2), 183-190. doi: https://doi.org/10.15446/ esrj.v24n2.87920 the widespread distribution of soft rock and soft soil in a hydrological wetland environment is a common geotechnical engineering problem encountered in coastal engineering construction. a study method for consolidation and deformation characteristics of soft rock and soft soil foundation in a hydrological wetland environment is proposed to solve this problem. the study area was limited to the k9+280-k11+120 section along the fu-nehe section of national highway 111. the consolidation and deformation characteristics and loading conditions of soft soil foundation under embankment filling load, treatment methods of the soft rock foundation, stratum conditions, temperature changes, and time effects are analyzed. the results show that although the wetland soft rock and soil layer is not thick, the settlement of soft rock and soil accounts for more than 80% of the total settlement. the negative temperature has a certain influence on the consolidation settlement of soft rock foundation, which is mainly manifested in the difference between the settlement process of the central separation zone and the roadbed soft soil foundation. the pore water pressure of soft rock foundation dissipates to varying degrees. according to the monitoring results of settlement and pore water pressure, bagged sand wells are more suitable for soft rock foundation engineering treatment in a hydrological wetland. the research results can provide a reference for the study, calculation, and design of the consolidation and deformation of soft rock foundation in a hydrological wetland. abstract consolidation and deformation characteristics of soft rock foundation in a hydrological wetland environment características de consolidación y deformación de una fundación de roca blanda en el entorno de humedales hidrológicos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87920 la distribución generalizada de roca blanda y suelo blando en el entorno de humedales hidrológicos es un problema común de ingeniería geotécnica en las zonas costeras de construcción. para resolver este problema se propone un método de estudio para las características de consolidación y deformación de rocas blandas y cimientos de suelos blandos en ambientes de humedales hidrológicos. como área de estudio se definió la sección k9+280-k11+120 a lo largo de la sección fu-nehe de la carretera nacional 111. se analizaron las características de consolidación y deformación y las condiciones de carga de los cimientos del suelo blando bajo la carga de relleno del terraplén, los métodos de tratamiento de los cimientos de rocas blandas, las condiciones del estrato, los cambios de temperatura y los efectos del tiempo. los resultados muestran que aunque la capa de suelo y roca blanda del humedal no es gruesa, el asentamiento de roca blanda y suelo representa más del 80 % del asentamiento total. la temperatura negativa tiene cierta influencia en el asentamiento de consolidación de los cimientos de rocas blandas, que se manifiesta principalmente en la diferencia entre el proceso de asentamiento de la zona de separación central y los cimientos de suelos blandos de la calzada; la presión de poro del agua de la base de roca blanda se disipa en diversos grados. de acuerdo con los resultados de monitoreo del asentamiento y la presión del agua de los poros, los sacos de arena son más adecuados para el tratamiento de ingeniería de cimientos de rocas blandas en humedales hidrológicos. los resultados de la investigación pueden proporcionar una referencia para el estudio, el cálculo y el diseño de la consolidación y la deformación de los cimientos de rocas blandas en humedales hidrológicos. resumen record manuscript received: 18/12/2018 accepted for publication: 11/03/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 183-190 bin liu1, xiugen jiang2* 1water conservancy project & civil engineering college, tibet agriculture & animal husbandry university, nyingchi 860000, china 2college of water resources & civil engineering, china agricultural university, peking 100083, china * corresponding author: xyfdgzc@sina.com g e o t e c h n ic a l e n g in e e r in g https://doi.org/10.15446/esrj.v24n2.87920 https://doi.org/10.15446/esrj.v24n2.87920 https://doi.org/10.15446/esrj.v24n2.87920 mailto:xyfdgzc@sina.com 184 bin liu, xiugen jiang introduction the shrinkage of soil under pressure is called the compressibility of soil (baral et al., 2018). the process of soil compression increasing with time and dissipation of excess pore water pressure is called consolidation of soil, which is generally divided into primary consolidation and secondary consolidation (huang et al., 2017). main consolidation refers to the compression process caused by the gradual dissipation of excess pore water pressure and the increasing effective force in soil (zhang, 2017), while secondary consolidation refers to the slow deformation caused mainly by creep of soil skeleton (tao et al., 2018) after the basic dissipation of excess pore water pressure. the characteristics of soft rock and soil deformation with time under load easily lead to uneven settlement and excessive post-construction settlement of roadbed built on soft rock foundation (zhang et al., 2019; zhang & wang, 2018). due to the different origins of soft rock and soil, their respective engineering properties also show some differences. hydrological wetland ecosystem is located in the transitional area between water ecosystem and terrestrial ecosystem (yang et al., 2017). specific hydrological conditions are the driving force of wetland formation and maintenance. small changes in hydrological characteristics will lead to changes in wetland ecosystem. therefore, hydrological wetland ecosystems are vulnerable to climate change and other human activities. the response and feedback mechanism of hydrological wetland ecosystem under the influence of global change and human activities is the focus and frontier issue of common concern in the international community (guo, 2018). consolidation and deformation characteristics of soft rock foundation in hydrological wetland environment are affected by many factors, such as physical and mechanical properties of soft rock and soil, sludge thickness of hydrological wetland, loading capacity, loading mode and field environment (xu et al., 2017; kotani et al., 2017). in the hydrological wetland area, in order to carry out the construction of soft rock foundation smoothly, cofferdams are built on both sides of most sections before foundation treatment, and counter-pressure barriers are built between cofferdams and roads. due to the influence of hydrological tide, cofferdam and back pressure guard, the consolidation degree of silt foundation in hydrological wetland area is difficult to meet the design requirements during the design preloading period. in order to study the consolidation and deformation of silt in this kind of area, a lot of field tests have been carried out. however, the parameters of soil layer, initial conditions and boundary conditions of field prototype tests are difficult to adjust, and the influence of each factor on the deformation characteristics cannot be accurately analyzed. with the development of computer technology, numerical calculation greatly strengthens the ability to solve complex geotechnical engineering problems. the numerical method is more and more used in soft foundation treatment projects, but the numerical analysis of consolidation and deformati on characteristics of silt roadbed in hydrological wetland area is rare. the overload preloading of bagged sand well is a common method in soft foundation treatment (zhu & wen, 2017). for the consolidation analysis of sand well foundation, it is a three-dimensional problem with too much calculation. generally, it is simplified to the plane strain problem of sand wall foundation (yan et al., 2017). in order to better reflect the actual situation of the project, in recent years, in the construction of high-grade highway, on-site monitoring has been carried out in many soft rock base sections. although the cost of onsite monitoring is high, which brings some inconvenience to the construction of the project, the monitoring results can be used to improve the calculation results and predict the post-construction settlement and final settlement of soft rock foundation. based on this, the research method for consolidation and deformation characteristics of soft rock foundation under hydrological wetland environment is put forward, which provides the main analysis basis for judging whether the safety and preloading of embankment construction progress in soft rock foundation of hydrological wetland have reached stability. materials and methods project survey the geological conditions along the fuyu-nehe section of national highway 111 are relatively complex. the test section is k9+280-k11+120, which is about 20 km away from the hometown of red-crowned crane zhalong nature reserve. the two layers of soft soil encountered are deposited by hydrological wetlands (ma et al., 2018). the first layer of soft rock is with the thickness of 0.3-0.5 m and is silty clay (guang-ze et al., 2018); the second layer of soft rock is with the thickness of 2-6 m and is silty clay, which has poor engineering properties. two layers of undisturbed soil samples of soft rock are tested in laboratory. the main physical and mechanical indexes are listed in table 1. table 1. basic physical properties of soft rock and soil determination index silty clay silty clay sampling depth/m 1.9—2.5 4.6—5.2 moisture content/% 60.01 53.03 density/g·cm-3 1.57 1.7 relative density 2.16 2.37 void ratio 1.21 1.13 saturation/% 100 100 liquid limit/% 58 50.27 plastic limit/% 30.57 26.78 liquid index 1.11 1.12 plastic index 27.43 23.49 in the test section, 8 sections are selected for field study, and the monitoring results of one of the typical sections are analyzed (k9 + 280 and the treatment method is to lay geogrids on bagged sand wells) (zhou et al., 2017). the geological conditions of the engineering in monitoring section and the buried position of the instruments and equipment are detailed as shown in figure 1. determination of permeability coefficient sand wells have changed the drainage boundary conditions of the original hydrological wetland foundation (hao et al., 2017), shortened the drainage distance, and the drainage conditions have been greatly improved compared with the natural foundation. radial drainage consolidation is the main factor in sand well foundation, and the vertical permeability coefficient can be obtained from the geotechnical test value. because of the heterogeneity of the sludge in the hydrological wetland area, the permeability coefficient obtained from indoor and outdoor tests is discrete, and the measured settlement curve can reflect the consolidation condition of the foundation comprehensively. the horizontal permeability coefficient (sun et al., 2017) can be calculated by using settlement data. the theoretical solution of average consolidation degree of soil under various drainage conditions can be summed up as a general expression: u e t= 1−α −β (1) where, t is time,  and  are parameters to be determined. according to the measured settlement-time curve, any three time t1, t2 and t3 after the load stops are selected and t3 ‒ t2 = t2 ‒ t1. it can be obtained that: β − − − = 1 2 1 2 1 3 2t t s s s s ln (2) in the formula, the settlement values at time t1, t2 and t3 are s1, s2 and s3 respectively. for the consolidation of shaft foundation (ideal shaft), there are: c d f n c c d h h e v h e = ( ) + β 2 2 2 2 8 4 π (3) 185consolidation and deformation characteristics of soft rock foundation in a hydrological wetland environment results analysis of layered settlement monitoring results figure 2 shows the change process of embankment filling height of test section. the starting time of abscissa is november 5. before the installation of the observation equipment, the test section has been filled with gravel drainage cushion and part of embankment fill, and geogrid has been laid. because of the influence of climate, the embankment has not been filled and the soft rock foundation has not been loaded during the whole winter (liu, 2018). the main loading time is concentrated in the following five, six and seven months, which lasts more than 90 days. the thickness of the embankment is 1.8 m and the height of the embankment is 4.0 m. figure 2. change process of embankment fill height in test section where, de is the diameter of equivalent circle, d l le = = 2 3 1 05 π . , l is the distance between sand wells, h is the thickness of unilateral drainage soil layer, f n n n n n n ( ) = ( ) 2 2 2 21 3 1 4   ln , n is the diameter ratio, cv and ch are the vertical and horizontal consolidation coefficients, respectively. transforming the permeability coefficient of sand well foundation into that of sand wall foundation starting from baron’s theory (wei et al., 2018), and considering the effects of smear, lateral deformation and vertical seepage, the equivalent method between plane and axisymmetric problems of sand well foundation can be obtained. this method can only adjust the permeability coefficient (zhang et al., 2017), and the spacing of sand wall can still be determined according to the need of meshing (yang et al., 2018). the transformation formula is as follows: k d k k d khp h h vp v v= =; (4) where, kh and kv are horizontal and vertical permeability coefficients of sand well foundation, khp and kvp are equivalent horizontal and vertical permeability coefficients of sand wall foundation, and dh and dv are adjustment coefficients. the vertical permeability coefficient kv of silt roadbed is 2.82×10 -7 cm/s in laboratory test, and the horizontal permeability coefficient kh=2.60×10 -7 cm/s is calculated according to settlement data (wang et al., 2018). the horizontal permeability coefficient and vertical permeability coefficient are khp=8.92×10 -7 cm/s and kvp=2.54×10 -7 cm/s respectively when the sand well treatment area is converted into sand wall foundation. figure 1. geological bubble surface of test section, i.e. instrument burial map 186 bin liu, xiugen jiang figure 3 and figure 4 show the settlement curves of the magnetic rings on the subsidence tube of the test section. the buried position of each subsidence pipe and each magnetic ring on each subsidence pipe is shown in figure 1. figure 3. settlement curves of magnetic rings in no. 1 settling pipe of test section figure 4. settlement curves of magnetic rings in no. 2 settling pipe of test section figure 3 shows the settling process curves of the magnetic rings of no. 1 settlement pipe in the test section. the layered settlement pipe is buried on both sides of the embankment fill in the test section, i.e. at the shoulders of both sides of the embankment. the characteristics of the settlement of the magnetic rings are as follows: from november 19 to january 9, there is no obvious settlement process; from january 9 to february 17, the settlement of the magnetic rings has increased, and then until may 1, that is, before the embankment filling load, the settlement of the magnetic rings of the settlement pipe at the shoulder has not increased significantly. after may 1, the embankment filling began to load, and the settlement of the magnetic rings of the settlement pipe at the shoulder increased obviously. the main reason for this settlement process is that the temperature decreases from november 19 to january 9, and the embankment fill gradually freezes in depth. from january 9 to february 17, it is the coldest season in the region (issaka & ashraf, 2017; liu & baghban, 2017). the freezing depth reaches the maximum. the embankment fill forms a shell-like freezing layer with the thickness of 2.5 m, which has good integrity and high strength. it can transfer the load from the middle to the edge of the embankment, resulting in the edge soil. the pressure increases, which is consistent with the observation results of the increase of the earth pressure in the process of embankment freezing. figure 4 shows the settling process curves of the magnetic rings of the no. 2 settlement pipe in the test section. the layered settlement pipe is located in the central partition zone of the embankment fill. from november 19 to february 17, the following year, during the freezing process of embankment fill in winter, there is no obvious settlement of soft soil foundation at the central separation zone of the two test sections. from february 17 to may 1, it is the season when the temperature of embankment filling gradually rises. the shelllike freezing layer of embankment filling gradually melts. the pressure on the soft soil foundation at the central separation zone increases gradually. the settlement of the magnetic rings of the subsidence tube increases, especially the settlement of the magnetic rings near the surface of the soft rock foundation. every year from may to september is the construction season of highway construction in the northern region. the embankment filling speed is faster, especially concentrated in may, june and july. the settlement of soft rock foundation has an obvious acceleration process. the more the closer to the surface settlement of soft rock foundation is, the greater the settlement of soft soil is, and the settlement of soft soil accounts for more than 80% of the total settlement. in addition, the treatment method of soft soil foundation for bagged sand well is adopted in the test section. from the settlement data of the test section, it can be seen that the slope foundation of the settlement curve of the soft soil foundation treated by the bag sand well is small at the later stage, which indicates that the treatment of the shallow soft rock foundation of hydrological wetland is more conducive to drainage consolidation of the soft rock foundation and reduce the post-construction settlement. analysis of pore water pressure monitoring results tables 2 and 3 describe the variation curve of pore water pressure in the test section. tables 2 and 3 show that pore water pressure dissipates in varying degrees during the freezing process of embankment fill. during the whole winter, from feb. 17 to pre-loading, the observed pore water pressure increases gradually, but the rising amplitude and rate are not obvious. the main reasons are: (1) the temperature gradually rises, snow melts and rainfall influences, and the groundwater level gradually rises. (2) due to the melting of embankment fill, the frozen crust layer begins to melt. except near the slope angle, the soft rock foundation is subjected to the increase of the base pressure caused by embankment fill. since the beginning of may, embankment filling begind to be constructed. the additional stress of soft rock foundation increased under the action of filling loading, resulting in a significant increase in excess pore water pressure in the foundation soil. the change process of excess pore water pressure is very synchronous with the loading process of filling. that is, the excess pore water pressure increases obviously during loading. the faster the loading rate is, the larger the loading is, the faster the excess pore water pressure increases. the excess pore water pressure dissipates obviously after stopping loading. from the curve of pore water pressure change of two test sections, it can be concluded that the drainage consolidation process of soft rock foundation is greatly influenced by geological conditions. normally, the shallower the pore water pressure probe is buried, the larger the excess pore water pressure is generated in the initial stage of loading, because the excess pore water pressure is caused by additional stress. the more the nearer to the surface of soft rock foundation is, the greater the additional stress caused by embankment filling load is. from tables 2 and 3, it is easy to find that the excess pore water pressure dissipation of the deep buried probes no. 5 and no. 1 is very fast, basically after loading, the excess pore water pressure dissipation has been completed, because the two probes are buried in the well permeable gravel layer. through the curve of pore water pressure change in the test section, it can be concluded that the excess pore water pressure dissipates quickly in 187consolidation and deformation characteristics of soft rock foundation in a hydrological wetland environment soft soil foundation treated by bagged sand wells. this shows once again that the use of bagged sand wells to treat the soft rock foundation of hydrological wetland is more conducive to the drainage consolidation of soft rock and soil, and increases the stability of soft rock foundation. effect of water level variation on consolidation and deformation characteristics of soft rock and soil consolidation and deformation characteristics of water level with the same change range figure 5 is the settlement time history curve, and figure 5 (a) is the whole settlement time curve of surcharge preloading. figure 5 shows that settlement deformation mainly occurs during loading and decreases with the increase of depth. sediment curve on top of silt takes eight data points per day, while other curves only take one data point per day. settlement value changes periodically every day, and the change range is basically unchanged. from figure 5 (b), it can be clearly seen that the periodic variation of settlement caused by hydrological tides is the same as that caused by tidal frequencies, and its influence decreases rapidly with the increase of depth. the amplitudes of settlement cycle change are 35 mm and 5 mm at the top and 16 m below the silt surface, respectively. figure 6 is a pore pressure-time history curve, from which the effect of tide on pore pressure is found. figure 6 (a) is the pore pressure-time history curve in whole process. at the initial stage of loading, because of the small load, the hydrological tide has a great influence on pore pressure. the pore pressure increases with the depth. however, with the development of loading process, the pore pressure decreases with the depth increasing, indicating that the pore pressure caused by loading is greater in the shallow part than in the deep part, but when the drainage consolidation process is almost finished, pore pressure increases again with the increase of depth. it can be seen from figure 6 (b) that hydrological tide causes periodic variation of pore pressure. the frequency of variation is the same as that of hydrological tide, and the law of variation of pore pressure is opposite to that of hydrological tide. this is because pore pressure is influenced by both hydrological tides and the upper load caused by hydrological tides, and the variation of pore pressure with depth is small. effect of water level variation on drainage consolidation the actual situation of consolidation of silt roadbed in hydrological wetland area is very complicated. even if the external hydrological tide changes the same, because of the difference of the elevation of the silt top surface itself, the change of water level in the roadbed will be different, and most sections are filled with cofferdams, whose permeability varies greatly. therefore, the influence of hydrological tide on pore pressure and settlement in the measured values of each section is also quite different. the actual effect of hydrological tide in engineering is to change the range of water level change at the top of silt, and the range of influence is different at different sections. from figure 7, it can be seen that the larger the variation range of water level on the top of silt is, the larger the variation range of settlement is. this effect is a linear relationship. from figure 7, it can also be found that when the hydrological tide falls below table 2. pore water pressure variation curve of no. 2 hole in test section time no. 8 void hydraulic probe (initial buried depth 3.6 m)/kpa 7 void hydraulic probe (initial buried depth 4.7 m)/kpa no. 2 void hydraulic probe (initial buried depth 5.6 m)/kpa no. 5 void hydraulic probe (initial buried depth 7.4 m)/kpa november 5th 5.89 17.43 31.78 36.54 december 15th 2.76 15.04 30.65 33.99 january 25th -2.31 13.03 28.89 30.94 february 27th -2.24 12.4 30.23 30.23 april 7th 1.95 13.86 32.06 33.15 may 17th 3.6 17.66 33.74 35.42 june 26th 11.22 23.79 42.73 46.91 august 5th 7.05 17.68 33.15 35.06 september 15th 6.97 19.07 29.66 35 table 3. pore water pressure change curve of test section 3 time no. 6 void hydraulic probe (initial buried depth 3.6 m)/kpa no. 3 void hydraulic probe (initial buried depth 4.3 m)/kpa no. 9 void hydraulic probe (initial buried depth 4.9 m)/kpa no. 1 void hydraulic probe (initial buried depth 6.0 m)/kpa november 5th 3.98 11.03 21.18 32 december 15th 3.99 8.57 18.26 30.1 january 25th 4.02 6.02 10.05 27.95 february 27th 4.1 5.11 7.6 16.24 april 7th 5.17 7.58 7.93 29.17 may 17th 6.46 9.74 12.15 31.03 june 26th 15.03 15.37 19.34 38.86 august 5th 11.2 14.19 16.82 28.62 september 15th 9.85 14.3 16.91 27.44 188 bin liu, xiugen jiang the top of the silt, the settlement varies with the variation of water level. the larger the variation of water level is, the smaller the settlement is. this indicates that the hydrological tide will reduce the consolidation settlement. the larger the variation range of silt top water level is, the greater the influence on pore pressure variation range is. the influence relationship is also a linear relationship, as shown in figures 8, 9 and 10. when the water level is the same as the top of the silt, the variation range of water level has little effect on pore pressure, but from the whole pore pressure-time history curve in front, it can be seen that the variation range of pore pressure curve caused by each variation range of water level increases with the development of consolidation and drainage process. therefore, attention should be paid to the influence of hydrological tide on settlement and pore pressure in site measurement during construction of hydrological wetland areas. discussion considering the role of hydrological tide in the silt roadbed in hydrological wetland, the horizontal permeability coefficient of sand well silt roadbed is calculated based on the field measured data, and the three-dimensional sand well roadbed is converted into the plane strain sand wall roadbed, and the equivalent permeability coefficient is determined. the consolidation and deformation characteristics of soft rock foundation under hydrological wetland environment are calculated as follows: (1) at the shoulder position on both sides, the settlement of embankment fills is not obvious at the initial stage of freezing. in the coldest season, the settlement increases. then, until the embankment fills are loaded, the settlement near the slope does not increase significantly; at the central partition zone of (a) settlement-time curve of preloading with surcharge (b) settlement-time curve 4 days before construction commencement figure 5. settlement-time curve (a) the whole process (one calculation per day) (b) former 4d figure 6. pore pressure-time curve 189consolidation and deformation characteristics of soft rock foundation in a hydrological wetland environment embankment fills, there is no obvious settlement in the whole winter. as the temperature rises gradually, the settlement increases, and the settlement near the surface of the soft rock foundation increases obviously. during the construction season, the settlement of the soft rock foundation increases obviously, and the closer to the surface the settlement is, the larger the settlement of the soft rock soil accounts for more than 80% of the total settlement. the slope of the settlement curve of soft rock foundation treated by bagged sand well is smaller in later stage, which indicates that bagged sand well treatment method is beneficial to drainage and consolidation of shallow soft rock foundation of wetland in cold region, and reduces post-construction settlement of soft rock foundation. (2) during the whole winter, pore water pressure dissipates in varying degrees; the change of excess pore water pressure is very synchronous with the loading process of filling soil, and the excess pore water pressure increases obviously during loading. the faster the loading rate is, the larger the loading is, and the faster the excess pore water pressure increases. the excess pore water pressure dissipates obviously after stopping loading. the excess pore water pressure of soft rock foundation treated by bagged sand wells dissipates quickly, which is more conducive to drainage consolidation of soft rock and soil, and increases the stability of soft rock foundation. (3) during the freezing process of embankment fill, the earth pressure near the central separation zone decreases gradually, while the earth pressure near the embankment slope increases gradually. during the melting process of embankment fill, the earth pressure near the central separation zone increases gradually, while the earth pressure near the embankment slope decreases gradually. (4) hydrological tides and back pressure guards play an important role in drainage and consolidation of silt roadbed. the regularity of settlement and figure 7. settlement-time relationship curve under different water level variation ranges figure 8. pore pressure-time relationship curve under different water level variation ranges figure 9. settlement variation mud top water level variation figure 10. pore pressure variation range mud top water level variation range 190 bin liu, xiugen jiang deformation caused by hydrological tide changes. the settlement-time history curve changes periodically with the increase of hydrological tide, and its influence decreases with the increase of silt depth. the pore pressure dissipation curve is also affected by hydrological tides. with its periodic variation, the frequency of variation is the same as that of hydrological tides. the law of variation of pore pressure is contrary to that of hydrological tides, and its influence decreases with the increase of depth. (5) when the variation of silt top water level caused by tide is different, the periodic variation range of settlement and pore pressure is linearly related to the variation range of silt top water level. when the hydrological tide is lower than the silt top, the variation range of silt top water level has no effect on pore pressure, but it has obvious effect on settlement. the larger the variation range of water level is, the smaller the settlement is, indicating that the hydrological tide will decrease the settlement at consolidation stage. conclusions the engineering geological problems faced by the construction of highgrade highway on soft rock foundation are mainly the settlement and stability of soft rock foundation. the engineering properties of soft rock and soil are different because of different causes. in this paper, the research method of consolidation and deformation characteristics of soft rock foundation in hydrological wetland environment is proposed. the correlation of engineering properties, treatment methods and time effect of soft rock foundation in hydrological wetland is studied by combining on-site monitoring with numerical calculation. in the future study and work, we should carry out further work in this area, especially the experimental study on the structure of soft rock and soil and the compression and strength characteristics of soft rock and soil at different depths, as well as on-site monitoring of deformation of soft rock foundation in hydrological wetland during embankment filling. by evaluation of post-construction consolidation settlement and stability, construction loading rate is controlled by using monitoring data in further study, to ensure that the stability of soft soil foundation can guide construction more effectively. acknowledgements the study was supported by “the national natural science foundation of china, china (no. 11672362)”, and “natural science foundation of tibet autonomous region, china (no. xz 2018 zr g-32(z))”. references baral, p., rujikiatkamjorn, c.,, indraratna, b., & kelly, r. (2018). radial consolidation characteristics of soft undisturbed clay based on large specimens. journal of rock mechanics and geotechnical engineering, 10, 33-41. guang-ze, z., chun-yang, c., zhang, s., & gong, j. f. (2018). the discussion of boundary condition for large deformation of soft rockmass and its control countermeasures. journal of railway engineering society, 35, 27-31. guo, j. (2018). research on analysis of temperature variation and engineering practice based on super-long concrete structures. automation & instrumentation, 46-49. hao, l. i., bai, h. b., jian-jun, w. u., meng, q. b., wang, y. c., zhao, h. m., huang, w. x., ma, k., & zheng, l. (2017). elasto-plastic solution of deformation in circular soft rock tunnel based on double yield surface model. journal of china coal society, 42, 833-841. huang, x., liu, q., liu, b., liu, x., pan, y., & liu, j. (2017). experimental study on the dilatancy and fracturing behavior of soft rock under unloading conditions. international journal of civil engineering, 15, 1-28. issaka, s., & ashraf, m. a. (2017). impact of soil erosion and degradation on water quality: a review. geology, ecology, and landscapes, 1(1), 1-11. kotani, a., hiyama, t., ohta, t., hanamura, m., kambatuku, j. r., awala, s. k., & iijima, m. (2017). impact of rice cultivation on evapotranspiration in small seasonal wetlands of north-central namibia. hydrological research letters, 11, 134-140. liu, z. (2018). economic analysis of energy production from coal/biomass upgrading; part 1: hydrogen production. energy sources part b-economics planning and policy, 13(2), 132-136. liu, z., & baghban, a. (2017). application of lssvm for biodiesel production using supercritical ethanol solvent. energy sources part a-recovery utilization and environmental effects, 39(17), 1869-1874. ma, q., zhao, z. h., gao, x. j., chen, s. j., & tan, y. l. (2018). numerical survey on the destabilization mechanism of weakly cemented soft rock roadway considering interlayer effect. geotechnical & geological engineering, 37, 1-11. sun, l., yang, b., sun, c., li, x., & wang, z. (2017). experimental research on mechanism and controlling of floor heave in deep soft rock roadway. journal of mining & safety engineering, 34, 235-242. tao, z., zhu, c., zheng, x., wang, d., liu, y., he, m., & wang, y. (2018). failure mechanisms of soft rock roadways in steeply inclined layered rock formations. geomatics natural hazards and risk, 9, 1186-1206. wang, h., an, x., & zhang, z. (2018). effect of advanced treatment on ammonia nitrogen contained in secondary effluent from wastewater treatment plant. fresenius environmental bulletin, 27(4), 2043-2050. wei, x. h., cui, h. l., & li, h. l., & bai, x. (2018). cloud data cache technology supporting rapid visualization of large-scale seismic exploration data. journal of jilin university (science edition), 56, 1147-1155. xu, q., ding, w., zhu, h., tang, z., & li, y. (2017). study on progressive unloading failure characteristics of superlarge tunnel in soft and weak rock mass. china civil engineering journal, 50, 104-114, 132. yan, x., mao, h., ai, q. i., xu, x., & xiong, y. (2017). numerical analysis of seismic response and shaking table test of mega frame structure with isolated sub-structure. journal of building structures, 38, 43-51. yang, a. w., kong, l. w., & guo, f. (2017). accumulative plastic strain characteristics and growth model of tianjin binhai soft clay under cyclic loading. rock & soil mechanics, 38, 979-984. yang, x. b., lv, s. g., & jiang, j. p. (2018). experimental study on dc partial magnetic noise of 500 kv single-phase transformer in no-load condition. journal of china academy of electronics and information technology, 13, 446-451. zhang, j. (2017). theoretical analysis on failure zone of surrounding rock in deep large-scale soft rock roadway. journal of china university of mining & technology, 46, 292-299. zhang, x., jun, l. i., liu, j., & peng, w. (2017). research on deformation characteristics of strong expansive soft rock roadway and its control strategy. journal of china university of mining & technology, 46, 493-500. zhang, y. f., & wang, l. (2018). simulation of bearing capacity prediction of stratum in urban building. computer simulation, 35, 450-454. zhang, h., li, h., zhang, t., wang, q., wang, w., wang, x., zhao, h., chang, t., & guo, t. (2019). research and engineering application of high pre-stressed resistance enhancement large deformation bolt in deep soft rock roadway. meitan xuebao/journal of the china coal society, 44, 409-418. zhou, j., zheng, w. y., & liu, g. (2017). frequency adaptive nk±m order harmonic repetitive controller and its application in four-leg apf. journal of power supply, 15, 10-16. zhu, j., & wen, x. h. (2017). intelligent power monitoring system of building equipment based on internet of things. chinese journal of power sources, 41, 1775-1777. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201805020 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201805020 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201805020 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201804015 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201804015 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201804015 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201802097 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201802097 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201704002 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201704002 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201704002 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201712032 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201712032 keywords: fuzzy statistics; rock slope; stability; finite element analysis palabras clave: estadísticas difusas; pendiente de roca; análisis de elementos finitos. how to cite item wang, z., & lin, m. (2021). finite element analysis method of slope stability based on fuzzy statistics. earth sciences research journal, 25(1), 123-130. doi: https://doi. org/10.15446/esrj.v25n1.93320 in order to reduce the risk of slope stability evaluation due to the fuzziness of calculation parameters, a finite element analysis method of slope stability based on fuzzy statistics is proposed. based on the principle of quasi-static method and with the help of the finite element software comsol multiphysics, this paper studies the stability of the gravel soil accumulation rock slope under the action of different seismic acceleration. by analyzing the displacement, plastic zone and safety factor of the rock soil slope, the stability of the rock soil slope is analyzed. the research results show that the fuzziness of mechanical parameters of rock and soil slope will lead to the fuzziness of position displacement and stress analysis results of rock and soil slope, and the analysis of rock and soil slope with the method of fuzzy finite element analysis can strengthen the comprehensive understanding of position displacement, stress and safety of rock and soil slope by engineers and technicians, and reduce the stability of rock and soil slope due to the fuzziness of calculation parameters to a certain extent evaluate the risk qualitatively. abstract finite element analysis method of slope stability based on fuzzy statistics método de análisis de los elementos finitos para estabilidad de taludes con base en estadísticas difusas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.93320 con el fin de reducir el riesgo de evaluación en la estabilidad de taludes debido a la falta de claridad de los parámetros de cálculo, en este estudio se propone un método de análisis de elementos finitos con base en estadísticas difusas. basado en el principio del método cuasi-estático y con la ayuda del software multifísico de elementos finitos comsol, este trabajo estudia la estabilidad del talud de roca con acumulación de grava en suelo bajo la acción de diferentes aceleraciones sísmicas. al analizar el desplazamiento, la zona plástica y el factor de seguridad, se analiza la estabilidad del talud del suelo rocoso. los resultados de la investigación muestran que la falta de claridad de los parámetros mecánicos de la pendiente de la roca y del suelo conducirá a la falta de claridad del desplazamiento de posición y los resultados del análisis de tensión de la pendiente de la roca y el suelo. el análisis de la pendiente de la roca y el suelo con el método de análisis difuso de elementos finitos puede fortalecer la comprensión integral del desplazamiento de posición, el estrés y la seguridad de la pendiente de la roca y el suelo por parte de ingenieros y técnicos, y reducir la estabilidad de la pendiente de la roca y el suelo debido a la falta de claridad de los parámetros de cálculo y, hasta cierto punto, evaluar el riesgo cualitativamente. resumen record manuscript received: 13/11/2019 accepted for publication: 01/12/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 123-130 zhongjie wang1,*, min lin2 1department of architectural engineering, yantai vocational college, yantai, 264670, china 2yantai architectural design and research co. ltd, yantai, 264000, china *corresponding author: jzgcxjyk@163.com; yt_yt@126.com; g e o t e c h n ic a l e n g in e e r in g https://doi.org/10.15446/esrj.v25n1.93320 https://doi.org/10.15446/esrj.v25n1.93320 https://doi.org/10.15446/esrj.v25n1.93320 mailto:jzgcxjyk@163.com mailto:yt_yt@126.com 124 zhongjie wang, min lin introduction small and medium-sized landslides often occur in rock and soil slopes, most of which occur in a short time after heavy rain (canli et al., 2018). landslides undoubtedly pose a huge hidden danger to traffic safety. due to the lack of understanding of the characteristics of the special regional soil and the rock soil slope composed of it, the design and construction personnel are often unable to reasonably evaluate the stability of the rock soil slope and take reasonable protection and reinforcement measures (norouzi et al., 2019). therefore, it is necessary to systematically study the deformation mechanism and law, stability evaluation method, protection and reinforcement of rock and soil slope. because the parameters used in geotechnical analysis have a certain type of fuzzy uncertainty, the uncertainty or quantity is scattered. this is due to many reasons, such as test methods, etc., people can not accurately obtain the calculation mechanical parameters needed for calculation and analysis, and the value of mechanical parameters at a certain point in the calculation model can only be given a general range, for example, the measured value of friction angle or cohesive force in the rock and soil slope can never be a fixed value (tamate & hori, 2018). each experiment will lead to some data that is different from the previous values, thus showing a certain distribution in a certain range. in the past, the uncertainty (fuzziness or randomness) of each calculation parameter is eliminated according to the statistical analysis method before calculation, and a certain value is selected to evaluate the stability of rock and soil slope. the rigid body limit equilibrium method is widely used in the stability analysis of rock and soil slope. however, due to the limitations of the method itself and the uncertainty of many factors affecting the stability of the rock slope, the results of the solution are difficult to be directly used by the engineers, and its accuracy often depends on the experience of the users (karray et al., 2018). this results in a lot of phenomena that are judged to be stable in the calculation conditions, but in the actual situation, instability and failure occur. in view of this kind of problems, this paper studies the finite element analysis method of rock and soil slope stability. with the wide application of the finite element technology in the engineering field, the research on the application of the finite element method to evaluate the stability of the rock and soil slope has gradually increased. in these studies, all kinds of parameters are analyzed after being treated as deterministic parameters. the influence of the fuzzy uncertainty of the calculation mechanical parameters on the stress and displacement is also very significant, and the impact on the project cannot be ignored (naderiboldaji et al., 2018; xiao et al., 2018). due to the strong water permeability, wide particle size distribution and large pores of the accumulated gravelly soil, it is very easy to lose stability and induce landslide disaster under the action of rainfall or earthquake (chinkulkijniwat et al., 2019; provost et al., 2018). therefore, it is very important to deepen the understanding of the failure mode of the rock and soil slope and analyze the stability of the rock and soil slope for the prevention and control of the rock and soil slope disaster, which has certain research value and engineering practical significance (cao & wen, 2019). in this paper, the fuzzy finite element method is used to study the stability of rock slope. design of the finite element analysis method for the stability of rock and soil slope based on fuzzy statistics establishment of numerical model of rock and soil slope the finite element model is established based on the geotechnical slope data. the specific geometric dimensions of the geotechnical slope model are shown in figure 1. in the simulation, it is covered with gravelly soil and phyllite. model parameters: the slope height is 56m, the slope is about 19 °, the calculation area is 130m long, and the total height of the model is 75m. the model boundary is a fixed constraint, i.e. the displacement in x and y directions on the bottom boundary is 0; the left and right boundaries are roller supports, i.e. the displacement in the direction perpendicular to the boundary (normal) is zero, but the boundary can move freely in the tangent direction; the slope surface is a free boundary, and the specific boundary conditions are shown in figure 2. the parameters of rock and soil used for the stability calculation of rock and soil slope are determined according to the physical and mechanical test values provided in the rock and soil test report, as shown in table 1. table 1. rock and soil parameters material science gravels phyllite natural severe / (kn/m3) 18.4 23.5 modulus of elasticity /(mpa) 80 840 poisson’s ratio 0.35 0.24 cohesive force / (kpa) 17.6 45.00 internal friction angle / (°) 24.00 30.00 according to the statistics, the strength parameters c and  of gravelly soil on the slope, and the change interval of fuzzy statistical coefficient is 0.05, and the statistical results are shown in table 2. table 2. fuzzy statistics of strength parameters of gravelly soil fuzzy statistical coefficient cohesionc/ kpa internal friction angleψ/ (°) 1.00 17.60 24.00 1.05 16.68 22.86 1.10 16.00 21.82 1.15 15.30 20.87 1.20 14.67 20.00 1.25 14.08 19.20 1.30 13.54 18.64 1.35 13.03 17.78 1.40 12.58 17.14 1.45 12.13 16.55 1.50 11.73 16.00 figure 1. numerical model of geotechnical slope figure 2. boundary conditions of geotechnical slope 125fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment in the finite element method, mesh generation has a great influence on convergence and solution speed. the free triangulation grid provided by comsol multiphysics software (kwan et al., 2019; bouida et al., 2018) is adopted for the geotechnical slope. the whole grid consists of 11086 triangle cells, 431 edge cells, 41 top cells, 0.5314 minimum cell mass, 0.9844 average cell mass, 0.05057 cell area ratio, 6703m2 grid area, 2.247 maximum growth rate and 1.095 average growth rate. calculation of the load on the slope it is necessary to calculate the safety degree of the fuzzy point before calculating the load on the slope. the fuzzy point safety factor ks is an index used to evaluate the safety degree of each point in the slope land. to some extent, it can reflect the local stability of the slope. the fuzzy limit equilibrium method can be used to study the fuzzy evaluation of the overall stability of geotechnical slope (sajjad et al., 2018). the safety factor ks of fuzzy point is a ratio of fuzzy stress and shear strength of rock and soil. according to the mohr coulomb criterion, the calculation formula of point safety factor is: k c f s = + + + −      − σ σ σ σ α σ σ α 1 2 1 2 1 2 2 2 2 2 2 cos sin (1) α σ σ σ σ = − −      + +      −1 2 2 2 1 1 2 1 2 cos f c f (2) in the calculation of anti slope stability, because the structure of rock and soil slope has no bottom, the soil in the slope cannot participate in the anti slope work. according to the research results on the structure of rock and soil slope, 70% of the weight of soil in the slope can be taken as the effective weight of anti slope stability, then the expression of gravity g is: g g h blt i i= + ( )∑0 7. γ (3) where: gt is the weight of the slope, i and hi are the weight and depth of the i layer soil in the slope, b is the width of the simplified structure, and l is the distance between the centerline of the rock soil slope structure and the centerline of the subgrid bin. it is assumed that the earth pressure on both sides of the slope will reach the limit value when the immersed rock soil slope structure reaches the ultimate failure state. according to the analysis in 1.1, the active earth pressure on the landside of the slope is multiplied by the rankine active earth pressure resultant force on the imaginary vertical plane by the fuzzy statistical coefficient of 0.75, and the passive earth pressure on the sea side of the slope is multiplied by the rankine passive earth pressure resultant force on the imaginary vertical plane by the coefficient of 1.2. the specific expression is: f l e dya a h t = + ∫0 75 0 0 . (4) e h k c k qka i i a a a=( ) − +γ 2 (5) f l e dyp p t = ∫1 2 0 0 . (6) e h k c kp i i p p=( ) +γ 2 (7) in the formula, h is the height of the non embedded part of the slope, t0 is the depth of the slope, ea and ep are the active earth pressure and passive earth pressure values at a certain depth, fa and fp are the active and passive earth pressure resultant values of the wall, i and hi are the weight and depth of the soil in the i layer, l is the distance between the centerline of the geotechnical slope ground structure and the centerline of the sub compartment. according to the american api code, the friction force f ei d= / of soft clay to the unit area of rock soil slope is not greater than the undrained shear strength cu of soft clay; the maximum end resistance stress (soil reaction at the bottom of slope) is 9cu. assuming that when the rock soil slope structure reaches the ultimate failure state, the side friction and bottom reaction of the slope body reach the limit value, then: f l c dyu t  = ∫ 0 0 (8) f l c dyu h t  = + ∫ 0 0 (9) f c l d xr u= +( )9 β (10) among them, the value of cu is derived from the shear strength formula and c and . at a point where the depth from the ground is y, the undrained shear strength of soft clay is: c k u h cu t i i= ( ) +∑0 γ ϕtan (11) where: γi ih( )∑ is the weight of all soil layers from above y to the ground, and the meaning of other symbols is the same as before. at the depth of slope bottom, the undrained shear strength of soft clay is: c u h cu t i i= ( ) +∑ γ ϕtan (12) where: γi ih( )∑ is the weight of all soil layers above the slope bottom, and other symbols have the same meaning. in the formula, c is the cohesion of rock mass; f ei d= / is the friction coefficient  1 of rock;  2 is a positive value at compressive stress. finite element analysis of the stability of rock soil slope the key factor to determine the stability of rock soil slope is the embedment of foundation soil. in the limit state, it is considered that the actual shear stress produced by the external load of rock soil slope is equal to the minimum shear strength that the soil can play to resist the external load, that is, when the actual strength index of soil is reduced continuously until reaching the ultimate failure state, the shear strength of soil is equal to the external load the strength of shear stress produced by loading is equal. the ultimate failure state of rock soil slope can be simulated by weakening the strength of soil. especially for the geotechnical slope structure, the external load it bears is complex, including the wave force, the earth pressure generated by the filler behind the wall, the earth pressure generated by the variable load on the wharf surface, the residual water pressure caused by the water level difference, the stacking load, the flow mechanical load and other ground service loads (saneie et al., 2018). at the same time, in the stability analysis of geotechnical slope structure, 126 zhongjie wang, min lin the load combination under various adverse conditions must be considered. therefore, by using the fuzzy statistical method, the stability and safety factor of the structure can be obtained simply by weakening the strength parameters of the soil and simulating the instability and failure of the rock soil slope structure. in this paper, the fuzzy statistical method is used, and the large-scale finite element analysis software is used abaqus, a three-dimensional elasticplastic overall analysis model of interaction between rock soil slope structure and soil mass is established. the basic contents include: selection of model calculation domain, model boundary conditions, selection of model calculation unit, constitutive model of structure and soil mass, model of contact surface between soil mass and structure, simulation of initial stress field of foundation, etc. (kundu et al., 2018). taking the structure of rock and soil slope as an example, the typical plan and section of the structure are shown in figure 3 and figure 4 respectively. figure 3. plan of geotechnical slope structure (unit: mm) figure 4. cross section of geotechnical slope structure the structure of rock and soil slope is formed by a series of continuous arrangement of slopes, with arcade walls and obvious spatial bearing characteristics. when the action direction of wave load is perpendicular to the wall axis of the structure, the structure is symmetrical about the plane parallel to the action direction of wave force. therefore, for the convenience of analysis, it is assumed that the action direction of wave load is perpendicular to the wall axis of rock soil slope structure. using the spatial symmetry of load and structure, half of a slope and a certain range of soil before and after the wall axis are taken as the area to establish the finite element model, and symmetrical boundary conditions are set on the symmetrical boundary. among them, the calculation domain of soil mass is: in the direction perpendicular to the wall axis, take the area within 5 times of the diameter of the geotechnical slope on both sides; in the direction of the structural depth of the geotechnical slope, take the area within 3 times of the depth of the slope below the slope (chiorescu et al., 2019). the calculation domain plan of the overall model is shown in figure 5. front side wave direction right side structure of rock soil slope slope top left side rear fill 14m 26m 130m 130m calculation area of finite element model rear side figure 5. plan of finite element analysis calculation area in the setting of boundary conditions in the finite element calculation domain, the influence of boundary conditions on the finite element analysis domain should be reduced as much as possible. therefore, in the finite element analysis model of geotechnical slope structure, the basement surface is fixed boundary, the surface is free boundary, the left and right sides are symmetrical boundary, the front and rear sides are side boundary (cifuentes et al., 2019). generally, the structure of rock soil slope is made of steel. the strength and rigidity of the structure are far greater than that of the soil. the displacement and instability of the whole model system mainly depend on the deformation and bearing capacity of the foundation soil. therefore, in the overall analysis model, the slope structure is simulated by elastic model, while the constitutive model of the soil is simulated by abaqus mohr coulomb model in finite element analysis software is used for simulation (kalateh, 2019). for the three-dimensional elastic-plastic finite element model, due to the existence of soil with incompressible plastic deformation, which to some extent limits the choice of model element type. in order to avoid over constraint and reduce calculation cost as much as possible, eight node three-dimensional reduced integral solid element is used for soil and rock slope structure elements in this paper. in order to simulate the phenomenon of adhesion, slippage and separation between soil and rock slope structure under the action of wave force, residual water pressure, stacking load and other external loads, a master-slave contact surface is set up in the area where the structure contacts the soil. considering that the elastic modulus of the structure is far greater than the elastic modulus of the soil, the contact surface on the structure is designated as the main contact surface, and the contact surface on the soil is from the contact surface. the normal direction of the contact constitutive model is hard contact, and the tangential direction is coulomb friction constitutive model. according to the american api code, it is set that the friction between cohesive soil and slope wall per unit area f ei d= / is not greater than the undrained shear strength cu of clay (talesnick, 2013). according to the principle of fuzzy statistics, the stability analysis steps of rock soil slope structure under various external load combinations are as follows: step 1: establish an elasto-plastic three-dimensional finite element analysis model of interaction between rock soil slope structure and soil mass; step 2: according to the actual situation of the project, calculate the external load of the geotechnical slope structure, and according to its load distribution form, set the corresponding load distribution in the finite element analysis model to solve the stress and displacement of the soil and structure system under the design load; 127fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment step 3: the strength parameters (including cohesion c and tangent value tan  of internal friction angle) of all soils in the overall finite element analysis model are calculated step by step to solve the stress and displacement of soil and structure system under the corresponding soil strength until the finite element calculation does not converge; step 4: draw the relationship curve between the fuzzy statistical coefficient of soil strength parameters and the structural displacement; step 5: determine the criteria of structural instability; step 6: according to the criteria of structural instability, the strength reduction coefficient corresponding to structural instability is determined, which is defined as the safety coefficient of structural stability of geotechnical slope (chen et al., 2019; feng et al., 2019). comparative analysis of experiments construction of experimental model the stratum of the slope is from triassic to jurassic, in which triassic is a set of thick to very thick sandstone, mainly siltstone and fine sandstone, with weak argillaceous rock and soft hard inter bedded rock group of coal line; jurassic is a set of soft rock with hard rock group of siltstone and mudstone as main inter bedded sandstone. the artificial high rock and soil slope is 1300.0m long, with the maximum excavation elevation of 425.00m and the maximum slope height of 150.0m. in order to study the artificial high rock slope of the ship lift section on the left bank of xiangjiaba, the uncertainty analysis of plane stress and plane deformation of typical section 4 (maximum section height) is carried out by using the fuzzy finite element theory. the finite element mesh of rock slope is shown in figure 6. figure 6 finite element mesh division of rock slope the physical and mechanical parameters of rock mass are shown in table 3. the values outside the brackets in the table are calculated parameter values based on the statistical mean, while the values inside the brackets are the range values of each physical quantity. boundary conditions: the left and right boundaries of the experimental model are respectively constrained by the horizontal x-direction hinge displacement, and the lower boundary of the model is constrained by the vertical y-direction hinge displacement, forming the displacement boundary conditions. quadrilateral 4-node element is used for calculation. table 3. mechanical parameters of rock mass in slope land formation lithology t3 2-6 、 t3 4 sandstone t3 3 siltstone, siltstone with argillaceous rock argillaceous rock with sandstone strong weathering rock mass type iv iv iv f  0.75(0.62-0.81) 0.65(0.52-0.73) 0.55(0.42-0.65) c' / mpa 0.60(0.52-0.73) 0.50(0.45-0.64) 0.30(0.23-0.41) rock density /(g·cm-3) 2.50(2.44-2.61) 2.46(2.4-2.55) 2.40(2.35-2.52) modulus of elasticity / gpa 5.0(4.68-5.72) 4.2(3.78-5.1) 3.3(2.7-4.0) poisson’s ratio 0.25(0.23-0.26) 0.28(0.24-0.31) 0.30(0.25-0.32) moderate weathering rock mass type iii iii iii f  1.05(0.94-1.13) 0.95(0.89-1.02) 0.8(0.68-0.96) c' / mpa 1.0(0.88-1.11) 0.9(0.84-0.97) 0.7(0.66-0.79) rock density /(g·cm-3) 2.57(2.51-2.62) 2.58(2.40-2.67) 2.56(2.39-2.61) modulus of elasticity / gpa 12.0(11.3-13.1) 9.0(8.3-10.2) 5.0(4.6-5.3) poisson’s ratio 0.23(0.21-0.245) 0.25(0.24-0.263) 0.28(0.265-0.3) breeze rock mass type ii ii ii f  1.2(1.01-1.3) 1.0(0.85-1.17) 0.9(0.85-1.1) c' / mpaz 1.4(1.13-1.56) 1.0(0.87-1.18) 0.8(0.73-0.89) rock density /(g·cm-3) 2.60(2.55-2.75) 2.62(2.54-2.73) 2.58(2.5-2.7) modulus of elasticity / gpa 19.0(17.4-21.7) 12.0(10.2-13.7) 8.0(7.3-9.1) poisson’s ratio 0.22(0.2-0.24) 0.23(0.21-0.24) 0.25(0.23-0.27) 128 zhongjie wang, min lin load condition: self weight, excavation unloading. in order to simulate the excavation and unloading of rock and soil slope, loading is carried out in two steps. firstly, the natural rock and soil slope before excavation is discretized, and the whole model is discretized into 1372 nodes and 1294 units. after the calculation, the node forces on the excavation boundary are reversely applied on the corresponding nodes of the calculation grid after excavation to simulate the unloading process in the excavation area of rock and soil slope. after excavation, there are 1255 nodes and 1178 elements in the finite element model. during the experiment, it is considered that the displacement under the action of self weight has been completed, so the displacement caused by excavation is only the displacement caused by unloading of the slope, that is to say, when calculating the deformation caused by excavation of the slope, the displacement under the condition of self weight is deducted, and the stress state caused by self weight is retained. test steps the following simplified calculation method is established for the structure of rock and soil slope, and the solution process is realized by fortran programming language: step 1: set the stability safety factor value k required by the geotechnical slope structure; step 2: according to the basic principle of the finite element strength reduction method, the strength parameters c and  of all soils are reduced to the soil parameters c' and ' under the condition of the stability safety factor, and are substituted into the load calculation formula; step 3: combine the equation and the calculation formula of reaction fr at the bottom of slope to find the depth of rock and soil slope that meets the set safety factor: a. let x0 = 0, solve the equation, and get the initial value t0 of the depth of the slope. b. from (1) the t0 value is used to solve all the load values, and the equation is used to get the reaction value of rock soil slope; c. the fr value obtained from (2) is used to calculate the distance x between the turning point and the centerline by the following formula: f c l d xr n= +( )9 2β β (13) d. according to the relationship between x and x0 , the x0 value under the condition of meeting x obtained in (3) is obtained, which is: x d d x d x 0 2 2 = − +( ) = − β β β (14) if | x'0−x0| <  and  are given accuracy requirements, calculation will be stopped; otherwise, continue to the next step. e. let x x x 0 0 0 2 = − , solve the equation, and obtain the value t0 of the depth of the slope under this condition. continue (2) and the following steps until the end of the calculation. at this time, the calculated t0 is the depth of the rock slope to meet the set stability safety factor value, and x is the distance between the turning point of the rock slope and the central axis. analysis of experimental results a. analysis of the characteristic values of fuzzy stress and fuzzy displacement at the measuring points of geotechnical slope figure 7 is a schematic diagram of the location of geotechnical survey points. see table 4 for the characteristic value of stress and displacement at the geotechnical survey point. figure 7. location of geotechnical survey points table 4. characteristic values of stress and displacement at measuring points of geotechnical slope point number most probable value minimum value maximum value p/l h/p 1 horizontal displacement /mm -0.55 -0.44 -0.78 1.26 1.42 vertical displacement /mm -0.42 -0.34 -0.76 1.21 1.84 shear stress / kpa -2.34 0.97 -6.33 -2.41 2.71 maximum principal stress /kpa 14.71 12.66 66.73 1.16 4.53 minimum principal stress / kpa -25.7 -11.42 -84.89 2.26 3.29 2 horizontal displacement /mm 0.93 0.91 1.92 1.02 2.05 vertical displacement /mm -2.39 -2.19 -4.32 1.09 1.18 shear stress / kpa 2.20 1.44 3.84 1.53 1.74 maximum principal stress /kpa 13.30 8.30 42.99 1.60 3.23 minimum principal stress / kpa -19.55 -11.03 -64.57 1.77 3.30 3 horizontal displacement /mm 0.85 0.85 1.80 1.00 2.11 vertical displacement /mm -2.77 -2.55 -4.94 1.09 1.78 shear stress / kpa -1.79 -1.00 -3.96 1.79 2.21 maximum principal stress /kpa 10.63 7.06 32.42 1.50 3.05 minimum principal stress / kpa -20.49 -11.30 -71.76 1.81 3.50 4 horizontal displacement /mm 1.89 1.85 3.94 1.02 2.08 vertical displacement /mm -5.21 -4.68 -9.42 1.07 1.77 shear stress / kpa -16.28 -14.14 -37.89 1.15 2.33 maximum principal stress /kpa -11.41 -6.24 -129.76 1.83 11.37 minimum principal stress / kpa -58.68 -51.73 -336.01 1.13 5.73 5 horizontal displacement /mm -0.78 -0.57 -1.17 1.37 1.50 vertical displacement /mm -2.36 -2.19 -4.14 1.08 1.75 shear stress / kpa -8.18 -6.46 -19.27 1.27 2.36 maximum principal stress /kpa -28.34 -12.92 -146.71 2.19 5.18 minimum principal stress / kpa -52.18 -30.43 -240.683 1.71 4.61 129fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment it can be seen from table 2 that due to the complexity of the geometric boundary of the experimental section and the fuzziness of the material, the distribution state of the earth stress and displacement of the rock slope may change greatly. when the fuzziness of materials is small, that is, the value of mechanical parameters of materials is close to the most possible value (mean value), and the deviation of calculation results is small. when the fuzziness of materials is large, that is, the value of mechanical parameters of materials is far from the most possible value (mean value), and the deviation of calculation results is large. the ratio of possible value to minimum value (p / l) and the ratio of maximum value to possible value (h / p) in table 2 show this one o’clock. table 2 shows that the deviation of the maximum principal stress can reach 11.37 times. the calculation also shows that the change of the position shift of the rock slope caused by the fuzziness of the rock slope material is smaller than the change of the earth stress of the rock slope. the ratio of the maximum to the possible value (h / p) is less than 2.0, and the ratio of the maximum to the possible value (h / p) is more than 2.0. it can be seen that the fuzziness of rock and soil slope material has a great influence on the stress estimation of rock and soil slope. b. fuzzy displacement analysis of excavated slope figure 8-11 shows the contour map for calculating the displacement of rock slope. due to the high level of membership (0.8, 0.9, 1.0), it can be found that the distribution of horizontal displacement and vertical displacement are basically the same under the condition of excavation and unloading. the horizontal displacement of the slope shows an inward movement above 320m and an outward movement below 320m. the displacement model of rock slope is suitable for the material composition of rock mass and the geometric properties of material interface. due to the formation of the free surface at the foot of the slope, the downward vertical displacement occurs after unloading. the displacement of rock and soil slope caused by excavation and unloading is small, and its magnitude is millimeter. the maximum value of vertical displacement under different membership levels appears at the top of the slope, and the value of vertical displacement gradually decreases from top to bottom. the maximum value of horizontal displacement under different levels of subordination degree generally appears on the rock slope excavated at the elevation of 450 m. conclusions in this paper, based on fuzzy statistics, the finite element analysis method of rock and soil slope stability is proposed. the two-dimensional finite element analysis method of rock slope stability is established by introducing the fuzzy statistical method. based on the principle of quasi-static method and with the help of the finite element software comsol multiphysics, the stability of rock and soil slope of gravel soil accumulation under different seismic acceleration is studied. by analyzing the displacement, plastic zone and safety factor of the rock soil slope, the stability of the rock soil slope is analyzed. the research results show that the fuzziness of mechanical parameters of rock and soil slope will lead to the fuzziness of position displacement and stress analysis results of rock and soil slope, and the analysis of rock and soil slope with the method of fuzzy finite element analysis can increase the comprehensive understanding of position displacement, stress and safety of rock and soil slope by engineers and technicians, and reduce the stability evaluation of rock and soil slope caused by the fuzziness of calculation parameters price risk. figure 8. upper limit of horizontal displacement of geotechnical slope profile 4 (μ = 0.8) figure 9. lower limit of horizontal displacement of geotechnical slope profile 4 (μ = 0.8) figure 10. lower limit of vertical displacement of geotechnical slope profile 4 (μ = 0.8) figure 11. lower limit of vertical displacement of geotechnical slope profile4 (μ = 0.8) 130 zhongjie wang, min lin references bouida, n., bouchikhi, a. s., megueni, a., & gouasmi, s. (2018). a finite element analysis for evaluation of j-integral in plates made of functionally graded materials with a semicircular notch. journal of failure analysis and prevention, 18, 1573–1586. canli, e., loigge, b., & glade t. (2018). spatially distributed rainfall information and its potential for regional landslide early warning systems. natural hazards, 91(1), 103-127. cao, j., & wen, f. h. (2019). the impact of the cross-shareholding network on extreme price movements: evidence from china. journal of risk, 22(2), 79–102. chen, c., wang, w., & lu, h. y. (2019). stability analysis of slope reinforced with composite anti-slide pile model. rock and soil mechanics, 40(08), 3207-3217. chinkulkijniwat, a., tirametatiparat, t., supotayan, c., yubonchit, s., horpibulsuk, s., salee, r., & voottipruex, p. (2019). stability characteristics of shallow landslide triggered by rainfall. journal of mountain science, 16(9), 2171-2183. chiorescu, d., chiorescu, e., nagîţ, g., & olaru, s. c. (2019). the study of the radius of connection of the die for deep drawing using analysis with finite element. materials science forum, 957, 103-110. cifuentes, h., montero-chacón, f., galán, j., cabezas, j., & martinez-de la concha, a. (2019). a finite element-based methodology for the thermomechanical analysis of early age behavior in concrete structures. international journal of concrete structures and materials, 13(1), 41. feng, w. x., ma, j. s., & he, j. (2019). fitting method of pressure subsidence characteristics based on rbf surrogate model. computer simulation, 36(09), 5-9. kalateh, f. (2019). dynamic failure analysis of concrete dams under air blast using coupled euler-lagrange finite element method. frontiers of structural & civil engineering, 13(1), 15-37. karray, m., hussein, m. n., delisle, m. c., & ledoux, c. (2018). framework to assess the pseudo-static approach for the seismic stability of clayey slopes. canadian geotechnical journal, 55(12), 1860-1876. kundu, j., sarkar, k., singh, p. k., & singh, t. n. (2018). deterministic and probabilistic stability analysis of soil slope – a case study. journal of the geological society of india, 91(4), 418-424. kwan, j. s., sze, e. h. y., & lam, c. (2019). finite element analysis for rockfall and debris flow mitigation works. canadian geotechnical journal, 56(9), 1225-1250. naderi-boldaji, m., hajian, a., ghanbarian, d., & bahrami, m. (2018). finite element simulation of plate sinkage, confined and semi-confined compression tests: a comparison of the response to yield stress. soil and tillage research, 179(1), 63-70. norouzi, e., moslemzadeh, h., & mohammadi, s. (2019). maximum entropy based finite element analysis of porous media. frontiers of structural and civil engineering, 13(2), 364-379. provost, f., malet, j. p., gance, j., helmstetter, a., & doubre, c. (2018). automatic approach for increasing the location accuracy of slowmoving landslide endogenous seismicity: the apoloc method. geophysical journal international, 215(2), 1455-1473. sajjad, m., joy, j. a., & jung, d. w. (2018). finite element analysis of incremental sheet forming for metal sheet. key engineering materials, 783, 148-153. saneie, h., alipour-sarabi, r., nasiri-gheidari, z., & tootoonchian, f. (2018). challenges of finite element analysis of re-solvers. ieee transactions on energy conversion, 34(2), 973 983. talesnick, m. (2013). measuring soil pressure within a soil mass. canadian geotechnical journal, 50(7), 716-722. tamate, s., & hori, t. (2018). monitoring shear strain in shallow subsurface using mini pipe strain meter for detecting potential threat of slope failure. geotechnical testing journal, 41(2), 413-424. xiao, s., guo, w. d., & zeng, j. (2018). factor of safety of slope stability from deformation energy. canadian geotechnical journal, 55(2), 296-302. _gjdgxs keywords: enhanced electrodynamic method; high concentration chromium; contaminated soil method; solubility palabras clave: método electrodinámico mejorado; cromo de alta concentración; método de suelo contaminado; solubilidad how to cite item guo, l., zhang, x., chen, y., & liu, j. (2021). remediation of high concentration chromium contaminated soil by enhanced electrodynamic method. earth sciences research journal, 25(2), 247-253. doi: https://doi.org/10.15446/ esrj.v25n2.93896 high concentration of chromium salt has caused serious pollution to the environment since its production. the longstanding chromium residue has polluted the soil, and the total chromium concentration of some polluted soil has reached 30000 mg / kg. for the remediation of chromium contaminated soil, the enhanced electrodynamic method was proposed. in order to improve the efficiency of electrokinetic remediation of chromium contaminated soil, two enhanced electrokinetic remediation technologies were proposed: electrokinetic oxidation enhanced remediation technology and electrokinetic enhanced remediation technology. cr (iii) in soil was polluted by oxidant chromium oxide to increase the content of dissolved cr (vi),so as to improve the efficiency of electric repair, in order to find a suitable prb medium as a breakthrough to improve the repair efficiency. the experimental results show that compared with the traditional electrodynamic technology, the enhanced electrodynamic method can effectively improve the removal rate of total chromium in the soil, and provide technical support for the enhanced electrodynamic remediation of chromium contaminated soil. the ph control system makes the ph of anode and cathode electrolyte always maintain acidic, and the h+ migration speed is faster than that of oh-, and the ph of soil near the anode is lower than that near the cathode. abstract remediation of high concentration chromium contaminated soil by enhanced electrodynamic method remediación de suelos contaminados con cromo de alta concentración mediante un método electrodinámico mejorado issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n2.93896 la alta concentración de sal de cromo ha causado una grave contaminación al medio ambiente desde su producción. el residuo de cromo de larga data ha contaminado el suelo y la concentración total de cromo de algunos suelos contaminados ha alcanzado los 30000 mg / kg. para la remediación de suelos contaminados con cromo se propuso el método electrodinámico mejorado. para mejorar la eficiencia de la remediación electrocinética de suelos contaminados con cromo se propusieron dos tecnologías mejoradas de remediación electrocinética: tecnología de remediación mejorada por oxidación electrocinética y tecnología de remediación mejorada por oxidación electrocinética. cr (iii) en el suelo fue contaminado por óxido de cromo oxidante para aumentar el contenido de cr(vi) disuelto, para mejorar la eficiencia de la reparación eléctrica, para encontrar un medio prb adecuado como un gran avance para mejorar la eficiencia de la reparación. los resultados experimentales muestran que, en comparación con la tecnología electrodinámica tradicional, el método electrodinámico mejorado puede mejorar eficazmente la tasa de eliminación del cromo total en el suelo y proporcionar apoyo técnico para la remediación electrodinámica mejorada del suelo contaminado con cromo. el sistema de control de ph hace que el ph del electrolito del ánodo y del cátodo siempre se mantenga ácido, y la velocidad de migración del h+ es más rápida que la del oh-, y el ph del suelo cerca del ánodo es más bajo que el del cátodo. resumen record manuscript received: 23/01/2019 accepted for publication: 10/10/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 2 (june, 2021): 247-253 lin guo1*, jun liu1, yao chen1, xiaocun zhang2 1electronic information and electrical engineering college, shangluo university, shangluo, 726000, china 2college of urban, rural planning and architectural engineering, shangluo university, shangluo, 726000, china corresponding author: linguo0303@126.com so il s mailto:linguo0303@126.com 248 lin guo, xiaocun zhang, yao chen, jun liu introduction the production of various chromium compounds such as potassium dichromate and chromium trioxide is characterized by backward technology, small scale and scattered layout. chromium pollution accidents occur frequently, resulting in a large number of water-soluble chromium cr (vi) slag, which has seriously threatened human health and caused serious environmental pollution. chromium enterprises have gradually become the focus of the government and all walks of life. for the purpose of source control, a large number of seriously polluted chromium enterprises have been shut down in recent years, and most of the chromium slag has been reasonably treated by the end of 2012 (jourdan et al., 2018). however, due to a large number of abandoned sites of closed chromium production enterprises have not taken necessary environmental protection measures, chromium residue has been piled up for a long time, and a large number of cr(vi) has entered into the soil through the effect of wind and rain, causing the soil chromium content to exceed the standard seriously, and chromium contaminated soil has become a key problem of soil pollution. the survey found that there were more than one third of the total determination points, about 271 points, and chromium exceeding standard was one of the important indicators (ojanen et al., 2018). after investigation, the pollution characteristics of excessive chromium in soil are as follows: from the analysis of valence state, the content of cr (iii) in soil chromium pollution is significantly higher than that of cr (vi), the content of cr (vi) in chromium residue site is only more than ten percent of cr (iii); from the analysis of morphology, the proportion of residual chromium is the largest. in reference (cleveland & townsend, 2019) suggests that the increase of nutrients in tropical rainforests will lead to the loss of a large amount of carbon dioxide from soil to the atmosphere. the terrestrial biosphere atmosphere carbon dioxide (co) exchange is mainly in tropical forests, where the absorption of photosynthetic carbon (c) is considered to be limited by phosphorus (p).phosphorus can also limit the decomposition of organic matter and the loss of soil carbon in phosphorus deficient tropical forests. the effects of phosphorus fertilizer on soil respiration in costa rica lowland tropical rain forest were studied through field fertilization experiments. in the early stage of wet season, when the input of soil soluble organic matter was large, the soil respiration rate increased significantly with the application of phosphorus. the results showed that the mechanisms driving the observed respiratory responses to nitrogen and phosphorus increase may be different, but nitrogen and phosphorus fertilization stimulated a large amount of co loss. in reference (jurkevitch et al., 2019), the characteristics of predation range, ribotyping, soil and rhizosphere diversity of bdellovibrio isolated from plant pathogenic bacteria were proposed. thirty strains of bdellovibrio were isolated from agricultural soil and plant rhizosphere. by means of molecular and culture methods, it was found that bdellovibrio in soil contains different biological subpopulations from that in rhizosphere. 13 strains of soil and 7 strains of soybean rhizosphere bdellovibrio were isolated from pseudomonas corrugata, and 7 strains and 2 strains of soil were isolated from erwinia carinii subsp. by analyzing the predation range of 5 isolates of bdellovibrio in soil and rhizosphere, it was found that most of the 22 substrate species were plant pathogens and plant growth promoting bacteria, which had unique utilization patterns and differences among closely related strains. although some progress has been made in the above research, there are some limitations in the remediation of high concentration chromium contaminated soil. therefore, it is proposed to strengthen the research on the remediation of high concentration chromium contaminated soil by electrodynamic method. remediation of high concentration chromium contaminated soil by enhanced electrodynamic method there are two main ways to remediate chromium contaminated soil: a. remove chromium from the soil to make the retained concentration of chromium close to or reach the soil background value; b. change the existing form of chromium in the soil, reduce hexavalent chromium to trivalent chromium, and reduce its migration ability and bioavailability in the soil (watson et al., 2019; zehe et al., 2018; walker et al., 2018). the removal of chromium from soil mainly focuses on chemical leaching, microbial leaching and phytoremediation. the optimization of chemical leaching agents is the research focus of chemical leaching technology, and the research and development of high-efficiency environmental protection chemicals is the future development direction. citric acid, edta、glda and other chemicals used in the chemical leaching of chromium contaminated soil show good removal effect, and even clean water can effectively remove hexavalent chromium in soil (rinella & reinhart, 2020; ser et al., 2020). microbial leaching uses organic acids or surfactants produced by microorganisms to acidify or complexate with chromium to separate it from soil. compared with chemical leaching, the cost of microbial leaching can be reduced by about 80%, which has little disturbance to the environment and will not cause secondary pollution. therefore, it has been widely concerned in recent years (shi et al., 2018; zehe et al., 2018). some scholars have successfully leached 52% 68% of cr from heavy metal contaminated soil by autotrophic microorganism thiobacillus. heterotrophic microorganisms such as bacillus, pseudomonas and mold also have a good leaching effect on cr. at present, there are few studies on the removal of chromium from soil by microbial adsorption, mainly due to the difficulty of separation from soil (zohdi, 2018; frey et al., 2019; marchiori et al., 2019). there are many studies on the remediation of heavy metal contaminated soil by hyperaccumulation plants, but there are relatively few studies on the removal of chromium. the research on stabilization of chromium contaminated soil mainly focused on chemical stabilization and microbial stabilization. chemical stabilization technology has been widely used in the remediation of chromium contaminated soil, and the current research focus has gradually shifted from the optimization of chemical agents to the long-term safety assessment of the soil after remediation (frey et al., 2019; wang et al., 2018). the research on microbial stabilization began with the discovery of a cr (vi) reducing pseudomonas strain in the 1970s. at present, dozens of genera including bacteria, actinomycetes, fungi and molds have been found to have cr (vi) reducing ability. combined with other technologies. there are many cases of electric technology combined with other technologies to repair contaminated soil, mainly in the following ways: a. electric technology combined with prb. the principle of this technology is to transform cr (vi) into cr (iii), and the research content is to fill prb medium with zero valent iron and ferric oxide (nan & yan, 2019; sakae et al., 2018; heckmann et al., 2018). the results show that the latter is better; b. ultrasonic enhanced electrokinetic method can improve the migration ability of pollutants without any substances (you et al., 2018). at present, this method is often used to remove heavy metals from the soil; c. electrokinetic and biological (plants, microorganisms) joint remediation of chromium residue contaminated soil. direct current was applied to make indian mustard absorb heavy metals in the soil polluted by compound heavy metals (yang et al., 2019); d. using this technology to remove total petroleum hydrocarbons in soil, it is found that the removal rate of total petroleum hydrocarbons by electrokinetic method alone is 55%, and the removal rate of total petroleum hydrocarbons by fenton electrokinetic method combined with iron electrode is 97%. materials and methods sources and properties of the tested soils cr(iii)preparation of contaminated kaolin the experimental kaolin was purchased from the superfine calcined kaolin of datong jinyuan kaolin co., ltd. accurately weigh 6.00kg of the kaolin in the polyethylene box; weigh 15.37g of chromium trichloride hexahydrate, dissolve and fix the volume to 2.00l volumetric flask, then add the solution and 2l deionized water into the polyethylene box with kaolin, mix evenly and balance for one week. chromium residue contaminated soil before the use of the soil, the chromium residue collected from a certain place in henan province by the laboratory polluted the soil. after air drying, the gravel and other debris were screened out, and the soil was powdered by zn-08l small-scale pulverizer, and then the tested soil samples were screened through a 100 mesh screen for standby. the basic properties of chromium contaminated soil are shown in table 1. it can be seen from table 1 that the total chromium concentration of the tested soil sample is especially high, nearly 30000 mg/kg. chromium mainly 249remediation of high concentration chromium contaminated soil by enhanced electrodynamic method exists in the form of cr (vi) and cr (iii), accounting for 17% and 83% respectively. by using the improved bcr extraction method, the proportion of various forms in the tested soil samples is 25.1% in acetic acid extractable state, 5.5% in reducible extractable state, 13.7% in oxidizable state and 55.7% in residual state. experimental equipment experimental device the electrokinetic oxidation and electrokinetic prb experiments were carried out by using the electrokinetic device. in the electrokinetic prb experiment, a prb medium with a width of 5cm was added at 5cm away from the anode. the device is made of organic glass, which mainly includes the following three parts: anode electrolysis chamber, cathode electrolysis chamber and soil chamber. the size of anode and cathode electrolysis chamber is: length width height cm cm cm × × = × ×20 10 10 , and the size of soil chamber is: length width height cm cm cm × × = × ×30 20 10 . the same rod-shaped graphite electrode and electrolyte were placed in the anode and cathode electrode chambers, and the soil samples were evenly placed in the soil chamber. in addition, the device is connected with dc power supply (adjustable voltage), ammeter and peristaltic pump. the alkali solution tank and acid solution tank continuously add sodium hydroxide and acetic acid solution to the cathode chamber through peristaltic pump to balance the experimental ph; the 250 mesh filter screen is mainly used to block the electrode chamber and soil chamber to prevent soil from entering the electrode chamber; the collection tank is mainly used to collect anode and cathode electrolyte during and after the electromotion. instruments for experiment the instruments and equipment required for the experiment are shown in table 2: table 2. instruments and equipment required for experiment serial number name of instrument and equipment model 1 electric experimental device 30×20×10cm plexiglass column 2 ammeter 1~500ma 3 digital acidity meter fe20 4 conductivity meter s230 5 electrothermal constant temperature blast drying oven dh-101-3 bs 6 high speed centrifuge lg10-2.4a 7 atomic absorption spectrophotometer aa6300 8 electronic balance al204 9 peristaltic pump bt100-2j 10 electric heating plate eg20a plus 11 magnetic heating stirrer 78-1 12 thermostatic oscillator hz-9511k 13 small crusher zn-08l 14 visible spectrophotometer 722sp 15 electronic scale uwa-k015 16 ultraviolet visible spectrophotometer 752 experimental process batch test of potassium permanganate oxidation of cr (iii): the batch test was carried out in a 100ml conical flask with stopper, in which 100ml crcl3•6h2o of certain concentration was added, the ph value was adjusted (0.1mol/l hcl and naoh were used to adjust the ph value), and then a certain amount of kmno4 was added. put the conical flask into the shaking table (rotate speed 110r / min) to vibrate the reaction, take samples regularly, and determine the absorbance of cr (vi) generated by oxidation. fenton reagent oxidation of cr (iii) batch test: the oxidant was replaced by fenton reagent, and the others were the same as potassium permanganate oxidation of cr (iii). electric oxidation enhanced repair experiment: the 250 mesh filter cloth was fixed on the inner side of the porous filter screen of the reactor with a long tail clamp, and then the balanced 5 mesh filter cloth was accurately weighed with an electronic scale kgcr (iii) soil preparation or treated chromium slag contaminated soil was placed in the soil sample chamber of the electric device, the overflow port in the electrolysis chamber was blocked, 0.1mol/l kcl solution was added into the anode and cathode electrode chambers, and a certain concentration of potassium permanganate solution was continuously added into the cathode by using a circulating pump according to the conditions. after placing the electric soil and electrolyte, insert the carbon rod electrode connected to the digital display power into the electrolyte, power on, and apply a dc voltage of 1vgcm at both ends of the soil sample. during the test, acetic acid solution was dripped into the cathode electrode chamber by peristaltic pump to control the ph of the cathode solution to be acidic, and the current was recorded at intervals. after four days of electric driving, the electrolyte and soil samples were collected, and the soil samples were divided into 10 parts according to figure 1. take 100-200g of each sample for drying, grinding, passing through 100 mesh sieve, and determine ph, conductivity, cr (vi) and total chromium. figure 1. soil sample sampling diagram after oxidation electromotive test electric prb enhanced remediation of chromium residue contaminated soil experiment: 5 kg of chromium residue contaminated soil was accurately weighed by electronic scale after grinding and screening, put into polyethylene container, added 1.12 mol / l citric acid and deionized water, stirred evenly, acidified for about 4 days. 5 cm thick prb medium was added into the soil chamber 5 cm away from the anode. 3.5 kg chromium slag was added to the contaminated soil in other parts of the soil chamber, and 0.1 mol / l kcl solution was added into the electrolysis chamber. after 9 days, samples were taken as shown in figure 2 to determine ph, conductivity, cr (vi), total chromium and other indicators. accurately weigh 4g of soil sample in a 100ml beaker, add 50ml of certain concentration of potassium chloride, conduct electromagnetic stirring for 5min, centrifugation, pour the supernatant into a volumetric flask of certain volume, repeat the above steps, combine the supernatant and fix the volume. the concentration of cr (vi) in solution was determined by diphenylcarbazide spectrophotometry (gb 7467-87), and cr (vi) in soil was calculated. take 100 mesh 0.5000 g (accurate to 0.0001 g) soil sample, put it in a polytetrafluoroethylene crucible, digest it with hydrochloric acid, nitric acid, hydrofluoric acid and perchloric acid on the electric heating plate, transfer table 1. physical and chemical properties of chromium residue contaminated soil physical and chemical unit ph conductivity(ms/cm) saturated water content (%) w(cr(t))(mg/kg) w(cr(vi))(mg/kg) size 9.3-10.2 6.1-8.5 40 29291±50 4980±50 250 lin guo, xiaocun zhang, yao chen, jun liu it to a 50 ml colorimetric tube, add 5 ml 10% ammonium chloride solution, fix the volume, shake well, dilute, and inject the sample to determine the total chromium content. take 10g soil sample into centrifuge tube, add 25ml deionized water, mix well, stand still, use digital acidity meter to measure the ph of solution. after several days of electric driving, the following formula is used for calculation: e u idt= ∫ (1) in the formula, represents the power consumption of electric repair (kw·h), represents the voltage (u), represents the current intensity (a), and represents the operation time (h). take 5g soil sample into centrifuge tube, add 25ml deionized water, mix well, stand still, and measure soil conductivity with conductivity meter. the batch experiments of potassium permanganate and fenton oxidation of cr (iii) were carried out in order to select suitable oxidants and oxidation reaction conditions. the experimental conditions include reaction ph, time, temperature, oxidant concentration, etc. the reaction kinetics of potassium permanganate oxidation of cr (iii) was also studied. on this basis, the electrokinetic oxidation enhanced repair experiment was carried out. analysis of experimental results the effects of ph, time, temperature and oxidant concentration on the oxidation of cr (iii) were studied by the reaction of 50.00 mg/l cr (iii) with a certain concentration of potassium permanganate. the reaction kinetics of cr (iii) oxidation by potassium permanganate was also studied. effect of ph on oxidation of cr (iii) by potassium permanganate at relatively high ph, permanganate reacts to form manganese dioxide, which is expressed as follows: mno e h o mno oh ph4 2 23 2 4 3 5 12 − − −+ + = + = . (2) potassium permanganate can react to form mn2 + at lower ph. the reaction equation is as follows: mno e h m h o4 2 25 4 − − ++ = + >+8 n+ ph 3.5 (3) when ph is higher than 12, the following reactions will occur: mno e mno ph4 4 2 12− − ++ = > (4) it can be seen that potassium permanganate has strong oxidizability under acidic conditions. in addition, when ph is higher than 7, cr (iii) will exist in the form of precipitation, which makes it more difficult for potassium permanganate to oxidize cr (iii). therefore, in order to study the optimum ph for the oxidation of cr (iii) by potassium permanganate, the range of ph greater than 7 was directly omitted, and the experimental ph range was set as 1 ~ 7. the traditional electrokinetic remediation technology only migrates and concentrates the mobile target pollutants into the electrolytic cell, and the removal efficiency of most of the difficult to migrate cr (iii) in the chromium residue contaminated soil is low. therefore, in order to improve the removal rate of cr (iii) and total cr in the soil polluted by chromium slag, and to repair the soil polluted by chromium, the actual soil polluted by chromium slag was selected as the test soil. two enhanced electrokinetic remediation methods, electrokinetic oxidation enhanced remediation technology and electrokinetic prb enhanced remediation technology, were studied in order to provide technical support for enhanced electrokinetic remediation of chromium residue contaminated soil. a. a series of batch tests were carried out to select the suitable oxidant and the optimum oxidation reaction conditions. the effects of ph, time, temperature and oxidant concentration on the oxidation of cr (iii) by potassium permanganate and fenton reagent were studied. b. potassium permanganate oxidation enhanced electrokinetic remediation of cr (iii) contaminated kaolin was carried out, in which the prepared kaolin cr (iii) concentration was 500 mg · kg-1. the experiment mainly studied the feasibility of potassium permanganate enhanced electrodynamic chromium removal, and analyzed the current, ph, soil moisture content, potassium permanganate concentration and other factors. c. the experiment of potassium permanganate oxidation enhanced electrokinetic remediation of chromium residue contaminated soil was carried out to study the changes of ph, moisture content, cr (vi) and total chromium of chromium residue contaminated soil after electrokinetic oxidation. d. in order to further study the mechanism of enhanced remediation of chromium residue contaminated soil by electrokinetic prb, fe3o4, fe3o4 + activated carbon, fe3o4 + hydroxy alumina were used as prb media, respectively. the test results are shown in figure 3. figure 3. effect of ph on cr (vi) production it can be seen from figure 3 that under acidic conditions (ph = 1.0-6.0), the amount of cr (vi) is more and the difference is not big. the highest cr (vi) concentration was 47.4 mg / l at ph 3.0. the oxidation rate of cr (iii) reached 94.8% from the initial cr (iii) concentration of 50.00 mg / l before the reaction and the conservation of element mass. the lowest cr (vi) formation rate was found at ph 6 in acid condition, but it also reached 41.55% the results show that the oxidation rate of cr (iii) is 83.1%, and the cr (vi) concentration is 32.33 mg / l under neutral condition. therefore, the optimal ph of the experiment is 1-6. although the h + generated at ph 6 may be oxidized by permanganate to increase the consumption of potassium permanganate, considering the economy, the formation rate of cr (vi) and the follow-up experiment, the follow-up experiment chooses ph 6. figure 2. soil sample sampling after electric prb joint repair test 251remediation of high concentration chromium contaminated soil by enhanced electrodynamic method effect of different initial cr (iii) concentration on oxidation four different initial concentrations of cr (iii) (10.00 mg/l, 20.00 mg/l, 35.00 mg/l, 50.00 mg/l) were set to react with 2 g/l potassium permanganate at ph 6, temperature 20 °c and rotating speed 120 r/min. the experimental results are shown in figure 4. figure 4. variation of the amount of cr6 + generated by cr3 + at different initial concentrations with time it can be seen from figure 4 that with the progress of time, the formation of cr (vi) in four different reactions is significantly accelerated in 0 ~ 60 min, that is, cr (iii) and potassium permanganate react rapidly in this period. after 60 min, the amount of cr (vi) increased, but the change was not obvious. at this time, the reaction rates of the above four different initial cr (iii) concentrations were more than 80%. therefore, the sample was taken at 60 min in the followup experiment. the kinetics of potassium permanganate oxidation of cr (iii) was studied. different sampling time (0, 5, 10, 20, 25, 30, 45, 60 min) was set in 0-60 min. the sampling analysis results are shown in figure 5. figure 5. effect of potassium permanganate concentration on the first order kinetic constant it can be seen from figure 5 that the cr (vi) concentration gradually increases with the reaction. the pseudo first-order dynamic simulation was carried out during this period. the oxidation of cr (iii) by potassium permanganate followed pseudo first order kinetics under the condition of 2g / l potassium permanganate, the results are shown in formula (5): − ( ) = ( ) + + d c dt c r k robs 3 3 (5) in the formula, cr3 is the concentration of cr (iii) at time t, that is, ct. therefore, r2 and kobs (pseudo first order kinetic constants) can be obtained by plotting lnct against time t . the results are shown in table 3. table 3. pseudo first order kinetic parameters of potassium permanganate oxidation of cr (iii) initial cr3 concentration 10mg/l 20mg/l 35mg/l 50mg/l kobs 0.035 0.038 0.098 0.205 r2 0.924 0.969 0.966 0.965 it can be seen from table 3 that when cr (iii) concentrations are 10.00 mg / l, 20.00 mg / l, 35.00 mg / l and 40.00 mg / l respectively, adding potassium permanganate with concentration of 2g / l, the pseudo first-order kinetic constants of the reaction are small, but they slowly increase, and the r2 of four different initial cr (iii) concentrations are all above 0.9. therefore, the pseudo first order kinetics can be used to describe the reaction of cr (iii) with excessive potassium permanganate. in order to further study the effect of temperature on the oxidation of cr (iii) by potassium permanganate, four different temperatures of 10 °c, 20 °c, 30 °c and 40 °c were set. 50 mg/l cr (iii) was oxidized by 2 g/l potassium permanganate. the reaction was carried out at ph 6 and 120 r / min. samples were taken at different intervals. the results are shown in figure 6. figure 6. variation of cr (vi) formed by oxidation of cr (iii) by potassium permanganate with time at different temperatures temperature has a great influence on the oxidation of cr (iii) by potassium permanganate. it can be seen from figure 6 that at 313 k, the reaction end point is close to 10 min. in comparison, at 283 k, the reaction begins to approach the end point after 90 min. that is to say, the higher the temperature is, the more favorable the reaction is. therefore, the oxidation of cr (iii) by potassium permanganate is an endothermic reaction. 252 lin guo, xiaocun zhang, yao chen, jun liu enhanced remediation of chromium residue contaminated soil by electrokinetic prb because of the wide source and low price of aluminum salt, and through the experiment of adsorption of cr (vi) by hydroxy alumina, we know that hydroxy alumina has a strong adsorption performance for cr (vi). therefore, hydroxyl alumina can be used in prb medium in the experiment of enhanced remediation of chromium contaminated soil by electrokinetic prb. in order to further study the method of enhanced electrodynamic chromium removal and the specific mechanism of electrodynamic prb enhanced remediation, the design conditions of electrodynamic remediation for a chromium residue contaminated soil in henan province are shown in table 4. the length, width and height of the filled prb medium are about 20 cm, 5 cm and 10 cm respectively. before electrokinetic remediation, the ph value of chromium residue contaminated soil was above 9.5. according to the experimental conditions of electrokinetic prb enhanced remediation of chromium residue contaminated soil as shown in table 4, after electrokinetic remediation for 9 days, the ph value of each part of chromium residue contaminated soil of three groups of electrokinetic prb enhanced remediation decreased, ranging from 9.0 to 9.25. the specific distribution of ph in soil is shown in figure 7. figure 7. ph changes of soil at different distances from anode after electric driving before electrokinetic remediation, the ph value of chromium residue contaminated soil was above 9.5. according to the experimental conditions of electrokinetic prb enhanced remediation of chromium residue contaminated soil as shown in table 4, after electrokinetic remediation for 9 days, the ph value of each part of chromium residue contaminated soil of three groups of electrokinetic prb enhanced remediation decreased, ranging from 9.0 to 9.25. the specific distribution of ph in soil is shown in figure 7. the ph distribution in figure 7 is mainly due to h+ produced by anode electrolysis. h+ moves to soil pores under the action of electroosmosis and electromigration, which makes the ph of soil near anode lower than that near cathode. although oh is produced by cathode electrolysis, the ph of electrolyte is always acidic due to the existence of ph control system, which makes most of ohproduced by cathode neutralize with acetic acid, so that the ph of soil near cathode does not exceed the initial ph of soil. in addition, the ph control system makes the ph of anode and cathode electrolyte always maintain acidic, and the migration speed of h+ is faster than that of oh-, but the ph of the soil is still alkaline after electrokinetic, which is mainly due to the strong alkaline of the tested soil itself. conclusions and prospect a. based on the results of oxidation of cr (iii) by potassium permanganate and fenton reagent, the amount of cr (vi) produced by fenton reagent is only about 9 mg / l when the initial cr (iii) is 200 mg/l and the condition is relatively suitable, which shows that the ability of fenton reagent to oxidize cr (iii) is poor. relatively speaking, when the initial cr (iii) concentration is 50 mg / l, 20 °c, ph is 6, potassium permanganate is 2g/l, and the reaction time is 60 min, the cr (vi) production has reached about 45 mg/l, and the removal rate is as high as 90%. therefore, potassium permanganate is selected as the oxidant in the subsequent electric remediation of chromium residue contaminated soil. b. the oxidation of cr (iii) by potassium permanganate occurred under acidic conditions. when the ph was 3, the concentration of cr (vi) was 47.4 mg/l, and the oxidation rate of cr (iii) was 94.8%; c. the oxidation of cr (iii) by potassium permanganate is greatly affected by the concentration and temperature of potassium permanganate. with the increase of potassium permanganate concentration and reaction temperature, the formation of cr (vi) first increases and then tends to be stable. the reaction between potassium permanganate and cr (iii) is a pseudo first order reaction depending on the concentration of potassium permanganate. when the concentration of potassium permanganate is not excessive, the reaction kinetics conforms to the second order kinetics the ph control system makes the ph of anode and cathode electrolyte always keep acidic, and the migration speed of h+ is faster than that of oh-, and the ph of soil near anode is lower than that near cathode. enhanced electrodynamic method can improve the removal efficiency of total chromium to a certain extent, but there are also some problems as follows a. potassium permanganate added to the soil will have a certain impact on the physical and chemical properties of the soil, and damage the quality of the soil to a certain extent. therefore, what kind of soil is suitable for electrooxidation technology and how to properly treat the soil added with potassium permanganate after electrooxidation are worthy of further study. b. in the process of electrooxidation, there is still a lot of research space in the form of cr (iii) oxidized by potassium permanganate and the change of soil cr form after electrooxidation. c. in the electrokinetic test, fe3o4, hydroxyl alumina and activated carbon are used as prb media, how long the repair effect can last, and it needs to increase the electrokinetic time for further research; when the prb media fails, how to regenerate and realize the recycling of the media is also a direction of further research. acknowledgement the research is supported by: shaanxi provincial natural science foundation: study on contaminated soil remediation and comprehensive utilization of resources in tailing ponds based on opto-mineral-agri complementary technology, no.2019sf-246; national natural science foundation of china: water adaptation mechanism of plants with different root depth to thin purple soil slope, no.41471188. table 4. experimental results of enhanced remediation of chromium residue contaminated soil by electric prb number time/d voltage/v cathode liquid ph control of anode and cathode prb media anode solution 1 9 30 0.1m/l kcl ph control test of naoh / acetic acid fe3o4 0.1m/l kcl 2 9 30 0.1m/l kcl ph control test of naoh / acetic acid fe3o4+hydroxyl aluminum(mass ratio 2:1) 0.1m/l kcl 3 9 30 0.1m/l kcl ph control test of naoh / acetic acid fe3o4+activated carbon(mass ratio 2:1) 0.1m/l kcl 253remediation of high concentration chromium contaminated soil by enhanced electrodynamic method references cleveland, c. c., & townsend, a. r. (2019). nutrient additions to a tropical rain forest drive substantial soil carbon dioxide losses to the atmosphere, proceedings of the national academy of sciences of the united states of america, 103(27), 10316-10321. frey l. j., vorländer, d., rasch, d., ostsieker, h., muller, b., schulze, m., schenkendorf, r., mayr, t., grosch, j. h., & krull, r. (2019). novel electrodynamic oscillation technique enables enhanced mass transfer and mixing for cultivation in micro-bioreactor, biotechnology progress, 35(5), 2827-2835. heckmann, j., pufahl, k., franz, p., grosse, n. b., li, x., woggon, u. (2018). plasmon-enhanced nonlinear yield in the otto and kretschmann configurations, physical review. b, condensed matter and materials physics, 98(11), 1-8. jourdan, p. m., lamberton, p. h. l., fenwick, a., & addiss d. g. (2018). soil-transmitted helminth infections, lancet (london, england), 391(10117), 252-265. jurkevitch, e., minz, d., ramati, b., & barel, g. (2019). prey range characterization, ribotyping, and diversity of soil and rhizosphere bdellovibrio spp. isolated on phytopathogenic bacteria, applied and environmental microbiology, 66(6), 2365-2371. marchiori, a., li, y., & evans, j. (2019). design and evaluation of iotenabled instrumentation for a soil-bentonite slurry trench cutoff wall, infrastructures, 15(12), 59-64. nan, f., & yan, z. (2019). silver-nanowire-based interferometric optical tweezers for enhanced optical trapping and binding of nanoparticles, advanced functional materials, 29(7), 1-8. ojanen, p., minkkinen, k., alm, j., penttila, t. (2018). corrigendum to “soil– atmosphere co 2, ch 4 and n 2 o fluxes in boreal forestry-drained peatlands” [for. ecol. manage. 260 (2010) 411–421], forest ecology & management, 412(15), 95-96. rinella, m. j., & reinhart, k. o. (2017). (2020). mixing soil samples across experimental units ignores uncertainty and generates incorrect estimates of soil biota effects on plants: response to cahill etal. (2017) ‘no silver bullet: different soil handling techniques are useful for different research questions, exhibit differential type i and ii error rates, and are sensitive to sampling intensity’, the new phytologist, 216(1), 15-17. sakae, y., straub, j. e., & okamoto, y. (2018). enhanced sampling method in molecular simulations using genetic algorithm for biomolecular systems, journal of computational chemistry, 12(15), 58-65. ser, h. l., ab mutalib, n. s., yin, w. f., chan k. g., goh, b. h., & le, l. h. (2020). evaluation of antioxidative and cytotoxic activities of streptomyces pluripotens musc 137 isolated from mangrove soil in malaysia, frontiers in microbiology, 6(10), 1398-1406. shi, y., li, y., xiang, x., sun, r., yang, t., he, d., zhang, k., ni, y., zu, y. g., adams, j. m., & chu, h. (2018). spatial scale affects the relative role of stochasticity versus determinism in soil bacterial communities in wheat fields across the north china plain, microbiome, 6(1), 27-34. walker, x. j., baltzer, j. l., cumming, s. g., day, n. j., johnstone, j. f., rogers, b. m., solvik, k., turetsky, m. r., & mack, m. c. (2018). soil organic layer combustion in boreal black spruce and jack pine stands of the northwest territories, canada, international journal of wildland fire, 27(2), 125-134. wang, k., lorenz, r. d., & baloch, n. a. (2018). enhanced methodology for injection-based real-time parameter estimation to improve back emf self-sensing in induction machine deadbeat-direct torque and flux control drives, ieee transactions on industry applications, 54(6), 6071-6080. watson, g. k., & cain, r. b. (2019). microbial metabolism of the pyridine ring. metabolic pathways of pyridine biodegradation by soil bacteria, the biochemical journal, 146(1), 157-172. yang, x., xia, x., zhang, z., nong, b., zeng, y., wu, y., xiong, f., zhang, y., liang, h., pan, y., dai, g., deng, g., & li, d. (2019). identification of anthocyanin biosynthesis genes in rice pericarp using pcamp. plant biotechnol journal. 17(9), 1700-1702. you, z., li, l., lu, j., & ge, h. (2018). integrated tempering enhanced sampling method as the infinite switching limit of simulated tempering, the journal of chemical physics, 149(8), 15-23. zehe, e., blume, t., & bloschl, g. (2018). the principle of ‘maximum energy dissipation’: a novel thermodynamic perspective on rapid water flow in connected soil structures, philosophical transactions of the royal society of london, series b, biological sciences, 365(1545), 13771386. zohdi, t. i. (2018). electrodynamic machine-learning-enhanced fault-tolerance of robotic free-form printing of complex mixtures, computational mechanics, 12(12), 89-95. https://xueshu.baidu.com/s?wd=author:(dg addiss) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(kg chan) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(l learn-han) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(l learn-han) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person gjdgxs 30j0zll 1fob9te 3znysh7 tyjcwt 4d34og8 26in1rg lnxbz9 35nkun2 1ksv4uv 44sinio 2jxsxqh z337ya keywords: gps; gnss; point positioning; ann; estimation of the l2 carrier observables. palabras clave: gps; gnss; posicionamiento; red artificial neuronal; estimación de los observables l2. how to cite item carletti negri, c. v., & lima segantine, p. c. (2020). estimation of gps l2 signal observables using multilayer perceptron artificial neural network for positional accuracy improvement. earth sciences research journal, 24(1), 97-103. doi: https://doi.org/10.15446/esrj.v24n1.78880 in recent decades, due to the increasing mobility of people and goods, the rapid growth of users of mobile devices with location-based services has increased the need for geospatial information. in this context, positioning using data collected by the global navigation satellite systems (multi-gnss) has gained more importance in the field of geomatics. the quality of the solutions is related, among other factors, to the receiver’s type used in the work. to improve the positioning with low-cost devices and to avoid additional user expenses, this work aims to propose the implementation of an artificial neural network (ann) to estimate the gps l2 carrier observables. for this, a network model was selected through the cross-validation (cv) technique, the observations were estimated, and the accuracy of the solutions was analyzed. the cv technique demonstrated that a multilayer perceptron with four intermediate layers and one with one intermediate layer are the most appropriate configurations for this problem. the dual-frequency rinex processing (with artificial data) revealed significant improvements. for some tests, it was possible to comply with the rural property georeferencing regulations of the brazilian national institute of colonization and agrarian reform (incra). the results indicate, therefore, that the methodological proposal of the present investigation is very promising for approximating the quality of positioning reachable using a dual-frequency receiver. abstract estimation of gps l2 signal observables using multilayer perceptron artificial neural network for positional accuracy improvement estimación de los observables de señal gps l2 utilizando redes neuronales artificiales de perceptrón multicapa para mejorar la precisión posicional issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.78880 en las últimas décadas, debido a la creciente movilidad de personas y bienes, el rápido crecimiento de los usuarios de dispositivos móviles con servicios basados en la ubicación ha aumentado la necesidad de información geoespacial. en este contexto, el posicionamiento utilizando los datos recopilados por los sistemas globales de satélite de navegación (multi-gnss) ha ganado más importancia en el campo de la geomática. la calidad de las soluciones está relacionada, entre otros factores, con el tipo de receptor utilizado en el trabajo. para mejorar el posicionamiento con dispositivos de bajo costo y evitar gastos adicionales del usuario, este trabajo tiene como objetivo proponer la implementación de una red neural artificial (ann) para estimar los observables del operador gps l2. para esto, se seleccionó un modelo de red a través de la técnica de validación cruzada (cv), se estimaron las observaciones y se analizó la precisión de las soluciones. la técnica cv demostró que un perceptrón multicapa con cuatro capas intermedias y uno con una capa intermedia son las configuraciones más apropiadas para este problema. el procesamiento rinex de doble frecuencia (con datos artificiales) reveló mejoras significativas. para algunas pruebas, fue posible cumplir con las regulaciones de georreferenciación de propiedad rural del instituto nacional de colonización y reforma agraria (incra). los resultados indican, por lo tanto, que la propuesta metodológica de la presente investigación es muy prometedora para aproximar la calidad de posicionamiento accesible utilizando un receptor de doble frecuencia. resumen record manuscript received: 04/04/2019 accepted for publication: 15/11/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 97-103 cassio vinícius carletti negri, paulo cesar lima segantine university of são paulo (usp) * corresponding author: cassiovcn@usp.br g ps m e a su r e m e n t s 98 cassio vinícius carletti negri, paulo cesar lima segantine introduction the increasing mobility of people and goods, as well as the rapid growth of users of mobile devices with location-based services, significantly expands the need for geospatial information gathered from artificial earth satellites. to address this challenge, the integrity of the multi-gnss (us american gps, russian glonass, european galileo, and chinese beidou) propagated into national or international positioning services is of great importance. the gps (global positioning system) can be highlighted here due to the stability of the system over a long period of time and the robustness of its signals. initially, its satellites emitted the signals in two frequencies (l1 and l2) only. over the years, on the basis of its modernization program, the outdated satellites of the space segment of the gps were replenished by satellites of new generations emissions as well new navigation signals that were added to the existing signals (leick, rapoport, and tatarnikov, 2015; segantine, 2005). however, one of the advantages of dual or triple-frequency receivers is the possibility of making linear combinations between the observables of different carriers to improve the positioning accuracy by reducing the ionospheric effect and by also eliminating the errors of receiver and satellite clocks. in recent years, to avoid the high costs associated with the use of double or multifrequency receivers, some studies have been conducted to assess the possibility of using superficial artificial neural networks (anns) to generate l2 carrier observations from data gathered by a single-frequency receiver within a continuous monitoring network (mazzetto, 2017; c. a. u. da silva, 2003). although some unsatisfactory results were obtained in certain tests during the learning process of the networks, in general, the relative positioning solutions estimated using the generated l2 observables were more accurate than those using the original l1 data only. this suggests that it is possible to model the behavior of gps observations for up to 30 minutes to generate the l2 carrier observables, improving the positioning accuracy of low-cost single-frequency receivers. according to hornik, stinchcombe and white (1989), a shallow multilayer perceptron (mlp) can approximate any function at a particular accuracy level. however, bengio and lecun (2007) argue that although networks with one hidden layer have advantages, deeper models can represent some systems more efficiently because they can learn lower-level abstractions (characteristics of the characteristics). in addition, goodfellow, bengio and courville (2016) have shown from empirical tests that anns with two or more hidden layers have a better generalization capacity than the simpler models. thus, this paper aims to implement anns with more than one hidden layer, through the tensorflow library in the python computer language, to estimate the gps l2 carrier observables. in addition, the present investigation seeks to validate a network topology through the cross-validation (cv) technique by random sampling and compare the results with those obtained in other research. literature review this research requires the union of two areas of knowledge: geodesy, with a gps problem in geomatics, and computer science, with anns being a possible solution. due to this association, it became necessary to detail the state of the art of anns and their applications in the area of geomatics and similar areas. starting with the ann theme, mcculloch and pitts (1943) developed the concept of the first artificial model of a biological neuron, exhibiting its computational capacities. posteriorly, rosenblatt (1958) proposed a new model named perceptron. in the following years, widrow and hoff (1960) created the delta learning rule to minimize the errors of an ann based on gradient descent. in 1982, hopfield’s proposition of a recurrent ann gave the area quite prominence, and in the following years, the backpropagation algorithm creation (rumelhart, hinton, and williams, 1986) and computational advances have consecrated the ann studies. several practical applications have since been performed with anns. among the research developed, there is a current that relates them to the satellite positioning segment with the gnss. in mosavi (2004), anns were used in the estimation of observations corrections for differential global positioning system (dgps). furthermore, kubik and zhang (1997) and indriyatmoko et al. (2008) also applied the ann tool to predict the dgps corrections. jwo and chin (2002) used a superficial ann to determine the precision geometric dilution (gdop) parameter, and jwo and lai (2007) conducted tests with several architectures and topologies, trying to select the ideal configuration to estimate gdop values. the results showed that all candidate networks met expectations regarding the accuracy of their outputs, but those trained by the backpropagation algorithm presented a longer learning process. in addition, a recurrent wavelet neural network (rwnn) was implemented in mosavi (2007) to reduce gps receiver time data noise and to model the errors of this source. anns were also applied to improve positioning during gps signal blocking. semeniuk and noureldin (2006), for instance, created an alternative method based on an ann that estimates the speed and position errors of an inertial navigation system to correct and improve positioning in locations where the gps system is not available. more recently, a hybrid between a wavelet neural network (wnn) and the strong tracking kalman-filter (stkf) was proposed in an attempt to improve the positioning by satellites during this type of situation (chen et al., 2013). the combination of these two tools in the work of chen et al. (2013) was based on the wnn outstanding performance in data prediction and on the capacity of the stkf to estimate inertial navigation system (ins) errors. thus, it has been proposed to represent a sequential process because when there is no blockage in the gps signal, the wnn is trained using the predicted prior positioning errors, which were provided by the stkf, and it generates, as an output, the current positioning error. approaching the topic of this paper, silva (2003) sought to improve the positioning with single-frequency receivers, creating a method to estimate the gps l2 carrier observables using an ann. for estimating the l2 carrier phase, an mlp with one hidden layer and 6 artificial neurons was chosen; for estimating the l2 carrier code, the author adopted the same architecture but with 30 artificial neurons in one hidden layer. both configurations were implemented in matlab 6.1 and trained by the levenberg-marquardt algorithm, reaching up to 5,000 iterations. the results of silva (2003) indicate that the tool can successfully model the observable behavior. in another investigation, leandro, silva and santos (2007) evaluated the best method for feeding an ann with gps observations. in addition, mazzetto (2017), adopting the same topologies indicated by silva (2003), proposed a method in which only the signals with higher strength obtained in four stations near the receiver, where it is desired to estimate the observables, should be used because, theoretically, they experience the same atmospheric effects. in this research, ann training processes required up to 8 minutes and 15,000 iterations to complete. although so much time has been expended, some validation mse values were quite high and not converging as expected. despite these factors, he was able to increase the prediction time of the gps observables and improve the positioning accuracy in certain tests performed. the literature review showed that the studies used only an intermediate layer ann to estimate the gps l2 carrier observables. in addition, there is no ann topology validated through the cv technique to estimate the gps l2 carrier observables. therefore, the use of an appropriate ann model to estimate the l2 carrier data, capable of extracting the characteristics of the process more efficiently than a shallow mlp and being tolerant to the noises present in the input signals, represents an important step towards the enhancement of postprocessed solutions with single-frequency devices. materials and methods in this investigation, it was decided to implement the network in the tensorflow library through the python language. designed initially for machine learning applications, the low-level structure tool is currently being used in different fields and by several large companies (tensorflow, 2018). to select the appropriate ann model, the cv technique was applied, bypassing the training and validation samples, which were obtained in the dual-frequency rinex files generated in the inconfidentes/mg (mgin), piracicaba/sp (sppi), lins/sp (spli) and jaboticabal/sp (spja) stations. then, the ann was tested in the l1 carrier observables generated in the receiver fixed in são carlos/sp (eesc). figure 1 shows this configuration, chosen to compare the results obtained with those obtained in previous works, 99 estimation of gps l2 signal observables using multilayer perceptron artificial neural network for positional accuracy improvement in the receiver. in addition, the positioning solutions originated based on data postprocessing. for the extraction of the gps observables from the rbmc station rinex files, the rinex adapter software was used. this program was developed in the research of mazzetto (2017). in the creation of new rinex files with the artificial observables, the software deletes the original l2 carrier data and multiplies the l1 carrier phase values in the first epoch (from each satellite) by a factor k, or any other desired value, to determine the initial number of the l2 carrier phase since the ambiguity term is unknown. the generation of observations occurred differently for each type of observable, as these do not behave equally over time. thus, two ann models were implemented: one that used the phase differences of the l1 carrier as the input signal and the phase differences of the l2 carrier as the output. the second model used the c/a code observations as input signal and the p2 code data as the output, using the reverse truncation process. a similar method for treating observable gps for anns was used in leandro, silva and santos (2007). the topology for each ann, that is, the optimal number of hidden layers and neurons, the optimization algorithm and other parameters, was determined with tests that used the random subsampling cross-validation technique, where the available dataset was randomly subdivided into training and validation subsets five times (cv = 5), seeking to analyze the performance of each candidate towards data drawn differently for each test (haykin, 2008; i. n. silva, spatti, and flauzino, 2010). the new rinex files produced with the artificial observables were processed in geo office software v. 5, from leica, available at the department of transportation engineering, at são carlos school of engineering (university of são paulo), by the relative method to the eesc station with the other four stations as references. the positioning solutions were evaluated for the accuracy of the coordinates and compared with those produced by previous research. table 2 shows the official coordinates of the eesc rbmc station used to determine the accuracy of the positioning solutions. to evaluate the performance of the new model using exactly the same data used by mazzetto (2017). next, table 1 shows the test dates and times. the variation in the ionospheric refraction mainly depends on the latitude of the station, the angle of elevation, and the time of observation. near the magnetic equator, the ionospheric refraction varies more rapidly in time. hence, in our example, we use data of satellites that were received on different days between 01:00 and 03:30 pm due to strong solar activities. the mse values were calculated according to equation 1. mse n d yi i i n = ⋅ − = ∑1 1 2( ) (1) where: n: number of samples; di: observed output i (true value); yi: predicted output i. it is important to highlight that this proposed method for generating the l2 carrier observables was applied after the l1 carrier data were collected; that is, the estimation of these values was not performed in real-time with the ann figure 1. arrangement of rbmc stations used adapted from mazzetto (2017) table 1. test dates and times adapted from mazzetto (2017) test date gps time 1 09/15/2014 01:00 – 01:30 pm 2 01/10/2015 02:30 – 03:00 pm 3 06/10/2015 01:30 – 02:00 pm 4 01/20/2016 03:00 – 03:30 pm 100 cassio vinícius carletti negri, paulo cesar lima segantine the basic processing settings adopted in the geo office 5.0 software were l1 and l1 + l2 frequencies; hopfield tropospheric model; leica geo office standard (standard) v. 5 ionospheric model; broadcast ephemeris; elevation mask: 15º; and gps system data only. the gnss device models used in this work are the following: 1) eesc à receiver: leica gr10; antenna: leica ar10 (773758); 2) spja à receiver: trimble netr8; antenna: gnss choke ring (trm59800.00); 3) spli à receiver: trimble netr8; antenna: gnss choke ring (trm59800.00); 4) sppi à receiver: trimble netr8; antenna: gnss choke ring (trm59800.00); 5) mgin à receiver: trimble netr5; antenna: zephyr gnss geodetic model 2 (trm55971.00). the spatial deviations (sd) were calculated according to equation 2. sd n n e e a ao p o p o p= − + − − −( )² ( )² ( )² (2) where n0 = the official north coordinate, in meters; e0 = the official east coordinate, in meters; a0= the official geometric altitude coordinate, in meters; np = the processed north coordinate, in meters (relative positioning method); ep = the processed east coordinate, in meters (relative positioning method). ap = the processed geometric height coordinate, in meters (relative positioning method). results and discussions in this section, we first present the tests and results regarding the choice of the topologies of each ann model, validated by the cv technique by random sampling. then, the training times and the validation mse of each test are shown. finally, the spatial deviations of each positioning solution using the artificial observations were compared with those obtained in other works and with the single-frequency files. ann implementation the initial step was to develop the generic algorithm of a network, from the input data loading and normalization to the determination of the number of hidden layers, neurons and mathematical models. it is known that the optimal topology selection of an ann occurs through attempts and errors. each problem requires a different model. with the generic mlp, the tests were started with the observations obtained in the spja, spli, sppi and mgin rbmc stations. the evaluation of the best model was due to the search for lower validation mse values in the four experiments. carrier code ann model first, the intervals in which the data should be normalized was tested, choosing the limits between 0 and 1. this step was performed through the class minmaxscaler, provided by the package scikit-learn (scikit-learn, 2018). the restriction of the network inputs in relation to signal strength (0, 5 and 7), as successfully proposed in mazzetto (2017), was applied in the present investigation. however, inserting all the observations of the code without eliminating those of lower quality proved to be the best option, unlike his conclusion. this situation can be explained by the noise tolerance in the input signals that the newly implemented ann has developed. regarding the number of hidden layers and the number of neurons, several topologies were investigated. figure 2 shows the average validation mse values for each tested configuration. although the results are from one date, the other learning processes and the results obtained on other days were similar to those shown in the image below. 0 0.02 0.04 0.06 0.08 0.1 0 25 50 75 100 125 150 a ve ra ge v al id at io n m se (c v = 5 ) training epochs 25 30 35 15-3 23-7 25-10 25-15 18-10-5 25-12-7 14-8-4-3 18-8-7-5 20-10-7-5 figure 2. comparison between models with different numbers of hidden layers and artificial neurons day 01/20/2016 – fourth test day (03:00 03:30 pm) in black (30), the topology chosen in silva (2003), applied in mazzetto (2017), and in red (18-8-7-5), the configuration chosen in this investigation are shown. in relation to the tests performed to choose the activation function, figure 3 shows a test that analyzed the rectified linear unit (relu) and logistics (sigmoid) functions. it is observed that the relu produced lower validation mse values, although in some evaluations, both functions performed satisfactorily, but with relu converging faster. géron (2017) states that in practice, this function generally produces excellent outputs, is fast in processing, and works well in deep networks. 0.00 0.05 0.10 0.15 0.20 0 25 50 75 100 125 150 a ve ra ge v al id at io n m se (c v = 5 ) training epochs relu sigmoid figure 3. comparison of activation functions 01/10/2015 – second test day (02:30 03:00 pm) to select the optimization algorithms available in the tensorflow library, only the two most popular algorithms were used: adam and the stochastic gradient descent (sgd). based on figure 4 (mse values reached for table 2. official coordinates of the são carlos rbmc station (eesc) – ibge (2018a) station geographic coordinates (sirgas 2000) precision (m) eesc latitude 22 ° 00 ‘ 17.81599 “ s 0.001 longitude 47 ° 53 ‘ 57.04968 “ w 0.001 geometric height (m) 824.587 0.005 101 estimation of gps l2 signal observables using multilayer perceptron artificial neural network for positional accuracy improvement 09/15/2014), it is possible to verify that the former presented higher efficiency, converging rapidly. this result was expected because kingma and ba (2014) already demonstrated the effectiveness of the algorithm in training mlps, comparing it with four others. 0.00 0.05 0.10 0.15 0.20 0 25 50 75 100 125 150 a ve ra ge v al id at io n m se (c v = 5 ) training epochs adam sgd figure 4. comparison between optimization algorithms 09/15/2014 first test day (01:00 01:30 pm) the method for initializing the synaptic weights in each layer was also tested by choosing the variance scaling type, whose objective is to keep constant the variance scale of the input signals to the final layer of the network. figure 5 presents the average validation mse values for tests performed on 06/10/2015. 0.00 0.20 0.40 0.60 0 25 50 75 100 125 150a ve ra ge v al id at io n m se (c v = 5 ) training epochs glorot uniform random normal random uniform variance scaling figure 5. comparison between the initialization methods of synaptic weights day 06/10/2015 – third test day (01:30 02:00 pm) carrier phase ann model for the carrier phase, again using the minmaxscaler class from the scikit-learn library (scikit-learn, 2018), the data were normalized between the intervals -1 and 1, reducing the propagation of the error, as the normalization factor is smaller (leandro et al., 2007). in addition, at this stage of the research, it was chosen to restrict signal strength (intensity equal to 5), training the ann only with the highest quality observations. the tests on the number of hidden layers and neurons were limited to evaluating only one or two intermediate layer topologies since, above that, overfitting of the samples was verified. figure 6 shows the results obtained for the fourth test day (01/20/2016). in red (8), the topology selected in this work is highlighted, and in black (6), the configuration adopted by silva (2003) and used in mazzetto (2017) is shown. regarding the activation functions, two types were investigated: linear and hyperbolic tangent (tanh). figure 7 shows the results. based on all tests, it was found that the tanh function converged faster and produced the best positioning solutions and, for these reasons, was adopted for the ann of the carrier phase. regarding the optimization algorithms, adam and sgd were tested again, and the first algorithm was chosen because it provided the best results. figure 8 shows one of the tests carried out that corroborates this conclusion. 0.00 0.10 0.20 0.30 0.40 0.50 0 25 50 75 100 125 150 a ve ra ge v al id at io n m se (c v = 5 ) training epochs adam sgd figure 8. comparison between optimization algorithms 06/10/2015 – third test day (01:30 02:00 pm). finally, the last analysis verified which initialization methods of the synaptic weights generate the lowest validation mse values. according to 0.00 0.02 0.04 0.06 0.08 0 25 50 75 100 125 150 a ve ra ge v al id at io n m se (c v = 5 ) training epochs 5 6 8 10 11 3-2 5-2 5-3 6-2 6-4 figure 6. comparison between models with different numbers of hidden layers and artificial neurons day 01/20/2016 – fourth test day (03:00 03:30 pm) 0.00 0.04 0.08 0.12 0.16 0.20 0 25 50 75 100 125 150 a ve ra ge v al id at io n m se (c v = 5 ) training epochs linear function tanh function figure 7. comparison of activation functions day 09/15/2014 – first test day (01:00 01:30 pm). 102 cassio vinícius carletti negri, paulo cesar lima segantine figure 9, it is observed that starting weights randomly and with a uniform distribution generated good results. 0.00 0.03 0.05 0.08 0.10 0 25 50 75 100 125 150a ve ra ge v al id at io n m se (c v = 5 ) training epochs glorot uniform random normal random uniform variance scaling figure 9. comparison between the initialization methods of synaptic weights 01/10/2015 – second test day (2:30 3:00 pm). ann training and validation with the network configurations and all parameters selected, the l2 carrier phase and code observations of the eesc station were generated for the four days. table 3 shows the total sum of training epochs of the anns, the learning time and the validation mse for each observable.. table 3. total sum of the number of training epochs and time for generating the l2 carrier observables test date validation mse (p2 code) validation mse (l2 phase) training epochs duration (s) 1 09/15/2014 0.002882 0.003958 369 36.77 2 01/10/2015 0.004782 0.007323 219 29.28 3 06/10/2015 0.001514 0.034828 88 91.77 4 01/20/2016 0.000095 0.210098 23 4.95 it was verified that the mse performance parameter of the code presented very low values, close to zero, unlike the ones exposed in previous works (mazzetto, 2017), where these numbers were not even below the unit. in general, the number of training periods of anns, considering both models, represents only 1.9% of the number of iterations that were performed in past research for the same sample data, while the total process time equals only 8.1%. these numbers point to the optimization of the process of generating observations with anns, being considerably faster, even when adopting models with more than one hidden layer. gps data processing as the main objective of this paper was to evaluate the possibility of improving positioning solutions using anns, this section shows the results regarding the processing of gps data with artificial observables. parallels were made between the deviations obtained with the single-frequency solutions; the l1 + l2modif data, obtained in mazzetto (2017); and the l1 + l2a observables generated in the present investigation, aiming to prove the validity of the proposed method. figure 10 shows the spatial deviations (sd) obtained in processing for the first test day. for the second test (01/10/2015), spatial deviations are shown in figure 11. next, figure 12 shows the spatial deviations obtained on 06/10/2015 (test 3). finally, the spatial deviations of the processing performed on 01/20/2016 (test 4) are shown in figure 13. as the above figures demonstrated, applying the l1 carrier observations together with those of the l2 carrier, which were generated in this work, it was possible to reduce the positioning errors in several baselines (on average, a 40.2% reduction) when compared to the results obtained with only the original l1 data. therefore, it can be stated that the process of gps l2 carrier observables estimation using anns can improve relative positioning solutions with data obtained in single-frequency receivers. 1.348 2.016 0.936 1.0831.386 1.355 1.030 0.955 0.349 0.543 0.762 0.541 0.00 0.50 1.00 1.50 2.00 2.50 mgin-eesc spja-eesc spli-eesc sppi-eesc sp at ia l d ev ia tio n (m ) baseline l1 l1/l2modif mazzetto (2017) l1/l2a figure 10. spatial deviations from relative processing for 09/15/2014 (test 1) 1.655 0.664 1.782 0.7820.637 0.498 0.463 0.492 0.717 0.609 0.954 1.086 0.00 0.50 1.00 1.50 2.00 2.50 mgin-eesc spja-eesc spli-eesc sppi-eesc sp at ia l d ev ia tio n (m ) baseline l1 l1/l2modif mazzetto (2017) l1/l2a figure 11. spatial deviations from relative processing for 01/10/2015 (test 2) 2.141 2.182 1.136 1.911 0.969 1.026 0.938 0.699 1.060 0.275 0.497 0.91 0.00 0.50 1.00 1.50 2.00 2.50 mgin-eesc spja-eesc spli-eesc sppi-eesc sp at ia l d ev ia tio n (m ) baseline l1 l1/l2modif mazzetto (2017) l1/l2a figure 12. spatial deviations from relative processing for 06/10/2015 (test 3) 2.399 1.230 2.226 1.123 0.451 0.506 0.448 0.401 2.427 0.611 0.589 1.523 0.00 0.50 1.00 1.50 2.00 2.50 mgin-eesc spja-eesc spli-eesc sppi-eesc sp at ia l d ev ia tio n (m ) baseline l1 l1/l2modif mazzetto (2017) l1/l2a figure 13. spatial deviations from relative processing for 01/20/2016 (test 4) 103 estimation of gps l2 signal observables using multilayer perceptron artificial neural network for positional accuracy improvement conclusion dual-frequency receivers help to improve the positioning accuracy by enabling the formation of linear combinations between the observables of different carriers. this allows us to use differencing techniques between the phase or code measurements in data processing to reduce the ionospheric effect and to also eliminate the errors of receiver and satellite clocks. those so-called geodetic receivers are associated with high costs and, therefore, are waived by the service provider in many cases, especially in large-scale networks where a high number of these types of instruments are needed. in recent years, some studies have been carried out to check the achievable accuracy using the generated l2 carrier observations based on data gathered by a single-frequency receiver. this study aims to determine the quality of the gps l2 carrier observables estimated by applying anns to answer the question of whether a dual-frequency receiver can be replaced by a cost-efficient single-frequency receiver approaching its achievable accuracy. for this purpose, according to the literature review, for the first time, two ann topologies were validated and selected through the cv technique to estimate the gps l2 data, among a wide universe of possible configurations. concisely, the mlps selected in this research contain up to 4 hidden layers and 38 processing units. despite this, the training and validation process of the models was extremely fast, only a few seconds, and represented less than 10% of the time spent in other research. regarding the validation mse values, all numbers were very close to zero. it was verified that by applying the observations estimated by the anns, together with the first carrier originals, it was possible to achieve significant reductions of the spatial deviations in several baselines when compared to those obtained with only the l1 observable. in addition, some vectors presented smaller errors than those presented in mazzetto (2017), and others have complied with the incra rural georeferencing standards, which determine a plane deviation of less than 50 cm. however, it should be noted that it was not possible to apply the l3 (iono-free) combination in the relative processing of the dual-frequency rinex files, similar to what has been evidenced in previous research. for future research, the use of genetic algorithms in the search for an even better ann configuration for this problem is indicated, as they can analyze a vast number of candidates in a much shorter period of time than a human would need. additionally, as future research, it is proposed that an analysis be performed to identify which factors influence the final results of the gps data processing because, in some tests, the ann could not successfully estimate the observations of the l2, resulting in positioning accuracy degradation when using the artificial observables. acknowledgments this study was financed in part by the coordenação de aperfeiçoamento de pessoal de nível superior brasil (capes) finance code 001 (no. 1680400). references bengio, y., & lecun, y. (2007). scaling learning algorithms towards ai. large scale kernel machines, (1), 321–360. chen, x., shen, c., zhang, w. bin, tomizuka, m., xu, y., & chiu, k. (2013). novel hybrid of strong tracking kalman filter and wavelet neural network for gps/ins during gps outages. measurement, 46(10), 3847–3854. https://doi.org/10.1016/j.measurement.2013.07.016 géron, a. (2017). hands-on machine learning with scikit-learn and tensorflow (1st ed.). sebastopol: o’reilly media. https://doi. org/10.3389/fninf.2014.00014 goodfellow, i., bengio, y., and courville, a. (2016). deep learning (adaptive computation and machine learning series). adaptive computation and machine learning series. cambridge: mit press. haykin, s. (2008). neural networks and learning machines (3rd ed.). upper saddle river: pearson prentice hall. hopfield, j. j. (1982). neural networks and physical systems with emergent collective computational abilities. proceedings of the national academy of sciences, 79(8), 2554–2558. https://doi.org/10.1073/ pnas.79.8.2554 hornik, k., stinchcombe, m., & white, h. (1989). multilayer feedforward networks are universal approximators. neural networks, 2(5), 359– 366. https://doi.org/10.1016/0893-6080(89)90020-8 ibge. (2018). ibge | portal do ibge. retrieved may 16, 2018, from https:// www.ibge.gov.br/index.php indriyatmoko, a., kang, t., lee, y. j., jee, g. i., cho, y. b., & kim, j. (2008). artificial neural networks for predicting dgps carrier phase and pseudorange correction. gps solutions, 12(4), 237–247. https://doi. org/10.1007/s10291-008-0088-x jwo, d. j., & chin, k. p. (2002). applying back-propagation neural networks to gdop approximation. the journal of navigation, 55(1), 97–108. https://doi.org/10.1017/s0373463301001606 jwo, d. j., & lai, c. c. (2007). neural network-based gps gdop approximation and classification. gps solutions, 11(1), 51–60. https:// doi.org/10.1007/s10291-006-0030-z kingma, d. p., & ba, j. l. (2014). adam: a method for stochastic optimization. international conference on learning representations, 1–15. leandro, r. f., silva, c. a. u., & santos, m. c. (2007). feeding neural network models with gps observations: a challenging task. dynamic planet, 130, 186–193. leick, a., rapoport, l., & tatarnikov, d. (2015). gps satellite surveying (4th ed.). hoboken: johnwiley & sons. mazzetto, j. v. b. (2017). estimation of l2 observables of the gps system through the use of artificial neural networks. são carlos school of engineering university of são paulo (eesc-usp), são carlos. mcculloch, w. s., & pitts, w. (1943). a logical calculus of the ideas immanent in nervous activity. the bulletin of mathematical biophysics, 5(4), 115–133. https://doi.org/10.1007/bf02478259 mosavi, m. r. (2004). a wavelet based neural network for dgps corrections prediction. wseas transactions on systems, 3(10), 3070–3075. mosavi, m. r. (2007). gps receivers timing data processing using neural networks: optimal estimation and errors modeling. international journal of neural systems, 17(5), 383–393. https://doi.org/10.1142/ s0129065707001226 rosenblatt, f. (1958). the perceptron: a probabilistic model for information storage and organization in the brain. psychological review, 65(6), 386–408. https://doi.org/10.1037/h0042519 rumelhart, d. e., hinton, g. e., & williams, r. j. (1986). learning representations by back-propagating errors. nature, 323(6088), 533– 536. https://doi.org/10.1038/323533a0 sang, j., kubik, k., & zhang, l. (1997). prediction of dgps corrections with neural networks. in knowledge-based intelligent electronic systems. ieee. https://doi.org/10.1109/kes.1997.619409 scikit-learn. (2018). scikit-learn: machine learning in python. retrieved october 25, 2018, from http://scikit-learn.org/ segantine, p. c. l. (2005). gps: global positioning system (1st ed.). são carlos: eesc/usp. semeniuk, l., & noureldin, a. (2006). bridging gps outages using neural network estimates of ins position and velocity errors. measurement science and technology, 17(10), 2783–2798. https://doi. org/10.1088/0957-0233/17/10/033 silva, c. a. u. da. (2003). a method to estimate gps data observables using artificial neural networks. são carlos school of engineering university of são paulo (eesc-usp), são carlos. silva, i. n., spatti, d. h., & flauzino, r. a. (2010). redes neurais artificiais para engenharia e ciências aplicadas (2nd ed.). são paulo: artliber. tensorflow. (2018). tensorflow. retrieved may 16, 2018, from https://www. tensorflow.org/ widrow, b., & hoff, m. e. (1960). adaptive switching circuits. neurocomputing: foundations of research, 123–134. keywords: multiple linear regression; the xisha islands; water area; landforms; heterogeneity; fractal characteristics. palabras clave: regresión lineal múltiple; las islas xisha; área de agua; accidentes geográficos; heterogeneidad; características fractales. how to cite item wang, x. (2021). fractal characteristics of geomorphological heterogeneity in xisha islands waters under multiple linear regression. earth sciences research journal, 25(1), 41-48. doi: https://doi.org/10.15446/esrj. v25n1.94076 based on the obvious anisotropy of xisha islands waters topography, the uneven fractal characteristics of xisha islands waters topography are studied by using multivariate linear expression. using multiple linear regression analysis method to extract the projection of heterogeneous characteristic factors, the geometric heterogeneous characteristics of xisha islands waters are obtained. the fractal feature of landform is studied based on projection coverage method, and the non-isotropic body irregularity in xisha islands waters is measured. experimental results show that this method can effectively analyze the non-uniform fractal characteristics of the xisha islands waters. the fractal dimension of different types of landforms in xisha islands in the corresponding region is high mountain area > medium and low mountain area > basin area, which reflects the difference of surface roughness or complexity of different types of landforms. the calculated fractal dimension is helpful to reveal the heterogeneity of general geomorphological features in different development stages. abstract fractal characteristics of geomorphological heterogeneity in xisha islands waters under multiple linear regression características fractales de la heterogeneidad geomorfológica en las aguas de las islas xisha bajo regresión lineal múltiple issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.94076 con base en la anisotropía obvia de la topografía de las aguas de las islas xisha, se estudian las características fractales desiguales de la topografía de las aguas de las islas xisha mediante el uso de una expresión lineal multivariante. utilizando el método de análisis de regresión lineal múltiple para extraer la proyección de factores característicos heterogéneos, se obtienen las características geométricas heterogéneas de las aguas de las islas xisha. la característica fractal de la forma del terreno se estudia con base en el método de cobertura de proyección y se mide la irregularidad del cuerpo no isotrópico en las aguas de las islas xisha. los resultados experimentales muestran que este método puede analizar eficazmente las características fractales no uniformes de las aguas de las islas xisha. la dimensión fractal de diferentes tipos de accidentes geográficos en las islas xisha en la región correspondiente es: zona de alta montaña > zona de media y baja montaña > zona de la cuenca, que refleja la diferencia de rugosidad o complejidad de la superficie de diferentes tipos de accidentes geográficos. la dimensión fractal calculada es útil para revelar la heterogeneidad de las características geomorfológicas generales en diferentes etapas de desarrollo. resumen record manuscript received: 23/06/2020 accepted for publication: 21/12/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 41-48 xuli wang* department of management engineering, inner mongolia vocational college of chemical engineering, hohhot 010070, china *email of corresponding author: wangxuli1981@163.com g e o m o r ph o l o g y https://doi.org/10.15446/esrj.v25n1.94076 https://doi.org/10.15446/esrj.v25n1.94076 https://doi.org/10.15446/esrj.v25n1.94076 mailto:wangxuli1981@163.com 42 xuli wang introduction for human beings, the most common phenomena in nature, such as the varied landforms and the turbulence in the river, are unpredictable and lack rules. since pangu’s founding, human beings have been tirelessly exploring this magical and wonderful nature (bagley et al., 2017). in the process of understanding nature and transforming nature, human beings are uncovering their veil layer by layer to explore their true faces (cai et al., 2016). it should be said that physicists spend a lot of energy in analyzing the universe and basic particles. with the creation of newton’s classical mechanics, einstein’s theory of relativity, and the development of quantum mechanics, human beings have made brilliant achievements in natural science (liu & jiang, 2020; arangurenriano et al., 2018). natural numbers, triangles, rectangles, squares, circles and so on, as practical techniques or tools of mathematics, are gradually created in the development of human social practice, and form arithmetic, algebra, geometry, trigonometry and calculus. mathematics gradually matures from simple to complex (yuan et al., 2016). in mathematics, geometry is the first subject to rise from empirical knowledge to theoretical knowledge, and its typical representative is euclid’s “geometric original” (padisak et al., 2016). it is the first work in the history of mathematics to systematically discuss geometry by using axioms and methods. since then, for quite a long time, geometry, as the science of describing the form of things, has been regarded as the only real mathematics and the model of scientific rigor (gu et al., 2018; hu et al., 2018; kong et al., 2018; zhang, 2020). euclidean geometry is a subject about intuitionistic spatial shape relations. its studies ordered, smooth and feature length geometry (gao et al., 2016). no matter euclidean geometry, or cartesian geometry analytic geometry and differential geometry developed later, the lines, curves and surfaces studied by them all have a common assumption that these lines and surfaces are smooth, meet the differentiability, and are abstract. however, the geometry of nature is complex. the points, lines and surfaces of nature are not all in the geometry category of mathematicians (riordan et al., 2016). nature is more irregular, disordered and discontinuous. as long as people pay a little attention to a large number of non-linear irreversible phenomena in the surrounding environment, they will find that people know little about these phenomena and have no idea about many problems. take the phenomenon that we are used to in the sky as an example. when you look up at the sky, you can often see a group of clouds floating in it, which is poetic (ozguc, 2018). but if you look at the clouds with different telescopes, you will find that the shape of the clouds seems to have nothing to do with the magnification of the telescope. no matter how large the magnification, its shape is always the same (li et al., 2017). for non-linear science, classical mechanics, quantum mechanics and relativity are useless. there must be new theories to study these scientific problems that all human wisdom can’t solve. in the mid-s, mandelbrot, a researcher in the physics department of the corporate research center of the united states and a professor in the mathematics department of harvard university, published an article on “how long is the coastline of britain” in science magazine of the united states. in the paper, he stressed that the length of the uk coastline is uncertain and depends on the scale used in the survey (skorek, 2018; wu, 2018). the smaller the scale used to measure the length of coastline, the longer the corresponding length of coastline. therefore, the traditional length scale is not a good measure of coastline, and a new parameter or measure must be found to better characterize the coastline. in the in-depth study of this problem, he brought out the concept of fractal. how long is the british coastline? it is considered to be the first epoch-making paper on fractal (chen et al., 2016). the term “fractal” was first proposed by mandelbrot. its original meaning is “irregular, fractional and fragmented” object. at present, fractal is a frontier topic in nonlinear science. in different literatures, fractal is given different names, such as “hausdorff measure set”, “irregular set” and “set with fine structure”, etc. generally, fractal can be regarded as the state of large and small fragments gathering. due to its rapid development in many sciences, fractal has become a discipline to describe the regularity of many irregular things in nature (luo, 2016). based on the above background, this paper proposed the research method for heterogeneous fractal characteristics of landforms in the water of the xisha islandas based on multiple linear expression, and analyzes the heterogeneous fractal characteristics of landforms in the water of the xisha islands (gao et al. 2017). materials and methods multiple linear regression analysis of projection pursuit based on extracting the factors of heterogeneous characteristics the projection pursuit multiple linear regression analysis method based on the extraction of heterogeneous characteristics factors. firstly, according to the common characteristics of landforms in the water of the xisha islands, this method divides the affiliation of independent variables to the heterogeneous characteristics (lyu et al., 2018). then, the projection pursuit technology is used to classify the independent variables into the characteristic factors. finally, the heterogeneous characteristics factor is used as a new explanatory variable to complete the multiple linear regression analysis of landforms in the water of the xisha islands. the advantage of this method is to solve the multicollinearity problem in the case of small samples. at the same time, by extracting the factors of the independent variable, the independent variable screening is more consistent with the research significance of the problem itself, so as to distinguish the traditional independent variable screening method (liang & pei, 2016). the affiliation division of independent characteristics when modeling specific problems with multiple regression analysis, it often involves dozens or hundreds of independent variables because of the complexity and specialization of the problems. traditional methods, such as correlation coefficient analysis and stepwise regression, can achieve a good fitting effect if they are only selected from the perspective of data analysis, but the regression model is often lacking persuasion and research significance. aiming at the multi-level phenomenon that landforms in the water of the xisha islands show common characteristics when there are too many independent variables or they belong to a specific subject according to their professional meanings, the independent variables of landforms in the water of the xisha islands are classified as the heterogeneous characteristics factors according to their affiliation, which are the new explanatory variables of multiple regression analysis (schuur et al., 2016). projection pursuit the main feature of projection pursuit is to solve the problem of “dimension disaster” caused by the sparse high-dimensional spatial data of landforms in the water of the xisha islands. by projecting the high-dimensional data of landforms in the water of the xisha islands into the 1-3-dimensional subspace, it can maximize (or minimize) the projection index and find the projection that can reflect the original high-dimensional data structure or characteristics, so that the projection data is analyzed later (abuzied & alrefaee, 2017). the steps of dimensionality reduction for the high-dimensional data of landforms in the water of the xisha islands are as follows. step1: data normalization of landforms in the water of the xisha islands. because there are variables in the water of the xisha islands sample data, there are great differences in the dimensions and orders of magnitude between variables, which makes the variable indicators lack consistency measurement. in order to facilitate the analysis, and ensure the equivalence and order of landforms in the water of the xisha islands data, it is necessary to normalize the original data of landforms in the water of the xisha islands. the normalization method of landforms in the water of the xisha islands is as follows. as larger as better, the indicators are: x i j x i j x j x j x j ( , ) *( , ) ( ) ( ) ( ) min max min = − − (1) as smaller as better, the indicators are: 43fractal characteristics of geomorphological heterogeneity in xisha islands waters under multiple linear regression x i j x j x i j x j x j ( , ) ( ) *( , ) ( ) ( ) max max min = − − (2) where, xmax( j ) and xmin( j ) are maximum and minimum value of the jth indicator, x ( i, j ) is a sequence of normalized values of index eigenvalue. step 2: construct projection index function q (a). pp method is to aggregate the p dimensional data {x ( i, j ) ; j = 1, 2, ... p} of landforms in the water of the xisha islands into {a (1), a (2), a (3),..., a (p)} as the one-dimensional projection value z ( i ) of the projection direction: z i a j x i j i n j n ( ) ( ) ( , )( , ,..., )= = = ∑ 1 2 1 (3) step 3: construct projection index function. in order to find the heterogeneity characteristics of landforms in the water of the xisha islands data in the high-dimensional data of landforms in the water of the xisha islands, when synthesizing the projection index values, it is required that the distribution characteristics of the projection values should meet the requirements that the local projection points should be as dense as possible, and it is better to agglomerate into several point clusters, while the whole projection point clusters should be as scattered as possible (wilson et al., 2016). therefore, the projection index function can be expressed as: q a s dz z( )= (4) where, sz is the standard deviation of the projection value; dz is the local density of the projection value z(i). s z e z nz i i n = −( ) −( )           = ∑ ( ) / . 2 1 1 0 5 (5) d r r i j u r r i jz i n j n = −( ) −( ) = = ∑ ∑ 1 1 ( , ) ( , ) (6) where, e ( z ) is the average value of the sequence z ( i ); r is the window radius of the local density, which can be taken as 0.1; r i j z z ui j( , )= − ( ) represents a unit order function, if r − r ( i , j ) ≥ 0 its value is 1, then if r − r ( i , j ) < 0 its value is 0. step 4: optimize the projection index function to determine the best projection direction. different projection directions reflect different characteristic structures of landforms in the water of the xisha islands. the best projection direction is to reflect the projection directions of some kind of characteristic structures of landforms in the water of the xisha islands as much as possible. therefore, the best projection direction can be estimated by solving the problem of maximizing projection index function, namely: maxq a s dz z( )= (7) s t a j j p . . 2 1 1( )= − ∑ (8) this is a complex nonlinear optimization problem with optimization variables {a ( j ) ; j =1,2,..., p}, which is difficult to deal with by traditional optimization methods. therefore, in this paper, real coded accelerated genetic algorithm (raga) is applied to solve the problem of high-dimensional global optimization of its heterogeneous characteristics of landforms in the water of the xisha islands. raga the standard genetic algorithm (sga) code is binary, its process is tedious, and the accuracy is limited by the length of string, the calculation is large, and premature convergence may occur, which is not convenient to develop genetic operators for special problems. to solve this problem, this paper uses an improved accelerated genetic algorithm based on real coding, which greatly enhances the optimization performance of the algorithm (otoperalias, 2018). the steps of raga modeling are as follows: (1) the following linear transformation is used to optimize the real number coding of landforms variables in the water of the xisha islands: x j a j y j a j j p( )= ( )+ ( ) ( )( ) =( )1 2, ,..., (9) where f is the objective function of the optimized landforms in the water of the xisha islands, p is the number of optimized variables. equation (9) corresponds the j-th variables to be optimized x ( j ) aaa and a ( j ) of the landforms in the water of the xisha islandson on the initial variable interval [0, 1] to the real number y ( j ) on the interval [0, 1]. y ( j ) is the genetic gene of raga. (2) the initialization of the parents of landforms in the water of the xisha islands. suppose that the size of the parents of landforms in the water of the xisha islands is n, then n groups uniform random number on the interval of [0,1] are generated. there are p in each group, namely u i j j p i n, ; , ,..., ; , ,...,( ) = ={ }1 2 1 2 . the individual y ( i , j ) is substituted into equation (9) to obtain the optimal variable value x ( i , j ), and then the corresponding objective function value f i( ) is obtained through equation (7). according to the order from small to large, f i i n( ) ={ }; , ,...,1 2 corresponding individuals y ( j, i) are sorted. the smaller the objective function value of landforms in the water of the xisha islands is, the stronger the individual’s adaptability is. after sorting, the first individual is the excellent individual, so that it can enter the next generation directly. (3) the fitness evaluation of the parents in the water of the xisha islands. the evaluation function based on the order is to distribute the regeneration according to the order of chromosome, which is not suitable for the actual target value. set parameter α ∈( )0 1, , and the evaluation function based on order is: eva y j i a a i ni( , ) , ,...,( )= −( ) =( )−1 1 21 (10) here i = 1 means the chromosome is the best, i = n means the worst. (4) the first generation y ( j , i ) of landforms in the water of the xisha islands was generated by selection. the selection process is based on rotating the wheel n times. each rotation selects a chromosome for the new population. the selection process is described as follows: the cumulative probability qi ( i = 1,2,..., n ) of each chromosome y ( j , i ) is: q q eval y j ii j p i n j i 0 12 12 1 0= = ( )( )         = =( ) = ∑ , , ,..., ; , ,...,      (11) a random number r is generated from the interval [0,qi]; if qi-1 < r ≤ qi y ( j, i ), the i-th chromosome is selected. repeat step 2 and step 3 for n times, so that the n duplicated chromosomes can be obtained and a new generation of individuals can be formed. (5) the parent population of landforms in the water of the xisha islands was hybridized. to determine the parent of the crossover operation, repeat the following procedure from i = 1 to i = n: the random number r is generated from [0,1], if r < pc , then y ( j, i) is a parent. y'1( j, i) andy'2( j, i) were used to represent the selected parents, and they were randomly matched. when the parents are odd, one chromosome can be removed or another chromosome can be selected to ensure pairing (ginde et al., 2016). the arithmetic crossover method is adopted, i.e. firstly, a random number c is generated from (0, 1), then 44 xuli wang the crossover operation is carried out between y'1( j, i) and y'2( j, i) according to the following form, and two offspring x and y are generated: x cy j i c y j i y c y j i cy j i = ( )+ −( ) ( ) = −( ) ( )+ ( )       ' , ' , ' , ' , 1 2 1 2 1 1  (12) the second generation {y2( j, i) ; j =1,2,..., p; i=1,2,..., n} of landforms in the water of the xisha islands was obtained by the above hybridization. (6) mutation operation. the process of parent selection of mutation is similar to that of cross operation. from i = 1 to n, the following process is repeated: the random number r is generated from the interval [1, 0]. if r < pm is selected, then chromosome y ( j, i) is selected as the variant parent, and y ( j, i) is selected as the variant parent to correspond to each selected parent, which is represented by y'3( j, i). perform the variation as follows. in r n, the direction of variation d was selected randomly, then y'3( j, i) + md. a new generation {y3( j, i) ; j =1,2,..., p; i=1,2,..., n} of landforms in the water of the xisha islands was obtained by mutation. (7) evolutionary iteration. from the previous step, we will get 3n progeny individuals, and sort them according to their fitness function from large to small. the first (n-k) progeny is selected as the parent population of the new landforms in the water of the xisha islands. the algorithm turns to step (3) to carry out the next round of evolution, reevaluation, selection, hybridization and mutation. this is repeated until the objective function value of the optimal individual yn ( j, i) is less than a certain set value or the algorithm reaches a predetermined number of times of acceleration, and the algorithm ends. at this time, the optimal individual in the current population is regarded as the optimization result of raga, which are the heterogeneous characteristics of landforms in the water of the xisha islands (zhuo, 2016). research method for fractal characteristics of landforms based on projection covering method basic principle of projection covering method the projection covering method is to directly quantify the heterogeneous characteristics of landforms in the water of the xisha islands by establishing the corresponding relationship between the surface of the water of the xisha islands and the projection surface, projecting the landforms surface onto the two-dimensional plane. in this method, the complex landforms in the water of the xisha islands are projected vertically onto the projection plane of square grid with side length r (juban et al., 2016). let a, b, c and d be the four corners of any square grid on the projection plane of the water of the xisha islands, corresponding to the points on the surface at different heights. if the surface area enclosed by these four points is approximately obtained, the total area of the whole surface of landforms in the water of the xisha islands is approximately: s r s r y j ik k n r n( )= ( ) ( ) = ( ) ∑ 1 , (13) where s(r) is the total surface area corresponding to the square grid side length r of the projections of the lands in the water of the xisha islands; sk(r) is the surface area corresponding to the k-th square grid in the projection surface of the xisha islands; n (r) is the total number of meshes projected from the ground surface of the water of the xisha islands to the projection surface of the square grid; r is the mesh side length of the projection face of the square mesh; k is the mesh sequence number of the projection face. when the side length r of the square grid of the planes in the water of the xisha islands is changed, the total surface area s(r) is also changed. especially when r → 0, the real surface area of landforms in the water of the xisha islands is close to s(r). if the surface has fractal characteristics, according to the fractal theory, the relationship between s(r) and r is: d ins r inrs r = − → 2 0 lim ( ) (14) where ds is the fractal dimension of the landforms surfaces in the water of the xisha islands. irregularity analysis of heterogeneous body of landforms in the water of the xisha islands based on the landforms fractal dimension the basic principle of the fractal dimension of lands surface in the water of the xisha islands is based on the projection covering method. using the spatial analysis function provided by arcgis software, the dem (tin format) of in the water of the xisha islands obtained from each measurement is transformed into grid data format (grid format) of different sizes. grid sizes are 3, 6, 9, 12, 15, 93, 96, 99mm, 33 species in total (iacobucci et al., 2016). then the surface area s(r) of the different grid sizes corresponding to the landforms in the water of the xisha islands is calculated. finally, take in(r) as the abscissa and in[s(r)] as the ordinate, and collect the calculation results of different scales r into the double logarithmic coordinate system. the least square method is used to fit the straight line and determine the scale-free interval. the fractal dimension of the surface of landforms in the water of the xisha islands can be determined by the slope  of the fitting straight line, namely ds = 2 −. based on this, we can measure the irregularity of landforms in the water of the xisha islands. the implementation process is shown in figure 1: results shidao is a rock island in the xisha islands, located in the northeast of yongxing island, the largest island in the xisha islands, with a distance of more than 1 km. with an altitude of 15 meters, it is the top of xisha islands. the rock on the island stands up, and the features of landforms on the eroded coast are very obvious, which is in sharp contrast with other grey sand islands surrounded by the crystal white jade belt beach. the geologists who set foot in xisha have paid attention to its own unique characteristics, all of which focus on the geological history of the xisha islands since the late period. in this paper, the objective of the study is to study the terrestrial characteristics of landforms in the shidao waters of the xisha islands. dynamic change of fractal dimension of landforms table 1 shows the surface fractal dimensions of the landforms in the shidao waters of the xisha islands at different development stages calculated by this method: table 1. fractal dimension calculation results of xisha islands water area based on projection coverage method measuring order the correlation coefficient (r2) geomorphic fractal dimension 1 0.9915 2.0142 2 0.9923 2.0254 3 0.9902 2.0326 4 0.9922 2.0534 5 0.9908 2.0575 6 0.9908 2.0654 7 0.9907 2.0664 8 0.9914 2.0818 9 0.9908 2.0745 note: 9 significance tests were all extremely significant. it can be seen from the table that in the scale-free range, there is a good linear relationship between in[s(r)] and in(r), the correlation index r2 is above 0.99, and the significance test reaches the extremely significant level 45fractal characteristics of geomorphological heterogeneity in xisha islands waters under multiple linear regression (=0.01). this shows that the landforms of the shisha islands waters have obvious fractal characteristics in this region. with the change of measurement sequence, the surface fractal dimension of the landforms in the shidao water area of the xisha islands increased from 2.0142 in the first measurement to 2.0818 in the eighth measurement, and the surface fractal dimension increased greatly from the first measurement to the fourth measurement, and then decreased. at the 9th measurement, the fractal dimension of the landforms surface in the shidao waters of the xisha islands decreased to 2.0745, which was lower than that of the 8th measurement. this reflects that the dynamic evolution process of the simulation is an accurate expression of the dynamic changes of the overall characteristics of the landforms in the shidao waters of the xisha islands. comparisons of fractal dimensions of landforms in the shidao water area of the xisha islands by different methods in order to further analyze the advantages of this method for the shidao water area of the xisha islands, the fractal dimension of the landforms surface in the shidao water area of the xisha islands calculated by the method in this paper is compared with the results obtained by the more commonly used calculation methods (table 2). among them, take the increment of landforms fractal dimension as the comparison object. that is to say, the fractal dimensions of the landforms in the shidao water area of the xisha islands calculated by the above two methods are obtained by subtracting the corresponding geometric topological dimensions of the fractal objects. the corresponding fractal objects of this method and box counting dimension method are landforms surface and terrain contour, and the corresponding values of geometric topological dimension are 2 and 1. it can be seen from table 2 that the maximum value of the fractal dimension increment of landforms calculated by the method in the shidao water area of the xisha islands is 0.0818 and the minimum value is 0.0142. during the first to the eighth measurements, the average increment of fractal dimension of landforms in the shidao water area of the xisha islands increased significantly. the maximum increment of fractal dimension of small watershed landforms calculated by box counting dimension method is 0.4254 and the minimum is 0.1521. it can be seen that the fractal feature quantification results of landforms in the shidao water area of the xisha islands are more conducive to reveal the heterogeneity of the overall characteristics of landforms in different development stages. fractal dimension values of different regions for a large number of statistical fractals existing in nature, they are not infinite generation like strict mathematical fractals, but can be established in a certain scale range. a certain form is closely related to a certain mechanism of action, and it is precisely because of the fractal characteristics of the multidimensional domain in the forms of landforms that provides an excellent opportunity for us to study the mode, intensity and scale range of the internal and external forces. divide the shidao water area of the xisha islands into six areas, and calculate the fractal dimension values of different areas by the method in this paper. see table 3 for the calculation results of the fractal dimension values of different areas: figure 2 is the fractal structure of the multi-dimensional domain in the shidao water area of the xisha islands. figure 1. realization process of geomorphic fractal feature research method based on projection coverage method table 2. fractal dimension increment comparison of the landforms in shidao waters of xisha islands by different calculation methods measuring order ds dc 1 0.0142 0.1521 2 0.0254 0.2138 3 0.0326 0.2779 4 0.0534 0.3293 5 0.0575 0.3755 6 0.0654 0.3998 7 0.0664 0.4217 8 0.0818 0.4254 9 0.0745 0.3516 table 3. calculation results of fractal dimension values in different regions scale range i ii iii high mountain watersa 2.0863±0.0017 2.03973±0.0009 2.00749±0.0007 high mountain watersb 2.05295±0.0012 2.02402±0.00077 2.00545±0.000087 medium and low mountain watersc 2.03543 2.01972±0.0007 2.00263±0.00033 medium and low mountain watersd 2.00961 2.00415±0.0008 2.00089±0.00002 basin watere 2.00298 2.00110±0.000079 2.00034±0.000012 basin waterf 2.0012 2.00005±0.0000007 2.00002±0.000004 46 xuli wang figure 2. multi-dimension fractal structure of geomorphologic research area from the above results, we can see that: in the shidao water area of the xisha islands, each research area can be roughly divided into three obviously different linear intervals (scale from small to large is called i, ii, iii respectively), that is, multi-scale fractal. it shows the local self-similar or selfaffine replication relationship in the shidao water area of the xisha islands. with the increase of the observation scale, the fractal dimension value of each interval decreases in turn. this shows that with the increase of the observation scale, the roughness of the shape in the shidao water area of the xisha islands is decreasing, and the fluctuation of the smaller scale is ignored. the process of controlling the action of landforms is from the dominant role of external forces, the struggle between internal and external forces, to the mechanism that is basically replaced by internal forces. in the shidao water area of the xisha islands, the fractal dimension values of different types of landforms in the corresponding interval are as follows: water area in high mountain area > water area in middle and low mountain area > water area in basin area, which reflects the difference of surface roughness or complexity of different types of landforms. therefore, the fractal dimension value calculated by this method can be used as an important index of the erosion intensity of the surface external forces in the shidao water area of the xisha islands. in the shidao water area of the xisha islands, the distribution of multifractal spectral values of different landforms is shown in figure 3. it can be seen from figure 3 that in the shidao water area of the xisha islands, the overall characteristics of the water area in the high mountain area > the water area in the middle and low mountain area > the water area in the basin area, thus reflecting the heterogeneity degree of the probability distribution of the surface area of different types of landforms in the shidao water area of the xisha islands. in the shidao water area of the xisha islands, different study areas of the same landforms also show different degree of uneven distribution of undulating surface area. among them, the distribution of heterogeneity is significant in the c study area and the f study area of basin area in the middle and low mountain area, on the contrary, their corresponding d and e areas tend to be uniform and single fractal characteristics. discussion in the shidao water area of the xisha islands, the study of landforms can be said to be a new research field. we believe that the problems that need further improvement and in-depth study at present are mainly from the following aspects: (a) high mountain waters (b) medium and low mountain waters (c) basin water figure 3. multifractal spectral range distribution of different geomorphic types 47fractal characteristics of geomorphological heterogeneity in xisha islands waters under multiple linear regression (1) the internal mechanism of the characteristics of the fractal segments of landforms many initial researches only focus on the discovery and description of the fractal phenomena in the landforms phenomenon. however, the internal formation mechanism or external dynamic conditions of the fractal properties of these landforms, as well as whether the fractal characteristics can reflect the nature of a particular landforms process, and how the regional characteristics of the fractal characteristics of landforms are, remain to be further explored. at the same time, the existing research shows that many dynamic factors of landforms, such as geography, geology, geophysical field (gravity field, etc.) and even geological structure (fault), can have fractal structure. on this basis, the landforms can also be fractal, so what is the mathematical relationship between these two kinds of fractal? (2) the relationship between fractal dimension of landforms and landforms process many researches on the fractal characteristics of landforms reveal that the fractal law is only a static characteristic (i.e. the “instantaneous” state of landforms nowadays), while others try to associate the fractal characteristics with the development of landforms, but it is still difficult to explain the relationship between the fractal characteristics and the landforms process. therefore, the following problems need to be studied: 1) is the fractal dimension of landforms constant or changing? 2) how long can the fractal dimension of modern landforms represent the geo history stage? 3) the study shows that the climate and lithology conditions are basically the same, and the development of landforms reaches the basin in the middle age or old age (the area is large enough), and the fractal dimension value of the river is close to 0. does this mean that a stable (balanced) landforms type fractal value tends to a definite value (constant)? 4) if the fractal dimension of landforms in a certain stage is a constant, how long does it take from the beginning of its development to reaching this constant? 5) can fractal dimension reflect the stages of the development of landforms? that is, do different landforms have their own typical fractal dimension? what are the reasons? 6) although fractal dimension can be used as the index of watershed erosion period division, what is the relationship between the division of water system development stage marked by the formation of fractal water network and the traditional landforms stage division based on landforms profile and other morphological factors? 7) does the geological structure, lithology, climate and vegetation affect the fractal dimension of this stage? (3) the relationship between the gradient free interval and the scale range of the main landforms processes the fractal features of landforms are usually only statistical features in a certain scale range, and are this scale range related to the scale range of the main landforms processes? at the same time, the same landforms process or action is different on different scales, and the fractal feature is a kind of cross-scale geometric feature in a certain scale range (i.e. scale-free range). what does this fractal feature mean? what’s more, does a certain kind of landforms follow the statistical fractal feature, which means that the mechanism of its leading landforms forces is scale-free or cross scale? the answer to this question is very important in the reliability judgment of the results of the indoor micro landforms experimental model and the determination of the extension or representative range (including common hydraulic engineering landforms simulation such as gezhouba project, etc.). (4) deepening the research of landforms mode there are many successful mathematical rules to simulate the generation of fractal terrain, but mathematicians also admit that this terrain does not reflect any information of the formation process of landforms. other studies have found that the natural landforms landscape can be simulated by using a simple near linear rule. but what is the relationship between this simple rule and the multifactor complex process in the development of natural landforms? does it reflect some characteristics of the complex landforms process, or just conform to some rules of people’s visual judgment? how can we effectively use this mathematical rule to establish the landforms development and evolution model with the participation of internal and external dynamics? what is the difference between simulated terrain and natural terrain? obviously, the interpretation of fractal landforms model is also an important aspect. (5) the feasibility study on the application of time and space fractal law in the prediction of landforms process landforms events, such as earthquakes, floods, mudslides, etc., can be time fractal, which means that they have certain predictive significance. for example, if the distribution of flood water is time fractal, it means that the ratio of the number of floods with 100 year return period to the number of floods with 10 year return period can represent the ratio of the flood with 1000 year return period to the flood with 100 year return period. conclusions gouide, a famous landforms scientist, pointed out in his book the technology of landforms research that once a new technology can be used, the whole scientific process can develop around it and enter a higher level of research and interpretation. as a new technology, fractal theory is also applied to landforms. the application of fractal theory in the study of landforms can be very extensive, including the study of landforms of various landforms or watershed as a whole and the study of landforms process. fractal theory provides a new research method for landformers, which enables them to discover the new cross-scale quantitative characteristics of the traditional complex and disordered landforms phenomenon. at the same time, fractal dimension has become a new parameter in landforms. the application of fractal theory also makes it possible for the research of landforms to connect the simple formula obtained from field observation with the complex landforms and establish the model. therefore, fractal theory provides a possible way for landforms scholars to explain complex terrain. therefore, this paper puts forward the research method for heterogeneous fractal characteristics of xisha islands water lands based on multiple linear expression, and studies the forms of heterogeneous characteristics in the shidao water area of the xisha islands. there is a good linear relationship between in[s(r)] and in(r), the correlation index r2 is above 0.99, and the significance test reaches the extremely significant level (= 0.01). this shows that the landforms of the shisha islands waters have obvious fractal characteristics in this region. the shidao water area of the xisha islands is divided into six regions, and the fractal dimension values of different regions are calculated by the method in this paper. the overall characteristics are as follows: mountain water area > middle and low mountain water area > basin water area. this shows the degree of heterogeneity of the probability distribution of rough surface area in different types of landforms. references abuzied, s. m., & alrefaee, h. a. (2017). mapping of groundwater prospective zones integrating remote sensing, geographic information systems and geophysical techniques in el-qaà plain area, egypt. hydrogeology journal, 25(7), 1-22. aranguren-riano, n. j., guisande, c., shurin, j. b., jones, n. t., barreiro, a., & duque, s. r. (2018). amino acid composition reveals functional diversity of zooplankton in tropical lakes related to geography. taxonomy and productivity. oecologia, 187(46), 1-12. bagley, r. k., sousa, v. c., niemiller, m. l., & linnen, c. r. (2017). history, geography and host use shape genomewide patterns of genetic variation in the redheaded pine sawfly (neodiprion lecontei). molecular ecology, 26(4), 1022-1044. cai, a. h., he, j. w., & wang, p. (2016). experiment and error analysis of path planning for sonobuoy array deployment. journal of china academy of electronics and information technology, 11(1), 21-24. chen, b. y., yuan, h., li, q., shaw, s. l., lam, w. h. k., & chen, x. (2016). spatiotemporal data model for network time geographic analysis in the era of big data. international journal of geographical information science, 30(6), 1041-1071. 48 xuli wang gao, j. s., you, x. m., & zheng, j. (2016). estimation of soc by ann and ah method. chinese journal of power sources, 40(9), 1842-1844. gao, s., janowicz, k., montello, d. r., & hu, y. (2017). a data-synthesisdriven method for detecting and extracting vague cognitive regions. international journal of geographical information science, 31(6), 1-27. ginde, g., saha, s., mathur, a., venkatagiri, s., vadakkepat, s., narasimhamurthy, a., & daya sagar, b. s. (2016). scientobase: a framework and model for computing scholastic indicators of nonlocal influence of journals via native data acquisition algorithms. scientometrics, 108(3), 1-51. gu, c. y., zhao, p., wang, l., & guo, h. (2018). technical application of petroleum logging instruments in marine logging. polish maritime research, 25(3), 22-28. hu, y., xu, m., & li, y. (2018). simulation model and state analysis of ship transmission line. polish maritime research, 25(3), 36-42. iacobucci, d., schneider, m. j., popovich, d., & bakamitsos, g. a. (2016). mean centering, multicollinearity, and moderators in multiple regression: the reconciliation redux. behavior research methods, 49(1), 1-2. juban, r., ohlsson, h., maasoumy, m., poirier, l., & kolter, z. (2016). a multiple quantile regression approach to the wind, solar, and price tracks of gefcom2014. international journal of forecasting, 32(3), 1094-1102. kong, d. s., deng, m. x., liu, y., & tan, x. (2018). study of the force and deformation characteristics of subsea mudmat-pile hybrid foundations. polish maritime research, 25(3), 43-53. li, y. t., yang, x. l., & li, s. b. (2017). analysis of disaster impact factors based on spatial autocorrelation. automation & instrumentation, 3, 4-6. liang, y. m., & pei, x. h. (2016). artificial fish-swarm algorithm optimized by particle swarm algorithm. computer simulation, 33(6), 213-217. liu, b., & jiang, x. (2020). consolidation and deformation characteristics of soft rock foundation in a hydrological wetland environment. earth sciences research journal, 24, 183-190. luo, c. h. (2016). active contour model based on gradient vector flow. journal of jilin university (science edition), 05, 1103-1108. lyu, n., zhao, j., & lyu, m. (2018). exploration of land development intensity index of port container logistics park based on quantitative algorithm and pent analysis method. polish maritime research, 25(3), 61-67. oto-peralias, d. (2018). what do street names tell us? the ‘city-text’ as sociocultural data. journal of economic geography, 18(1), 187-211. ozguc, o. (2018). estimation of buckling response of the deck panel in axial compression. polish maritime research, 25(4), 98-105. padisak, j., vasas, g., & borics, g. (2016). phycogeography of freshwater phytoplankton: traditional knowledge and new molecular tools. hydrobiologia, 764(1), 3-27. riordan, e. c., gugger, p. f., ortego, j., smith, c., gaddis, k., thompson, p., & sork, c. l. (2016). association of genetic and phenotypic variability with geography and climate in three southern california oaks. american journal of botany, 103(1), 73-85. schuur, j. d., baker, o., freshman, j., wilson, m., & cutler, d. m. (2016). where do freestanding emergency departments choose to locate? a national inventory and geographic analysis in three states. annals of emergency medicine, 69(4), 383-392. skorek, g. (2018). study of losses and energy efficiency of hydrostatic drives with hydraulic cylinder. polish maritime research, 25(4), 114-128. wilson, c. a., matthews, k., pulsipher, a., & wang, w. h. (2016). using geographic information systems to determine site suitability for a lowlevel radioactive waste storage facility. health physics, 110(2 suppl 1), s17-s25. wu, d. (2018). impact of green total factor productivity in marine economy based on entropy method. polish maritime research, 25(3), 141-146. yuan, h., ding, x. p., wang, b. r., mou, w., & yang, c. (2016). research on effect on dc-dc converter caused by parasitic resistance. journal of power supply, 14(1), 86-94. zhang, q. (2020). study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area. earth sciences research journal, 24, 35-44. zhuo, w. (2016). a large-area image sensing and detection system based on embedded thin-film classifiers. ieee journal of solid-state circuits, 51(1), 281-290. keywords: landform types; loess landform; karst; urban and rural areas; spatial distribution; characteristics. palabras clave: tipos de relieve; forma de relieve loess; karst; zonas urbanas y rurales; distribución espacial; características. how to cite item li, j., cai, z., & duan, l. (2020). spatial distribution characteristics and optimum layout of urban and rural areas in different geomorphological types. earth sciences research journal, 24(3), 267-275. doi: https:// doi.org/10.15446/esrj.v24n3.89450 taking jinghe river basin in the loess geomorphological area and guangnan county in the karst geomorphological area as the study area, the spatial distribution characteristics of urban and rural areas of different geomorphological types are analyzed. by using gis and related statistical analysis software, this paper summarizes three basic urban and rural types: river channel type, plateau surface type, and loess terrace horizon prototype in the loess landscape jinghe river basin. it is known that most towns in the loess plateau gully area are in the jinghe river basin. according to the spatial distribution characteristics of urban and rural areas, the optimal layout based on the main structure of five districts, nine river corridors, and four plates is proposed. using the dem module of arcgis to divide the elevation and gradient of guangnan county, we know that the density of urban and rural settlements in guangnan county is low and the spatial distribution is dispersed, and the distribution of urban and rural settlements shows a strong elevation orientation. the distribution of urban and rural settlements has a normal distribution relationship with the elevation. the largest number of urban and rural settlements is between 2.1° and 25°. according to the present situation of settlement distribution, this paper puts forward some optimization strategies, such as appropriate settlement scale, settlement space development monitoring, and so on. abstract spatial distribution characteristics and optimum layout of urban and rural areas in different geomorphological types características de distribución espacial y disposición óptima de áreas urbanas y rurales en diferentes tipos geomorfológicos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n3.89450 este estudio analiza las características de distribución espacial de áreas urbanas y rurales de diferentes tipos geomorfológicos al tomar como área de estudio la cuenca del río jinghe, en el área geomorfológica de loess, y el condado de guangnan, en el área geomorfológica kárstica. mediante el uso de sistemas de información geográfica y aplicaciones de análisis estadístico relacionado, este documento resume tres tipos básicos urbanos y rurales: tipo de canal de río, tipo de superficie de meseta y prototipo de horizonte de terraza de loess en la cuenca del río jesshe. en esta cuenca se encuentran asentamientos humanos principalmente ubicados sobre depósitos de loess. de acuerdo con las características de distribución espacial de las áreas urbanas y rurales, este estudio propone un diseño óptimo basado en la estructura principal de cinco distritos, nueve corredores fluviales y cuatro placas. al usar el módulo dem de arcgis para dividir la elevación y el gradiente del condado de guangnan se estableció que la densidad de los asentamientos urbanos y rurales en este condado es baja, la distribución espacial es dispersa y la distribución de los asentamientos urbanos y rurales muestra una fuerte orientación de elevación. la distribución de los asentamientos urbanos y rurales tiene una relación de distribución normal con la elevación. el mayor número de asentamientos urbanos y rurales está entre 2.1° y 25°. de acuerdo con la situación actual de la distribución de asentamientos, este documento presenta algunas estrategias de optimización, como la escala de asentamiento apropiada, el monitoreo del desarrollo del espacio de asentamiento, entre otros. resumen record manuscript received: 12/03/2019 accepted for publication: 11/03/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 3 (september, 2020): 267-275 jing li*1,2,3, zhongyuan cai4, lianru duan2 1northwest institute of historical environment and socio-economic development, shaanxi normal university, xi'an 710048, china 2school of urban planning and municipal engineering, xi’an polytechnic university, xi’an 710048, china 3state key laboratory of green building in western china, xi'an university of architecture and technology, xi’an 710048, china 4school of architecture, xi'an university of architecture and technology, xi'an 710055, china *corresponding author: lijing982426@163.com g e o m o r ph o l o g y https://doi.org/10.15446/esrj.v24n3.89450 https://doi.org/10.15446/esrj.v24n3.89450 https://doi.org/10.15446/esrj.v24n3.89450 mailto:lijing982426@163.com 268 jing li, zhongyuan cai, lianru duan introduction china has a vast territory, and there are great differences in the historical basis, natural environment, and economic development of different regions. therefore, the trend of urban-rural interaction development in different geomorphic types of regions is also different (zhang & fan, 2017). geomorphology is the undulating form of the lithospheric surface, and it is also an integral part of the natural geographical elements. in the evolution of urban and rural spatial form, on the one hand, geomorphology directly or indirectly acts on the development and expansion of urban and rural space, causing different responses and changes (aumaître et al., 2017). on the other hand, geomorphology is the main place for energy exchange and transformation between urban and rural space and other elements and becomes the basis for the development of urban and rural space. moreover, geomorphology itself has undergone profound changes under the influence of other elements, including urban and rural spatial expansion (samadzadegan et al., 2017). in a word, landform and urban-rural space interact and promote each other, making the whole living environment change constantly. correctly measuring the spatial distribution characteristics and optimizing the layout of urban-rural interactive development level is the premise of taking effective measures to coordinate urban-rural relations and narrow the gap between urban and rural areas. it is also the basis of choosing a reasonable model of urban-rural development and formulating optimal countermeasures for urban-rural interactive development in various regions. many scholars at home and abroad have carried out relevant studies on the spatial distribution characteristics of urban and rural areas in different geomorphic regions. under the background of rapid urbanization, american scholars pak and brecko analyzed the impact of land use structure change, population size, and industrial structure adjustment on the development of urban and rural settlements in suburbs. scholar liu xiantao studied the influence of topography, roads, and rivers on the spatial distribution characteristics of urban and rural settlements by using the method of cv value of the voronoi map. it was shown that topography was the dominant factor affecting the spatial distribution of urban and rural settlements, while road traffic was the second factor affecting the spatial distribution of urban and rural settlements. zhang xia, a scholar, used spatial analysis and landscape index analysis to analyze the spatial distribution pattern of urban and rural settlements in hilly areas of chongqing and its influencing factors. it was concluded that there were many urban and rural settlements in hilly areas, but the scale of urban and rural settlements was generally small. the concentration of spatial distribution was low, and with the increase of terrain gradient, the area and average scale of urban and rural settlement land showed a relatively decreasing trend. therefore, topography was the most direct factor affecting the distribution pattern of urban and rural settlements, which restricted the land use, scale, shape, and distribution characteristics of urban and rural settlements (li et al., 2017). studying the spatial distribution characteristics and optimization of urban and rural areas will help to understand the spatial distribution law of urban and rural interactive development in an all-round way and provide a decisionmaking basis for promoting urban and rural interactive development in different regions. based on the relationship between landform types and urban-rural development, this paper focuses on the spatial distribution characteristics of urban and rural areas in loess and karst landform regions, compares the spatial distribution characteristics of urban and rural areas under different landform types, and puts forward corresponding optimization measures. taking jinghe river basin as a typical area of loess landform type, this paper uses gis and related statistical analysis software to superimpose and analyze the terrain and landform, urban and rural patches of jinghe river basin, summarizes three basic urban and rural types: river channel type, plateau type, and loess terrace prototype, and quantitatively studies the spatial distribution pattern of urban and rural areas in jinghe river basin, according to the suitable index of human settlement environment and life span of jinghe river basin. from the perspective of spatial distribution characteristics of eco-spatial environment, social and economic development status, conforming to the future needs of coordinated development of environment and economy, and establishing eco-friendly “low-carbon” urban-rural system spatial structure, this paper elaborates the optimization of urban-rural spatial distribution characteristics in loess landform type regions. taking guangnan county of yunnan province as a typical karst landform, this paper uses dem module of arcgis software to analyze the basic characteristics of the spatial distribution of urban and rural settlements in guangnan county of yunnan province and explores the influence of natural factors on the spatial distribution of urban and rural settlements in karst mountain areas, so as to provide a basis for optimizing the pattern of urban and rural settlements in karst landform type regions. material and methods overview of loess geomorphology research area located in the north and west of china, the loess plateau covers a total area of 642,000 km2, of which 454,000 km2 is the most serious area of soil erosion and the most fragile ecological environment in china and even in the world. 80% of villages and towns in the loess plateau are located in small watersheds or related to small watersheds, and nearly 60% of the population live in them. the contradiction between the fragile ecological environment and the rapid development of urbanization in the loess plateau area, the contradiction between the comprehensive management of soil and water conservation ecological environment and the lagging concept of architecture discipline, gave birth to the concept of ecological living environment (lu et al., 2017). jinghe river is the largest tributary of weihe river, the first tributary of the yellow river. it originated at the eastern foot of liupan mountain in ningxia. it has successively formed radial water systems, such as namalian river, puhe river, and heihe river, and injected the weihe river near chenjiatan, gaoling county of shaanxi province. the total area of the basin is 46538 square kilometers. due to the cutting of the jinghe river and its tributaries, the jinghe river basin has formed four landforms: loess plateau-gully area, loess hilly-gully area, loess rocky mountain area, and loess terrace plain area. among them, the most representative is the loess plateau-gully landscape in the central part of the jinghe river basin and the loess hilly-gully landscape in the northern part of the jinghe river basin (yasuhara et al., 2017). therefore, the spatial distribution of urban and rural areas in the jinghe river basin on the loess plateau has unique regional characteristics. jinghe river basin is one of the most densely populated areas in the loess plateau, and the whole region contains four basic types of landforms in the loess plateau. it is also the main area for the construction of human settlements and the prevention of ecological damage and environmental pollution in the loess plateau (nematollahi et al., 2018). landform characteristics of loess the surface of the jinghe river basin decreases gradually from northwest to southeast. the ground elevation is mainly between 1000 and 1600 meters above sea level. the vast area of the jinghe river basin is the loess geomorphic area. the loess layer is deep, forming a typical loess layer profile (mori, 2018). because of the cutting of the jinghe river and its tributaries, the landform landscape of the loess with the source-gully alternation is shown in figure 1. research method of spatial distribution in this paper, based on the human settlements environment assessment index of jinghe river basin, combined with the overall goal of the development of jinghe river basin, the knowledge of ecology, geography, human settlements environment, and other related disciplines are applied, the elements of future traffic planning, industrial layout, population development prediction, etc. are added to determine the idea and overall direction of regional planning. using arcgis geographic information system software and envi remote sensing processing software, spss analysis and statistics software, and other methods and means, the development strategic planning program of the urban-rural system of jinghe river basin is formulated (yang et al., 2018). the plan is based on ecological protection, which changes the traditional planning method that takes economic benefits as the main factor and ignores ecological benefits. 269spatial distribution characteristics and optimum layout of urban and rural areas in different geomorphological types of the spatial distribution of urban and rural settlements in regional planning and hopes to provide theoretical guidance for the construction of new socialist urban and rural areas in the loess landform region (devos et al., 2018). results spatial distribution characteristics and optimal layout of urban and rural areas in loess landform area types and forms of urban and rural areas the results of the relationship between urban and rural types in different forms and topography in the jinghe river basin of the loess plateau are shown in figure 2. figure 2. relations between urban-rural types and terrain and geomorphology in different forms river channel type: the urban and rural areas are distributed on both sides or on one side of a wider river channel. due to the limitation of topography, the urban and rural land extends along the river banks or valleys, forming a zonal layout. the main characteristic of the banded layout is that its plane structure develops in the same direction with traffic lines and traffic flows, and has a strong directionality (peng et al., 2017). land tends to develop along rivers. residential, commercial services and industries are arranged on one or both sides of the main traffic lines. in the early stage of urban and rural development in chuandao, because of the small area, small scale, population concentration, low degree of regional differentiation, and less restrictive effect of terrain on urban and rural areas, the urban and rural agglomeration and compactness are high, showing a crumby structure. with the growth of urban and rural population, the expansion of urban and rural scale, and the differentiation of urban and rural functions, the spatial structure also changes. landform has a decisive impact on the expansion of urban and rural areas. urban and rural expansion generally chooses better conditions for the development of the river channel, and urban and rural areas begin to develop in a zonal manner. plateau surface type: urban and rural location is the plateau surface located in the loess plateau gully area and residual plateau gully area. generally, the large plateau surface (dongzhi plateau) is 400-500 meters higher than the valley, while the small plain (shili plateau) is 200-300 meters higher than the valley (wan mahari et al., 2018). the erosion around the plateau is active and serious, while the loess overlay on the plateau is deep, the land is flat and the soil quality is good, mainly homogeneous arable land. the cultivated land area is large, the farmland forest network is grid-like, and the patches of villages and towns are mosaic. in qingyang, dongzhi, and xiaojin towns on dong zhi plateau, changwu county on changwu plateau, majia town, and shili township on shili plateau, miaoqu town and mengba town on mengba plateau, urban and rural morphology mostly develop into blocks in the plateau survey of karst geomorphology research area research area guangnan county is located in the southeastern part of yunnan province, at the junction of yunnan, guangxi, and guizhou provinces, adjacent to guangxi zhuang autonomous region, northeastern wenshan prefecture. it is located at east longitude 104°31′105°39′, and northern latitude 23°29′-24°28′. its total land area is 7730.23 km2, including 11 townships and 166 administrative villages. it connects funing county and guangnan county to the east, xichou county, and malipo county to the south, qiubei county and inkshan county to the west, and xilin county of guangxi zhuang autonomous region to the north. it is located in the northeast of the karst plateau in southeastern yunnan, the terrain is high in the north and low in the south. the highest elevation point is barley land in the southwest, with an elevation of 2035 m. the lowest point is in the xiyang river area of banbeng township, with an elevation of 480 m. in the north, eroded low mountains, low-medium mountains, and mid-mountains are the main areas of potential rocky desertification; in the south, karst areas are serious rocky desertification, densely covered with karst rocks, mostly carbonate rocks, with thousands of karst landforms (külköylüoğlu et al., 2017). the study area is a sub-hot plateau monsoon climate with an average annual temperature of 16.6°c. at the end of 2011, the total population of guangnan county was 661.07 million, of which 660.784 million people were agricultural, accounting for 99.96% of the total population, which is a typical agricultural-dominated region. the county’s gdp is 545.5814 million yuan, per capita gdp is 824.54 yuan, and ethnic minorities are 406.39 million, accounting for 61.47% of the total population, which is a typical ethnic minority region. the per capita gdp is lower than the average level of the whole province, and the economic development is relatively backward. it is a key poverty alleviation county in yunnan province. influenced by karst landforms, topography, rivers, highways, economic development level, ethnic minorities, and other factors, there is spatial differentiation in the distribution of urban and rural settlements (zhiqiang et al., 2017). research method of spatial distribution using dem module of arcgis software, the distribution characteristics of urban and rural settlements in different areas of karst landform are analyzed by dividing the elevation and slope of guangnan county (dai et al., 2017), to explore the distribution and development trend of urban and rural settlements hidden behind the image, and to provide a research basis for further exploring the formation mechanism and historical evolution of regional settlements of karst landform type. this paper provides a scientific basis for the optimization figure 1. landscape of loess geomorphology source-gully intersection 270 jing li, zhongyuan cai, lianru duan topography. urban and rural development extends from inside to outside, mostly along traffic trunk lines. traffic trunk lines often become the guide of settlement development and evolution. the loess terrace plain type: it has a flat terrain, fertile soil, abundant water sources, convenient irrigation and drainage, convenient transportation, better soil texture, and better agricultural production conditions. the reclamation rate and cropping index are high, and agriculture and animal husbandry are relatively developed. for jingyang county, luqiao town of sanyuan county, liquan county, fenghuo town, zhao town, and so on, these common forms of urban and rural areas are centralized, compact land use, which is a compact and efficient layout form. cluster form is a common form of terrace plain in urban and rural areas. the land-use layout is compact, the road system presents a network pattern, and the development axis of urban and rural areas is not obvious. after the enlargement of urban and rural scale in the initial stage of belt development, the road system gradually evolved into a more complex network pattern, making urban and rural land more compact and saving land use. spatial characteristics of urban-rural system by the end of 2015, there were 327 urban and rural areas in jinghe river basin, including 2 prefecture-level cities, 21 county towns, and 304 townships. urban density is 0.4/10,000 square kilometers, urban and rural density (district and county level) is 4.9/10,000 square kilometers, and township density is 70.3/10,000 square kilometers. the statistical results of urban and rural quantity in the jinghe river basin are shown in table 1. table 1. quantitative statistics of urban and rural areas in the jinghe river basin grade prefecturelevel cities county town towns (townships) number of urban and rural areas 2 21 304 names of urban and rural areas qingyang city and pingliang city jingyang county, liquan county, qian county, yongshou county, chunhua county, xunyi county, bin county and changwu county in xianyang city; qingcheng county, zhengning county, ning county, heshui county, zhenyuan county, huachi county, zhenyuan county, lingtai county, jingchuan county, chongxin county and huating county in pingliang city; jingyuan county and pengyang county of guyuan city gaozhuang township, chongwen township, chonghuang township, majiawan township, arctic township, mazhai township, longgao township, etc. according to the topographic and geomorphological features, there are 170 townships in the loess plateau gully area, accounting for 55.9% of the total, while there are only 12 townships in the loess rocky mountain area, accounting for 4.6% of the total. according to the location of river channel, plateau and terrace, there are 142 townships on the surface of loess plateau, accounting for 46.7% of the total; there are only 26 villages and towns in the leung of the loess plateau, accounting for 8.6% of the total. the results of the analysis of the distribution and location of urban and rural areas according to the topography and geomorphology are shown in table 2. optimizing the spatial distribution characteristics of urban-rural system (1) according to the suitable index of human settlements environment and the spatial differentiation characteristics of the ecological space environment, the current situation of social and economic development and the need for the coordinated development of environment and economy in the future, the basic model of large dispersion and small aggregation is adopted. great decentralization maintains the institutions and functions of ecosystems in the form of zero integration. small agglomeration refers to the formation of larger and more agglomerated urban and rural areas on a large plateau and broad river roads. the main problem of the population on the loess plateau is that it is too dispersed (hua et al., 2013). the rational reorganization of urban and rural spatial structure can be achieved by guiding the population in the gully area to flow orderly to the urban and rural areas of the plateau and river channel. the urban-rural spatial system of the jinghe river basin is the main structure of five districts, nine river corridors, and four plates. five areas refer to ecological protection areas, ecological control areas, ecological coordination areas, human settlements control areas, and human settlements construction areas. the corridor of the jiujiang river system refers to the jinghe river, malian river, puhe river, heihe river, and its main tributaries, ruhe river, daxi river, milan river, malan river, and rouyuan river. the fourth plate refers to the dongzhiyuan plate dominated by qingyang city, the upper reaches of jinghe river dominated by pingliang city, the lower reaches of binchang xunjing river dominated by binchang county (changwu plateau) plate and the weihe river stepped plainly dominated by jingyang county and liquan county. the urban-rural system planning of jinghe river basin, in accordance with the principles of urban-rural coordination, perfection of functions, development of small cities and towns, and highlighting special features, promotes the functions of the core cities of pingliang city and qingyang city, optimizes the urban form, cultivates such regional sub-central cities as bin county, changwu county, jingyang county, qian county, liquan county, qingcheng county, ning county, zhenyuan county, huan county and jingchuan county, and improves the urban and rural functions of yongshou county, chunhua county, xunyi county, lingtai county, chongxin county, zhengning county, heshui county, huachi county, jingyuan county and pengyang county, so as to accelerate the key towns building, form a urban and rural system of jinghe river basin with regional characteristics of the loess plateau, and construct a five-level urban and rural system of core cities, regional sub-central cities, general cities, key towns and general towns. the scale of urban and rural classes in the jinghe river basin is shown in table 3. (2) carbon emissions in urban and rural areas can be reduced through industrial restructuring, healthy lifestyle and technological innovation, but these measures cannot change the traffic trip and its corresponding energy consumption and emissions caused by the layout of urban and rural spatial structure (jiang et al., 2013). once the urban and rural spatial structure determined by urban and rural planning is established, the structural adjustment of traffic trip will be very difficult. because of the lock-in effect of spatial structure of urban and rural systems on carbon emissions, the total amount of energy consumed and co2 and other greenhouse gases emitted by transportation increases rapidly and is very difficult to control. therefore, the establishment of eco-friendly urban-rural system spatial structure with “low-carbon” is the key to achieve the goal of ecological urban-rural construction in loess landform. table 2. distribution and location of urban and rural areas in terrain and landform (unit: number) position chuan dao tableland liang mao terrace total topographic features hilly-gullied loess region 55 0 24 0 79 loess plateau gully area 28 142 0 0 170 loess rocky mountain area 12 0 2 0 14 loess terrace plain area 0 0 0 41 41 total 95 142 26 41 304 271spatial distribution characteristics and optimum layout of urban and rural areas in different geomorphological types 1) principles of optimizing regional urban-rural spatial structure integrating the main body of urban and rural spatial planning according to loess landform characteristics, the control and guiding role of planning is strengthened; combining with the landform of jinghe river basin, a regional urban and rural spatial pattern with multi-center, network and public transport corridor-type is constructed, corridor-type development with track or regional public transport guidance is adopted, and the coordination of regional public transport network and spatial layout pattern are emphasized. functional urban areas are organized through the new functional division of labor between urban and rural areas in the region (ji & he, 2019). the adjustment of urban and rural spatial structure should be coordinated with the planning of employment and residence in order to achieve the “low-carbon” urban development goal, and the efficient public transport system should be used to effectively link urban and rural areas together to form a regional “functional balance”; the corridor-oriented spatial organization model guided by inter-city rail transit and regional public transport should replace the satellite-oriented and centripetal-oriented spatial organization model commonly used in current planning. the theoretical basis of spatial structure of urban and rural areas should shift from central geography theory to multipolar network nested theory model. the construction of largescale public facilities should be combined with a public transport hub to control the use of cars. the regional development model with weak control mainly by road network caused by the growth control mode of green belt will eventually promote the use of cars. more encouraging is the green wedge-spaced urban spatial expansion of public transport, focusing on new development of a public transport hub, to achieve a controlled compact relief. 2) strategies for optimizing regional urban-rural spatial structure according to the above theory, at the regional level, it is suggested that the urban-rural spatial structure of the loess landform type areas should follow the corridor-type development model guided by the important trunk roads such as railways and expressways, take the polarized development of urban and rural areas in key areas as the core, and combine the points and axes, form the urbanrural spatial network structure in series. in most other areas, the ecological environment is harnessed and restored to build a green base for urban and rural development (hong et al., 2018). under the restriction and influence of topographic and geomorphological conditions and water resources conditions, urban and rural construction land is mostly distributed in jinghe river. jinghe river basin is the first region to develop human settlements. apart from the military defense function of urban and rural areas, open terrain and adequate water resources are also important ecological support conditions. the main development of urban and rural areas is directly affected by the valley landform. urban and rural construction land develops along the terrain and its rivers under the stress of landform conditions, mostly on the terrace of surface runoff. among them, rural and urban reclamation, landscape, buildings and so on are distributed on the first terrace of jinghe river, while the second terrace is distributed on the low-density urban and rural construction, and the main urban and rural construction is distributed on the third terrace of jinghe river. regionally, the distribution of human settlements is uneven, and there are different distribution characteristics along with the difference of geomorphology. on the whole, the population gradually changes with the geomorphic combination characteristics, which are high in table 3. scale of urban and rural levels in the jinghe river basin scale grade level 1 urban and rural areas level 2 urban and rural areas level 3 urban and rural areas level 4 urban and rural areas level 5 urban and rural areas scale standard (person) 60 ten thousand about 10-25 ten thousand 5-10 ten thousand 3-5 ten thousand 1-3 ten thousand names of urban and rural areas central district of qingyang city and central district of pingliang city binxian county, changwu county, jingyang county, qian county, liquan county, qingcheng county, ningxian county, zhenyuan county, huating county, jingchuan county yongshou county, chunhua county, xunyi county, lingtai county, chongxin county, zhengning county, heshui county, huachi county, jingyuan county, pengyang county taicun town and malan town in xunyi county; runzhen town and fangli town in chunhua county; xinmin town, arctic town and shuikou town, binxian county; xianggong town and tingkou town, changning town, yongshou county; changning town, yongshou county; chagan town, yanxia town and caodong town in liquan county; fengyang town, qianxian county; yunyang town and wangqiao town in jingyang county; shizi town and chaona town in lingtai county; ankou town and maxia town in huating county; wangcun town and yaodian town of jingchuan county; xinyao town of chongxin county; kongtong town, baishui town and shilipu town in kongtong district; jingheyuan town and liupanshan town in jingyuan county; wang wa town and baiyang town in pengyang county; tunzi town, mengba town and pingquan town, zhenyuan county; gonghe town and yulinzi town in zhengning county; changqing bridge town, hesheng town and zaosheng town in ningxian county; old town and country in heshui county; xiaojin town and dong zhi town in xifeng district; yima town, sanlipu town and maling town, qingcheng county; huachi county yuancheng town and country; quzi town and mubo town of huanxian county other towns 272 jing li, zhongyuan cai, lianru duan the south and low in the north, and high in the east and low in the west. it also gradually decreases with the divergence of the main trunk of the jinghe river basin system to the end. spatial distribution characteristics and optimum layout of urban and rural areas in karst geomorphological area spatial distribution of urban and rural settlements in guangnan county (1) overall characteristics of spatial distribution of urban and rural settlements based on the vector distribution map of urban and rural settlements in 2015, arcgis software is used to select the comparative statistical analysis of landscape indices, such as the number of urban and rural settlements, the total area of settlements, the average area of settlements, the density of spots, the minimum area of spots and the maximum area of spots. the landscape indices of urban and rural settlements in guangnan county in 2015 are shown in table 4. table 4. landscape index of urban and rural settlements in guangnan county in 2015 scenery landscape index number of plaques (number) 14016 patch density (per /km2) 0.98% total patch area (m2) 76606672.8 minimum plaque area (m2) 399.02 maximum plaque area (m2) 266582.08 average patch area (m2) 5465.66 the density of urban and rural settlements in guangnan county is obtained by using arc info function in arcgis. generally speaking, the spatial distribution of urban and rural settlements is scattered, but not random and regular. under the influence of the unique geographical environment such as slope and elevation, the areas with moderate elevation and certain gradient are the main districts of urban and rural settlements in guangnan county. the urban and rural settlements in the areas with elevation above 1500 m and slope above 25° are sparse; the scale of urban and rural settlements along rivers is relatively large due to the influence of river system; and affected by the degree of transportation convenience and other factors, the urban and rural settlements along the highway are more intensive. (2) spatial distribution characteristics of urban and rural settlements at different elevations from 460 to 2020 m above sea level in guangnan county, the dem module of arcgis software is used to divide the altitude into five elevation zones by the equal grading method. the attribute data of urban and rural settlements are superimposed with the attribute data of elevation by overlay module, and the distribution area of urban and rural settlements in different elevation zones is obtained as shown in table 5. from table 5, it can be seen that, with the rising altitude, urban and rural settlements gradually become sparse and the climate has obvious changes, the higher the altitude is, the worse the agricultural production conditions are; the main urban and rural settlements in guangnan county live below 1500 m. adequate electricity, convenient transportation and water use are the advantages of low-altitude areas, which are convenient for villagers to live in for a long time. with a relatively large population, urban and rural settlements show a strong elevation orientation. to sum up, the distribution of urban and rural settlements and elevation shows a normal distribution relationship, as shown in figure 3. (3) spatial distribution characteristics of urban and rural settlements on different slopes using the dem module of arcgis software, the gradient is divided into five gradient zones according to the critical gradient grading method. overlay module is used to overlay the attribute data of urban and rural settlements and the related attribute data of gradient, and the distribution area of urban and rural settlements in different gradient zones is obtained, as shown in table 6. from table 6, it can be seen that the largest number of urban and rural settlement patches is between 2.1° and 25°, accounting for 89% of the total. in areas with slopes between 0° and 20°, the total proportion of urban and rural settlements is less than 10%, and the number of urban and rural settlements table 5. statistics of urban and rural settlements in different elevation zones in 2015 altitude (m) 460-955 956-1175 1176-1365 1365-1555 1556-2020 plaque (individual) 1122 2142 4922 3955 2131 number ratio (%) 7.97 15.24 35.06 26.03 15.16 plaque area (m2) 6871562.52 11636443.21 23835251.18 21549560.58 12824410.06 area ratio (%) 8.85 15.06 31.06 28.08 16.61 (a) number of patches on the map of urban and rural settlements at different elevations (b) patch area of urban and rural settlements map at different elevations figure 3. the relationship between urban and rural settlements and elevation in 2015 273spatial distribution characteristics and optimum layout of urban and rural areas in different geomorphological types above 25° is less than 5%. however, the area of urban and rural settlements in areas with slopes above 15° accounted for 76% of the total area. this is because guangnan county is a karst mountain area. limestone in the karst area is easy to be eroded by water to form groundwater. groundwater will be stored in the area with the lowest slope. soil in the rocky desertification area in the guangnan area is easy to be washed down by rainwater to the area with the lowest slope. therefore, residents will use the area with the lowest slope as cultivated land and settlements will choose the place with a certain slope. to sum up, we can get the relationship between the distribution of urban and rural settlements and the size of slope, as shown in figure 4. (a) number of patches on urban and rural settlements maps with different slopes (b) patch area of urban and rural settlement maps with different slopes figure 4. the relationship between urban and rural settlements and slope in 2015 optimizing the spatial distribution characteristics of urban and rural settlements from the above, it can be seen that karst landform determines the landscape index, spatial distribution characteristics and spatial distribution characteristics of urban and rural settlements at different elevations and slopes in guangnan county (soto et al., 2017). therefore, the planning of socialist urban and rural construction in karst geomorphic regions, the optimization of spatial structure of urban and rural settlements and the development of small and medium-sized urban and rural areas should be based on the full investigation of spatial structure and distribution of karst geomorphic settlements. on the other hand, karst environment is a very fragile ecological environment. the premise of every work is to harness, protect and improve the ecological environment, so we must optimize the settlement space structure based on the ecological concept, respect and follow the natural process. (1) respect the spatial self-organization model of existing urban and rural settlements the present settlement is a long-term adaptation to the fragile karst ecological environment and the result of the continuous harmony and unification of the contradictions between human and land. therefore, when optimizing the living space, we must respect the spatial self-organizing mode of the existing settlement, and consider the natural conditions, social and economic development and residential culture of the karst geomorphological type area (martín-díaz et al., 2018). (2) moderate settlement scale karst is a very fragile ecological environment with many mountains and few flat lands. therefore, the environmental capacity of urban and rural settlements development is very limited, and the appropriate scale of settlements must be considered. appropriate settlement scale in karst urban and rural areas should be based on population, land use of settlements, the relationship between them and regional carrying capacity of cultivated land resources, land use structure, water resources, rocky desertification and soil erosion, labor employment, population migration, social and economic development and other factors. regardless of the population of settlements or the scale of residential land, there are huge spatial differences in karst-plateau-basin area, plateau-mountain area and plateau-canyon area. it is necessary to evaluate comprehensively from various aspects and complete the investigation and prediction of the current situation with different methods in order to formulate an appropriate and reasonable scale of urban and rural settlements and optimize the spatial structure of urban and rural settlements in karst landform type areas. (3) monitoring of settlement space development in recent years, urban and rural settlements in karst areas have many new characteristics in spatial extension, such as the empty and waste of old settlements, the strip extension of new buildings along highways, etc. (yan et al., 2018). in addition, the changes of settlement land use, regional land use type and structure, especially the changes of rocky desertification and soil erosion, have an important impact on the development of settlement space, and directly penetrate into residential culture. therefore, it is necessary to use the dem module of arcgis software to monitor real-time in order to investigate the problems in settlement development, predict the trend of settlement evolution, and make a good spatial planning of settlement development. (4) optimizing the residential space of settlements on the basis of the existing settlement spatial pattern, we should optimize the spatial pattern and focus on the control and restoration of the ecological environment, especially the control of rocky desertification and soil erosion. on this basis, we should realize the rational utilization of land and the optimization of spatial structure, strive for more and more reasonable land for urban and rural table 6. statistics of urban and rural settlements in different slope zones in 2015 slope (°) 0-2 2.1-6 6.1-15 15.1-25 >25 plaque (individual) 111002 418703 605243 458222 81206 number ratio (%) 6.62 24.89 36.03 27.25 4.76 pplaque area (m2) 576393931.50 275426853.04 984562766.30 2477144677.65 3416321408.14 area ratio (%) 7.35 3.45 12.63 32.04 44.09 274 jing li, zhongyuan cai, lianru duan settlement spatial development, and develop and protect the cultivated land, forest land and garden land on which we depend for survival. emphasis should be placed on supporting facilities construction, environmental improvement and transportation facilities construction (hong, 2017). discussion the regional topography and physiognomy of loess landform types play a decisive role in the spatial distribution of urban and rural areas. this paper uses gis and related statistical analysis software to analyze the overlap and correlation of the topography and physiognomy of jinghe river basin, urban and rural patches, and summarizes three basic urban and rural types, namely, river channel type, plateau surface type and terrace type. at the same time, the spatial distribution characteristics of urban and rural systems are analyzed in detail. according to the suitable index of human settlements environment and the spatial differentiation characteristics of ecological space environment in jinghe river basin, the present situation of social and economic development and the need of coordinated development of environment and economy in the future, the main structure of urban and rural space system, corridor of five districts, nine rivers system and four plates is put forward under the background of the new round of urbanization in the jinghe river basin, in order to solve the contradiction between urban and rural development and ecological environment deterioration in jinghe river basin, provide guidance for the coordinated development of regional natural environmental protection and social economy, and promote the scientific and orderly development of urban and rural areas. arcgis is used to extract the urban and rural settlement patches and generate the density map of urban and rural settlement in guangnan county. the total area of the patches, the average area of the patches, the density of the patches and other landscape indices are obtained for comprehensive analysis. elevation map and slope map are generated by dem module, and then overlapped with urban and rural settlement patches by overlay module. through the above analysis, the following conclusions can be drawn: (1) due to the high environmental vulnerability and sensitivity of karst mountainous areas and the constraints of human activities, rocky desertification in guangnan county is serious, and the area of rocky mountains is wide and dispersed, which results in the extremely dispersed distribution of settlement patches. because of the difference of natural conditions (elevation, elevation, slope) in guangnan county, there are obvious spatial differences in the distribution of urban and rural settlements. (2) with the change of different elevation zones, the pattern of urban and rural settlements also changes. the pattern of urban and rural settlements at 1100-1500 m elevation is the most, which indicates that this elevation zone is most suitable for the location of urban and rural settlements. (3) from the view of urban and rural settlement patches in different slope zones, there are obvious differences and uniqueness between slope and plain. the number of settlement patches is normal distribution with slope, but the area of urban and rural settlement patches is proportional to slope. conclusions natural environment is the material basis for human survival and social and economic development in urban and rural areas. different geographical units limit the spatial environment for urban and rural development. distribution characteristics and trends of urban and rural space in different geomorphic regions are not only influenced by socio-economic development and national cultural customs, but also by the surrounding natural environment conditions, which have a far-reaching impact on them, and even affect the appearance, size, layout and trend patterns of urban and rural settlements. therefore, in view of the spatial distribution characteristics of urban and rural areas of loess and karst landforms, this paper puts forward the optimal layout strategy, which provides the basis for choosing the reasonable model of urban and rural development and formulating the optimal countermeasures of urban and rural interactive development. there are some shortcomings in this study. because of the different data sources, the results of spatial feature analysis may be different. with the continuous updating and improvement of data sources, this study will be more accurate. in addition, this paper only analyses the natural environmental factors, but fails to carry out relevant research based on the regional socio-economic data of different geomorphological types. therefore, it fails to mine deeply into the characteristics, structure and laws of urban and rural settlements, which will become the direction of further research in the future. acknowledgments this work is supported by national natural science foundation of china, study on the green construction of rural settlement units in northwest china (51678481); shaanxi provincial education department special research project (17jk0460); xi’an university of architecture and technology, 2019 special research fund for new urbanization, (general items); xi’an polytechnic university, the project of general project of philosophy and social science (2019zxsk02). references aumaître, g., bourlès, d. l., & keddadouche, k. (2017). deglaciation in the central pyrenees during the pleistocene–holocene transition: timing and geomorphological significance. quaternary science reviews, 162, 111-127. dai, c. y., yuan, j. h., zhang, f. h., chen, x., & zhang, h. (2017). study on output characteristics of photovoltaic array based on dmppt. chinese journal of power sources, 41, 253-256. devos, a., fronteau, g., lejeune, o., sosson, c., chopin, e., & barbin, v. (2018). influence of geomorphological constraints and exploitation techniques on stone quarry spatial organisation: example of lutetian underground quarries in rheims, laon and soissons areas. engineering geology, 115, 268-275. hong, s. c., magarey, r., borchert, d. m., vargas, r. i., & souder, s. (2018). site-specific temporal and spatial validation of a generic plant pest forecast system with observations of bactrocera dorsalis (oriental fruit fly). neobiota, 27, 37-67. hong, y. (2017). structural characteristics and applications of hilbert’s type integral inequalities with homogenous kernel. journal of jilin university (science edition), 55, 189-194. hua, s., li, x., jie, j., mao, x., & qun, w. (2013). copper spatial distribution characteristics in typical mining area base on gis technology. information technology journal, 12(21), 6345-6349. ji, n., & he, g. r. (2019). texture feature extraction method of kiwifruit shape scar image based on gray level recognition. automation & instrumentation, 231, 165-168. jiang, m., li, x., & peng, t. (2013). a grey-box modeling approach for the reduction of spatially distributed processes using new basis functions. information technology journal, 12(22), 7019-7023. külköylüoğlu, o., yılmaz, s., & yavuzatmaca, m. (2017). comparison of ostracoda (crustacea) species diversity, distribution and ecological characteristics among habitat types. fundamental & applied limnology, 190, 63-86. li, s. d., bai, y. h., jiang, y., wang, z. h., wei, w. h., li, x., & liu, l. n. (2017). explicit statistic meteorological early-warning model of geological hazards in xinjiang, china based on the genesis theory of endogenic and exogenic coupling. journal of earth sciences & environment, 39, 286-300. lu, z. j., liu, x., qin, y. g., pan, y., guo, j. g., & huang, s. (2017). an adaptive multi-sensor management method for cooperative perception. journal of china academy of electronics and information technology, 12, 353-358. martín-díaz, j., palma, p., golijanin, j., nofre, j., oliva, m., & cengic, n. (2018). the urbanisation on the slopes of sarajevo and the rise of geomorphological hazards during the post-war period. cities, 72, 60-69. mori, t. (2018). increasing returns in transportation and the formation of hubs. japanese journal of public health, 46, 113-21. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dianyjs201702024 about:blank about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201702002 about:blank about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201901044 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201901044 about:blank about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201704005 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201704005 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201704005 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201704005 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201704005 about:blank about:blank 275spatial distribution characteristics and optimum layout of urban and rural areas in different geomorphological types nematollahi, m. r., fathipour, y., talebi, a. a., karimzadeh, j., & zalucki, m. p. (2018). sampling procedure and temporal-spatial distribution of the cabbage aphid, brevicoryne brassicae l. (hemiptera: aphididae), on canola. journal of agricultural science & technology, 16, 1241-1252. peng, w. x., wang, l. s., mirzaee, m., ahmadi, h., esfahani, m. j., & fremaux, s. (2017). hydrogen and syngas production by catalytic biomass gasification. energy conversion and management, 135, 270-273. samadzadegan, f., hasani, h., & reinartz, p. (2017). toward optimum fusion of thermal hyperspectral and visible images in classification of urban area. photogrammetric engineering & remote sensing, 83, 269-280. soto, m. v., sarricolea, p., sepúlveda, s. a., rodolfi, g., cabello, m., & maerker, m. (2017). assessment of hydro-geomorphological hazard potentials in the chilean semiarid coastal range and its impacts on la serena city, coquimbo region. natural hazards, 88, 1-22. wan mahari, w. a., chong, c. t., lam, w. h., anuar, s. t., ma, n. l., ibrahim, m. d., & lam, s. s. (2018). microwave co-pyrolysis of waste polyolefins and waste cooking oil: influence of n2 atmosphere versus vacuum environment. energy conversion and management, 171, 1292-1301. yan, h., yu, k., shi, q., lin, z., zhao, m., tao, s., & liu, g. (2018). airsea co 2 fluxes and spatial distribution of seawater p co 2 in yongle atoll, northern-central south china sea. continental shelf research, 165, 71-77. yang, y., zhang, x., & li, m. (2018). grid impedance detection method under unbalanced and harmonic grid voltage based on multiple complex coefficient filters. journal of power supply, 16, 69-75. yasuhara, m., tittensor, d. p., hillebrand, h., & worm, b. (2017). combining marine macroecology and palaeoecology in understanding biodiversity: microfossils as a model. biological reviews, 92, 199-215. zhang, x. j., & fan, d. y. (2017). tilt license plate image edge fuzzy feature recognition method research. computer simulation, 34, 372-375. zhiqiang, l., pang, r., zhu, j., deng, r., & shen, l. (2017). the correlation analysis of surface thermal environment and settlement characteristics in the hilly city. journal of chongqing university, 40, 60-71. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201802011 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201802011 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201802011 about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201701080 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201701080 about:blank keywords: combination analysis; urban agglomeration development zone; land resources; utilization potential; combination empowerment; land intensive. palabras clave: análisis de combinación; zona de desarrollo de aglomeraciones urbanas; recursos de la tierra; potencial de utilización; empoderamiento combinado; uso intensivo del suelo. how to cite item miao, j., & lu, q. (2020). an empirical study on the potential of land resources utilization in urban agglomeration development zones based on combination analysis. earth sciences research journal, 24(4), 439-447. doi: https:// doi.org/10.15446/esrj.v24n4.92069 to alleviate the contradiction between human and land caused by the limitation of land resources and the irreversibility of land use in the development zone of urban agglomeration, 24 indexes were selected from four aspects: land use status, resource security, economic feasibility, and social acceptability to establish the level evaluation index system for the sustainable use of land resources in baoding high-tech zone; also, the combination weighting method was used to determine the development level. the index weight of land resource utilization potential in the development zone was established. the subjective and objective linear combination was also established to obtain the proportion of subjective and objective. the multi-factor comprehensive evaluation model was used to evaluate the sustainable utilization degree of land in the high-tech zone. the comprehensive analysis of land-intensive use potential was made from five aspects: expansion potential, structure potential, intensity potential, management potential, and the number of years of land available; the analysis of the limiting factors and the combination of these factors of the land in the development zone was made to obtain the natural index after the renovation, and the natural quality potential of the cultivated land renovation was obtained according to the difference of the natural index before and after the renovation. the empirical results show that the high-tech zone’s land-use rationality index is relatively high during 2014-2018. during the evaluation period, the index continued to rise; the high-tech zone's resource security score was low, and there was no change during the evaluation period, which is basically at a low level. during the five years, the high-tech zone’s land-use sustainability has experienced the fluctuation process of first rising and then declining, which is in the stage of medium sustainable development. in the future, the potential for improvement is enormous. the prospect of the high-tech zone’s exploitable land structure is small, and the land structure is relatively reasonable, which is in line with the development orientation of the high-tech site. but the industrial land intensity in the zone also has a specific potential for excavation. the high-value areas of the natural quality potential of the land improvement in the high-tech site are mainly distributed in the west, northeast, and a few in the north and south. abstract an empirical study on the potential of land resources utilization in urban agglomeration development zones based on combination analysis estudio empírico sobre el potencial de utilización de los recursos de la tierra en una zona de desarrollo urbano con base en el método de análisis de combinación issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.92069 este estudio se realiza con el fin de conocer mejor el paradigma entre humanos y el uso de la tierra, mediado por la limitación de los recursos y la irreversibilidad de su uso en zonas de desarrollo de la aglomeración urbana. con este fin se seleccionaron 24 índices de cuatro aspectos: estado de uso de la tierra, seguridad de los recursos, viabilidad económica y aceptabilidad social, para establecer el sistema de índice de evaluación de nivel para el uso sostenible de los recursos de la tierra en la zona de alta tecnología de baoding; además, se utilizó el método de ponderación de combinación para determinar el nivel de desarrollo. se estableció el índice de ponderación del potencial de aprovechamiento de los recursos de la tierra en la zona de desarrollo, y se estableció la combinación lineal subjetiva y objetiva para obtener la proporción de subjetivo y objetivo. el grado de utilización sostenible de la tierra en la zona de alta tecnología se evaluó mediante el modelo de integral multifactorial. el análisis integral del potencial de uso intensivo del suelo se realizó a partir de cinco aspectos: potencial de expansión, potencial de estructura, potencial de intensidad, potencial de manejo y el número de años de tierra disponible. luego se realizó el análisis de los factores limitados y la combinación de los factores limitados del suelo en la zona de desarrollo para obtener el índice natural después de la renovación, y el potencial de calidad natural de la renovación del suelo cultivado se obtuvo según la diferencia del índice natural antes y después de la renovación. los resultados empíricos muestran que la racionalidad del estado de uso de la tierra de la zona de alta tecnología es relativamente alta durante 2014-2018. durante el período de evaluación, el índice siguió aumentando, el puntaje de seguridad de los recursos de la zona de alta tecnología fue bajo y no hubo cambios durante el período de evaluación, que básicamente se encuentra en un nivel bajo. durante los cinco años, la sostenibilidad del uso del suelo de la zona de alta tecnología experimentó el proceso de fluctuación de primero alza y luego declive, lo que se encuentra en la etapa de desarrollo sostenible medio, y en el futuro el potencial de mejora es grande; el potencial de la estructura de suelo explotable de la zona de alta tecnología es pequeño, y la estructura del suelo es relativamente razonable, lo cual está en línea con la orientación de desarrollo de la zona de alta tecnología, pero la intensidad de suelo industrial en la zona también tiene un cierto potencial de excavación. las áreas de alto valor del potencial de calidad natural del mejoramiento territorial en la zona de alta tecnología se distribuyen principalmente en el oeste, noreste y algunas en el norte y sur. resumen record manuscript received: 06/01/2019 accepted for publication: 07/07/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 439-447 jianjun miao, qian lu college of economics and management, nanjing university of aeronautics and astronautics, nanjing 210000, china *corresponding author: miaojianjun98@163.com; 18520204606@163.com e n v ir o n m e n ta l g e o l o g y https://doi.org/10.15446/esrj.v24n4.92069 https://doi.org/10.15446/esrj.v24n4.92069 https://doi.org/10.15446/esrj.v24n4.92069 mailto:miaojianjun98@163.com mailto:18520204606@163.com 440 jianjun miao, qian lu introduction since the establishment of the economic and technological urban agglomeration development zones in china during 1984, the cause of urban agglomeration development zone has been booming. remarkable achievements have been made in infrastructure construction, foreign capital, economic development, export expansion, talent cultivation, pilot reform, and service to the host city (wang & chen, 2016). it has also achieved good results in playing a window to the outside and radiation to the inside. however, there are too many problems in the construction of the development zone, mainly manifested in two aspects: the excessive establishment of the development zone and the rapid expansion of the development zone land, which increase the pressure of cultivated land protection. according to the regulations, the establishment of all kinds of development zones shall be subject to the twolevel examination and approval system of the state council and the people’s governments of provinces, autonomous regions, and municipalities directly under the central government. people’s governments at all levels below the provinces, autonomous territories, and municipalities directly under the central government shall not examine and approve the establishment of all kinds of development zones. some cities and counties set up all types of development zones without respecting regulations, resulting in severe waste and repeated land use (cao et al., 2016; dong et al., 2016). also, some provincial development zones set a low threshold for enterprises to enter and cannot realize the intensive use of land, which is also one reason for the rapid expansion of land use in the development zone. simultaneously, many development zones in the construction process, too much pursuit of “large and complete,” the land is going on the extensive development road of quantitative expansion. according to the ministry of land and resources, there are 6015 development zones (parks) in china, with a total planning area of 35400 square kilometers, equivalent to the entire construction land area of 660 cities and all construction towns in china. besides, a large amount of cultivated land, especially high-quality and high-yield cultivated land, is occupied in the development zone’s construction process, which causes tremendous pressure on the protection of cultivated land in china (li et al., 2017). blind expansion of development zones and extensive land use in the zones have severely damaged the development zones and urban construction, which has dramatically damaged the land resources already in short supply. therefore, evaluating the potential of land use in urban agglomeration development zone is the internal requirement of protecting cultivated land, saving urban construction investment, and strengthening land management (ding, 2019; price et al., 2019). in this paper, the index weight of land resource utilization potential in baoding high-tech development zone is determined using the combination weighting method. based on the index weight, the sustainable utilization level of land resources in the high-tech development zone is objectively evaluated by the multi-factor comprehensive evaluation model. the potential of landintensive utilization in five aspects of expansion potential, structural potential, and intensity potential are combined and analyzed. the potential of land can be changed through the development zone. the combination of unrestricted factors and restrictive factors can calculate the natural quality potential of land remediation in the high-tech zone and put forward the countermeasures for sustainable utilization of regional land, which provides an essential theoretical basis fore planning new villages in the high-tech area. basic definitions regional overview regional natural conditions: baoding’s high-tech zone is located in the ancient city of baoding, with flat terrain, an average elevation of 18.5m, and a slope of 1.0 ‰. the foundation’s bearing capacity is 18-20t per square meter, with a seismic fortification degree of 7. it belongs to a warm temperate continental semi-arid monsoon climate with four distinct seasons and mild weather. the annual average temperature is 12.3 °c; the coldest month is january, the average temperature is -4.2 °c, the warmest month is july, the average temperature is 26.7 °c, and the annual frost-free period is about 210 days. the precipitation is mainly from july to september, with an average yearly rainfall of 575mm. four distinct seasons, dry and windy in spring, hot and rainy in summer, cool in autumn, cold and little snow in winter. regional location: the central area, east area, and west area of baoding national high-tech industrial development zone are the primary areas. the central area is located in the north of baoding city, starting from torch road in the east, zhichang road in the south, xiangyang north road, luxin south road, luxin east road, luxin north road, jianshe north road in the west, and the north outer ring road in the north, with an area of 5.849km2. the eastern region is located in the east of baoding city, adjacent to expressway in the east, qiyi road in the south, baoding teachers college in the west and north, with an area of 0.185km2; the western region and the experimental site are located in baoding. in the west of the city, the west district area is 1.689km2, 800m from the flood control dike in the east to the west (li et al., 2017), 337 regiment family house in the west, jiangcheng road in the north. the experimental area is 2.890km2, from jianshe road in the east to the flood control dike in the west, jiangcheng road, baoshen railway, fuchang road in the south, kangzhuang road in the north. evaluation of land resource utilization potential based on combination weighting evaluation system the selected indicators should follow the principles of objectivity, scientifically, representativeness, comprehensiveness, dynamism, and operability to ensure the rationality of the evaluation results (arneth et al., 2018; song & deng, 2017). combined with the actual situation of land use in high-tech zone and the characteristics of sustainable land use in the county and referring to the evaluation standards of sustainable land use in china and abroad, an evaluation system of sustainable land use with the overall goal of sustainable land use in a high-tech zone is finally established, which includes four criteria levels: land-use level, resource security, economic feasibility, and social acceptability, as shown in table 1. index weight determination due to the subjective weighting method’s lack of objectivity, the weight determined by the objective weighting method will change with the environment’s change. this paper uses the subjective and objective combination method to determine the weight, the combined weighting method, to avoid the shortcomings of the subjective and objective weighting. the analytic hierarchy process (ahp) is a decision-making method that decomposes the decision-making elements into objectives, criteria, schemes, and other qualitative and quantitative analysis levels. in this paper, the ahp method is used to weight the indicators subjectively, and the results are shown in table 2. on this basis, the variation coefficient method is used to weight its indicators objectively, and the result of variation coefficient weighting is shown in table 3. the variation coefficient method directly uses the information contained in each index to calculate the weight of the index. the combination weighting is then carried out; that is, the subjective and objective linear combination is established to obtain the proportion of subjective and objective. the combination weighting value is also shown in table 3. evaluation model of land sustainable use degree factors comprehensive evaluation model and fuzzy comprehensive evaluation model are used to evaluate the sustainable use of land. because many factors affect the sustainable use of land, and applying a multi-factor comprehensive evaluation model is relatively mature, this paper uses a multifactor comprehensive evaluation model to evaluate the degree of sustainable land use in the high-tech zone (maranguit et al., 2017). suppose that there are n sub-objectives b t b t b tn1 2( ) ( ) … ( ), , , in the evaluation index system, and each sub-objective is composed of several indexes x ti ( ), where t is the year, and the constructor: a t w t b tj n j m ( ) = ( ) ( ) = ∑ 1 (1) 441an empirical study on the potential of land resources utilization in urban agglomeration development zones based on combination analysis calculation of the potential of intensive use of land in the high-tech zone (1) calculation of expansion potential the calculation formula of expansion potential is as follows: the area of available land = the land’s total area in the development zone the area of supplied state-owned construction land non-available land. (2) calculation of structural potential structural potential = area of built-up urban construction land × (ideal value of industrial land rate current value of industrial land rate). (3) calculation of strength potential strength potential = area of completed industrial and mining storage land × i i ii p i( ) / , where i i is the ideal value of comprehensive plot ratio of industrial land, building density of industrial land, input intensity of fixed assets of industrial land, output intensity of industrial land, and ip is the current value of index. b t p t x tn i i i n ( ) = ( ) ( ) = ∑ 1 (2) in formula 1, b tn ( ) is the annual evaluation index of the n sub-goal; a t( ) is the comprehensive evaluation value of annual sustainability; x ti ( ) is the standardized value of the i index, that is, the value of each index unified to dimensionless 0 ~ 1. in formula (2), p ti ( ) is the relative weight of the subgoal, and w tj ( ) is the weight of the total goal. it is necessary to set the land use potential level’s judgment standard to evaluate the degree of land use potential vividly and intuitively. according to some researches, the sustainable land use is divided into five grades, and the corresponding comprehensive evaluation values of each stage are < 0.25 (very low), 0.25-0.45 (low), 0.450.65 (medium), 0.65-0.85 (high), and > 0.85 (very high). table 1. criteria and indicators of sustainable land use evaluation system in high-tech zones criterion level land use status(b1) resource security(b2) economic feasibility(b3) social acceptability(b4) index level c11 land productivity (10,000 yuan/ha) c21 population density (person/ square kilometre) c31 per capita gdp (round) c41 per capita road length (km/ person) c12 proportion of effective irrigation area of cultivated land c22 per capita urban public green space area (square meters/person) c32 agricultural output value as a proportion of gdp c42 annual investment in education as a proportion of gdp c13 crop multiple cropping index c23 per capita arable land area (hectare/person) c33 the proportion of secondary industry output value to gdp c43 number of doctors per 10,000 people c14 the proportion of cultivated land c24 per capita food occupancy (kg/person) c34 proportion of output value of the tertiary industry in gdp c44 tap water penetration rate c12 proportion of effective irrigation area of cultivated land c25 annual reduction rate of cultivated land c35per capita net income of farmers (yuan) c45 growth rate of per capita net income of agricultural population c13 crop multiple cropping index c36 per capita gdp (10000 yuan/ha) c46 non-agricultural employment labor force ratio c14 the proportion of cultivated land c37 completion amount of fixed assets investment per unit area (10000 yuan / ha) c47 natural population growth rate c38 elasticity coefficient of land income growth table 2. subjective empowerment results index subjective weight value index subjective weight value c11 0.1560 c34 0.0074 c12 0.0593 c35 0.0108 c13 0.0928 c36 0.0324 c14 0.0288 c37 0.0228 c21 0.0166 c38 0.0442 c22 0.0237 c41 0.0190 c23 0.0971 c42 0.0316 c24 0.0641 c43 0.0939 c25 0.0368 c44 0.0470 c31 0.0156 c45 0.0686 c32 0.0035 c46 0.0097 c33 0.0050 c47 0.0134 table 3. objectively weighted and combined weighted index c11 c12 c13 c14 c21 c22 c23 c24 weighting of coefficient of variation 0.1097 0.0016 0.0033 0.0006 0.0012 0.0208 0.0012 0.0062 combined weight value 0.1328 0.0305 0.0481 0.0147 0.0089 0.0223 0.0492 0.0351 index c25 c31 c32 c33 c34 c35 c36 c37 weighting of coefficient of variation 0.1986 0.0609 0.0836 0.0347 0.0025 0.0307 0.038 0.0265 combined weight value 0.1177 0.0383 0.0435 0.0198 0.0050 0.0207 0.0352 0.0247 index c38 c41 c42 c43 c44 c45 c46 c47 weighting of coefficient of variation 0.1184 0.0541 0.0195 0.0208 0.0079 0.0673 0.0057 0.0864 combined weight value 0.0813 0.0365 0.0256 0.0573 0.0274 0.0679 0.0077 0.0499 442 jianjun miao, qian lu (4) measurement of management potential since the establishment of the high-tech development zone, the land management has been strictly controlled, and the idle land has been identified and disposed of in time (meng et al., 2018; houghton & nassikas, 2017; paulemercado et al., 2017). the land for paid use is generally supplied according to the legal maximum service life. at present, there is no idle land that has not been disposed of in the high-tech development zone, and there is no paid use of the expired land. (5) calculation of available years table 5 shows the calculation of the number of years that baoding hightech zone can provide land for expansion potential. years of land available i = expansion potential / annual average area of supply land in the previous three years; years of land available ii = expansion potential/area of supply land in the previous five years, maximum value. calculation of natural quality potential of land in the high-tech zone most of the traditional factor combination methods directly combine all factors of agricultural land classification. however, this method ignores the removability of factors, which leads to a large amount of calculation (yang et al., 2017), complex combination process, but also causes interference to the evaluation results. according to td/t1013-2013 code for acceptance of land remediation projects, adequate soil layer thickness, salinization degree, soil organic matter content, soil ph value, surface soil texture, drainage conditions, and irrigation assurance rate are the ranges for acceptance of cultivated land remediation quality. after the land remediation project, its properties can mutate, and other factors in the remediation process are all in funds. under the background of technology status and ecological environment protection, it is challenging to develop shortly; the environment is seriously damaged (chen et al., 2017) and not suitable for large-scale promotion. based on this theory, the authors could combine the land renovation factors to find the limiting factors. combination of modifiable factors according to the results of supplementary improvement of cultivated land quality in 2010 in the high-tech zone, the cultivated land quality grading factors in the zone include irrigation assurance rate, surface soil texture, soil organic matter content, and profile configuration, with 4, 3, 5 and 7 grading levels respectively (table 4), and three modifiable factors. corresponding codes express the score of modifiable factors of the agricultural land grading unit. the factor scores (10-100 points) correspond to 10 codes of 1, 2, 3, 4, 5, 6, 7, 8, 9 and + respectively. the three factors’ score codes are combined according to the order of grading factors in table 4 to be the combination type identification of modifiable factors of natural quality assessment unit. for example, if the combination type of modifiable factors in an evaluation unit is marked as “398”, it means that the irrigation assurance rate score of the unit is 30 points, corresponding to no irrigation conditions; the texture score of the surface soil is 90 points, corresponding to the surface soil is mainly clay; the quality score of the soil organic matter is 80 points, and the quality score of the corresponding organic matter is more than 2%. by analogy, all combination types of three factors of all evaluation units are obtained. combination of limiting factors because crop growth depends on the satisfaction degree of the living environment to its optimal growth environment factor value (marked by factor score in this paper), the small satisfaction degree is the main limiting factor affecting crop growth; and the main reason concerning the production capacity of agricultural land. the first limiting factor, i.e., the largest limiting factor, is the least satisfied, followed by the second limiting factor, and so on. according to this principle and grading standard, when the grading factor’s score is 50 or below, the factor will have a major limiting effect on crop growth. if the content of soil organic matter is less than 0.6% (i.e., the score of organic matter is 20), it is poor in nutrients; the texture of surface soil is gravelly soil, the salinization of soil above heavy level and the condition of no irrigation and drainage have seriously affected the cultivation of agricultural land. therefore, the factor scores between 10 and 50 (the identification code is 1, 2, 3, 4, and 5) are regarded as the limiting factors. the combination containing the limiting factors is selected from this combination of the modifiable factors to determine the type of the combination of the modifiable limiting factors, as the research object. calculation of natural quality potential according to the light and temperature production potential index and yield ratio coefficient of each designated crop, combined with the grading factor weight and factor index score of the selected crop, calculate the natural quality index before remediation following the regulations. then, combined with the study area’s actual situation, the qualitative and quantitative methods such as gis and remote sensing are comprehensively used to analyze the factors that can be transformed, and the degree of transformation. after remediation, the natural quality of cultivated land after remediation is reevaluated to obtain the index of natural quality. finally, the potential of the natural quality of cultivated land remediation is obtained according to the difference of natural index before and after remediation. the calculation formula is as follows: r a w fij tj j s k k k r j=         = = ∑ ∑ 1 1 10000/  (3) ′ = ′         = = ∑ ∑r a w fij tj j s k k k r j 1 1 10000/  (4) r r ri ij ij= ′  (5) among them, k is the number of grading factors; s is the number of designated crops; r is the number of grading factors; wk is the weight of the table 4. agricultural land grading and other factors in high-tech zones factor score factor score code irrigation guarantee rate topsoil texture soil organic matter content/% profile configuration 100 +(10) fully satisfy loam loam/clay/ soil 90 9 satisfy clay soil/sticky/ sticky, soil/ sand/soil, sand/sticky/ sticky 80 8 >2.0 stick/sand/ stick, whole body stick 70 7 general satisfaction sand sand / clay / sand 60 6 1.5~2.0 soil/sand/ sand, shallow viscous 50 5 1.0~1.5 stick/sand/ sand 40 4 0.6~1.0 sand and gravel 30 3 no irrigation conditions 20 2 <0.6 10 1 443an empirical study on the potential of land resources utilization in urban agglomeration development zones based on combination analysis k grading factors; j is the yield ratio coefficient of the j crop; atj is the light and temperature production potential index of the j crop; fk and ′fk are the index scores corresponding to the modifiable limiting factors before and after the treatment; rij, ′rij and ri respectively represent the natural quality indexes and potential of the j crops before and after treatment. results data preprocessing taking baoding national high-tech development zone as the research area, 400m×400m grid division is adopted. through the fishnet tool in arcgis software, the grid is created. after spatial connection and spatial superposition, the spatial relationship between the grid and multi-source data such as land resources, linear features is established to prepare for the grid computing of evaluation indexes. the research area is divided into 4338 grid units, as shown in figure 1. figure 1. grid unit for the quality potential evaluation of land resources in the high-tech zone evaluation results of land resource utilization potential of baoding hightech zone according to the data of baoding statistical yearbook (2014-2018) and statistics bureau of high-tech zone (2014-2018), combined with the evaluation model, the land sustainable utilization degree of the high-tech zone in 20142018 was calculated by levels, and the evaluation results are shown in table 5. on this basis, figure 2 was obtained. table 5. evaluation results of sustainable land use in high-tech zones from 2014 to 2018 criteria level evaluation particular year land use status resource security economic feasibility social acceptability comprehensive evaluation results 2014 0.48 0.21 0.34 0.42 0.43 2015 0.49 0.23 0.45 0.65 0.56 2016 0.71 0.21 0.55 0.57 0.61 2017 0.77 0.22 0.49 0.77 0.57 2018 0.81 0.21 0.46 0.61 0.51 figure 2. indicator schematic map of sustainable land use in high-tech zones from 2014 to 2018 in conclusion, the evaluation results of sustainable land use in the hightech zone are as follows: ① the rationality of land use in the area is relatively high, and the index continues to rise during the evaluation period, from 0.48 in 2014 to 0.81 in 2018, which is at a sustainable high level. ② in terms of resource security, the high-tech zone score is low, there is no change during the evaluation period, and it is basically at a low level. the main reason for the low resource security index is that the per capita urban public green space area (0.12 m2/person), per capita cultivated land area (0.034684 ha/person), per capita grain occupation (300 kg/person), and other indexes are far from the national standard value. ③ although the economic feasibility index has moved from a lower level to a medium level, it fluctuates greatly. after reaching a high of 0.55 in 2016, it shows a downward trend in the following two years. ④ the social acceptability is more reasonable. during the evaluation period, the index moved from the lower to the middle and higher levels, but its fluctuation is too large. ⑤ comprehensively, in the past five years, the sustainable degree of land use in the high-tech zone has experienced a fluctuating process of rising first and then declining, but it is basically in the stage of medium sustainable development, with great potential for future improvement. potential measurement results according to the calculation methods of expansion potential, intensity potential, structure potential, and years of land available, the current land resource potential data of baoding national high-tech development zone can be obtained as follows: 712.96 hm2 of urban construction land has been built in the high-tech development zone, and 65.91 hm2 of land has been supplied but not yet built. also, the area of non-construction land is 26.82 hm2. according to the calculation method of expansion potential, the available land in baoding high-tech development zone is 256.16 hm2. among them, according to the land use planning of high-tech development zone, 47.38 hm2 of land can be allocated, mainly for road, green space, public management, and public service; 208.78 hm2 for compensation, including 90.68 hm2 of land for industrial and mining storage; and 118.10 hm2 of another land can be used for compensation. according to the spatial distribution of the available land, it is mainly located in the third phase of the central area, the west area, and the experimental area, and all the land in the east area has been supplied. the completed industrial and mining storage land area in baoding national high-tech development zone is 427.01 hm2; the current value of the comprehensive plot ratio of industrial land is 0.60, and the ideal value is 0.65. according to the formula, the land area for improving the comprehensive plot ratio of industrial land is 32.85 hm2; the current value of industrial land construction coefficient is 41.88%, the ideal value is 45%, and the area of potential land that can be tapped by increasing the building density of industrial land is 29.61 hm2. the current value of industrial land fixed asset input intensity is 57.3557 million yuan/hm2, the ideal value is 58.5 million yuan/hm2, the area of potential land that can be tapped by increasing the input intensity of industrial land fixed assets is 8.21 hm2; the current value of industrial land output intensity is 12271. 380000 yuan/hm2, the ideal value is 123000000 yuan/hm2, and the area of potential land to increase the output 444 jianjun miao, qian lu intensity of industrial land is 0.99 hm2; the constructed urban construction land in baoding national high-tech development zone is 712.96 hm2, the current value of industrial land use rate is 59.89%, and the ideal value of industrial land use rate is 62%. according to the calculation method of expansion potential, through the adjustment of land structure, the area of industrial and mining storage land can be increased to 15.04 hm2; from 2016 to 2018, the average annual supply land of baoding high-tech zone in the first three years is 28.50 hm2, and the yearly average supply land of industrial and mining storage in the first three years is 7.60 hm2, from which it can be calculated that the number of available land years i is 8.99; and the number of available industrial and mining storage years i is 11.94. within the high-tech zone’s evaluation scope, the maximum area of land supplied in the first five years is 74.28 hm2, and the maximum area of the land provided for industrial and mining storage in the first five years is 16.19 hm2. therefore, it is calculated that the number of years of land available ii is 3.44 years, and the number of years of land available for industrial and mining storage ii is 5.60 years. combined with the above analysis, table 6 was made. table 6. estimation of the number of years available for the expansion potential of baoding national high-tech zone land-use type expansion potentialhm2 average annual supply land area in the first three years//hm2 maximum supply land area in the first five years// hm2 years of land available i//years years of land available ii//years remarks available land 256.16 28.50 74.28 8.99 3.44 including 69.74 hm2 completed rural construction land available for industrial and mining storage land 90.68 7.60 16.19 11.94 5.60 based on the current land resource potential data of baoding national high-tech development zone, the potential expansion calculation, potential strength calculation, potential structural calculation and available land year calculation, and the summary of land-intensive utilization potential were obtained, as shown in table 7. it can be seen from table 7 that the structure potential of the high-tech zone is relatively small, and the area of industrial and mining storage land can be increased by structural adjustment to 15.05hm2, which further proves that the structure of the land in the high-tech zone has been relatively reasonable and conforms to the development orientation of the high-tech site; the area of industrial and mining storage land that can be explored by improving the comprehensive plot ratio of industrial land and the building density of industrial land is 32.86 and 29.62hm2, respectively. it shows that the industrial land intensity in the zone also has an absolute mining potential; the industrial inputoutput power of intensive land use in the high-tech area has a small mining potential. by increasing the input intensity of industrial land fixed assets and industrial land output, the land volume for industrial and mining storage can be increased by 8.21hm2 and 0.99hm2, respectively. according to the calculation formula of expansion potential and baoding high-tech zone’s situation, the available land area of the evaluation scope is 256.16hm2, including 90.68hm2 of the available industrial and mining storage area. according to the calculation of the annual average supply land and the average yearly supply industrial and mining storage area of the previous three years, the actual available land number of baoding high-tech zone is 8.99 years, and the available industrial and mining storage area is determined comprehensively. according to the maximum land supply in the first five years and the maximum land supply for industrial and mining storage, it was calculated that the number of years that the high-tech zone can supply is 3.44 years, and the land supply for industrial and mining storage is 5.60 years. natural quality potential combination of modifiable limiting factors in theory, the combination type of modifiable factors of agricultural land classification unit in baoding national high-tech development zone should be 60 kinds of combination of three factors at different levels. through summary analysis, it is found that there are only 19 types of all factor combination of modifiable factors, among which eight types (3 + 6, 3 + 8, 375, 376, 378, 396, 398, 775) are qualified, and their distribution is shown in figure 3. it can be seen from figure 3 that the distribution of the combinations of the restrictions on the removability of agricultural land in the high-tech zone is relatively scattered. six of the eight combinations belong to the single restriction on the guaranteed rate of agricultural irrigation. one belongs to the single restriction on the content of soil organic matter (red area in the figure), and one belongs to the double restriction factors of “the guaranteed rate of irrigation + the content of soil organic matter.” the double restriction factors are concentrated in baizhigan town and dongcheng town. at the junction of fangzhen, the area is relatively small (purple area in the picture). therefore, it can be concluded that the natural conditions of the cultivated land in the hightech zone are mostly affected by the two limiting factors of irrigation guarantee rate and organic matter content. the irrigation guarantee rate is the leading limiting factor, and the whole area is affected by different degrees, so the land remediation project should focus on it. potential measurement and classification through the overlay analysis tool of arcmap, the overlay comparison analysis between the layers of the modifiable limiting factor unit and the non-limiting factor unit shows that 90% of the cultivated land in baizhigan town, dongchengfang town, shuangta street, yihezhuang are generally full, that is to say, the factor score code is 7. at the same time, the corresponding irrigation assurance rates of the other 11 towns are mostly basic. satisfied, that table 7. summary of land-intensive utilization potential of baoding national high-tech industrial development zone hm2 potential types potential constitution potential scale remarks expansion potential available land area 256.16 69.74 hm2 has been built into rural construction land among them: the area of warehousing land under available working conditions 90.68 structural potential adjustment of land use structure can increase the area of industrial and mining storage land 15.04 strength potential increasing the comprehensive volume ratio of industrial land and exploiting the potential land area 32.85 improving the building coefficient of industrial land and digging the potential land area 29.61 improving the intensity of investment in fixed assets of industrial land and exploiting potential land area 8.21 increasing the output intensity of industrial land and tapping potential land area 0.99 management potential land area used for compensation and expired but not disposed of 0.00 the area of idle land should be recovered 0.00 445an empirical study on the potential of land resources utilization in urban agglomeration development zones based on combination analysis is, the factor score code is 9, according to which, the score of limiting factors of irrigation assurance rate in the above four villages and towns is increased from code 3 to 7. that is, 1 level is increased, and 11 villages and towns such as songlindian are increased from 3 to 9, that is, two levels are increased. the distribution of limiting factors of organic matter is relatively centralized, the area is relatively small, and the organic matter content of the surrounding cultivated land is higher than 20%; that is, 1 level is higher. according to the calculation formula of natural quality potential, calculate the natural equivalent index and promotion potential of cultivated land in the whole area before and after the renovation. table 8 showed the combination type and natural equivalent index of some factors before and after the promotion. in table 8, the average natural equivalency index changed from 13782521 to 1851-2521. and the natural equivalency index increased by 174 on average, that is, 0.87 provincial natural classes. the national natural index increased by 211 on average, that is, 0.53 national natural grades. according to the classification method of natural discontinuities, the results are shown in figure 4. in figure 4, level 1 indicates the natural equal index’s potential range is 0-90, level 2 is 90-275, and level 3 is 275-473. the high-value areas of the natural quality potential of land remediation in the high-tech zone are mainly distributed in the west, northeast, and small in the north and south. the towns involved are dongchengfang town, wharf town, yihezhuang town, dongxianpo town, diaoweo town, linjiatun town, and haoguanzhuang town. in other areas, the promotion potential is relatively low and scattered, so it can be targeted to improve the old infrastructure such as irrigation and drainage, improve farmers’ awareness of green planting, encourage the use of organic fertilizers, and other measures. conclusions in this paper, the potential of land resource utilization in baoding hightech zone is evaluated by the method of the potential of land resource utilization based on the combination empowerment, and the potential of land intensive utilization is evaluated from the five aspects of the potential of expansion, structural potential and intensity potential of land resources in the high-tech zone. at the same time, the nature after renovation is obtained through the combination of the limited factors and the limited factors of land removability in the development zone. the analysis results show that from 2014 to 2018, the sustainable utilization index of land resources in baoding high-tech zone is fluctuating on the whole, but it is basically in the stage of medium sustainable development, with great potential for future improvement. from 0.43 in 2014 to 0.51 in 2018, the overall land utilization is in the stage of medium sustainable utilization. however, land use status has the greatest impact on the level of sustainable land use, followed by social acceptability, resource security and economic feasibility. the main reasons that restrict the improvement of the level of sustainable utilization of land resources are economic feasibility, social acceptability and resource security. these indicators are still at a medium or low level, especially the resource security is at a low level of development. therefore, in order to make a better sustainable use of land resources in the high-tech zone, the high-tech zone should start from the following aspects. (1) strengthen land consolidation and development and improve land use efficiency. although the land use situation of the high-tech zone is good, in order to make more scientific and rational use of land, we must make efforts to improve the basic conditions of agricultural production, increase the intensity of land reclamation and consolidation, increase the effective cultivated land area, and improve the efficiency of land production. (2) improve the resource security of land use and realize the transformation to the sustainable development mode. the evaluation result of land resource security of high-tech zone is still at a low level, which is not consistent with the figure 3. distribution of restrictive factor combination types table 8. combination of restrictive factors and natural index before and after renovation before rectification after renovation promotion factors combination of modifiable constraints natural equivalence index combination of modifiable constraints natural equivalence index 3+6 1577 7+6 1980 irrigation guarantee rate 3+6 1577 7+6 2190 irrigation guarantee rate 3+8 1642 7+8 2050 irrigation guarantee rate 3+8 1642 9+8 2255 irrigation guarantee rate 375 1642 775 1787 irrigation guarantee rate 375 1642 975 2040 irrigation guarantee rate 376 1443 776 1851 irrigation guarantee rate 376 1443 976 2088 irrigation guarantee rate 378 1507 778 1916 irrigation guarantee rate 378 1507 978 2120 irrigation guarantee rate 396 1779 796 2187 irrigation guarantee rate 396 1779 996 2392 irrigation guarantee rate 398 1843 798 2252 irrigation guarantee rate 398 1843 998 2456 irrigation guarantee rate 375 1642 376 1443 irrigation guarantee rate 775 1787 776 1851 irrigation guarantee rate 446 jianjun miao, qian lu level of social and economic development, which requires that in the process of economic development, high-tech zone should pay attention to improving the number of public green space and cultivated land area. (3) according to the regional characteristics, improve the level of economic development and social acceptability. formulate reasonable development strategies, change the industrial structure, increase the proportion of the second and third industries, increase the disposable income of residents, improve the land utilization rate, and promote economic development. at the same time, we should continue to improve the per capita net income of farmers, increase investment in education, improve infrastructure, increase medical and health facilities and the number of doctors, so as to improve social satisfaction. acknowledgments 1.research on measurement and reduction of undesirable output of national urban agglomeration (16bgl210) 2.empirical research on efficiency of urban land utilization base on desirable and undesirable output (15yja790046) references arneth, a., sitch, s., pongratz, j., stocker, b. d., ciais, p;., poulter, b., bayer, a. d., bondeau, a., calle, l., chini, l. p., gasser, t., fader, m., friedlingstein, p., kato, e., li., w., lindeskog, m., nabel, j. e. m. s., pugh, t. a. m., robertson, e., … zaehle, s. (2018). historical carbon dioxide emissions caused by land-use changes are possibly larger than assumed. nature geoscience, 10, 79-84. cao, y., bai, z., wei, z., & zhang, x. (2016). characteristic analysis and pattern evolution on landscape types in typical compound area of mine agriculture urban in shanxi province, china. environmental earth sciences, 75, 585. chen, c., liu, w., jiang, x., & wu, j. (2017). effects of rubber-based agroforestry systems on soil aggregation and associated soil organic carbon: implications for land use. geoderma, 299, 13-24. ding, z. j. (2019). the simulation of urban land use spatial optimization model based on ternary theory. computer simulation, 36, 235-238. dong, y., varquez, a. c.g ., & kanda, m. (2016). global anthropogenic heat flux database with high spatial resolution. atmospheric environment, 150, 276-294. houghton, r. a., & nassikas, a. a. (2017). global and regional fluxes of carbon from land use and land cover change 1850–2015. global biogeochemical cycles, 31, 456-472. li, y. n., wu, j. n., gao, y. h., gao, q., & liu, s. x. (2017). effects of fluorescence characteristics on water extractable organic matter in soda saline-alkali soil under different land use patterns. journal of jilin university (science edition), 55, 733-738. li, z., liu, c., dong, y., chang, x., nie, x., liu, l., xiao, h., lu, y., & zeng, g. (2017). response of soil organic carbon and nitrogen stocks to soil erosion and land use types in the loess hilly–gully region of china. soil & tillage research, 166, 1-9. maranguit, d., guillaume, t., & kuzyakov, y. (2017). land-use change affects phosphorus fractions in highly weathered tropical soils. catena, 149, 385-393. meng, x. l., wu, l. d., & yu, s. b. (2018). research on resource management mechanism of space information network based on virtualization. journal of china academy of electronics and information technology, 13, 47-52. paule-mercado, m. a., lee, b. y., memon, s. a., umer, s. r., salim, i., & lee, c. h. (2017). influence of land development on stormwater runoff from a mixed land use and land cover catchment. science of the total environment, 600, 2142-2155. figure 4. distribution of natural quality potential 1.research 2.empirical 447an empirical study on the potential of land resources utilization in urban agglomeration development zones based on combination analysis price, e. l., perić, m. s., romero, g. q., & kratina, p. (2019). land use alters trophic redundancy and resource flow through stream food webs. journal of animal ecology, 88, 677-689. song, w., & deng, x. (2017). land-use/land-cover change and ecosystem service provision in china. science of the total environment, 576, 705-719. wang, s., & chen, b. (2016). energy–water nexus of urban agglomeration based on multiregional input–output tables and ecological network analysis: a case study of the beijing–tianjin–hebei region. applied energy, 178, 773-783. yang, l., feng, q., yin, z., wen, x., si, j., li, c., & deo, r. c. (2017). identifying separate impacts of climate and land use/cover change on hydrological processes in upper stream of heihe river, northwest china. hydrological processes, 31, 1100-1112. issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n2.94786 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (june, 2022): 183 196 h y d r o l o g y drought monitoring using mowcatl data mining algorithm in aras basin, turkey emre topçu, kafkas university, faculty of engineering and architecture, 36100, kars, turkey emre.topcu@kafkas.edu.tr abstract: drought is a natural phenomenon that occurs frequently and has some adverse effects on the ecosystem and humanity. determination of drought beforehand is vital for optimal management of water resources. many different methods have been developed to detect drought. sequential association analysis is used for the data series analysis containing time information and is one of the methods used to determine the drought. a correlation can be established between the values taken by the data at different times when determining association rules with this method. the primary purpose of this study is to determine the sequential association patterns between precipitation and climate oscillation index for aras basin. the aras basin is a region where irrigation and animal husbandry are common. today, many dams and hydroelectric power plants, together with the increasing population, meet the water and energy needs. a possible drought event in this region will adversely affect the living things in the basin. therefore, the study focused on this basin. finding sequential associations between precipitation and climate oscillation index can determine the temporal correlations between these parameters and specifically detect drought. the mowcatl (minimal occurrences with constraints and time lags) algorithm was used to detect sequential associations, and the j-measure was used to evaluate the patterns in the study. sequential association patterns were determined by applying this method to the precipitation data obtained from 6 meteorology stations in the aras basin. ao (arctic oscillation) index, mei (multivariate enso) index, nao (north atlantic oscillation) index, oceanic niño index (oni), pdo (pacific decadal oscillation) index, pna (pacific/north american), and soi (southern oscillation index), followed by the 1, 3, 6 and 12-month agricultural standardized precipitation index (a-spi) were used in sequential association. the study results revealed that the antecedent parameters were ineffective in detecting arid conditions in ardahan and doğubeyazıt stations, and they were influential on drought conditions, especially in a-spi-3 and a-spi-12 month periods at other stations. although the altitude and geographical features are different, similar climatic patterns have been detected in some stations. as a result, it has been determined that climatic oscillations generally bring about typical situations in terms of drought for the aras basin. keywords: data mining; drought index; drought; oceanic index monitoreo de la sequía a través del algoritmo mowcatl de minería de datos para la cuenca de aras, turquía resumen: la sequía es un fenómeno natural que ocurre muy frecuentemente y que tiene efectos negativos en los ecosistemas y en la humanidad. la definición de la sequía, de antemano, es especialmente necesarua para la administración óptima de los recursos de agua. muchos métodos se han desarrollado para detectar la sequía. uno de estos métodos es el análisis de asociación secuencial que se usa para el análisis de series de datos con información de tiempo. se puede establecer una correlación entre los valores tomados en diferentes períodos cuando se determinan las reglas asociativas con este método. el propósito principal de este estudio es determinar los patrones de asociación secuencial entre precipitación y el índice de oscilación climática para la cuenca de aras, en turquía. esta cuenca es una región donde la irrigación y la agricultura son comunes. al día de hoy, muchas presas e hidroeléctricas, junto con el incremento de la población, demandan estos recursos hidrológicos. un evento de sequía en la región afectaría a estos seres que dependen de la cuenca. por esta razón, el estudio se enfoca en la cuenca de aras. encontrar las asociaciones secuenciales entre precipitación y el índice de oscilación climática puede determinar las correlaciones temporales entre estos parámetros y, específicamente, detectar la sequía. en este trabajo se usó el algoritmo mowcatl (ocurrencias mínimas con restricciones y retrasos, literal del inglés minimal occurrences with constraints and time lags) para detectar las asociaciones secuenciales y la medición j se usó para evaluar estos patrones. los patrones de asociación secuencial se determinaron al aplicar este método a la información de precipitación obtenida de seis estaciones meteorológicas en la cuenca de basin. los índices de oscilación ártica, enso multivariado, oscilación del atlántico norte, oceánico del niño, oscilación decadal del pacífico, pacífico/norte américa, y de oscilación del sur, seguidos por el índice de precipitación agrícola estandarizado (a-spi) en los meses 1, 3, 6 y 12 se utlizaron en la asociación secuencial. los resultados del estudio revelan que estos parámetros no son efectivos para detectar las condiciones áridas en las estaciones de ardahan y doğubeyazıt pero si son efectivos en las condiciones de sequía, especialmente en los períodos a-spi-3 y a-spi-12 en otras estaciones. a pesar de que las condiciones de altitud y geográficas son diferetes, patrones climáticos se han detectado en algunas estaciones. como resultado se determinó que las oscilaciones climáticas generalemente provocan situaciones normales de sequía en la cuenca de aras. palabras clave: mineria de datos; índice de sequía; sequía; índice oceánico. record manuscript received: 30/03/2022 accepted for publication: 14/06/2022 how to cite item: topcu, e. (2022). drought monitoring using mowcatl data mining algorithm in aras basin, turkey. earth sciences research journal, 26(2), 183-196. https://doi.org/10.15446/esrj.v26n2.94786 https://doi.org/10.15446/esrj.v26n2.94786 mailto:emre.topcu@kafkas.edu.tr https://doi.org/10.15446/esrj.v26n2.94786 184 emre topçu introduction drought is a natural phenomenon that causes irreversible damage to the ecosystem with its alarming and negative effects for millions of human, animal and plant species on earth. the geographical conditions, climate and anthropogenic impact of a region contribute to the emergence of drought. understanding the emergence of drought, its duration and when it will end is one of the most important work areas for hydrologists. today, climate change is worsening in many parts of the world (liming & yaonan, 2016). therefore, there is an increasing interest in sustainable water resources management. in fact, the aim here is to prevent future water resources wars that threaten species (rostam et al., 2020). the effects of climate change vary from region to region. it causes differentiation, especially in the local precipitation amount. therefore, it is vital to understand regional dry and rainy periods (meilin et al., 2020). global warming causes negative changes on the water cycle and spatialtemporal characteristics of water resources. this significantly disrupts the balance in water resources and dry and wet conditions, especially in semi-arid and arid regions (yanfeng et al., 2015). drought is responsive to environmental change and altogether affects financial, ecological and farming exercises. exploring the spatial-temporal properties of drought will help specify the correlation of drought with meteorological, hydrological and ecological cycles, and to further assess the effect of climate change on humanity (yudan et al., 2020). although drought is not easy to define precisely, it can simply figure 1. location of the study area table 1. geographical features of the stations station/no elevation (m) latitude longitude period ardahan/17046 1827 41.1061 42.7055 1980-2020 arpaçay/17656 1720 40.8539 43.3178 1980-2020 doğubeyazıt/17720 1640 39.5396 44.018 1980-2020 horasan/17690 1540 40.0383 42.1705 1980-2020 iğdır/17100 856 39.9227 44.0523 1980-2020 kars/17097 1777 40.6042 43.1073 1980-2020 be considered as periodic rainfall and shortage of water resources comparing average conditions. drought can be categorized as meteorological, agricultural, and hydrological. lack of rainfall in a certain period indicates meteorological drought, while a decrease in agricultural production due to insufficient soil moisture is an indicator of agricultural drought, and a decrease in underground and aboveground resources is an indicator of hydrological drought (yürekli et al., 2012). meteorological drought is the primary type, which is the impetus behind other droughts, bringing about a decrease in the normal long term precipitation. with the meteorological drought and its negative effects occur and different kinds of drought will eventually develop, and the combined adverse effects may cause harmful effects in an area (alamdarloo et al., 2020). several indexes are developed to detect meteorological, agricultural, and hydrological drought. these indexes are created using the parameters of the hydrological cycle and are used to minimize the negative effects of drought on water resources. when the index values indicating that this drought will occur, certain restrictions can be made in the use of water for precautionary purposes, general cleaning and irrigation operations can be minimized by the local resident. the drought index most commonly used by hydrologists today is the standardized precipitation index (spi). spi measures the lack of precipitation at various time intervals in regions with different climates (edwards & mckee, 1997). examining drought over a wide span of time according to climatological, 185drought monitoring using mowcatl data mining algorithm in aras basin, turkey atmospheric and oceanic parameters may help moderate future drought impacts on society by improving our comprehension of the drought hazard. the data analysis reveals that climate change scholars have demonstrated that in areas where precipitation dwindles in mid-elevation during summer or where snowfall decreases, the combined impact may bring about significant increases in drought recurrence in the twenty-first century (rind et al., 1990). determining the correlation between drought and different climatic parameters using large scale data sets can be used in drought prediction. it is critical to extract essential information from large databases and infer drought risk. today, it is technical data mining that focuses on such operations (tadesse et al., 2004). data mining utilizes an array of data analysis instruments to determine patterns and relationships between physical factors in various data sets (two crows, 1999). this method is utilized in multidisciplinary fields. for instance, this technique is utilized for business applications by numerous organizations for planning key advantages to increase profit (groth, 1998). data mining algorithm and models, for example, decision trees clustering, associations, categorization, regression, time series forecasting and sequential patterns can possibly determine drought examples and properties. for instance, time series data mining can be applied in determining drought patterns. the minimal occurrences with constraints and time lags (mowcatl) algorithm was used in this study to find the correlations between drought and atmospheric/oceanic indexes by considering time lags of their occurrences (harms et al., 2002; tadesse, 2002). data mining was used to specify the complex interactions including oceanic and atmospheric parameters that potentially lead to droughts over the examined stations in aras basin, turkey. the aras basin is in the first place in turkey, especially in livestock breeding. agriculture, irrigation and animal husbandry activities carried out in the summer months require a large amount of water. for this reason, many large and small dams have been built in the region today. the increasing population in the region has led to an increase in the need for clean water. in recent years, the warm and dry winter months in the region have caused a decrease in water resources. the study of drought in this region has become essential. this is an important reason why this study focuses on this region. materials and methods study area the study was performed in aras river basin, which is one of turkey’s 26 river basins. aras river basin in northeastern turkey is the coldest part because of the altitude. six meteorological stations within the basin with long annual precipitation data were selected. some climatic and meteorological indexes for the years 1980-2020 were examined in the study. as for the study area, summer months are rainy and cool, except for the iğdır station. it can be determined that iğdır station is the driest station with conventional methods. however, drought analysis with climatic oscillations has not been performed before in the region. precipitation data were obtained from the directorate general of meteorology. the characteristic features of the selected stations are shown in table 1. the location of the study basin is shown in figure 1. method the most used atmospheric and oceanic indexes were chosen for this analysis. the arctic oscillation (ao) is a climate index of the atmospheric circulation condition over the arctic. it comprises a positive stage, including below-average geopotential heights, known as negative geopotential height anomalies, and a negative stage wherein the opposite is valid (daculaweather, 2021). mei (multivariate enso index) is determined based on six fundamental observed factors over the tropical pacific. these six factors are sea-level pressure, zonal and meridional components of the surface wind, sea surface temperature, surface air temperature, and total cloudiness fraction of the sky (wolter & timlin, 1993). a positive mei is related to el niño, and negative values show la niña conditions. the nao (north atlantic oscillation) index is specified as the normalized pressure difference between a station on the azores and one in iceland (hurrell, 1995). since it is calculated based on stations toward the north (iceland) and south (azores) of the middle latitude westerly flow, it might be considered as a measure of the strength of these winds. a positive value indicate a strong mid-latitude westerly flow while the negative values show the weak mid-latitude westerly flow (hurrell, 1995). the oceanic niño index (oni) is the essential indicator for observing el niño and la niña, which are opposite phrases of the climate pattern called the el niño-southern oscillation, or shortly “enso”. el niño conditions to be available when the oceanic niño index is +0.5 or higher, demonstrating the east-central tropical pacific is significantly warmer than expected. la niña conditions exist when the oceanic niño index is 0.5 or lower revealing that the area is cooler than expected. (national oceanic and atmospheric administration, 2021). the pdo (pacific decadal oscillation) index is referred to as the main component of north pacific monthly sea surface temperature variability poleward of 20n (francis & hare, 1994). the positive values indicate the warm period of the north pacific sea surface temperature, and the negative ones show the cold phase. the pna (pacific/north american) index indicates the upper atmosphere conditions (overland et al. 2002). monthly pna averages are utilized in this analysis. the soi (southern oscillation index) is calculated through monthly sea level pressure average anomalies at tahiti (t), french polynesia, and darwin (d), australia. the normalized monthly average ocean level pressure anomaly soi [t-d] is an index that combines the southern oscillation into one sequence (trenberth & hoar, 1996). a positive soi implies la niña and a negative soi demonstrates el niño conditions. this study aims to present that inconstancy of the global atmospheric, and oceanic indexes can be utilized as an antecedent to local drought by formulation association rules regarding the drought index. agricultural standardized precipitation index (a-spi) the spi was formulated by mckee et al. (1993) for the classification and observation of drought. it is a comfortable, adaptable, global index, requiring just the precipitation data, which are the primary reasons behind its wide adaptability in different drought-related applications around the world. despite the fact that it is basically considered as a meteorological drought index, it is also utilized in numerous examinations on hydrological and agricultural drought. the lack of a soil-water balance part in spi is a restricting element for its utilization in vegetation-oriented applications. however, such a component is reasonably consolidated in the effective precipitation parameter. accordingly, the replacement of the total precipitation with the effective precipitation prompts a more solid definition of spi for agricultural drought and increases its adaptability for analyzing drought impacts on vegetation. the effective precipitation can be evaluated through an array of methods, for example, assessing the soil moisture changes with lysimeters or through the simulation of the water dynamics (water balance models) (bos et al., 2008; ebrahimpour et al., 2015; feddes et al., 1988; patwardhan et al., 1990). in spite of their precision, such techniques primarily used in local or experimental applications since they require many input and estimation installations, not generally accessible for operational use or in long time series, as needed for drought characterization. moreover, the effective precipitation assessment through the above mentioned techniques, in light of on the precipitation data only, offers the primary benefit of the original index permits to use even in regions with restricted information access. the most suitable strategy for effective precipitation assessment in the proposed updated index, specifically the agricultural standardized precipitation index (a-spi), is based on the area and the analyzed framework. the a-spi holds a similar standardization method to spi, through the fitting of the effective precipitation time series to an appropriate statistical distribution, which is then transformed into normal distribution. this study applied a basic empirical methodology utilized by the fao (food and agriculture organization of the united nations), which can be used in areas with a maximum slope of 4–5% (tigkas et al., 2019). the a-spi was analyzed in 1, 3, 6 and 12-month periods. the minimal occurrences with constraints and time lags (mowcatl) algorithm mowcatl algorithm was utilized to reveals the correlations between sequences in the multiple data sets, where a lag in time between the antecedent and the consequent exists. besides the conventional frequency and support limitations in sequential data mining, mowcatl utilizes separate antecedent and consequent inclusion constraints, separate antecedent and consequent maximum window widths, to determine the antecedent and consequent patterns that are isolated by a time lag (harms et al., 2002). an association 186 emre topçu rule is characterized as “if x then y,” where x is the rule antecedent and y is its consequent. the support of the antecedent, signified as support {x}, is the number of events that every occasions in the antecedent episode x happen together within a user-defined window width. the antecedent support (support {x}) is likewise called the rule coverage. an episode is regarded as frequent if its support meets or surpasses a user-identified threshold. the frequency and support of a consequent episode are also defined in this manner. the mowcatl algorithm primarily scans the data file in a database storing the occasions of the single events for the antecedent and consequent events individually. the algorithm only seeks for occurrences of events that meet the inclusion limitations. then it excludes the episodes that do not satisfy the user-specified support threshold. the single-event episodes are paired into episodes of two events, for pairs of events that happen together within the specified window width, and event records of the episode in the dataset. if there are no more events to pair, this process is completed. the frequent episodes are specified with the antecedent and the consequent independently, then mowcatl combines the frequent episodes to establish an episode rule (harms et al., 2002). the algorithm establishes episodic rules where the antecedent episode happens within a prescribed window width, the consequent episode occurs within a specified window width, and the start of the consequent follows an antecedent within a user-specified time lag. for instance, let’s assume that episode x is the events a and b, and episode y is the events c and d, the prescribed antecedent window width is 2 months, and the consequent window width is 3 months, with a time lag of 2 months. the parallel rule formed would indicate that if a and b occur within 2 months, then within 3 months c and d will occur together within 2 months (tadesse et al., 2005). in drought research, several climatic and oceanic datasets might be required to identify drought and associations of drought with specific values in these datasets. for example, the oceanic parameters referred to as the ao, mei, nao, oni, pdo, pna, soi and several oceanic indexes can be generated in a database so that the algorithm might form the associations with the drought index to monitor it. mowcatl calculation is utilized in the current examination for extricating rules between extraordinary scenes and climatic files, since this calculation can be utilized for various groupings and furthermore, this will catch without anyone else the slack between the events of climatic records and precipitation occasions. the data-mining context necessitates selecting the best rule in a certain model depending on the “interestingness measure.” the concepts of “interestingness” or “goodness” of the rules are defined to select and compare the better rules among the generated ones (bayardo & agrawal, 1999; silberschatz & tuzhilin, 1995). despite the support and confidence values’ significance in selecting rules, there are alternative methods and algorithms to use in quantifying interestingness measures (padmanabhan & tuzhilin, 1999; shokoohi-yekta et al., 2015). these methods include smyth and goodman’s j-measure (smyth and goodman, 1992) used in this study to quantify the interestingness of the rules for the mowcatl algorithm. smyth and goodman’s j-measure is specified as the average information content of a probabilistic classification rule and applied to identify the best rules relating discrete-valued attributes. a probabilistic classification rule is a logical implication of “if x then y” with some probability p (hilderman & hamilton, 1999). the j value interval is between 0 to 1. higher j values indicate better measurements. mowcatl is covered in detailed in (harms & deogun, 2004). experiment the drought was monitored as the time-series data of meteorological, climatological, and oceanic parameters generated rules that could detect the probable drought occurrence. the justification of oceanic parameter measurements to monitor drought relies on the general assumption that ocean-atmosphere relationships affect drought. as the fluctuations in oceanic parameters develop more slowly than surface meteorological parameters, the trend of the oceanic parameters better are more observable than the surface parameter trends such as precipitation deficits and surface temperature changes (mcpadeh, 1998). therefore, oceanic parameters might be regarded as antecedents and droughts as consequents in finding their correlation (tadesse et al., 2004). the atmospheric and oceanic indexes applied in this paper include ao, nao, oni, mei, pna, pda, soi. a-spi values for each station were also utilized to generate the rules. the monthly values from the ocean and a-spi were converted into discrete representations and classified into seven categories. assuming a normal distribution of this data for 40 years, each oceanic and atmospheric parameter value is divided into 0.5, 1, and 1.5 standard deviations from the normal frequency distribution. the drought index categories are: extremely dry (1), severely dry (2), moderately dry (3), normal (4), moderately wet (5), severely wet (6), and extremely wet (7). the drought classification thresholds of indexes are detailed in table 2. table 2. threshold values used to classify oceanic and climatic indexes category indexes 1 2 3 4 5 6 7 a-spi ≤ -2 -2<x≤-1.5 -1.5<x≤-1 -1<x≤1 1<x≤1.5 1.5<x≤2 > 2 ao ≤ -1.9 -1.9<x≤-0.8 -0.8<x≤-0.1 -0.1<x≤0.5 0.5<x≤1.2 1.2<x≤2.6 > 2.6 mei ≤ -1.4 -1.4<x≤-0.7 -0.7<x≤-0.2 -0.2<x≤0.3 0.3<x≤0.9 0.9<x≤2.1 > 2.1 nao ≤ -1.9 -1.9<x≤-1.1 -1.1<x≤-0.2 -0.2<x≤0.3 0.3<x≤1 1<x≤1.7 > 1.7 oni ≤ -1.4 -1.4<x≤-0.5 -0.5<x≤0.2 0.2<x≤0.7 0.7<x≤1.3 1.3<x≤2.3 > 2.3 pdo ≤ -1.7 -1.7<x≤-1 -1<x≤-0.2 -0.2<x≤0.4 0.4<x≤0.9 0.9<x≤1.6 > 1.6 pna ≤ -2.2 -2.2<x≤-1.3 -1.3<x≤-0.5 -0.5<x≤0.1 0.1<x≤0.6 0.6<x≤1.4 > 1.4 soi ≤ -1.9 -1.9<x≤-0.8 -0.8<x≤-0.2 -0.2<x≤0.3 0.3<x≤1 1<x≤1.8 > 1.8 discussion the rules in this study with a confidence level above 0.90 are presented in the tables. since drought and flood are so sporadic, the j values are so small that all values greater than 0.025 are to be included. mowcatl algorithm is utilized in the study for extracting rules between extreme episodes and climatic indexes. this algorithm’s applicability to multiple sequences and its ability to capture the lag between the occurrences of climatic indexes and rainfall events by itself were the main reasons behind this preference. as presented in table 3, the established rules determined neither dry nor wet conditions, but normal conditions, at all stations for 1 month a-spi (a-spi1) values. arid conditions could not be detected in a-spi1 consequent, as rules with lower levels of confidence were not included. a high level of confidence increases the validity of the rules. table 3 shows that ao, mei, pda and pna are more related in forming normal conditions. the other tables (table 4, 5, 6, 7 and 8) shows that the standard situation, which is generally in the normal drought category, is observed. 187drought monitoring using mowcatl data mining algorithm in aras basin, turkey table 3. sample rules generated for a-spi1 using mowcatl algorithm for serial episodes, antecedent and consequent windows of 2-months, and the 3-months maximum time lag between the start of the antecedent and the start of the consequent a-spi1 antecedent consequent confidence j measure ardahan nao(5)-pna(3) spi1(4) 0.962 0.037 mei(3)-pdo(5) spi1(4) 0.941 0.037 nao(5)-oni(2) spi1(4) 0.907 0.036 arpaçay mei(3)-pna(3) spi1(4) 0.938 0.042 pdo(3)-pna(3) spi1(4) 0.920 0.033 ao(5)-pdo(3) spi1(4) 0.917 0.038 kars mei(2)-nao(2) spi1(4) 0.958 0.034 mei(3)-pna(6) spi1(4) 0.933 0.031 oni(2)-pna(6) spi1(4) 0.913 0.036 mei(3)-oni(2) spi1(4) 0.906 0.068 oni(2)-soi(4) spi1(4) 0.903 0.042 horasan oni(3)-pna(3) spi1(4) 0.923 0.036 pdo(3)-pna(3) spi1(4) 0.920 0.033 doğubeyazit oni(3)-pna(3) spi1(4) 0.9231 0.0363 pdo(3)-pna(3) spi1(4) 0.9200 0.0332 iğdir pdo(3)-soi(6) spi1(4) 0.9500 0.0546 oni(1)-soi(6) spi1(4) 0.9211 0.0358 as presented in table 4, it is determined that for a-spi3, only normal conditions were found in all stations as a result of various algorithm rules. table 4. sample rules generated for a-spi3 using mowcatl algorithm for serial episodes, antecedent and consequent windows of 2-months, and the 3 months maximum time lag between the start of the antecedent and the start of the consequent a-spi3 antecedent consequent confidence j measure ardahan pdo(3)-soi(6) spi3(4) 0.950 0.055 oni(1)-soi(6) spi3(4) 0.921 0.036 arpaçay oni(2)-pdo(2) spi3(4) 0.962 0.090 mei(2)-nao(3) spi3(4) 0.957 0.075 soi(5)-pdo(2) spi3(4) 0.955 0.035 pna(6)-pdo(2) spi3(4) 0.950 0.030 mei(2)-pna(6) spi3(4) 0.938 0.042 mei(2)-pdo(2) spi3(4) 0.926 0.061 mei(3)-pna(3) spi3(4) 0.917 0.048 soi(6)-pna(5) spi3(4) 0.912 0.032 nao(3)-oni(2) spi3(4) 0.905 0.065 kars pdo(2)-pna(6) spi3(4) 0.950 0.040 oni(1)-pdo(3) spi3(4) 0.944 0.034 soi(4)-pdo(6) spi3(4) 0.912 0.046 horasan mei(4)-nao(6) spi3(4) 0.964 0.052 nao(6)-oni(4) spi3(4) 0.964 0.052 ao(2)-soi(2) spi3(4) 0.929 0.035 oni(4)-pna(5) spi3(4) 0.921 0.043 mei(4)-pna(5) spi3(4) 0.913 0.048 188 emre topçu a-spi3 antecedent consequent confidence j measure doğubeyazit nao(6)-mei(4) spi3(4) 0.964 0.052 nao(6)-oni(4) spi3(4) 0.964 0.052 ao(2)-soi(2) spi3(4) 0.929 0.035 oni(4)-pna(5) spi3(4) 0.921 0.043 pna(5)-mei(4) spi3(4) 0.913 0.048 iğdir nao(2)-soi(3) spi3(4) 0.962 0.048 nao(2)-pdo(2) spi3(4) 0.955 0.038 nao(2)-mei(5) spi3(4) 0.950 0.033 pna(3)-mei(4) spi3(4) 0.941 0.050 mei(4)-nao(3) spi3(4) 0.926 0.067 nao(3)-pdo(3) spi3(4) 0.912 0.036 mei(5)-oni(3) spi3(4) 0.906 0.032 oni(3)-pna(3) spi3(4) 0.904 0.050 looking at the rule of the antecedents and consequents in table 5 and table 6, ao, nao, mei and oni were more effective in the occurrence of normal situations for a-spi6. table 5. sample rules generated for a-spi6 using mowcatl algorithm for serial episodes, antecedent and consequent windows of 2-months, and the 3 months maximum time lag between the start of the antecedent and the start of the consequent a-spi6 antecedent consequent confidence j measure ardahan nao(2)-soi(3) spi6(4) 0.9615 0.0484 nao(2)-pdo(2) spi6(4) 0.9545 0.0379 mei(5)-nao(2) spi6(4) 0.9500 0.0328 mei(4)-pna(3) spi6(4) 0.9412 0.0505 mei(4)-nao(3) spi6(4) 0.9259 0.0673 nao(3)-pdo(3) spi6(4) 0.9118 0.0359 mei(5)-oni(3) spi6(4) 0.9063 0.0316 oni(3)-pna(3) spi6(4) 0.9038 0.0498 arpaçay ao(2)-mei(2) spi6(4) 0.9583 0.0448 ao(5)-pdo(2) spi6(4) 0.9545 0.0394 ao(5)-soi(6) spi6(4) 0.9333 0.0426 mei(2)-pdo(1) spi6(4) 0.9231 0.0329 nao(2)-ao(2) spi6(4) 0.9211 0.0470 mei(2)-nao(4) spi6(4) 0.9167 0.0423 nao(3)-pdo(5) spi6(4) 0.9118 0.0377 nao(5)-mei(2) spi6(4) 0.9091 0.0473 kars mei(7)-pdo(6) spi6(4) 0.9545 0.0551 mei(7)-soi(2) spi6(4) 0.9444 0.0413 oni(1)-soi(7) spi6(4) 0.9444 0.0413 ao(6)-pdo(3) spi6(4) 0.9375 0.0345 mei(1)-nao(3) spi6(4) 0.9375 0.0345 oni(1)-pdo(1) spi6(4) 0.9375 0.0345 pdo(2)-soi(4) spi6(4) 0.9091 0.0368 pna(6)-mei(2) spi6(4) 0.9063 0.0521 189drought monitoring using mowcatl data mining algorithm in aras basin, turkey table 6. (continue) a-spi6 antecedent consequent confidence j measure horasan oni(2)-pdo(6) spi6(4) 0.9643 0.0551 oni(5)-soi(4) spi6(4) 0.9583 0.0442 mei(7)-pdo(6) spi6(4) 0.9545 0.0389 pdo(7)-soi(2) spi6(4) 0.9545 0.0389 ao(4)-pna(2) spi6(4) 0.9500 0.0336 nao(5)-pdo(6) spi6(4) 0.9375 0.0469 mei(4)-nao(6) spi6(4) 0.9286 0.0371 nao(6)-oni(4) spi6(4) 0.9286 0.0371 ao(2)-soi(3) spi6(4) 0.9063 0.0327 doğubeyazit nao(4)-soi(5) spi6(4) 0.9583 0.0515 nao(5)-oni(6) spi6(4) 0.9500 0.0395 oni(2)-pna(7) spi6(4) 0.9444 0.0336 mei(3)-pna(2) spi6(4) 0.9231 0.0392 iğdir pdo(2)-soi(3) spi6(4) 0.9444 0.0316 pna(2)-soi(4) spi6(4) 0.9444 0.0316 oni(2)-soi(3) spi6(4) 0.9286 0.0417 ao(5)-soi(5) spi6(4) 0.9250 0.0573 ao(5)-pna(6) spi6(4) 0.9231 0.0365 mei(3)-pna(3) spi6(4) 0.9167 0.0629 as presented in table 7 and table 8, normal conditions were detected for a-spi12. table 7. sample rules generated for a-spi12 using mowcatl algorithm a-spi12 antecedent consequent confidence j measure ardahan mei(6)-pdo(6) spi12(4) 0.974 0.122 oni(1)-pdo(2) spi12(4) 0.958 0.068 oni(1)-soi(6) spi12(4) 0.947 0.098 oni(3)-pdo(2) spi12(4) 0.941 0.083 mei(5)-oni(3) spi12(4) 0.938 0.076 nao(4)-pdo(2) spi12(4) 0.938 0.038 mei(2)-nao(5) spi12(4) 0.932 0.100 ao(6)-pdo(2) spi12(4) 0.929 0.031 ao(6)-soi(6) spi12(4) 0.929 0.031 pdo(2)-soi(4) spi12(4) 0.909 0.041 arpaçay pdo(3)-oni(3) spi12(4) 0.981 0.126 soi(4)-pdo(3) spi12(4) 0.978 0.108 mei(4)-nao(6) spi12(4) 0.964 0.056 pna(2)-pdo(4) spi12(4) 0.958 0.045 ao(5)-pdo(2) spi12(4) 0.955 0.040 190 emre topçu a-spi12 antecedent consequent confidence j measure arpaçay nao(6)-soi(4) spi12(4) 0.933 0.043 nao(2)-soi(3) spi12(4) 0.923 0.033 ao(5)-pna(5) spi12(4) 0.921 0.048 mei(2)-nao(4) spi12(4) 0.917 0.043 oni(3)-soi(3) spi12(4) 0.912 0.077 pna(3)-soi(4) spi12(4) 0.909 0.048 kars mei(2)-oni(1) spi12(4) 0.974 0.120 soi(6)-oni(1) spi12(4) 0.974 0.120 oni(1)-mei(1) spi12(4) 0.964 0.082 mei(6)-nao(5) spi12(4) 0.962 0.074 nao(2)-pdo(2) spi12(4) 0.955 0.059 ao(6)-pna(2) spi12(4) 0.944 0.044 ao(4)-oni(1) spi12(4) 0.944 0.044 mei(1)-pdo(2) spi12(4) 0.944 0.044 soi(7)-oni(1) spi12(4) 0.944 0.044 ao(6)-pdo(3) spi12(4) 0.938 0.037 mei(1)-nao(3) spi12(4) 0.938 0.037 oni(1)-pna(4) spi12(4) 0.938 0.037 table 8. (continue) spi12 antecedent consequent confidence j measure horasan mei(2)-pdo(1) spi12(4) 0.962 0.056 oni(1)-pna(5) spi12(4) 0.955 0.044 oni(6)-pna(5) spi12(4) 0.944 0.032 oni(6)-soi(2) spi12(4) 0.941 0.059 nao(6)-pna(6) spi12(4) 0.929 0.043 nao(4)-soi(4) spi12(4) 0.923 0.075 mei(6)-ao(4) spi12(4) 0.921 0.054 ao(6)-nao(6) spi12(4) 0.917 0.032 mei(6)-oni(6) spi12(4) 0.913 0.060 mei(6)-soi(2) spi12(4) 0.903 0.072 doğubeyazit mei(5)-pdo(3) spi12(4) 0.947 0.077 oni(2)-pna(7) spi12(4) 0.944 0.036 ao(5)-pna(6) spi12(4) 0.923 0.042 mei(5)-pna(6) spi12(4) 0.905 0.056 oni(3)-pdo(3) spi12(4) 0.904 0.069 mei(5)-nao(6) spi12(4) 0.900 0.038 iğdir mei(4)-nao(6) spi12(4) 0.964 0.063 pna(5)-soi(4) spi12(4) 0.947 0.072 nao(6)-soi(5) spi12(4) 0.944 0.033 ao(5)-mei(4) spi12(4) 0.933 0.049 mei(4)-pdo(2) spi12(4) 0.923 0.039 ao(4)-nao(6) spi12(4) 0.917 0.033 mei(5)-oni(3) spi12(4) 0.906 0.040 191drought monitoring using mowcatl data mining algorithm in aras basin, turkey according to table 9, only when the mei (7) and nao (6) conditions occur together at the kars station, mildly severe drought may occur after 2 or 3 months. the confidence level for this rule is 0.83. this value is very close to 0.90. both severe wet (category 6) and slightly arid (category 3) conditions could be determined for a-spi6. a-spi6 (6) result was obtained for ao (1) -mei (6) and ao (1) -oni (6) conditions, and a-spi6 (3) result was obtained for ao (6) -oni (7) conditions at arpaçay station. a-spi6 (3) result was found for mei (4) -nao (1) conditions at the kars station. slightly wet conditions for a-spi12 were detected at arpaçay station. ao, nao, oni, pda and pna were effective for the detection of this situation. both severe wet, slightly wet and slightly arid conditions have been identified for horasan station. while ao, oni, mei and pdo were effective in determining wet conditions, mei and pdo were effective for dry conditions. similar situations are also valid for doğubeyazıt station. based on the a-spi results, among the stations, the station experiencing the most severe drought was identified as iğdır station. the formulated algorithm confirmed that the iğdır station is dry according to classical methods. severe drought has been observed under the influence of nao, mei, pdo and pna. especially in the seventh category of these oscillations, a severe drought occurred for a-spi12 (category 2). it can be concluded that not all climatic oscillations cause drought or marshiness in all conditions based on the above tables. drought or wet conditions can be determined only when the rules specified in the tables occur. similar rules occur in different meteorological stations are listed in table 10. according to table 10, the least common number of rules is seen at arpacay-kars supports (4 rules), while the highest number of similar rules is seen at horasan-doğubeyazıt stations (22 rules). what is interesting here is that while the arpaçay and kars stations are close to each other, a low number of similar rules are obtained, while the horasan and doğubeyazıt stations are far from each other, but a large number of similar rules are obtained. these stations are not very different in altitude, but may have different or similar precipitation regimes due to the surrounding mountains. arpaçay, horasan and kars stations are relatively flat areas, but there are very high mountains near doğubeyazıt station and these mountains shape the climate of the region where they are located. table 9. association rules obtained excluding normal conditions station antecedent consequent confidence j measure arpaçay ao(1)-mei(6) spi12(5) 0.8333 0.0746 ao(1)-oni(6) spi12(5) 0.8333 0.0746 mei(6)-ao(1) spi12(5) 0.8333 0.0746 nao(7)-pdo(2) spi12(5) 0.8333 0.0746 oni(6)-ao(1) spi12(5) 0.8333 0.0746 pdo(2)-nao(7) spi12(5) 0.8333 0.0746 pdo(1)-pna(6) spi12(5) 0.8333 0.0746 pna(6)-pdo(1) spi12(5) 0.8333 0.0746 ao(1)-mei(6) spi6(6) 0.8333 0.0357 ao(1)-oni(6) spi6(6) 0.8333 0.0357 ao(6)-oni(7) spi6(3) 0.8333 0.0899 mei(6)-ao(1) spi6(6) 0.8333 0.0357 oni(6)-ao(1) spi6(6) 0.8333 0.0357 oni(7)-ao(6) spi6(3) 0.8333 0.0899 kars mei(4)-nao(1) spi6(3) 0.8333 0.0816 nao(1)-mei(4) spi6(3) 0.8333 0.0816 mei(7)-nao(6) spi3(3) 0.8333 0.0922 nao(6)-mei(7) spi3(3) 0.8333 0.0922 horasan ao(3)-oni(1) spi12(5) 0.8000 0.0348 oni(1)-ao(3) spi12(5) 0.8000 0.0348 mei(1)-pdo(4) spi12(5) 0.8333 0.0497 mei(2)-pdo(6) spi12(3) 0.8333 0.0797 pdo(4)-mei(1) spi12(6) 0.8333 0.0497 pdo(6)-mei(2) spi12(3) 0.8333 0.0797 iğdir mei(1)-pdo(4) spi12(6) 0.8333 0.0479 mei(7)-pna(7) spi12(2) 0.8333 0.0826 mei(7)-pna(5) spi12(2) 0.8333 0.0326 pdo(4)-mei(1) spi12(6) 0.8333 0.0379 pdo(7)-nao(7) spi12(2) 0.8333 0.0426 pna(5)-mei(7) spi12(2) 0.8333 0.0526 192 emre topçu table 10. legend list for table 11 and table 12 color rules combination number of same rules arpaçay-horasan-doğubeyazıt 9 ardahan-arpaçay 6 ardahan-kars 8 arpaçay-kars 4 arpaçay-iğdır 6 horasan-doğubeyazıt 22 table 11 and table 12 show the frequency of occurrence of the same rule at different stations. table 11. similar rules table for aras basin stations coeff. rules ardahan arpaçay kars horasan doğubeyazıt iğdır spi1 >0.90 oni(3)-pna(3) = spi1(4) + + >0.90 pdo(3)-pna(3) = spi1(4) + + + >0.80 pna(3)-oni(3) = spi1(4) + + >0.80 pna(3)-pdo(3) = spi1(4) + + + spi3 >0.75 pdo(2)-oni(2) = spi3(4) + + >0.80 ao(2)-nao(3) = spi3(4) + + >0.90 ao(2)-soi(2) = spi3(4) + + >0.80 ao(5)-pdo(4) = spi3(4) + + >0.75 mei(3)-pna(3) = spi3(4) + + >0.80 mei(4)-nao(6) = spi3(4) + + >0.80 mei(4)-pna(5) = spi3(4) + + >0.80 nao(3)-ao(2) = spi3(4) + + >0.75 nao(6)-mei(4) = spi3(4) + + >0.90 nao(6)-oni(4) = spi3(4) + + >0.80 oni(2)-pdo(2) = spi3(4) + + >0.80 oni(4)-nao(6) = spi3(4) + + >0.90 oni(4)-pna(5) = spi3(4) + + >0.75 pdo(2)-pna(6) = spi3(4) + + >0.75 pdo(4)-ao(5) = spi3(4) + + >0.80 pna(3)-mei(3) = spi3(4) + + >0.80 pna(5)-mei(4) = spi3(4) + + >0.80 pna(5)-oni(4) = spi3(4) + + >0.75 pna(6)-pdo(2) = spi3(4) + + >0.80 soi(2)-ao(2) = spi3(4) + + table 12. (continue) coeff. rules ardahan arpaçay kars horasan doğubeyazıt iğdır spi6 >0.80 nao(2)-pdo(2) = spi6(4) + + >0.80 nao(2)-soi(3) = spi6(4) + + >0.75 nao(3)-oni(3) = spi6(4) + + + >0.80 oni(2)-pdo(6) = spi6(4) + + >0.80 oni(3)-nao(3) = spi6(4) + + + >0.75 pdo(2)-nao(2) = spi6(4) + + >0.80 pdo(6)-oni(2) = spi6(4) + + >0.80 soi(3)-nao(2) = spi6(4) + + 193drought monitoring using mowcatl data mining algorithm in aras basin, turkey coeff. rules ardahan arpaçay kars horasan doğubeyazıt iğdır spi12 >0.75 mei(6)-nao(5) = spi12(4) + + >0.80 mei(5)-oni(3) = spi12(4) + + + >0.80 mei(2)-pdo(2) = spi12(4) + + + >0.80 ao(5)-oni(3) = spi12(4) + + >0.80 ao(5)-pna(6) = spi12(4) + + >0.75 ao(6)-pdo(2) = spi12(4) + + >0.80 mei(2)-pna(5) = spi12(4) + + >0.80 mei(3)-soi(4) = spi12(4) + + + >0.75 mei(4)-nao(6) = spi12(4) + + >0.80 mei(5)-soi(2) = spi12(4) + + >0.75 mei(6)-pdo(6) = spi12(4) + + >0.80 nao(2)-soi(3) = spi12(4) + + >0.80 nao(3)-oni(3) = spi12(4) + + + >0.80 nao(5)-mei(6) = spi12(4) + + >0.80 nao(6)-mei(4) = spi12(4) + + >0.80 nao(6)-soi(4) = spi12(4) + + >0.75 oni(1)-pna(4) = spi12(4) + + >0.80 pdo(6)-pna(4) = spi12(4) + + + >0.75 pna(5)-mei(2) = spi12(4) + + >0.75 pna(6)-ao(5) = spi12(4) + + >0.80 soi(2)-mei(5) = spi12(4) + + >0.75 soi(6)-pdo(2) = spi12(4) + + >0.80 soi(4)-nao(6) = spi12(4) + + support and lift values for mowcatl algorithm for a-spi12 (category 4) are seen in table 13. the support of the rule is the number of times the rule holds in the database (harms and deogun 2004). the lift was obtained from the confidence and the occurrence probability of the consequent. table 14 shows the correlation values between some oscillation indexes and meteorological index in series for many years. therefore, it can be understood whether there is a correlation between which meteorological series and climatic oscillation values at which station. for example, at arpaçay station, pdo is the most relevant climatic oscillation, followed by mei, oni and soi oscillations, respectively. mei oscillation is apparent in iğdır, which is the driest station. table 13. support and lift values for mowcatl algorithm a-spi(4) rules support lift rules support lift a r d a h a n mei(6)-pdo(6) 0.026 1.029 h o r a sa n mei(2)-pdo(1) 0.019 1.032 oni(1)-pdo(2) 0.017 1.086 oni(1)-pna(5) 0.021 1.355 oni(1)-soi(6) 0.030 1.200 oni(6)-pna(5) 0.015 1.160 oni(3)-pdo(2) 0.023 1.054 oni(6)-soi(2) 0.026 1.052 mei(5)-oni(3) 0.023 1.054 nao(6)-pna(6) 0.021 1.065 nao(4)-pdo(2) 0.011 1.018 nao(4)-soi(4) 0.038 1.032 mei(2)-nao(5) 0.036 1.204 mei(6)-ao(4) 0.028 1.020 ao(6)-pdo(2) 0.011 1.018 ao(6)-nao(6) 0.019 1.032 ao(6)-soi(6) 0.011 1.164 mei(6)-oni(6) 0.034 1.037 pdo(2)-soi(4) 0.019 1.047 mei(6)-soi(2) 0.045 1.010 194 emre topçu a-spi(4) rules support lift rules support lift a r pa ç a y pdo(3)-oni(3) 0.041 1.027 d o ğ u b e y a z it mei(5)-pdo(3) 0.030 1.066 soi(4)-pdo(3) 0.038 1.011 oni(2)-pna(7) 0.017 1.108 mei(4)-nao(6) 0.028 1.265 ao(5)-pna(6) 0.021 1.088 pna(2)-pdo(4) 0.019 1.095 mei(5)-pna(6) 0.036 1.177 ao(5)-pdo(2) 0.019 1.194 oni(3)-pdo(3) 0.038 1.015 nao(6)-soi(4) 0.026 1.031 mei(5)-nao(6) 0.026 1.075 nao(2)-soi(3) 0.023 1.071 iğ d ir mei(4)-nao(6) 0.028 1.292 ao(5)-pna(5) 0.030 1.076 pna(5)-soi(4) 0.028 1.020 mei(2)-nao(4) 0.030 1.076 nao(6)-soi(5) 0.015 1.160 oni(3)-soi(3) 0.051 1.001 ao(5)-mei(4) 0.028 1.211 pna(3)-soi(4) 0.034 1.062 mei(4)-pdo(2) 0.021 1.065 k a r s mei(2)-oni(1) 0.032 1.282 ao(4)-nao(6) 0.019 1.032 soi(6)-oni(1) 0.030 1.196 mei(5)-oni(3) 0.026 1.052 oni(1)-mei(1) 0.019 1.044 mei(6)-nao(5) 0.019 1.124 nao(2)-pdo(2) 0.015 1.033 ao(6)-pna(2) 0.013 1.083 ao(4)-oni(1) 0.013 1.083 mei(1)-pdo(2) 0.013 1.083 soi(7)-oni(1) 0.013 1.083 ao(6)-pdo(3) 0.013 1.083 mei(1)-nao(3) 0.011 1.015 oni(1)-pna(4) 0.011 1.015 drought analysis studies using data mining method are becoming widespread today. since several complex factors are involved in the emergence of drought, methods and indexes with many or few parameters are being developed. sequential association analysis can be considered as an alternative to these conventional methods. the impact of climatic oscillation indexes on the meteorological index was investigated using many parameters in this study. a high level of confidence has resulted in detecting fewer drought and wetness situations. dry or wet conditions could not be revealed in ardahan and doğubeyazıt stations. in the provinces of ardahan and doğubeyazıt, dry periods do not occur due to the heavy rainfall in the past. for this reason, it has remained unanswered that drought does not occur at certain times and how many months later the wet time will come. due to the higher altitudes of ardahan and doğubeyazıt compared to other areas in the aras region, it receives a lot of snowfall, so the regions are wet in many times of the year. also in terms of analysis, this may be due to the high confidence level and the threshold values preferred in data discretization. however, climatic oscillations, which are frequently effective in determining climatic features at other stations, were revealed. conclusions drought can occur anywhere in the world from time to time. in this study, the mowcatl algorithm developed by harms et al., was used as an alternative to conventional drought methods for meteorological stations in the aras basin. since it is difficult to fully analyze the drought due to the factors that constitute it, the algorithm used has been able to advance the study to a certain extent. the formulated algorithm confirmed that the iğdır station is dry according to classical methods. however, the method applied in this study, more comprehensive studies can be performed by changing the antecedents and consequents. the arid and wetland situations that the data series have stored can be further illustrated. the result of the study also showed that stations with different geographical features may have similar climatic rules, while stations with similar geographical features may have different climatic features. this may be due to the partially unpredictable climatic parameters. this study conducted in the aras basin examined the effects of parameters such as climatic oscillations on the precipitation regime that causes drought. in the future, it is foreseen that the development of the region and the awareness of the local people will be engaged in more agriculture, and this type of work is important in turkey regarding the climatic situation of the region. it is inevitable that the algorithm used in this study will contribute to the region and will shed light on the studies to be done with other methods and algorithms. this study shall contribute to the scholarship on the climate regarding the study area and its surroundings. 195drought monitoring using mowcatl data mining algorithm in aras basin, turkey table 14. correlation chart of indexes ardahan/correlations aspi1 aspi3 aspi6 aspi12 ao mei nao oni pdo pna soi aspi1 1 .612** .464** .336** -.106* -.054 -.054 -.031 -.023 .061 .089 aspi3 1 .764** .568** -.087 -.084 -.068 -.043 -.038 .084 .111* aspi6 1 .764** .002 -.104* -.080 -.043 -.105* .035 .112* aspi12 1 -.018 -.213** -.107* -.152** -.180** .029 .178** arpaçay/correlations aspi1 aspi3 aspi6 aspi12 ao mei nao oni pdo pna soi aspi1 1 .634** .517** .422** -.073 -.077 -.070 -.083 -.155** -.009 .082 aspi3 1 .811** .689** -.028 -.119** -.008 -.122** -.200** .042 .144** aspi6 1 .864** .012 -.095* -.041 -.098* -.235** -.037 .116* aspi12 1 .019 -.093* -.042 -.103* -.271** -.022 .095* doğubeyazit/correlations aspi1 aspi3 aspi6 aspi12 ao mei nao oni pdo pna soi aspi1 1 .563** .423** .299** -.062 -.019 -.110* -.015 -.032 .036 .024 aspi3 1 .751** .510** -.041 -.019 -.090 .001 -.065 .017 .021 aspi6 1 .743** -.011 .012 -.049 .025 -.119** -.001 -.050 aspi12 1 -.022 -.007 -.059 -.040 -.190** .040 -.041 horasan/correlations aspi1 aspi3 aspi6 aspi12 ao mei nao oni pdo pna soi aspi1 1 .563** .423** .299** -.062 -.019 -.110* -.015 -.032 .036 .024 aspi3 1 .751** .510** -.041 -.019 -.090 .001 -.065 .017 .021 aspi6 1 .743** -.011 .012 -.049 .025 -.119** -.001 -.050 aspi12 1 -.022 -.007 -.059 -.040 -.190** .040 -.041 iğdir/correlations aspi1 aspi3 aspi6 aspi12 ao mei nao oni pdo pna soi aspi1 1 .536** .338** .247** -.109* -.034 -.172** -.019 -.005 .024 .086 aspi3 1 .664** .463** -.014 -.101* -.031 -.054 -.025 .029 .111* aspi6 1 .694** .069 -.125** .008 -.090 -.086 -.040 .102* aspi12 1 .077 -.185** .021 -.133** -.155** -.003 .115* kars/correlations aspi1 aspi3 aspi6 aspi12 ao mei nao oni pdo pna soi aspi1 1 .582** .463** .356** -.132** -.017 -.131** .020 -.027 .019 .033 aspi3 1 .736** .594** -.074 -.042 -.059 .033 -.009 .119* .061 aspi6 1 .793** -.005 -.070 -.077 .021 -.059 .067 .071 aspi12 1 .009 -.088 -.061 -.003 -.081 .072 .050 **: correlation is significant at the 0.01 level (2-tailed). *: correlation is significant at the 0.05 level (2-tailed). 196 emre topçu references alamdarloo, h. e., khosravi, h., nasabpour, s., & gholami, a. (2020). assessment of drought hazard, vulnerability and risk in iran using gis techniques. journal of arid land, 12, 984-1000. https://doi.org/10.1007/ s40333-020-0096-4. bayardo, r. j., & agrawal, r. (1999). mining the most interesting rules. proceedings of the fifth acm sigkdd international conference on knowledge discovery and data mining. san diego, ca, association for computing machinery, 145–154. bos, m. g., kselik, r. a., allen, r. g., & molden, d. (2008). water requirements for irrigation and the environment. springer science & business media. https://doi.org/10.1007/978-1-4020-8948-0. daculaweather (2021). https://www.daculaweather.com/4_ao_index.php#:~:text=the%20arctic%20oscillation%20(ao)%20is,which%20the%20 opposite%20is%20true. (access date: 02.02.2021). ebrahimpour, m., rahimi, j., nikkhah, a. & bazrafshan, j. (2015). monitoring agricultural drought using the standardized effective precipitation index. journal of irrigation and drainage engineering, 141(1), 04014044. https://doi.org/10.1061/(asce)ir.1943-4774.0000771 edwards, d. c., & mckee, t. b. (1997). characteristics of 20th century drought in the united states at multiple time scales. climatology report number, 97-2, colorado state university, fort collins, co. feddes, r. a, kabat. p., van, b. p., bronswijk, j. j. b., & halbertsma, j. (1988). modelling soil water dynamics in the unsaturated zone—state of the art. journal of hydrology, 100(1), 69–111. https://doi.org/10.1016/00221694(88)90182-5 francis, r. c., & hare, s. r. (1994). decadal-scale regime shifts in the large marine ecosystems of the northeast pacific: a case for historical science. fisheries oceanography, 3, 279–291. https://doi.org/10.1111/j.1365-2419.1994. tb00105.x gao, l., & zhang, y. (2016). spatio-temporal variation of hydrological drought under climate change during the period 1960–2013 in the hexi corridor, china. journal of arid land, 8(2), 157–171. https://doi.org/10.1007/ s40333-015-0022-3 groth, r. (1998). data mining: a hands-on approach for business professionals. prentice hall, new jersey, 264 pp. harms, s. k., deogun, j. & tadesse, t. (2002). discovering sequential rules with constraints and time lags in multiple sequences. proceedings of the 13th international symposium on foundations of intelligent systems, 432441, ismis, lyon, france. doi:10.1007/3-540-48050-1_47 harms, s. k., & deogun, j. (2004). sequential association rule mining with time lags. journal of intelligent information systems, 22(1), 7-22. https://doi. org/10.1023/a:1025824629047 hilderman, r. j., & hamilton, h. j. (1999). knowledge discovery and interestingness measures: a survey. computer science, university of regina, 28 pp. hurrell, j. w. (1995). decadal trends in the north atlantic oscillation: regional temperatures and precipitation. science, 269, 676–679. mckee, t. b., doesken, n. j., & kleist, j. (1993). the relationship of drought frequency and duration to time scales. 8th conference on applied climatology, american meteorological society, anaheim, california, 17-22 january. mcphaden, m. j., busalacchi, a. j., cheney, r., donguy, j. r., gage, k. s., halpern, d., ji, m., julian, p., meyers, g., mitchum, g. t., niiler, p. p., picaut, j., reynolds, r. w., smith, n., & takeuchi, k. (1998). the tropical ocean-global atmosphere observing system: a decade of progress. journal of geophysical research, 103(c7), 14,169–14,240. https://doi. org/10.1029/97jc02906 national oceanic and atmospheric administration (noaa). (2021). https:// www.climate.gov/news-features/understanding-climate/climate-variability-oceanic-ni%c3%b1o-index#:~:text=the%20oceanic%20ni%c3%b1o%20index%20(oni,or%20%e2%80%9censo%e2%80%9d%20 for%20short (access date: 02.02.2021). overland, j. e., niebauer, h. j., adams, j. m., bond, n. a., & mcnutt, s. l. (2002). causes of variability in the aleutian low: a project for the arctic research initiative. journal of climate, 12(5), 1542-1548. padmanabhan, b. & tuzhilin, a. (1999). unexpectedness as a measure of interestingness in knowledge discovery. decision support systems, 27, 303– 318. https://doi.org/10.1016/s0167-9236(99)00053-6 patwardhan, a. s., nieber, j. l., & johns, e. l. (1990). effective rainfall estimation methods. journal of irrigation and drainage engineering, 116(2), 182–193. rind, d., goldberg, r., hansen, j., rosenzweig, c., & ruedy, r. (1990). potential evapotranspiration and the likelihood of future drought. journal of geophysical research, 95, 9983–10.004. https://doi.org/10.1029/jd095id07p09983 rostam, g. m., sadatinejad, j. s., & malekian, a. (2020). precipitation forecasting by large-scale climate indices and machine learning techniques. journal of arid land, 12(5), 854–864. https://doi.org/10.1007/s40333-0200097-3 shokoohi-yekta, m., chen, y., campana, b., hu, b., zakaria, j., & keogh, e. (2015). discovery of meaningful rules in time series. proceedings of the 21th acm sigkdd international conference on knowledge discovery and data mining, 1085–1094. https://doi.org/10.1145/2783258.2783306 silberschatz, a., & tuzhilin, a. (1995). on subjective measures of interestingness in knowledge discovery. proceedings of the first international conference on knowledge discovery and data mining, montreal, canada, association for computing machinery, 275–281. smyth, p., & goodman, r. m. (1992). an information theoretic approach to rule induction from databases. ieee transactions on knowledge and data engineering, 4(4), 301–316. doi: 10.1109/69.149926 tadesse, t. (2002). identifying drought and its associations with climatic and oceanic parameters using data mining techniques. ph.d. dissertation, university of nebraska, lincoln. tadesse, t., wilhite, d., harms, s., hayes, m., & goddard, s. (2004). drought monitoring using data mining techniques: a case study for nebraska, usa. natural hazards, 33, 137-159. tadesse, t., wilhite, d. a., hayes, m. j., & goddard, s. (2005). discovering associations between climatic and oceanic parameters to monitor drought in nebraska using data-mining techniques. (2005). journal of climate, 18, 1541-1550. tigkas, d., vangelis, h. & tsakiris, g. (2019). drought characterisation based on an agriculture-oriented standardized precipitation index. theoretical and applied climatology, 135, 1435–1447. https://doi.org/10.1007/ s00704-018-2451-3 trenberth, k. e., & hoar, t. j. (1996). the 1990–1995 el niño-southern oscillation event longest on record, geophysical research letters, 23(1), 57–60. https://doi.org/10.1029/95gl03602 two crows corporation. (1999). introduction to data mining and knowledge discovery. third edition. two crows corporation, postmac, md. wang, y., kong, y., chen, h., & ding, y. (2020). spatial-temporal characteristics of drought detected from meteorological data with high resolution in shaanxi province, china. journal of arid land, 12(4), 561–579. https:// doi.org/10.1007/s40333-020-0066-x wolter, k., & timlin, m. s. (1993). monitoring enso in coads with a seasonally adjusted principal component index. proceedings seventh annual climate diagnostic workshop, norman, oklahoma, march 1993, pp. 52–57. wu, y., batur, b., zhang, j., & rasulov, h. (2015). spatio-temporal patterns of drought in north xinjiang, china, 1961–2012 based on meteorological drought index. journal of arid land, 7(4), 527–543. https://doi. org/10.1007/s40333-015-0125-x yang, m., yu. y., zhang, h., wang, q., gan. m., & yu, r. (2020). tree ring based drought variability in northwest tajikistan since 1895 ad. journal of arid land, 12(3), 413–422. https://doi.org/10.1007/s40333-020-0062-1 yurekli, k., sattari, t. m., anlı, s. a., & hınıs, a. m. (2012). seasonal and annual regional drought prediction by using data-mining approach. atmósfera, 25(1), 85-105. https://doi.org/10.1007/s40333-020-0096-4 https://doi.org/10.1007/s40333-020-0096-4 https://doi.org/10.1007/978-1-4020-8948-0 https://doi.org/10.1061/(asce)ir.1943-4774.0000771 https://doi.org/10.1016/0022-1694(88)90182-5 https://doi.org/10.1016/0022-1694(88)90182-5 https://doi.org/10.1111/j.1365-2419.1994.tb00105.x https://doi.org/10.1111/j.1365-2419.1994.tb00105.x https://doi.org/10.1007/s40333-015-0022-3 https://doi.org/10.1007/s40333-015-0022-3 https://doi.org/10.1023/a:1025824629047 https://doi.org/10.1023/a:1025824629047 https://doi.org/10.1029/97jc02906 https://doi.org/10.1029/97jc02906 https://doi.org/10.1016/s0167-9236(99)00053-6 https://doi.org/10.1029/jd095id07p09983 https://doi.org/10.1007/s40333-020-0097-3 https://doi.org/10.1007/s40333-020-0097-3 https://doi.org/10.1145/2783258.2783306 https://doi.org/10.1007/s00704-018-2451-3 https://doi.org/10.1007/s00704-018-2451-3 https://doi.org/10.1029/95gl03602 https://doi.org/10.1007/s40333-020-0066-x https://doi.org/10.1007/s40333-020-0066-x https://doi.org/10.1007/s40333-015-0125-x https://doi.org/10.1007/s40333-015-0125-x https://doi.org/10.1007/s40333-020-0062-1 n1260 keywords: aghbolagh; oshnavieh; involved fluids, la/y ratio, skarn, vein. palabras clave: aghbolagh; oshnavieh (irán); inclusiones fluidas; proporción lantano/itrio; skarn; veta how to cite item dana, k., nezafati, n., & abedini, m. v. (2020). study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran. earth sciences research journal, 24(2), 231-243. doi: https://doi. org/10.15446/esrj.v24n2.83340 aghbolagh copper deposit is located in the southwest of west azarbaijan province, 10 km north of oshnavieh city. most of the existing rock outcrops in the area include cambrian deposits (carbonated and clastic deposits). these deposits are metamorphosed by intrusive-granite masses (skarn-genesis). the infiltration of the mineralization solution at distances beyond contact and inside the sandstone sequences has resulted in the formation of copper ore veins. for microscopic and geochemical studies, 21 thin sections, and 22 doubly-polished cross-sections, 22 xrd specimens, 22 icp-ms specimens, and three samples for analysis of 10 oxides were analyzed by zarazma corporation. in the study of the fluids involved, measurements were made using a linkham thmsg600 heater and freezer plate, mounted on a polarized microscope with a separate stage and olympus model. studies on quartz minerals (3 samples) and garnet minerals (2 samples) were performed on five polished double sections by micro-thermometry measurements on 30 inclusion in quartz mineral and 15 inclusion in garnet mineral at rizkavan land company. the results show that the la/y ratio, which is a proper indicator for the ph conditions of the formation environment of ore deposit, in aghbolagh deposit is a range from 0.09 to 2.26. the granite mass has the highest amount, and the skarn zone represents two types of conditions. type 1: section where the la/y> 1 ratio observed in samples close to the skarn mass and ore. type 2: part where the ratio la/y <1 is close to marble lithology. in the case of a vein deposit, the samples have mineralization under alkaline conditions, and other samples show acidic conditions. the examination of fluid inclusions determined two main groups of fluids in this mineralization system. group 1: supersaturated fluids of salt, which is the origin of halite multiphase inclusions in the garnet ore during the progressive skarn alteration process. group 2: salt-saturated fluids, which are the source of fluid-rich two-phase intermediates in the garnet and quartz minerals, and have been effective in regressive alteration in the skarn system. abstract study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran estudio de inclusiones fluidas y geoquímica de elementos tierras raras del depósito cuprífero de aghbolagh, al norte de oshnavieh, en la provincia iraní de west azerbaijan issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.83340 el depósito cuprífero de aghbolagh se localiza en el sudeste de la provincia de west azerbaijan, 10 kilómetros al norte de la ciudad iraní de oshnavieh. los afloramientos rocosos en el área incluyen depósito del cámbrico (carbonatados y clásticos. estos depósitos fueron transformados por masas intrusivas de granito (génesis del skarn). la infiltración de soluciones de mineralización a distancias más allá del punto de contacto y al interior de secuencias de arenisca resultaron en la formación de vetas de cobre. para los estudios microscópicos y geoquímicos se analizaron 21 secciones delgadas y 22 secciones transversales doblemente pulidas, 22 muestras para análisis de difracción de rayos x, 22 muestras para análisis de espectrometría de masas por plasma acoplado inductivamente, y tres muestras para el análisis de 10 óxidos, que fueron procesados en los laboratorios de zarazma corporation. en el estudio de la inclusiones fluidas, las mediciones se hicieron con una bandeja de calentamiento y enfriamiento linkham thmsg600, montada en un microscopio polarizado modelo olympus. para los estudios en minerales de cuarzo (tres muestras) y de granate (dos muestras) se utilizaron cinco secciones doblemente pulidas donde se hicieron las medidas de microtermometría para treinta inclusiones en minerales de cuarzo y quince inclusiones en minerales de granate en los laboratorios de rizkavan land company. los resultados muestran que la proporción de lantano/itrio, que es un indicador apropiado para establecer las condiciones de alcalinidad en el ambiente de formación del depósito mineral, en el depósito de aghbolagh se encuentra en un rango de 0.09-2.26. la masa de granito registra una mayor proporción, mientras que en la zona del skarn se presentan dos condiciones diferentes. tipo 1: sección donde la proporción lantano/itrio es superior a 1 en las muestras cercanas a la masa de skarn. tipo 2: la sección donde la proporción lantano/itrio es inferior a 1 en la litología del mármol. para el caso de la veta, las muestras tienen una mineralización bajo condiciones alcalinas mientras que otras muestras tienen condiciones de acidez. el examen de las inclusiones fluidas determinó dos grupos principales de fluidos en este sistema de mineralización. grupo 1: fluidos supersaturados de sal, que es el origen de las inclusiones multifásicas de halita en el granate durante el progresivo proceso de alteración del skarn. grupo 2: fluidos saturados de sal, que son la fuente de fluidos ricos de dos fases intermedias en los minerales de cuerzo y granate, y que han sido efectivos en la alteración regresiva del sistema de skarn. resumen record manuscript received: 04/11/2019 accepted for publication: 24/01/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 231-243 kamal dana, nima nezafati, mansour vosoughi abedini department of geosciences and geophysics, science and research branch, islamic azad university, tehran, iran * corresponding author: nima.nezafati@gmail.com o r e g e o l o g y https://doi.org/10.15446/esrj.v24n2.83340 https://doi.org/10.15446/esrj.v24n2.83340 https://doi.org/10.15446/esrj.v24n2.83340 232 kamal dana, nima nezafati, mansour vosoughi abedini introduction the aghbolagh copper deposit in the southwest of west azarbaijan province and 10 km north of oshnavieh city, in the range of 45°,01′ to 45°, 09′ east length, and 37°, 09′ to 37°, 13′ in the northern latitude in the contact of the oshnavieh granitic mass was created with the cambrian carbonate rocks and in the sanandaj-sirjan zone. also, infiltration of the mass and mineralization solution at distances away from contact and within the sandstone sequences resulted in the formation of quartzite rocks and copper ore veins (figure 1). the oshnavieh intrusive assemblage has emerged of three intrusive families of diorite, granite, and alkali syenite-alkali granite (as-ag) in the northwestern part of the sanandaj-sirjan zone. the granite family (one of the three oshnavieh intrusive mass) with the distinctive features of s-type granites is the result of the partial melting of the polite-gray wake stones (ghalamghash et al., 2011). oshnavieh intrusive masses are one of the actual intrusions in the sanandaj-sirjan zone. this area has previously been investigated by researchers such as naghizadeh et al. (2004) to prepare a geological map 1: 10000 of oshnavieh (ghalamghash et al., 2011). from the perspective of isotopic age determination, the k-ar method has been investigated. due to the formation of region diorites and their synchronization of formation with granite rocks 100 million years ago, the oshnavieh intrusive assemblage appears to be the result of the subduction of the neotethys oceanic crust beneath central iran. dating by the k-ar method on granitic biotites has shown periods of 100 + 1.9 and 98.9 + 1.5 million years (late early cretaceous) (ghalamghash et al., 2011). one of the most useful ways to interpret the petrogenetic of intrusive masses is to use geochemical data and to study changes and ratios of major and trace elements in these masses (tabatabaiemanesh et al., 2011). experimental proof of the origin of hydrothermal solutions responsible for the formation of skarn deposits is one of the most essential stages of studies of these types of deposits. the condensation of rare earth elements, la to lu, in hydrothermal solutions provides useful information on the origin of the constituent elements of the deposit containing the basic elements (kato, 1999). one of the significant advances in ore deposits researches in recent years is the development of geochemical models related to various stages of transport and deposition of the ore-bearing fluids. these models, which are widely used in the design of exploration strategies, are primarily based on information obtained from fluid inclusion. fluid inclusions are the only source of information on the density of mineralized fluids. the eutectic point for the nacl + h2o system is approximately te <-20.8co. however, the eutectic point or te in some of the samples studied is lower than -20.8 co, indicating the presence of salts other than nacl in the mineralizing fluid. materials and methods for the microscopic study of samples, 21 thin sections and polished sections were prepared. after microscopic studies of samples and determination of tissue relationships between minerals and for geochemical works, 22 samples were analyzed by xrd and 22 samples of icp-ms and three samples for analysis of 10 oxides analyzed by zarazma company (table 2). for investigating the involved fluids, measurements were done using the linkham thmsg600 heater and freezer plate and mounted on a polarized microscope with the olympus model and separate from staging. fluid inclusion studies have been performed on quartz (3 samples) and garnet minerals (2 samples). for the study of fluid inclusions, five polished doubly sections were prepared. after performing petrographic fluid inclusions on them, five samples were selected for micro-thermometers measurements, and micro-thermometers measurements on 30 inclusions in quartz ore and 15 inclusions in garnet ore was performed. geology the aghbolagh copper deposit is located in the northwestern end part of the sanandaj-sirjan zone in terms of structural position. according to map 1:100,000 (naghizadeh, 2004), oshnavieh is partially separated and located in the aghbolagh copper deposit, including cambrian geological units. barut formation: alternating limestone and gray dolomite crystallized with shale; soltanieh formation: dolomite, creamy dolomitic limestone containing chert, marble; mila formation: limestone and dolomite locally associated with dark-gray shale; equivalence of zagun-lalun formation: shale-siltstone and sandstone in green, red-gray archaeozoic sandstone. the permian unit includes ruteh formation: limestone with an interlayer of dolomite, shale and quartzite sandstone; dord formation: red quartzite sandstone with shale and limestone interlayers, granite intrusive masses, diorite monzogabbro are coarse grains (figure 1). the most prominent geological units in the order of old to new consist of cambrian units (quartz sandstone, metamorphosed carbonate rocks (dolomite and metamorphic limestone), shale, and siltstone altered), permian units (quartzite rocks) and sandstone and dolomite, which are from quaternary formations. the most important economic and geological features of this area are cu mineralization at the site of granitoid contact with marble and outward copper mineralization within quartzite sandstones and veins (figure 1). aghbolagh granite mass in field surveys, the granite mass is in the central part of the range. these medium to coarse-grained granites have a bright, white appearance (figure 2a and 2d). this mass is leucocratic, containing significant quartz minerals, feldspar, plagioclase, and biotite, muscovite, sericite, and dark minerals, mainly pyrite (figure 3). these granites are poor in terms of the percentage of dark minerals, which is a characteristic feature of these granitoid, reducing their color coefficient. the low rate of mafic minerals in these rocks can be called leucogranite (moein vaziri and ahmadi, 2001). the granite intrusions are weathering, and have been altered at the surface. the color of the mass is predominantly white with fineto medium minerals and crystal-coarse in some places. the anorthosite minerals were visible in the central, coarse-crystalline sections, and towards the contact, the size of the minerals decreases, and it takes on aplitic texture. the dark minerals of this mass are mainly pyrite (fig. 2b), and in areas where more alteration has taken place, the mass has been reddish to brown (fig. 2c). this unit is the leading cause of the formation of skarn, and the resulting solutions have formed mineralized veins. for naming, determination of the magmatic type, series, and tectonic position of aghbolagh intrusive mass have been discussed. in this regard, two samples from different sections and one from the contact boundary were taken and sent to zarazma company for analysis of ten oxides (table 1). based on the diagram (cox et al., 1979), the intrusive mass object of this study has a granite composition, and according to the diagram (de la roche et al., 1980), they have a combination of granite to alkali granite. and based on the diagram (peccerillo and taylor, 1976), the samples are in the range of highpotassium calc-alkaline series (figure 4b). based on the work of shand (1943), modified by villaseca et al. (1998), the granite mass is in the pre-alumina class (figure 4d). and also, by using sio2 diagram against a/cnk (taylor, 1976), the mass can be classified into s-type granites. the position of the intrusive mass is based on diagrams (pearce et al., 1984) of volcanic arc granites (vags), which is formed simultaneously as the syn-collisional. according to the diagrams (schandl and gorton, 2002), the studied granites are related to the active margin of the continents, consistent with the nature of the sanandaj-sirjan zone (figure 5). table 1. analysis results of ten oxides of intrusive masses of aghbolagh-oshnavieh copper deposit l.o.ip2o5mnotio2mgok2ona2ocaofe2o3al2o3sio2sample 0.64<<0.130.33.624.520.421.613.8174.74aq-gt-sh2 2.03<<0.09<0.410.010.162.9717.7464.89ag-sg-2 0.77<<0.080.083.853.70.50.9412.9976.79aq-sg-6 233study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran figure 1. location and geological map of the study area (aghbolagh-oshnavieh copper deposit) figure 2. a: a view of the light granite mass, b: a view of the granite mass with pyrite minerals, c) a view of the alteration of the iron ore and its discoloration, d: a view of the aghbolagh copper ore deposit’s granite outcrop. geology of skarn unit the skarn unit is continuously expanding in the northern part of the intrusive mass along the west to east. this unit is fully covered and can only be identified with garnite outcrops (figure 6a) and hornfels and gossans (iron hat). mineralogy is only visible in a small part of the western range, but the presence of iron warhead indicates the presence of sulfide ores below (figure 6c). the minerals that can be identified in the hand sample in this section are quartz, garnet, epidote, cordierite, calcite, and minerals, including chalcopyrite, bornite, malachite, and azurite (figures 6 and 7). geology of vein ore deposit these deposits in particular contain sedimentary rock (dolomite, limestone, sandstone and shale) that have been affected by the intrusions of stoke, dyke and middle to felsic igneous floods. (figure 8) (morris, 1986; cox, 1986). the sandstone zone is located about 200 meters from the intrusive mass behind marble rocks. the ore solutions enter through the marble through the seams and crevices in the sandstones (the seam and crevice was probably caused by the igneous mass intrusion), which is a good place to deposit soluble materials. . in the meantime, veins of several centimeters to 3 meters thick are formed in these rocks. the veins are quartzitic and contain minor amounts of microcline, muscovite, sercite, hematite, goethite and sulfide and carbonate minerals based on xrd and light analyzes. soil space filling structures and dog teeth construction are abundant in this area. chalcopyrite, chalcocite, malachite, azurite and quartz are minerals in this part of the deposit (figure 9). discussion and investigation rare earth element geochemistry (ree) most studies of the behavior of rare earth elements (rees) in hydrothermal environments in active systems have been reviewed. so far, 234 kamal dana, nima nezafati, mansour vosoughi abedini figure 3. microscopic images of igneous rocks of the area: a) amorphous quartz minerals and intrusions of sericite, b) polysynthetic micelle in plagioclase showing little alteration, c) minerals forming granite, chlorite, feldspar potassium altered to sericite, quartz, and plagioclase, d) altered plagioclase and feldspar potassium, e) birefringence of muscovite with plagioclase and quartz, f) biotite and chlorite (biotite chlorite). qtz=quartz, pl=plagioclase, kfs= k feldspar, bt= biotite, chl= chlorite, ser= sericite, mo= muscovite figure 4. diagrams used to determine the magmatic series of agbolagh granitic mass a) the position of the intrusive samples in the afm diagram (a = (na2o + k2o), f = feot, m = mgo), (irvine and baragar, 1971) b) the position of the samples in the sio2 diagram versus k2o (peccerillo and taylor, 1976) c) position of the specimens in the a/cnk-a/nk aluminum diagram (shand, 1943) d) the position of the samples in the b-a plot diagram (a = al(k + na + 2ca), b = fe + mg + ti (villaseca et al, 1998) figure 5. the position of the specimens related to the aghbolagh granites mass in the diagrams (a) rb against nb, (b) y + nb against rb, (c) y against ta, (d) ta+yb against yb, to determine its tectonic position. (adapted from pearce et al., 1984). relatively few studies have been performed on the distribution processes of rare earth elements in fossilized hydrothermal systems or leachate metasomatism products from igneous, metamorphic or hydrothermal fluid associated with leached skarn deposits in carbonate rocks and adjacent silicates. extensive metasomatism of rare earth elements can occur in lowto medium-depth hydrothermal systems, and the geochemistry of these elements can be useful in identifying the origin of skarns and related deposits (adriana et al., 2009). the chondrite-normalized ree pattern for the granite mass, skarn, marble, and vein deposit indicates that although the rees pattern is slightly uniform, the lrees overall show more enrichment than the hrees in different lithologies (figure 9). the range of numbers normalized to chondrite in aglabagh deposit varies considerably, (2.58 to 141.93 for la and 0.14 to 27.27 for yb). this indicates that distinct differences in the increase and decrease of lrees and hrees occurred during the formation of the aghbolagh ore deposits. to evaluate the separation coefficient, between rees, different ratios of rees including (la/yb)n’, (la/sm)n and (gd/yb)n are used. (la/ yb)n calculates the separation coefficient of lrees from hrees during geochemical processes (aubert et al., 2001; yusoff et al., 2013) and calculates two other ratios of separation coefficients between lrees and mrees and between mrees and hrees. (yusoff et al., 2013).the extent of these ratios, range from (0.42 to 50.56) for (la / yb) n and (0.25 to 125/80) for (la / sm) n and (0.006 to 8.48) for (gd / yb) is n). the highest separation occurred between lrees and hrees (more than 125) and the least separation occurred between mrees and hrees during the scanning process. the highest coefficient of these ratios is in marble zone and only the coefficient (gd / yb) n in skarn has the highest coefficient, indicating that the highest separation in marble zone is not only due to hree deficiency but also due to hree and mree deficiency. 235study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran it is a zone. the la / y ratio well indicates the ph conditions of the deposit formation environment. (la / y> 1) indicates alkaline conditions and (la / y <1) represents acidic conditions (crinci and jurkowic, 1990). this ratio in the aghbolagh deposit ranges from (0.09 to 2.26). intrusive mass shows the highest (la / y) ratio of 2.26. the granite mass shows the highest amount of la / y (2.26). the skarn zone (endoskarn and exoskarn and ore present in exoskarn) show two types of conditions. type 1: section where la/y> 1 ratio is observed in samples close to the mass and skarn ore is also in this range. option 2: part where the ratio la/y<1 is close to marble lithology. these divisions are indicative of acidic conditions and hydrothermal alteration during the skarnification process. this is also observed in the veins of the aghbolagh deposit. according to the diagram of the deposit from the beginning of the granitic mass to the middle of the exoskarn zone (sample aq-ses7), alkaline conditions were dominant and from this point to the end of the host rock (marble) acidic conditions were present. in the case of vein deposit, the sample also contains alkaline mineralization and other samples show acidic conditions. hence, we can predict that in an environment affected by hydrothermal solutions where acidic and hot solutions containing ree are found due to interaction with dilute carbonate rocks and they are found to be alkaline state. the eu / eu * anomaly in the aghbolagh copper deposit ranges from (0.62 to 12.14) (figure 2 and table 2). more than one is called a positive anomaly and less than one is called a negative anomaly. the eu / eu * anomaly was positive in the intrusive mass, negative in the skarn zone, and again positive in the marble section. so in this section we have two positive anomalies on the mass side and host rock and one negative anomaly is observed in the skarn zone. these anomalies can be interpreted by the presence of plagioclase at the mass side and the presence of albite minerals in the marble zone and vein-section. in the skarn zone, the anomaly is negative due to the absence of plagioclase (zamanian and radmard, 2016). the positive eu anomaly indicates the stages of rock change in an acidic environment (kikawada, 2001). positive anomalies of eu and (la / lu) n also indicate high temperature acidic liquids (wood, 1990). based on two parameters eu / eu * and (pr / yb) n in the skarn zone, two types of hydrothermal solutions can be identified that affect the host rock: 1) solutions of low ree concentration with anomaly eu positive and acidic conditions and oxidizing property. 2) ree-rich solution with negative eu anomaly indicates alkaline conditions of such a solution (figure 10). the presence of a high ratio of exoskarn to endoskarn and extension of the exoskarn zone indicates low depth conditions (less than 5 km) and intense oxygen activity (oxidation conditions) (zamanian, 2003). involved fluids examination of endoskarn unit sample sample number: ag-sg-2 sample feature: this sample consists of quartz and carbonate. large crystalline quartzes are observed between the faceted carbonate crystals. figure 6. macroscopic images of the aghbolagh copper ore deposit exo-skarn zone a. view of garnet’s hand sample, epidote, and calcite of the exoskarn zone; b. view of garnet’s hand sample in light brown, epidote, calcite, and magnetite vein; c. view of the iron cap showing is a supergene zone; d. view of the mineralization in the exo-skarn zone with magnetite, malachite, azurite, and chalcopyrite minerals. grt= garnet, mag= magnetite, epi=epidote, mal= malakite, hem=hematite, cal=calcite figure 7. microscopic images of the exo-skarn zone of aghbolagh region: a) garnet mineral in normal light and its alteration to epidote; b) garnet mineral, epidote, and quartz in cross light; c) epidote mineral of pseudomorph garnet, chlorite in normal light; d) pseudomorph epidote after garnet. qtz=quartz, grt=garnet, epi= epidote, chl=chlorite figure 8: schematic image of a substituted polymetallic deposit, that showing different types and host rocks. (morris, 1986; cox, 1986) figure 9: (a): overview of quartz samples containing chalcopyrite mineralization, malachite, (b): quartz vein containing copper mineralization (chalcopyrite, chalcocite, malachite and azurite), (c): view of azurite and chrysocolla mineralization, (d): a view of the quartz vein containing mineralization 236 kamal dana, nima nezafati, mansour vosoughi abedini figure 9. trends of changes of (la/yb) n, (la/sm) n and (gd/yb) n in different zones of aghbolagh deposit figure 10. trends of changes in la/y ratios in different zones of aghbolagh deposit table 2. values of rare earth elements of aghbolagh ore deposit samples and ratios between rare earth elements for conditions evaluation of ore formation sample a qs g 2 a qs g 6 a qg t sh 2 a qs n s8 a qe ps h 4 a qs e s1 0 a qg o 20 a qs h s9 a qe p2 3 a qs e s7 lree la-nd ppm 2.12 50.56 84.06 154.74 121.88 114.48 24.21 127.37 47.62 121.54 mree sm-tb ppm 0.28 2.77 4.42 14.95 7.11 8.72 1.34 11.06 5.12 13.79 hree er-lu ppm 1.1 1.91 2.37 9.24 4.58 4.09 3.34 6.76 2.24 10.54 ∑ree (la-lu) ppm 3.5 55.24 90.85 189.93 133.57 127.29 28.89 145.19 54.98 145.87 (ln/lu)n 1.30 12.61 17.53 14.28 15.58 2.60 5.77 6.49 11.69 2.60 la/y 0.15 1.77 2.27 1.38 1.89 1.76 1.71 1.29 1.12 0.95 eu/eu* 14.12 1.68 1.29 0.76 0.75 0.62 1.53 0.96 0.93 0.76 (la/sm)n 62.90 7.13 7.65 4.28 6.35 3.94 9.21 4.48 3.57 3.14 (la/yb)n 2.25 16.37 20.23 6.45 9.67 10.11 2.89 7.64 9.74 3.55 (gd/yb)n 0.08 0.59 1.05 0.92 0.04 0.24 0.78 8.48 1.40 0.81 (pr/yn)n 0.11 3.57 5.25 3.18 4.16 5.12 0.50 3.41 3.85 1.97 sample a qe x sh 1 a qg a e p a qs m 3 a qs m 17 a qs m 18 a qv 15 a qv 16 a qv t 8 a qv t 9 lree la-nd ppm 5.23 31.83 5.43 21.53 5.43 49.35 2.03 4.23 25.31 mree sm-tb ppm 0.9 5.47 0.2 2.97 0.4 1.74 0.2 0.33 2.46 hree er-lu ppm 4.11 2.64 0.22 0.72 0.74 0.85 0.22 0.42 2.95 ∑ree (la-lu) ppm 10.24 39.94 5.85 25.22 6.57 51.94 2.45 4.98 30.72 (ln/lu)n 1.30 10.39 5.19 6.49 8.62 19.48 1.04 3.90 11.42 la/y 0.53 0.67 0.53 0.27 2.24 0.32 0.48 0.78 eu/eu* 1.84 0.87 14.12 0.66 5.34 3.64 14.12 5.77 0.89 (la/sm)n 4.27 1.20 125.81 2.28 17.97 7.93 50.32 31.45 6.12 (la/yb)n 0.75 1.69 44.95 33.71 13.48 50.57 17.98 10.11 4.12 (gd/yb)n 0.85 1.37 0.01 0.76 1.06 0.20 0.81 0.12 0.34 (pr/yn)n 0.01 1.22 1.71 3.94 0.51 15.85 1.71 0.26 0.96 figure 11. chart of changes of eu / eu * ratio in different zones of aghbolagh deposit figure 12. graph of changes in the ratio of ∑ree to eu / eu * in the exoskarn zone and aghbolagh ore deposit marble. 237study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran gaseous single phase phases are abundant in quartz crystals. these inclusions are elliptical and elliptic and vary in size from 2 to 10 microns. these inclusions also coincide with fluid-rich biphasic inclusions, which is evidence of boiling. liquid-rich biphasic inclusions are often observed as dispersed aggregates and to a lesser extent as single ones. these inclusions are very high in sample (figure 14). homogenization temperatures to liquid phase of fluid-rich two-phase inclusions are from 190 to 325.2 ° c. most of the measurements were done at temperatures of 310 to 340 ° c. the salinity of these inclusions varies from 7.31 to 20.67 % weight percent of salt equivalent and most of the measurements are between 9 and 12 % weight percent of salt equivalent. in the salinity versus homogeneous temperature diagrams for a fluid-rich two-phase inclusions, an isothermal mixing process is observed (figure 15). by examining the fluid inclusions in the quartz mineral in this sample, a main fluid was identified as a fluid with high temperature and low to medium figure 13. spider diagram of samples of different zones of aghbolagh copper ore deposit, normalized to chondrite (boynton, 1984) figure 14. left: accumulation of liquid-rich biphasic inclusions with irregular shapes. right: a single fluid-rich two-phase inclusion. salinity (temperature 310 to 340 ° c and salinity about 9 to 12 % weight percent of salt equivalent). the mixing process of fluid temperatures affected the evolution of the fluid. the modified source diagram of kesler, 2005 was used to determine the source of the fluids, based on which the fluid-rich two-phase inclusions are located within the seawater (figure 16). examination of skarn unit specimens sample number: ag-ga-21-b sample feature: this sample consists of a massive garnet with quartz. transparent quartz crystals are formed in the space between the faceted garnet crystals. fluid inclusions have been identified in garnet and quartz crystals. types of fluid inclusions in garnet: single-phase gas-liquid, liquidrich two-phase short-lived gas-phase single-phase inclusions are often found 238 kamal dana, nima nezafati, mansour vosoughi abedini in garnet crystals. these inclusions are tear-shaped, irregular, and elliptical, varying in size from 2 to 10 microns. in garnet, liquid-rich biphasic inclusions consist of a liquid phase and a vapor bubble. in these inclusions, the vapor phase volume is about 20 to 30 percent of the inclusion volume. these inclusions range in size from 2 to 10 microns and are often elliptical. the abundance of these inclusions in garnet is low (figure 17). figure 15. distribution of homogenization temperature against salinity for liquid-rich two-phase inclusions in ag-sg-2 sample. the isothermal mixing process is observed. trends from wilkinson (2001). figure 16. chemical composition of hydrothermal fluids based on the type of hydrothermal system in the salinity diagram against homogenization temperature (modified after kesler, 2005). most of the samples are within the seawater range. figure 17. left: accumulation of single phase gas inclusions, in garnet. right: a single liquid-rich two-phase inclusions in garnet. in this sample micrometer studies were performed only on garnet minerals. liquid phase homogenization temperatures are fluid-rich two-phase inclusions from 295.3 to 460 ° c. most of the measurements were performed at temperatures of 310 to 350 ° c. the salinity of these inclusions varies from 11.58 to 18.96 % weight percent of salt equivalent and the highest salinity frequency is from 13 to 19 % weight percent of salt equivalent. in the diagram of the homogenization temperature against salinity for fluid-rich two-phase inclusions, a distinct process of fluid isothermal mixing is observed. figure 18. distribution of homogenization temperature against salinity for liquid-rich two-phase inclusions in ag-ga-21b sample. the specific mixing process of fluid temperature is observed. trends from wilkinson, 2001. by examining the fluid inclusions in the garnet mineral in this sample, a high temperature fluid (310 to 350 ° c) and a salinity ranging from 13 to 19 % weight percent of salt equivalent were identified. the mixing process of the fluid temperature justifies the relatively large changes in fluid salinity. sample number: ag-ga-21-a sample feature: this sample consists of a semi-faceted mass garnet with coarse-grained quartz. quartz fills the space between garnet crystals. the garnet and quartz minerals host fluid inclusions in this sample (figure 20). types of fluid inclusions in garnet minerals: single-phase liquid inclusions, single-phase inclusions, liquid-rich two-phase inclusions, and halite multiphase inclusions (figures 21). 239study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran types of fluid inclusions in quartz minerals: single-phase gaseous intermediate, liquid-rich two-phase inclusions (two generations) in garnet, there is a low frequency single phase liquid inclusions ranging from 5 to 10 microns, and most of them about 5 microns, often with irregular shapes. single-phase gas inclusions are abundant and often have elliptical and lacrimal forms. these inclusions range in size from 5 to 15 microns and are dark in color because of the rich gas. figure 19. chemical composition of hydrothermal fluids based on the type of hydrothermal system in the salinity graph against homogenization temperature (modified after kesler, 2005). the position of the data in this sample is not in a certain position due to fluid mixing. liquid-rich biphasic inclusions consist of a liquid phase and a vapor bubble. in these inclusions, the vapor phase volume is about 30 to 40 percent of the inclusion volume. these inclusions range in size from 5 to 25 microns and often have irregular, lacrimal and elliptical shapes. the frequency of these inclusions in the sample is low. multi-phase halite inclusions were observed in the garnet mineralization. these inclusions have little frequency. these inclusions are irregular and irregular. these inclusions range in size from 10 to 30 microns. the vapor phase volume in these inclusions is 30-40% of the short-circuit volume and the halite phase volume is about 40% of the short-circuit volume. most of them have one or more solid or dark solid phases. single-phase gas quartz mineral inclusions are observed as dark inclusions due to the presence of gas phase in them. these inclusions are elliptical and tearshaped, and their abundance is low in quartz. two generations of liquid-rich biphasic inclusions were observed in quartz. generation 1, which is a liquid-rich two-phase inclusions in which the volume of the vapor phase is about 40 to 50 percent of the short-circuit volume. these inclusions are elliptical, rectangular, and lacrimal. these inclusions range from 3 to 15 microns. these inclusions are abundant in quartz. this inclusions is observed with single-phase gas inclusions, which can be evidence of fluid boiling. generation 2, which is a liquid-rich biphasic inclusion whose vapor phase volume is about 20% of the inclusion volume. the vapor phase brownian motion is seen in these inclusions. these inclusions are 3 to 10 microns in size. the abundance of these inclusions is low in quartz. in this sample, two-phase liquid-rich inclusion and halite-rich of garnet minerals were measured by micro-thermometers. liquid phase homogenization temperatures are fluid-rich two-phase inclusion from 270.3 to 510.2 ° c, with the most frequent measurements being carried out at temperatures of 270 to 310 ° c. the salinity of these inclusions varies from 13.17 to 21.68 % weight percent of salt equivalent and the highest salinity is 16 to 22 % weight percent of salt equivalent. there is no particular trend in the salinity diagram for the homogenization temperature for liquid-rich two-phase inclusion (figure 22). figure 20. left: transparent quartz containing fluid inclusions between garnet (isotropic) crystals. ppl light. right: transparent quartz with fluid inclusions between garnet crystals (isotropes). xpl light. figure 21. left: a halitized multiphase inclusion in garnet. right: a halitized multiphase inclusion with clear and dark solid phases in garnet. 240 kamal dana, nima nezafati, mansour vosoughi abedini in this sample, due to the low frequency of halite inclusions in garnet mineral, only 3 inclusion of this type were analyzed. liquid phase homogenization temperatures are halite-rich multiphase inclusion ranging from 395.2 to 490.3 ° c. the salinity of these inclusions varies from 46.9 to 58.5 % weight percent of salt equivalent and the dissolution temperature of the salts in the studied multi-phase inclusion varies from 395.2 to 490.3 ° c. examination of the fluid inclusions in the garnet mineral revealed two fluids in this sample: a high temperature fluid (about 500 ° c) and a salinity of about 50 wt% salt equivalent and a fluid with a temperature of 270 to 310 ° c. c and salinity 22-22% by weight percent of salt equivalent. the initial formation of the garnet is likely to have occurred with the first fluid and the subsequent fluid with lower temperature and salinity has affected the garnet, causing the garnet to calcite and quartz alteration. the modified fluid diagram of kesler, 2005 was used to determine the source of the fluids, based on which the fluid forming most of the fluid inclusions in this sample has magmatic water origin (fig. 23). wilkinson, 2001 diagram was also used to determine the type of fluid genesis in the study sample. in this diagram, fluid-rich, two-phase inclusions and fluid phases are in the fluid range associated with skarn deposits and with tin and tungsten granitoid deposits. since the fluid inclusions data in this sample are obtained from garnet ore, the type of this sample is related to skarn ore deposit. aghbolagh vein deposit section: sample number: ag-v-16 this sample is a vein composed of quartz and ore matte. large crystalline quartzes are observed between the faceted carbonate crystals. the readable fluid inclusions were observed only in quartz crystals. liquid-rich biphasic inclusions consist of a liquid phase and a vapor bubble. in these inclusion the vapor phase volume is about 30% of the inclusions volume. these inclusion range in size from 2 to 15 microns and often have irregular, lacrimal and elliptical shapes. the frequency of these inclusions is high in quartz (figure 24). by examining the fluid inclusions in the quartz mineral in this sample, two fluids were identified: a fluid with higher temperature and salinity (temperature 280 to 310 ° c and salinity 12 to 15 % weight percent of salt equivalent) and a fluid with low temperature and salinity low (190 to 220 ° c and salinity 9 to 12 % weight percent of salt equivalent). the modified source diagram of kesler, 2005 was used to determine the source of the fluids, based on which the fluid constituting this sample has seawater and basin salinity (figure 25). the modified fluid diagram of kesler, 2005 was used to determine the fluid genesis type in the studied sample, based on which the fluid constituting this sample has seawater and basin salinity (figure 1). 25). wilkinson, 2001 diagram was also used to determine the type of fluid genesis in the study sample. in this diagram, fluid-rich two-phase inclusions are in the fluid range with epithermal and skarn deposits. figure 22. distribution of homogenization temperature against salinity for liquid-rich two-phase inclusion in ag-ga-21a sample. there is no clear trend. trends from wilkinson, 2001. figure 23. chemical composition of hydrothermal fluids based on the type of hydrothermal system in the salinity diagram against homogenization temperature (modified after kesler, 2005). the position of most of the data in this sample is in the magmatic water range. figure 24. left: accumulation of liquid-rich biphasic inclusion with irregular shapes. right: a single gas single-phase inclusion. 241study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran discussion in the samples of these studies, micrometers of fluid inclusions were performed on the garnet and quartz minerals. in the study of liquid-rich biphasic fluid inclusions in a garnet mineral, a fluid with a temperature of 270 to 350 ° c with an average salinity of 16 to 19 wt.% salt equivalent was identified (figures 26 and 27). this fluid is probably related to a retrograde alteration in the scarification process that resulted in the alteration of garnets and the formation of low-temperature minerals such as quartz and calcite from the garnet. the evolution of this fluid mixing process has been influenced by the fluid temperature. garnet minerals also identified halitoneic inclusions that have a temperature of about 500 ° c and a salinity of about 50% by weight of salt, which can be related to the progressive fluidization of the skarn systems and lead to the formation of skarn systems. garnet is included in this system. in the investigation of fluid-rich biphasic fluid inclusions in quartz minerals, two fluids were identified, with the first fluid having a temperature of 260 to 300 ° c and a salinity of about 15 to 12 % weight percent of salt equivalent (figure 28). this fluid is most likely equivalent to a retrograde alteration fluid in the garnet mineral and reflects the initial quartz formation temperature. quartz mineralization also revealed a fluid with a temperature of 180 to 220 ° c and a salinity of 9 to 12 % weight percent of salt equivalent, which is a delayed fluid in the system. in the evolution of these two fluids in quartz, fluid mixing processes with lower and cooler salinity, fluid temperature mixing, and cooling process have a significant impact. for halite multiphase inclusion, in all the measured inclusion, the salt dissolution temperature is equal to the temperature of the liquid phase homogeneity. the location of these inclusion near the th = ts midline can be one of the signs of boiling in the mineralization system. examination of the fluid inclusion from the study sample revealed that two main groups of fluids were present in the formation of fluid inclusion: group 1: salt supersaturated fluids that have been the source of the formation of halitite multiphase intermediates in the garnet mineral during the progressive skarn alteration process. group 2: salt-saturated fluids that are the source of fluid-rich two-phase interconnects in the garnet and quartz minerals in this skarn system and have been effective in retrograde alteration in the skarn system. figure 25. chemical composition of hydrothermal fluids based on the type of hydrothermal system in the salinity graph against homogenization temperature (modified after kesler, 2005). the location of the data in this sample is in the seawater and basin salinity range. bulk software (bakker 2003) was used to calculate of the formation pressure of liquid-rich biphasic inclusion. the homogenization temperature of 350° c with an average salinity of 18 % weight percent of salt equivalent was used to calculate the average pressure of the formation of liquid-rich two-phase inclusion in garnet. according to this software, the average pressure formation in the sample is 15.7 mpa (mpa), which indicates moderate pressure and formation at moderate depths (proportional to the regressive alteration in the skarn system). homogeneous temperature of 300° c and average salinity of 12 wt% salt were used for liquid rich two-phase inclusions in quartz. accordingly, the mean pressure of the formation of inclusion in this sample is 8.1 mpa (mpa), indicating moderate pressure and formation at low depths (proportional to retrograde alteration).wilkinson, 2001 diagram was used to determine the type of fluid genesis in the studied samples. in this graph the fluids associated with the garnet ore are within the fluid range associated with the skarn deposits (fig. 29). also, the quartz mineral fluids are partly in the skarn range and the lower temperature and salinity are in the epithermal range (fig. 29). finally, it is necessary to integrate geological information, alteration and mineralization in order to precisely determine the fluid genesis of the area. figure 26. frequency distribution of homogenization temperature in liquidrich two-phase inclusions in garnet minerals in studied samples. figure 27. frequency distribution of salinity in fluid-rich two-phase inclusions in garnet minerals in studied samples. 242 kamal dana, nima nezafati, mansour vosoughi abedini references: adriana, h., paul, g. s., graham, s. t., colin, h. h., & wolfgang, r. l. (2009). geochemistry of garnet-rich rocks in the southern curnamona province, australia, and their genetic relationship to broken hill-type pb-zn-ag mineralization. economic geologists, 104, 687-712. aubert, d., stille, p., & probst, a. (2001). ree fractionation during granite weathering and removal by waters and suspended loads: sr and nd isotopic evidence. geochimica et cosmochimica acta, 65, 387–406. bakker, r. j. (2003). package fluids 1. computer programs for the analysis of fluid inclusion data and for modelling bulk fluid properties. chemical geology, 194, 3–23. boynton, w. v. (1984). cosmochemistry of the rare earth elements: meteorite studies. in: henderson, p., (ed.), rare earth element geochemistry, elsevier, amsterdam, 63-114. cox, d. p. (1986). descriptive model of zn-pbskarn deposits, in cox, d.p. and singer, d.a. (eds.) mineral deposit models: u.s. geological survey bulletin 1693, p. 90. cox, k. g., bell, j. d., & pankhurst, r. j. (1979). the interpretation of igneous rocks. george, allen and unwin, london. crinci, j., & jurkowic, i . (1990). rare earth elements in triassic bauxites of croatia yugoslavia. travaux 19, 239–248. de la roche, h., leterrier, j., grandcloud, p., & marchel, m. (1986). a classification of volcanic and plutonic rocks using r1–r2 diagrams and major elements relationship with current nomenclature. chemical geology, 28, 183– 210. ghalam, g. j., houshmand, m. s., & vosoughi, a. m. (2011). geology, geochemistry and petrogenesis of the oshnavieh intrusion (northwest of iran). journal of earth sciences, 88, 219232. irvine, t. n., & baragar, w. r. a. (1971). a guide to the chemical classification of the common volcanic rocks. canadian journal of earth sciences, 8, 523-547. kato, y. (1999). rare earth elements as an indicator to origins of skarn deposits, example of the kamioka zn-pb and yoshiwara-sannotake (cu-fe) deposit in japan. resource geology, 49, 183–198. kesler, s. e. (2005). ore-forming fluids. elements, 1, 13–18. kikawada, y. (2001). experimental studies on the mobility of lanthanides accompanying alteration of andesite by acidic hot spring water. chemical geology, 176, 137–149. moein, v. h., & ahmadi, a. (2001). petrography and petrology of igneous age of veins. tarbiat moallem university, tehran. morris, h. t. (1986). descriptive model of polymetallic replacement deposits. in: cox, d. p., & singer, d. a., (eds). mineral deposit models: u.s. geological survey bulletin 1693, p. 99-100. naghizadeh (2004). 1: 100000 geological map of oshnavieh, geological organization and minaral discoveries of county. pearce, j. a., harris, b. w., & tindle, a. g. (1984). trace element discrimination diagrams for the tectonic interpretation of granitic rocks. journal of petroleum science and engineering, 25, 956-983. peccerillo, a., & taylor, s. r. (1976). geochemistry of eocene calcalkaline volcanic rocks from the kastamonu area, northern turkey. contributions to mineralogy and petrology, 58, 63–81. schandl, e. s., & gorton, m. p. (2002). application of high field strength elements to discriminate tectonic settings in vms environments. economic geology, 97, 629-642. shand, s. j. (1943). eruptive rocks. their genesis, composition, classification and their relation to deposits. thomas murby and co, london, pp.448. tabatabaiemanesh, s. m., mirlohi, a., & movahedi, m. (2011). petrology and mineral chemistry of ochestan granitoids (south of mahallat, markazi province). petrology 2(7), 51-76 (in persian). taylor, b. e. (1976). origin and significance of c-o-h fluids in the formation of ca-fe-si skarn, osgood mountains, humboldt county, nevada. unpublished ph.d. thesis, stanford, ca, stanford university, 149 p. villaseca, c., barbero, l., & rogers, g. (1998). crustal origin of hercynianperaluminous granitic batholiths of central spain: petrological, geochemical and isotopic (sr, nd) constraints. lithos, 43, 55-79. figure 28. left: frequency distribution of salinity in fluid-rich two-phase inclusions in quartz minerals in the studied samples. right: frequency distribution of homogenization temperature in liquid-rich two-phase inclusions in quartz minerals in studied samples and identification of two fluid groups. figure 29. salinity diagram against homogenization temperature for all types of hydrothermal deposits: most of the garnet ore fluid inclusions in the studied samples are in the fluid range associated with skarn deposits and quartz ore fluid inclusions are in the range of skarn and epithermal deposits. limits from wilkinson, 2001. 243study of involved fluids and geochemistry of rare earth elements of aghbolagh copper deposit, in north oshnavieh, west azerbaijan province of iran wilkinson, j. j. (2001). fluid inclusions in hydrothermal ore deposits. lithos, 55, 229-272. wood, s. a. (1990). the aqueous geochemistry of the rare-earth elements and yttrium. theoretical predictions of speciation in hydrothermal solutions to 350°c at saturation water vapor pressure. chemical geology, 88, 99–125. yusoff, z. m., ngwenya, b. t. & parsons, i. (2013). mobility and fractionation or ree during deep weathering of geochemically contrasting granites in a tropical setting, malaysia. chemical geology, 349–350, 71–86. zamanian, h. (2003). iron mineralization related to the almoughlagh and south ghorveh batholiths with specific refrenceti the baba ali and gelali deposits. university of pune, 220 pp. zamanian, h. & radmard, k. (2016). geochemistry of rare earth elements in the baba ali magnetite skarn deposit, western irana key to determine conditions of mineralization. geologos, 22(1), 33-47. keywords: gis remote sensing technology; tourist attractions; heavy metal pollution; pollution tracing; coefficient of variation; correlation analysis; palabras clave: tecnología de teledetección sig; atracciones turísticas; contaminación por metales pesados; rastreo de contaminación; coeficiente de variación; análisis de correlación; how to cite item mo, j., tian, x., & shen, w. (2021). tracing the source of heavy metal pollution in water sources of tourist attractions based on gis remote sensing. earth sciences research journal, 25(2), 207-214. doi: https://doi.org/10.15446/ esrj.v25n2.84631 to effectively prevent heavy metal pollution in water sources in tourist attractions, clarify the degree of control of heavy metal pollution sources, and improve the accuracy of tracing results, a gis-based remote sensing method of heavy metal pollution in tourist attractions is proposed. using gis spatial analysis method, the dem elevation data monitored by remote sensing is obtained, the watershed geographic information is compiled, and the gps obtains the longitude and latitude coordinates to locate the source of heavy metal pollution. the plug-in application framework is designed, and the watershed geographic information and plug-in application framework are integrated to build the pollution tracing platform. according to the mixing direction of pollutants after entering the water source, the migration and diffusion coordinate system of heavy metal pollution in the water source is established. the spatial-temporal distribution function model of heavy metal pollutants in water sources is constructed through the migration, transformation, and concentration of heavy metal pollutants in water sources. the tracing results of heavy metal pollution in water sources of scenic spots are obtained. the results showed that the order of variation coefficient of heavy metal pollution elements was cr > cd > cu > ni > zn > pb. the spatial distribution of heavy metal pollution elements was extremely uneven. there was a certain positive correlation between ni and cr, and the correlation coefficient of cu and zn was 0.78. the positive correlation was very significant, and the homology was very strong. moreover, the identification result of the proposed method is very close to the real value, which can accurately trace the source of heavy metal pollution in the water source of tourist attractions, with small tracing error and high accuracy of tracing result evaluation. abstract tracing the source of heavy metal pollution in water sources of tourist attractions based on gis remote sensing trazabilidad de la contaminación por metales pesados en fuentes de agua de atracciones turísticas a través del sistema de información geográfica issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n2.84631 para prevenir eficazmente la contaminación por metales pesados en las fuentes de agua en las atracciones turísticas, aclarar el grado de control de las fuentes de contaminación por metales pesados y mejorar la precisión de los resultados de rastreo, se propone un método de detección remota basado en sig de la contaminación por metales pesados en las atracciones turísticas. utilizando el método de análisis espacial sig, se obtienen los datos de elevación dem monitoreados por teledetección, se compila la información geográfica de la cuenca y se obtienen las coordenadas de longitud y latitud mediante gps para ubicar la fuente de contaminación por metales pesados. el marco de la aplicación de complemento está diseñado, y la información geográfica de la cuenca hidrográfica y el marco de la aplicación de complemento están integrados para construir la plataforma de rastreo de la contaminación. de acuerdo con la dirección de mezcla de los contaminantes después de ingresar a la fuente de agua, se establece el sistema de coordenadas de migración y difusión de la contaminación por metales pesados en la fuente de agua. a través de la migración, transformación y concentración de contaminantes de metales pesados en la fuente de agua, se construye el modelo de función de distribución espaciotemporal de los contaminantes de metales pesados en la fuente de agua y se obtienen los resultados de rastreo de la contaminación por metales pesados en la fuente de agua de los lugares escénicos. los resultados mostraron que el orden del coeficiente de variación de los elementos contaminantes por metales pesados fue cr> cd> cu> ni> zn> pb. la distribución espacial de los elementos contaminantes por metales pesados fue extremadamente desigual. hubo cierta correlación positiva entre ni y cr, y el coeficiente de correlación de cu y zn fue de 0,78. la correlación positiva fue muy significativa y la homología fue muy fuerte. además, el resultado de la identificación del método propuesto está muy cerca del valor real, que puede rastrear con precisión la fuente de contaminación por metales pesados en la fuente de agua de las atracciones turísticas, con un pequeño error de rastreo y una alta precisión en la evaluación del resultado del rastreo. resumen record manuscript received: 10/09/2019 accepted for publication: 19/01/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 2 (june, 2021): 207-214 jianghong mo1, xinling tian2, wei shen2,3,* 1. school of tourism management, guilin tourism university, guilin 541006 china 2. college of tourism management, enshi polytechnic, enshi 445000 china; 3. school of business administration, zhongnan university of economics and low, wuhan 430074 china; *corresponding author: 201701080025@stu.zuel.edu.cn r e m o t e s e n si n g https://doi.org/10.15446/esrj.v25n2.84631 https://doi.org/10.15446/esrj.v25n2.84631 https://doi.org/10.15446/esrj.v25n2.84631 mailto:201701080025@stu.zuel.edu.cn 208 jianghong mo, xinling tian, wei shen introduction at present, with the rapid development of the industrial economy and the continuous development of tourist attractions, the point pollution and surface pollution in the river basin and around the scenic spot are increasing day by day. the water source of the scenic spot is affected mainly by the precipitation and runoff process, water and soil loss in the upper reaches, and the pollution brought by the surrounding areas of the scenic spot, which makes the pollution load of the scenic spot increase sharply and the water quality of the scenic spot deteriorate (smith et al., 2018). due to the prominent problem of eutrophication of water sources in tourist attractions, people pay more attention to the study of nutrients in surface runoff, while ignoring the threat of other heavy metal substances. the potential harmfulness of heavy metal pollution has brought serious threat to the water quality of water sources in tourist attractions. the accumulation of heavy metals in water sources not only directly affect the physical and chemical properties of water sources, reduce the biological activity of water sources, hinder the effective supply of nutrients, but also enter the food chain and groundwater through plant absorption and leaching, posing a serious threat to the ecological environment and human health (liu et al., 2020). there is not only a problem of heavy metal pollution in surface water sources. there are prominent water quality problems in urban groundwater to varying degrees, and urban domestic pollution, industrial pollution, irrigation of farmland with untreated sewage and deterioration of geological conditions are the main causes of pollution (wang et al., 2018). therefore, the problem of heavy metal pollution in water sources of tourist attractions needs to be solved. in view of the prevention and control of heavy metal pollution in water sources, it is of great significance to accurately trace the source of heavy metal pollution in water sources of tourist attractions by using reasonable scientific research methods. tracing the source of heavy metal pollution in water sources of tourist attractions is to trace the sources of heavy metal pollutants in water sources of tourist attractions. it is the key technology and prerequisite for scientifically and accurately predicting the pollution of any water area. at present, great progress has been made in the study of pollution tracing. liu et al. (2019) proposed a method for tracing thallium pollution in sediments of the northern pearl river basin in south china. an improved irmm sequential extraction procedure for selected depth profiles showed that the enrichment of thallium in geochemical flow components was extremely high. the near-range quantitative source analysis using lead isotope fingerprint technology showed that along the depth profile 80%~90% of the thallium pollution is from lead-zinc smelting waste. the results emphasize the key role of smelting activities in the discharge of large amounts of geochemical flow thallium into the sediments. wang et al. (2018) proposed a method for tracing the source of heavy metal pollution in reservoirs in hilly areas of southern china, and studied the pollution characteristics, sources and health risks of soluble heavy metals in 19 representative reservoirs in hilly areas of southern china. the results show that due to the interference of human activities, the concentration of heavy metals in the studied reservoirs has great spatial heterogeneity. however, the pollution tracing error of the above methods is relatively large, and the evaluation accuracy of the tracing results is low. in view of the above problems, a method based on gis remote sensing is proposed to trace the source of heavy metal pollution in water sources of tourist attractions. using gis spatial analysis method, compiling remote sensing monitoring basin geographic information, using gps to obtain longitude and latitude coordinates, positioning source of heavy metal pollution, designing plug-in application framework, integrating watershed geographic information and plug-in application framework, and building pollution tracing platform. based on the analysis of the migration and transformation mechanism of heavy metal pollution in the water source of tourist attractions, the migration and diffusion coordinate system of heavy metal pollution in water sources is established. through the migration, transformation and concentration of heavy metal pollutants in water source, the spatial-temporal distribution function model of heavy metal pollution in water source is constructed, and the error of tracing results of heavy metal pollution in water source of tourist attractions is obtained. the identification results of the proposed method are very close to the real value, and can accurately trace the source of heavy metal pollution in the water source of scenic spots. the tracing error is small, and the evaluation accuracy of the tracing results is high. establishment of pollution tracing platform based on gis remote sensing technology gis remote sensing technology geographic information system gis is a computer technology system which takes the spatial data with geographical location attribute as the research object, takes the spatial database as the core, adopts the method of spatial analysis and spatial modeling, and provides various spatial and dynamic resources and environmental information timely. it is the product of the combination of traditional disciplines and modern science and technology, and a general technology for analyzing and processing massive spatial data (kong et al., 2018). it can be used for dynamic description of spatial information, simulate and demonstrate the development and change process of things through space-time modeling, so as to provide services for scientific research, management and decision-making. its biggest characteristic is that it can organically combine all kinds of information in social life with the graphic information reflecting geographical position, thus making it possible to solve complex spatial problems scientifically. global positioning system gps is a system that uses gps positioning satellites to locate and navigate the world in real time. it is called the global satellite positioning system. gps is a kind of satellite navigation system with omni-directional, all-weather, alltime and high-precision developed by the u.s. department of defense. it can provide global users with low-cost, high-precision three-dimensional position, speed and precise timing navigation information. it is an application model of satellite communication technology in the field of navigation. it greatly improves the information level of the earth society and effectively promotes the development of the digital economy. remote sensing technology remote sensing is a comprehensive technology of earth observation. in a broad sense, it refers to all non-contact remote detection, including the detection of electromagnetic field, force field, mechanical wave, acoustic wave, seismic wave, etc. in practical work, the detection of gravity, magnetic force, acoustic wave and seismic wave is classified as geophysical exploration, so only electromagnetic wave detection belongs to the category of remote sensing (ahmed et al., 2018). in the narrow sense, it refers to the comprehensive detection technology that the electromagnetic wave characteristics of the target are recorded from a distance without contact with the detection target by using the detection instrument, and the characteristics and changes of the object can be revealed through analysis. remote sensing technology can provide geometric and physical information of a large area and its surrounding environment in real time and quickly. the massive spectral information of ground observation provides accurate qualitative and quantitative data for target recognition and scientific law detection. platform design based on gis plug-in and application framework the construction of this platform adopts the design concept of application framework and is based on arc gis engine series products of arc gis company. the application framework has five main features: modularity, reusability, extensibility, simplicity and maintainability. the application framework can be logically divided into multiple levels and modules, so as to improve the aggregation of applications and reduce the coupling of applications. each independent module communicates and interacts through a unified pipeline or protocol. when a module changes significantly, as long as its exit and entrance remain unchanged, the effectiveness and reliability of the system will not be affected. extensibility is one of the significant characteristics of application framework, which means that the function of application framework has the ability to grow. the extensibility of framework enables us to realize different functions based on one platform and meet different functional needs. 209tracing the source of heavy metal pollution in water sources of tourist attractions based on gis remote sensing compiling watershed geographic information in order to improve the effect of trace analysis of heavy metal pollution in water sources of tourist attractions, it is necessary to use gis spatial analysis method to analyze water source direction through dem elevation data monitored by remote sensing, so as to obtain complete and detailed water source data. taking the water source basin of a tourist attraction as an example, the digital water source drainage network is extracted. the data source comes from the remote sensing elevation data obtained by the us space shuttle columbia when it takes off. the data source covers more than 80% of the world and the whole territory of china (xu et al., 2019). this article uses geographic information software for the arc gis series software released by esri, uses arc gis desktop version to process it, and stitches the two areas together to obtain a digital elevation map of a tourist attraction. use the hydrology hydrology analysis tool in spatial analyst tools in the arc gis desktop system toolbox to analyze, obtain the water source basin information of the entire tourist attraction, and complete the compilation of the basin geographic information. locate the source of heavy metal pollution in water sources in order to trace the sources of heavy metal pollution in water sources of scenic spots more quickly and conveniently, it is necessary to calibrate the data of heavy metal pollution sources in water sources of scenic spots on the basis of the compilation of watershed geographic information. the longitude and latitude coordinates obtained by gps are used to locate the source of heavy metal pollution in water sources, which provides theoretical support for the construction of water source heavy metal pollution tracing platforms. through the actual investigation, combined with the census data of pollution sources, the names and coordinates of pollution sources were investigated. by using the add xy data tool in arc gis tools, the coordinates in the map coordinate system were automatically captured, the information of pollution sources was displayed, and each pollution source was marked according to the name. the gis map built by the platform was obtained, and the location of heavy metal pollution sources in water sources was completed. design plug-in application framework because the whole pollution source tracing platform takes into account information query, pollution source tracing algorithm, geographic information positioning, decision support and other functions, its platform structure is complex and its requirements are changeable. therefore, application development is a highly targeted work. the module design, hierarchical division, language selection and system deployment of the program all depend on the specific needs of the end user. in this paper, the plug-in programming is based on the interface. the framework program is established. the attributes and methods of each excuse are analyzed one by one. after analyzing all the interfaces, the plug-in and the whole framework are linked to complete the whole framework. integrating gis plug-ins and frameworks in order to further maintain the extensibility of the plug-in framework design, the plug-in set is encapsulated. after the plug-ins are written, they are compiled into the form of dynamic link library. enter the dynamic link library file into the folder specified in the framework compilation. after the framework runs, it automatically traverses the dynamic link library plug-ins and interprets them on the framework to realize the corresponding functions. through the use of geographic information arc gis series software, the digital elevation map of a tourist attraction is obtained, and the hydrological analysis tool in arc gis is used to analyze the water source basin information of the whole scenic spot, and the geographic information of the river basin is compiled. based on the longitude and latitude coordinates obtained by gps, the source of heavy metal pollution in water source is located, and the plug-in application framework is designed. the plug-in set is encapsulated and compiled into the form of dynamic link library. the plug-in application framework is integrated to realize the construction of pollution tracing platforms. basic theory of heavy metal pollution in water sources pollution characteristics and toxicity of heavy metals in water sources heavy metals in aquatic ecosystems are distributed in various components of the system, which have an impact on the components of aquatic ecosystems. when the accumulation of heavy metals in organisms reaches a certain degree, the symptoms of injury will appear, physiological obstruction, development stagnation, and even death. at the same time, the structure and function of the entire aquatic ecosystem will be damaged or even collapsed (kang et al., 2019). the pollution characteristics are as follows: (1) trace heavy metals in natural water can cause biological poisoning. for example, the heavy metals cadmium and mercury can produce toxic effects only in the concentration range of 0.001-0.01mg/l. (2) heavy metals are enriched in organisms. the content of heavy metals in the food chain can be increased to tens of thousands of times in the last level of the food chain through the biological amplification of the food chain. therefore, heavy metals may cause greater harm to people and the environment. for example, when the concentration of mercury in seawater is 0.0001mg/l, the mercury content in plankton reaches 0.01~0.002 mg/l, that in small fish is 0.2~0.5 mg/l, and that in large fish is 1~5 mg/l. the mercury content in large fish is 10000-60000 times higher than that in seawater. (3) not only can heavy metals not be degraded by microorganisms, on the contrary, certain heavy metal elements can be converted into metal organic compounds under the action of microorganisms. for example, the methylation process of mercury is to convert inorganic mercury into organic mercury through bacterial strains. (4) there are various ways for heavy metals to enter the human body. for example, they can be ingested into the human body through breathing, drinking water, food, etc., which have toxic effects on human health. due to the cumulative effect of heavy metals in the human body, some heavy metals often take decades to realize their harmful effects on the human body (merinomascorro et al., 2018). in a word, heavy metals are potentially hazardous pollutants. the water source is a dynamic, open and complex multi-phase system. the water environment medium, water quality physical and chemical factors are easy to affect the water source. the solid-phase material contains a large number of clay minerals, organic matter, metal oxides and their hydrogen oxygenates that can adsorb and invade its interior. the harm of heavy metals to the water environment will not show in a certain period of time, but when the environmental conditions of the water source change or the accumulated amount of toxic substances exceeds its bearing capacity, all kinds of ligands in the water source will suddenly activate to form chelates and chelates, thus causing serious ecological harm. the harm of heavy metals in water can be divided into three categories: (1) it is harmful to aquatic organisms and organs. (2) algae, which are harmful to aquatic plants, play a very important role in the food chain of aquatic ecosystems. once algae absorb heavy metals, they will cause disorder of their physiological function, growth and metabolism. the decrease of cytochrome and inhibition of photosynthesis lead to cell distortion, tissue necrosis and even poisoning death, which changes the composition and species of algae in the natural environment. therefore, the sensitivity of aquatic organisms and algae is often used as a heavy metal pollution detector. (3) harmful to human health, heavy metals entering the human body are not easy to excrete and accumulate gradually, causing various and multi-level hazards to human health. its toxicological effects are mainly manifested in the influence on the normal development of the fetus, reproductive disorders, and the reduction of human quality. migration and transformation mechanism of heavy metal pollution in water source in order to effectively trace the source of heavy metal pollution in the water source of tourist attractions, it is necessary to study and analyze the migration and transformation mechanism of heavy metal pollutants in the water source of tourist attractions, master some laws of migration and diffusion 210 jianghong mo, xinling tian, wei shen of heavy metal pollutants in water sources of tourist attractions, and then, according to these laws, study and operate relevant mathematical tracing methods to realize the tracing and tracing of heavy metal pollution in water sources of tourist attractions. establishment of migration and diffusion coordinate system of heavy metal pollution in water source after a heavy metal pollution event occurs, the heavy metal pollutants in the water source need to move for a period of time before they can be completely mixed in the water source. that is, any type of pollution must go through the following three mixing stages in sequence during this period to achieve complete mixing with the water source body: (1) the mixing stage along the water depth in the water source, that is, the vertical mixing stage, has a range of influence called the near zone. the movement of pollutants in this area is particularly complex. the location of the pollution source, the form of discharge, and the hydraulics of the water body are the parameters. the main influencing factor of the mixing distance x1 in the direction of the water flow in the stage. (2) from the completion of the first stage to the fully mixed stage of the water source, that is, the mixing stage, the intensity of the pollution source, water area conditions, and hydrological factors are the main factors affecting the mixing distance x2 and concentration distribution of the water flow direction in this stage. (3) the area where the pollutants are fully mixed in the water source is the complete mixing stage, also known as the vertical mixing stage. the type and nature of the pollutants, the flow rate of the water source, and microorganisms are the main factors affecting the concentration changes in the area. it can be seen that after heavy metal pollutants enter the water source, they show vertical, horizontal and vertical mixing zones as figure 1. figure 1. various mixing stages after heavy metal pollutants enter the water source in figure 1, the migration and diffusion patterns of pollutants in each mixing region are different. in order to accurately describe the migration and diffusion of pollutants in water source, the coordinate system of pollutant migration and diffusion should be constructed according to the mixing direction of pollutants after entering the water source. for example, the direction of migration and diffusion of pollutants entering the water source in the lateral mixing zone is the x direction, that is, the x direction is defined as the direction of water flow along the water source (liu et al., 2018). the direction of migration and diffusion of pollutants in the longitudinal mixing zone is the y direction, that is, the y direction is defined as the direction perpendicular to the water surface. the direction of migration and diffusion of pollutants in the vertical mixing zone is the z direction, that is, the z direction is defined as the direction of gravity. through analysis, it is found that the pollutants are along the direction of the water flow (x direction). because the thrust of the water flow is dominant, the migration of the pollutants in this direction is mainly carried out with the water flow. in the horizontal plane perpendicular to the direction of water flow (y direction), since the concentration of pollutants has a gradient, in this direction pollutants mainly migrate from areas with high concentrations to areas with low concentrations. along the direction of gravity (z direction), due to the exchange of momentum, heat and mass between the pollutant and the water source, the migration and diffusion in this direction is more complicated, and the pollutant migration speed is affected by the size of the contact area between the pollutant and the water source. migration and transformation of heavy metal pollution in water source colloidal substances existing in natural water sources are deposited in suspended particulate matter of water sources. the main colloidal substances include organic and inorganic, including inorganic polymer compounds composed of hydrated oxides and soil minerals, and organic polymer compounds composed of organic matter, humus and protein (jia et al., 2020). colloid has a huge specific surface area. because of the surface energy and charge, a large number of ions can be adsorbed on the surface of various colloidal particles, such as pesticides and heavy metal elements in the water source, and a variety of physical and chemical reactions occur at the solid-liquid interface, which makes heavy metals transfer from aqueous phase to solid phase. therefore, the migration and transformation of heavy metal ions are closely related to the existence of colloidal substances. the migration and transformation of heavy metal ions transferred from liquid phase to solid phase water source are closely related to organic carbon, sulfide, ph, redox potential, carbonate, particle size, and primary minerals in sediment. under the influence of these conditions, it has adsorption, coordination, precipitation and dissolution with the water source and the substances in the water source to realize the transformation of heavy metal ions in the water source. therefore, heavy metal ions in the water source have complicated physical and chemical reactions. therefore, the pollutants in the water source have the characteristics of position change and attenuation reaction in various forms under the joint action of water flow, gravity and various resistances, including push flow migration, dispersion dilution and transformation and migration: (1) push flow migration: also known as flow migration, refers to the behavior of pollutants moving in the x, y, z direction under the action of the water body after entering the water body. however, this migration behavior of pollutants only changes its spatial position in the water body, and does not reduce its concentration value. in general, if the pollutants in the water source only migrate with the flow, the product of the concentration of pollutants in the water body and the flow velocity component in the x, y, z direction is equal to the pollutant migration flux in the x, y, z direction, namely: p u c p u c p u cx x y y z z  , , (1) in formula (1), px , py , pz represents the flux of pollutants in the x, y, z direction, ux , uy , uz represents the water velocity component in the x, y, z direction, and c represents the concentration of pollutants. (2) dispersion and dilution: due to the uneven flow in all directions of the water source and the large gradient of pollutants entering the water source, the pollutants entering the water source will not only migrate in the flow direction, but also have diffusion behaviors such as molecular diffusion, turbulent diffusion or shear flow dispersion (liu et al., 2019). therefore, any type of pollutant entering the water source exists dispersion dilution, that is, dispersion dilution is a basic movement form of pollutant in water source. molecular diffusion: diffusion refers to the phenomenon that pollutants migrate to homogenization due to the existence of gradient, which is a common phenomenon after pollutants enter the water body. among them, the molecular diffusion of pollutants refers to the phenomenon of random dispersion of pollutants caused by the irregular movement of pollutant molecules, that is, when the concentration of pollutants in water is not completely mixed, the pollutant molecules will move from the high concentration area to the low concentration area. therefore, the molecular diffusion process of pollutants in water source obeys fick’s first law: i e c x i e c y i e c zx m y m z m =− ∂ ∂ =− ∂ ∂ =− ∂ ∂ , , (2) 211tracing the source of heavy metal pollution in water sources of tourist attractions based on gis remote sensing in the formula (2), ix , iy , iz represents the molecular diffusion flux in the x, y, z direction, c̄ represents the concentration of heavy metal pollutants in the water source, em represents the molecular diffusion coefficient of the pollutants, and the partial derivative term on the right side of the equation represents the concentration gradient along the x, y, z direction. the negative sign indicates that the diffusion of pollutant molecules is consistent with the negative concentration gradient. turbulent diffusion: turbulence refers to the random pulsating diffusion of various physical quantities contained in any pollutant point in the water flow field with time variation. turbulent diffusion, also known as turbulent diffusion, refers to the phenomenon of pollutant migration from high concentration area to low concentration area caused by turbulent field of water body. its size is closely related to the whirling motion degree of water flow. therefore, the turbulent diffusion law of pollutants in water source can be expressed by fick’s first law: u e c x u e c y u e c zx x y y z z =− ∂ ∂ =− ∂ ∂ =− ∂ ∂ , , (3) in formula (3), ux , uy , uz represents the turbulent diffusion flux of pollutants along the x, y, z direction, and ex , ey , ez represents the turbulent diffusion coefficient of pollutants along the x, y, z direction. shear flow dispersion: the pollutants entering the flowing water body are usually transported along with the water flow, and the dispersion coefficient is closely related to the average cross-section velocity and pollutant concentration (ren et al., 2020). however, due to the influence of perimeter and water viscosity, the concentration and velocity distribution of pollutants on the same section are often non-uniform, that is, there will be shear stress and velocity gradient in the water source. similarly, the first law can be used to characterize the mass flux of pollutants produced by dispersion: o d c x o d c y o d c zx x y y z z =− ∂ ∂ =− ∂ ∂ =− ∂ ∂ , , (4) in formula (4), ox , oy , oz respectively represents the pollutant diffusion flux generated by dispersion in the x, y, z direction, and dx , dy , dz represents the dispersion coefficient along the x, y, z direction. the difference between molecular diffusion and turbulent diffusion is that only after pollutants enter the water source does the shear flow dispersion occur, and the coefficient magnitude is generally 10~104m2/s. (3) transformation and transportation: generally speaking, the pollutants that enter the water body are usually adsorbed by suspended substances such as sediment and aquatic organisms in the water body. the role of the water body makes the pollutant concentration in the water body change in time and space by means of aggregation, precipitation and resuspension, that is, the pollutant has a transformation and transportation effect in the water body type. generally, different types of pollutants will be transformed and transported in different ways after entering the water source. for example, particulate heavy metal pollutants generally move with water in the form of ions, molecules or colloids dissolved in water. dissolved heavy metal pollutants are transformed and transported through food chain. the insoluble pollutants such as na, k, ci and f can only migrate with the flow because of no transformation and attenuation. speciation and bioavailability of heavy metal pollution in water source usually, the toxicity and bioavailability of heavy metals in water sources are studied together. the toxicity of heavy metals is closely related to the speciation of heavy metals in water sources. therefore, the bioavailability of heavy metals in water sources also depends on their speciation. due to the variety of heavy metal forms, some of the heavy metals in the water source are easy to form insoluble hydroxides, carbonates and sulfides, and some of them can be transferred into the bottom water source through adsorption and polymerization with coarse particles. therefore, the existing forms of heavy metals in water sources mainly include the following two aspects: (1) the form of heavy metals in natural water and the form of heavy metals in water sources. water source water source can not only provide nutrients for aquatic organisms, but also store toxic substances. as an important part of the water environment, it is not only a carrier of water source pollutants, but also an indicator of water source pollution. (2) the environmental behavior and toxic effects of heavy metals are not only related to the content of heavy metals, but also depend on the existing forms, migration and transformation characteristics and bioavailability of heavy metals in environmental media. therefore, in the process of studying the pollution of heavy metals in water sources, simply analyzing the content of heavy metals can not truly reflect the pollution degree of heavy metals. combined with its morphological analysis, the pollution level is comprehensively considered (sun et al., 2018). through the analysis of the form of heavy metals to study its geochemical process and toxic effect has a great role in promoting, based on this to evaluate its potential migration law and environmental hazards, so that the toxicity, bioavailability, chemical activity and re migration of heavy metals can be better revealed. there are different forms of heavy metals in water sources, and their migration ability and toxic effects are quite different. exchangeable state is easily dissolved in water or exchanged by other polar ions when environmental conditions change. because of the instability of its existing form, it is an important part of water quality. carbonate bound state is especially easy to re-enter the water source when the water source environment changes. the strong oxidation of iron will be partly released into the water source due to environmental changes. the organic state is relatively stable and is not easy to be absorbed and utilized by organisms. the residual form is the most stable of all forms, and it is not easy to release into the water environment for a long time. for organisms, the toxicity of metal elements in dissolved state is the largest, while that of particulate metal is relatively small. if a kind of compound which is completely insoluble in water can be formed and firmly bound in the water source, it will be safer and more ideal. tracing method of heavy metal pollution in water source taking the heavy metal pollution incident of water sources in tourist attractions as an example, a mathematical model that can grasp the relevant information of pollution sources is established, that is, tracing the source of heavy metal pollution in water sources of tourist attractions as figure 2. figure 2. source tracing of heavy metal pollution in water sources of scenic spots if the geometry of the water source canal remains unchanged, a number of observation points qj are arranged in the calculation area [0, l], and q(t ) represents the pollutant concentration distribution function observed during the t period. therefore, the tracking and tracing of such sudden water pollution is how to determine the model parameters, the location of the pollution source, the time of discharge and the intensity based on the known information. the migration and transformation equation of pollutants in sudden water pollution incidents can be approximated to: ∂ ∂ =− ∂ ∂ + ∂ ∂ ∂ ∂      + + c t u c x x c x s cλ δ (5) in formula (5), u represents the average water flow velocity of the water source,  represents the longitudinal dispersion coefficient of the water source,  represents the attenuation coefficient of the pollutant in the water source, and s represents the source term of the pollutant in the water source. 212 jianghong mo, xinling tian, wei shen if f  0 is used to represent the location of the pollution event, and the water source area [0, l] is taken as the research object, the pollutants entering the water source section will move with the flow, longitudinal dispersion and attenuation under the combined action of water flow and shear force. if it is assumed that the pollutant concentration of the upstream inflow and downstream outflow of the research object are both 0, then: c x x l c t s t t t c l t t t , , , 0 0 0 0 0 0 0 ( )= < < ( )= ( ) < < ( )= < <         (6) in formula (6), s(t ) represents the intensity of pollutants entering the canal. to construct a water source heavy metal pollution tracing model based on the above formula, that is, the time and space distribution function model of heavy metal pollutants in water sources is: c x t s t x t x u t t t t , exp( )= ( ) −( ) − − −( )  −( ) − −( )   ∫0 32 1 1 4 1 1 πλ λ δ   dt (7) in addition, due to the accident site, monitoring equipment, sampling, etc., there is an observation error in the pollutant concentration observation sequence value q(t ), and this error plays a decisive role in the accuracy of the tracing result. therefore, the pollutant concentration time series value of the observation point can be expressed as: q q q j mj noisy j exact j exact, , , , ,...,= + =εω 1 2 (8) in the formula (8), qj, noisy , qj, exact represents the observation value with error and the accurate observation value,  represents the error level,  represents a set of random numbers conforming to the standard distribution, and m represents the number of observations. the error of the tracing result of heavy metal pollution in this type of water source can be expressed as: e n s ss e k t k k n = ( ) −( )  = ∑ 1 2 1 (9) in formula (9), es represents the error value of water source heavy metal pollution tracing results, st represents the actual value of water source heavy metal pollution tracing, se represents the tracing result of water source heavy metal pollution, and n represents the output number of water source heavy metal pollutant emission intensity and water source heavy metal pollutant concentration spatial and temporal distribution. through the above steps, according to the mixing direction of pollutants entering the water source, the migration and diffusion coordinate system of heavy metal pollution in the water source is established. through the migration, transformation and concentration of heavy metal pollutants in the water source, the spatial-temporal distribution function model of heavy metal pollutants in the water source is constructed, and the tracing error of heavy metal pollution in the water source of tourist attractions is obtained. experimental analysis overview of the study area the research area is a tourist attraction with a total water source of about 6km. the terrain belongs to the typical low mountain and hilly area. the terrain in the study area fluctuates greatly. the general trend of the terrain is high in the southeast and low in the northwest. the basic landform is erosion erosion landform, denudation landform and accumulation landform. the average altitude of the scenic spot is 1000~1500m, and the total area of the scenic spot is 16,400 km2. the climate of the tourist attraction is a typical warm temperate zone and semi-humid continental monsoon climate with four distinct seasons. summer is hot and rainy, winter is cold and dry, and spring and autumn are short. the annual average temperature is 13.6 °c, the annual frost-free period is 180~200 days, and the annual average rainfall is 624.8mm. data acquisition and processing in this study, spss13.0 was used for basic statistical analysis, gs7.0 was used to analyze the spatial variability of heavy metal content data. spatial interpolation module of arc gis desktop 9.2 software was used to make and output the distribution of heavy metal index sample points in water source, as shown in figure 3. figure 3. distribution of sample points of heavy metal index content in water sources the samples were dried in a vacuum freeze dryer. the dried samples were filtered with a 4-mesh standard screen to remove the large stones, garbage and other debris, and then sealed in a dry and clean self sealing bag. put the appropriate amount of sample into agate mortar and grind it through 100 mesh sieve. mix the screened samples evenly, put them into self sealed bags and label them. after that, the samples were placed in a refrigerator at 4 °c and kept away from light to avoid the degradation of light and microorganisms and ensure the reliability of the samples. take a proper amount of treated sample, continue to grind and make it pass through 200 mesh screen. these samples are used to determine the speciation of heavy metals. analysis of human disturbance on heavy metal elements coefficient of variation is a variable that describes the spatial distribution. through the study of the coefficient of variation, we can understand the volatility of heavy metal content changes, and then analyze the degree of heavy metal pollution by human beings. the larger the coefficient of variation is, the greater the degree of heavy metal pollution is affected by human beings. that is to say, heavy metals with obvious external pollution can be identified and analyzed. its calculation formula is as follows: c s xv d (10) in formula (10), sd represents the standard deviation of the sampled sample, and x represents the sample mean. the classification of the degree of variation is shown in table 1. 213tracing the source of heavy metal pollution in water sources of tourist attractions based on gis remote sensing table 1. classification of variation degree coefficient of variation degree of variation <15% low variation 15%~36% moderate variation >36% high variation the coefficient of variation of heavy metal elements in water sources of tourist attractions is as table 2. table 2. variation coefficient of heavy metal elements in water sources of tourist attractions heavy metal elements coefficient of variation degree of variation zn 56.24% high variation cd 96.29% high variation cu 66.54% high variation cr 16.87% moderate variation ni 43.09% high variation pb 58.35% high variation according to the data in table 2, the order of variation coefficient of heavy metal elements in water sources of tourist attractions is cd > cu > pb > zn > ni > cr. except for cr, the other metal elements are highly variable, especially cd, with a coefficient of variation of 96.29%, indicating that its spatial distribution is extremely uneven and highly interfered by human beings. the variation coefficient of heavy metal pollution elements in water sources of tourist attractions is as table 3. the order of variation coefficient of heavy metal pollution elements in water sources of tourist attractions is cr > cd > cu > ni > zn > pb. except for pb, the variation of other elements is high. the variation degree of ni and cu is 56%, while that of cr and cd is about 70.5%. therefore, ni-cu and cr-cd have certain similarities in spatial distribution. compared with the data in table 2 and table 3, the results of coefficient of variation of cr and pb are quite different. in the analysis of the coefficient of variation of heavy metal elements in the water source of tourist attractions, cr belongs to moderate variation, while in the variation coefficient table of heavy metal pollution elements in tourist attractions, it is highly variable. the coefficient of variation of heavy metals in water sources of tourist attractions is 58.35%, but the variation coefficient of heavy metals pollution in water sources of tourist attractions is the lowest, only 32.24%. the results show that the concentration of pb in the sediment of water source polluted by heavy metals in scenic spots is relatively close, and the concentration of pb does not exceed the national secondary standard and remains at a low level. it can be concluded that the discharge of heavy metal pollution in the water source of tourist attractions is not the main source of pb pollution. the reasons for the difference are: based on the principle of average distribution, the sampling results reflect the overall pollution status of water sources in a long period of time, while for the heavy metal pollution of water sources in tourist attractions, the samples are taken after field investigation. the nature of heavy metal pollution elements in water sources of tourist attractions depends on the types of pollution sources. therefore, the results of the coefficient of variation are quite different. table 3. variation coefficient of heavy metal pollution elements in water sources of tourist attractions heavy metal elements coefficient of variation degree of variation pb 32.24% moderate variation ni 56.48% high variation cu 56.87% high variation zn 46.09% high variation cr 73.62% high variation cd 68.35% high variation correlation analysis of heavy metal elements generally, there is a significant correlation between heavy metals as one of the sources of heavy metals. therefore, in order to find homologous heavy metals, it is necessary to carry out correlation analysis. the specific steps of correlation analysis are as follows: (1) perform z-score standardization on all data first to prevent interference with the analysis results due to the difference in concentration of different types of heavy metals. (2) the correlation analysis of heavy metals and organic matter content (loi) was carried out. (3) in order to ensure the accuracy of the correlation analysis results, the bilateral significance level test was carried out on the correlation analysis results. the results of correlation analysis of heavy metal elements are as table 4. the results of the significance level test are as table 5. table 4. correlation analysis results of heavy metal elements z-score (ni) z-score (cu) z-score (zn) z-score (cr) z-score(cd) z-score(loi) z-score(ni) 1.000 z-score(cu) -.045 1.000 z-score(zn) -.045 .768 1.000 z-score(cr) .517 -.287 -.405 1.000 z-score(cd) .058 .187 -.108 .385 1.000 z-score(loi) .479 .139 .167 -.175 .195 1.000 table 5. significance level test results z-score(ni) z-score(cu) z-score(zn) z-score(cr) z-score(cd) z-score(ni) z-score(cu) .872 z-score(zn) .839 .002 z-score(cr) .047 .294 .135 z-score(cd) .848 .503 .713 .165 z-score(loi) .093 .634 .578 .548 .487 214 jianghong mo, xinling tian, wei shen according to the results in table 4 and table 5, there is a certain positive correlation between ni and cr, the correlation coefficient is 0.517, and the p value in the test results is 0.048 (< 0.05), so the two are significantly positive correlation. the correlation coefficient of cu and zn is 0.78, and their significance level is 0.01 (< 0.05), which means that the positive correlation between cu and zn is extremely significant and the homology is very strong. there is a positive correlation between ni and organic matter content, but their significance level is 0.095 (> 0.05), which indicates that organic matter content has a certain influence on ni, but it is not the main influencing factor. trace error analysis of heavy metal pollution in order to verify the effectiveness of the tracing method of heavy metal pollution in tourist attractions based on gis remote sensing, it is assumed that the error follows the gaussian distribution   0.01, and the pollution source parameter g(x1, x2, s1, s2) is uniformly distributed and independent of each other. among them, x1, x2 obeys uniform distribution h (0,100), s1, s2 obeys uniform distribution h (0,10). using the proposed method, the liu et al. (2019) method and the wang et al. (2018) method for iterative calculation, the tracing errors of different methods are obtained as table 6. according to the data in table 6, the absolute error, relative error and standard deviation of the proposed method are less than 0.2%, 0.4% and 0.15 respectively, and the absolute error, relative error and standard deviation of pollution source intensity are less than 0.02%, 0.4% and 0.06 respectively, and the tracing value is in good agreement with the real value. it can be seen that compared with the liu et al. (2019) method and the wang et al. (2018) method, the tracing identification result of the proposed method is very close to the true value, and the accuracy of the prior distribution of the unknown parameter is higher under the same random sampling frequency. the tracing error is smaller, and the tracing assessment accuracy is higher. conclusion this paper proposes a method based on gis remote sensing to trace the source of heavy metal pollution in the water source of tourist attractions, which can accurately trace the source of heavy metal pollution in the water source of tourist attractions, with small tracing error and high accuracy of tracing results evaluation. however, it has not formed a clear definition of the trace source of heavy metal pollution in water sources, and lacks a set of systematic and complete tracing theory. therefore, a clear and complete tracing theory is proposed in the next research to provide more accurate theoretical and methodological support for the tracing and tracing of heavy metal pollution in water sources. references ahmed, w., lobos, a., senkbeil, j., peraud, j., gallard, j., & harwood, v.j. (2018). evaluation of the novel crassphage marker for sewage pollution tracking in storm water outlets in tampa, florida. water research, 131(15), 142. doi: 10.1016/j.watres.2017.12.011 jia, x., fu, t., hu, b., shi, z., & zhu, y. (2020). identification of the potential risk areas for soil heavy metal pollution based on the source-sink theory. journal of hazardous materials, 393(2), 122424. https://doi. org/10.1016/j.jhazmat.2020.122424 kang, m. j., kwon, y. k., yu, s., lee, p. k., park, h. s., & song, n. (2019). assessment of zn pollution sources and apportionment in table 6. trace error of heavy metal pollution by different methods the proposed method the reference [4] method the reference [5] method x1 x2 s1 s2 x1 x2 s1 s2 x1 x2 s1 s2 true value 78 36 8.6 4.5 78 36 8.6 4.5 78 36 8.6 4.5 recognition value 78.8 35.9 8.61 4.53 72.8 32.5 7.91 4.23 69.8 31.9 7.68 3.86 absolute error 0.08 0.10 0.01 0.02 0.15 0.26 0.07 0.08 0.23 0.29 0.09 0.10 relative error /% 0.10 0.29 0.12 0.38 0.19 0.32 0.16 0.45 0.21 0.37 0.19 0.54 sample standard deviation 0.12 0.10 0.05 0.06 0.17 0.13 0.09 0.08 0.20 0.17 0.12 0.13 agricultural soils impacted by a zn smelter in south korea. journal of hazardous materials, 364(15), 475-487. https://doi.org/10.1016/j. jhazmat.2018.10.046 kong, j., guo, q., wei, r., strauss, h., zhu, g., li, s., song, z., chen, t., song, b., zhou, t., & zheng, d. (2018). contamination of heavy metals and isotopic tracing of pb in surface and profile soils in a polluted farmland from a typical karst area in southern china. science of the total environment, 637-638(1), 1035-1045. https://doi.org/10.1016/j. scitotenv.2018.05.034 liu, b., wang, j., xu, m., zhao, l., & wang, z. (2019). spatial distribution, source apportionment and ecological risk assessment of heavy metals in the sediments of haizhou bay national ocean park, china. marine pollution bulletin, 149(11), 110651.1-110651.8. https://doi. org/10.1016/j.marpolbul.2019.110651 liu, j., wang, j., tsang, d. c. w., xiao, t., chen, y., & hou, l. (2018). emerging thallium pollution in china and source tracing by thallium isotopes. environmental science & technology, 52(21), 11977-11979. https://doi.org/10.1021/acs.est.8b05282 liu, j., yin, m., luo, x., xiao, t., wu, z., li, n., wang, j., zhang, w., lippold, h., belshaw, n.s., feng, y., & chen, y. (2019). the mobility of thallium in sediments and source apportionment by lead isotopes. chemosphere, 219(3), 864-874. https://doi.org/10.1016/j. chemosphere.2018.12.041 liu, y., gao, t., xia, y., wang, z., liu, c., li, s., wu, q., qi, m., & lv, y. (2020). using zn isotopes to trace zn sources and migration pathways in paddy soils around mining area. environmental pollution, 267(9), 115616. https://doi.org/10.1016/j.envpol.2020.115616 merino-mascorro, j. a., hernandez-rangel, l. g., heredia, n., & garcia, s. (2018). bacteroidales as indicators and source trackers of fecal contamination in tomatoes and strawberries. journal of food protection, 81(9), 1439-1444. doi: 10.4315/0362-028x.jfp-18-073 ren, y., guo, x., & ma l. (2020). feature extraction simulation of contaminated soil radar image based on enhancing operator. computer simulation, 37(04), 5-8, 61. smith, s. d. a., banister, k., fraser, n., edgar, r. j. (2018). tracing the source of marine debris on the beaches of northern new south wales, australia: the bottles on beaches program. marine pollution bulletin, 126(1), 304-307. doi: 10.1016/j.marpolbul.2017.11.022 sun, j., yu, r., hu, g., su, g., & zhang y. (2018). tracing of heavy metal sources and mobility in a soil depth profile via isotopic variation of pb and sr. catena, 171, 440-449. https://doi.org/10.1016/j.catena.2018.07.040 wang, q., chen, q., yan, d., & xin, s. (2018). distribution, ecological risk, and source analysis of heavy metals in sediments of taizihe river, china. environmental earth sciences, 77(16), 569. https://doi.org/10.1007/ s12665-018-7750-6 wang, x., zhang, l., zhao, z., & cai, y. (2018). heavy metal pollution in reservoirs in the hilly area of southern china: distribution, source apportionment and health risk assessment. science of the total environment, 634(1), 158-169. doi: 10.1016/j.scitotenv.2018.03.340 xu, d., wang, r., wang, w., ge, q., zhang, w., chen, l., & chu, f. (2019). tracing the source of pb using stable pb isotope ratios in sediments of eastern beibu gulf, south china sea. marine pollution bulletin, 141(4), 127-136. https://doi.org/10.1016/j.marpolbul.2019.02.028 https://xueshu.baidu.com/s?wd=author:(j peraud) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(j gallard) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person 10.1016/j.watres https://xueshu.baidu.com/s?wd=author:(z shi) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1016/j.jhazmat.2020.122424 https://doi.org/10.1016/j.jhazmat.2020.122424 https://xueshu.baidu.com/s?wd=author:(pk lee) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(hs park) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(n song) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1016/j.jhazmat.2018.10.046 https://doi.org/10.1016/j.jhazmat.2018.10.046 https://www.researchgate.net/scientific-contributions/guangxu-zhu-2142964878 https://www.researchgate.net/profile/si-liang_li https://www.researchgate.net/scientific-contributions/tongbin-chen-38483224 https://doi.org/10.1016/j.scitotenv.2018.05.034 https://doi.org/10.1016/j.scitotenv.2018.05.034 https://xueshu.baidu.com/s?wd=author:(z wang) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1016/j.marpolbul.2019.110651 https://doi.org/10.1016/j.marpolbul.2019.110651 https://xueshu.baidu.com/s?wd=author:(t xiao) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(y chen) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(l hou) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1021/acs.est.8b05282 https://xueshu.baidu.com/s?wd=author:(t xiao) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(z wu) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(n li) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(j wang) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(w zhang) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(h lippold) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://xueshu.baidu.com/s?wd=author:(ns belshaw) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1016/j.chemosphere.2018.12.041 https://doi.org/10.1016/j.chemosphere.2018.12.041 https://www.researchgate.net/profile/chengshuai_liu https://www.researchgate.net/scientific-contributions/shehong-li-2182338477 https://www.researchgate.net/scientific-contributions/qiqi-wu-2172463348 https://doi.org/10.1016/j.envpol.2020.115616 https://xueshu.baidu.com/s?wd=author:(s garcia) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person 10.4315/0362-028x.jfp https://xueshu.baidu.com/s?wd=author:(edgar, robert j) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person 10.1016/j.marpolbul https://xueshu.baidu.com/s?wd=author:(y zhang) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1016/j.catena.2018.07.040 https://xueshu.baidu.com/s?wd=author:(s xin) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person https://doi.org/10.1007/s12665-018-7750-6 https://doi.org/10.1007/s12665-018-7750-6 https://xueshu.baidu.com/s?wd=author:(y cai) &tn=se_baiduxueshu_c1gjeupa&ie=utf-8&sc_f_para=sc_hilight=person 10.1016/j.scitotenv https://doi.org/10.1016/j.marpolbul.2019.02.028 remote sensing quantitative retrieval of ecological water (eco-water) has been foundational in systemic and quantitative research for water resources. eco-water resource levels indicate conservation ability for the eco-water layer and influence of this on precipitation transformation and runoff regulation. the remote sensing quantitative inversion retrieved the mec (modulus of eco-water conservation) of the upper minjiang river basin study area in 1994 and 2001, and combined with climate data between 1990 and 2005, the influence of conservation water on the eco-water layer on runoff was then analyzed. results revealed significant efficacy for flood control and water supply during the drought from the hydrologic cycle of ecowater. thus protection and restoration of the eco-water layer for flood and drought prevention are crucial. el sondeo remoto del agua ecológica (del inglés eco-water, agua conservada en la superficie terrestre) es indispensable en la investigación sistemática y cuantitativa de las fuentes de agua. los niveles de suministros de agua ecológica indican la capacidad de conservación de la capa de agua ecológica y la influencia de esta en la transformación de precipitación y la regulación de escorrentía. la inversión cuantitativa por sondeo remoto estableció el módulo de conservación de agua ecológica (mec, del inglés modulus of eco-water conservation) para el área de estudio en la cuenca alta del río minjiang entre 1994 y 2001, y combinada con la información climática de entre 1990 y 2005, se analizó la influencia de conservacion de agua en la capa ecoacuática. los resultados mostraron una gran eficacia en el control de inundaciones y en el suministro de agua durante la sequía a lo largo del ciclo hidrológico. por esta razón, la protección y restauración de la capa de agua ecológica para la prevención de inundaciones y sequía es necesaria. earth sciences research journal earth sci. res. j. vol. 20, no. 3 ( september, 2016 ) : e1 e6 abstract resumen keywords: eco-water, remote sensing quantitative retrieval, upper minjiang river basin, runoff palabras clave: agua ecológica, sondeo remoto cuantitativo, cuenca alta del río minjiang, escorrentía. the influence of eco-water retrieved by quantitative remote sensing on runoff in upper minjiang river basin record manuscript received: 15/01/2016 accepted for publication: 21/11/2016 how to cite item huang, j., yang, w., yang, x., & deng, b. (2016). the influence of eco-water retrieved by quantitative remote sensing on runoff in upper minjiang river basin. earth sciences research journal, 20 (2), e1-e6. doi:http://dx.doi.org/10.15446/esrj.v20n2.55177 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.55177 jin huang, wunian yang*, xin yang*, bing deng key laboratory of geoscience spatial information technology, ministry of land and resources of the p.r. china, chengdu university of technology, chengdu, sichuan, 610059, p. r. china correspondent authors: wunian yang and xin yang equally contributed to this work e-mail: jin huang (first author) jinjin0522@163.com wunian yang (correspondent author) ywn@cdut.edu.cn xin yang (correspondent author) johnny0701@qq.com bing deng dengbing2000@163.com h y d r o l o g y influencia del agua ecológica en la escorrentía de la cuenca alta del río minjiang medida a través de teledetección cuantitativa jin huang, wunian yang, xin yang, bing denge2 1 introduction the concept of “eco-water” was introduced by yang wunian et al. in 2001, to investigate the impact of water conserved on the earth’s surface in forest vegetation on the hydrologic cycle and the local ecological environment (yang et al., 2001; wan et al., 2004). eco-water is water closely related to the earth’s surface vegetation and is conserved by the vegetation, the vegetation humus layer and the root soil layer after conversion of the atmospheric precipitation forms the “ecological water layer” centered on vegetation. ecowater is not directly involved in the hydrological cycle process yet significantly contributes as a buffer and runoff regulator as rainfall is intercepted and groundwater recharged in the eco-water layer. eco-water resource levels are difficult to extract and quantify with conventional methods. quantitative remote sensing inversion theory and algorithm, however, facilitate quantification as developments in high sensitivity remote sensing imaging related to vegetation and soil have occurred in recent years (li et al., 2009; jian et al., 2012). runoff influence resulting from the vegetation canopy, vegetation humus layer, and the soil layer has consistently presented challenges in hydrological research. extensive hydrological effect experimentation has followed involving respective sample regions producing substantial runoff information (liu et al., 1989; ogee and brunet, 2002; rao et al., 2005; hu et al., 2014). causative mechanisms from the influence of vegetation canopy, vegetation humus layer and soil layer on runoff are exhibited independently, yet interaction effectiveness is the same—conserving water, buffering and adjusting flow. the eco-water layer then utilizes the three combined as an “independent converter” in the hydrologic cycle. the capability of converting eco-water quantitatively, as expressed by the remote sensing quantitative retrieval method, simplifies water resource quantification while concurrently contributing to increased practicality for determining the influence of eco-water on runoff. 2 eco-water remote sensing quantitative retrieval 2.1 study area and data upper minjiang river basin, located in the transition region from eastern qinghai-tibet plateau to sichuan basin, is an essential tributary to the upper yangtze river and also a vital water conservation area for the chengdu plain. vegetation of the upper minjiang river has been referred to as the “natural barrier and green reservoir of the land of abundance”, but due to massive deforestation for hundreds of years, deterioration of this ecological environment has created severe reductions in water conservation capacity as forest vegetation has decreased and lead to reduction in dry season runoff. the relation between spectral data and biochemical parameters of vegetation and soil should be determined respectively for the mec modeling. different plants and soil specimens were collected and biochemical parameters then measured. the water content of vegetation and soil was computed from the biochemical parameters, while the spectral data was acquired with field spec pro fr, the full spectrum portable spectrometer of asd (analytical spectral devices). landsat5 tm images were utilized in this study with the product type referred to as “l1t”, provides systematic radiometric and geometric accuracy by incorporating group control points while employing a dem (digital elevation model) for topographic accuracy. the study area consists of three scenes; the ranks numbers are 13037, 13038 and 13039. preprocessing is required before quantitative image retrieval: mosaic, subset, radiation calibration and atmospheric correction. image quality and imaging time for eco-water remote sensing quantitative retrieval are critical factors for accuracy. cloud, snow, and fog appearance in images will, for example, significantly interfere with the ecological water quantitative inversion if not controlled within 10%. consistent seasons should be compared to the amount of eco-water resources in different years as vegetation and precipitation vary among seasons producing large differences in vegetation and soil water content. images for mec retrieval were acquired on june 26, 1994, and on june 13, 2001, in this study. two phase images are with high quality and similar imaging time that makes inversion results more accurate and comparable. 2.2 modulus of eco-water conservation modeling the amount of water conserved by the eco-water layer was represented by the modulus of eco-water conservation (mec), an integral parameter of ecological water. mec is related to vegetation and soil water content and represents the water conservation ability of the earth’s surface based on vegetation. remote sensing technology with high sensitivity and accessibility allowed to extract the vegetation and soil information. thus quantification of each was a precondition for calculating mec. the basic research unit of remote sensing image digital processing is a single pixel with mec following by also utilizing a single pixel as the primary research unit. according to the definition of mec that the mec model of pixel (i,j) is as follows: (1) where ewtc(i,j) is vegetation water content of the pixel(i,j), w(t)(i,j) is soil water content of the pixel(i,j). if there are m lines and n rows of the remote sensing image of the study area, the amount of ecological water resource is: (2) the principle of remote sensing quantitative inversion is the reflection spectrums of surface features exhibit particular characteristics (zhang et al., 2010). vegetation water content has an obvious response to spectral characteristics in the wavelength range from near infrared to intermediate infrared. 1600nm indicates a high moisture absorption band while 820nm indicates a weak moisture absorption band, allowing the vegetation index sr (simple ratio) established by the ratio effectively to highlight the vegetation water content information (zhang et al., 2010; wang et al., 2011; yi et al., 2012). the modeling process of vegetation water content is demonstrated in figure 1. advantages and disadvantages of soil moisture estimation models with optical sensor differ, and after analyzing the characteristics and regression effects with actual soil water content measurement, smmrs (soil moisture monitoring by remote sensing) based on soil line was chosen. this is a straight line in a twodimensional spectral space of red and near infrared generated from a series of pure soil pixels ranged by reflectance. the line is a comprehensive reflection of the soil’s characteristic in various water conditions (zhan et al., 2007; qin et al., 2012). the modeling process of soil water content is demonstrated in figure 2. the influence of eco-water retrieved by quantitative remote sensing on runoff in upper minjiang river basin e3 figure 1. the vegetation water content modeling flow figure 2. the soil water content modeling flow 2.3 results of eco-water remote sensing quantitative retrieval sr is the ratio of infrared wavelengths (center wavelength of 1676 nm) and nearly red band (center wavelength of 840 nm) in remote sensing image inversion of ewtc. the soil line of the image must be derived initially to compute smmrs for the remote sensing image reversal of w(t). the modulus of eco-water conservation of the study area was then retrieved according to the mec model based on the already retrieved ewtc and w(t). ecological water-related parameters inversion results are displayed in table 1. table 1. eco-water remote sensing inversion results jin huang, wunian yang, xin yang, bing denge4 3 the influence analysis of eco-water on runoff 3.1 change trend of runoff and climate weather conditions during the same period should also be considered when studying the influence of ecological water on runoff. climate data collected for this study was related to precipitation and temperature of year and month in the upper minjiang river basin between 1990 and 2005. precipitation and temperature data were collected from 9 national surface meteorological stations (songpan, zoige, barkam, hongyuan, xiaojin, dujiangyan, pingwu, mianyang, ya'an) which were within or adjacent to the river basin and then was interpolated to derive raster data of the study area. runoff data of year and month in the upper minjiang river basin between 1990 and 2005 was collected from zipingpu hydrologic stations. according to the data analysis results, during the period 1990-2005 in upper minjiang river basin, the annual runoff depth, and the annual precipitation both revealed a decreasing trend (figure 3); annual average temperature was on the rise (figure 4); minimum monthly average depth in dry season appeared nearly consistent in february; and february monthly runoff depth exhibited a decreasing trend (figure 5). figure 3. annual precipitation and runoff depth change trend between 1990 and 2005 of upper minjiang river basin figure 4. annual average temperature change trend of between 1990 and 2005 of upper minjiang river basin figure 5. monthly runoff depth of february change trend between 1990 and 2005 of upper minjiang river basin 3.2 variation of eco-water two phases of mec were retrieved with quantitative remote sensing method. the mean of mec in 1994 was 1.6206 g/cm2, converted to a depth of 16.206 mm. in 2001 was 1.4635 g/cm2, converted to a depth of 14.635 mm. the depth of eco-water was 1.6mm less in 2001 than in 1994; thus water conservation capability decreased in 2001 from 1994 levels. subtraction of the two phases of mec may reveal the change in eco-water levels, extent and spatial distribution of recovery or degradation, and the impact of eco-water on runoff. a high ratio overlap of the mec maximum reduction area and the deforestation area occurred (figure 6), revealing that deforestation likely resulted in a reduction of water conservation capability in the basin. various levels of mec reduction took place in the lower altitude valley area, with slight increases in the relatively smooth terrain areas of songpan and heishui counties. the mec changes closely correlated to the vegetation level with reduction of mec due to vegetation degradation resulting from both natural and human factors. temperature rise and decreasing precipitation caused aridification of the river valley, leading to vegetation degradation (wu et al., 2003), while deforestation and overgrazing generated the same result (sansom, 1999). measures for plant protection and restoration were implemented to achieve an increase in mec as deforestation have been prohibited since 1999 and farmland has been converted back to forests or grasslands with mitigative forest planting continuously and actively promoted. recovery of mec remains much less than a reduction, as the mec recovery area was significantly less than the reduction area with the greatest degree of improvement at 0.6178 g/ cm2, much less than the most significant reduction level of 2.6820 g/cm2. the influence of eco-water retrieved by quantitative remote sensing on runoff in upper minjiang river basin e5 table 2. depth contrast of precipitation, runoff, and eco-water eco-water is a vegetation-centered water resource that decreases with vegetation destruction and increases with vegetation restoration. the year 1999 was a turning point for vegetation change in the upper minjiang river basin as deforestation practices were prohibited and implementation of plant protection measures followed. the changing trend is illustrated as eco-water was monitored between 1990 and 2005, first decreasing until 1999 and then slightly increasing. change direction of february runoff depth during the same period was analyzed and found to exhibit an obvious decreasing trend between 1990 and 1999 as the correlation coefficient r2 reached 0.7037 (figure 7a). the declining trend was weakened between 1999 and 2005 as the correlation figure 6. eco-water mec changes and comparison with deforestation area between 1994 and 2001 of upper minjiang river basin 3.3 the influence analysis of eco-water on runoff a comparison of mean mec with annual precipitation depth, annual runoff depth and runoff depth in the dry season was conducted to determine the eco-water influence on runoff. the effect of eco-water on runoff is displayed more clearly by water recharging abilities in the dry season as the formation of flood peak is so closely aligned with precipitation. influence of eco-water on runoff focuses on exploring water recharging capability for a runoff in the dry season in this study. comparing the data from the year 2001 with the year 1994 (table 2), the annual precipitation depth had increased by 8.7% and the annual runoff depth had increased by 9%, but the runoff depth in dry season had decreased by 5.6%, indicating transformation of precipitation directly into surface runoff increased, while intercepted and conserved precipitation fell. the mean of mec depth had dropped8.9%, explaining the reduction in water conservation capacity of the basin. comparing the two phases for the depth of runoff and eco-water reveals the vital influence of eco-water on runoff as when the amount of eco-water was reduced, the water supply capacity of eco-water to flow in dry season also decreased. jin huang, wunian yang, xin yang, bing denge6 coefficient r2 was only 0.1116 and the decreasing slope increased from -0.3358 to -0.1952 (figure 7b). the trend change comparison further proved that ecowater exerted significant influence on runoff in the upper minjiang river basin. trends demonstrated a pattern showing that when the amount of ecowater decreased, eco-water water recharge capability to a runoff in dry season also decreased, while under eco-water recovery conditions, water recharge capability also exhibited a degree of recovery. figure 7a. february runoff depth change trend between 1990 and 1999 of upper minjiang river basin figure 7b. february runoff depth change trend between 1999 and 2005 of upper minjiang river basin 4 conclusion and prospect (1) as a product of vegetation canopy, vegetation humus layer, and the soil layer, eco-water may be assumed to be independent of the hydrologic cycle, allowing the study of water storage and recharge capability from the three sources to be systematic and quantitative and simplifying the study of the related hydrological effects. (2) eco-water is related to vegetation at the earth’s surface. thus changes of eco-water resource amounts are closely related to vegetation alterations. deforestation is central to decreases in the supply of eco-water resources making plant protection and restoration for the eco-water layer protection and restoration crucial. (3) the comparison of change revealed that eco-water exerted influence on runoff in the upper minjiang river basin. a vast supply of eco-water increased capability of water recharge for a runoff in the dry season and a decreasing eco-water supply weakened the capacity. eco-water is integral to the hydrological cycle in determining the allocation of water resources. water resources will attain a dynamic and natural balance in well-developed eco-water areas; however, when the eco-water layer suffers damage, precipitation will begin to transform into surface runoff directly and instantaneously, contributing to an imbalanced setting off flooding, droughts, and ecological deterioration. further research may produce sufficient eco-water data retrieved over time to be analyzed with precipitation and temperature for specifying more accurate effect and action mechanisms of eco-water on runoff and for contributing to improved flood and drought management. references jian, j., yang, w. n., jiang, h., wan, x. n., li, y. x., & peng, l. (2012). a model for retrieving soil moisture saturation with landsat remotely sensed data. international journal of remote sensing, 33(14), 4553-4566. doi: http://dx.doi.org/10.1080/01431161.2011.652313 li, y. x., yang, w. n., tong, l., jian, j., & gu, x. f. (2009). remote sensing quantitative monitoring and analysis fuel moisture content based on spectral index. acta optica sinica, 29(5), 14031407. doi: 10.3788/aos20092905.1403 liu, x. d., wu, q. x., & su, n. h. (1989). studies on rainfall interception in canopy, litter and soil hydrological characteristics of forests in liupanshan mountains. scientia silvae sinicae, 25(3), 220-227. ogee, j., & brunet, y. (2002). a forest floor model for heat and moisture including a litter layer. journal of hydrology, 255(1), 212-233. doi: http://dx.doi.org/10.1016/s0022-1694(01)00515-7 qin, q. m., you, l., zhao, y., zhao, s. h., & yao, y. j. (2012). soil line automatic identification algorithm based on two-dimensional feature space. transactions of the csae, 28(3), 167-171. rao, l. y., zhu, j. z., & bi, h. x. (2005). hydrological effects of forest litters and soil in the simian mountain of chongqing city. journal of beijing forestry university, 27(1), 33-37. sansom, a. l. (1999). upland vegetation management: the impacts of overstocking. water science and technology, 39(12), 85-92. wan, x. n., yang, w. n., wu, b f., sun, w. d., & huang, q. (2004). conception of eco-water sphere and its application. advance in earth sciences, 6(19), 117–121. wang, l. t., wang, s. x., zhou, y., liu, w. l., & wang, f. (2011). vegetation water content retrieval and application of drought monitoring using multi spectral remote sensing. spectroscopy and spectral analysis, 31(10), 2804-2808. wu, y., su, z. x., & fang, j. y. (2003). study on causes and ecological renewal of arid and warm valley of upper minjiang river. journal of china west normal university (natural sciences), 24(3), 276-281. yang, w. n., yuan, p. x., & wan, x. n. (2001). the study report: the ecological environment integrative survey and evaluation in the minjiang upper river based on remote sensing. chengdu: chengdu university of technology archives, 863-308-21(6). yi, q. x., bao, a. m., luo, y., & zhao, j. (2012). measuring cotton water status using water-related vegetation indices at leaf and canopy levels. journal of arid land, 4(3), 310-319. zhang, j. h., xu, y., yao, f. m., wang, p. j., guo, w. j., li, l. & yang, l. m. (2010). advances in estimation methods of vegetation water content based on optical remote sensing techniques. science china technological sciences, 53(5), 1159-1167. zhang, r. h., tian, j., li, z. l., su, h. b., chen, s. h., & tang, x. z. (2010). principles and methods for the validation of quantitative remote sensing products. science china (earth sciences), 53(5), 741-751 zhan, z. m., qin, q. m., ghulan, a., & wang, d. d. (2007). nir-red spectral space based new method keywords: system structure; fuzzy mathematics; total quaintly control of point source; total quaintly control of unit; water quality planning. palabras clave: estructura del sistema; matemática difusa; control total de la fuente puntual; control total de la unidad; planificación de la calidad del agua. how to cite item liu, r. (2020). research on control method of pollutant total amount of water quality based on fuzzy mathematics. earth sciences research journal, 24(2), 191-199. doi: https:// doi.org/10.15446/esrj.v24n2.87248 given the scale-up of pollutants discharged into water bodies, control via reducing part of the discharge is turning out ineffective. this article proposes a method based on fuzzy mathematics for controlling water from the total pollutant amount. firstly, a control framework and process of the total pollutant amount was built. the total amount of pollutants discharged into several water bodies in china was studied to achieve the predetermined environmental objective. then, water pollution of different regions was evaluated via water quality model based on the fuzzy mathematics in the area or essential protection domain with severe pollution and a concentrated pollution source. this procedure makes a comprehensive evaluation of the water quality pollution more scientific. finally, the control of the total pollutant amount was completed via the optimized combination of point source control and unit control of the total amount. experimental results show that the method is scientific, objective, and reasonable while controlling the pollutant discharge. it controls the total pollutant amount excellently. abstract research on control method of pollutant total amount of water quality based on fuzzy mathematics investigación sobre el método de control de la cantidad total de contaminantes de la calidad del agua basada en las matemáticas difusas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87248 con el aumento de los contaminantes descargados en el agua, es difícil controlar la contaminación del medio ambiente mediante la reducción de parte de la descarga de contaminantes. por lo tanto, se propone el método de control de la cantidad total de contaminante de la calidad del agua basado en matemáticas difusas. en primer lugar, se construyó un marco de control y un proceso de la cantidad total de contaminantes. la cantidad total de contaminantes descargados en esta región se controló dentro de una cierta cantidad para lograr el objetivo ambiental predeterminado. luego, se evaluó la contaminación del agua de diferentes regiones mediante un modelo de calidad del agua basado en las matemáticas difusas de la región o en el dominio de protección clave con contaminación severa y una fuente de contaminación concentrada, lo que hace que la evaluación integral de la contaminación de la calidad del agua sea más científica. finalmente, el control de la cantidad total de contaminantes se completó mediante la combinación optimizada del control de fuente puntual y el control de la unidad de la cantidad total. los resultados experimentales muestran que el método es científico, objetivo y razonable durante el control de la descarga de contaminantes. controla la cantidad total de contaminantes de forma excelente. resumen record manuscript received: 13/11/2019 accepted for publication: 07/04/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 191-199 rui liu* institute of mathematics and computer science, yan'an university, yan'an, 716000, china * corresponding author: hgkyugyu123@163.com m at he m at ic s a pp li ed t o hy dr ol og ic p ro bl em s https://doi.org/10.15446/esrj.v24n2.87248 https://doi.org/10.15446/esrj.v24n2.87248 https://doi.org/10.15446/esrj.v24n2.87248 mailto:hgkyugyu123@163.com 192 rui liu introduction recently, the trend of environmental conditions deterioration of our country has not been improved, especially for the water environment pollution (shi et al., 2017). hundreds of organisms have been detected in main rivers, lakes, and reservoirs in china, and some water areas have been severely polluted by the organisms. the increasing trend is shown in the national water source pollution. nearly 90% of urban water is contaminated severely, while water quality of water sources of 50% of principal towns is substandard. the water pollution has deteriorated from east to west, extended from tributaries to trunk stream, spread from city to village, permeated from the earth’s surface to underground, and diffused from area to watershed. according to statistics, there is universal organic pollution in main rivers in china, and the non-point source pollution is increasingly outstanding (wang et al., 2018). the liaohe river and haihe river are polluted severely. the water quality of the huaihe river is reduced, while the one of the yellow river is not optimistic. the water quality of the songhua river is not bad, while the one of the zhujiang river and the changjiang river is right in total. organic pollution of surface water of seven river basins is universal, and water quality of 58% of fracture surfaces in the trunk stream of each drainage basin belongs to iv grade. in contrast, one of 22% of the fracture surfaces belongs to iv grade and 20% to v grade or inferior v grade. lake pollution in china is also severe. the water body of most lakes is eutrophication. mainly, the contamination levels in taihu lake, chaohu lake, and dian lake (the sanhu for short) are the most serious. for years, the pollutant discharged into the lake from the non-point source pollution on land and factory sewage increases with the rapid development of the economy and growth of population, which results in increasingly outstanding water environment problems of the sanhu. in addition, many lakes that are close to the populated regions such as cities and towns have been degenerated into the wastewater reservoir in the drainage basin, especially for the dirty pond and stream of the village, which is alarming. to control the worsening trend of environmental degradation, the decision of the state council on several issues of environmental protection requires the prevention and control of water pollution in river basins. the total quantity control of critical pollutants and the verification system should be implemented according to the law for the water bodies whose discharge of water pollutants has reached the standard but cannot meet the water environment quality standards stipulated by the state. the total control of water pollutants is mainly based on environmental statistics (ge et al., 2017), supplemented by measures such as pollution declaration and registration to verify the current situation of regional water pollution and discharge. based on the pollution type, according to requirements of specific water environment quality target and water quality target of boundary section, local environmental capacity, economic development scale, pollution control technology, environmental management level, and other factors are considered synthetically to reckon the maximum discharge of pollutant allowed by the target of water environment quality. according to the different geographical locations, technical level, pollutant discharge quantity, and pollution control capacity of each source, the responsibility of pollutant reduction is optimized and coordinated. also, the permitted emission pollutant index and reduction quantity are limited in the form of pollutant discharge allowance to achieve the target of water environment quality finally. there have been some reports on the research of total emission control. a heuristic method has been used to research pollutant control. the method mainly uses dynamic programming algorithms, which are particularly attractive in this respect. however, the code of dynamic programming is complex, which needs many computer resources. in addition, using algorithms based on mathematical programming still exists resistance in practical application. therefore, there is a certain difference between theory and practice (wang et al., 2017) in the system analysis of water quality management. according to the probability restriction condition, the method researches the problem via the optimization model and describes an optimization model that can reflect the uncertainty of water quality. the adopted technology is the chance constraint plan, where the certainty equivalent replaces the probability restriction in optimization problems of water quality. first-order uncertainty analysis is used to represent the intrinsic uncertainty of random factors (mohamad & boumahdaf, 2017). the total amount of the polluted water was controlled via a linear programming model. in the condition of lateral mixing and variation in river flow generated by tributary confluence with different pollutant levels, the optimization assignment model of the garbage load of the drainage basin is proposed. on the premise of meeting the water quality environmental standards of the water intake along the river, the model regards the flow of the river system as a random variable. it is applicable to the river where three tributaries converge at the same point. the flow frequency of the flow group obtained by the statistical method represents the flow frequency of the basin approximately. the priority of sewage treatment transfers from one tributary to another tributary (huang et al., 2017). however, in general, these control methods are still immature and imperfect. further research and discussion are still needed. this research applied the fuzzy mathematical method (li et al., 2017) to carry out a preliminary attempt for the comprehensive evaluation of water quality (brix et al., 2017). the measured value reflecting various problems of water quality pollution was translated into the quality value, which indicates superiority and inferiority of water quality via function relationship. our method considered the fuzziness of the boundary line of the water quality and overcame the insufficient clearness in other evaluation methods of water quality pollution. therefore, the comprehensive evaluation method of fuzzy mathematics is more scientific while evaluating the water quality pollution in various regions. this article used the weight matrix and membership relation to build the model of water quality (gao et al., 2019). the total quality of the contribution of sewage exits was used to calculate the allowable discharge capacity of each sewage exit (molinos-senante et al., 2017), which would improve the total amount of pollution discharge. overall control frame and primary function the control frame of the total quantity is comprised of numerous indexes. the control frame is divided into four sub-domains to facilitate the analysis, as shown in figure 1. each subsystem was analyzed via the regional analysis method as follows. (1) the subsystem of the pollution source the subsystem included point source system (eckley et al., 2017), surface source system (fan et al., 2018), and pollutant system. the pollutant system included indexes such as codcr (chen et al., 2017), bod5 (bodiga et al., 2017), ammonia nitrogen, and volatile phenols. as the pollutant of total quantity control, it could be selected from the standard-exceeding pollutant of water quality according to the requirement of plan and management. (2) the subsystem of water function zone the subsystem included systems of the zone of sewage discharge control, upstream functional zone, transitional zone, and downstream functional area. the zone of sewage discharge control was the river reach where the pollutant from sewage draining exit and surface source were converged intensively. the zone of sewage discharge control was the vital zone for total quantity control of the received pollutant water and the sewage-draining exits, and the complete quantity control was conducted in this zone. the scope of this zone was 100m~500m of upstream from the first sewage exist, namely taking the termination section of its upstream function zone as the starting section and the section where the pollutants discharged from the sewage outlet were mixed with the river water as the termination section. in the control zone, there were not only physical processes of dilution and diffusion of pollutants, but also the processes of chemistry, optics, and biology self-purification (wang, 2019). the transitional zone was set up to ensure the pollutants discharged from the sewage control zone could satisfy the water quality target of the next functional region after being diluted and self-purified by the river. the starting section of this area is the termination section of the sewage control area, while the termination section is the starting section of the next function area. if the distance between the termination section of the sewage control area and water intake of the following function area is within 2000 meters, the transitional zone did not have to be set, and the termination section of the sewage control area could be placed at the starting section of the downstream functional area. the migration and transformation process of pollutants in this section was mainly the self-purification. the starting section of the upstream functional area was the section that is 1000m from the first water intake in the upstream region, and the termination section was the one that is 100m~500m from the first sewage 193research on control method of pollutant total amount of water quality based on fuzzy mathematics and two-dimensional models while the upstream function area via the input and output model directly. the analog computation was carried out for the concentration of the pollutant of the termination section in the sewage control area via the transformation model of the one-dimensional model using the control target of water quality of the termination section in the sewage control area as the basis of calculation. then, the concentration of the pollutants was multiplied by the designed water quantity of this river. thus, the total amount of contaminants allowed to flow into the sewage control area was obtained. the contribution of pollutants from the upstream functional area in the sewage control area to the termination section was subtracted, the total amount of contaminants allowed to flow into the part of the sewage control area and the allowable discharge capacity of sewage draining exit was obtained. then, the allowable discharge capacity of each sewage-draining exit was calculated via the transformation model of the two-dimensional model and hydrological parameters under the condition of design water quantity. method and procedure of total quantity control the total quantity control work starts from the survey of regional water environment and ends up with the water resources protection plan. as shown in figure 2, the entire process consists of seven workflows. survey of regional water environment the survey mainly investigated the distribution of sewage exits, the distance between sewage exits, the discharge of wastewater, and the types and concentrations of pollutants. intake quantity of water of each intake, intake time and intake location were also investigated. to determine the major regional water pollutants, the regional water quality was investigated and evaluated. the function of upstream and downstream reach of sewage exit was surveyed. so were the distance between the last sewage exitand the first water intake of downstream reach, the water quality control target of the functional area, and the hydrological characteristics of the regional reach. according to the requirement of compilation of total quantity control, the investigation data were classified systematically. the pollutants of total quantity control were determined, so were the water quality mathematical model adopted by total quantity control, exit in the upstream sector. the river reach is the pollutant input area of the sewage control area in the upstream region. the scope of the downstream functional area was the river reach that is below 1000m from the first intake in the upstream of the downstream of the transitional zone. the target of water quality control of the initial section of this area was an essential basis for calculating the total afflux amount of pollutants allowed to enter the sewage control area. (3) the subsystem of the water quality model the subsystem included water quality model, parameters of the model, and hydrological parameters. the water quality model was built via the fuzzy mathematical method. the model mainly included the input and output mathematical model, one-dimensional mathematical model, and a twodimensional numerical model that could reflect the input response relationship between pollution sources and water quality targets (abubakari, 2017). in conditions of inflow of tributaries, the two-dimensional analytical model and the input and output model were used for the upstream function area of the sewage control area while the two-dimensional model for the transition area (yao et al., 2017). if the pollution control area was conditional, the numerical solution of a two-dimensional partial differential equation could be carried out for the analog computation of water quality. parameters of the water quality model included a longitudinal diffusion coefficient, transverse diffusion coefficient, and selfpurification coefficient. the longitudinal and transverse diffusion coefficients were calculated via empirical formulas according to hydrological parameters, and the self-purification coefficient was calculated via the measured data of hydrology and water quality or simultaneous monitoring data of water quality and water quantity. river hydrological parameters included average flow velocity, average river width, average depth, and design flow. (4) the subsystem of total control the subsystem was composed of the total amount of pollutants allowed to flow into the sewage control area, the allowable discharge capacity of sewage draining exit, and the simulation model of water quality. the simulation model of water quality was composed of the mathematical model established after parameter estimation. analog computation was carried out for the transition area and sewage control area via the transformation model of one-dimensional figure 1. structure chart of the overall control frame 194 rui liu figure 2. flow of total quantity control water quality simulation scheme and total quantity analysis method and the total quantity control scheme was put forward. according to the function of the total quantity control area, the river reach in the area was divided into sewage control area and transition area, and the scope was determined. meanwhile, the scopes of upstream reach of the sewage control area and downstream reach of the transition area were determined. so were the function of water and the target of the water quality control. fuzzy mathematics method to control the water pollution the work plays a key role in establishing the input response relationship between pollution sources and the target of water quality control. the fuzzy mathematics method uses the weight matrix and the maximum membership principle to consider the factors or main factors related to the evaluated things and carries out the total quantity control for the factors. comprehensive judgment of the fuzzy mathematics method the membership function is the precondition for building the fuzzy mathematical model. to select a model truly reflecting the self-purification law of pollutants in the water body is very important using the “lower semi-trapezoidal fuzzy number”, “triangle fuzzy number” and “upper semi-trapezoidal fuzzy number” as the gist for building membership function. generally, the model is selected as the following principles. the model is selected according to the needs of dilution and diffusion, self-purification process and total quantity control of pollutants in water. the dimensions of the selected model are determined according to the existing hydrological data, water quality data, and the difficulty of evaluating parameters in the model. the model should be strived to be simple, applicable, with clear physical concept, and simplified according to the actual situation; the application conditions, application scope, advantages and disadvantages of the model should be understood in order to make the model have a good simulation property. according to the protection target and the purpose of water area use, integrated with the “groundwater quality standard” (read et al., 2017) (gb 3838-2002), the membership function of water quality evaluation is divided into 5 categories by function. i class is mainly applicable to the source of water and state reserves, while ii class applicable to the first-grade protection zone of centralized drinking water source, habitat of rare aquatic organism, fish protection zone, spawning ground of fish and shrimp, feeding ground of larvae, etc. iii class is suitable for the second-grade protection zone of centralized drinking water source, wintering ground of fish and shrimp, migration channels, fishery waters and swimming areas of aquaculture. iv class is applicable to general industrial water zone and general industrial water zone and recreational water zone that is non-direct contact with humans, while v class applicable to agricultural water zone and water zone with general landscape requirements. figure 3. membership function of water quality evaluation figure 4. the modified membership function of water quality evaluation corresponding to functions of the above five categories of water areas of surface water, fundamental items of environmental quality standards of surface water were divided into five categories, the corresponded standard values of the categories were carried out for different functional categories respectively. the standard value with higher functional category of water area was stricter than that with lower functional category. if the same water area had multiple use functions, the standard value corresponding to the highest function category was performed. it was the same meaning to realize the function of water area and reach the standard of function category. as shown in fig. 3, membership functions of index factors the polluted water was built according to the classification standards of water quality. it could be seen from the classification standard of single quality of the polluted water (table 1) that standard value of iv class and v class water quality was a critical value (namely the standard values of si4 and si5 were equal, that were critical values of iv class and v class water quality). in fact, they were only distinguishing value and 195research on control method of pollutant total amount of water quality based on fuzzy mathematics the transition from iv class to v class was absent, resulting in the repetition and discontinuity of the membership function. parts of evaluation results of water quality were v class, however belonged to iv class actually. it was obvious that the water quality evaluation could not build the membership function (si4-si5=0), leading to distortion of the evaluation result. in view of the membership function with iv-grade connection degree was symmetric about xi = si4, it was modified via triangular fuzzy number, restraining si4 si5 = s4 si3, which could build the membership functions of water quality of iv class and v class. table 1. evaluation standard of single water quality/ mg•l-1 category/ standard value i class ii class iii class iv class v class total hardness ≤150 ≤300 ≤450 ≤450 ≤550 total dissolved solids ≤300 ≤500 ≤10000 ≤1000 ≤2000 sulfate ≤50 ≤150 ≤250 ≤250 ≤350 chloride ≤50 ≤150 ≤250 ≤250 ≤350 iron ≤0.1 ≤0.2 ≤0.3 ≤0.3 ≤1.5 manganese ≤0.05 ≤0.05 ≤0.1 ≤0.1 ≤1.0 copper ≤0.01 ≤0.05 ≤1.0 ≤1.0 ≤1.5 zinc ≤0.05 ≤0.5 ≤1.0 ≤1.0 ≤5.0 nitrate ≤2.0 ≤5.0 ≤20 ≤20 ≤30 nitrite ≤0.001 ≤0.01 ≤0.02 ≤0.02 ≤0.1 ammonia nitrogen ≤0.02 ≤0.02 ≤0.2 ≤0.2 ≤0.5 fluoride ≤1.0 ≤1.0 ≤1.0 ≤1.0 ≤2.0 mercury ≤0.00005 ≤0.0005 ≤0.001 ≤0.001 ≤0.001 chromium ≤0.005 ≤0.01 ≤0.05 ≤0.05 ≤0.1 bicarbonate radical ≤100 ≤200 ≤300 ≤300 ≤400 the relationship of subordination of i v type of water quality was continuous via the modified membership function (fig.4) and conformed to the evaluation standard of the “groundwater quality standard” (gb/t 14848—93). the expression of the modified membership function was shown in formulas 2-6. a set of pollution factors for a given section was known µ = µ µ µ1 2, , ,  n{ }, where elements µi i 2, , n=( )1,  were factors influencing the evaluation object. usually the importance of each factor was different, hence a corresponding weight ai i 2, , n=( )1,  should be given for each factor i. a i nn i i i n =( ) = ≥ =    = ∑a a a a a 1 2 1 1 0 1 2 , , , , ,,   (1) element i is an important degree of membership of factor i therefore, weight set a is a fuzzy subset of factors . general evaluation sets v v v vm= …{ }1 2, , , was established. the elements vj j m= …( )1 2, , , were various possible evaluation results, which could be fuzzy or non-fuzzy, but the relationship to “v” was clear. the evaluation started from a i to confirm the degree of membership rij j m= …( )1 2, , , of the evaluation object to the element vj of the evaluation set, which was called single-factor fuzzy evaluation. evaluation result r of the factor i was the single-factor fuzzy evaluation set. when i belongs to the i class, the expression of membership function was shown in formula 2. ri i i i i i i i i i i i 1 2 2 2 2 1 0 = ≤ − − < < >      ? ? x s s x s s s x s x s l l l  (2) when i belongs to the ii class, the expression of membership function was shown in formula 3. 0 3 2 1 2 3 3 ? x s x s x s s s s x s s x s s l l l i i i i i i i i i i i i i i i ≤ > − − < < − − ∪ 22 2 3s x si i i< >      (3) when i belongs to the iii class, the expression of membership function was shown inf formula 4. ri i i i i i i i i i i i i i i 3 2 4 2 3 2 2 3 4 0 = ≤ > − − < < − ? x s x s x s s s s x s s x s ∪ 44 3 3 4− < >      s s x s i i i i (4) when i belongs to the iv class, the expression of membership function was shown in formula 5: ri i i i i i i i i i i i i i i 4 3 5 3 4 3 3 4 5 0 = ≤ > − − < < − ? x s x s x s s s s x s s x s ∪ 55 4 4 5− < >      s s x s i i i i (5) when μi belongs to the v class, the expression of membership function was shown in formula 6: ri i i i i i i i i i i i 5 4 5 5 4 4 5 5 0 1 = ≤ − − < < >    ? ? x s s x s s s x s x s    (6) si si si si si1 2 3 4 5, , , , in formulas 2-6 corresponded respectively with the standard limiting value of i,ii, iii, iv, v grades in water quality standard, where s s s si i i i5 5 4 5= + − . the xi was the measured value of sampling point in the research area and r r r r ri i i i i1 2 3 4 5, , , , were degree of membership of i,ii, iii, iv, v class water. degree of membership r r r r r ri i i i i i=( )1 2 3 4 5, , , , of factor i to evaluate v was confirmed via the fuzzy mathematics. expression of fuzzy relationship from u to v was defined as r r r ri i i i={ }1 1 1, , ,  γ , j n= …( )1 2, , , , which is further expressed via matrix representations as formula 7 shows. 196 rui liu r r r r r r r r r r j j i i ij =         11 12 1 21 22 2 1 2 � � � � � � �           (7) it is obvious that the fuzzy evaluation of single factor only reflects the influence of one factor on the evaluation object, but does not reflect the comprehensive influence of all factors. so the comprehensive evaluation result cannot be obtained. therefore, the influence of all factors must be considered comprehensively, which is called fuzzy comprehensive evaluation. the set of comprehensive assessment r is the fuzzy subset of the v, which can be expressed as follow: r a v r r r r r r n m m� �i� � � � � � � = =( )a a a1 2 11 12 1 21 22 2, , , � � � r r r b b b i i im m 1 2 1 2                   =( ), , , (8) where b v r j mi i n ij= ( ) ==1 1 1 2a λ , , , , , m is the evaluation index. it considers the influence of all factors comprehensively. obviously, the line i of b represents the influence degree of ifactor staking each evaluation element on the evaluation object. the column j represents the degree of all factors taking i evaluation element on the evaluation object. hence, each column element is multiplied by the corresponding factor weight i n= …( )1 2, , , , which more more reasonably reflect the comprehensive influence of all factors. establishment of water quality model and calculation of evaluation results given the condition of u v r, ,{ } evaluation space, operational rule was built via weight matrix w and membership relation matrix r. thus, the membership matrix y of fuzzy comprehensive relation between evaluation object of water quality and each evaluation order was obtained as shown in formula 9. y w r c c cj= =( )� �1 2, , , (9) where y was the comprehensive evaluation results.  represented the operational rule. c c c j l nj1 2 2 3, , , , , , ,  =( )was the membership degree. common fuzzy operational rules are the type of main factors determine the i type of main factors highlighted, the ii type of main factors highlight, type of weighted average model, which were shown from formula 10 to formula 13. (1) main factors determines y v w r w ri n i ij i n i ij= ( )= ( )  = =1 1λ max min , (10) formula 10 adopted the principle of “small to large” operation, namely the weight matrix w took the minimum value from the evaluation element (index) corresponding to each column in the membership relation matrix r, then took the maximum. the advantage of the algorithm is to remove the repeated factors in membership degree and weight index. therefore, it avoided the repeated consideration of the membership degree and weight of indexes to take the minimum value first, which highlighted the role of extreme value. (2) i type of main factors highlight y v wr w ri n i ij i n i ij= ( )= ( )  = =1 1max min , (11) formula 11 adopted the principle of “multiply to take the maximum”, namely the weight matrix w took the minimum value after multiplying the element corresponding to each column in the membership relation matrix r. the algorithm not only considered the interaction effect of index membership degree and weight on evaluation results, but also highlighted the role of extreme value. (3) ii type of main factors highlight y w r w r w ri n i ij i ij i n i ij i n =⊕ ( )= ( )         = ( )= = = ∑ ∑1 1 1 1λ λ λmin , (12) formula 12 adopted the principle of “adding small values “, namely the weight matrix w took the minimum value from the element corresponding to each column in the membership relation matrix r, then added together. this algorithm removed the important factors in the membership degree and weight of the index. however, the overlapping budgets might cover the influence of the maximum pollution index on groundwater. when the concentration of some indicators was particularly large, the comprehensive evaluation results were unreasonable. (4) type of weighted average model y w ri ij i n = ( ) = ∑ λ 1 (13) formula 13 adopted the principle of “multiplication and addition”, namely the weight matrix w multiplied the element corresponding to each column in the membership relation matrix r, then carried out a linear weighted average. the algorithm reflected the comprehensive influence of all factors, which was a comprehensive evaluation with a membership weight type in fact. the establishment of weight matrix plays an important role in building water quality model, which directly affects the application of model and the calculation result of total quantity control. to construct the weight matrix, the tracking and monitoring data of water mass should be obtained and proceeded according to certain methods, which requires a large amount of time and funds. it is difficult in establishing the process of the protection planning of water resources. in condition of the limited funds and time, establishment of the weight matrix and regression method can be used to estimate the parameters. due to different importance degrees of µi i l n=( ), , , ,2 3  in the comprehensive evaluation of water quality and environmental quality in subset u, such as function, contribution, and influence, the index factors participating in the evaluation should be weighted. different weights influence the evaluation results in varying degrees. there are two major methods for weight definition, subjective weighting method and objective weighting method. the latter is usually used in practical evaluation. our research used the method of rate of pollution contribution to confirm the weight. it is a dynamic/variable weight method, which can not only highlight the impact of major pollution indexes, but consider the comprehensive impact of individual pollution indexes on the environment. the weight calculation was shown in formulas 14-16, concretely. w x s i l n j li i j s ij= = =( )=∑/ , , , ; , , , 1 5 2 2 51   (14) w w w i l ni i j n i  = =( )=∑/ , , ,1 2  (15) w w w w wl n i n i={ } ==∑, ,2 1 1 (16) in formulas 14-16, xi is the measured values of water quality samples in the research area. sij is the standard limiting value of i,ii, iii, iv, v grade in water quality standard and wi is the weight of underground water indexes in different places obtained by normalization. confirming the target of water quality control the main purpose is to determine the target of water quality control of the functional areas in the upstream sewage control area and downstream transition area. the determination of the target of water quality control is mainly based on the river function and management requirement of total quantity control. therefore, it has certain timeliness. calculation of discharge total amount allowed in the control area according to the water quality control target and designed water quantity of the initial section of the downstream functional area of the transition zone, the 197research on control method of pollutant total amount of water quality based on fuzzy mathematics pollutant concentration value of the final section of the sewage control area was calculated via the water quality model. the concentration value was multiplied by the designed water quantity and the discharge total amount was obtained. the required designed water quantity, river width, average water depth, river velocity and other parameters could be calculated according to the established relevant model and hydrological calculation methods. according to the total amount of allowable pollutant discharge, the established water quality model was used to calculate the allowable pollutant discharge for each sewage exit. according to the development of control section, pollution control situation, sewage discharge and water pollution treatment situation, through the actual data of the specific control unit, the maximum allowable discharge of control unit, cod discharge and concentration of each point were calculated. total discharge amount (q)=total quantity (qtotal)  fuzzy evaluation matrix (r) (17) examination of method the total quantity control scheme was examined. if there is any shortage, adjustment should be carried out for the design of control scheme or target of water quality control and calculation should be proceeded by a working routine until the results are satisfied. the achievement was brought into the program of water resource protection. simulation experiment and results the main technology of total quantity control is to establish the water quality model via fuzzy mathematical method, and to calculate the total quantity of pollutants allowed to flow into the sewage control area and the allowable total quantity of sewage outlets. this paper takes the pilot project of total quantity control in the water function zone in a certain area as an example to illustrate the main technical methods of total quantity control. general situation of region water environment the river section in this area is 98 kilometers long, and there are hydrological and water quality stations on the main stream. the monitoring data of the hydrological and water quality station are shown in table 2. there are many tributaries in the middle of the river and 5 sewage exits. the length of sewage exits is 3.2 km. the calculation and comprehensive evaluation results of the membership function based on the fuzzy mathematics were shown as follows. the membership functions of each grade water are shown from formula 18 to formula 27. there are many tributaries in the middle of the river and 5 sewage exits. the length of sewage exits is 3.2 km. the calculation and comprehensive evaluation results of the membership function based on the fuzzy mathematics were shown as follows. the membership functions of each grade water are shown from formula 18 to formula 27. r x x x x i i i i 1 0 0 5 0 5 1 0 5 5 0 7 0 7 0 = −( ) ≤ < < >      . . . . . . (18) r x x x x x x i i i i i 2 0 0 5 0 5 0 5 3 0 3 0 5 0 7 0 3 0 5 = − −( ) −( ) ≤ < < < < . . . . . . . . ∪ 7.0 ..0      (19) r x x x x x x i i i i i 3 0 0 5 0 5 2 0 2 0 3 0 5 0 2 0 3 0 = − −( ) −( ) ≤ < < < < . . . . . . . . ∪ > 5.0     (20) r x x x x x x i i i i i 4 0 3 0 1 0 1 0 2 0 3 0 1 0 2 0 = − −( ) −( ) ≤ < < < <    . . . . . . . ∪ > 3.0   (21) r x x x x i i i i 5 0 2 0 1 1 0 1 0 2 0 2 0 = − −( ) ≤ < < >      . . . . . (22) membership function of bod to each grade water r x x x x xi i i i1 0 3 0 1 0 0 5 1 0 1 5 2 0= − −( ) −( ) ≤ ≥ < <      . . . . . . (23) r x x x xi i i2 0 0 2 1 5 3 0 1 5 1 0 1 5 2 0= − −( ) − −( ) ≤ < <      . . . . . . . ∪ (24) r x x x x x x i i i i i 3 0 2 0 0 5 5 0 2 0 5 0 2 0 3 0 3 0 5 0 = −( ) − −( ) ≤ > < < < < . . . . . . . . . ∪     (25) r x x x x x x i i i i i 4 0 3 0 0 33 0 8 3 0 8 0 3 0 5 0 5 0 8 = − −( ) − −( ) ≤ > < < < < . . . . . . . . ∪ ..0      (26) r x x x x i i i i 5 0 0 33 0 5 1 5 0 5 0 8 0 8 0 = − −( ) ≤ < < ≥      . . . . . . (27) table 2. synchronous detection data of hydrology and water quality station name time water temperature flow velocity flow pollutant (°c) (m/s) (m³/s) codct ammonia nitrogen a 10:40 0.2 0.15 15.7 77.1 3.90 12:40 0.14 17.2 85.1 3.76 14:40 0.14 16.9 81.1 3.97 16:40 0.15 16.2 89.1 3.99 b 0:50 0.2 0.42 19.9 73.2 3.74 3:40 0.44 20.8 81.1 3.74 5:30 0.42 20.6 77.1 3.93 7:10 0.43 20.5 78.7 3.77 8:10 0.42 20.4 69.2 3.96 9:10 0.42 20.1 65.2 3.94 c 8:20 0.2 0.74 17.3 85.1 3.80 10:40 0.77 16.6 65.2 3.77 12:30 0.84 18.0 66.0 3.85 14:10 0.79 17.9 69.2 3.87 15:10 0.77 17.8 69.2 3.80 16:10 0.84 18.0 64.4 3.91 d 14:00 0.2 0.29 16.4 61.2 3.77 16:00 0.28 16.9 57.3 3.77 18:30 0.28 16.9 65.2 3.77 20:00 0.27 16.8 63.6 3.87 21:00 0.28 16.9 65.2 3.87 22:00 0.26 16.7 75.7 3.71 198 rui liu the membership functions of other comment factors to each grade water are analogized in order. according to the above membership calculation formula, membership functions of each evaluation factor for 5 grades are established to obtain membership degrees respectively. the single-factor evaluation matrix r is established: r = 1 0 0 0 0 1 0 0 0 0 1 0 0 0 0 0 0 38 0 62 0 0 1 0 0 0 0 0 48 0 52 0 0 0 0 0 90 0 10 0 0 0 1 0 . . . . . . 00 0 1 0 0 0 0 0 0 1 0 0 1 0 0 0 0 1 0 0 0 0 0 94 0 06 0 0 0 0 0 14 0 86 0 0 1 0 0 0 0 . . . .                                                                 (28) according to the above weight calculation method, the normalized weight matrix w is established after normalization. the results are shown in table 3. table 3. weight calculation results do bod cod cn phenol concentration 2.69 2.50 7.73 0.004 0.0076 w x si i ij j i s = = ∑/ 1 5 3.6 3.9 5.6 0.0076 0.0134 standard concentration 0.75 0.64 1.38 0.53 0.57 w wi i j i n / = ∑ 0.19 0.17 0.36 0.14 0.15 w = 0 1024 0 0442 0 0117 0 1286 0 0007 0 0701 0 1197 0 0178 0 0227 0 . . . . . . . . . .11572 0 0063 0 0163 0 0842 0 1589 0 0592. . . . .        (29) as shown in table 3, y = 0.3855 and comprehensive evaluation grade is ri3. it means that the level of water quality in the region is iii grade. the main pollutants are codcr, bod5, ammonia nitrogen, and volatile phenol, which are all the non persistent pollutants. the discharge of the above pollutants from the five sewage exits in this area is shown in table 4. table 4. discharge situation of waste water and pollutant of sewage exits number sewage exit discharge amount of wastewater (million t/a) codct bod5 ammonia nitrogen volatile penol 1 a 179 167 78.9 28.3 0.03 2 b 5.26 5.55 1.46 0.428 0.0005 3 c 1640 3290 1050 358 0.41 4 d 1480 397 59.3 7.16 0.015 5 e 3550 4270 1470 547 0.60 summation 6854.26 8129.55 2659.66 940.88 1.0555 experimental procedure and results step 1: total amount of pollutants allowed to follow into the sewage control area the total amount of pollutants allowed to import into the sewage control area refers to the total amount of pollutants allowed to import into the termination section of the sewage control area. based on the concentration index of regional water quality control, the pollutant concentration was calculated via the established model. then, the concentration was multiplied by the design of water to obtain the total amount of pollutants. the target of water quality control in the region is iii class water. thus, the concentration of codcr is 15 mg/l. the concentration of ammonia nitrogen is 0.5mr/l. according to the calculation, the concentration of codcr of the section is 15.96 mg/l, while the ammonia nitrogen is 0.503 mg/l. the calculation concentration is multiplied by the designed water quality (7.47m3/s). the allowable total amount of codcr is 3759.8 t/a, while the ammonia nitrogen is 118.5t/a. step 2: quantity allocation of allowable pollutant discharge of sewage exit the total amount of pollutants allowed to flow into the section consists of two parts. one part is the amount of pollutants discharged from estuary to the section, and the other part is the amount of pollutants discharged from the five discharge exits in the sewage control area. in the calculation of the quantity allocation of allowable pollutant discharge, the amount of pollutant imported from the first part should be subtracted, and the rest is the total contribution of the discharge exits. then the total contribution amount of each exit is used to calculate the allowable discharge amount. the specific steps are as follows: (1) contribution of the imported pollutants to the pollution at the section for the pollutants afflux of non-point source discharge, there are only four rivers containing the pollutants, which are named as a, b, c, d respectively. the pollutants and river water have been mixed evenly at this section. the targets of water quality control of the section of main stream estuary of a river and d river are ii class while the one of b river and c river are iii class. the standards of water quality control and designed duty of water are shown in table 5. table 5. standards of water quality control and designed duty of water section and tributary designed duty of water water quality control standard (mg/l) name (m³/s) codct ammonia nitrogen a river a sewage exit 91.5 0.76 b river b sewage exit 5.55 0.428 c river c sewage exit 1760.35 9.005 d river d sewage exit 211.6 0.056 the pollutants in this section are mainly self-purification process and the mixing process after the import of tributaries. when the tributaries flow into the main stream, the concentration of the pollutants after the afflux is calculated via the model. when there is no tributary inflow, the model is used to calculate the self-purification process of pollutants. the data in table 3 are substituted into the model. the pollutants concentration value of the termination section of the sewage control area is calculated via the water quality model. the concentration value is multiplied by the designed water amount of the section, which is the total amount of pollutants allowed to import into the sewage control area. after computation, the contribution concentration of codcr is 4.94mg/l and the contribution amount is 1163.7t/a. the concentration of ammonia nitrogen is 0.4mg/l, and the contribution amount is 94.23t/a. the total contribution of codcr from the sewage exits is 2596.1t/a and the concentration was 11.02mg/l. the total contribution of ammonia nitrogen is 24.27t/a and the concentration is 0.103mg/l. (2) quantity of allowable pollutant discharge of sewage exit according to the total contribution of pollutants from the discharge exits at the section and the proportion of pollutant discharge from each discharge 199research on control method of pollutant total amount of water quality based on fuzzy mathematics exit in the total discharge, simulation calculation is carried out for the water quality via the transformation formula of the established water quality model, and the allowable pollutant discharge amount of each discharge exit is obtained respectively. the calculation results are shown in table 6. it can be seen that the control results are improved apparently. table 6. calculation results of quantity of allowable pollutant discharge of sewage exit number sewage exit quantity of allowable pollutant discharge (t/a) codct ammonia nitrogen 1 a 91.5 0.76 2 b 5.55 0.428 3 c 1760.35 9.005 4 d 211.6 0.056 5 e 2263.3 14.921 summation 4332.3 25.17 conclusion (1) the framework and functions of total quantity control are analyzed, and the workflow and technical method of total quantity control are put forward, which is of practical significance for the current work of water resources protection planning. (2) the comprehensive evaluation method is normative and applicable, and the existing hydrology and water quality data can be used to evaluate the water quality, which has good practicability. (3) the membership function and weight value are used to establish the water quality model to calculate the allowable discharge amount of pollutants from the sewage exit. and the existing hydrology data are used for calculation. the method is used in the calculation of the allowable discharge amount from the sewage exit for the first time, which is simple, applicable and highly simulated. total amount control of pollutant is to make a control plan and allocate total amount of pollutant after in-depth investigation and research and application of scientific method. to some extent, it is the dynamic planning of regional pollution sources, scientific, quantitative and standardized management of pollution sources, rational use of environmental capacity, and reducing pollution pressure to ensure the realization of regional environmental quality goals. fuzzy comprehensive evaluation is comprehensive, concrete, objective and scientific in comprehensive evaluation and planning control of main pollutants judging and pollution trend tracing. it can be seen from the calculation results that the method can integrate the specific situation and characteristic, and has excellent accuracy and adaptability. it can be applied to the control of water pollution extensively. references abubakari, a., raymond, s., jo, h. s. (2017). optimal planar array architecture for full-dimensional multi-user multiple-input multiple-output with elevation modeling. electronics and telecommunications research institute journal, 39, 234-244. bodiga, v. l., thokala, s., kovur, s. m., & bodiga, s. (2017). zinc dyshomeostasis in cardiomyocytes after acute hypoxia/ reoxygenation. biological trace element research, 179, 117-129. brix, k. v., deforest, d. k., tear, l., grosell, m., & adams, w. j. (2017). use of multiple linear regression models for setting water quality criteria for copper: a complementary approach to the biotic ligand model. environmental science & technology, 51, 5182-5192. chen, c. h., tsai, c. r., liu, y. h., hung, w. l., & wu, a. y. (2017). compressive sensing (cs)-assisted low-complexity beamspace hybrid precoding for millimeter-wave mimo systems. ieee signal processing bulletin, 65, 1412-1424. eckley, c. s., gustin, m., miller, m. b., & marsik, f. (2017). scaling nonpoint-source mercury emissions from two active industrial gold mines: influential variables and annual emission estimates. environmental science & technology, 45, 392-399. fan, l., yuan, y., ying, z., lam, s. k., liu, l., zhang, x., liu, h., & gu, b. (2018). decreasing farm number benefits the mitigation of agricultural non-point source pollution in china. environmental science and pollution research, 26, 464-472. gao, w., wang, w. f., & dimitrov, d. (2019). toughness condition for a graph to be all fractional (g,f,n)-critical deleted. filomat, 33(9), 2735-2746. ge, b., wang, z., lin, w., xu, x., li, j., ji, d., & ma, z. (2017). air pollution over the north china plain and its implication of regional transport: a new sight from the observed evidences. environmental pollution, 166, 29-38. huang, w., kai, s., qi, j., & ning, j. (2017). optimize the configuration of dynamic reactive power supply by using voronoi diagram method of linear programming. ieee power systems reporting, 1-1. li, y., shuai, s., & tong, s. (2017). adaptive fuzzy control design for stochastic nonlinear switched systems with arbitrary switchings and unmodeled dynamics. ieee transactions on cybernetics, 47, 403-414. mohamad, d. a., & boumahdaf, a. (2017). semiparametric bicomponent mixed model with linear constraints. ieee information theory bulletin, 64, 1-36. molinos-senante, m., maziotis, a., & sala-garrido, r. (2017). assessment of the total factor productivity change in the english and welsh water industry: a färe-primont productivity index approach. water resources management, 31, 2389-2405. read, e. k., carr, l., cicco, l. d., dugan, h. a., hanson, p. c., hart, j., kreft, j., read, j. s., & winslow, l. (2017). water quality data for nationalscale aquatic research: the water quality portal. water resources research, 53, 1735-1745. shi, y., xu, g., wang, y., engel, b. a., peng, h., zhang, w., cheng, m., & dai, m. (2017). modelling hydrology and water quality processes in the pengxi river basin of the three gorges reservoir using the soil and water assessment tool. agricultural water management, 182, 24-38. wang, d. a. (2019). critical review of cellulose-based nanomaterials for water purification in industrial processes. cellulose, 26, 687-701. wang, y., huang, q., lemckert, c., & ma, y. (2017). laboratory and magnetic field evaluation of heavy metal pollution in shilaodan beach. marine pollution bulletin, 117, 291-301. wang, y., yang, j., liang, j., qiang, y., fang, s., gao, m., fan, x, yang, g., zhang, b., & feng, y. (2018). analysis of the environmental behavior of farmers for non-point source pollution control and management in a water source protection area in china. general environmental science, 227, 1126-1135. yao, y., verginelli, i., & suuberg, e. m. (2017). two-dimensional analytical model of vapor intrusion involving vertical heterogeneity. water resources research, 53, 4499-4513. yu, s., & lu, h. (2018). multi-level and step-by-step optimization of integrated watershed management for large-scale total water pollutant allocation. environmental geosciences, 77, 373. zhou, l., li, h., & sun, k. (2017). teaching performance evaluation by means of a hierarchical multifactorial evaluation model based on type-2 fuzzy sets. applied intelligence, 46, 34-44. slowness surface calculation for different media using the symbolic mathematics language maple® earth sciences research journal earth sci. res. j. vol. 8, no. 1 (dec. 2004) : 63 67 manuscript received july 2004. 63 paper accepted october 2004. slowness surface calculation for different media using the symbolic mathematics language maple® miguel duarte1, carlos piedrahita1, trino salinas1, hernando altamar 2 and karen pachano 2 1 instituto colombiano de petróleos , ecopetrol. piedecuesta – colombia. e-mail: mudarte@ecopetrol.com.co; tsalinas@ecopetrol.com.co. 2 universidad industrial de santander -uis. bucaramangacolombia abstract starting from the equation in different media, we obtain the different type of waves that can exists in such media. the evaluation of the eigenvalues and eigenvectors let us construct the slowness surfaces. in general complex calculations case have to be made. in this work, routines were implemented in the symbolic language maple® and the slowness surfaces were plotted. this work is relevant for the modelling of equivalent media that simulates natural fractured reservoirs, like those common in the colombian foothills. it is important to understand the seismic response of this reservoirs for exploration of this areas. key words: christoffel matrices, slowness vector, cubic symmetry, hexagonal symmetry, isotropic, symmetry, slowness surfaces, eigenvalues, eigenvectors resumen partiendo de la ecuación en diferentes medios, obtenemos los diferentes tipos de ondas que pueden existir en tales medios. la evaluación de los valores propios y los vectores propios permite construir las superficies de lentitud. en general, es necesario hacer complejos casos de cálculos. en este trabajo se implementaron algunas rutinas en el lenguaje simbólico maple® y se señalaron las superficies de lentitud este trabajo es pertinente para el modelado de medios equivalentes que simulan los depósitos fracturados naturales, como los que aparecen en las zonas de piedemonte colombianos. es importante entender la reacción sísmica de estos depósitos para la exploración de estas áreas. palabras clave: isotropía, matriz de christoffel, simetría cúbica, simetría hexagonal, superficies de lentitud, valores propios, vectores propios, vector de lentitud. © 2004 esrj -unibiblos. miguel duarte et al. methodology this work was developed entirely in the field of the linear theory of elasticity. for homogeneous media the solution of the general elastodynamic equation reduces to the solution of the christoffel matrix eigenvalues problem. the christoffel equation is solved numerically using the symbolic mathematics programming language maple®. the phase velocity is obtained as a function of the anisotropic parameters and the propagation direction. the phase velocity 1/v is used to obtain the slowness, which is plotted using the standard plotting maple function in polar coordinates. the resulting graphic shows the relation between the slowness vector and the propagation angle. formulation of the problem and results theory research – fundamentals the deformation laws for a linear elastic body are determined by the linear momentum equation (1), the constitutive equation or the hooke’s law (2) and the small strain – displacement linear relation (3) (sen, 2002), (krishnaswamy, 2004), (pšenčík, 1994): 2 ij i j u u x t τ τ ρ ρ ∂ ∂ ∇⋅ = ↔ = ∂ ∂ && (1), : ij ijkl klc cτ ε τ ε= ↔ = (2), ( )1 1 2 2 k l kl l k u u u u x x ε ε ⎛ ⎞∂ ∂ = ∇ + ∇ ↔ = +⎜ ⎟ ∂ ∂⎝ ⎠ (3). where: τ is second order stress tensor, ρ is density u is displacement vector c is fourth order elasticity tensor or stiffness tensor ε is second order linearized finite strain tensor or tensor of small strain , ,i j kx x x – cartesian coordinates. strain and stress tensor are symmetric. this assumption leads us to a contracted form of the above equations, the voigt notation. equation (3) defines 6 different equations for only three components of the displacement vector ui, therefore some other conditions (4) need to be imposed over the small strain components in order to guarantee the existence of the displacement field (mase, 1970). 2 22 2 0 0ij jmkm ik k m i j j i kx x x x x x x x ε εε ε ε ∂ ∂∂ ∂ ∇× ×∇ = ↔ + − − = ∂ ∂ ∂ ∂ ∂ ∂ ∂ ∂ (4) thus, combining (1), (2), (3) and (4) the most general form of the elastodynamic equation is obtained: 2 2 k ij j l u u c i x x t ρ ⎛ ⎞∂ ∂∂ =⎜ ⎟ ∂ ∂ ∂⎝ ⎠ (5) note that the stiffness tensor reduces to a 6x6 symmetric matrix c i j k lc ij due to the symmetry of the stress and strain tensors. the condition (5) gives a simpler strain – displacement relation. the stiffness matrix can be obtained by combination of coordinate transformations corresponding to the material lattice symmetry properties. in the most general case the stiffness matrix has got 21 independent elastic constants. this is the monoclinic material. hexagonal symmetry, for example, has got 5 different elastic constants. this system has got one of the axes of symmetry such that a rotation around this axis does not change the stiffness matrix. for homogeneous materials in which elastic constants and density are constant, (5) reduces to (6): 2 2 2 k i ij j l u u c x x t ρ ∂ ∂ = ∂ ∂ ∂ (6) first consider harmonic plane waves moving through homogeneous anisotropic medium. then the solution must be sought in the next form: ( ) ( ) ( ) ( ), , i iik vt l xik vt l x i i iu x t ue u x t u e − −− −= ↔ = (7) where: u is the displacement amplitude vector, integration constant, i is an imaginary unit, 64 2 k π λ = is the wave number, λ is wavelength, v is phase velocity, t is time, l is unit vector along the propagation direction, x is position vector. replacing the displacement vector (7) in the elastodynamic equation (6) and applying the fourier transform, we get: 2 2 0ij jk kl i ll c l u uk ρω − = (8) where: ijl is director cosines matrix of the propagation vector, 2 v kv π ω λ = = – angular frequency. the equation (8) can be rewritten as: ( )2 0ik ik kv uδ ργ − = (9) where: il ij jk kll c lγ = ⋅ ⋅ – christoffel matrix (10) the slowness 1 p v = is obtained by solving the next equation for the three values of v2. ( 2det 0ik ik vδ ργ − =) (11) therefore, three types of plane harmonic waves along the propagation direction l are obtained, with three different phase velocities v(i). the corresponding displacement vectors are mutually orthogonal; they are also called polarization vectors. the vectors ( )i ju l and can be parallel, in that case the corresponding wave to that displacement direction is called pure longitudinal wave. perpendicular directions of displacement correspond to pure shear waves. ( )i ju body of the programming routine this routine plots the wave surfaces for all cases of anisotropy > restart: with(linearalgebra): unprotect(gamma,gamma): stiffness matrices stiffness matrix in the voigt form > c:=matrix(6,6,shape=symmetric): now the user must choose the anisotropy case: isotropic case (2 elastic constants) > c[1,1]:=lambda+2*mu: c[2,2]:=lambda+2*mu: c[3,3]:=lambda+2*mu: c[4,4]:=mu: c[5,5]:=mu: c[6,6]:=mu: c[1,2]:=lambda: c[1,3]:=lambda: c[2,3]:=lambda: evalm(c); cubic anisotropy (3 elastic constants) > c[1,1]:=c11: c[2,2]:=c11: c[3,3]:=c11: c[4,4]:=c44: c[5,5]:=c44: c[6,6]:=c44: c[1,2]:=c12: c[1,3]:=c12: c[2,3]:=c12: evalm(c); hexagonal anisotropy, vti (5 elastic constants) > c[1,1]:=c11: c[2,2]:=c11: c[3,3]:=c33: c[4,4]:=c44: c[5,5]:=c44: c[6,6]:=c66: c[1,2]:=c11-2*c66: c[1,3]:=c13: c[2,3]:=c13: eval(c); orthorhombic anisotropy (9 elastic constants) > c[1,1]:=c11: c[2,2]:=c22: c[3,3]:=c33: c[4,4]:=c44: c[5,5]:=c55: c[6,6]:=c66: c[1,2]:=c12: c[1,3]:=c13: c[2,3]:=c23: evalm(c); monoclinic anisotropy (13 elastic constants) 65 miguel duarte et al. 66 > c[1,1]:=c11: c[1,2]:=c12: c[1,3]:=c13: c[1,4]:=c14: c[1,5]:=c15: c[1,6]:=c16: c[2,2]:=c22: c[2,3]:=c23: c[2,4]:=c24: c[2,5]:=c25: c[2,6]:=c26: c[3,3]:=c33: c[3,4]:=c34: c[3,5]:=c35: c[3,6]:=c36: c[4,4]:=c44: c[4,5]:=c45: c[4,6]:=c46: c[5,5]:=c55: c[5,6]:=c56: c[6,6]:=c66: evalm(c); ray director cosines matrix > l:=matrix(6,3): l[1,1]:=l[1]: l[2,2]:=l[2]: l[3,3]:=l[3]: l[4,2]:=l[3]: l[4,3]:=l[2]: l[5,1]:=l[3]: l[5,3]:=l[1]: l[6,1]:=l[2]: l[6,2]:=l[1]: evalm(l); christoffel matrix > gamma:=transpose(l).c.l; eigenvalues and eigenvectors the elastic constants should be declared according to the chosen anisotropy case: consts:=c11=16.57*10^10,c12=6.39*10^10,c44 =7.956*10^10,rho=2332: (cubic anisotropy) > consts:=c11=3.46e10,c13=1.07e10,c33 =2.84e10,c44=8.36e9,c66=1.26e10,rho=2500: (vti elastic constants) >dir_l:=l[1]=sin(theta),l[2]=0,l[3]=cos(theta): >ch:=characteristicmatrix(subs(dir_l,gamma), k); > d:=determinant(ch); > factor(d); > k:=[solve(d,k)]; > map(simplify,%,trig); > a:=subs(k=k[3],ch); > a:=map(simplify,subs(k=k[3],ch),trig); > solve({a[1,1]*u1+a[1,2]*u2=0},{u1}); > simplify(%,trig); > solve({a[2,1]*u1+a[2,2]*u2=0},{u1}); > simplify(%,trig); > rank(a); > eg:=map(simplify,columnspace(a)); > plot([subs(consts,dir_l,sqrt(rho/k[1])),subs(consts ,dir_l,sqrt(rho/k[2])),subs(consts,dir_l,sqrt(rho/k[ 3]))],theta=0..2*pi,color=[red,blue,green],coords= polar); > plot([subs(consts,dir_l,sqrt(k[1]/rho)),subs(consts ,dir_l,sqrt(k[2]/rho)),subs(consts,dir_l,sqrt(k[3]/r ho))],theta=0..2*pi,color=[red,blue,green],coords= polar); > phi_1:=theta+arctan(diff(sqrt(k[1]/rho),theta)/sqrt (k[1]/rho)); > v_1:=sqrt(sqrt(k[1]/rho)^2+diff(sqrt(k[1]/rho),the ta)^2); > phi_2:=theta+arctan(diff(sqrt(k[2]/rho),theta)/sqrt (k[2]/rho)): > v_2:=sqrt(sqrt(k[2]/rho)^2+diff(sqrt(k[2]/rho),the ta)^2): > phi_3:=theta+arctan(diff(sqrt(k[3]/rho),theta)/sqrt (k[3]/rho)): > v_3:=sqrt(sqrt(k[3]/rho)^2+diff(sqrt(k[3]/rho),the ta)^2): > plot({[subs(consts,dir_l,v_1),subs(consts,dir_l,phi _1),theta=0..2*pi],[subs(consts,dir_l,v_2),subs(co nsts,dir_l,phi_2),theta=0..2*pi],[subs(consts,dir_l, v_3),subs(consts,dir_l,phi_3),theta=0..2*pi]},colo r=[red,blue,green],coords=polar); > plot({subs(consts,dir_l,sqrt(k[3]/rho)),[subs(const s,dir_l,v_3),subs(consts,dir_l,phi_3),theta=0..pi/2] },theta=0..pi/2,color=[red,blue],coords=polar); in the anterior routine the term vti means vertical and transversally isotropic media. routine work examples this program plots the three wave surfaces for any anisotropic material: slowness surface, phase velocity surface and group velocity surface. the program was tested using the next stiffness matrix for the vti case: 10 3 3.4 0.9 1.07 0 0 6 0.94 1.07 0 0 0 4 3.46 1.07 0 0 0 1.07 2.84 0 0 0 10 , 2500 0 0 0.836 0 0 0 0 0 0.836 0 0 0 0 0 0 1.26 kg c pa m ρ ⎛ ⎞ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟ = ⋅⎜ ⎟ ⎜ ⎟ ⎜ ⎟ ⎜ ⎟⎜ ⎟ ⎝ ⎠ ij = figure 1. stiffness matrix for sandstone (thomsen, 2000) conclusions the theoretical study of anisotropic media is important for studying the effect of different type of media on wave propagation. particularly the effect that fractures, typical of colombian foothills. we expect to continue our study of this concepts and the development of computational tools that simulate wave propagation in general transversal inhomogeneous media. figure 2. slowness surfaces for sandstone. horizontal axis x3. vertical axis x1. slowness given in s/m. figure 3. slowness surfaces for sandstone. horizontal axis x3. vertical axis x1. phase velocity given in m/s. acknowledgments the authors wish to thank the support received from the colombian institute of petroleum (icp), the research division of ecopetrol, in particular the research unit uin, which had let accessed laboratory data as well as information related to anisotropy. references sen, mrinal. (2002) lectures: “seismic wave propagation in anisotropic media”. the university of texas at austin. krishnaswamy, sridhar. (2004) “acousto-optical methods of materials characterization”. northwestern university. pšenčík, ivan. (1994) “introduction to seismic methods”. pppg/ufba, el salvador. mase, george e. (1970) “theory and problems of continuum mechanics”. schaum’s outline series, mcgraw-hill thomsen, leon. (2000) “weak elastic anisotropy”. applied seismic anisotropy: theory, background, and field studies. geophysics reprint series no. 20, seg. 67 with a burial depth of 1000 m as the demarcation, the coal reservoir in south yanchuan block, china is divided into deep reservoir and shallow reservoir regions. a combination of coalbed methane well production data, well logging interpretation, coalbed methane numerical simulations and reservoir properties were used to research various production characteristics at different depths. the results indicate that coal thickness and gas content are not key factors that influence methane production. the shallow reservoir is located in a tension zone, while the deep reservoir is located in both a transformation zone and a compression zone. although the reservoir and closure pressures increase with the burial depth, the pressures fluctuate in the deep reservoir, especially in the transformation zone. this fluctuation influences the opened degree of the fractures in the reservoir. the effective stress is lower in the deep reservoir than in the shallow reservoir, leading to higher permeability in the deep reservoir. this difference in effective stress is the key factor that influences the methane production. the combination of coal thickness and gas content also significantly influenced the methane production. influenced by the reservoir and closure pressures, the type iii coal in the shallow reservoir is more developed, while the deep reservoir contained more developed type i and type ii coal. the permeability increases exponentially with increasing thickness of type i and type ii coal, which determines the high reservoir permeability in the deep reservoir. the development of type iii coal leads to the poor reservoir hydraulic fracturing effect. however, a reservoir with thick type i and type ii coal can have a positive effect. con una profundidad de enterramiento de 1000 metros, el yacimiento de carbón del bloque yanchuan sur, en china, se divide en dos: el depósito profundo y el depósito superficial. este trabajo combina los datos de la información de producción de gas metano asociado carbón, la interpretación de registros de pozo, las simulaciones numéricas de metano asociado a carbón y las propiedades del reservorio para encontrar las características de producción a diferentes profundidades. los resultados indican que el espesor del carbón y el contenido de gas no son factores que alcancen a influir en la producción de metano. el depósito superficial se encuentra en una zona de tensión, mientras el depósito profundo está ubicado en una región tanto de transformación como de compresión. aunque el reservorio y la presión de cierre se incrementan con la profundidad de enterramiento, las presiones fluctúan en el depósito profundo, especialmente en la zona de transformación. esta fluctuación influye en el grado de apertura de las fracturas en el depósito. la tensión efectiva es más baja en el depósito profundo, lo que significa una mayor permeabilidad. la diferencia en la tensión efectiva es el factor clave que incide en la producción de metano. afectado por las presiones de cierre y del yacimiento, el carbón tipo iii en el depósito superficial está más desarrollado, mientras que el depósito profundo contiene carbón tipo i y tipo ii más desarrollado. la permeabilidad se incrementa exponencialmente con el incremento del espesor en el carbón tipo i y tipo ii, lo que determina la alta porosidad en el depósito profundo. el desarrollo de carbón tipo iii lleva a un pobre efecto de la fractura hidráulica en el depósito. sin embargo, un depósito con carbón tipo i y tipo ii espeso podría tener un efecto positivo. earth sciences research journal earth sci. res. j. vol. 20, no. 3 (september, 2016) : d1 d9 abstract resumen keywords: burial depth, cbm productivities, reservoir pressure, closure pressure, permeability palabras clave: explotación en profundidad, producción de gas metano asociado a carbón, presión de depósito, presión de cierre, permeabilidad. influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china record manuscript received: 13/06/2016 accepted for publication: 02/09/2016 how to cite item li, t., & wu, c. f. (2016). influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china. earth sciences research journal, 20(3), d1-d9. doi:http://dx.doi.org/10.15446/esrj.v20n2.58000 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.58000 teng li, cai-fang wu h y d r o c a r b o n s influencia de la presión, la estructura del carbón y su permeabilidad sobre la productividad de gas metano de carbón en profundidades de enterramiento del bloque yanchuan sur, china teng li, cai-fang wud2 1 introduction the production of coalbed methane from a single well is an important target to evaluate the success or failure of the coalbed methane development. however, different coalbed methane wells within the same block have different productivities. research on the factors that influence coalbed methane production show that it is a result of a coupling action, and the major factors can be restricted to three types: geological, development techniques and production management. geological factors include the reservoir burial depth, thickness, gas content, permeability, reservoir pressure, tectonic and hydrogeological conditions, all of which can radically influence the reservoir properties (kaiser et al., 1994; pashin and groshong jr., 1998; karacan et al., 2008; yao et al., 2009a; lv et al., 2012). permeability is a major factor used to evaluate the production of coalbed methane from a single well (li et al., 2011). the natural fractures in the coal are passageways for gas movement (perera et al., 2013a). the movement of gas is the result of a coupling of gasliquid-solid flows (wu et al., 2014), which has a great deal to do with the gas pressure and concentration (thararoop et al., 2012). the coal matrix swells with increasing gas pressure, especially under low gas pressure conditions (approximately 3.5 mpa), leading to the reduction in coal permeability (perera and ranjith, 2012). although the coal matrix becomes weaker with higher gas pressure, the newly created micro-fractures provide space for gas adsorption, which further enhances the coal matrix swelling (viete and ranjith, 2006). however, the increased liquid pressure may result in shrinking of the coal matrix (perera and ranjith, 2012). tectonic movements can promote coalbed methane preservation and secondary hydrocarbon generation (yao et al., 2009a), but the deformation of the coal mass structure associated with these movements should not be ignored (li, 2001; li et al., 2003). the thin-skinned structure can influence the production of coalbed methane wells (pashin and groshong jr., 1998), and the impermeable faults also affect the flow of gas and water from the coal reservoir to the well bores (karacan et al., 2008). aside from tectonic movements, gas pressure can also deform the coal mass structure, with higher gas pressure being associated with more serious deformation of the coal structure (perera et al., 2013b). the mechanical properties of the coal also contribute to the permeability, mainly as a result of the dynamic young’s modulus during the adsorption of gas, especially regarding the adsorption of co2 (perera et al., 2013c). when the gas desorbs from the surface of the coal matrix, the elastic modulus controls the coal-stress sensitivity (karacan, 2009; tao et al., 2012). research on the pore structure of 34 coal samples with various maturity from north china has shown that coal permeability is close to the coal pore fractal, especially for middleand high-ranked coal (ro,max = 1.47-4.21 %). this permeability may be influenced by the seepage pores and degree of coalification (yao et al., 2009b; meng and li, 2013). the production of coalbed methane is a coupled process, that involves the desroption and flow of methane, deformation of the coal structure, dynamic change of coal porosity and permeability (mazumder et al., 2012; tao et al., 2012; liu and harpalani, 2013; peng et al., 2014; qu et al., 2014). additionally, any external fluid could influence the reservoir permeability during fracturing (gentzis et al., 2009). essentially, dynamic permeability implies that the state of the coal reservoir varies from the initial to the final time periods (qu et al., 2014). during the production process, the effective stress of the coal reservoir and the shrinking of the coal matrix will influence the absolute permeability of the reservoir (durucan and edwards, 1986; laubach et al., 1998; chen et al., 2010; liu et al., 2010; tao et al., 2012; liu and harpalani, 2013). the permeability of the fairway wells in the san juan basin varies by approximately 100 times over the production period (palmer, 2009; mazumder et al., 2012). a study on the production of the coalbed methane wells in the south qinshui basin showed that the burial depth of the coal reservoir, the bottom bore pressure and the dewatering rate are the three main factors that control the effective stress of the coal reservoir, which further influences the absolute permeability (tao et al., 2012). when methane outputs from the bottom of the bore, the extensive underground water does not favor a reduction in pressure, and this water contributes to the groundwater flexibility energy (wu et al., 2007). this energy can keep the fractures in the coal reservoir open, thereby increasing the permeability of the reservoir, and increasing the performance of the coalbed methane wells (somerton et al., 1975; st. george and barakat, 2001; yao et al., 2009b). the decrease of bottom bore pressure promotes the desorption of coalbed methane, while the rapid decrease of pressure may lead to the gas locking in the reservoir, resulting in reduced flow rates (tao et al., 2012). the effective stress increases with increasing burial depth (mazumder et al. 2012), and permeability is more reduced in a deep reservoir is higher than that in a shallow reservoir (tao et al., 2012). conversely, lv et al. (2012) reported that methane production in a deep coal reservoir is higher than that in a shallow coal reservoir, mostly as a result of the coal thickness and gas content. however, this study did not consider the change in stress state with increasing reservoir burial depth. in the south yanchuan block, with a burial depth of 1000 m serving as the demarcation, the reservoir stress characteristics have changed from σv > σhmax > σhmnin to σhmax > σv> σhmnin (wu, 2012). this finding means that the reservoir stress state has changed from tensile to compressive, and there has been an obvious change in coalbed methane production. the methane production at various depths is differs, and this may be influenced by the change in reservoir stress state. in this paper, using the south yanchuan block as an example, the reservoir properties that influence coalbed methane production at various depths are discussed, such as pressure and the production characteristics of coalbed methane wells, characteristics and differences of coal reservoir and closure pressures, and coal structure and permeability. 2 methods 2.1 geological setting the south yanchuan block is situated in the southeast of ordos basin, china. the study area is a monoclinal structure with a northeast trend and a western dip. although the ordos basin has experienced repeated geologic history, specifically the indosinian movement, the yanshan movement and the himalayan movement, the structure in the study area is simple. the structure can be divided into five third-class structural units, the xibaigou gentle slope belt, the bai’e fault nose structure belt, the tanping gentle slope belt, the baishansi fault nose structure belt, and the wanbaoshan gentle slope belt are distributed from east to west (fig. 1) (cai et al., 2014; li et al., 2015; yan et al., 2015). the south yanchuan block was compressed during the three main tectonic movements: ns trending, nw-ne trending, and ne-sw trending movements. the ns trending stress from the indosinian period established the basement structure. the nw-ne trending stress from the yanshanian period mainly controlled the structural characteristics, including the four large-scale faults and numerous small-scale faults. the magma activity in yanshanian period was active, albeit insignificant in the south yanchuan block (li et al., 2015). the hydrodynamics in the study area are complex, including a strong runoff and recharge zone, a weak runoff zone, a stagnant zone, a strong runoff zone, a weak stagnant zone and a deep stagnant zone, situated from the east to west (cai et al., 2014). the lower permian shanxi formation (p1s) no. 2 coal seam is the target stratum for the development of coalbed methane. influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china d3 2.2 samples test and data processing there have been approximately 60 coalbed methane wells in the research area. however, most of these wells are in initial production stages, exhibiting unstable productivity. in order to ensure that the research is scientificity, coalbed methane wells that have stably produced for an excess 150 days were selected. the proximate analysis of coal samples in this study utilizes the instrumental method of proximate analysis of coal (gb/t 307322014, chinese national standard). the ro, max tests were carried out using whole-rock maceral analysis (sy/t 6414-1999, chinese petroleum and natural gas industry standard), and at least 800 points were tested. reservoir permeability can be acquired from injection/fall-off well tests, experimental test, history matching, geophysical log and constructing permeability prediction models (palmer, 2009; li et al., 2011; ma et al., 2011; thararoop et al., 2012; meng and li, 2013; perera et al., 2013b; peng et al., 2014). in this paper, the coal reservoir permeability is acquired from the injection/falloff well test, and the fitting permeability is acquired from the comet3 coalbed methane numerical simulation software. comet3 is a three-dimensional, threecomponent, two-phase, dual-porosity simulator for modeling gas and water production from a desorption-controlled coal reservoir (advanced resources international, 2005). the diffusion, desorption theory and dual-porosity/ single-permeability model are used for the coalbed methane simulation. the following equations are used to calculate the conservation of gas and water. conservation of gas: (1) conservation of water: (2) where, f indicates fracture systems, ∆ is the gradient operator, ∆• is the divergence operator, bn (n = g or w) is the gas or water shrinkage factor, γn (n = g or w) is the gas or water gradient, rsw is gas solubility in water, φ is effective fracture porosity and z is the elevation. during the simulation process, certain reservoir parameters should be inputted, including coal thickness, burial depth, reservoir pressures, porosity, permeability, gas content, sorption time, langmuir volume, langmuir pressure, and aquifers. all of these parameters can be acquired from engineering and experimental tests. if the permeability from the injection/ fall-off well test is influenced by the external fluid, then the permeability is the domain parameter that will be fitted in the simulation software. coal structure of reservoir was explained using logging data (fu et al., 2009). the coal structure is divided into three types, type i coal is primary texture coal, type ii coal is cataclastic texture coal, and type iii coal is tectonic coal. 3 results and discussion 3.1 coalbed methane well production characteristics fifteen coalbed methane production wells are studied in this paper. as shown in fig. 2, different wells have various methane and water production characteristics. according to the research of cai et al. (2014), when methane wells are located in the underground water weak-runoff and stagnant zones, the influence of underground water may be weak. the syc-1 and syc-6 wells wxhibit high methane production, with the maximum methane production of syc-6 exceeding 2000 m3/d, while the methane productions of syc-2 and syc10 are low. the output of water decreases the reservoir pressure, allowing for the methane to desorb from the coal. although the water production fluctuates, the dominant trend is decreasing. it was found that there were significant depressions in the liquid levels at syc-1 and syc-6 after the bores began to produce methane, followed by a leveling-out of the liquid levels. however, the liquid levels of syc-2 and syc-10 remained steady, indicating that the seepage of liquid is not smooth, a trait that may be related to the coal reservoir permeability. figure 2. production curves of typical coalbed methane wells in the south yanchuan block (the magenta line is the liquid level, the red line is the average methane production, and the blue line is the water production) figure 1. location of coalbed methane block 3.3 characteristics of reservoir and closure pressures influenced by depth generally, the crustal stress will increase with increasing burial depth. due to the shift in the stress state at 1000 m, the reservoir and closure pressures present typical characteristics. the reservoir in the research area is generally under pressure, with the shallow reservoir pressure is particularly low (fig. 4a). these low pressures combined with low gas saturation do not favor high production of coalbed methane. at depths of 1050 to 1150 m, the reservoir pressure fluctuates, indicating that depth is within the stress transformation zone (fig. 4a). additionally, the reservoir pressure gradient is low, and this is mainly teng li, cai-fang wud4 3.2 influence of coal properties on methane production as mentioned above, there are many factors that can influence the production of coalbed methane wells, particularly the permeability and thickness of coal and the gas content (kaiser et al., 1994; lv et al., 2012). generally, the higher the methane content and coal thickness, the higher the coalbed methane production (lv et al., 2012); however, that is not obvious in south yanchuan block. as shown in fig. 3d, the relationship between gas content and coal thickness in terms of average methane production is weak, meaning that these two factors are not the key controllers of the methane production. with increasing burial depth and coal reservoir permeability, the average methane production increases (fig. 3a, c), especially when the data from syc-3 and syc8 are removed (fig. 3b). the low methane production of these two bores will be explained later. when the burial depth exceeds 1000 m, the methane production is quiet high (fig. 3a, b). this phenomenon may be influenced by the shift of the stress state. to facilitate the discussion, the reservoir burial depth above 1000 m is referred to as the shallow reservoir, while the reservoir burial depth beyond 1000 m is referred to as the deep reservoir. the average methane production in the shallow reservoir is less than 700 m3/d, while the methane production always exceeds 500 m3/d in the deep reservoir, where it reaches 1759.37 m3/d (table 1). figure 3. relationship between burial depth, permeability, coal thickness and gas content with the average methane production types wells burial depth (m) coal thickness (m) gas content (m3/t, ad) ro, max permeability (md) mad (%) aad (%) vad (%) average methane production (m3/d) shallow reservoir syc-1 936.20 4.90 8.28 2.38 0.1735 1.48 14.58 11.40 1018.10 syc-2 898.80 5.10 7.60 2.30 0.0605 0.83 7.49 10.50 118.08 syc-4 896.00 5.10 6.98 2.19 1.09 26.04 11.16 679.96 syc-5 876.50 4.60 6.87 2.30 1.02 4.98 9.77 1253.36 syc-7 983.50 5.30 4.80 2.06 0.0490 0.64 12.86 10.20 154.36 syc-10 895.80 6.10 8.56 2.24 0.0170 0.68 14.76 10.86 267.70 syc-11 890.80 5.10 9.40 1.96 0.0281 252.95 syc-12 907.10 6.90 4.14 0.86 31.27 10.09 67.13 syc-13 928.70 5.28 8.02 2.35 0.0188 0.95 11.01 11.79 564.66 syc-14 916.40 4.28 9.50 2.46 0.98 9.30 9.46 768.21 syc-15 892.40 4.90 8.31 2.24 0.75 9.12 8.55 138.33 deep reservoir syc-3 1169.10 5.00 18.21 2.52 0.1698 0.85 11.91 8.81 544.89 syc-6 1259.40 4.80 11.60 2.57 0.1698 0.67 14.25 8.75 1759.37 syc-8 1242.40 4.30 5.10 2.02 0.89 8.81 9.67 182.75 syc-16 1064.58 5.82 14.44 2.74 0.9504 1.49 11.00 8.70 1288.86 syc-17 1498.10 3.90 10.63 3.08 0.4132 0.99 8.58 10.98 syc-18 1111.60 2.40 6.44 0.7529 0.81 7.32 8.17 syc-20 1129.31 3.94 10.74 2.65 0.0044 0.94 27.53 9.50 note: ad, the air-dry basis; -, no data. table 1. statistics of reservoir property parameters in south yanchuan, block due to the deep stagnant underground water zone (fig. 4a) (cai et al., 2014). the closure pressure also presents the same relationship with the burial depth, where the closure pressure in the transformation zone is quite fluctuant, and the maximum closure pressure of syc-20 is even higher than in deep reservoir (fig. 4b). in the deep reservoir, while the closure pressure has increased, it is not obvious (fig. 4b). where the burial depth of the reservoir exceeds 1000 m, especially in the compression zone, the crustal stress turns to compressive stress, and the closure and reservoir pressure are both high in the deep reservoir. this high reservoir pressure can maintain high stratum energy (wu et al., 2007), and the high closure pressure can keep the fracture open (perera and ranjith, 2012). with increasing the reservoir burial depth, the reservoir effective stress influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china d5 increases, and the fracture network system tends to close (laubach et al., 1998; yao et al., 2009a), which would lead to an exponential decrease in permeability (somerton et al., 1975; gray, 1987). the permeability decreases with the increase in effective stress (fig. 5a), due to the coupling of reservoir and closure pressures. the effective stress is also divided into three sections (fig. 5b). the low closure pressure gradient indicates high permeability in the deep reservoir. under the interaction between the low closure pressure gradient and the high reservoir pressure gradient, the permeability of the deep reservoir is generally more than 0.1 md. this observation has validated the conjecture, which means the coalbed methane geological condition is superior in the deep reservoir. 3.4 coal structure characteristics influenced by burial depth 3.4.1 coal structure proportion the statistical results show that in the shallow reservoir, the type i coal proportion reaches 38.18 %, the type ii coal proportion reaches 26.41 %, and the type iii coal proportion reaches 35.41 % (table 2). in the deep reservoir, the type i coal proportion reaches 25.25 %, the type ii coal proportion reaches 39.69 %, and the type iii coal proportion reaches 35.06 % (table 2). the type iii coal is mainly located in the transformation zone coal reservoir. if this section of the coal reservoir is ignored, the proportion of type i, ii and iii coal in the deep reservoir will reach 30.14 %, 50.14 % and 19.71 % respectively. high proportions of types i and ii coal is a good basis for high permeability in the deep reservoir (li et al., 2015). although the tectonics are simple in the research area, the deep reservoir (especially in the compression zone) coalbed methane wells lie in monoclinal structure. however, the shallow coalbed methane wells lie in the core of the secondary syncline or the wing figure 4. relationship between reservoir and closure pressures with the burial depth figure 5. relationship between effective stress and reservoir permeability of the secondary anticline, which would decrease permeability, and can be a disadvantage to the coalbed methane production. the development of type i coal can maintain the coal mass structure, while the development of type ii coal can increase the permeability of the coal reservoir (li et al., 2015), and this effect is the main reason for the low methane production at syc-3 and syc-8. the deep reservoir is mainly composed of the type i and ii coal structures, it has high stratum energy, and the production of coalbed methane is steady. with aslow increase of the effective stress, the fluid can discharge steadily, and the methane production effect is positive (table 3). coalbed methane production in the shallow reservoir varies, and the low reservoir pressure results in high effective stress. with the continuing production, the effective stress increases with decreasing reservoir pressure, and the fractures are rapidly closed. additionally, the development of type iii coal can make the migration of methane difficult, and can result in the methane wells’ having low to no productivity (table 3). 3.4.2 coal thickness generally, the higher the coal thickness, the larger the gas content the coal seam can maintain (yao et al., 2009a; lv et al., 2012). however, there is no marked correlation between coal thickness and coalbed methane production (fig. 3). in the shallow reservoir, the relationship between coal thickness and coalbed methane production is even negative (fig. 6a); mainly due to the development of type iii coal. the development of type iii coal can damage the reservoir permeability (li et al., 2015), and desorption of methane adsorbed at the surface of the coal matrix is difficult, resulting in low methane production. when the coal thickness is adjusted to the thickness of type i and ii coal, the methane production increases exponentially (fig. 6b). teng li, cai-fang wud6 table 2. statistics of coal structure with the logging data in south yanchuan, block types wells type i coal type ii coal type iii coal fitting permeability (md)thickness (m) ratio (%) thickness (m) ratio (%) thickness (m) ratio (%) shallow reservoir syc-1 4.90 100.00 0.00 0.00 0.00 0.00 syc-2 0.00 0.00 0.00 0.00 5.10 100.00 0.12 syc-4 0.00 0.00 0.00 0.00 5.10 100.00 syc-5 2.70 58.70 1.90 41.30 0.00 0.00 3.22 syc-7 0.00 0.00 4.10 100.00 0.00 0.00 syc-10 0.00 0.00 3.30 55.93 2.60 44.07 0.19 syc-11 5.10 100.00 0.00 0.00 0.00 0.00 1.71 syc-12 0.65 10.48 2.00 32.26 2.20 57.26 syc-13 5.30 100.00 0.00 0.00 0.00 0.00 0.81 syc-15 0.00 0.00 1.60 41.03 2.30 58.97 0.52 deep reservoir syc-3 4.50 100.00 0.00 0.00 0.00 0.00 0.29 syc-6 0.00 0.00 4.80 100.00 0.00 0.00 4.23 syc-8 0.70 17.28 1.75 43.21 1.60 39.51 0.8 syc-16 2.72 41.72 3.80 58.28 0.00 0.00 5.18 syc-17 0.00 0.00 2.10 53.85 1.80 46.15 syc-18 0.00 0.00 0.00 0.00 4.20 100.00 syc-20 0.00 0.00 0.00 0.00 3.40 100.00 0.08 note: -, no data. table 3. coalbed methane production characteristics in south yanchuan, lock types wells water drainage characteristics methane production characteristics shallow reservoir syc-1, syc2, syc-4, syc5,syc-11, syc-13, syc-14, syc-15 high maximum water production (4.02-10.13 m3/d), and water production is unstable. the water production should be regulated, then the methane production can be maintained. normally, with the decrease of the water production, the methane production gets down (118.08-1253.36 m3/d). syc-7, syc-10, syc-12 the maximum and average water produciton is low, 2.3-14.37 m3/d and 0.83-1.41 m3/d respectively. no stable methane production, even non methane produciton (67.13-267.7 m3/d). deep reservoir syc-8 high maximum and average water production, 10.80 m3/d and 5.15m3/d respectively. no stable methane production (182.75 m3/d). syc-3, syc-6, syc-16 high maximum water production (4.02-12.4 m3/d), low average water production (0.27-3.07 m3/d). methane production is stable and high (544.89-1759.37 m3/d). figure 6. relationship between coal thickness and average methane production the coal thickness also influences gas content, and the coalbed methane content unit thickness (cbmcut) can be used (lei et al., 2012). (3) where, is cbmcut, m3/(t·m); c is gas content, m3/t; h is apparent thickness, m; and α is dip angle, °. figure 7. relationship between cbmcut and average methane production cbmcut is a combined reflection of coal thickness and gas content; it increases exponentially with average methane production (fig. 7). cbmcut also has a relationship with the coal structure, and decreases with increasing coal thickness. this effect is mainly due to influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china d7 the resistance of tectonic deformation of thick coal; with thinner coal (1-2 m) more easily deformed than thick coal (> 5 m) (lei et al., 2012). 3.5 permeability in the shallow reservoir, the development of type iii coal results in the low permeability, as values are always below 0.1 md (table 1, fig. 5). the permeability of syc-11 and syc-13 coalbed methane wells with the type i coal is still low, mainly due to low reservoir pressure, and insufficient stratum energy to keep the fracture open. the deep reservoir has a single coal structure is, and the permeability is higher than that in the shallow reservoir. the development of type i and ii coal can maintain high permeability, which indicates that the type ii coal can promote the increase of permeability (fig. 8). figure 8. relationship between coal structure and reservoir permeability there is a positive exponential correlation between the fitting permeability and the thickness of type i and ii coal (fig. 9). type iii coal in the reservoirs of syc-2, syc-4, syc-7, syc-10 wells is more developed. the coal structure is complicated, and the incremental multiplier of reservoir permeability is limited after the hydrofracture, which means the fracturing effect is not obvious with the development of tectonic coal (table 2). with the development of type i coal in the reservoir of syc-11 and syc-13 wells, the permeability increased dramatically (table 2). except for the syc-3 well, the fitting permeability in the deep reservoir increased, and the development of a dirt band in the reservoir decreases the fracturing effect. this high permeability means higher methane production (fig. 2). figure 9. relationship between thickness of type i and ii coal and permeability teng li, cai-fang wud8 4 conclusions the average methane production in south yanchuan block increases with increasing coal reservoir burial depth, and coal thickness and gas content are not the key factors influencing methane production. the shift of the stress state makes the reservoir and closure pressure fluctuate, which leads to various coal structure characteristics in the shallow and deep reservoirs, and finally influences the coal reservoir permeability that controls the coalbed methane production. the reservoir in the research area is under pressure, and the shallow reservoir pressure is drastically low, which results in high effective stress. high closure pressure in the deep reservoir keeps the fracture open, and the reservoir permeability is superior. however, the reservoir and closure pressures in the transformation zone in the deep reservoir are fluctuant. the reservoir and closure pressures also influenced the coal structure. the type iii coal in the shallow reservoir is quite developed, while there is type i and ii coal in the deep reservoir. the combination of coal thickness and gas content can significantly influence the average methane content, and this also has a great deal to do with the coal structure. there is a positive exponential correlation between the fitting permeability and the thickness of type i and ii coal. the coalbed methane numerical simulation results show the development of type iii coal can decrease the fracturing effect, while it increases with the thickness of type i and ii coal. conversely, the development of type iii coal causes the methane production to be unsteady, which is a disadvantage to methane productivity. acknowledgements this work was supported by the national major special project of science and technology of china (2016zx05044001), the natural science foundation of china (41272178, 41572140), the fundamental research funds for the central universities (2014zdpy26, 2015xkzd07), the qing lan project and the research and innovation project for college graduates of jiangsu province (grant no. kylx15_1392). references advanced resources international (2005). users manual comet3 version 2.0. advanced resources international, washington, dc, usa. cai, y., liu, d., zhang, k., elsworth, d. yao, y. & tang, d. (2014). preliminary evaluation of gas content of the no. 2 coal seam in the yanchuannan area, southeast ordos basin, china. journal of petroleum science and engineering, 122, 675-689. chen, z., liu, j., elsworth, d., connell, l. d. & pan, z. (2010). impact of co2 injection and differential deformation on co2 injectivity under in situ stress conditions. international journal of coal geology, 81(2), 97-108. durucan, s. & edwards, j. s. (1986). the effects of stress and fracturing on permeability of coal. mining science and technology, 3(3), 205-216. fu, x., qin, y., wang, g. g. x. & rudolph, v. (2009). evaluation of coal structural and permeability with the aid of geophysical logging technology. fuel, 88, 2278-2285. gentzis, t., deisman, n. & chalaturnyk, r. j. (2009). effect of drilling fluids on coal permeability: impact on horizontal wellbore stability. international journal of coal geology, 78(3), 177-191. gray, i. (1987). reservoir engineering in coal seams: part 1 the physical process of gas storage and movement in coal seams. spe reservior engineering, 2, no. 1, 28-34. kaiser, w. r., hamilton, d. s., scott, a. r., tyler, r. & finley, r. j. (1994). geological and hydrological controls on the producibility of coalbed methane. journal of the geological society, 151, 417-420. karacan, c. o. (2009). elastic and shear moduli of coal measure rocks derived from basic well logs using fractal statistics and radial basis functions. international journal of rock mechanics and mining sciences, 46(8), 1281-1295. karacan, c. o., ulery, j. p. & goodman, g. v. r. (2008). a numerical evaluation on the effects of impermeable faults on degasification efficiency and methane emissions during underground coal mining. international journal of coal geology, 75(4), 195-203. laubach, s. e., marrett, r. a., olson, j. e. & scott, a. r. (1998). characteristics and origins of coal cleat: a review. international journal of coal geology, 35(1-4), 175-207. lei, b., qin, y., gao, d., fu, x., wang, g. g. x., zou, m. & shen, j. (2012). vertical diversity of coalbed methane content and its geological controls in the qingshan syncline, western guizhou province, china. energy exploration & exploitation, 30(1), 43-58. li, h. (2001). major and minor structural features of a bedding shear zone along a coal seam and related gas outburst, pingdingshan coalfield, northern china. international journal of coal geology, 47(2), 101-113. li, h., ogawa, y. & shimada, s. (2003). mechanism of methane flow through sheared coals and its role on methane recovery. fuel, 82(10), 1271-1279. li, j., liu, d., yao, y., cai, y. & qin, y. (2011). evaluation of the reservoir permeability of anthracite coals by geophysical logging data. international journal of coal geology, 87, no. 2, 121-127. li, t., wu, c. & liu, q. (2015). characteristics of coal fractures and the influence of coal facies on coalbed methane productivity in the south yanchuan block, china. journal of natural gas science and engineering, 22, 625-632. liu, j., chen, z., elsworth, d., miao, x. & mao, x. (2010). evaluation of stress-controlled coal swelling processes. international journal of coal geology, 83(4), 446-455. liu, s. & harpalani, s. (2013). permeability prediction of coalbed methane reservoirs during primary depletion. international journal of coal geology, 113, 1-10. lv, y., tang, d., xu, h. & luo, h. (2012). production characteristics and the key factors in high-rank coalbed methane fields: a case study on the fangzhuang block, southern qinshui basin, china. international journal of coal geology, 96-97, 93-108. ma, q., harpalani, s. & liu, s. (2011). a simplified permeability model for coalbed methane reservoirs based on matchstick strain and constant volume theory. international journal of coal geology, 85(1), 43-48. mazumder, s., scott, m. & jiang, j. (2012). permeability increase in bowen basin coal as a result of matrix shrinkage during primary depletion. international journal of coal geology, 96-97, 109-119. meng, z. & li, g. (2013). experimental research on the permeability of high-rank coal under a varying stress and its influencing factors. engineering geology, 162, 108-117. palmer, i. (2009). permeability changes in coal: analytical modeling. international journal of coal geology, 77(1-2), 119-126. pashin, j. c. & groshong, jr., r. h. (1998). structural control of coalbed methane production in alabama. international journal of coal geology, 38, no. 1-2, 89-113. peng, y., liu, j., zhu, w., pan, z. & connell, l. (2014). benchmark assessment of coal permeability models on the accuracy of permeability prediction. fuel, 132, 194-203. perera, m. s. a. & ranjith, p. g. (2012). carbon dioxide sequestration effects on coal’s hydro-mechanical properties. international journal of energy research, 36(10), 1015-1031. perera, m. s. a., ranjith, p. g. & choi, s. k. (2013a). coal cleat permeability for gas movement under triaxial, non-zero lateral strain condition: a theoretical and experimental study. fuel, 109, 389-399. perera, m. s. a., ranjith, p. g., choi, s. k. & bouazza, a. (2013c). a parametric study of coal mass and cap rock behavior and carbon dioxide flow during and after carbon dioxide injection. fuel, 106, 129-138. perera, m. s. a., ranjith, p. g. & viete, d. r. (2013b). effects of gaseous and super-critical carbon dioxide saturation on the mechanical properties of bituminous coal from the southern sydney basin. applied energy, 110, 73-81. qu, h., liu, j., pan, z. & connell, l. (2014). impact of matrix swelling area propagation on the evolution of coal permeability under coupled multiple processes. journal of natural gas science and engineering, 18, 451-466. somerton, w. h., soylemezoglu, i. m. & dudley, r. c. (1975). effect of stress on permeability of coal. international journal of rock mechanics and mining sciences & geomechanics abstracts, 12(5-6), 129-145. st., george, j. d. & barakat, m. a. (2001). the change in effective stress with shrinkage from gas desorption in coal. international journal of coal geology, 45(2-3), 105-113. tao, s., wang, y., tang, d., xu, h., lv, y., he, w. & li, y. (2012). dynamic variation effects of coal permeability during the coalbed methane development process in the qinshui basin, china. international journal of coal geology, 93, 16-22. thararoop, p., karpyn, z. t. & ertekin, t. (2012). development of a multimechanistic, dual-porosity, dual-permeability, numerical flow model for coalbed methane reservoirs. journal of natural gas science and engineering, 8, 121-131. viete, d. r. & ranjith, p. g. (2006). the effect of co2 on the geomechanical and permeability behaviour of brown coal: implications for coal seam co2 sequestration. international journal of coal geology, 66(3), 204-216. wu, c. f., qin, y. & fu, x. h. (2007). stratum energy of coal-bed gas reservoir and their control on the coal-bed gas reservoir formation. science in china series d: earth sciences, 50(9), 1319-1326. wu, c. f., qin, y. & zhou, l. g. (2014). effective migration system of coalbed methane reservoirs in the southern qinshui basin. science china earth sciences, 57(12), 2978-2984. wu, q. (2012). study on in-situ stress condition of coal reservoir and effects on cbm well production in yanchuannan block, ordos basin. ph.d. thesis, department of geology, northwest university, xi’an, china. yan, t., yao, y. & liu, d. (2015). critical tectonic events and their geological controls on gas generation, migration, and accumulation in the weibei coalbed methane field, southeast ordos basin, journal of natural gas science and engineering. 27, 1367-1380. yao, y., liu, d., tang, d., tang, s., che, y. & huang, w. (2009a). preliminary evaluation of the coalbed methane production potential and its geological controls in the weibei coalfield, southern ordos basin, china. international journal of coal geology, 78(1), 1-15. yao, y., liu, d., tang, d., tang, s., huang, w., liu, z. & che, y. (2009b). fractal characterization of seepage-pores of influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china d9 articulo 5.indd earth sciences research journal archaeology earth sci. res. sj. vol. 15, no. 2 (december, 2011): 115 120 quaternary dating by electron spin resonance (esr) applied to human tooth enamel eduar carvajal1, luis montes2 and ovidio a. almanza1 1 departamento de física, universidad nacional de colombia, cra 30 nº 45-03, bogotá 2 departamento de geociencias, universidad nacional de colombia. cra 30 nº 45-03, bogotá. corresponding author: ovidio almanza m, tel. (57 1) 3165000 ext 13048; fax (57 1) 3165635. e-mail oaalmanzam@unal.edu.co abstract this paper presents the results obtained from using electron paramagnetic resonance (epr) to analyse tooth enamel found at the aguazuque archaeological site (cundinamarca, colombia), located on the savannah near bogota at 40 37’ north and 74°17’ west. it was presumed that the tooth enamel came from a collective burial consisting of 23 people, involving men, women and children. the tooth enamel was irradiated with gamma rays and the resulting free radicals were measured using an electron spin resonance (esr) x-band spectrometer to obtain a signal intensity compared to absorbed doses curve. fitting this curve allowed the mean archaeological dose accumulated in the enamel during the period that it was buried to be estimated, giving a 2.10 ± 0.14 gy value. rosy software was used for estimating age, giving a mean 3,256 ± 190y before present (bp) age. these results highlight epr’s potential when using the quaternary ancient ruins dating technique in colombia and its use with other kinds of samples like stalagmites, calcite, mollusc shells and reefs. resumen en este trabajo se presentan los resultados obtenidos al datar por resonancia paramagnética electrónica muestras de esmalte dental provenientes del sitio arqueológico aguazuque, ubicado en la sabana de bogotá en coordenadas 40 37’ de latitud norte y 74°17’ de longitud oeste de greenwich. se presume que el esmalte dental proviene de un entierro colectivo compuesto de 23 individuos entre mujeres, hombres y niños. el esmalte dental fue irradiado con rayos gamma y los radicales libres producidos fueron medidos usando un espectrómetro de resonancia del spin electrónico (esr), banda x, con el fin de realizar una curva de intensidad de la señal versus la dosis absorbida por el esmalte. el ajuste de esta curva permitió obtener la dosis media arqueológica acumulada en el esmalte (d a), durante el periodo de enterramiento de la muestra. el valor obtenido fue de 2.10 ± 0.14 gy. para estimar la edad de la muestra se usó el software rosy, obteniendo una edad media de las muestras de esmalte de 3256 ± 190y bp. los resultados obtenidos muestran el potencial uso de la técnica epr como herramienta de datación de muestras cuaternarias en colombia y abre la posibilidad de usarla en otro tipo de muestras como estalagmitas, calcitas, conchas de moluscos y corales. palabras claves: epr, esmalte dental, datación arqueológica. keywords: epr/esr, tooth enamel, colombia, archaeological dating. record manuscript received: 11/05/2011 accepted for publications: 30/11/2011 introduction electron spin resonance (esr) analysis has recently become an alternative c14 and thermo-luminescence dating method which can be applied to a variety of problems in geology, archaeology and paleoanthropology (renfrew c et al., 1992; jonas 1997; walker 2005; grun 1989). esr spectroscopy is the only method for detecting, identifying and quantifying free radicals (halliwell & gutteridge 2007; punchard & kelly 1996). for instance, this method is used for estimating the free radical level and paramagnetic centres produced in some materials by ionising radiation. the some free radicals’ long mean life (above 100 my) make it possible to reveal them in materials as old as 2 my without the risk of overcoming the sample’s saturation limits. esr allows some materials to be dated because free radical concentration is a measurement of the total radiation dose absorbed by the sample during the time it was exposed to radiation and hence their burial time (jonas 1977). esr has been used for dating the formation and more recent re-crystallisation of three types of gypsum samples: massive, bedded and fracture filling gypsum; in this case, the estimated upper miocene–pliocene age agreed with that soil’s stratigraphy (ülkü 2004). molodkov (2001) used esr to date eight terrestrial shell samples from treugolnaya (triangular) cave (northern caucasus), ages ranging from 583,000 to 393,000y old. shigueo et al., (2003) dated a calcite formation found on a rock wall painting at toca da bastiana, rock shelter in serra da capivara national park, piaui, brazil, obtaining an age of 35,000 to 43,000y, indicating that humans lived there before 35,000 years ago (this being the date for the first humans to arrive in brazil). otheduar carvajal, luis montes and ovidio a. almanza116 er papers dating stalagmites, calcite, mollusc shells and reefs are renfrew et al., (1992), jonas (1997), walker (2005) and grun (1989). the present study was aimed at using the esr method for dating tooth enamel from the aguazuque archaeological site in soacha, colombia. theoretical aspect the aguazuque archaeological site (cundinamarca, colombia) is located on the savannah near bogota at 40 37’ north and 74°17’ west (figure 1). the human tooth used in this work was extracted from a skeleton (labelled ag-i) taken from a collective burial site where 23 people were found arranged in a foetal position; men, women and children were identified. the burial site was circular shaped (4.5 m diameter) and as burnt bones and lacerated skeletons were found outside the circle, partially painted red and black, it was deduced that cannibal practices were common by the group responsible for the burial (correal 1990). archaeological research concerning the early and middle holocene period has led to figure 1. location of aguazuque archaeological site (cundinamarca, colombia, 40 37' north and 74°17' west). (taken from groot, a.m., 1992). figure 2. energy levels of an electron placed in a magnetic field. the arrow shows the transitions induced by microwave radiation. reconstructing the way of life for human groups in the region, thereby showing a transition from hunters and gatherers to agricultural societies (groot 1992). a way of life based on hunting several sized mammals has been recorded during a lapse of 11,000 to 5,000 years before today, with rock shelters providing favourites housing places, such as the archaeological sites of el abra, tequendama and nemocón 4 in cundinamarca (correal 1979). sites such as chía i (ardila 1984) and aguazuque (correal 1990) have been investigated and the aforementioned research has led to establishing changes in the form of settlement and adaptation mechanisms (groot, 1992). electron spin resonance (esr) consists of the resonant absorption of electromagnetic energy during electron-spin transitions. a static magnetic field should be applied to resolve different electron-spin levels. unpaired electrons from free radicals have spin equal to ½. there are two magnetic levels (+½ and –½) in a magnetic field, both having different energy (figure 2); the transition between these two levels is possible in the following resonance condition: hν=gβh where ν is resonance frequency, h is plank’s constant, g is the g-factor (which is a constant = 2 for ½ spin), β is the bohr magneton which is an electron magnetic dipole moment and h is the magnetic field. an important conclusion derived from this formula is the linear dependence between the applied magnetic field and resonance frequency. the most frequently used microwave energies drop in what is called the x band. this is because it is a good compromise between sensitivity, sample size and water content effects; ν ≈ 9.8 ghz and h ≈ 350 mt for the x band. esr dating and dosimetry are based on the fact that ionising radiation produces paramagnetic centres having long lifetimes in a number of materials. such centres’ concentration in a given sample is a measure of the total radiation dose to which a particular sample has been exposed. this effect can be used to determine the length of time of such exposure and hence, in many cases, a sample’s burial time (jonas 1997). esr has been used with tooth enamel for determining the age of palaeolithic archaeological sites, mainly because tooth enamel is ubiquitous in these contexts and because it allows dating beyond the 14c dating time range (40–45 ka max.). it is thus also very useful in important sites regarding human evolution studies, sites having palaeontological interest within a quaternary time frame and because many sites lack the volcanic materials needed for k – ar and 40ar/39ar dating (rink 1997). (1) quaternary dating by electron spin resonance (esr) applied to human tooth enamel 117 the aim is to determine the accumulated dose (da) to which a sample has been exposed whilst being buried; the exposure time is then determined by comparing d a to the average dose rate over such period: t= da (d) however, no significant difference in the accuracy of dose reconstruction has yet been found using different calibration methods (wieser et al., 2000). each sample’s radiation sensitivity is individually calibrated for dose assessment by the additive dose method; each sample is incrementally irradiated with at least 14 additional doses up to 100 gy or more (avoiding the sample becoming saturated). a linear regression analysis is applied to the esr measurements at each dose increment and the original measurement (laboratory-un-irradiated sample). the original d a in the sample is obtained from the negative intercept of the regression line with the dose axis; is estimated as being the total radiation provided by all radioactive elements present in a sample and soil and also cosmic radiation (ikeya 1993; jonas 1997; walker 2005; grun 1989). materials and methods the dentine was initially removed from teeth using a dental drill and water cooling. the enamel was then placed in a 30% naoh solution for one day to disinfect it and separate any remaining dentine (iaea report 2002); 1.4 mm mean enamel thickness was obtained for ag-i. the sample was air-dried at room temperature for three days; a dental drill was used to strip around 50 ± 5 μm from the inside and outside of the enamel surface to ensure that alpha radiation had no effect. the molar enamel powder was divided into nine aliquots. each aliquot was placed inside a plastic capsule for irradiation using 60c standard radiation equipment at room temperature and 8gy/min mean dose rate; the doses ranged from 0 to 740 gy. after irradiation, the esr signal was measured with a bruker esp300 x-band spectrometer. the g  and g  signals’ peak to peak height was used for calculating amplitude. the spectrometer parameters used were: 336 mt central field, 0.02 t scan range, 0.1 m t amplitude modulation, 100 khz modulation frequency, 0.328 s time constant and 20 mw microwave power. the aliquots were then irradiated with additional doses and the same process was repeated until getting 9 data points from esr signal amplitude as a function of γ-ray dose. soil samples were also collected from the aguazuque site and sent for u, th and k content analysis by gamma spectrometry. radiometric analysis is widely used for determining natural radioisotopes in geological samples by means of spectroscopic methods and, especially, for quantifying the uranium, thorium and potassium present in samples. these samples were analysed at ingeominas’ nuclear technology group’s environmental radiometrics’ laboratory which has a low-level gamma spectrometry system with 15 cm lead shielding and high-resolution gehp hyperpure germanium detector (canberra model gc2019), having 1.89% absolute detection efficiency for 661.6 kev. results and discussion figure 3 shows the esr spectra for enamel sample ag-i having an irradiation dose ranging from 54 to 324 gy. a linear model was used for fitting esr signal intensity to dose, as has been done by other authors (baffa et al., 2000; hefne et al., 2002; kinoshita et al., 2001a, 2008b; mascarenhas et al., 1982); very good agreement with linear fit was shown (figure 4). the archaeological d a obtained using this method was da = 2.10 ± 0.14 gy. rosy software (brennan et al., 1999) was used for calculating the enamel sample’s age, based on the given d a, cosmic radiation dose rate (determined at 251 ± 15 μgy/year), u, th and k content obtained from figure 3. esr spectra of tooth enamel ag-i in a range of 54 to 324 gy. (2) eduar carvajal, luis montes and ovidio a. almanza118 soil sample analysis and enamel (1.4 mm ag-i thickness and 5 mm dentine thickness) in sample. soil water content was 12.1%. the software’s default values were used for the remaining parameters. the cosmic radiation dose rate was obtained using the equation for luminescence and esr dating (prescott & hutton 1994). finding the dose rate to convert d a into an age is a crucial step after da has been found. rosy software allows age to be calculated from an enamel sample by taking da due to radioisotopes from adjacent layers into account. the software computes age based on three uranium absorption models (linear, early uptake and a combination of both). table 1 shows the sample concentration (ppm) 238u 232th 40k aguazuque 1.86 ± 0.19 3.72 ± 0.37 1.66 ± 0.09 alpha beta gamma cosmic gamma+cosmic dentine 0.00 0.00 enamel 0.00 0.00 cementum 0.00 0.00 sediment 0.00 46.55 347.31 uranium 0.00 34.54 188.50 thorium 0.00 12.01 158.81 potassium 0.00 0.00 total 0.00 46.55 347.31 251.00 598.31 (alpha+beta) dose rate (µgy/y) = 46.55 total dose rate (µgy/y) = 644.86 table 1. uranium, thorium and potassium concentration as determined by gamma spectrometry. table 2. dose rate obtained from radioisotope concentration, local cosmic radiation and total dose rate. alpha radiation from sediment was not used in the calculation, as about 50 μm was removed from the outer side of the tooth. figure 4. signal versus dose growth showing the linear fitting used. archaeological dose for ag-i was ad = 2.91 ± 0.14 gy; correlation coefficient was 0.9937. soil analysis results. the present work did not determine u, th and k concentration in enamel and dentine as their percentages have been negligible for calculating age in previous work, especially in dates for the holocene period (kinoshita et al., 2008; hefne et al., 2002). typical soil density, enamel and dentine values were used for calculating age (2.00, 3.00 and 2.82 g/cm3, respectively). the layer of enamel stripped off from each side by acid etching was about 50±5 mm; this prevented a possible contribution towards the dose having been made by alpha particles deposited in the sample. using these values and the radioisotope concentrations listed in table 1, rosy software was used for obtaining the alpha, beta and gamma radiation dose rate contribution for each radioisotope in sediment (the results are listed in table 2). these results were the same for any uranium absorption model. considering a 251 ± 15 μgy/year dose rate for cosmic radiation, a 3,256 ± 190y before present (bp) age was thus found for the ag-i sample. it should be noted that rosy software calculates these ages using three uranium accumulation models: linear, early and a combination of both (ikeya, 1993; grun, 1989; fattibene et al., 2010). the way that uranium accumulates in a tooth is established by these models; accumulation increases at the same rate as time in the linear model and accumulation occurs in a short period of time in the early model, relative to a tooth’s age and then remains constant after this short period of time. the tooth’s age was the same when calculated by these three methods; this result was consistent if taken into account that the accumulation of radioactive elements inside the tooth was negligible. conclusions age determined at many countries’ archaeological sites is the result of radiocarbon (14c) analysis of particular items found around the burials and not direct dating of individuals found there. the results regarding either physical or geological phenomena may not be chronologically equal, hence the importance of having techniques like esr which can directly quaternary dating by electron spin resonance (esr) applied to human tooth enamel 119 and absolutely determine the age of individuals of interest. the age of a human burial site found in colombia was determined in this work from a single tooth, a mean age of 3,256 ± 190y bp being found. this result agreed with stratigraphic analysis at sites providing soil ages ranging from 2,800 to 3,500y bp for 50 and 80 cm depths (correal, 1990). esr dating accuracy was obtained relatively straightforwardly, using just a few grams of tooth enamel, thereby highlighting some of the advantages and characteristics of the esr dating method. the results so obtained have contributed towards joining links in the long chain of colombian and american history (where there is still much to do, due to the rich pre-hispanic and colonial past). this paper thus represents a pioneering effort aimed at promoting the dating of our ancestors. acknowledgments we would like to thank the dib, universidad nacional de colombia, for providing financial support (project 8003348), m. peña from ingeominas for soil sediment analysis, the universidad nacional de colombia’s physical anthropology laboratory and its director, j.v. rodriguez, for providing facilities and allowing the use of his samples, as well as am groot (archaeologist) for her support during this research. references a.f. borgonovea, a. kinoshitaa,b, f. chena, p. nicoluccia, o. baffaa, (2007): energy dependence of different materials in esr dosimetry for clinical x-ray 10mv beam. radiation measurements, 42: 1227– 1232. anatoly molodkov; (2001): esr dating evidence for early man at a lower palaeolithic cave-site in the northern caucasus as derived from terrestrial mollusc shells; quaternary science reviews 20: 1051 -1055 ardila, g.i, (1984): chía, un sitio precerámico en la sabana de bogotá. fundación de investigaciones arqueológicas nacionales. banco de la república. bogotá d.c. baffa, o., brunetti, a., karmann, i., neto, c.m.d., (2000): esr dating of a toxodon tooth from a brazilian karstic cave. appl. rad. isot., 52: 1345-1349. brennan, b.j., rink, w.j., rule, e.m., schwarcz, h.p., prestwich, w.v., (1999): the rosy esr dating program, ancient. tl. 17: 45-53. correal, g., (1979): investigaciones arqueológicas en abrigos rocosos de nemocón y sueva. fundación de investigaciones arqueológicas nacionales. banco de la república. bogotá d.c. correal, g., (1990): aguazuque : evidencias de cazadores, recolectores y plantadores en la altiplanicie de la cordillera oriental. fundación de investigaciones arqueológicas nacionales, banco de la república. bogotá d.c. groot, a.m., (1992): checua. una secuencia cultural entre 8500 y 3000 años antes del presente. fundación de investigaciones arqueológicas nacionales. banco de la república. bogotá. grün, r., (2006): direct dating of human fossils. yearb. phys. anthropol. 49: 2-48. halliwell b. and gutteridge j. 2007. free radicals in biology and medicine. oxford university press, usa. hefne, j., yamani, a., al-dayel, o., ikeya, m., al-osaimi, s., (2002): esr dating of tooth from pre-islamic sitin saudi arabia. advances in esr applications. 18: 119-121. iaea report., (2002): use of electron paramagnetic resonance dosimetry with tooth enamel for retrospective dose assessment. report of a coordinated research project. iaea-tecdoc-1331. vienna. ikeya, m., (1993): new applications of electron spin resonance: dating, dosimetry and microscopy, world scientific publishing company, singapore. jonas, m., (1997): concepts and methods of esr dating. radiat. meas. 27: 943-973. kinoshita a., braga f.j.h.n., graeff c.f.o., baffa o., (2001): esr dosimetry of 89srand 153sm-in bone. appl. rad. isot. 54: 269-274. kinoshita, a., figueiredo, a.m.g., felice, g.d., lage, m.c.s.m., guidon, n. baffa, o., (2008): electron spin resonance dating of human teeth from toca da santa shelter of são raimundo nonato, piauí, brazil. nucl. instrum. meth b. 266: 635-639. mascarenhas s., baffa filho o., ikeya m., (1982): electron spin resonance dating of human bones from brazilian shell-mounds (sambaquis). am. j. phys. anthropol. 59, 413-417. prescott, j.r., hutton, j.t., (1994): cosmic ray contributions to doserates for luminescense and esr dating: large depths and long-term time variations. radiat. meas. 23, 497-500. punchard n. and kelly f. 1996. free radicals: a practical approach. oxford university press, usa. renfrew, c. y bahn, p. arqueología. teorías, métodos y práctica (1992): ediciones akal s.a. pp 107147, rink, w.j., (1997): electron spin resonance (esr) dating and esr applications in quaternary science and archaeometry. radiat. meas. 27: 975-1025. shigueo watanabe, walter elias feria ayta and henrique hamaguchi; (2003): some evidence of a date of first humans to arrive in brazil. journal of archaeological science 30: 351–354. ülkü ulusoy; (2004): esr studies of anatolian gypsum; spectrochimica acta part a 60: 1359–1365. walker m. (2005): quaternary dating methods. john wiley & sons ltd, the atrium, southern gate, chichester. chapters 1, 2, 3 and 4. wieser, a., mehta, k., amira, s., aragno, d., bercea, s., brik, a.,bugai, a., callens f., chumak v., ciesielski b., debuyst r., dubovsky s., duliu o.g., fattibene p., haskell e.h., hayes r.b., ignatiev e.a., ivannikov a., kirillov v., kleschenko e., nakamura n., nather m., nowak j., onori s., pass b., pivovarov s., romanyukha a., scherbina o., shames a.i., sholom, s., skvortsov v., stepanenko v., tikounov d.d., toyoda s. (2000): the second international intercomparison on epr tooth dosimetry. radiat. meas, 32, 549–557. keywords: hj-1a his; landsat-8 oli; fusion method; wetland classification. palabras clave: hj-1a his; landsat-8 oli; método de fusión; clasificación de humedales. how to cite item zhu, d., wan, l., & gao, w. (2019). fusion method evaluation and classification suitability study of wetland satellite imagery. earth sciences research journal, 23(4), 339346. doi: https://doi.org/10.15446/esrj. v23n4.84350 based on hj-1a hsi data and landsat-8 oli data, rs image fusion experiments were carried out using three fusion methods: principal component (pc) transform, gram schimdt (gs) transform and nearest neighbor diffusion (nnd) algorithm. four evaluation indexes, namely mean, standard deviation, information entropy and average gradient, were selected to evaluate the fusion results from the aspects of image brightness, clarity and information content. wetland vegetation was classified by spectral angle mapping (sam) to find a suitable fusion method for wetland vegetation information extraction. the results show that pc fusion image contains the largest amount of information, gs fusion image has certain advantages in brightness and clarity maintenance, and nnd fusion method can retain the spectral characteristics of the image to the maximum extent; among the three fusion methods, pc transform is the most suitable for wetland information extraction. it can retain more spectral information while improving spatial resolution, with classification accuracy of 89.24% and kappa coefficient of 0.86. abstract fusion methods evaluation and classification suitability study of wetland satellite imagery evaluación del método de fusión y estudio de idoneidad de clasificación de imágenes satelitales de humedales issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.84350 con base en los datos hj-1a hsi y landsat-8 oli, los experimentos de fusión de imágenes rs se llevaron a cabo utilizando tres métodos de fusión: transformación de componente principal (pc), transformación de gram schimdt (gs) y algoritmo de difusión vecina más cercana (nnd). se seleccionaron cuatro índices de evaluación, desviación estándar, entropía de información y gradiente promedio, para evaluar los resultados de fusión de los aspectos de brillo de imagen, claridad y contenido de información. la vegetación de humedales se clasificó por mapeo de ángulo espectral (sam) para encontrar un método de fusión adecuado para la extracción de información de vegetación de humedales. los resultados muestran que la imagen de fusión de pc contiene la mayor cantidad de información, la imagen de fusión gs tiene ciertas ventajas en el mantenimiento del brillo y la claridad, y el método de fusión nnd puede retener las características espectrales de la imagen al máximo. entre los tres métodos de fusión, la transformación de pc es la más adecuada para la extracción de información de humedales. puede retener más información espectral al tiempo que mejora la resolución espacial, con una precisión de clasificación del 89,24% y un coeficiente kappa de 0,86. resumen record manuscript received: 20/05/2019 accepted for publication: 27/08/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 339-346 danyao zhu1,2,3, luhe wan1,2*, wei gao1,2 1heilongjiang province key laboratory of geographical environment monitoring and spatial information service in cold regions, harbin normal university, harbin 150025, china 2college of geographical science, harbin normal university, harbin 150025, china 3department of geography, mudanjiang normal university, mudanjiang 157011, china * corresponding author: wanluhe@163.com g eo ph y si c s 340 danyao zhu, luhe wan introduction with the development of remote sensing technology, a large number of high-resolution satellites have appeared, such as irs-1c/1d, hj-1a/b, ikonos, quickbird, orbview, gf-1/2. these satellites with different band ranges, spatial resolution and spectral resolution have led to an increasing number of remote sensing data of different types in the same region (wang, li, & li, 2001) . how to integrate these remote sensing data and obtain the most applicable remote sensing image has become a difficulty in remote sensing technology. image fusion technology, by integrating the spatial and spectral advantages of remote sensing images, has continuously expanded its application field, and researchers have introduced the fusion image concept into wetland vegetation classification. wu, he and wang (2015) applied the fusion image of cbers-3 to the classification of wetlands by using four fusion methods, namely, gihs transform, wavelet transform, contourlet transform, nsct-gihs transform, and got better results. dong, lin, sun, qiu & zhang (2013) studied the applicability of wetland classification by using landsat-5 and spot-5 based on brovey transform, gram-schmidt transform and principal component (pc) transform. the results show that the fusion of 543 bands of landsat-5 and the third principal component of spot-5 are most suitable for the extraction of wetland information. wei, li, tan, & xun (2011) used the fusion image of ndvi (normalized difference vegetation index) extracted -36°chris high spectrum image and 0°chris image, adopted the method of spectral angle mapping (sam) to extract wetland vegetation type information, the experimental results show that the accuracy of vegetation classification in hyperspectral images is improved. zhu (2012) used the fusion image of hyperspectral image of hj-1a and ccd image, and extracted the typical vegetation distribution in zhalong region by using spectral angle mapping, and obtained higher classification results. the above mentioned image fusion methods are mainly focuses on the neighborhood processing of digital images. the focus of the research is how to improve the spatial resolution and spectral resolution of images through the improvement of algorithm. most scholars who do digital image fusion research tend to study in the background of computer science, but there are not many studies from the perspective of geography, especially the application of hyperspectral fusion image to wetland vegetation classification, so this paper conducts the research from this perspective. in this study, the wuyiling national nature reserve in xiaoxing’an mountains in northeast china was taken as the research area, and the hj-1a remote sensing image in july 2015 and landsat-8 remote sensing image in august 2015 were taken as the data sources. by analyzing the spatial resolution, spectral resolution, brightness, clarity, information content and the spectral features, the best fusion method applicable to the extraction of wetland information was found. the fuson image is classified by spectral angle mapping method, and the accuracy of the fusion method is verified to compare the classification accuracy of typical wetland vegetation. methods study site the images used in this study is located in wuyiling national nature reserve, geographic coordinates are 129°00′e-129°28′e,48°33′n-48°50′n, east of the xiaoxing’an mountains peak, northeast heilongjiang province. hyperspectral images can make up for the lack of distinguishing ground object information in wide-band remote sensing images (yan, 2008). however, due to the low spatial resolution of hj-1a data source (100 m), by fusion with the high-resolution landsat-8 oli full-color image (15 m), the spatial resolution of the data source was improved, so as to obtain more features of ground objects (table 1). the auxiliary data include the vector boundary map of the research area and the measured data of the field sampling points. table 1. data sources used in the study area time satellite designation sensor type 20150719 hj-1a hsi 20150815 landsat-8 oli data preprocessing hj-1a data. was preprocessed with envi5.3 software, which hj-1a /1btools tool is used to convert the data format, after calculating the signal to noise ratio of each band image, remove the first 20 bands with low signal to noise ratio. due to the existence of severe tilted stripes and bad lines in hsi data, the replace bad line function is used to repair bad lines and the global method is used to remove image stripes. radiation calibration of hsi secondary data has been carried out, and flaash atmospheric correction is still needed, and landsat8 oli images of the same period were taken as reference, select gcps: image to image tool was used for geometric precision correction, finally, the image is mosaic and cropped (fig. 2). fusion methods due to the large number of bands in hj-1a hyperspectral image, three fusion methods with unlimited band number are adopted in this paper for fusion operation. (1) principal component (pc) transform pc transform is a multidimensional orthogonal linear transformation based on statistical characteristics (ma, feng, he, & zhang, 2007). it concentrates the common information of multi-band images into the first principal component, while the unique spectral information that does not overlap with any other bands is put into other principal components, then, the panchromatic band data are stretched to have the same mean and variance with the first principal component, and which are used to replace the first principal component for principal component inverse transformation to restore to the original space and generate multi-band fusion images with high spatial resolution. (2) gram schimdt (gs) transform gs transformation is essentially a phase recovery algorithm (ma & wang, 2016). the main steps of the algorithm are as follows: (1) use the multispectral image with low spatial resolution to generate simulated panchromatic image with low spatial resolution; (2) the simulated image (as the first band) is superimposed on the multi-spectral image, and the reconstructed multi-spectral image is carried out gs positive transformation; (3) replace the first component of gs transform with high-resolution panchromatic image, then which was carried out gs inverse transformation on the multi-spectral image to get the final fusion image. figure 1. study site on xiaoxing’an mountains in heilongjiang province 341fusion method evaluation and classification suitability study of wetland satellite imagery g mn m i n j i ix y = = = +( )∑ ∑1 1 1 2 2 2   (3) where: m, n are the number of rows and columns of the image; ix , iy are the first-order difference of x direction and y direction, respectively; i, j are the line and column number of each pixel in the image of the same band. information entropy, an evaluation index of spatial detail information based on information quantity, reflects the amount of image informati on quantity. the larger the value is, the larger the information content of the fusion image be. the expression is: h l i ppi i= − =∑ 1 0 � (4) where: l represents the overall gray scale series in images; pi represents the probability of gray value of pixels in images. results and analysis the pca transform, gs transform and nnd algorithm results of hj-1a his image are shown in the figure (fig. 3.) visual interpretation it can be seen from the comparison between the fusion image and the original image that the fusion image maintains the spectral characteristics more greatly in terms of maintaining spectral information. pc fusion image and gs fusion image have moderate color contrast, and all kinds of ground objects in the fusion image show standard spectral detail features. visually speaking, there is no significant difference in spectral features between the two fusion images; the color change of nnd fusion image is serious, and spectral deformation of all kinds of ground objects has occurred, in which the construction land is green and the spectral deformation is serious. sphagnum and its surroundings appear purplish, which is quite different from the original image. it can be seen from the spatial information of the image that the fusion image is clearer and more distinguishable, with the same spatial resolution and clearer texture structure as the landsat-8 panchromatic image. in the nnd fusion image, the texture of construction land is clearer, and the edges of ground (3) nearest neighbor diffusion (nnd) algorithm nnd algorithm was proposed by sun, chen and david (2014). the main theory is that the new spectrum of high-resolution fusion images can be expressed as the weight combination of adjacent hyperpixel spectra of current pixels in low-resolution multispectral images. the fusion method based on nearest neighbor spreading can maintain accurate spectral information of multi-spectral images identified by spatial features, and is applicable to the application of accurate spectral information other than traditional visual inspection (ocalan, erdogan, tunalioglu, & durdag, 2016; gao, 2015). evaluation methods in this paper, four evaluation indexes including mean, standard deviation, information entropy and average gradient were selected to objectively evaluate the fusion image from the perspectives of image brightness, clarity and information content (fusun, saygin, mustafa, & filiz, 2017). mean, which is the gray mean value of pixels in remote sensing images, which can be expressed as: v mn m i n j f i j= = = ( )∑ ∑1 1 1 , (1) where: m, n are the width and height of the image respectively; f(i,j) is the dn value function of all pixels in the fusion image; i and j represent the row and column numbers of raster cells. standard deviation, which reflects the dispersion of dn values of all pixels and their dn means, can be used to evaluate the brightness of images. the higher the value, the more information the image contains. the specific formula is: std mn m i m j f i j v= = = ( ) −( )∑ ∑1 1 1 , (2) where: v is mean, m, n are the number of rows and columns of the image; f(i,j) is the dn value function of all pixels in the fusion image; i and j represent the row and column numbers of raster cells. average gradient, with the ability to sensitively reflect the image contrast to small details and texture transformation expression. the larger the average gradient value is, the more layers the image has and the clearer the image will be. the expression is: figure 2. preprocessed hj-1a hsi image (a. false color composite of band 111, 91, 71) and landsat-8 oli image (b. false color composite of band 5, 4, 3) 342 danyao zhu, luhe wan objects are prominent, which makes it easier to identify. however, serious fuzzy phenomena appear at the boundaries of sphagnum and marsh, the fused image is not clear enough. the color of construction land in pc fusion image and gs fusion image is relatively bright and dazzling, and the vegetation features in gs fusion image are obvious. quantitative evaluation according to the above quantitative evaluation indexes, with the support of envi/idl8.5, the eigenvalues of fusion images obtained by the three fusion algorithms used in this study are extracted, and the calculated results are shown in table 2. table 2. fusion image evaluation index calculation results mean value standard deviation average gradient information entropy original hyperspectral image 115.498 77.508 4.392 4.158 pc fusion image 111.143 92.902 12.358 4.652 gs fusion image 117.078 81.412 14.542 4.647 nnd fusion image 117.971 77.215 6.142 4.515 in terms of the brightness information of the image, the difference between the mean value of the fused image and the original image is relatively small. descending means of fusion image: nnd fusion image > gs fusion image > pc fusion image. the results show that the brightness of gs fusion image is the closest to the original image, indicating that gs fusion image has certain advantages in maintaining brightness information. from the aspect of image sharpness, average gradient can be used to evaluate the effect of improving sharpness in objective evaluation indexes. it can be seen from the table that the average gradient of gs fusion image is the largest, followed by pc fusion image, indicating that gs fusion image and pc fusion image have advantages in terms of sharpness. in terms of the information content of the image, the standard deviation and information entropy of the fused image can evaluate the improvement of the information content. the standard deviation and information entropy of pc fusion image are better than the other two fusion images. the information content of pc fusion image is the largest, followed by gs fusion image and nnd fusion image. spectral variation analysis in hyperspectral images, the radiation information of each pixel in each band can be extracted into a continuous spectral curve, reflect the spectral characteristics of ground objects, and realize the fine classification of ground objects, which is the most significant advantage of hyperspectral images (moon, choi, lee, & an, 2017; adam, mutanga, rugege, 2010). if the spectral curve of fusion image is distorted, the spectral information also changes. therefore, to evaluate which fusion method can retain the spectral information of the original image to the greatest extent, it is necessary to analyze the spectral curve variation degree of fusion images. from the fusion images and the original images, typical features of 9 ground objects samples, which concluded construction land, lakes, farmland, sphagnum, carex tato, salix brachypoda, white birch, pinus koraiensis and spruce, were selected to calculate the average spectral curve (fig. 4.). in the spectrum curve of construction land, the spectrum curve of pc, gs and nnd fusion image is similar. before 700 nm, the nnd fusion image is lower than the original image spectrum curve, and after 720 nm, the spectrum curve of the three kinds of fusion image is higher than the original image. in comparison, the spectral fidelity of nnd fusion image is better. in lake spectral curve, before 700nm, spectrum curves of three kinds of fusion image are almost the same, after 700 nm, spectrum curves of three kinds of fusion image are higher than the original image. it indicates that the spectral information of the water body of the three fusion images is seriously distorted, in which the change of the original spectral information by gs fusion is relatively small, while the change of the original spectral information by nnd fusion is relatively large; the sphagnum spectral curves of the three fusion images almost coincide with that of the original images, indicating that the fusion images have little influence on the sphagnum spectra; the spectral curve of the carex tato before and after fusion had the same trend before 700 nm, indicating that spectral fidelity was better. after 700 nm, the spectrum curve of nnd fusion image is lower than that of the original image, while the spectrum curve of pc fusion image and gs fusion image is higher than that of the original image, and keep the same trend; the spectral curves of farmland before and after fusion, after 700 nm, the spectral curves of three fusion images and the original image presented significant differences and serious spectral deformation. the reflectance of the spectral curves of the three fusion images was much higher than that of the original image and show the same trend; the spectral curves of white birch, pinus koraiensis and spruce before and after fusion, the spectral curves of the three fusion images and the original images almost overlapped before 750 nm, indicating that the fusion images retained good spectral characteristics. after 750 nm, the original spectral curve of white birch was higher than that of the three fusion images, while the original spectral curve of pinus koraiensis and spruce was lower than that of the three fusion images, indicating that there was a certain degree of spectral deformation in the fused images. classification results and accuracy evaluation in this study, the spectral angle mapper (sam) classification method with high accuracy was selected to extract the information of wetland types in wuyiling national nature reserve. sam is a hyperspectral classification algorithm designed based on the spectral shape characteristics of hyperspectral data. this algorithm has a simple design idea and does figure 3. results of different fusion methods (false color composite of band 111, 91, 71) 343fusion method evaluation and classification suitability study of wetland satellite imagery figure 4. comparison of spectral curves between fusion images and row image 344 danyao zhu, luhe wan not require data compression of hyperspectral remote sensing data. it is a relatively mature classification algorithm in hyperspectral image classification. sam classification processing is applied to fusion images of different methods (fig. 5). meanwhile, the accuracy of classification results is evaluated by using field sampling data in 2015 (tables 3-5). according to the classification results, sam was used to classify the fusion image. the accuracy of pc fusion image classification was higher and that of gs fusion image classification was lower. among them, the mapping accuracy of pinus koraiensis is 79.23% to 83.56%, and that of spruce is 94.08% to 99.15%. there are often misclassification between pinus koraiensis and spruce. in the commission errors, the values of spruce and white birch are relatively high. among the three fusion images, the commission errors of spruce and white birch in pc fusion images are relatively low. in the omission errors, the values of pinus koraiensis is larger, and the classification results of other ground objects are better because of the large difference in spatial details. conclusion and discussion in view of the low spatial resolution of hyperspectral remote sensing image of hj-1a satellite, landsat-8 panchromatic image was applied to carry out fusion experiments on three fusion methods of maintaining image’s band number, namely, pc transform, gs transform and nnd algorithm, and the fusion results was evaluated. the main conclusions are as follows: (1) compared with the original image, the fusion image is clearer and more distinguishable. the three fusion images maintain the spectral characteristics to a large extent, among which pc fusion image and gs fusion image have moderate color contrast, while nnd fusion image has serious color change. the texture of fusion image construction land is obvious, and the edges of construction land are prominent, while the edges of different vegetation types are seriously blurred. the high reflectivity of construction land in pc fusion image and gs fusion image makes it difficult to distinguish.gs image fusion significantly enhanced the vegetation characteristics. in conclusion, gs figure 5. spectral angle mapping classification results 345fusion method evaluation and classification suitability study of wetland satellite imagery fusion image and pc fusion image have advantages in sharpness. pc fusion image contains the largest amount of information, followed by gs fusion image. gs fusion image has a better performance in brightness information. (2) in the spectral variation analysis, except for the serious spectral information distortion of the lake, the spectral curves of other features are similar in the pc fusion image and the gs fusion image; the variation degree of spectrum curve in nnd fusion result is slightly less than that of pc and gs fusion results; in a variety of typical ground objects, the spectral information of lake and farmland is seriously distorted in the near infrared band, and the spectral information of salix brachypoda, spruce and sphagnum is best preserved. (3) the sam was applied to extract typical vegetation types from fusion images, and the accuracy of classification results was evaluated. the optimal overall classification accuracy reached 89.24%, and the kappa coefficient was 0.86. the classification accuracy of pc fusion image is the highest, and contains the largest amount of information; nnd fusion image have certain advantages in spectral fidelity and higher classification accuracy. the classification accuracy of gs fusion image is relatively low. table 3. pc fusion image classification accuracy confusion matrix commission errors (%) omission errors (%) mapping accuracy (%) user accuracy (%) overall accuracy (%) kappa coefficient farmland 5.29 3.11 96.89 94.71 89.24 0.86 sphagnum 2.81 2.56 97.44 97.19 carex tato 0.52 1.21 98.79 99.48 salix brachypoda 1.38 3.43 96.57 98.62 white birch 10.54 0.54 99.46 89.46 pinus koraiensis 7.15 16.44 83.56 92.85 spruce 41.92 0.85 99.15 58.08 table 4. gs fusion image classification accuracy confusion matrix commission errors (%) omission errors (%) mapping accuracy (%) user accuracy (%) overall accuracy (%) kappa coefficient farmland 7.17 3.22 96.78 92.83 87.27 0.83 sphagnum 8.54 3.85 96.15 91.46 carex tato 1.91 2.25 97.75 98.09 salix brachypoda 1.89 6.23 93.77 98.11 white birch 17.87 1.09 98.91 82.13 pinus koraiensis 5.78 20.44 79.56 94.22 spruce 43.38 5.92 94.08 56.62 table 5. nnd fusion image classification accuracy confusion matrix commission errors (%) omission errors (%) mapping accuracy (%) user accuracy (%) overall accuracy (%) kappa coefficient farmland 4.70 2.07 97.93 95.30 87.93 0.84 sphagnum 2.56 2.56 97.44 97.44 carex tato 2.87 0.52 99.48 97.13 salix brachypoda 1.22 4.70 95.30 98.78 white birch 16.86 0.54 99.46 83.14 pinus koraiensis 6.17 20.77 79.23 93.83 spruce 46.14 0.85 99.15 53.86 acknowledgements add a third grant fund: this study was supported by the national natural science foundation of china (41671100), a grant from the education department of heilongjiang province (1352msyzd003), and a grant from mudanjiang normal university (gg2017005). references adam, e., mutanga, o., rugege, d. (2010). multispectral and hyperspectral remote sensing for identification and mapping of wetland vegetation: a review. wetlands ecology and management, 18, 281-296. dong, q. l., lin, h., sun, h., qiu, l., & zhang, y. (2013). application of multisource remote sensing data fusion method in wetland classification. journal of central south university of forestry and technology, 33(1), 52-57. 346 danyao zhu, luhe wan fusun, b. s., saygin, a., mustafa, t. e., & filiz, s. (2017). evaluation of image fusion methods using palsar, radarsat-1 and spot images for land use/ land cover classification. journal of the indian society of remote sensing, 45(4), 591-601. gao, l. (2015). landsat8 oli remote sensing image fusion based on the nearest neighbor diffusion method. proceedings of the 2015 annual conference of jiangsu surveying and mapping geographic information society. ma, x. x. & wang, j. l. (2016). the basic research of phase retrieval algorithm. optik-international journal for light & electron optics, 127(4), 1561-1566. ma, y. p., feng, z. k., he, y. j., & zhang, z. h. (2007). research on image fusion of quickbird based on erdas imagine software. journal of beijing forestry university, (s2), 181-184. moon, w., choi, j., lee, s., & an, k. (2017). in situ nutrient-spiking bioassays for determining phosphorus and light limitation in a wetland ecosystem. journal of environmental biology, 38(4), 631-639. ocalan, t., erdogan, b., tunalioglu, n., & durdag, u. m. (2016). accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment. earth sciences research journal, 20(4), d1-d9. sun, w. h., chen, b., & david, w. (2014). nearest-neighbor diffusion-based pansharpening algorithm for spectral images. optical engineering, 53(1), 13-17. wang, z. j., li, d. r., & li, q. q. (2001). application of multiple wavelet theory in spot and tm image fusion. journal of wuhan university (information science edition), 1, 24-28. wei, w., li, z. y., tan, b. x., & xun, h. s. (2011). research on remote sensing classification method of longbaotan wetland based on multi-angle hyperspectral chris image. forest science research, 24(2), 159-164. wu, r. j., he, x. f., & wang, j. (2015). evaluation of fusion quality of multispectral and panchromatic images of wetland on resource 3 satellit. high-tech communication, 25(2), 157-162. yan, z. (2008). research on multi-source remote sensing image fusion technology. wuhan university of technology. zhu, z. x. (2012). zhalong wetland vegetation classification based on hyperspectral remote sensing image. harbin normal university. keywords: the yellow river estuary; sediment transport; numerical simulation; summary and prospect. palabras clave: estuario del río amarillo; transporte de sedimentos; simulación numérica; resumen y perspectiva. how to cite item jian, j., & chen, h. (2019). advance of research on the numerical simulation of sediment transport in the yellow river estuary. earth sciences research journal, 23(4), 379-383 doi: https:// doi.org/10.15446/esrj.v23n4.84100 the numerical simulation of sediment transport in the yellow river estuary is an important method to quantitative analyze the problems by water and sediment movement, including sediment deposition and river bed evolution, beach deposition and development, the formation and development of the delta, and so on. making an intensive study of the problems is useful to study the principles of sediment movement and the evolution law of the estuary bed, and it is also important significance to scientific demonstrate of the estuary sediment deposition, sea water intrusion prevention treatment scheme and solve the problems of engineering. based on the mathematical modeling of sediment transport and our research work, this paper analyzes the research statement, the unsolved issues and the developed trend of this kind problem, to provide the theoretical reference on more accurate numerical simulation of sediment transport in the yellow river estuary and the scientific reference on the demonstration and decision of estuarine regulation scheme. abstract advance of research on the numerical simulation of sediment transport in the yellow river estuary avance de la investigación sobre la simulación numérica del transporte de sedimentos en el estuario del río amarillo issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.84100 la simulación numérica del transporte de sedimentos en el estuario del río amarillo es un método importante para analizar cuantitativamente los problemas por el movimiento del agua y los sedimentos, incluida la deposición de sedimentos y la evolución del lecho del río, la deposición y el desarrollo de las playas, la formación y el desarrollo del delta, y demás. hacer un estudio intensivo de los problemas es útil para estudiar los principios del movimiento de sedimentos y la ley de evolución del lecho del estuario, y también es importante demostrar científicamente la deposición de sedimentos del estuario, el esquema de tratamiento de prevención de intrusiones de agua de mar y resolver los problemas de ingeniería. basado en el modelo matemático del transporte de sedimentos y en nuestro trabajo de investigación, este estudio analiza la declaración de la investigación, los problemas no resueltos y la tendencia desarrollada de este tipo de problemas, para proporcionar la referencia teórica sobre una simulación numérica más precisa del transporte de sedimentos en el estuario del río amarillo y la referencia científica sobre la demostración y decisión del esquema de regulación estuarina. resumen record manuscript received: 27/04/2019 accepted for publication: 25/09/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 379-383 jinfeng jian*, huanzhen chen 1school of mathematics and statistics, shandong normal university, jinan 250014, china * corresponding author: jjf_sdnu@163.com h y d r o lo g y 380 jinfeng jian, huanzhen chen introduction the mathematical model of sediment transport of estuary is a branch of river simulation. it is an important method to quantitatively predict the problems caused by water and sediment movement, including river bed evolution, formation and development of deltas, expansion of siltation areas, transport and deposition of silt and sediment, and so on. the yellow river estuary is a strong accumulative estuary with frequent fluctuation of weak tides and much sediment. its evolution is influenced not only by the little water and much sediment of upper reaches, but also by the main factors such as tidals and ocean dynamic characteristics in coastal areas, including tides, currents and waves. at the same time, the influence of interaction between cohesive sediment and turbulence, the flocculation of cohesive sediment and the ocean chemistry should be considered. the problem is complex, the simulation is difficult and the cost is high. with the rapid development of modern computing technology, especially parallel computing technology, compared with traditional simulation methods such as measured model and physical similari ty model, the numerical simulation of sediment transport by using mathematical model has the advantages of low investment, short cycle, repeatability, and easy improvement, which has become an important method of estuary simulation. the thorough study of this kind of problem can help people to more clearly grasp the mechanism and form of the sediment transport in the estuary. it will also be helpful for the scientific demonstration and engineering solution of the estuary control schemes such as governing estuarine siltation, slowing down estuarine extension velocity, preventing seawater intrusion, which has a positive and important significance for the effective implementation of the national strategies of “development plan of high efficiency ecological economic zone in yellow river delta” and “development plan of shandong peninsula blue economic zone”. there are many mathematical models describing the sediment transport problem of the yellow river. different hydrodynamic conditions, different demands and different boundary conditions can lead to subtle differences in the models. the one-dimensional model can simulate the siltation and deformation of riverbed sediment in long-term and non-uniform sand, and it is often used in the study of river channel simulation in the long reach. the two-dimensional model can describe the horizontal or vertical distribution of velocity field and sediment field in water area and it is often used in the study of riverbed deformation in short-term river sections such as estuaries and ports. however, the sediment content in the practical problem is generally non-uniformly distributed along the water depth, and the sediment transport presents a typical three-dimensional property. at this time, the three-dimensional mathematical model can simulate the sediment transport and sedimentation law more accurately and completely. however, the development of three-dimensional sediment mathematical model is slow, the main reasons are: (1) the study of the three-dimensional sediment movement law is still not perfect, the sediment transport rate, resistance and other sediment parameters still adopt onedimensional sediment transport calculation formula. (2) the existing numerical calculation format has a complex structure, large calculation amount, and it is not easy to be visualized and softwareized. (3) the boundary conditions are not easy to be obtained. because of the complexity of riverbed terrain and the complex and variable dynamic conditions of rivers and oceans, the real initial boundary conditions are not easy to be obtained. in recent years, hydraulic engineers and computational mathematicians at home and abroad have proposed one-dimensional and two-dimensional mathematical models of sediment transport in estuaries under certain special simplifying assumptions, and realized the numerical simulation of finite difference method, finite element method and finite volume method, which have been applied to solve large-scale engineering sediment problems. however, in the general sense, the mathematical model and numerical simulation of sediment transport in the yellow river estuary, especially the three-dimensional problem, are still rare, and the systematic and rigorous theory of numerical analysis has not been reported. on the basis of referring to relevant literature and combining with our research work, this paper attempts to make a brief analysis and comment on the current situation of numerical simulation of sediment transport problem in estuaries, and put forward its own views and opinions on the future development, with a view to benefiting. two-dimensional mathematical model of sediment transport in the water era of the yellow river estuary, the sediment movement is mainly suspended load, and the planar flow velocity is significantly higher than the vertical flow velocity. considering the influence of turbulence, tidal current, wave and the assumption of hydrostatic pressure distribution, by performing reynolds average, wave period average and water depth average on the threedimensional sediment transport model, a planar two-dimensional sediment transport model composed of hydrodynamic model (water flow continuous equation, water flow movement equation) and sediment transport model (sediment continuous equation, riverbed deformation equation) can be obtained (yang, 1993; dou, dong, dou & li, 1995; cao & wang, 1993). water flow continuous equation van rijin, tan (1) in the equation, h represents the water depth, (u, v) represents the component of the vertical average velocity along the direction of x and y, and it is understood as different composite velocity fields in different water areas (for example, estuary water area is understood as a composite field of wave, runoff and tidal current velocity ). water flow movement equation ∂ ∂ + ∂ ∂ + ∂ ∂ + ∂η ∂ + τ ρ − τ ρ − + ρ ∂ ∂ + ∂ hu t huu x hvu y gh x fv s x s bx m wx m m xx ( ) ( ) 1 xxy by m wx m y hu hu t huu x hvv y gh y ∂ ε ∂ ∂ + ∂ ∂ + ∂ ∂ + ∂η ∂ + τ ρ − τ ρ       = ( ) ( ) v −− + ρ ∂ ∂ + ∂ ∂ ε fu s x s y hv m yx xy 1       = v (2) in this equation,  represents the water level,  = h + zb, zbrepresents the riverbed elevation, m represents the fluid density, hx, by represents the component of the shear stress on the bottom of the riverbed in the direction of x and y, wx, wy represents the component of the surface wind shear stress in the direction of x and y, f = 2 sin  represents the coriolis force coefficient reflecting the effect of the earth’s rotation,  represents the angular velocity of the earth’s rotation,  represents the local latitude, sxx, sxy, syx, syy represents the wave radiation stress term, i.e. the residual momentum flux caused by fluctuations, and  is the turbulent viscous coefficient. sediment continuous equation ∂ ∂ +υ ∂ ∂ + ∂ ∂ − s t s x v s y d s f h sv = (3) in this equation, s represents the sediment concentration, fs represents the function of sediment scouring and silting on the riverbed, generally taken as fs = (s* − s ),  represents the sediment settling velocity,  represents the recovery saturation coefficient, s* represents the sediment transport capacity, and d represents the sediment diffusion coefficient. this is a uniform sediment model. if the non-uniform sediment is taken into account, the sedi ment concentration of each group of sediment gradation satisfies this equation. suspended load riverbed deformation equation ′ =γ ∂ ∂ − z t fb s (4) in this equation, ´is the dry bulk density of sediment. 381advance of research on the numerical simulation of sediment transport in the yellow river estuary the windward finite element, mass concentrated windward finite element and other methods are produced. these methods are effective, but due to the introduction of excessive artificial viscosity, the discrete accuracy is reduced and the discontinuous screeding phenomenon is caused. petrov-galerkin and streamline diffusion methods can better avoid numerical dispersion and non-physical numerical oscillations, but they still have problems when dealing with interrupted flow. high-resolution format finite element method can better avoid the high frequency oscillation of the interrupted flow and the false oscillation caused by the convection dominant by introducing minimum and sufficient consumption in the stiffness matrix, but no strict mathematical proof has been given. at present, the models used in the engineering are mainly tabs-2 developed by the us army corps of engineers waterway experimental station, which can be used for the simulation of rivers, reservoirs and estuaries, and the cche2d established by jia,wang ,which can simulate the secondary flow effect formed in the curve. finite volume method (guo, han & he, 1996; pan, lu & yu, 2009; hu & tan, 1995; wu & wang, 2012; benkhaldoun, elmahi & seaid, 2010) the finite volume method is more suitable for complex regions than the finite difference method, and its calculation amount is lower than that of the finite element, so it has been widely used. there are two key problems to be solved in the finite volume method, one is the arrangement of variable nodes in the integral unit, the other is the calculation method of boundary flux of the integral unit. the traditional finite volume method can not deal with shock wave problem, but it has the ability to deal with discontinuous problem when using riemann solution operator, which can accurately capture the shock wave. the finite volume method can use unstructured grid at the same time. in the volume method of second-order and higher accuracy, the limiter function should be introduced to eliminate numerical oscillation in the process of flux reconstruction. at present, the finite volume method is the mainstream method to solve the problem of sediment transport. it mainly uses the body-fitted grid to establish the fast2d model in the curvilinear coordinate system to simulate the sediment transport and topographic dynamics the alluvial channels; the sutrench-2d model established by vanrijin,tan is used to simulate the sediment transport at the bottom of the sedimentary river bed and the change of the riverbed elevation under the effect of quasi-steady-state water flow and wind-induced waves, and there is also no strict mathematical proof. characteristic finite element method (li & chen, 2011; wang & liu, 2003; liu, li & chen, 2013; zhou & chen, 2007; luo, zhu, zeng & xie, 2007; dawson & martinez, 2000) since the convection terms of the first three equations of the sediment transport model can lead to numerical dispersion and non-physical oscillation, and have hyperbolic characteristics, numerical dispersion along the characteristic line can improve the stability of the format. at the same time, the truncation error about time is smaller, so the larger time step can be adopted. the characteristic line method is suitable for the strong estuary areas where the hydrodynamic changes are severe. firstly, the characteristic finite element scheme of shallow water equation is proposed by, dawson,monica, and the suboptimal error estimation is obtained. wang jiwen and liu ruxun get the cn-galerkin method based on the characteristic direction of the shallow water wave equation, and obtain the suboptimal error estimation; zhou zhaojie and chen huanzhen apply it to the planar two-dimensional water and sediment model, and a suboptimal error estimation is obtained. mixed finite element method (luo, zhu, zeng & xie, 2004; lu, 2007; xia, 2013) the mixed finite element can approximate the function and its gradient with high accuracy at the same time, which overcomes the shortcomings of the standard finite element method in calculating the gradient by solving itself, and the requirement for the solution of space smoothness becomes lower. luo zhendong and others give a mixed finite element method for two-dimensional sediment transport model, give the existence and uniqueness of generalized solution and the mixed finite element solution, and obtain the optimal order error estimation by increasing the number of interpolations in the test function space of flow velocity field and sediment field. lu xiuying and xia kaifeng the conditions of initial boundary values. initial water level (water depth) usually starts from downstream water level, and the initial value of the water level at each point is given by linear interpolation according to the average river bed gradient. the initial flow velocity and sediment concentration are given if there are measured values. otherwise, they can be taken as 0. with the process of numerical calculation, the influence of initial values will gradually disappear. the initial siltation elevation is usually taken as zero. on this basis, the calculated positive value is the siltation thickness, while the negative value is the scouring thickness. the inflow boundary generally gives the flow or water level, and for the suspended load, the sediment concentration should also be given. at the outflow boundary, the sediment flux is generally considered to be zero. for the boundary conditions of the fixed shore, the condition of no slip of flow is generally adopted, and the velocity field is taken as 0. when studying the sea area, due to the periodicity of the tide level, the edge beach changes with the water level, which leads to the change of boundary. at this time, the boundary “freezing” technology can be used to process the complex moving boundary into the fixed boundary. the determination of parameters. in the two-dimensional sediment transport model, the determination of the parameters describing sediment characteristics and hydrodynamic forces, such as shear stress at the bottom of the riverbed, shear stress of surface wind, wave radiation stress, sediment transport capacity of current, recovery saturation coefficient, sedimentation velocity and dry bulk density of sediment, have not been effectively expressed in the twodimensional motion theory. most of them directly use the corresponding data or regular formulas of one-dimensional sediment transport. numerical simulation method the numerical simulation mainly includes the discrete format construction, the mathematical demonstration of the format (the uniqueness, stability, and convergence of the existence of the solution), the program implementation of the format, and the practical application of the format. in terms of format construction methods, there are mainly finite difference, finite volume element, finite element, mixed finite element, characteristic finite element, discontinuous finite element and other methods. at the same time, in order to better deal with the problem of discontinuous (shock) flow, sometimes the technique of windward and adding artificial dissipative items is combined. finite difference method (lu & peng, 1985; zheng, 2003; spasojevic & holly, 1990; chang, 1998) the finite difference method which is widely used in the planar twodimensional sediment transport model is the alternating direction implicit format method (adi ). the traditional difference method has the advantages of simple structure and easy implementation. however, it is inflexible and the accuracy is difficult to improve when dealing with irregular boundary and complex boundary conditions. although it can basically adapt to irregular boundary after using the curve coordinate technology, there are still some shortcomings. in order to solve the numerical oscillation that occurs when the shock current is generated, the windward adi method is formed in combination with the windward. at present, the main model used in engineering is mobed2 model established by spasojevic and holly in curvilinear coordinate system, which can simulate sediment deposition in waterways, estuaries and coastal areas; the fluvial12 model established by chang in rectangular coordinate system, which can simulate the changes of riverbed section and sediment gradation caused by river bending; the delft-2d model used to simulate wave and current movement can describe the influence of wave motion, but these models do not give strict mathematical proof. finite difference method (li & zhang, 1999; zhang & yin, 2002; chippada, dawson, martinez & wheeler, 1998; chippada, dawson, martinez & wheeler, 1998; dawson & martinez, 2000) the finite element method can deal with irregular boundaries and complex boundary conditions better, and it can improve the calculation accuracy more conveniently, but the storage amount and calculation amount are larger than the finite difference method. both the finite difference method and the finite element method can generate numerical dispersion due to the existence of convection terms. combined with the windward technology, 382 jinfeng jian, huanzhen chen propose a characteristic mixed finite element method for shallow water equation and sediment transport equation, and obtain suboptimal error estimation. discontinuous finite element method (shi, 2013; dawson & proft, 2004) discontinuous finite element not only retains the advantages of the finite element which can be applied to irregular boundary and easy to improve accuracy, but also has the advantages of easy processing discontinuous problems, easy carrying out parallel computation, good stability and entropy compatibility. dawson,proft use the methods of discontinuous finite element and finite element coupling to study the shallow water equations, gives the calculation format, and obtains the suboptimal error estimation of the l2 model. shi baohai proposes a semi-discrete format for the two-dimensional sediment transport model by using discontinuous finite element method, estimates the errors, and obtains suboptimal error estimation of the l2 model. further research and outlook at present, although the research on the establishment of mathematical models and the technology of numerical simulation for sediment transport problems have made great progress, some major problems such as the construction and simulation of mathematical models in three dimensions need to be further studied and explored. (1) research on the basic laws of sediment movement. the factors that strongly influence the law of sediment diffusion and transport include riverbed resistance, sedimentation, sedimentation velocity and recovery saturation coefficient in the sediment transport model, and flocculation caused by seawater temperature and salinity changes. these laws do not necessarily conform to the classical gauss distribution, and may conform to some algebraic distribution. this needs to be further explored in mathematical theory and physical experiments to form a constitutive equation that reflects the mathematical physical nature of practical problems, which can more accurately describe the process of sediment transport. (2) research on the basic laws of hydrodynamics. sediment deposition or seawater erosion in the waters of the estuary coast is affected by a variety of hydrodynamic factors. therefore, we should consider the wavecurrent coupling model formed by runoff, tidal current and wind, and pay attention to the shape, function and development law of shear peak and pinnacle peak, which are caused by runoff and ocean dynamics, and the mechanism of sediment accumulation and capture, so as to form an accurate hydrodynamic system. (3) research on numerical simulation technology. based on the hydrodynamics and laws of sediment movement in the estuary, we should consider factors such as strong convection, nonlinearity, strong coupling, and data roughness in practical problems to establish computable models and mathematical theories, which provide a solid theoretical basis for a more accurate simulation of sediment transport in the yellow river estuary, and provide a reliable scientific basis for the demonstration and decisionmaking of the control scheme of the yellow river estuary. acknowledgement the paper is supported by the national natural science foundation of china (no. 11471196, 11301311, 10971254) and the natural science foundation of shandong province (no. 2014zrb01849). references benkhaldoun, f., elmahi, i., & seaid, m. (2010). a new finite volume method for flux-gradient and source-term balancing in shallow water equations. computer methods in applied mechanics and engineering, 199, 49-52. cao, z. & wang, g. (1993). numerical simulation of wave-induced sediment lifting and tidal sediment transport. acta oceanologica sinica, 15(1), 107-118. chang, h. h. (1998). generalized computer program: users’ manual for fluvial-12: mathematical model for erodible channels. san diego. chippada, s., dawson, c. n., martinez, m. l., & wheeler, m. f. (1998). finite element approximations to the system of shallow water equations i: continuous time a priori error sstimates. siam journal on numerical analysis, 35(2), 692-711. chippada, s., dawson, c. n., martinez, m. l., & wheeler, m. f. (1998). finite element approximations to the system of shallow water equations ii: discrete-time a priori error estimates. siam journal on numerical analysis, 36(1), 226-250. dawson, c. n., & proft, j. (2004). coupled discontinuous and continuous galerkin finite element methods for the depth-integrated shallow water equations. computer methods in applied mechanics and engineering, 193(1), 289-318. dawson, c. n. & martinez, m. l. (2000). a characteristic-galerkin approximation to a system of shallow water equations. numerische mathematik, 86, 239-256. dawson, c. n. & martinez, m. l. (2000). finite element approximations to the system of shallow water equations, part iii: on the treatment of boundary conditions. siam journal on numerical analysis, 38(1), 149-159. dou, g., dong, f., dou, x., & li, t. (1995). study on mathematical model of coastal sediment in estuary. science in china (series a), 25(9), 995-1001. guo, q., han, q., & he, m. (1996). mathematical model of two-dimensional tidal current and sediment. journal of sediment research, 1, 48-55. hu, s. & tan, w. (1995). numerical modeling of two-dimensional shallow water flow on unstructured grids. advances in water science, 6(1), 1-9. li, d. & zhang, h. (1999). finite element method for simulation of twodimensional flow and sediment movement in the lower yellow river. journal of sediment research, 4, 59-63. li, w. & chen, h. (2011). a finite element numerical simulation for 2-d non-homogeneous current and silt sedimentation model. applied mathematics a journal of chinese universities, 26(2), 169-178. liu, m., li, w., & chen, h. (2013). multistep characteristic finite element method for flat non-homogeneous current and silt sedimentation model. acta mathematicae applicatae sinica, 36(5), 870-880. lu, x. (2007). numerical simulation of two-dimensional shallow water equation. jinan: school of mathematical science, shandong normal university. lu, x. & peng, r. (1985). calculation of suspended sediment scouring and silting -solution of implicit finite difference chasing method. journal of sediment research, 1, 32-43. luo, z., zhu, j., zeng, q., & xie, z. (2004). mixed finite element methods for the shallow water equations including current and silt sedimentation the continuous-time case. applied mathematics and mechanics, 25(1), 74-84. luo, z., zhu, j., zeng, q., & xie, z. (2004). mixed finite element methods for the shallow water equations including current and silt sedimentation the discrete-time case along characteristics. applied mathematics and mechanics, 25(2), 166-180. pan, c., lu, h., & yu, p. (2009). numerical simulation of sediment transport in discontinuous shallow water flows with triangle grids. journal of hydrodynamics, 24(6), 778-785. shi, b. (2013). an up-wind discontinuous galerkin method for current and silt sedimentation in two space dimensions. jinan: school of mathematical sciences, shandong normal university. spasojevic, m. & holly, f. m. (1990). mobed2: numerical simulation of twodimensional mobile-bed processes. technical report no.344, iowa institute of hydraulic research, university of iowa. wang, j. & liu, r. (2003). a characteristics based on galerkin method for the system of shallow water equations. acta mathematicae applicatae sinica, 26(3), 458-466. wu, h. & wang, j. (2012). a new high resolution finite volume method for solving two-dimensional shallow water equation. computer technology and development, 22(10), 55-58. 383advance of research on the numerical simulation of sediment transport in the yellow river estuary xia, k. (2013). a characteristic-mixed finite element method for the simulation of two-dimensional water and sediment transport model. jinan: school of mathematical science, shandong normal university. yang, g. (1993). river mathematical model. beijing: ocean press. zhang, x. b. & yin, r. l. (2002). planar 2-d flow and sediment mathematical modeling. advances in water science, 12 (6): 665-669. zheng, j. (2003). development and application of numerical models for flow motion and sediment transport in an orthogonal body-fitted coordinate system. marine science bulletin, 22(1), 1-8. zhou, z. & chen, h. (2007). a characteristic finite element simulation for current and silt sedimentation model in two space dimensions. numerical mathematics a journal of chinese universities, 29(3), 245-256. issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n1.81087 earth sciences research journal earth sci. res. j. vol. 26, no. 1 (march, 2022): 55 68 m in e r a l o g y mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone (nw iberia) sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero universidad de león, españa. salca@unileon.es how to cite item: alcalde-aparicio, s., vidal-bardan, m., & alonsoherrero, e. (2022) mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone (nw iberia). earth sciences research journal, 26(1), 55-68 . https://doi.org/10.15446/ esrj.v26n1.81087 record manuscript received: 19/07/2019 accepted for publication: 07/01/2022 abstract this paper presents detailed mineralogical results together with a geochemical characterization for a sequence of six natural soil profiles. bedrock samples (r series) and overlying soil samples (s series) were characterized. the soil profiles are distributed in a series of paleozoic lithological units from lower ordovician to upper carboniferous in age (iberian massif, nw iberia). the lithological influence on mineral properties and geochemical composition and, how different weathering may be occurring under very similar temperate and acidic conditions, have been studied. field observations together with laboratory analyses were indicative of differential weathering. so, a series of selected chemical indices and relations were applied to clarify this assumption. the mineralogy was analysed by scanning electron microscopy (semeds), x-ray diffraction (xrd) of rock powder and soil oriented aggregates. x-ray fluorescence spectrometry (xrf) and inductively coupled plasma mass spectroscopy (icp-ms) were applied to analyse chemical composition. the first results showed how major elements, sio2, al2o3 and fe2o3, slightly enriched in the soil profiles, are consistent with the dominant mineralogy: quartz, chlorite, muscovite and/or illite, together with kaolinite and albite. the bases k2o, na2o, cao and mgo are also coherent with mineral composition and experience little variation, but are gradually removed in the profiles. the mobility of major elements leads to a general loss of bases and, in general, a slight enrichment in silica and sesquioxides. sio2 is enriched, firstly accumulated in soils and partially depleted by dissolution as colloidal form. al2o3 in some soils is slightly less than in former rocks, so other physical processes are expected to take place, involving clay removal with consequent aluminium depletion too. a special emphasis has been given to albite coexisting with kaolinite, firstly supposed to be directly inherited from parent substrates when present, but finally the chemical index pia shows it was mainly due to mineral alteration of plagioclases. the best correlations to explain the albite alteration and kaolinitization progress were obtained with chemical indices pia, ciw, cia and al2o3/na2o ratio. this together with mineralogical signatures, suggest that kaolinite is the result of gradual dissolution due to the acid hydrolysis of albite in such acidic environments, which may also be attributed to the organic matter influence. keywords: geochemical relations; soil profiles; differential weathering; plagioclase index of alteration (pia); albite acid hydrolysis; mineralogía de arcillas y señales geoquímicas como indicadores rápidos de procesos de meteorización química diferencial en una litosecuencia silícea de perfiles de suelos este trabajo presenta resultados mineralógicos detallados junto con una caracterización geoquímica para una secuencia de seis perfiles de suelos naturales. se caracterizaron muestras de rocas (serie r) y muestras de suelos suprayacentes (serie s). los perfiles de suelo se distribuyen en una serie de unidades litológicas paleozoicas de edad desde el ordovícico inferior hasta el carbonífero superior (macizo ibérico, nw de iberia). se ha estudiado la influencia litológica en las propiedades mineralógicas y en la composición geoquímica y cómo puede ocurrir una alteración diferente en los perfiles bajo condiciones templadas y ácidas muy similares. las observaciones de campo junto con los análisis de laboratorio fueron indicativos de alteración diferencial. por lo tanto, se aplicaron una serie de índices químicos de meteorización y algunas relaciones seleccionadas para aclarar esta suposición. la mineralogía se analizó mediante microscopía electrónica de barrido (sem-eds), difracción de rayos x (xrd) de muestras de polvo de roca y agregados orientados del suelo. se aplicaron espectrometría de fluorescencia de rayos x (xrf) y espectroscopia de masas de plasma acoplado inductivamente (icpms) para analizar la composición química. los primeros resultados mostraron cómo los elementos mayoritarios, sio2, al2o3 y fe2o3, ligeramente enriquecidos en los perfiles de suelo, son consistentes con la mineralogía dominante: cuarzo, clorita, moscovita y/o ilita, junto con caolinita y albita. las bases k2o, na2o, cao y mgo también son coherentes con la composición mineral y experimentan poca variación, pero se eliminan gradualmente en los perfiles. la movilidad de los elementos mayoritarios lleva a una pérdida general de bases y, en general, un ligero enriquecimiento en sílice y sesquióxidos. el sio2, primero se acumula en los suelos y se pierde parcialmente por disolución en forma coloidal. el al2o3, en algunos suelos es ligeramente menor que en las rocas, por lo que se espera que parte se pierda también en forma coloidal o tengan lugar otros procesos físicos que involucran la remoción de la arcilla, con la consiguiente pérdida del aluminio también. se ha dado un énfasis especial a la presencia de albita que coexiste con la caolinita; en primer lugar, se supone que se heredó directamente de los materiales originales en los que está presente, pero finalmente el índice pia muestra que se debe principalmente a la alteración de las plagioclasas. las mejores correlaciones para explicar esta alteración de la albita y el progreso de la caolinita, se obtuvieron con los índices químicos pia, ciw, cia y la relación al2o3/na2o, que junto con firmas mineralógicas, sugieren que la caolinita es el resultado de la disolución gradual debido a la hidrólisis ácida de la albita en dichos ambientes ácidos, que también puede ser atribuido a la influencia de la materia orgánica. palabras clave: relaciones geoquímicas, perfiles de suelo, alteración diferencial, índice de alteración de las plagioclasas (pia), hidrólisis ácida albita resumen https://doi.org/10.15446/esrj.v25n1.74167 mailto:salca@unileon.es https://doi.org/10.15446/esrj.v26n1.81087 https://doi.org/10.15446/esrj.v26n1.81087 56 sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero56 introduction the processes controlling the alteration of primary minerals depend on the relative mobility of some major components (anderson & hawkes, 1958; johnson et al., 1968; white et al., 2001; anderson et al., 2002; white & brantley, 2003; wilson 2004; meunier et al., 2007; white & buss, 2014). basically, the unique genetic relationship between the soil and the parent material through the profiles helps to understand the role of rock weathering and soil formation. the influence of the parent material mineral composition on the genesis of the soils is still a question to be explored, as widely shown in the traditional literature (chesworth, 1973a). therefore, the main source of the bases is related to in situ chemical weathering of local bedrocks (munroe et al., 2007). special importance has been given to the removal of alkali and alkaline earth elements (na, ca, mgand k) during the continental weathering (chesworth, 1973b), due to the alteration of clay minerals and a high removal of na from dissolution of feldspars and plagioclases, followed by a progressive enrichment in silica and al and fe sesquioxides (wilson, 2004); occasionalyy, desilification occurs at the final weathering stage. during weathering the rock forming primary minerals, micas and feldspars, are depleted in bases; in this way, k release is considered a good measure of the degree of alteration of the aluminium silicate minerals (wilson, 2004). in addition, chemical weathering of silicate rocks through acid hydrolysis could take place and lead to an exchange of na, k, ca and mg for h and, perhaps, a loss of si (bain et al., 1990; ezzaım et al., 1999a, 1999b; papoulis et al., 2004). according to banfield & eggleton (1990). both processes, the loss of k and na and the enrichment in al, are in consonance with an advancing process of kaolinization and formation of iron and aluminium oxyhydroxides too. this fact has been explained by some authors for weathering soil profiles from the iberian massif (molina et al., 1990; vicente et al., 1991; vicente et al., 1997; molina & cantano, 2002; jiménezespinosa et al., 2007; núñez & recio 2007; fernández-caliani & cantano 2010; doval et al., 2012). other relevant factors regarding weathering in the traditional literature include the fundamental alteration mechanisms and the products derived from the primary minerals (banfield & eggleton, 1990; merriman et al., 1990; robertson & eggleton, 1991; robert & tessier, 1992; gardner & walsh, 1996; aspandiar & eggleton, 2002, spears, 2016). many secondary clays are commonly considered an alteration product in acidic soil environments as explained by pédro (1997). for example, the evolution of parent rocks controlled by an hydrolysis phenomenon, including bisialitisation, monosiallitisation and allitisation, within the formation of minerals from the smectite group, kaolinite group gibbsite and iron oxides group found in the pedological horizons. evaluating geochemical composition, not only releases basic information about the formation conditions, but also the evolution and the alteration stages of the rock substrates and soils. wilson (2004) noted that the differences in alteration mechanisms, products and the rates of weathering commonly observed among bedrock and soil environments can differ widely, even in the same geological units. nesbitt et al. (1980) already mentioned degradation and leaching dominate the early weathering stages, whereas during the advanced stages, ion exchange and adsorption onto clays are of most influence. banfield & eggleton (1990), fedo et al. (1995) and price & velbel (2003) suggested a sequential formation of secondary weathering products and pointed out the utility of evaluating weathering indices attending to simple profile depth variations. thus, evaluating chemical weathering indices and element relations between mobile or immobile elements help rapidly to understand weathering intensity, possible processes and, sometimes, soil grade of development and evolutionas proven for instance by gardner & walsh (1996). the main chemical alteration indices (ruxton 1968; parker 1970; nesbitt and young 1982; harnois 1988; jayawardena and izawa 1994; fedo et al. 1995) have deeply contributed to the understanding of weathering a great variety of geological substrates. even though, some of the conventional indices, mainly use to estimate weathering intensity, should be reformulated as an alternative approach to evaluate different rock weathering grades (guan et al., 2001; meunier et al., 2013). nevertheless, chemical indices of alteration resulted really helpful when evaluating the losses or enrichment of base cations from the rocks however, evaluating the release of silica and relative accumulation of iron and aluminum in soils from the rock weathering requires being adapted (meunier et al., 2013). additionally, several studies highlighted implications for paleoweathering conditions and provenance as pointed out by fedo et al. (1995). recent reviews on climatic and geochemistry (perri, 2020 or dinis et al., 2020) enlighten their strongly interactions using chemical indices as proxies. recently, dinis et al. (2020) have recognised these significant relationships between mineralogical and geochemical signatures of sediments and rainfall registers at catchment scale. moreover, some other factors take an important role in such complex context, particularly diverse physical factors driven by topography as suggested by tardy et al. (1973). this work characterizes the mineralogical and geochemical composition of a series of six natural profiles consisting of soils (s series) and parent rocks (r series) developed on palaeozoic lithologies from the iberian massif under similar weathering conditions in temperate and acidic environments. the influence of lithology on mineral and geochemical properties and the assumption of mineral alteration by differential weathering have been addressed. the specific aims of this research are: 1) characterize the major compositional mineralogy, particularly focusing on fine-grained and clay fractions in six selected weathering profiles; 2) identify the geochemical signatures according to the chemical indices of alteration; 3) understand the factors controlling the weathering processes and 4) explain possible differential chemical weathering mechanisms presumably taking place in such acidic environments. geological setting the profile sites studied are located in the nw part of the iberian massif (variscan orogenic belt) included in the west asturian–leonese zone (walz) (figure 1), specifically in the navia-alto sil domain (nasd) (pérez-estaún et al., 2004; marcos et al., 2004). the variscan orogeny in the walz was a polyphasic deformative process associated with synkinematic metamorphism under epizonal deformation conditions, mainly to greenschist facies (martínezcatalán et al., 1990; suárez et al., 1990), with mineral paragenesis associations typical of the chlorite zone. the walz, including the lower to upper paleozoic units, is characterized principally by the presence of a thick assemblage of cambrian-ordovician siliciclastic rocks, which shows an overall structure, formed largely by recumbent folds and thrusts. figure 1. weathering sampling profiles (sp-1, sp-2, sp-3, sp-4, sp-5, sp-6) a) location in the lithological units from the walz (nw iberian massif, spain). b) synthetic stratigraphic column modified from pérez-estaún (1990) showing formations and series form ages uc: upper carboniferous; s: silurian; uo: upper ordovician; mo: middle ordovician; lo: lower ordovician and cm: cambrian. map source: the national geological survey institute of spain (igme) modified from rodríguez fernández (1982). the lithostratigraphic sequence starts with the lower paleozoic materials of cambrian to ordovician age (pérez-estaún et al., 1990). it begins with detritic facies characterized by an alternation of quartzites with slates, siltstones and sandstones referred to as the los cabos series (cua). a range of later paleozoic materials, referred to as the transition layers, gradually gives way to the luarca formation (lua), an assemblage of fine grained black slates, rich in sulphides like pyrite and organic matter. this is immediately followed by the agüeira formation (agu), formed by alternating sandstones, siltstones and shales with a presence of turbidite depositional environment, which sometimes culminates in a quartzite layer. the silurian strata, poorly represented, are found in a gradually alternating section, including fully sandy phases with varying proportions of slates. these 5757mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone are discordantly overlain by upper carboniferous (carb) (stephanian b-c) synorogenic materials of the el bierzo coal basin (barba et al., 1994), which represent the development of a strong paleorelief. the carboniferous unit consists of a broad range of sediments such as alternating layers of sandstones, shales, siltstones and conglomerates combined with intermediate coalbed deposits (colmenero et al., 2002; wagner, 2004). materials and methods the sampling profiles (sp) are located in six different lithostratigraphic units within the paleozoic succession named as cua, lua, agu and carb (table 1): los cabos series (sp-6)-cua, luarca formation (sp-3 and sp-5)-lua, agüeira fm (sp-2)-agu, and carboniferous series (sp-1 and sp-4)-carb. samples including partially weathered or fresh rocks (r-series) were named as: r1 (figure 2a), r2 (figure 2e) , r3 (figure 2b), r4 (figure 2f), r5 (figure 2g) and r6 (figure 2h); other altered weathered geological materials in contact with the bedrocks beneath, including the soil (s-series) named as: s1 (figure 2a), s2.1 and s2.2 (figure 2e), s3.1 and s3.2 (figure 2b), s4 (figure 2f), s5 (figure 2g), s6 (figure 2h), were collected in the sampling sites from sp-1 to sp-6 as showed in figure 1.the correspondence between the profile sites and samples series is specified below (table 1): sp-1 (samples r1 and s1)-carb; sp-2 (samples r2, s2.1 and s2.2)-agu; sp-3 (samples r3, s3.1 and s3.2)-lua, sp-4 (samples r4 and s4)-carb; sp-5 (r5 and s5)-lua; sp-6 (r6 and s6)-cua. table 1. descriptive features of the investigated sampling profiles soil horizons (s-series) and bedrocks samples (r-series) location (ul: upper and ll: lower level of weathering) according to depth, general profile weathering degree descriptive material features (hd: high; md: moderate, ld: low degree), lithological units, parent rocks and soil series. *acronym for the geological unit’s name profile site u.t.m. coordinates (29n) sample rock/soil horizon depth (cm) level of weathering weathering degree geologic system/ formation lithological unit* parent material soil series sp-1 x:721.188.6 y:4.730.234.5 r1 r +18 ll hd carboniferous carb beige argillite carb-boezas1 ah 0-18 ul sp-2 x:722.255.7 y:4.735.448.2 r2 r + 25 ll md agüeira agu sandy slates and grey siltstones agu-igüeña s2.1 ah1 0-8 ul s2.2 ah2 8-25 ul sp-3 x:720.118.1 y:4.737.203.2 r3 r + 20 ll hd luarca lua black slates and shales lua-urdiales s3.1 ah 0-12 ul s3.2 a/r 12-20 ul sp-4 x:725.378.6 y:4.732.542.95 r4 r +25 ll md carboniferous carb sandstones and sandy siltstones carb-rodrigatoss4 ah 0-25 ul sp-5 x:729.825.63 y:4.736.191.3 r5 r +15 ll md luarca lua black slates and shales lua-tremors5 ah 0-15 ul sp-6 x:722.500.26 y:4.740.380.14 r6 r +15 ll ld los cabos series cua quartzites cua-colinas del campos6 ah 0-15 ul 58 sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero58 figure 2. soil profiles showing two different levels of weathering ul: upper level; ll: lower level where samples were taken: image a) profile sp1 (s1 and r1) and image b) profile sp3 (s 3.1, s3.2 and r3). images from e) to h) soil profile sites sp2, sp4, sp% and sp6 respectively. thin sections showing the rock samples microtextures and granulometry r1 (image a1) and r3 (image b1); r2 (image e1), r4 (image f1), r5 (image g1) and r6 (image h1). observational scale at 1 mm. sem bse images (a2 at 100 µm and b2 at 20 µm captured scale; e2, f2, g2 and h2 at 50 µm captured scale) sem bse detailed images for identification of muscovite (ms) and illite (ill) on r1 (image c) and quartz (qtz), muscovite (ms), chlorite (chl) and albite (ab) on r3 (image d). microscopy scanning electron microscopy with energy dispersive x-ray spectrometry (sem-eds) was carried out as a supporting tool to validate mineral petrographic analysis. thin sections from sample substrates were previously studied by optical microscopy (pom, transmitted and reflected light), using a nikon eclipse e600 pol petrographic microscope. the same polished thin sections were subsequently carbon coated and studied under a jeol jsm-6480 high resolution scanning electron microscope (sem), equipped with an oxford instruments d6679 energy dispersive x-ray spectrometry (eds) detector. for each specimen acceleration voltages range of 10 to 20 kev, working distance was 10 mm and count rate set at 15000 counts/s. back-scattered electron (bse) images were acquired and processed using inca energy software provided by oxford instruments. x ray diffraction representative portions of powdered rock samples were subjected to x-ray diffractometric analyses. air dried rock samples were pulverized in a restch pm400 automatic rotary planetary mortar mill, collecting the fraction 5959mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone <125 µm. xrd analyses were performed at room temperature and 400 c during 0.75 hours. representative samples were subjected to x-ray powder diffraction, using a siemens bruker d5000 series diffractometer system equipped with cu-k-alpha radiation at 40kv and 30 ma, 0.2 mm receiving slit, a range of 2°-58° 2θ and graphite secondary monochromator facilities. scanning step size was set at 0.02 º 2θ and the scanning rate 1 second per step, giving a scan speed of approximately 2º 2θ per minute and a total duration of 0.75 h. oriented aggregates of the <2.0 µm fine clay fraction were obtained from the soil samples, using suspension after hexametaphosphate dispersion, separation by sedimentation and settling techniques after 24 hours (klug & alexander, 1974). the clay samples were dried at 40°c, homogenized and manually powdered using an agate pestle and mortar. previously, the sample pretreatment was the organic matter removal with a hydrogen peroxide agent (30% v/v) at 25 c. duplicate samples were mixed and shaken in a tapered centrifuge tube. the remainder, which stayed in suspension, was removed and dropped on to a circular glass slide. this was then allowed to dry in the laboratory atmosphere, leaving an oriented clay film on the surface of the glass slide. each glass slide, with its clay film, was subjected to x-ray diffraction analysis using a panalytical empyrean ii diffractometer in the air-dried state, using cu-k-alpha radiation and in this case a range of 2°-46° 2θ for soil clay samples. the slides were then placed in a desiccator containing ethylene glycol, and exposed to the glycol vapour at room temperature for several days. the glycol-treated slides were then heated at 400 c for one hour and subjected to xrd analysis. identification of the minerals in the parent rocks and soil clays was carried out following the general indications of moore & reynolds (1989). chemical analyses chemical analyses were performed within duplicated samples including bulk chemistry and fine grained fractions <2.0 µm for clay soil samples. major element analyses were carried out by x ray fluorescence spectrometry (xrf) and, alternatively, some of the major and minor elements were measured by inductively coupled plasma mass spectrometry (icp-ms). the fused borosilicate disks were prepared with 95%pt/5%au crucibles, adding 0.500 g of a 1:1 mixed fusion flux of lithium tetraborate/metaborate (50% li 2b4o7/50% libo2), in a fusion melting induction furnace heated to 1000 ºc during 20 min, using a non-wetting agent of libr. they were analysed in a bruker-nonius s4 pioneer wavelength dispersive xrf spectrometer. the samples prepared following the specifications of the alkali fusion technique required acid dissolution with hno3 5% (v/v). they were measured by icp-ms in a thermo scientific quadrupole xseries2, with detection limits of 0.01% using as standard elements in, sc and rh. the major element oxides cao, mgo, na2o, k2o, sio2, fe2o3, al2o3, p2o5, mno and tio2 were expressed as weight percentages (wt %). previously, representative splits of each sample were oven-dried, and then heated at 975 c for 12h in a high temperature furnace hobersal12-pr400to obtain the loss on ignition (loi) percentage in each case. results mineralogy in the preliminary mineralogical identification for rocks the principal minerals are phyllosilicates of the mica group, mostly ms-ill (figure 2c) and chlorites (figure 2d) respectively, accompanied by quartz (3.34 å) and feldspars, according to the major peaks (3.19 å and 3.22-3.24 å) correspond to plagioclases (albite) as confirmed by sem analyses (figure 2d) or k-feldspars, in the case of r6 (figure 2h2). the microscopic and sem bse analyses can be found detailed on figure 2. it can be noted that albite is present on r3, r5 and r4 (figures b2, f2 and g2) and, clearly more abundant, on r2 (figure e2), all assessing the xrd analyses too. in addition, kaolinite (kln) is only present in some of the r-samples (table 2). there is a relative higher abundance of chlorite in samples r3, r5 and r2, the albite content is higher in r2, however kaolinite content is higher in r1 (table 2). the major minerals in the sample r1 are mica (muscoviteillite), quartz and kaolinite respectively (ms-ill, qtz and kln) as seen in figure 3. kaolinite is clearly identified in sample r1 (figure 3a). by contrast, an iron-rich chlorite identified in samples r2 to r5 as shown in (figure3b), sem analysis confirm that correspond to chamosite (figure 3b). from samples r2 to r6 the peaks at 10 å and 5.0 å in all patterns analysed are indicating mica, but finally the sem results shown enough evidence to be identified as muscovite, so it was also confirmed by sem. it is noted that the peak of 3.34 å can be also of mica-illite, so the reflection of 4.26 å would be enlightening. this also applies to sample r1 (figure3a), although overall intensity is relatively low for that pattern, suggesting a mica (muscovite) rather than a less well-ordered illite phase as confirmed by sem analysis (figure 2c). table 2. mineral identification in the rock samples and relative mineral abundance referred to the total mean quartz contents. abbreviations: qtz = quartz; ms = muscovite; ill = illlite; chl = chlorite; kln = kaolinite; pl = plagioclases; ab = albite; kfs = k-feldspars. conversion scaled weights: ****= 3.0 (very abundant) ***= 1.0 (abundant) **= 0.4-0.6 (frequent) *= 0.3-0.2 (present) tr <0.2 (traces) and – = not identified. profile site sample qtz ms ill chl kln pl kfs sp-1 r1 ** *** *** – ** – – sp-2 r2 ** *** ** – *** – sp-3 r3 ** *** ** * * – sp-4 r4 *** ** * ** – sp-5 r5 ** *** ** * * – sp-6 r6 **** tr – – tr tr figure 3. rock powder unoriented xrd pattern, air-dried and solvated at room temperature 25 c (bottom line) and heated at 550 c (top line). major minerals identified in a) sample r1 and b) sample r3. only the range from 5° to 35°2θ is shown in the pattern. the principal mineral phases present in the soil samples from the clay oriented aggregates patterns are given below in table 3. in particular, the sequences are essentially the same identified before (micaceous phases and illite, with iron rich chlorite, traces of quartz, feldspars and minor kaolinite with some accompanying iron oxides). particularly, only minor ferric amorphous species like lepidocrocite and iron oxides were identified as traces (figure 4a). a small peak at 6.2 å on glycol (and also air-dried diffractogram) that disappears 60 sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero60 on heating suggests a trace of iron oxides corresponding to lepidocrocite in s5 and s1 (figure 4a). the presence of ferroan chlorite has been confirmed in samples s2.1, s2.2, s3.1, s3.2 (figure 4b) and s5 through the strong peaks at 7.1 å and 3.53 å, with smaller peaks at 14 å and 4.7 å, suggesting an iron rich variety. small peak at 3.58 å on saturated patterns indicates additional minor kaolinite detected in samples s3.1, s3.2 (figure 4b), s4 and s5. peak at 7.1 å, not significantly affected by heating accompanied by a small peak at 3.58 å, especially on heating, suggests that kaolinite is also present (figure 3a). the absence of strong peaks at 14 å, 7.1 å and 3.53 å, especially with glycol, indicate that chlorite is not present in this particular sample (figure 4a). moreover, in the samples from s2 to s5 peaks at 14 å, 7.1 å and 3.53 å, glycol solvated (figure 4b), indicate presence of chlorite; a small peak at 3.58 å on heating, suggests that kaolinite is also present (figure 4b). quartz (peaks at 3.35 and 4.26 å) and k-feldspar (peak at 3.2 å) are also representative in the composition of the soil samples showing sharp peaks in reflections (figure 4b). table 3. total chemical composition determined in the six profiles for parent rocks and soil samples. contents of major components analysed by xrf and icp-ms (weight %), *loi (loss on ignition). profile site sample depth (cm) sio2 al2o3 fe2o3 k2o mgo cao na2o tio2 p2o5 mno loi* total sp-1 r1 +18 56.67 23.38 5.79 4.89 0.73 0.26 0.50 1.00 0.06 0.02 6.74 100.04 s1 0-18 67.01 17.58 7.31 3.85 0.63 0.41 0.48 0.79 0.16 0.02 1.15 99.39 sp-2 r2 + 25 66.99 15.69 5.42 2.45 1.93 0.42 2.17 0.99 0.02 0.05 4.36 100.49 s2.1 0-8 73.23 13.38 5.75 2.58 1.35 0.47 0.84 1.62 0.08 0.04 0.85 100.19 s2.2 8-25 68.46 15.30 6.40 2.98 1.64 0.42 0.88 2.09 0.08 0.04 1.11 99.40 sp-3 r3 + 20 59.69 20.03 6.64 3.13 1.69 0.52 1.42 1.10 0.07 0.07 5.68 100.04 s3.1 0-12 63.84 19.85 7.91 2.97 1.35 0.35 0.80 1.14 0.12 0.04 1.04 99.41 s3.2 12-20 65.22 19.82 7.50 2.15 1.53 0.34 0.70 1.24 0.10 0.05 0.76 99.41 sp-4 r4 +25 65.14 18.17 3.86 3.40 1.50 0.34 0.91 1.09 0.06 0.03 5.63 100.13 s4 0-25 82.11 9.92 3.04 1.73 0.49 0.23 0.34 0.54 0.06 0.01 0.94 99.41 sp-5 r5 +15 56.80 22.72 7.63 2.95 1.77 0.46 0.93 1.13 0.08 0.07 5.84 100.38 s5 0-15 60.80 22.11 9.30 1.94 1.06 0.32 0.86 1.61 0.15 0.17 1.07 99.39 sp-6 r6 +15 97.45 1.07 0.17 0.44 0.05 0.42 0.00 0.08 0.02 0.00 1.01 100.71 s6 0-15 95.03 1.63 0.59 0.55 0.06 0.19 0.05 0.45 0.01 0.00 1.42 99.98 figure 4. clay oriented aggregates a xrd pattern from a representative sample a) s1 in the profile sp-1 and b) s3 in the profile sp-3 showing k-glycol treatment (bottom line) and heated at 400 c (top line). only the range from 5° to 35°2θ is shown to clarify the interpretation in the pattern. 6161mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone rocks and soils geochemistry the contents of major elements in the samples are summarized in table 3. the contents found in the soils are coherent with those ones present in the rocks through the soil weathering profiles. sio2, fe2o3 and al2o3 contents are quite similar in soil profiles sp1 to sp5 being the most abundant elements. however in s6 from sp-6 the content of sio2 reaches about 95%, while in the rest of soil samples get around 65-70%. the bases k2o, mgo, na2o and cao are generally lower than those present in the rocks. k2o (2-3%) shows higher contents among the last ones and, together with mgo (1-2%), are more abundant than cao (0.2-0.4%) and na2o (0.5-0.9%). the content of na2o is slightly higher in the soil samples of sp-2 and sp-3 (table 3). regarding loi results, the highest values correspond to sp-1 where r1 reaches 6.74 in the carboniferous series and the lowest to sp-6, where r6 is 1.01 (table 3). the edaphic contents are little higher than the lithogenic ones for sio2, fe2o3 and al2o3 (table 3). the parent rocks show relatively important amounts of na2o, fe2o3 and mgo as well as k2o which are retained in the less altered minerals (table 3). so, the contents found in the residual soils show an enrichment of sio2, fe2o3 and al2o3 too (table 3). chemical indices of alteration and element ratios weathering indices were calculated using molar proportions, increasing alteration indicated by increasing cia (chemical index of alteration; nesbitt and young, 1982), ciw (chemical index of weathering; harnois, 1988) and pia (plagioclase index of alteration; fedo et al., 1995). ciw, cia and pia have reached higher values in sp-1, sp-3 and sp-5, all are in consonance with the geochemical base mobilization and depletion of major cations calculated (table 4). however, minor differences have been shown in the present data from the soils and underlying rocks (table 4). ciw and pia show quite good correlation fitting (r2=0.49) (figure 5a), although cia and pia reaches a better correlation r2=0.94. the fitting between ciw and cia reaches a positive r2=0.99. the higher values of the cia and ciw are indicating a possible high grade of alteration, mainly in the rock samples from profiles sp-1, sp-3 and sp-5. piw present high values specifically suggesting one of the most possible ways of primary mineral alteration (table 4). it is also assumed that loi, the only one not expressed as a ratio, increases with the degree of weathering. loi has also resulted higher in sp-1, which is in consonance with the rest of indices calculated. a r2=0.30 fitting in the correlation with ciw (figure 5b) showing higher correlation in the rock series than the soil ones. table 4. indices associated to the intensity of mineral alteration and weathering. abbreviations: cia: chemical index of alteration; ciw: chemical index of weathering; pia: plagioclase alteration index; compositional molar ratios calculated from major oxides related to chemical and mineral weathering indices. abbreviations: r2o3=(fe2o3+al2o3); bxo=(na2o+k2o+cao+mgo). profile sample depth (cm) cia ciw pia sio2/ al2o3 sio2/ r2o3 sio2/ tio2 sio2/ k2o bxo/ r2o3 bxo/ al2o3 k2o/ na2o al2o3/ na2o sp-1 r1 +18 77.99 94.71 93.27 4.11 3.55 75.03 18.13 0.31 0.76 6.39 28.22 s1 0-18 75.50 91.96 89.71 6.46 5.10 111.98 27.23 0.33 0.42 5.29 22.31 sp-2 r2 + 25 69.20 78.36 75.06 7.23 5.92 89.55 42.81 0.62 0.55 0.74 4.40 s2.1 0-8 72.68 85.65 82.54 9.27 7.28 59.80 44.48 0.50 0.63 2.01 9.64 s2.2 8-25 73.79 87.40 84.55 7.58 5.98 43.45 35.90 0.49 0.63 2.24 10.61 sp-3 r3 + 20 75.01 85.91 83.52 5.05 4.17 71.88 29.82 0.45 0.53 1.45 8.54 s3.1 0-12 79.34 91.03 89.48 5.45 4.34 74.48 33.64 0.34 0.43 2.45 15.13 s3.2 12-20 82.91 91.85 90.86 5.57 4.49 69.50 47.51 0.32 0.40 2.03 17.27 sp-4 r4 +25 75.79 89.53 87.22 6.07 5.35 79.43 29.99 0.46 0.44 2.46 12.13 s4 0-25 77.67 90.99 89.12 14.01 11.72 201.79 74.41 0.35 0.41 3.33 19.37 sp-5 r5 +15 80.32 90.53 89.15 4.23 3.49 66.40 30.17 0.36 1.27 2.07 14.75 s5 0-15 84.35 91.69 90.90 4.66 3.67 50.20 49.09 0.24 0.31 1.48 17.70 sp-6 r6 +15 46.18 58.14 43.50 154.21 139.73 1546.77 345.99 1.16 0.72 0.00 0.00 s6 0-15 61.49 49.44 70.96 98.47 79.99 282.97 268.03 0.59 0.36 7.12 15.60 figure 5. correlation between chemical indices a) pia and cia b) loi and ciw a b 62 sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero62 selected molar ratios, based on the principle that the ratio between concentrations of mobile (such as sio2, cao, mgo, k2o and na2o) and immobile elements (al2o3, fe2o3, tio2), should increase with weathering have been studied. these molar ratios such as sio2/al2o3 (table 4) related to silica leaching result a dominant factor in the characterization of the weathering together with chemical weathering indices (ciw, cia and pia) related to weathering intensity. the main chemical indices are expected to increase over time as leaching progresses. some other selected ratios form the literature present a different behavior. for instance, sio2/k2o tends to show lower values when the weathering is higher (sp-1) (table 4). opposite to that, sio2/tio2 remains almost constant with no noticeable differences as well as sio2/al2o3 and sio2/r2o3 both are in consonance among the rocks and soil samples not showing remarkable differences (table 4). other ratios regarding the bases give an estimation of weatherable primary minerals remaining in soils, such as bxo/ r2o3 and bxo/al2o3 (table 4), decreases over time, so less weathered materials, for instance r6, present higher values (table 4). k2o/na2o and al2o3/na2o are both good indicators the weathering degree showing lower values related to less alteration (sp-2) and higher values for more altered phases (sp-1) (table 4). discussion geochemical relations, elements mobility and mineral alteration in general, the relative chemical composition variations are neither high nor too significant, comparing the upper and lower parts in the profiles, since the sampling depth is rather limited (table 3). generally, the profiles present an increasing content of some elements, noticeable in sio2, al2o3, and fe2o3, from the parent rocks to the upper levels of the soil profiles (table 3) while cations are leached. it is particularly important the relatively high k2o content found in all of soils, especially in sp-1 (figure 6). na2o, mgo and cao are present in soils, but not significantly enriched (figure 6). according to the results, there is a net gain of sio2, al2o3, and fe2o3 in the soils, in particular, from profiles sp-1, sp-3 and sp-5 (table 3). otherwise, k2o, mgo, cao and na2o are still relatively higher in the parent rocks (figure 6), this last one especially in sample r2 from sp-2 due to the abundance of albite (table 2). however, other leachable elements such as cao do not show this clear tendency, so the content is rather limited in both rocks and soils, since it is only enriched in sp-1 (figure 6). meanwhile, soil mgo gain seems to be more evident, in lower parts of sp-2 and sp-3 (figure 6) but limited in the rest. the k2o depletion from rocks and enrichment in soils, particularly shown in sp-1 and sp-6, is mainly the result of micas alteration (figure 6). the bases mobilities against iron sesquioxides are evident, but not high. the alteration causes hardly any variation in the relative amounts of na2o and mgo, but leaching in the upper layers preferentially eliminates mgo (figure 6). exactly the same happens with cao with respect to na2o. the mobility of k2o is always greater than na2o, and k2o, seems to be the most mobile, as there may be no minerals to retain it. in short, a gradual depletion of bases and enrichment in sio2 and iron and aluminium sesquioxides is observed. al2o3 contents found in soils are not higher than those ones in rocks (figure 6). al2o3 is slightly less common in soils, particularly in sp-4 and sp-1 and less noticeable in the rest (figure 6). indeed, the enrichment of silica, though not marked, is observed in most of soil profiles, especially in sp-4. sio2 accumulation is normally followed by sio2 mobilization and partial desilification, according to the removal as soluble colloidal form leached out of the profile. under these climatological conditions the intense leaching left a material poor in bases, since there is hardly any variation in their content. figure 6. enrichment and depletion in the soil profiles: composition of soil horizons showing mean concentrations and standard deviation of major components e xpressed in oxides (%). 6363mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone lower al2o3 found in soils than in underlying rocks must be related with other physical processes besides chemical decomposition related to the removal of fine-grained clay particles enriched in al2o3. dinis & oliveira (2016) explained that the composition of metasedimentary-derived sediments is not substantially affected by their grain size distributions. it could be explained by the dissolution leaching out of the profile, as colloidal particles clay fractions enriched in al2o3, as happens when sio2 migrates in the edaphic environment. this could explain the lower al2o3 contents found in some soils due to dissolution, leaching and final depletion out from the soil profile. as seen before, the profiles experience a gradual and little noticeable mobilization of bases from the destruction of the dominant primary minerals, mainly micas, fe-bearing chlorites and feldspars and plagioclases, which is followed by a depletion in the upper parts of the soil profiles as well, as seen in doval et al. (2012), ling et al. (2016) and (2018). these losses were particularly observed in the early stages of the soils pedological development in agreement with senol et al. (2016). it is common generalizing the genesis of highly weathered soils which results in intense loss of bases (na2o, cao, k2o and mgo), whereas less soluble al2o3 accumulates in newly formed solid phases such as secondary 1:1 phyllosilicates like kaolinite, or mineral forms of aluminium hydroxide, such as gibbsite too (romero et al. 1992). however, taboada & garcı́a (1999) showed that much evolution of the material is not necessary to explain the abundance of kaolin minerals and gibbsite, as these are compatible in many saprolites with absolutely fresh feldspars; in this case explaining the coexistence of albite and kaolinite supporting the evidence shown in this paper. taboada & garcia (1999) also reported that gibbsite may even be the mineralogical product formed from the plagioclases in the initial weathering stages when intense leaching occurs. figure 7. cross data plots within linear correlation and r2 fitting for a) al2o3/na2o and k2o/na2o and b) al2o3/na2o and pia 64 sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero64 albite alteration supported by chemical indices indices associated to the intensity of mineral alteration and chemical weathering (ciw, cia and pia) give a quantitative measure of feldspar weathering, by relating al2o3 enrichment in the weathering in contrast to na2o, cao and k2o, which should be removed from the soil profile during plagioclase or k-feldspar weathering (nesbitt and young, 1982; white et al., 2001; price and velbel, 2003). some element ratios biplots shown in figure 7, help to clarify the interpretation. the molar ratios al2o3/na2o and k2o/ na2o (figure 7a) show clearly differences, mainly in sp-2 sample r2 where albite composition is higher and the ratios for mobile/mobile elements k2o/ na2o and al2o3/na2o (figure 7a) show the opposite in sp-1 where kaolinite is dominant in composition, due to in r1 and s1 albite is not present. al2o3/ na2o with k2o/cao with a correlation fit r2=0.56, indicative of the possible weathering increase with a linear positive tendency of the plot (figure 7a). al2o3 and k2o enrichment and na2o mobilization are both indicative of the gradual plagioclase alteration, so chemical indices support the main hypothesis in this case of albite alteration. the cross-dot plots reflect a good fitting and indicate mineral composition, where albite unaltered is dominant or kaolinite is dominant (figure 7a). furthermore, al2o3/na2o result a key factor in the characterization of the grade of weathering (figure 7b), when plotting against plagioclase index of alteration (pia) (fedo et al., 1995) and with chemical weathering indices (ciw) (harnois, 1988). al2o3/na2o is considered a good indicator of the degree of weathering if we supposed lower values for less altered materials where plagioclases (albite) still remain in parent rocks. weathering chemical indices selected pia and ciw show a clear correlation, similar to the one obtained for al2o3/na2o and pia for rock and soil series (figure 7b). this positive correlation (r2=0.54) indicates a progressive weathering of albite while advancing the kaolinization in the most altered soil profiles sp-1, as the higher pia values suggest, and less alteration in sp-2 (figure 7b). in short, cations such as na, k and ca commonly generated by the weathering of feldspar and plagioclases into kaolinite, are rapidly release in the soil domain (fedo et al., 1995). pia, ciw and cia and other chemical relations such as al2o3/na2o help to understand the most reliable assumption that kaolinite comes from the gradual alteration of albite. despite the evidence of element leaching, although quite limited, during the data evaluation, geochemical indices such as pia and al2o3/na2o ratio are good indicators of possible differential weathering intensities in these profiles. differential weathering signatures the main findings point out some possible differential weathering signatures supporting the mineral alteration of albite, especially the clue given by the identification of primary phyllosilicates and their coexistence with other secondary phases. this finding gives some enlightening data indicative of possible different processes affecting the alteration profiles. particularly, special emphasis is given to the albite occurrence, which is found coexisting with kaolinite in the mineral sequences found in the profiles. this assumption may be explained by at least three particular conditions as shown in literature. kaolinite, firstly supposed to be directly inherited in soils from the parent substrates when formerly present (especially abundant in r1 from sp-1). however, this explanation was not enough in the profiles where kaolinite does not exist originally since the edaphic formation of this mineral from the weathering of silicates requires extreme alteration condition (molina et al., 1990; hernando massanet et al., 2004); not possible in the current studied soil conditions. secondly, the weathered plagioclases drive to the occurrence of derived secondary kaolinite associated to a pedogenic origin (driese et al., 2007; churchman & lowe, 2012). strong pedogenic dissolution and alteration of feldspars and mafic minerals to kaolinite, gibbsite and iron oxides, as well as geochemical losses of na, ca, mg and k (driese et al., 2007) must be necessary to understand this fact occurring under particular conditions. in this work, the results have shown certain amounts of amorphous species like lepidocrocite and other iron oxides which are also associated to organic phases and weathering process in slates and shales rich in sulphides and organic phases, which are the dominant parent rocks. adding to the previous, the hydrolytic weathering the main cause of a progressive transformation of affected components into clay minerals, ultimately into kaolinite (romero et al. 1992; taboada & garcı́a, 1999), following the silicate weathering generally known as monosialitization. this acid hydrolysis favors the mobilizations of some bases, including in later stages, and the release of ba and k from the partial alteration of k-feldspars, plagioclases and micas (fernández-caliani & cantano, 2010). in short, kaolinite has been widely shown as the stable weathering product of the feldspars, plagioclases (albite) and muscovite (núñez & recio, 2007). in such context, it could enhance to explain, at least partially, the formation of secondary clay minerals belonging to the kaolinite group and the expansive clays and 2:1 mixed layers, more abundant in areas where a dry period of intense evaporation generally occurs. finally, the last assumption may explain that kaolinite and albite could coexist in different size-fractions because weathering was still not able to destroy all albite. these rock substrates which are mainly made up of fine to very fine fractions, mainly silt and clay particles, are not supposed to be substantially affected by the grain size distribution of the sediment in agreement with dinis & oliveira (2016). moreover, very fine grained fractions in the profiles may be depleted due to loss process controlled by diverse physical factors driven by topography (tardy et al., 1973), remaining in the soils only the coarser fractions. this is directly linked to mechanical weathering and erosion rates. this last explanation associated to intense erosive activity requires great mineral similarity among weathered and fresh substrates, together with a low development in the soil profiles due to a continuous renovation-denudation in the recycling. no less important is the narrow relationship which links geochemical composition with the contrasting climatic conditions. dinis et al. (2020) and perri (2020) have recently showed how chemical indices act as good proxies for chemical weathering to reconstruct past climate conditions. dinis et al. (2016 and 2017) have also investigated the roles played by distinct processes on the weathering intensity and the palaeoclimatic conditions, demonstrating the composition of clay is strongly influenced by climatically driven weathering. otherwise, perri (2018) has shown higher abundance of kaolinite related to warm–humid conditions. moreover, perri (2020) released medium-low cia values for similar climatic areas, in terms of mean annual temperature and precipitation, which underwent moderate-weak chemical weathering similar to this study. the results do not differ a lot from the values found in this study for the ciw, cia and pia; quite similar too to those obtained in previous works of scarciglia et al. (2016) or perri (2016). in the light of the chemical indices and the linkages showing good relations with geochemical data, in this particular case, the albite hydrolytic alteration weathering by acid dissolution is still the main hypothesis supporting the presence of kaolinite in the most altered investigated soil profiles. perri et al. (2015) demonstrated in granitoid weathering simulations the co2-controlled dissolution of albite-rich plagioclase is the most important reaction, followed by the dissolution of k-feldspar, biotite, chlorite, and muscovite, in order of decreasing importance. in fact, differential chemical weathering occurring in the acidic solution of soils seems to better explain this way of alteration. the acidic solution from the oxidation of both, organic matter, quite important in the composition, and sulphides, abundant in the rocks, could explain the ways of alteration, supporting the presence of such abundant kaolinite. some authors have already demonstrated the influence of organic matter in the acid dissolution to explain kaolinite presence (gallardo lancho et al.1976; romero et al. 1992, fischer et al., 2009; li et al., 2014; emberson et al., 2016; ling et al., 2016). solute transfer and elements mobility by acidic percolation of soil solutions should be studied further in order to clarify this assumption. so that, solute transfer and elements mobility by acidic percolation of soil solutions and leaching data should be studied further to give a concluding explanation like in other studies supported by cronan (1985), bain (2007) or di figlia, et al. (2007). conclusions the identification of primary phyllosilicates and their coexistence with other secondary clays gives an enlighten clue indicative of possible different 6565mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone weathering processes affecting the soil profiles. particularly, special emphasis is given to the albite occurrence, which is found coexisting with kaolinite in most of the mineral sequences. kaolinite can be directly inherited in soils from parent rocks and partially due to mineral alteration, preferably from albite. in addition, since both could coexist in different size-fractions they are differently affected by weathering. so, it may be concluded that the geochemical composition must be also substantially affected by the grain size distribution of the sediment. moreover, good correlations between chemical indices and element ratios, may indicate kaolinite in this case should come from the alteration of plagioclases during weathering. the traditional chemical indices have also helped to understand different chemical weathering processes taking place. pia, ciw and cia have resulted an essential tool to infer some assessment due to the lack of other more specific data; giving a measure of plagioclase differential weathering; they together with, al2o3/na2o and k2o/na2o, have also supported that kaolinite comes from the gradual alteration of albite mainly due to acid hydrolysis. the differential weathering and albite alteration detected in these profiles is in agreement with the mobility of major elements leading to a general loss of bases (na2o, cao, k2o and mgo). in this work, the relations between the parent rocks composition and soils can easily assist in determining the bases loss and the k2o enrichment from the bottom to upper parts in the soil profiles. in general, a slight enrichment in silica and iron and aluminium sesquioxides attributed to the advancing process of weathering in agreement with most of findings show. however, al2o3 in some profiles is slightly less abundant in soils than in parent rocks, probably because al2o3 in the finest clay fractions is being lost by physical processes. similarly occurring with sio2, firstly enriched in the soil profiles when weathering advances and, later, leached out of the profile. hence, besides chemical weathering, other physical or mechanical processes, due to erosion and removal, are expected to take placein such favorable topographic conditions. this together with a low development detected in the soil profiles leads to think of a continuous renovation-denudation in the recycling. in short, several factors, not only data limited to geological source areas and the critical lithology control, mainly based on geochemical and mineralogical composition, but also morphology and topography, climatic registers, or even vegetation interactions, may be assessed on weathering research. acknowledgments the authors want to express their gratitude to the editors, editor-inchief and the anonymous reviewers for their exhaustive reviews and valuable suggestions on the manuscript which helped to enrich the scientific discussion. thanks for the support given by dr. fernando gómez-fernández on the performance of sem analysis and especially for his expertise on mineralogy interpretations. the authors would also like to thank the help given with the xrd analysis by the school of biological, earth and environmental sciences university of new south wales (unsw). we are also grateful for the help provided by the microscopy laboratory service (university of león, spain). the assistance and the technical support provided by the structural analysis unit sai-uae (university of coruña, spain) and the technical scientific services (university of oviedo, spain). this research was funded by the educational and research authority from castilla & león regional government (leoo3a08) and supported by the spanish ministry of education and science (edu/1262/2009), first author’s phd funding project. references anderson, d. h. & hawkes, h. e. (1958). relative mobility of the common elements in weathering of some schist and granite areas. geochimica et cosmochimica acta, 14(3), 204-210. https://doi.org/10.1016/00167037(58)90079-6 anderson, s. p., dietrich, w. e. & brimhall jr, g. h. (2002). weathering profiles, mass-balance analysis, and rates of solute loss: linkages between weathering and erosion in a small, steep catchment. geological society of america bulletin, 114(9), 1143-1158. https://doi.org/10.1130/0016-760 6(2002)114<1143:wpmbaa>2.0.co;2 aspandiar, m. f. & eggleton, r. a. (2002). weathering of chlorite: i. reactions and products in microsystems controlled by the primary mineral. clays and clay minerals, 50(6), 685-698. https://doi. org/10.1346/000986002762090227 bain, d. c., mellor, a. & wilson, m. j. (1990). nature and origin of an aluminous vermiculitic weathering product in acid soils from upland catchments in scotland. clay minerals, 25(4), 467-475. https://doi.org/10.1180/claymin.1990.025.4.05 bain, d.c. (2007). soil clay minerals and their relevance to environmental change. macla, 7, 57-59. banfield, j. f. & eggleton, r. a. (1990). analytical transmission electron microscope studies of plagioclase, muscovite, and k-feldspar weathering. clays and clay minerals, 38(1), 77-89. https://doi.org/10.1346/ ccmn.1990.0380111 barba, p., nozal, f., rodríguez, a. s. & suárez, a. (1994). ii: estratigrafía. in: itge (editor). mapa geológico de la provincia de león a escala 1:200000. instituto tecnológico geominero de españa, diputación de león, 13-90. chesworth, w. (1973a). the parent rock effect in the genesis of soil. geoderma, 10 (3), 215-225. https://doi.org/10.1016/0016-7061(73)90064-5 chesworth, w. (1973b). the residua system of chemical weathering: a model for the chemical breakdown of silicate rocks at the surface of the earth. journal of soil science, 24 (1), 69-81. https://doi.org/10.1111/j.1365-2389.1973.tb00742.x churchman, g. j. & lowe, d. j. (2012). alteration, formation, and occurrence of minerals in soils. in: huang, p. m.; li, y. & sumner, m. e. (editors). handbook of soil sciences. 2nd edition. vol. 1: properties and processes. crc press (taylor& francis), boca raton, fl, 20, 1-20. 72 pp. https:// hdl.handle.net/10289/9024 colmenero, j. r., fernández, l. p., moreno, c., bahamonde, j. r., barba, p., heredia, n., et al. (2002). carboniferous. in: gibbons, w. & moreno, t. (editors). geology of spain. london, the geological society of london, 120–153. cronan, c. s. (1985). chemical weathering and solution chemistry in acid forest soils: differential influence of soil type, biotic processes, and h+ deposition. in: the chemistry of weathering. springer, dordrecht, 175-195. https:// doi.org/10.1007/978-94-009-5333-8_11 di figlia, m. g., bellanca, a., neri, r. & stefansson, a. (2007). chemical weathering of volcanic rocks at the island of pantelleria, italy: information from soil profile and soil solution investigations. chemical geology, 246 (1-2), 1-18. https://doi.org/10.1016/j.chemgeo.2007.07.025 dinis, p. & oliveira, á. (2016). provenance of pliocene clay deposits from the iberian atlantic margin and compositional changes during recycling. sedimentary geology, 336, 171-182. https://doi.org/10.1016/j.sedgeo.2015.12.011 dinis, p. a., dinis, j. l., mendes, m. m., rey, j. & pais, j. (2016). geochemistry and mineralogy of the lower cretaceous of the lusitanian basin (western portugal): deciphering palaeoclimates from weathering indices and integrated vegetational data. comptes rendus geoscience, 348(2), 139-149. https://doi.org/10.1016/j.crte.2015.09.003 dinis, p., garzanti, e., vermeesch, p. & huvi, j. (2017). climatic zonation and weathering control on sediment composition (angola). chemical geology, 467, 110–121. https://doi.org/10.1016/j.chemgeo.2017.07.030 dinis, p. a., garzanti, e., hahn, a., vermeesch, p. & cabral-pinto, m. (2020). weathering indices as climate proxies. a step forward based on congo https://doi.org/10.1016/0016-7037(58)90079-6 https://doi.org/10.1016/0016-7037(58)90079-6 https://doi.org/10.1130/0016-7606(2002)114%3c1143:wpmbaa%3e2.0.co;2 https://doi.org/10.1130/0016-7606(2002)114%3c1143:wpmbaa%3e2.0.co;2 https://doi.org/10.1346/000986002762090227 https://doi.org/10.1346/000986002762090227 https://doi.org/10.1180/claymin.1990.025.4.05 https://doi.org/10.1180/claymin.1990.025.4.05 https://doi.org/10.1346/ccmn.1990.0380111 https://doi.org/10.1346/ccmn.1990.0380111 https://doi.org/10.1016/0016-7061(73)90064-5 https://doi.org/10.1111/j.1365-2389.1973.tb00742.x https://doi.org/10.1111/j.1365-2389.1973.tb00742.x https://doi.org/10.1007/978-94-009-5333-8_11 https://doi.org/10.1007/978-94-009-5333-8_11 https://doi.org/10.1016/j.chemgeo.2007.07.025 https://doi.org/10.1016/j.sedgeo.2015.12.011 https://doi.org/10.1016/j.sedgeo.2015.12.011 https://doi.org/10.1016/j.crte.2015.09.003 https://doi.org/10.1016/j.chemgeo.2017.07.030 66 sara alcalde-aparicio, m. vidal-bardan, e. alonso-herrero66 and sw african river muds. earth-science reviews, 201, 103039. https://doi.org/10.1016/j.earscirev.2019.103039 doval, m., martín-garcía, r., la iglesia, á. & alonso-zarza, a. m. (2012). clay minerals associations in palaeoweathering profiles from central spain: genesis and implications. clay minerals, 47(1), 117-129. https://doi. org/10.1180/claymin.2012.047.1.117 driese, s. g., medaris jr, l. g., ren, m., runkel, a. c. & langford, r. p. (2007). differentiating pedogenesis from diagenesis in early terrestrial paleoweathering surfaces formed on granitic composition parent materials. the journal of geology, 115(4), 387-406. https://doi.org/10.1086/518048 eberl, d. d., farmer, v. c. & barrer, r. m. (1984). clay mineral formation and transformation in rocks and soils. philosophical transactions of the royal society of london a: mathematical, physical and engineering sciences, 311 (1517), 241-257. https://doi.org/10.1098/rsta.1984.0026 emberson, r., hovius, n., galy, a. & marc, o. (2016). oxidation of sulfides and rapid weathering in recent landslides. earth surface dynamics, 4(3), 727-742. https://doi.org/10.5194/esurf-4-727-2016 ezzaım, a., turpault, m. p. & ranger, j. (1999a). quantification of weathering processes in an acid brown soil developed from tuff (beaujolais, france): part i. formation of weathered rind. geoderma, 87(3-4), 137-154. https://doi.org/10.1016/s0016-7061(98)00061-5 ezzaım, a., turpault, m. p. & ranger, j. (1999b). quantification of weathering processes in an acid brown soil developed from tuff (beaujolais, france): part ii. soil formation. geoderma, 87 (3-4), 155-177. https://doi. org/10.1016/s0016-7061(98)00061-5 fedo, c. m., wayne nesbitt, h. & young, g. m. (1995). unraveling the effects of potassium metasomatism in sedimentary rocks and paleosols, with implications for paleoweathering conditions and provenance. geology, 23(10), 921-924. https://doi.org/10.1130/0091-7613(1995)023<0921:uteopm>2.3.co;2 fernández-caliani, j. c. & cantano, m. (2010). intensive kaolinization during a lateritic weathering event in south-west spain: mineralogical and geochemical inferences from a relict paleosol. catena, 80 (1), 23-33. https:// doi.org/10.1016/j.catena.2009.08.005 fischer, c., schmidt, c., bauer, a., gaupp, r. & heide, k. (2009). mineralogical and geochemical alteration of low-grade metamorphic black slates due to oxidative weathering. geochemistry, 69 (2), 127-142. https://doi.org/10.1016/j.chemer.2009.02.002 gallardo lancho, j., camazano, m.s., alonso, j. s. & sánchez, y. g. (1976). influencia de la materia orgánica en la génesis de gibsita y caolinita en suelos graníticos del centro-oeste de españa. clay minerals, 11(3), 241249. https://doi.org/10.1180/claymin.1976.011.3.06 gardner, r. & walsh, n. (1996). chemical weathering of metamorphic rocks from low elevations in the southern himalaya. chemical geology, 127(1-3), 161-176. https://doi.org/10.1016/0009-2541(95)00089-5 guan, p., ng, c. w. w., sun, m. & tang, w. (2001). weathering indices for rhyolitic tuff and granite in hong kong. engineering geology, 59(1-2), 147159. https://doi.org/10.1016/s0013-7952(00)00071-5 harnois, l. (1988). the ciw index: a new chemical index of weathering. sedimentary geology, 55, 319-322. https://doi.org/10.1016/00370738(88)90137-6 harriss, r. c. & adams, j. a. (1966). geochemical and mineralogical studies on the weathering of granitic rocks. american journal of science, 264(2), 146-173. https://doi.org/10.2475/ajs.264.2.146 hernando massanet, m. i., barba carretero, a. m. & hernando costa, j. (2004). mineralogía de suelos de la sierra de guadarrama (madrid, españa). cuaternario y geomorfología, 18(1-2), 67-71. jayawardena, u. d. s. & izawa, e. (1994). a new chemical index of weathering for metamorphic silicate rocks in tropical regions: a study from sri lanka. engineering geology, 36(3-4), 303-310. https://doi.org/10.1016/00137952(94)90011-6 jiménez-espinosa, r., vázquez, m. & jiménez-millán, j. (2007). differential weathering of granitic stocks and landscape effects in a mediterranean climate, southern iberian massif (spain). catena, 70(2), 243-252. https:// doi.org/10.1016/j.catena.2006.09.001 johnson, n. m., likens, g. e., bormann, f. h. & pierce, r. s. (1968). rate of chemical weathering of silicate minerals in new hampshire. geochimica et cosmochimica acta, 32(5), 531-545. https://doi.org/10.1016/00167037(68)90044-6 klug, h. p. & alexander, l. e. (1974). x-ray diffraction procedures: for polycrystalline and amorphous materials. second edition, wiley-vch, u.s.a., 992 pp. li, s. l., chetelat, b., yue, f., zhao, z. & liu, c. q. (2014). chemical weathering processes in the yalong river draining the eastern tibetan plateau, china. journal of asian earth sciences, 88, 74-84. https://doi.org/10.1016/j. jseaes.2014.03.011 ling, s., wu, x., sun, c., liao, x., ren, y. & li, x. (2016). mineralogy and geochemistry of three weathered lower cambrian black shale profiles in northeast chongqing, china. geosciences journal, 20 (6), 793-812. https://doi.org/10.1007/s12303-016-0008-y ling, s., wu, x., zhao, s. & liao, x. (2018). evolution of porosity and clay mineralogy associated with chemical weathering of black shale: a case study of lower cambrian black shale in chongqing, china. journal of geochemical exploration,  188, 326-339. https://doi.org/10.1016/j.gexplo.2018.02.002 marcos, a., martínez-catalán, j. r., pérez-estaún, a., pulgar, j. a. & vera, j. a. (2004). características generales de la estructura de la zona asturoccidental-leonesa. in: vera, j. a. (ed). geología de españa, sge-igme, madrid, 54-55 pp. martínez-catalán, j. r., pérez-estaún, a., bastida, f., pulgar, j. a. & marcos, a. (1990) west asturian-leonese zone. structure. in: dallmeyer, e. & martínez-garcía (eds.). pre-mesozoic geology of iberia, r.d., springer-verlag, berlin, 103-114 pp. merriman, r. j., roberts, b. & peacor, d. r. (1990). a transmission electron microscope study of white mica crystallite size distribution in a mudstone to slate transitional sequence, north wales, uk. contributions to mineralogy and petrology, 106(1), 27-40. https://doi.org/10.1007/ bf00306406 meunier, a., sardini, p., robinet, j. c. & prêt, d. (2007). the petrography of weathering processes: facts and outlooks. clay minerals, 42(4), 415-435. https://doi.org/10.1180/claymin.2007.042.4.01 meunier, a., caner, l., hubert, f., el albani, a. & prêt, d. (2013). the weathering intensity scale (wis): an alternative approach of the chemical index of alteration (cia). american journal of science, 313(2), 113-143. https:// doi.org/10.2475/02.2013.03 molina, e., cantano, m., rodriguez, p. g. & vicente, m. a. (1990). some aspects of paleoweathering in the iberian hercynian massif. catena, 17(4-5), 333-346. https://doi.org/10.1016/0341-8162(90)90036-d molina, e., gonzález, m. g. & espejo, r. (1991). study of paleoweathering on the spanish hercynian basement montes de toledo (central spain). catena, 18(3-4), 345-354. https://doi.org/10.1016/0341-8162(91)90030-2 molina, e. & cantano m. (2002). study of weathering processes developed on old piedmont surfaces in western spain: new contributions to the interprehttps://doi.org/10.1016/j.earscirev.2019.103039 https://doi.org/10.1016/j.earscirev.2019.103039 https://doi.org/10.1180/claymin.2012.047.1.117 https://doi.org/10.1180/claymin.2012.047.1.117 https://doi.org/10.1086/518048 https://doi.org/10.1098/rsta.1984.0026 https://doi.org/10.5194/esurf-4-727-2016 https://doi.org/10.1016/s0016-7061(98)00061-5 https://doi.org/10.1016/s0016-7061(98)00061-5 https://doi.org/10.1016/s0016-7061(98)00061-5 https://doi.org/10.1130/0091-7613(1995)023%3c0921:uteopm%3e2.3.co;2 https://doi.org/10.1130/0091-7613(1995)023%3c0921:uteopm%3e2.3.co;2 https://doi.org/10.1016/j.catena.2009.08.005 https://doi.org/10.1016/j.catena.2009.08.005 https://doi.org/10.1016/j.chemer.2009.02.002 https://doi.org/10.1016/j.chemer.2009.02.002 https://doi.org/10.1180/claymin.1976.011.3.06 https://doi.org/10.1016/0009-2541(95)00089-5 https://doi.org/10.1016/s0013-7952(00)00071-5 https://doi.org/10.1016/0037-0738(88)90137-6 https://doi.org/10.1016/0037-0738(88)90137-6 https://doi.org/10.2475/ajs.264.2.146 https://doi.org/10.1016/0013-7952(94)90011-6 https://doi.org/10.1016/0013-7952(94)90011-6 https://doi.org/10.1016/j.catena.2006.09.001 https://doi.org/10.1016/j.catena.2006.09.001 https://doi.org/10.1016/0016-7037(68)90044-6 https://doi.org/10.1016/0016-7037(68)90044-6 https://doi.org/10.1016/j.jseaes.2014.03.011 https://doi.org/10.1016/j.jseaes.2014.03.011 https://doi.org/10.1007/s12303-016-0008-y https://doi.org/10.1007/s12303-016-0008-y https://doi.org/10.1016/j.gexplo.2018.02.002 https://doi.org/10.1016/j.gexplo.2018.02.002 https://doi.org/10.1007/bf00306406 https://doi.org/10.1007/bf00306406 https://doi.org/10.1180/claymin.2007.042.4.01 https://doi.org/10.2475/02.2013.03 https://doi.org/10.2475/02.2013.03 https://doi.org/10.1016/0341-8162(90)90036-d https://doi.org/10.1016/0341-8162(91)90030-2 6767mineralogy and geochemical signatures as indicators of differential weathering in natural soil profiles from the west asturian-leonese zone tation of the ‘‘raña’’ profiles. geomorphology, 42(3-4), 279-292. https:// doi.org/10.1016/s0169-555x(01)00091-5 moore, d. m. & reynolds, r. c. (1989). x-ray diffraction and the identification and analysis of clay minerals (vol. 322). oxford: oxford university press, 321 pp. munroe, j. s., farrugia, g. & ryan, p. c. (2007). parent material and chemical weathering in alpine soils on mt. mansfield, vermont, usa. catena, 70(1), 39-48. https://doi.org/10.1016/j.catena.2006.07.003 nesbitt, h. w., markovics, g. & price r. c. (1980). chemical processes affecting alkalis and alkaline earths during continental weathering. geochimica et cosmochimica acta, 44(11), 1659-1666. https://doi.org/10.1016/00167037(80)90218-5 nesbitt, h. w. & young, g. m. (1982). formation and diagenesis of weathering profiles. the journal of geology, 97(2), 129-147. https://doi. org/10.1086/629290 núñez, m. a. & recio, j. m. (2007). kaolinitic paleosols in the southwest of the iberian peninsula (sierra morena region, spain). paleoenvironmental implications. catena, 70(3), 388-395. https://doi.org/10.1016/j.catena.2006.11.004 papoulis, d., tsolis-katagas, p. & katagas, c. (2004). progressive stages in the formation of kaolin minerals of different morphologies in the weathering of plagioclase. clays and clay minerals, 52(3), 275-286. https://doi. org/10.1346/ccmn.2004.0520303 parker, a. (1970). an index of weathering for silicate rocks. geological magazine, 107, 501–504. https://doi.org/10.1017/s0016756800058581 pédro, g. (1997). clay minerals in weathered rock materials and in soils. in: soils and sediments. springer, berlin, heidelberg, 1-18pp. https://doi. org/10.1007/978-3-642-60525-3_1 pérez-estaún, a., bastida, f., martínez-catalán, j. r., gutiérrez marco, j. c., marcos, a. & pulgar, j. a. (1990). the west asturian-leonese zone. stratigraphy. in: dallmeyer, r. d. & martínez-garcía, e. (editors). pre-mesozoic geology of iberia, springer-verlag, berlin, 92-102. https://doi. org/10.1007/978-3-642-83980-1_3 pérez-estaún, a., bea, f., bastida, f., marcos, a., martínez-catalán, j. r., arenas, r., díaz garcía, f., azor, a., simancas, j. f. & gonzález lodeiro, f. (2004). la cordillera varisca europea: el macizo ibérico. in: vera, j. a. (editor). geología de españa, sge-igme, madrid, 17-25. perri, f. (2018). reconstructing chemical weathering during the lower mesozoic in the western-central mediterranean area: a review of geochemical proxies. geological magazine, 155(4), 944-954. https://doi.org/10.1017/ s0016756816001205 perri, f. (2020). chemical weathering of crystalline rocks in contrasting climatic conditions using geochemical proxies: an overview. palaeogeography, palaeoclimatology, palaeoecology, 556, 109873. https://doi.org/10.1016/j. palaeo.2020.109873 perri, f., scarciglia, f., apollaro, c. & marini, l. (2015). characterization of granitoid profiles in the sila massif (calabria, southern italy) and reconstruction of weathering processes by mineralogy, chemistry, and reaction path modeling. journal of soils and sediments, 15(6), 1351-1372. https://doi.org/10.1007/s11368-014-0856-x price, j. r. & velbel, m. a. (2003). chemical weathering indices applied to weathering profiles developed on heterogeneous felsic metamorphic parent rocks. chemical geology, 202(3-4), 397-416. https://doi.org/10.1016/j. chemgeo.2002.11.001 reynolds, r. c. (1980). interstratified clay minerals. in: brindley, g. w. and brown, g. (editors). crystal structures of clay minerals and their x-ray identification, monograph 5, mineralogical society, london, 49–303 pp. https://doi.org/10.1180/mono-5.4 robert, m. & tessier, d. (1992). incipient weathering: some new concepts on weathering, clay formation and organization. in: martini, v. & chesworth, w. (editors). developments in earth surface processes (vol. 2). elsevier, 71-105 pp. https://doi.org/10.1016/b978-0-444-89198-3.50009-x robertson, i. d. & eggleton, r. a. (1991). weathering of granitic muscovite to kaolinite and halloysite and of plagioclase-derived kaolinite to halloysite. clays and clay minerals, 39(2), 113-126. https://doi.org/10.1346/ ccmn.1991.0390201 rodríguez fernández, l. r. (1982). mapa geológico nacional (magna), serie 2ª, primera edición, eccala 1:50.000 hoja nº127 (noceda). instituto geológico y minero de españa (igme), madrid. romero, r., robert, m., elsass, f. & garcía, c. (1992). abundance of halloysite neoformation in soils developed from crystalline rocks. contribution of transmission electron microscopy. clay minerals, 27(1), 35-46. https:// doi.org/10.1180/claymin.1992.027.1.04 ruxton, b. p. (1968). measures of the degree of chemical weathering of rocks. the journal of geology, 76(5), 518-527. https://doi.org/10.1086/627357 scarciglia, f., critelli, s., borrelli, l., coniglio, s., muto, f. & perri, f. (2016). weathering profiles in granitoid rocks of the sila massif uplands, calabria, southern italy: new insights into their formation processes and rates. sedimentary geology, 336, 46-67. https://doi.org/10.1016/j.sedgeo.2016.01.015 senol, h., tunçay, t. & dengiz, o. (2016). geochemical mass-balance, weathering and evolution of soils formed on a quaternary-age basaltic toposequences. solid earth discuss (preprint), 105, 1-22. https://doi. org/10.5194/se-2016-105 spears, d. (2016). clay mineralogy of onshore uk carboniferous mudrocks. clay minerals, 41(1), 395-416. https://doi.org/10.1180/0009855064110201 suárez, o., corretgé, l. g. & martínez, f. j. (1990). the west asturian-leonese zone. distribution and characteristics of the hercynian metamorphism. in: dallmeyer, r. d. & martínez-garcía, e. (editors). pre-mesozoic geology of iberia, springer-verlag, berlín, 129-133. taboada, t. & garcia, c. (1999). pseudomorphic transformation of plagioclases during the weathering of granitic rocks in galicia (nw spain). catena, 35 (2-4), 291-302. https://doi.org/10.1016/s0341-8162(98)00108-8 tardy, y., bocquier, g., paquet, h. & millot, g. (1973). formation of clay from granite and its distribution in relation to climate and topography. geoderma, 10, 271–284. https://doi.org/10.1016/0016-7061(73)90002-5 vicente, m. a., molina, e. & espejo, r. (1991). clays in paleoweathering processes: study of a typical weathering profile in the hercynian basement in the montes de toledo (spain). clay minerals, 26(1), 81-90. https://doi. org/10.1180/claymin.1991.026.1.08 vicente, m. a., elsass, f., molina, e. & robert, m. (1997). palaeoweathering in slates from the iberian hercynian massif (spain): investigation by tem of clay mineral signatures. clay minerals, 32(3), 435-451. https://doi. org/10.1180/claymin.1997.032.3.06 wagner, r. h. (2004). the iberian massif: a carboniferous assembly. journal of iberian geology, 30, 93–108. white, a. f. & brantley, s. l. (2003). the effect of time on the weathering of silicate minerals: why do weathering rates differ in the laboratory and field? chemical geology, 202(3-4), 479-506. https://doi.org/10.1016/j. chemgeo.2003.03.001 https://doi.org/10.1016/s0169-555x(01)00091-5 https://doi.org/10.1016/s0169-555x(01)00091-5 https://doi.org/10.1016/j.catena.2006.07.003 https://doi.org/10.1016/0016-7037(80)90218-5 https://doi.org/10.1016/0016-7037(80)90218-5 https://doi.org/10.1086/629290 https://doi.org/10.1086/629290 https://doi.org/10.1016/j.catena.2006.11.004 https://doi.org/10.1016/j.catena.2006.11.004 https://doi.org/10.1346/ccmn.2004.0520303 https://doi.org/10.1346/ccmn.2004.0520303 https://doi.org/10.1017/s0016756800058581 https://doi.org/10.1007/978-3-642-60525-3_1 https://doi.org/10.1007/978-3-642-60525-3_1 https://doi.org/10.1007/978-3-642-83980-1_3 https://doi.org/10.1007/978-3-642-83980-1_3 https://doi.org/10.1017/s0016756816001205 https://doi.org/10.1017/s0016756816001205 https://doi.org/10.1016/j.palaeo.2020.109873 https://doi.org/10.1016/j.palaeo.2020.109873 https://doi.org/10.1007/s11368-014-0856-x https://doi.org/10.1016/j.chemgeo.2002.11.001 https://doi.org/10.1016/j.chemgeo.2002.11.001 https://doi.org/10.1180/mono-5.4 https://doi.org/10.1016/b978-0-444-89198-3.50009-x https://doi.org/10.1346/ccmn.1991.0390201 https://doi.org/10.1346/ccmn.1991.0390201 https://doi.org/10.1180/claymin.1992.027.1.04 https://doi.org/10.1180/claymin.1992.027.1.04 https://doi.org/10.1086/627357 https://doi.org/10.1016/j.sedgeo.2016.01.015 https://doi.org/10.1016/j.sedgeo.2016.01.015 https://doi.org/10.5194/se-2016-105 https://doi.org/10.5194/se-2016-105 https://doi.org/10.1180/0009855064110201 https://doi.org/10.1016/s0341-8162(98)00108-8 https://doi.org/10.1016/0016-7061(73)90002-5 https://doi.org/10.1180/claymin.1991.026.1.08 https://doi.org/10.1180/claymin.1991.026.1.08 https://doi.org/10.1180/claymin.1997.032.3.06 https://doi.org/10.1180/claymin.1997.032.3.06 https://doi.org/10.1016/j.chemgeo.2003.03.001 https://doi.org/10.1016/j.chemgeo.2003.03.001 white, a. f. & buss, h. l. (2014). natural weathering rates of silicate minerals. treatise on geochemistry, second edition. elsevier, oxford, 115-155. white, a. f., bullen, t. d., schulz, m. s., blum, a. e., huntington, t. g. & peters, n. e. (2001). differential rates of feldspar weathering in granitic regoliths. geochimica et cosmochimica acta, 65(6), 847-869. https://doi. org/10.1016/s0016-7037(00)00577-9 wilson, m. j. (2004). weathering of the primary rock-forming minerals: processes, products and rates. clay minerals, 39(3), 233-266. https://doi. org/10.1180/0009855043930133 young, g. m. & nesbitt, h. w. (1998). processes controlling the distribution of ti and al in weathering profiles, siliciclastic sediments and sedimentary rocks. journal of sedimentary research, 68(3), 448-455. https://doi. org/10.2110/jsr.68.448 https://doi.org/10.1016/s0016-7037(00)00577-9 https://doi.org/10.1016/s0016-7037(00)00577-9 https://doi.org/10.1180/0009855043930133 https://doi.org/10.1180/0009855043930133 https://doi.org/10.2110/jsr.68.448 https://doi.org/10.2110/jsr.68.448 archivo04.indd 35 manuscript received april 6, 2006. oil decontamination of bottom sediments experimental work results sergey v. lushnikov1, yulia a. frank1 and danil s. vorobyov2 1 research and technical association “priborservice” ltd., komsomolsky av., 70, tomsk, russia, 634041 2 tomsk state university, lenina av., 36, tomsk, russia, 634050 corresponding author: sergey v. lushnikov, e-mail: pribor@mail.tomsknet.ru abstract this article presents the results of experimental work during 2004-2005 on oil decontamination of bottom sediments of lake schuchye, situated in the komi republic (northern russia). the cause of the contamination were huge oil spills occurred after a series of accidental ruptures on the harjaga-usinsk and vozej-usinsk oil-pipe lines in 1994. flotation technology was used for the cleaning of bottom sediments. 157 tons of crude oil were removed during the course of 2-year experimental work from an area of 4,1 ha. the content of aliphatic and alicyclic oil hydrocarbons was reduced from 53,3 g/kg to 2,2 g/kg, on average. hydrobiological investigations revealed that bottom sediments started to be inhabited by benthos organisms, dominantly oligochaeta. besides oligochaeta, chironomidae maggots and bivalvia were detected. the appearance of macrozoobenthos organisms can serve as a bioindicator of water quality. key words: bioindication, bottom sediments, macrozoobenthos, oil contamination and decontamination of water ecosystems. resumen este artículo presenta los resultados del trabajo experimental realizado entre 2004-2005 en la descontaminación de petróleo de los sedimentos de fondo del lago schuchye en la república de komi (norte de rusia). la causa de la contaminación del lago fueron enormes derrames de petróleo ocurridos después de una serie de ruptures accidentales en los oleoductos de harjaga-usinsk y vosey-usink en 1994. la tecnología de flotación fue usada para limpiar los sedimentos de fondo. en el transcurso de los dos años de trabajo experimental, fueron removidas 157 toneladas de petróleo crudo en un area de 4,1 ha. el contenido de hidrocarburos alifáticos y alicíclicos se redujo en promedio de 53,3 g/kg a 2,2 g/kg. las investigaciones hidrobiológicas mostraron que los sedimentos de fondo empezaron a ser habitados por organismos bentónicos, predominantemente por oligohetas (oligohaeta). también fueron detectadas larvas de chironomides (chironomidae), y bivalvos (bivalvia). la aparición de organismos macrozoobenthos puede servir como un bioindicator de la calidad del agua. palabras claves: sedimentos de fondo, contaminación y descontaminación petrolífera de ecosistemas acuáticos, bioindicación, macrozoobenthos. earth sciences research journal earth sci. res. j. vol. 10, no. 1 (jun. 2006): 35-40 36 sergey v. lushnikov, yulia a. frank and danil s. vorobyov introduction professional ecologists are fully aware of the harmful impact the development of oil fields has on the environment. the monitoring research of these technogenic territories almost always reveals a negative effect on environment (laffon, et al., 2006; paasvirta, et al., 1982). serious accidents lead to irreparable damage to natural ecosystems, and mitigation of effects of pollution may last for decades (laffon, et al., 2006). in regions with lakes and rivers the aquatic ecosystems are heavily impacted by oil spills. oil and oil products are most intensively washed down into the river network during the spring tide when runoff is formed at all geomorphologic levels and in summer at the heaviest rain period (kondratyeva, 2000; zhadin & gerd, 1961). liquidation of oil spills on a water surface is carried out by booming, by means of different types of skimmers, dispersants and sorbents. additionally bacterial isolates are also used for such purposes (gentili, et al., 2006; mukherji, et al., 2004; reed, et al., 2004). if measures for localization and gathering the pollutant are not taken at the proper time, it may lead to oil and oil products sorption by suspended organic and mineral particles and aquatic flora, and in the long run to their sinking to the bottom of the reservoir, which produces a negative effect on bottom organisms associations causing their total degradation (ke, et al., 2005; reed, et al., 2004). oxygen deficiency, low photo-oxidation indices, low water temperature prevents the destruction of the pollutant. under anaerobic conditions, organic substances undergo fermentation but not oxidation, and sulfate reduction and methane formation take place (kondratyeva, 2000). at the same time toxic substances accumulate in water (petrov, 1978; vinogradov, 2002). at choosing technological schemes for cleaning of bottom sediments in lake schuchye several variants were taken into consideration. the first one is the use of pumping units for removing contaminated soil. pumping-out of bottom sediments contaminated with oil requires adding some technological stages for cleaning the contaminated soil which considerably complicates the decontamination work and increases its cost. in lake schuchye, the oil accumulated on the surface of bottom sediments in the form of layers with aggregates of different sizes. on the off thicknesses upper layer of bottom sediments where the “soil-oil” contact was contaminated. there is no doubt that if oil was removed from the upper contaminated layer, a large amount of pollutionless sediments would be pumped out, acting as a sorbent for oil while moving along the pipeline. removing the sediments layer with a thickness of 10 cm (of lake schuchye, with an area of 6,26 ha.) would result in more than 6000 m3 of contaminated soil, which should undergo further decontamination. therefore, application of this technological scheme was considered inexpedient. the second variant is more acceptable: hydropneumatic decontamination of bottom sediments, based on the ability of molecular adhesion of oil to the phase at the interface between air and water (flotation). the given scheme allows removing the oil from the surface of bottom sediments. determination of the necessary decontamination level of bottom sediments of the reservoir, after which mechanisms of natural self-purification start actively working, is very important for a wide application of flotation technology to oil decontamination of water bodies. methodology lake schuchye no.1 (figure 1), where experimental work on oil decontamination of bottom sediments was carried out, is a thermokarstic slowflow lake with an area of 6,26 ha. and a maximum depth of 7 m. it is a part of a system of 4 lakes (figure 1) and situated in the komi republic (northern russia). monthly, samples of water and bottom sediments were taken at monitoring stations and samples of bottom invertebrates were taken every ten days. surface water sampling was carried out according to the general technical conditions, (2002) (gost 17.1.5.04-81) from the depth of 0,4 m up. bottom sediments sampling was carried out by means of a petersen bottom sampler with a grip area of 1/80 m2. after lifting of the bottom sampler containing bottom sediments, water was poured out; bottom sediments were put into a polyethylene bag and labeled. for collection of zoobenthos samples, the petersen bottom sampler was also used. to eliminate probable errors resulting from hydrobiontes distribution in homogeneity each zoobenthos sample consisted of a set of three or four bottom sediments samples taken separately. 37 oil decontamination of bottom sediments experimental work results figure 1. map of russia, (top) showing the location of the komi republic and a map of the schuchye lake system (lakes no. 1-4). the lake 4, (1 ha.) by means of a flotation technology. in 2005 the equipment was improved, which made possible to reduce the oil content in bottom sediments and during the summer period of 2005, 92 tons of crude oil were extracted. during 2 seasons of cleaning work in 2004-2005, only the hydropneumatic cleaning scheme was used. the content of aliphatic and alicyclic oil hydrocarbons was reduced from 53,3±7,46 g/kg to 2,2±0,39 g/kg on average (figure 2a). reduction of oil content in the lake water mass during cleaning operations was not so significant (figure 2b). hydrobiological research carried out in the lake in 2005 showed that after two years of integrated decontamination work in the lake zones where the content of oil in soil did not exceed 3,3 g/kg, bottom sediments contained benthonic organisms. three groups of macrozoobenthos were found: there types of oligochaeta, chironomidae maggots, and bivalvia. the number of benthos came up to 2320 specimens per m2, and measured biomass was as high as 3,940 g/m2. oligochaeta were the basic benthos because of their quantity (90-100%) and biomass (80-100%) in lake schuchye. the samples were washed in a rinse bag made of fine gauze no. 28 to remove fine fractions of silt. residual contents were examined with the help of a magnifying glass, and zoobenthos organisms were taken out with tweezers, placed into glass containers with 70% ethyl alcohol and labeled (morduhaiboltovsky, 1975). oil contents in the water mass and bottom sediments were determined by means of an infrared spectrophotometrical method. the method allows the determination of aliphatic and alicyclic oil carbohydrates and is based on the extraction of oil by ccl 4, chromatographic fractionation with aluminum oxide and on quantitative determination in the infrared region using the «specord m 80». oil decontamination of bottom sediments and water was carried out according to the technology developed and patented by the authors of this paper (lushnikov & vorobyov, 2005). results in 2004, as much as 65 tons of oil were extracted from bottom sediments of the experimental zone of figure 2. oil content in bottom sediments (a) and in water mass (b) before remediation (june 2004) and after remediation (june 2005 and september 2005). ( ) ( ) 38 sergey v. lushnikov, yulia a. frank and danil s. vorobyov discussion in water reservoirs and waterways oil spreads mostly under the influence of water currents and wind. the currents of water are the basic factor of oil spread in waterways. relative proximity of the waterside and a winding river-bed allows oil to reach waterways quickly. the best conditions for oil accumulation exist in backwaters, creeks, and sections of rivers where there is no current. in these sections the content of oil and oil products in bottom sediments is high in comparison with that of riverbeds. sedimentation of contaminated suspended particles and their accumulation in soil exert a negative influence on bottom associations of organisms (rusanova & vorobyov, 2001). in lakes oil spreads on the water surface under the influence of wind. irregularity of the waterside leads to formation of zones with conditions favorable for oil accumulation. removing such oil patches presents no difficulties if special oil-gathering equipment is used, although the solution to this problem is versatile. at first, it is necessary to locate the oil patch. furthermore, the decontamination technology depends on availability of access roads to the reservoir and the need for transporting localized oil on water surface; in the case of large reservoirs the use of specialized oil-spill boats is optimal. gathering oil from the water surface and cleaning the waterside of a reservoir is complicated by the presence of higher aquatic vegetation. to a greater extent this refers to water-resistant plants which can be found in the littoral zone, on floating bogs, on swampy and highly moist shores, and to aquatic plants with aerial generative organs. after water plants die, oil kept by the vegetation on the surface of water accumulates in the bottom sediments of the waterside where its destruction slows down. therefore, contaminated vegetation is removed from the reservoir along with oil. figure 3 shows a part of the shore of lake schuchye shore with oil contaminated aquatic vegetation before decontamination and with recovered phytocenosis after removing oil and parts of the plants rising to the water surface. to avoid secondary pollution of decontaminated areas near the waterside it is necessary to use barriers to prevent further spread of oil. as a rule, secondary pollution occurs, in reservoirs where bottom sediments are highly contaminated with oil. in summer, when the temperature in the bottom layers of water rises, in shallow zones there occurs spontaneous raise of oval and globular-shaped oil aggregates to the surface. this phenomenon took place in the zones of the lake with a depth up to 2 m. in 2004 in the experimental zone of lake schuchye more than 12 tons of oil rose to the water surface. figure 3. the waterside of schuchye lake no.1 before (top) and after decontamination (bottom) (photo by d.s.vorobyov). fertilizer application, the use of aerators in the water area of the reservoir, and optimal temperature conditions in july 2004, the temperature in the bottom layers of water at a depth of 5 m reached 19-20 oc, were favorable to the intensive raise of oil. to reduce the content of oil absorbed by sediments during the second step of cleaning operations we activated microbiological destruction processes. the content of nutrients and oxygen in water is the key factor in the processes of microbiological decomposition, thus, when carrying out work in the lake we used aerators, set up active booms on the surface of water, and applied dosified amounts of fertilizers. the flotation technology applied allowed us to remove oil (in the form of a continuous layer and aggregates) from the surface of bottom sediments of the lake. during the cleaning up work period in 39 oil decontamination of bottom sediments experimental work results lake schuchye in 2004-2005 as much as 157 tons of oil were extracted from the bottom sediments. however, a large amount of oil is kept in the sediments due to the high sorption capacity of fine soils. it is well known that there is a connection between the granulometric composition of bottom sediments and the amount of oil absorbed by them – fine sediments possess higher sorption surface, hence they can retain more oil; 1 gram of bottom sediments under optimal conditions absorbs up to 12-25 mg of oil products (zyulkina, 1975), (i.e. the concentration of this pollutant can run up to 25 g/kg). hydrobiological research carried out in the lake in 2005 showed that after two years of integrated decontamination work in the lake zones where the content of oil in soil did not exceed 3,3 g/kg, bottom sediments contained benthonic organisms (in 20032004 benthonic organisms were not found there). the number of benthos measure up to 2320 specimens per m2, and the biomass was as much as 3,940 g/m2. many characteristics of zoobenthos (species diversity, quantity, biomass etc.) depend on the physical properties of the soil and sediments quality, and the amount of easily assimilated organic substance present in it transformed by bacteria. (vinogradov, et al., 2002) noted that the quality criteria of bottom sediments should involve parameters such as the level of quantitative development of zoobenthos. the three groups of macrozoobenthos in the lake schuchye correspond to a benthos group which can endure oil pollution best of all. oligochaeta feed from organic substances in sediments and play an important role in their formation. during a 24-hour period, a worm digests an amount of detritus which weighs 4 times as much as the worm itself (zhadin & gerd, 1961).). experiments carried out on another group of hydrobiontes, polychaeta, showed that aromatic structures of organic substances of excrements acquired a different character from that in initial forms. these hydrobiontes transformed oil products, namely, tailing (georga-kopulus & alemov, 1990).). oligochaeta also take part in the process of transformation of oil-contaminated soil. and although they do not play the main role in the self-purification processes (as compared with microbial associations), the whole activity of the complex biosystem of a reservoir provides a rapid recovery. conclusions the flotation technology allowed effective oil decontamination of the bottom sediments of schuchye lake situated in komi republic (northern part of russia). during the period of experimental cleaning work (2004-2005) a total of 157 tons of crude oil were removed. a significant amount of oil was kept by sediments because of the high sorptive capacity of fine soils. it was shown that the aliphatic and alicyclic oil hydrocarbons content in bottom sediments reduced in the course of the experiment from 53, 3±7, 46 g/kg to 2, 2±0, 39 g/kg. after a 2-year integrated decontamination work in the lake zones where the content of oil in soil did not exceed 3,3 g/kg bottom, sediments started to be inhabited by benthonic organisms. three groups of macrozoobenthos have been found oligochaeta, chironomidae maggots and bivalvia. oligochaeta were the basic benthos according to their quantity (90-100%) and biomass (80-100%). the total number of benthos came up to 2320 specimen per m2, and the biomass totaled 3,940 g/m2. the fact that macrobenthos organisms appear in the bottom sediments of lake schuchye is a strong bioindicative proof of an improving in the habitat quality of bottom organisms associations. currently we initiated the task of working out integral criteria for a quantitative assessment of recovery in water bodies using hydrochemical and hydrobiological tools. to solve this problem, monitoring research of the lake and research of other water bodies undergoing decontamination in different climatic zones will be continued. acknowledgements the project was put into practice by the research and technical association “priborservice” ltd. supported by “lukoil-komi” ltd., federal state department “komirybvod”, the department of water industry of komi republic and usinsk regional environmental protection committee. we also thank coworkers from department of ichthyology and hydrobiology of tomsk state university. references • gentili a.r., m.a., cubitto, m. ferrero and m.s. rodríguez (2006). bioremediation of crude oil polluted seawater by a hydrocarbon-degrading bacterial strain immobilized on chitin and chitosan flakes, international 40 sergey v. lushnikov, yulia a. frank and danil s. vorobyov biodeterioration and biodegradation. 57, no. 4, 222228. • georga-kopulus l.a., and s.v. alemov (1990). the role of nereis sp. in transforming oil tars in marine bottom sediments, hydrobiological journal. 26, no. 2, 60-64. (in russian). • ke, l., k.s.h. yu, y.s. wong, and n.f.y. tam (2005). spatial and vertical distribution of polycyclic aromatic hydrocarbons in mangrove sediments, science of the total environment. 340, no. 1-3, 177-187. • kondratyeva l.m. (2000). secondary pollution of aquatic ecosystems, water source. 27, no. 2, 221-231. (in russian). • laffon b., t. rábade, e. passaro, and j. méndez (2006). monitoring of the impact of prestige oil spill on mytilus galloprovincials from galician coast, environment international. 32, no. 3, 342-348. • lushnikov, s.v., and d.s vorobyov (2005). patent for the method for cleaning up water of reservoirs and bottom sediments from oil and oil products. no. 2260652, subm. 05.05.2004; publ. 20.09.2005, bull. no. 26. (in russian). • mukherji s., sh. jagadevan, g. mohapatra, and a. vijay (2004). biodegradation of diesel oil by an arabian sea sediment culture isolated from the vicinity of an oil field, bioresource technology. 95, no. 3, 281-286. • nature protection. hydrosphere. apparatus and mechanisms for selection, initial treatment and stering samples of natural waters (2002). general technical conditions (gost 17.1.5.04-81). (in russian). • paasvirta j., h. käärialäinen, m. lahtiperä, j. pellinen, and s. sinkkonen (1982). oil residues in baltic sediment, mussel and fish. study of the finnish archipelago 1979-1981, chemosphere. 11, no. 8, 811-821. • petrov, g.n. (1978). some physical processes of selfpurification of water from oil, hydrobiological journal. 14, no. 4, 53-54. (in russian). • reed, m., p. daling, a. lewis, m.k. ditlevsen., b. brors, j. clark, and d. aurand (2004). modelling of dispersant application to oil spills in shallow coastal waters, environmental modelling and software. 19, no. 7-8, 681-690. • rusanova a.i., and d.s. vorobyov (2001). peculiarities of bottom associations development in floodplain reservoirs of the river vasyugan, 8th congress of russian hydrobiological association russian academy of science, kaliningrad, 180-182. (in russian). morduhai-boltovsky, f.d. (1975). the technique of studying ecosystems of inland water reservoirs, “nauka” publishing house, moscow, 240 pp. (in russian). • vinogradov g.a., n.a. berezina, n.a. lapteva, and g.p. zharikov (2002). the use of structural characteristics of bacterioand zoobenthos for quality rating of bottom sediments (by the example of reservoirs of the upper volga basin), water source. 29, no. 3, 329-336 (in russian). • zhadin v.i., and c.v. gerd (1961). rivers, lakes and storage pools of the ussr: their fauna and flora, publishing house of the ministry of education of rsfsr, moscow, 600 pp. (in russian). • zyulkina v.y. (1975). the importance of suspended sediments and bottom sediments in self-purification of the river selenga from oil products, hydrobiological journal. 3, 63-65 (in russian). this work analyses soil loss over the last 150 years due to urban growth in one of the boroughs of highest environmental value around madrid (spain), i.e. colmenar viejo. the extent and patterns followed by urban extension are examined retrospectively using a present cartographic base layer. this evolutionary approach considering space and time has required a thorough compilation of cartographic sources that were created using classic topography, photogrammetry or remote sensing, some of which have been rarely used so far. these documents have been georeferenced, edited and handled using geographic information systems (gis) and remote sensing programs. urban growth in colmenar viejo has been steady, has registered exponential increases (soil sealing affected 0.14% of the total municipal territory in 1861 and covers 2.64% nowadays) and has followed similar patterns to other spanish towns influenced by metropolitan processes. abstract keywords: soil sealing; cartography; topography; photogrammetry; remote sensing; geographic information systems (gis). soil sealing in madrid (spain), study case of colmenar viejo issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n3.51450 andrés aristegui cortijo1*, maría eugenia pérez gonzález2 1*instituto nacional geográfico de españa. 2 universidad complutense de madrid earth sciences research journal earth sci. res. j. vol. 21, no. 3 (september, 2017): 111-116 este trabajo analiza la pérdida de suelo en los últimos 150 años debido al crecimiento urbano en uno de los municipios de mayor valor ambiental de madrid (españa), colmenar viejo. la extensión y los patrones seguidos por la expansión urbana se examinan retrospectivamente usando una capa base cartográfica actual. este enfoque evolutivo que considera el espacio y el tiempo, ha requerido una recopilación completa de fuentes cartográficas, que fueron creadas usando la topografía clásica, la fotogrametría o la teledetección, algunas de las cuales han sido raramente utilizadas hasta ahora. estos documentos han sido georeferenciados, editados y manipulados utilizando sistemas de información geográfica (gis) y programas de teledetección. el crecimiento urbano en colmenar viejo ha sido estable, ha registrado aumentos exponenciales (el sellado del suelo afectó al 0,14% del territorio municipal total en 1861 y cubre el 2,64% en la actualidad), y ha seguido patrones similares a otros pueblos españoles influenciados por procesos metropolitanos. resumen palabras clave: sellado del suelo; cartografía; topografía; fotogrametría; teledetección; sistemas de información geográfica. record manuscript received: 23/06/2015 accepted for publication: 30/06/2017 how to cite item aristegui, c. a., & perez, g. m. e. (2017). soil sealing in madrid (spain), study case of colmenar viejo. earth sciences research journal, 21(3), 111-116. doi: http://dx.doi.org/10.15446/esrj.v21n3.51450 g e o l o g ic a l e n g in e e r in g sellado de suelo en madrid (españa). el caso de estudio de colmenar viejo 112 andrés aristegui cortijo, maría eugenia pérez gonzález introduction major demographic changes have taken place in europe since the 19th century. this social evolution has resulted on a relevant urban growth and on significant transformations on land use and land cover (meirich, 2008). these variations can be approached from several disciplines since they are fostered by complex processes. however, in this specific study these changes have been analysed from a cartographic and geographic perspective. soil sealing is one of the main consequences of urban growth. this is a matter of grave concern as it entails ground waterproofing and leads to almost irreversible damages on soil ecosystems (nizeyimana, et al, 2001; montanarella, 2007; scalenghe & ajmone-marsan, 2009; eea, 2012; siedentop & fina, 2012). spatio-temporal dynamics of soil sealing have been regularly studied with the use of 20th century maps, aerial photographs, orthophotographs and satellite images (ridd, 1995; zhang, 2001; wu, 2004; moeller, 2005; kampouraki et al., 2006, etc.). however, the use of information predating the mid-20th century is rather unusual since only a few organisations have digitised ancient documents on a large scale and made them accessible to the general public. instead, this study has included 19th century maps provided by the national geographic institute of spain (ign). they have been used to analyse patterns of urban growth and the extent to which land has been lost to sealing over the last 150 years in a borough located in the outskirts of madrid. this is a rather innovative approach that enables this retrospective analysis to be extended in time an additional century to what is usually considered. the effects of soil sealing resulting from urban growth are intensely felt in spain on coastal regions (ojeda & villar, 2006; valera et al., 2011a) and on major metropolitan areas, like madrid (garcía & pérez, 2007, 2011 and 2016; garcía et al., 2014). this study has focused on a specific town called colmenar viejo, which is located in the northern outskirts of madrid (40º 39’ 31’’ n, 3º 45’ 58’’ w). this borough has been chosen because it has got both common characteristics to other spanish towns affected by metropolitan processes, but also distinctive attributes which are only to be observed in this territory, the most outstanding being: an environmentally significant and well preserved extension (18,256 ha). proximity to the city of madrid (30 km) and to several royal estates, i.e. el pardo y el soto de viñuelas, la granja de san ildefonso, san lorenzo de el escorial and aranjuez (fig. 1). colmenar viejo is well connected to the main city and to the surrounding towns by both road and rail. an urban character right from the beginning of the period under study as colmenar viejo has been a medium-size town since the 15th century. nevertheless, it is still nowadays physically apart from other built-up areas within the metropolitan area of madrid. this eases its separate study. various attempts to accomplish a scientific map of spain based on a geodetic network were carried out in the surroundings of colmenar viejo since the 1860´s. therefore, many maps of this area from the mid and late 19th century can be found at the national geographic institute of spain (ign), which is the only organisation that stores this type of scientific ancient cartography. this large amount of available ancient maps is most uncommon in spain, where some territories had not been scientifically surveyed even a century later. one of the main disadvantages of analysing soil sealing is that it is rarely confined to one administrative unit (borough, province, region or country) (seifert, 2010). as a result, analysis is more about how it has evolved in a specific territory than about comparing its evolution with other areas. instead, this study has aimed to show the spatio-temporal results of sealing at a municipal level so that it can be compared with other boroughs or territorial units. all information has been integrated in a gis so it can be measured and updated. figure 1. situation of colmenar viejo within the province of madrid and within spain sources a large variety of sources of information has been used to measure and analyse soil sealing. this variability results from the different techniques used for mapping over the last 150 years, i.e. classic topography (18601960 approx), photogrammetry (1960-2005 approx) and remote sensing (2005-2017 approx). this enables the use and integration of various cartographic techniques and its use in combination with gis programs, what makes this study rather innovative. in the first place, a careful selection of 365 ancient maps of colmenar viejo has been provided by the topographic archive at the national geographic institute of spain (ign). from 1860, when the spanish geodetic network started being calculated, to 1960, when photogrammetry started being widely used, several mapping projects using classic topographic methods were carried out in colmenar viejo. the material gathered for this lengthy period ranges from highly detailed land registry records at a scale of 1:500, to sheets from the national topographic map at a scale of 1:50,000. the characteristics of these old maps are summarised in table 1 and some examples of this ancient cartography are shown in fig. 2. should anyone be interested in going in depth in these sources, see muro et al., 1996; arístegui, 2014; urteaga & nadal, 2001; moreno, 2008. table 1. ancient maps of colmenar viejo created using classic topography [period 1860-1960] (source: national geographic institute of spain (ign)) 113soil sealing in madrid (spain), study case of colmenar viejo figure 2. examples of ancient maps of colmenar viejo created using classic topography [period 1860-1960] (source: national geographic institute of spain (ign)) in the second place, the period from 1960 to 2005 has been analysed by means of visually interpreting some aerial photographs. these images have been chosen from various photogrammetric flights, one per decade. they were captured at different scales, varying from 1:18,000 to 1:40,000, on analogue or digital formats, and on a colour or on a black & white scale, as shown on table 2. table 2. aerial photographs [period 1960-2005] (source: national geographic institute of spain (ign)) figure 3. base layer: each polygon represents an urban land plot within the builtup area as it stood in the year 2013 (source: spanish cadastral service (catastro)) finally, the different population censuses carried out in spain since the mid-19th century have been acquired at the national institute for statistics of spain (ine) (table 3). table 3. population registered in colmenar viejo on various censuses (source: national institute for statistics of spain (ine)) in the third place, the most recent period has been examined by means of remote sensing images from the year 2011. a panchromatic image (with a spatial resolution of 2.5 m) and a multispectral image (with a spatial resolution of 10 m) from spot 5 satellite (path 69, row 033) were analysed. these images have been supplied by the national geographic institute of spain (ign) over the national remote sensing plan (pnt). in the fourth place, some auxiliary sources have also been used. on the hand, an orthophoto from the year 2013 has been selected. it has been provided by the national geographic institute of spain (ign) over the aerial orthophoto national plan (pnoa). this layer has been used for the first step of the method, which has consisted of georeferencing all maps. on the other hand, an up-todate cadastral map of the built-up area has been chosen. this map has been provided by the spanish cadastral service. it is in a polygonal vector format (.shp) and represents urban land plots in the year 2013 (fig. 3). this map has been used as a base layer for the second stage of the method, which has consisted of editing this layer to help calculating soil loss by means of sealing. 114 andrés aristegui cortijo, maría eugenia pérez gonzález method the first step of the method has included changing all maps to a raster format and georeferencing them in datum etrs89, projection utm, zone 30 t, which is the official reference system in central spain at present. this process has been carried out with global mapper 14 gis program, using the 2013 orthophoto as a base layer. between 30 and 50 control points have been assigned to each map. these control points have been located in places that seem to have remained invariable over the last decades. afterwards a rubber-sheeting conversion with triangulation has been used and rotations, transformations and changes in local and directional scales have been applied. mean rms at control points has been 3.4 m. all control points with a rms above 6 m have not been taken into account. finally, georeferenced maps have been saved in .ecw format. the second stage of the method has included part of the method proposed by valera et al. (2011 b). first, the urban area has been analysed using maps dating from 1861, 1918 and 1952, photos from 1968, 1979, 1983, 1991 and 2002, and satellite images from 2011. second, the 2013 cadastral base layer, downloaded on global mapper 14 gis application, has been progressively edited in inverse chronological order by means of visually interpreting maps, photos and satellite images for each date. as the process has gone back in time, plot erasing has always been needed (it has never been necessary to add them) since the built-up area has always shown an increase with time, never a decrease. due to the differing nature of mapping techniques, certain decisions had to be taken. on the one hand, spectral analysis using erdas imagine 2011 remote sensing program has been used to edit the 2013 cadastral base layer with respect to the 2011 satellite image. on the other hand, to estimate the urban area in 1952, for which only rural maps were available, it has been necessary to erase those plots that appeared on the 1968 edited base layer, but not on the 1952 map. however, the territory comprising the urban area could not be modified because no information from 1952 was available. therefore they have been left the same as they were in 1968. also the variability in the different urban units represented on ancient maps had to be overcome. for instance, when it came to adjusting the base layer to the urban area in 1918, maps referred to city blocks and not to land plots. in other words, the available information was less detailed on the 1918 map than on the base layer, since each block includes several plots. less spatial information meant that blocks had to be edited as a whole as it was not possible to discriminate information related to each plot. the third level of the method has included analysing patterns and determining urban growth by means of using arcgis10.1 gis program. in this regard, pinho & oliveira, 2009 and tombolini et al., 2015 have been consulted. two aspects have been assessed: two-dimensional growth considering only the x and y axis. these calculations help examining the extent of soil sealing in quantitative terms. the sealed soil area (in hectares), the sealing increase between consecutive dates (in percentages), the mean annual sealing growth (in hectares per year) and the sealed soil among the total municipal territory (in percentages) have been calculated for each date. three-dimensional growth considering also the z axis. these figures help analysing the patterns followed by urban extension from a qualitative perspective. since building heights were not represented on maps and there was therefore a lack of information, this analysis has been based on auxiliary information related directly to vertical growth, i.e. population and population density. population of the town for each date from 1861 to 2011 has been obtained from a linear function between the two population censuses closest in time. population density for each date has been determined by dividing the estimated population by the urban area. three-dimensional urban growth has been calculated by means of the following parameters: population (in inhabitants), population increase between consecutive dates (in percentages), mean annual population growth (in inhabitants per year), population density (in inhabitants per sq km) and population density mean annual increase (in percentages). the fourth step of the method has consisted of examining the territory in situ and verifying the obtained figures and calculations. results the main calculations related to soil sealing and the characteristics of urban growth have been summarised in table 4 and fig. 4. table 4. two-dimensional and three-dimensional growth with respect to the two-dimensional growth of the urban area, which refers to the extent of the process from a quantitative perspective, the following should be noted: the old town was small in relation to the present-day built-up area. in 1861 it comprised 25.3 hectares, i.e. only 0.14% of the municipal territory. from 1861 to 1918 urban growth was subdued, with a total increase in the built-up area of 40%, which implies a mean annual growth of less than 0.20 hectares. this means that the loss of soil due to sealing in colmenar viejo affected hardly 0.19% of the municipal territory by 1918. from 1952 to 1991 a considerable urban extension took place and thus the built-up area of the town easily tripled. mean annual growth showed ever greater increases, going from 3 hectares/year at the initial time to nearly 6 hectares/year at the final stage. at the beginning of the 1990´s, 1.31% of the municipal territory was covered by built-up area. the greatest urban growth has taken place over the last two decades. from 1991 to 2011 the built-up area has doubled, with a mean annual growth of between 9 and 16 hectares. at the end of the period, the percentage of sealed soil has already reached 2.64%. therefore, the two-dimensional growth of the built-up area throughout the period under study has been both progressive, i.e. all periods have recorded positive increases, and exponential, i.e. the mean annual growth has increased over an ever larger area. the urban area has doubled in size during the period of fastest growth which comprises only twenty years, i.e. from 1991 to 2011. as a result, 2.64% of the municipal territory is now sealed as a consequence of urban processes. this percentage is lower than in other european cities, where soil sealed from 1990 to 2006 within a range of 6% to 9% (eea, 2012), and it is also lower than in other boroughs within the region of madrid, where these figures vary from 8.5% to 13.5% (garcía et al., 2014). these discrete figures are probably related to the proximity of colmenar viejo to some royal estates and the influence of their environmental protection. it could therefore be concluded that the data recorded in colmenar viejo should not give rise to much concern. however, it should also be borne in mind that since the borough under study covers the second largest 115soil sealing in madrid (spain), study case of colmenar viejo municipal territory within the region of madrid, small quantities in relative terms are highly significant in absolute terms. •1952-1968: detached houses for holiday home use •1979-1991: detached and semi-detached houses with ample gardens •2002-2011: vast urban extension and exorbitant soil sealing; extensive residential fabric with a predominance of houses over blocks conclusions this study has analysed the spatio-temporal dynamics and has enabled the measuring of the soil sealing process due to urban growth over the last 150 years in a borough located in the northern outskirts of madrid, i.e. colmenar viejo. this town is one of the boroughs of highest environmental value around the capital city of spain and has got both common characteristics to other spanish towns influenced by metropolitan processes and specific attributes that are only to be observed in this territory. the sources and the method used in this study are rather innovative. on the one hand, most soil sealing analysis only consider maps since the mid-20th century. instead, this study has enabled this retrospective study to go back in time an additional century with the use of 19th century maps obtained from the national geographic institute of spain (ign). on the other hand, integrating in a gis program a large volume of maps made by different cartographic techniques, i.e. classic topography, photogrammetry and remote sensing, has guaranteed the capture of a detailed and reliable urban cartography since the mid-19th century and has enabled the use and integration of various mapping techniques. furthermore, this study has aimed to show the spatio-temporal results of sealing at a municipal level so that it can be measured, updated and compared with other boroughs or territorial units. calculating the extension of urban growth from a quantitative perspective has been fulfilled by georeferencing ancient maps, digital maps, aerial photographs and satellite images, integrating them into a gis program and analysing some parameters regards the twodimensional urban extension. observing the patterns followed by urban growth from a qualitative point of view has been achieved by using population censuses accomplished since the mid-19th century and using an indicator system to analyse the three-dimensional urban growth. part of valera´s et al. (2011b) method for editing and adjusting maps to the urban area according to different dates has been used. on the one hand, always using the same layer as the base map avoids slivers. on the other hand, using a vector layer for a plot detail provides accurate and precise information for quantifying and analysing urban growth. results show that, in terms of quantity, urban growth has been both progressive, i.e. all periods have recorded positive increases, and exponential, i.e. the mean annual growth has increased over an ever larger area. along the whole period, urban extension has increased from 25.3 hectares to 482.5 hectares, i.e. from 0.14% to 2.64% of the municipal territory, what means multiplying 19-fold its size. the trend followed by urban growth has been therefore similar to other spanish towns influenced by metropolitan processes. however, the intensity of this process has remained slightly under the european and madrilian average, which is probably related to its proximity to some royal estates, what has conferred this territory a stronger environmental protection. in terms of quality, the patterns observed on urban extension comply with the general urban growth model recorded during the 20th and 21st centuries in the outskirts of major spanish cities. acknowledgments this paper has been supported by the science and innovation office of the spanish government under the research program cso-2012-34785, called “sellado de suelo en la comunidad de madrid y áreas limítrofes. estudio mediante teledetección y sistemas de información geográfica (soil sealing in madrid and surrounding areas. analysis using remote sensing programs and geographic information systems)”. figure 4. sealed soil in the borough of colmenar viejo from 1861 to 2011 with respect to the three-dimensional growth, which shows the patterns followed by urban processes from a qualitative perspective, the main features are: population growth has been steady throughout the whole period under study. the greatest mean annual growths were recorded in the 1970´s, with around 700 inhabitants per year, and during the final stage, with over 1,000 inhabitants per year. population density has varied constantly. it registered a plus sign over some periods, which shows a rise on population density, and a minus sign during some others, which denotes a drop on population density. variations in population density have remained modest, with mean annual changes within a range of ±1.5%. if only population statistics were to be taken into account, it could therefore be presumed that while population has risen steadily in numerical terms, it has not followed the same patterns in terms of population density. this infers that there have been some periods in which the built-up area has spread loosely and population density has dropped, and others in which population density has risen as a result of high building construction. however, if, in addition, these data are broken down into periods corresponding to available maps, and statistics are complemented with the ground truth, it can be concluded that the dominant processes are consistent with the general model of urban growth followed by metropolitan outskirts of major spanish cities. these can be summarised as follows: initial period from 1861 to 1918: modest yet steady extension, following the same patterns observed historically up to that time, i.e. narrow streets and one-storey buildings periods of vertical growth: • 1968-1979: tower blocks with very few gardens and leisure areas •1991-2002: residential estates of 3-4 storey buildings in landscaped surroundings periods of horizontal growth: •1918-1952: similar houses to those in the old town but including bigger courtyards and vegetable patches 116 andrés aristegui cortijo, maría eugenia pérez gonzález references arístegui, a. (2014). los trabajos topográfico-parcelarios de la junta general de estadística en boadilla del monte (madrid) [1863 1868]. ct catastro, 80, pp. 7-87. european environment agency (2012). documento de trabajo de los servicios de la comisión. directrices sobre mejores prácticas para limitar, mitigar o compensar el sellado del suelo. bruselas, 15.5.2012, swd (2012) 101 nal/2. garcía rodríguez, m. p. & pérez gonzález, m. e. (2007). changes in soil sealing in guadalajara: cartography with landsat images. science of total environment, 41. doi: 10.1016/j.scitotenv.2007.01.048. garcía rodríguez, m. p. & pérez gonzález, m. e. (2011). sellado de fluvisoles en la comunidad de madrid. análisis a partir de imágenes landsat. anales de geografía, 31 (2), 125–137. garcía rodríguez, m. p., pérez gonzález, m. e. & guerra zaballos, a. m. (2014). using tm images to detect soil sealing change in madrid (spain). geoderma, 214–215, 135–140. garcía rodríguez, m. p. & pérez gonzález, m. e. (2016). mapping of soil sealing by vegetation indexes and built-up index: a case study in madrid (spain). geoderma, 268, 100-107. kampouraki, m., wood, g. & brewer, t. (2006). the application of remote sensing to identify and measure sealed areas in urban environments. in: proceedings from isprs 1st international conference on objectbased image analysis (obis 2006) workshop proceedings, saltzberg, austria. meirich, s. (2008). mapping guide for a european urban atlas. gse land consortium. moeller, m. (2005). remote sensing for the monitoring of urban growth patterns. in: proceedings of the isprs joint conference urban/ urs. az, usa, tempe. montanarella, l. (2007). trends in land degradation in europe. in: climate and land degradation. germany, springer-heidelberg, sivakumar mvk, ndegwa n, edit., pp. 83-104. moreno, t. (2008). breve crónica de un siglo de catastro en españa (1906-2002). ct catastro, 63, pp. 31-60. muro, i., nadal, f. & urteaga, l. (1996). geografía, estadística y catastro en españa. barcelona, el serbal. nizeyimana, e. l., petersen, g. w., imhoff, m. l., sinclair jr., h. r., waltman, s.w., reed-margetan, d. s., levine, e. r. & russo, j. m. (2001). assessing the impact of land conversion to urban use on soils with different productivity levels in the usa. soil science society of america journal, 65, pp. 391-402. ojeda zújar, j. & villar lama, a. (2006). evolución del suelo urbano alterado en el litoral de andalucía. 1988-2002. revista internacional de ciencia y tecnología de la información geográfica, 7, pp. 73-99. pinho, p. & oliveira, v. (2009). cartographic analysis in urban morphology. environment and planning. b, planning & design, 36, 107-127. ridd, m. k. (1995). exploring a v–i–s (vegetation –impervious surface– soil) model for urban ecosystem analysis through remote sensing: comparative anatomy for cities. international journal of remote sensing, 16, 21652185. scalenghe, r. & ajmone-marsan, f. (2009). the anthropogenic sealing of soils in urban areas. landscape and urban planning, 90/1-2, 1-10. seifert, f.m. (2010). 11 improving urban monitoring toward european urban atlas. global mapping of human settlement: experiences, datasets, and prospects, 231. siedentop, s. & fina, s. (2012). who sprawls most? exploring the patterns of urban growth across 26 european countries. environment and planning, a44(11), 2765 – 2784. urteaga, l. & nadal, f. (2001). las series del mapa topográfico de españa, escala 1:50.000. madrid, centro nacional de información geográfica. tombolini, i., munafò, m. & salvati, l. (2015). soil sealing footprint as an indicator of dispersed urban growth: a multivariate statistics approach. urban research & practice, taylor & francis. valera lozano, a., añó vidal, c. & sánchez díaz, j. (2011a). crecimiento urbano (1956-2005) y sellado antropogénico del suelo en el municipio de alacant. serie geográfica, 17, 97-08. valera lozano, a., añó vidal, c. & sánchez díaz, j. (2011b). cincuenta años (1956-2006) de crecimiento urbano y degradación de suelos por sellado antropogénico en el término municipal de valencia. anales de geografía de la universidad complutense de madrid, 31/2, 177-191. wu, c. (2004). normalized spectral mixture analysis for monitoring urban composition using etm+ imagery”. remote sensing of environment, 93, 480-492. zhang, y. (2001). detection of urban housing development by fusing multisensor satellite data and performing spatial feature post classification. international journal of remote sensing, 22/17, 3339 3355. keywords: earthquake rayleigh waves; group velocity; s transform; multiple filter technique. palabras clave: ondas sísmicas rayleigh; velocidad de grupo; transformada s; técnica de filtrado múltiple (mft). how to cite item jiang, c., wang, y., & zeng, g. (2020). group velocity measurements of earthquake rayleigh wave by s transform and comparison with mft. earth sciences research journal, 24(1), 91-95. doi: https://doi.org/10.15446/ esrj.v24n1.85366 based upon the synthetic rayleigh wave at different epicentral distances and real earthquake rayleigh wave, s transform is used to measure their group velocities, compared with the multiple filter technique (mft) which is the most commonly used method for group-velocity measurements. when the period is greater than 15 s, especially than 40 s, s transform has higher accuracy than mft at all epicenter distances. when the period is less than or equal to 15 s, the accuracy of s transform is lower than that of mft at epicentral distances of 1000 km and 8000 km (especially 8000 km), and the accuracy of such two methods is similar at the other epicentral distances. on the whole, s transform is more accurate than mft. furthermore, mft is dominantly dependent on the value of the gaussian filter parameter α , but s transform is self-adaptive. therefore, s transform is a more stable and accurate method than mft for group velocity measurement of earthquake rayleigh waves. abstract group velocity measurements of earthquake rayleigh wave by s transform and comparison with mft mediciones de la velocidad de grupo de las ondas sísmicas rayleigh mediante la transformada s y comparación con la técnica de filtrado múltiple (mft) issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.85366 con base en ondas sintéticas rayleigh a diferentes distancias epicentrales y en ondas sísmicas rayleigh reales, en este trabajo se emplea la transformada s para medir la velocidad de grupo de este tipo de ondas y se compara con la técnica de filtrado múltiple (mft, por sus siglas del inglés, del nomre multiple filter technique), que es el método más utilizado para las mediciones de velocidad de grupo. cuando el periodo es mayor a 15 s, especialmente de 40 s, la transformada s tiene mayor precisión que mft, para todas las distancias epicentrales. cuando el periodo es menor o igual a 15 s, la precisión de la transformada s es menor que mft para distancias epicentrales de 1000 km y 8000 km (especialmente 8000 km). la precisión de estos dos métodos es similar para otras distancias epicentrales. en general, la transformada s es más precisa que mft. además, mft depende predominantemente del valor del parámetro de filtro gaussiano α, pero la transformada s es autoadaptativa. por lo tanto, la transformada s es un método más estable y preciso que mft para la medición de la velocidad de grupo de las ondas sísmicas rayleigh. resumen record manuscript received: 12/03/2019 accepted for publication: 17/12/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 91-95 chanjun jianga,b,c, youxue wanga,b*, gaofu zengd aschool of earth science, guilin university of technology, guilin 541006, china. bguangxi key laboratory of hidden metallic ore deposits exploration, guilin 541006, china. cbowen college of management, guilin university of technology, guilin 541006, china. dchina nonferrous metals (guilin) geology and mining co,. ltd, guilin 541006, china. * corresponding author: uxue.wang@glut.edu.cn g e o ph y si c s 92 chanjun jiang, youxue wang, gaofu zeng introduction earthquake surface wave tomography has become a powerful tool to image the deep structures of the crust and upper mantle due to its dispersion characteristics. measuring high-quality group velocities or phase velocities of surface wave, such as rayleigh wave, is a key foundation for surface wave tomography. currently, there are some methods to measure group velocities of surface wave, such as moving window analysis (mwa) (landisman et al., 1969), multiple filter technique (mft) (dziewonski et al., 1969) and continuous wavelet transform (cwt) (jiang et al., 2017), of which mft is the most commonly used one (ritzwoller & levshin, 1998; zhu et al., 2002; huang et al., 2003; cho et al., 2007; li et al., 2009; saygin & kennet, 2010; rindraharisaona et al., 2017; tang et al., 2018; lu et al., 2018; mechie et al., 2019). in 1969, mwa was proposed by landisman et al. using rectangular window in time domain. in the same year, mft was presented by dziewonski et al. using gaussian filter in frequency domain. when taking gaussian filter instead of rectangular window in time domain, mwa is equivalent to mft in frequency domain. mft, a commonly used method, is dominantly dependent on the value of gaussian filter parameter α, and much research has been done (inston et al., 1971; cara, 1973; nyman & landisman, 1977; herrmann & ammon, 2004; kolínsky, 2004; zhu et al., 2007; chen et al., 2014; jiang et al., 2019). especially, herrmann and ammon (2004) gave recommended values of gaussian filter parameter α with distance variation when the period is from 4 s to100 s. cwt was used to measure group velocities of earthquake rayleigh wave, compared to mft (jiang et al., 2017). in 1996, s transform was proposed by stockwell et al., which is an extension of cwt based on morlet wavelet (stockwell et al.,1996). s-transform combines advantages of short-time fourier transform (stft) and wavelet transform (wt). for instance, inverse s transform is directly related to fourier transform and it is a lossless transformation, linear transformation has no cross-term and high time-frequency resolution, the time-frequency resolution is related to the frequency of the signal, and the basic wavelet need not satisfy the admissibility condition (pinnegar & mansinha, 2003a, 2003b; zheng & wang, 2015). in recent years, s-transform has been widely used in geophysics (pinnegar & eaton, 2003, 2006; askari & siahkoohi, 2008; parolai, 2009; askari et al., 2011; tselentis et al., 2012; askari & ferguson, 2012; tu et al., 2013; zheng & wang, 2015). in this paper, s transform is introduced to measure group velocities of earthquake rayleigh wave and is compared with the commonly used mft. s transform if x(t) is a surface wave signal generated by natural earthquake, then the s transform of x(t) is defined as s f x t f t f i ft dtτ − −τ − −∞ ∞ , exp exp( ) = ( ) ( )         ( ) + ∫ 2 2 2 2 2 π π (1) where t is time, τ is the central point of the window function and can control the position of window function along time axis, and f is frequency. in s transform, gaussian function and basic wavelet are respectively defined as g t f t f f ( ) =       2 2 2 2 π exp − (2) w t f t f i ft g t i ftf f( ) =       = ( ) ( ) 2 2 2 2 2 2 π π πexp exp− − − (3) the basic wavelet in s-transform is the product of simple harmonics only scaled in time domain and gauss function scaled and shifted. time window width of s-transform varies inversely with frequency, i.e. the time window is wider in low frequency band, which leads to higher frequency resolution, while the time window in high frequency band is narrower, which leads to higher time resolution. therefore s transform is self-adaptive. inverse s transform is written as x t s f d e dfi ft( ) = ( )           ++ ∫∫ τ τ −∞ ∞ −∞ ∞ , 2π (4) comparison with mft at present, mft is the most commonly used method to measure group velocities of earthquake rayleigh wave. based upon the synthetic rayleigh wave at different epicentral distances and real earthquake rayleigh wave, s transform is compared with mft. mft choosing appropriate value of gauss filter parameter α can reduce the group-velocity measurement error of surface wave. the recommended values of α with epicentral distance variation in the period range from 4 s to 100 s given by herrmann and ammon (2004) are listed in table 1, in which the value of α at epicentral distance of 3 000 km is originally not available but is obtained by interpolation of the values at epicenter distances of 2 000 km and 4 000 km. table 1. values of gaussian filter parameter α epicentral distance (km) values of α 1000 25 2000 50 3000 75 4000 100 8000 200 synthetic rayleigh waves in order to check the effectivity of s transform in the measurement of group velocities and compare s transform with mft, synthetic data of rayleigh waves were generated by modal summation method (herrmann & ammon, 2004) based upon ak135 earth model (kennett et al., 1995), and the source parameters are listed in table 2. there are five stations with epicentral distances of 1 000 km, 2 000 km, 3 000 km, 4 000 km and 8 000 km, respectively. the azimuth angles of such stations are 45º and the sampling interval of data is 1 s. synthetic seismogram of rayleigh wave at epicentral distance of 3000 km and its amplitude spectra are shown in figure 1. table 2. the source parameters surface-wave-modes strike dip slip angle magnitude depth 5 0º 45º 0º 6.0 20 km and then s transform and mft are respectively applied to measure group velocities of the synthetic data . from the results by such two methods (fig. 2), the following conclusions can be drawn: 1) when the epicentral distance is 1 000 km, the accuracy of groupvelocity curve obtained by s-transform is slightly lower than that by mft in the period range of less than or equal to 15 s but much higher in the other period range, in the period range of greater than 40 s groupvelocity curves measured by s transform and mft both have large error and therefore the period to be measured should not be large when the epicentral distance is small; 2) when the epicentral distance is 2 000 km, 3 000 km or 4 000 km, the accuracy of group-velocity curve obtained by s-transform is similar as that by mft in the period range of less than or equal to 15 s but much higher in the other period range, especially when the period is greater than 40 s the group-velocity curve obtained by mft fluctuates obviously and has large error with the theoretical one while that by s transform fits well with the theoretical one; 93group velocity measurements of earthquake rayleigh wave by s transform and comparison with mft 3) when the epicentral distance is 8 000 km, the accuracy of group-velocity curve obtained by s-transform is lower than that by mft in the period range of less than or equal to 15 s but higher in the other period range. on the whole, the group-velocity curve gained by s transform fits better with the theoretical one than mft, therefore its accuracy is higher than that of mft. real earthquake rayleigh wave the real seismogram of earthquake rayleigh wave acquired from mobile station in guangxi of china is shown in fig. 3 and the parameters of the earthquake event are listed in table 3. the coordinates of the station are (110.50°e, 24.09°n) and epicentral distance is 3 551.22 km. for mft, the value of gauss filter parameter α at such epicentral distance is 87.5 by interpolation (table 1). figure 3. real seismogram of earthquake rayleigh wave acquired in guangxi of china. table 3. parameters of the earthquake event magnitude/mw greenwich mean time longitude/°e latitude/°n depth/km 7.1 2013/10/25 17:10:18.43 144.68 37.17 25.67 then s transform and mft are respectively applied to measure group velocities of such real data and the results are shown in fig.4. overall the group-velocity curves obtained by s transform and mft are similar. but when the period is greater than 40 s, the group-velocity curve obtained by mft fluctuates obviously while the curve by s transform is much smoother, hence s transform is more reliable than mft. conclusions in this paper, based upon the synthetic rayleigh wave at different epicentral distances and real earthquake rayleigh wave, the group velocities are measured by cwt and mft respectively. from the analysis and comparison, the conclusions can be drawn generally as following: 1) when the epicentral distance is small, such as 1 000 km, in the period range of greater than 40 s group velocities measured by s transform and figure 1. synthetic seismogram of rayleigh wave at distance of 3000 km (a) and its amplitude spectra (b) figure 2. comparison between s transform and mft based on synthetic data 94 chanjun jiang, youxue wang, gaofu zeng mft both have large error and are both not reliable and therefore the period to be measured should not be large in such case; 2) when the period is greater than 15 s, s transform has higher accuracy than mft at all epicenter distances, especially when the period is greater than 40 s the group-velocity curve obtained by mft fluctuates obviously and its accuracy is much lower than that by s transform; 3) when the period is less than or equal to 15 s, the accuracy of s transform is lower than that of mft at epicental distances of 1000 km and 8000 km (especially 8000 km), and the accuracy of such two method is similar at the other epicental distances. on the whole, s transform is more accurate than mft. furthermore, mft is dominantly dependent on the value of the gaussian filter parameter α, but s transform is self-adaptive. therefore, s transform is a more stable and accurate method than mft for group velocity measurement of earthquake rayleigh waves. acknowledgment funded by project 41574039 from nsf, key project 2016gxnsfda380014 from guangxi scientific foundation and project 2018ky0857 from guangxi education department. references askari, r., ferguson, r. j., & demeersman, k. (2011). estimation of phase and group velocities for multi-modal ground roll using the ‘phase shift’ and ‘slant stack generalized s transform based’ methods. crewes research report, 23, 1-11. askari, r., & ferguson, r. j. (2012). dispersion and the dissipative characteristic of surface waves in the generalized s-transform domain. geophysics, 77(1), v11-v20. askari, r., & slahkoohi, h. r. (2008). ground roll attenuation using the s and x-f-k transforms. geophysical prospecting, 56(1), 105-114. cara, m. (1973). filtering of dispersed wavetrains. geophysical journal of the royal astronomical society, 33(1), 65-80. cho, k. h., herrmann, r. b., ammon, c. j., & lee, k. (2007). imaging the upper crust of the korean peninsula by surface-wave tomography. bulletin of the seismological society of america, 97(1b), 198-207. dziewonski, a., bloch, s., & landisman, m. (1969). a technique for the analysis of transient seismic signals. bulletin of the seismological society of america, 59(1), 427-444. herrmann, r. b., & ammon, c. j. (2004). computer programs in seismology, 3.30[cp/ol]. http://www.eas.slu.edu/eqc/eqccps. html huang, z., su, w., peng, y., zheng, y., & li, h. (2003). rayleigh wave tomography of china and adjacent regions. journal of geophysical research: solid earth, 108(b2). inston, h. h., marshall, p. d., & blamey, c. (1971). optimization of filter bandwidth in spectral analysis of wavetrains. geophysical journal of the royal astronomical society, 23(2), 243-250. jiang, c. j., wang, y. x., xiong, b., & wang, h. y. (2019). measurements of surface-wave group velocity using mft and the value of gaussian filter parameter. journal of guilin university of technology, 39(2). jiang, c. j., wang, y. x., xiong, b., yun, p., xu, j. r., & nai, z. l. (2017). measurement of surface wave group velocity using wavelet transform. acta seismologica sinica, 39(3), 356-366. kennett, b. l. n., engdahl, e. r., & buland, r. (1995). constraints on seismic velocities in the earth from traveltimes. geophysical journal international, 122(1), 108-124. kolínsky, p. (2004). surface wave dispersion curves of eurasian earthquakes: the sval program. acta geodynamica et geomaterialia, 1(2), 165-185. landisman, m., dziewonski, a., & sato, y. (1969). recent improvements in the analysis of surface wave observations. geophysical journal of the royal astronomical society, 17(4), 369-403. li, h., su, w., wang, c. y., & huang, z. (2009). ambient noise rayleigh wave tomography in western sichuan and eastern tibet. earth and planetary science letters, 282(1-4), 201-211. lu, y., stehly, l., & paul, a. (2018). high-resolution surface wave tomography of the european crust and uppermost mantle from ambient seismic noise. geophysical journal international, 214(2), 1136-1150. mechie, j., schurr, b., yuan, x., schneider, f., sippl, c., minaev, v., gadoev, m., oimahmadov, i., abdybachaev, u., moldobekov, b., & orunbaev, s. (2019). observations of guided waves from the pamir seismic zone provide additional evidence for the existence of subducted continental lower crust. tectonophysics, 762, 1-16. nyman, d. c., & landisman, m. (1977). the display-equalized filter for frequency-time analysis. bulletin of the seismological society of america, 67(2), 393-404. parolai, s. (2009). denoising of seismograms using the s transform. bulletin of the seismological society of america, 99(1), 226-234. pinnegar, c. r., & eaton, d. e. (2003). application of the s transform to prestack noise attenuation filtering. journal of geophysical research, 108(b9). pinnegar, c. r., & mansinha, l. (2003a). the s-transform with windows of arbitrary and varying shape. geophysics, 68(1), 381-385. pinnegar, c. r., & mansinha, l. (2003b). the bi-gaussain s-transform. siam journal on scientific computing, 24(5), 1678-1692. pinnegar, c. r. (2006). polarization analysis and polarization filtering of threecomponent signals with the time-frequency s transform. geophysical journal international, 165(2), 596-606. rindraharisaona, e. j., tilmann, f., yuan, x., rümpker, g., giese, j., rambolamanana, g., & barruol, g. (2017): crustal structure of southern madagascar from receiver functions and ambient noise correlation: implications for crustal evolution. journal of geophysical research: solid earth, 122(2), 1179-1197. doi:10.1002/2016jb013565. figure 4. comparison between s transform and mft based on real data 95group velocity measurements of earthquake rayleigh wave by s transform and comparison with mft ritzwoller, m. h., & levshin, a. l. (1998). eurasian surface wave tomography: group velocities. journal of geophysical research: solid earth, 103(b3), 4839-4878. saygin, e., & kennett, b. l. n. (2010). ambient seismic noise tomography of australian continent. tectonophysics, 481(1-4), 116-125. stockwell, r. g., mansinha, l., & lowe, r. p. (1996). localization of the complex spectrum: the s transform. ieee transactions on signal processing, 44(4), 998-1001. tang, z., mai, p. m., chang, s. j., zahran, h. (2018). evidence for crustal low shear-wave speed in western saudi arabia from multi-scale fundamental-mode rayleigh-wave group-velocity tomography. earth and planetary science letters, 495, 24-37. tselentis, g., martakis, n., paraskevopoulos, p., lois, a., & sokos, e. (2012). strategy for automated analysis of passive microseismic data based on s-transform, otsu’s thresholding, and higher-order statistics. geophysics, 77(6), ks43-ks54. tu, r., wang, r. j., zhang, y., ge, m., & zhang, q. (2013). real-time coseismic velocity and displacements retrieving and de-noising processs by high-rate gnss. china satellite navigation conference(csnc), proceedings, lecture notes in electrical engineering, 244, 523-537. zheng, c. l., & wang, b. s. (2015). applications of s transform in seismic data processing. progress in geophysics, 30(4), 1580-1591. zhu, j. s., cao, j. m., cai, x. l., yan, z. q., & cao, x. l. (2002). high resolution surface wave tomography in east asia and west pacific marginal seas. chinese journal of geophysics, 45(5), 646-661. keywords: uav; remote sensing; sea area supervision; aerial photography; image processing. palabras clave: aeronaves no tripuladas; sensores remotos; supervisión de la zona marítima; fotografía aérea; procesamiento de imágenes. how to cite item chen, r. (2021). application of uav low altitude remote sensing system in sea area supervision. earth sciences research journal, 25(1), 65-68. doi: https://doi.org/10.15446/esrj. v25n1.94162 the sea area supervision is the premise and guarantee of safeguarding national security, protecting national sovereignty, and realizing the development of marine resources, and its importance is self-evident. to carry out the national sea area work more efficiently, this study designed low altitude-unmanned aerial vehicles (uav) remote sensing system applied to the sea area supervision and analyzed the remote sensing photography technology and remote sensing image processing technology. experiments verified the effectiveness of the system. the research results show that the uav-based low altitude remote sensing system can extract high-precision sea area information through aerial images’ interpretation. it is hoped that this study can provide some reference for improving the efficiency of current sea area supervision. abstract application of uav-low altitude remote sensing system in sea area supervision aplicación del sistema de teledetección con aeronaves no tripuladas de baja altitud en la supervisión de zonas marítimas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.94162 la supervisión del área marítima es la premisa y garantía de salvaguardar la seguridad nacional, proteger la soberanía nacional y desarrollar los recursos marinos. su importancia es evidente. para llevar a cabo el trabajo de manera más eficiente, este estudio diseñó un sistema de teledetección de baja altitud con aeronaves no tripuladas (uav) aplicado a la supervisión del área marítima y analizó la tecnología de fotografía de teledetección y la tecnología de procesamiento de imágenes de teledetección. la eficacia del sistema se verificó mediante experimentos. los resultados de la investigación muestran que el sistema de teledetección de baja altitud basado en vehículos aéreos no tripulados puede extraer información del área marina de alta precisión a través de la interpretación de imágenes aéreas. se espera que este estudio pueda proporcionar alguna referencia para mejorar la eficiencia de la supervisión actual del área marítima. resumen record manuscript received: 24/06/2020 accepted for publication: 21/12/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 65-68 rong chen1,2* 1department of urban construction and management, xiamen city university, xiamen 361008, china 2department of urban construction and management, the open university of china (xiamen branch), xiamen 361008, china *corresponding author: crxx0708@163.com g e o ph y si c s https://doi.org/10.15446/esrj.v25n1.94162 https://doi.org/10.15446/esrj.v25n1.94162 https://doi.org/10.15446/esrj.v25n1.94162 mailto:crxx0708@163.com 66 rong chen introduction with the rapid economic development of all countries in the world, land resources have been unable to meet the needs of national development, so the ocean with rich natural resources will become the focus of follow-up development of all countries in the world (lamchin et al., 2016; zhang et al., 2015). it is necessary to map the islands in the ocean to realize the search and development of marine resources (cao et al., 2016; gorji et al., 2017; nieuwhof et al., 2015). currently, the information acquisition method for islands and reefs in many countries is mainly naval manual measurement. however, this method is inefficient and accompanied by substantial operational risks, which can no longer meet current marine development needs (cummer & inan, 2016; liu & zhou, 2017). given this, this study proposes to use aeronaves no tripuladas (uav) instead of artificial, using uav remote sensing technology to detect islands in the sea area, hoping to provide a particular reference for the current sea area detection and supervision work. uav and gps-assisted uav remote sensing image acquisition and processing analysis of aerial triangulation the aerial triangulation method is one of the most commonly used methods of uav remote sensing measurement. this study obtains aerial image data through this method and carries out post-processing (ren et al., 2019; weixing et al., 2015). analytic aerial triangulation, also known as analytic aerial triangulation encryption, its calculation essence is to obtain the control point position through actual measurement within several air belts or in the lan composed of several air belts and carry out adjustment calculation for all image pair encryption points and external orientation elements of all images (amin et al., 2020; abboud & nofal, 2020; hoque et al., 2017). at present, three methods are mainly used in analytic aerial triangulation, which are an independent model method, flight belt method, and beam method (alfy, 2016; bi, 2020; xu, 2020). figure 1 is the schematic diagram of aerial triangulation with the beam method. in this method, the images captured by a uav remote sensing camera constitute a light beam, which is regarded as the basic adjustment calculation unit. the collinear equation of central projection is taken as the equation of adjustment calculation. after the beam’s rotation and translation, the light of the same name between the models can achieve the best intersection, and the whole region can be transformed into the ground coordinate system. then, the image’s exterior orientation elements and the ground coordinates of all the measured objects can be obtained through adjustment calculation. among the three analytical aerial triangulation methods mentioned above, the beam method has the highest accuracy and the most rigorous theory, so the beam method has become the most mainstream analytical aerial triangulation method. gps aided aerial triangulation in this study, the uav is in a low altitude flight state, its flight altitude is low, the fuselage is light, and the airflow over the ocean is large, and the airflow changes very fast. therefore, more heading and side overlap are needed in aerial measurement. at the same time, the single image area of the digital camera used by uav is small, and the field measurement workload will be doubled by using uav for aerial photography. in this case, reducing the number of ground image control points has become an urgent problem in uav remote sensing application. in order to reduce the difficulty of uav low altitude measurement, gps-aided aerial triangulation is introduced in this study. gps-assisted aerial triangulation refers to the installation of gps receiver on uav, and continuous observation of gps signal through airborne gps receiver and reference point receiver on the ground. through the gps carrier-phase differential positioning technology for data processing, the threedimensional space coordinates of uav at the time of shooting are obtained. the three-dimensional coordinate is taken as an additional observation value and introduced in the adjustment of the surveying area. the principle of converting the original ground control into air control is shown in figure 2. figure 2 is the schematic diagram of gps-aided aerial photogrammetry. when using this method, it is necessary to transform the aerial photography system and measure the eccentricity vector. the gps receiver antenna is installed on the aerial plane, and the eccentricity vector from the photographing center s to the gps receiver antenna phase center a is calculated. secondly, the gps receiver installed on the uav is used to obtain the fieldwork signal of aerial photography. according to the size of the area to be measured and the mapping scale, the ground reference stations are reasonably arranged by using differential gps positioning technology. next, the coordinates of the airborne gps station are calculated by gps observation data, and the coordinates are obtained by processing satellite ephemeris and gps observation data. finally, the photogrammetry and gps station coordinates are adjusted jointly. specifically, the spatial coordinates of the gps antenna phase center are used as photogrammetric data for regional network joint adjustment, and then the spatial coordinates and accuracy of the target object are calculated. at present, the increasingly mature uav remote sensing technology provides strong support for the sea area supervision, which can realize the rapid monitoring task of the sea area. however, the increasingly complex situation of marine supervision puts forward higher requirements for the processing efficiency of uav remote sensing photography. in this study, a variety of methods to improve the efficiency of the computer are combined to realize the rapid geometric processing of aerial photos. the image fast geometric processing process is divided into five steps. first, we need to match the image and get the corresponding connection point. under the condition of consistent images, the matching strength and speed of the algorithm are improved. secondly, combined with pos data, the rapid adjustment of the regional network is carried out, and the sparse matrix is figure 1. schematic diagram of beam method resolving aerial triangulation figure 2. schematic diagram of gps-assisted aerial photogrammetry 67application of uav-low altitude remote sensing system in sea area supervision introduced in the calculation process. after matching the connection point, the amount of adjustment calculation should be reduced as much as possible to improve the calculation efficiency. in the third step, the connection points are interpolated to obtain the dem in the survey area quickly, and the image ortho screening is carried out. next, on the basis of the face voronoi diagram image mosaic method, orthoimage mosaic is carried out to achieve less data redundancy and fast automatic mosaic. finally, automatic extraction of sea area elements from the image map of the monitoring area is carried out to obtain the location data of elements in the corresponding sea area. experimental research and analysis in order to verify the effectiveness of the uav low altitude remote sensing system, experiments were carried out to detect the coastline of islands in the sea area. it is very difficult to detect the coastline of islands by manual on-site measurement, and there are areas that are difficult to reach manually, so the fieldwork is very heavy. this study will detect the coastline of islands by uav remote sensing technology. the survey site was determined as an island bay in china. the vegetation coverage of the island bay is low, and the coastal topography is mainly hill with an altitude difference of less than 60m, which is suitable for uav low altitude aerial photography. in order to accurately grasp the location of the coastline, this study uses a three-dimensional laser to scan the coastal area of the island bay, and obtains centimeter-level fine dem. after the refined dem is obtained, gps rtk is used to measure a section of coastline in the island bay area, and the measured coastline is obtained. the coastline extracted on the basis of aerial photography and radar is compared with the field measured coastline, and the results are shown in figure 3. figure 3. coastline comparison of daowan bay figure 3 shows the comparison chart of island bay coastline results. (a), (b), (c) respectively show the comparison of general sandy shoreline, artificial shoreline, and bedrock shoreline. in figure 3(a), 15 feature points are selected for comparison; due to the difficulty of manual measurement, only 10 feature points are selected in figures 3(b) and 3(c). by comparing the coordinates of feature points obtained by different measurement methods, the measurement accuracy can be obtained. figure 4 shows the comparison results of the characteristic points of the sandy shoreline. as can be seen from figure 4, the sandy shoreline extracted from uav remote sensing images is quite different from that based on radar point cloud data. compared with the measured results, it is found that the sandy shoreline extracted from the uav remote sensing image is closer to the measured result, and the mean square deviation of both is less than 2 meters. figure 5 shows the comparison results of artificial shoreline and bedrock shoreline feature points. as can be seen from figure 5, the difference between the two methods based on uav remote sensing image and radar point cloud data is small, whether it is the coordinates of artificial shoreline feature points or bedrock shoreline feature points. through further calculation, it can be seen that the mean square deviation of the feature point coordinates is within 2.5 m in both cases. in conclusion, the uav-based low altitude remote sensing technology proposed in this study can interpret the information with high accuracy through aerial photography. this method can be combined with lidar technology in practical work, and then modify and offset the information extracted from remote sensing images to a certain extent. uav remote sensing system is expected to become an important tool for sea area monitoring. conclusion because the ocean contains rich natural resources, and it represents the national marine sovereignty, how to supervise the sea area has become the focus of attention of all countries. in this context, the study proposed to combine the sea area supervision and uav remote sensing technology, and verified the method by an example. the research results show that the sea area information obtained by the uav low altitude remote sensing system proposed in this study is closer to the measured information than lidar, and this method is expected to become an effective method to supervise the sea area. it is hoped that this study can contribute to the efficient development of marine regulatory work. in addition, there are still some problems in this study. when the uav is used to carry out aerial surveys in an area far away from the coast, the image obtained by the uav is likely to fall into the water, which will hinder the aerial triangulation to some extent, which will be improved in the future research. references abboud, i. a., & nofal, r. a. e. (2020). using remote sensing techniques to interpret geomorphological features along the east coast of the red sea, at yanbu, saudi arabia. arabian journal of geosciences, 13(14). alfy, m. e. (2016). assessing the impact of arid area urbanization on flash floods using gis, remote sensing, and hec-hms rainfall–runoff modeling. hydrology research, 47(6), 1142-1160. amin, m., khan, m. r., baig, m. h. a., baig, i. a., imran, m., & hanif, m. (2020). calculation and validation of actual evapotranspiration from satellite derived indices with observed data in delineated agro-climatic zones of punjab using remote sensing and gis figure 4. comparison of characteristic points of sandy shoreline figure 5. comparison of characteristic points between artificial shoreline and bedrock shoreline 68 rong chen techniques. applied ecology and environmental research, 18, 4637-4650. bi, t. (2020). optimal allocation algorithm of geological and ecological high-resolution remote sensing monitoring sampling points. earth sciences research journal, 24, 105-110. cao, g., zhou, l., & li, y. (2016) a new change-detection method in highresolution remote sensing images based on a conditional random field model. international journal of remote sensing, 37(5-6), 1173-1189. cummer, s. a., & inan, u. s. (2016). ionospheric e region remote sensing with elf radio atmospherics. radio science, 33(6), 1781-1792. gorji, t., sertel, e., & tanik, a. (2017). monitoring soil salinity via remote sensing technology under data scarce conditions: a case study from turkey. ecological indicators, 74, 384-391. hoque, a. a., phinn, s., roelfsema, c., & childs, i. (2017). tropical cyclone disaster management using remote sensing and spatial analysis: a review. international journal of disaster risk reduction, 22, 345-354. lamchin, m., lee, j. y., lee, w. k., lee, e. j., kim, m., lim, c. h., choi, h. a., & kim, s. r. (2016). assessment of land cover change and desertification using remote sensing technology in a local region of mongolia. advances in space research, 57(1), 64-77. liu, h., & zhou, q. (2017). accuracy analysis of remote sensing change detection by rule-based rationality evaluation with post-classification comparison. heilongjiang medicine journal, 25(5), 1037-1050. nieuwhof, s., herman, p. m. j., dankers, n., troost, k., & van der wal, d. (2015). remote sensing of epibenthic shellfish using synthetic aperture radar satellite imagery. remote sensing, 7(4), 3710-3734. ren, j., yang, w., yang, x., deng, x., zhao, h., wang, f., & wang, l. (2019). optimization of fusion method for gf-2 satellite remote sensing images based on the classification effect. earth sciences research journal, 23, 163-169. weixing, w., ting, c., sheng, l., & enmei, t. (2015). remote sensing image automatic registration on multi-scale harris-laplacian. journal of the indian society of remote sensing, 43(3), 501-511. xu, a. (2020). remote monitoring system of geological exploration in lava area based on gps. earth sciences research journal, 24, 207-214. zhang, s., li, f., li, t., yang, j., bu, k., chang, l., wang, w., & yan, y. (2015). remote sensing monitoring of gullies on a regional scale: a case study of kebai region in heilongjiang province, china. chinese geographical science, 25(005), 602-611. _30j0zll _1fob9te _3znysh7 _2et92p0 the concentration and distribution of aromatic biomarkers in coals and shales from five boreholes penetrating the maastrichtian mamu formation of the anambra basin, southeastern nigeria, were investigated by gas chromatography-mass spectrometry to assess the thermal maturity and organic matter input. the study focused on the variations of the relative abundances of naphthalenes, phenanthrenes, and monaromatic and triaromatic steroids identified on the mass fragmentograms. trimethylnaphthalene (tmn) is the most abundant member of the naphthalene family while methylphenanthrene (mp) is the most abundant phenanthrene family member. the total of phenanthrenes and their isomers was greater than that of naphthalenes. the distribution of these aromatic hydrocarbons and their akyl derivatives was strongly controlled by a selective expulsion mechanism and thermal maturation of organic matter. the low dibenzothiophene/phenanthrene (dbt/phen) ratios (0.01-0.06), as well as the enhanced concentrations of 1,2,5-tmn relative to 1,2,7tmn, indicates organic matter derived mainly from higher plants, and the extract ternary plot of c27, c28 and c29 monoaromatic steroids suggests a type iii and mixed type ii/iii kerogen. the calculated mean vitrinite reflectance (%rm), determined from the distributions of the isomers of methyldibenzothiophene ratio (mdr) in the rock extracts, ranged from 0.51 to 1.43. these maturity values indicate that the coal and shale extracts are marginally mature for hydrocarbon generation. la concentración y distribución de biomarcadores aromáticos en carbones y esquistos de cinco perforaciones en la formación maastrichtian mamu de la cuenca de anambra, en el sureste de nigeria, fueron analizados a través de un estudio de espectometría cromatográfico y de masa del gas para medir la madurez termal y la entrada de material orgánico. el estudio está enfocado en las variaciones de la abundancia relativa de naftalinas y fenantrenos, y en los esteroides monoaromáticos y triaromáticos identificados en los fragmentogramas de masas. la trimetinaftalina (tmn) es la más abundante de la familia de las naftalinas mientras el metilfenantreno (mp) es el más abundante de los fenantrenos. el tota de los fenantrenos y sus isómeros fue mayor que el de las naftalinas. la distribución de estos hidrocarbones aromáticos y sus alquilos derivados fue controlada ampliamente por un mecanismo de expulsión selectiva y de la maduración térmica de material orgánico. la baja proporción dibenziotofeno/fenantreno (dbt/ phen) (0.01-0.06), al igual que las concentraciones mejoradas de 1,2,5-tmn relativas de 1,2,7-tmn indican que la materia orgánica se deriva principalmente de plantas mayores, y del diagrama terniario de los esteroides monoaromáticos c27, c28 y c29 sugiere un tipo iii mezclado con tipos ii/iii de querógenos. el valor calculado de la reflectancia de vitrinita (%rm) determinado de la proporción de isómeros de metildibenziotofeno (mdr) en los extractos rocosos oscila de 0.51 a 1.43. estos valores de madurez indican que los extractos de carbones y esquistos son marginalmente maduros para la generación de hidrocarbono. earth sciences research journal eart sci. res. j. vol. 18, no. 1 (june, 2014): 51 62 abstract resumen key words: aromatic biomarkers, maturity, mamu formation, anambra basin palabras clave: biomarcadores aromáticos; madurez geológica; formación mamu, cuenca de anambra. using aromatic biological markers as a tool for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation, southeastern nigeria 1. jude e ogala, delta state university, nigeria. email: etunimogala@yahoo.com 2. mike i akaegbobi, university of ibadan, nigeria. email: izumike20022002@yahoo.com petroleum geology record manuscript received: 06/11/2012 accepted for publication: 29/04/2014 52 jude e. ogala & mike i. akaegbobi introduction aromatic biomarkers and their structural isomers have been found useful for the maturity assessment of oils, source rocks and coals (radke et al., 1982a; radke and welte, 1983; alexander et al., 1985; radke et al., 1986; garrigues et al., 1988; strachan et al., 1988; akaegbobi et al., 2000; akinlua et al., 2007; sonibare et al., 2008). complex mixtures of aromatic hydrocarbons have been identified in crude oils (cumbers et al., 1986), organic extracts of coals (radke et al., 1982a, b; puttman and villar, 1987) and ancient sediments (alexander et al., 1985; garrigues et al., 1988, 1990). these complex mixtures of polyaromatic hydrocarbons (pahs) are usually not generated through biosynthetic processes by living organisms in the sediments. however, some living organisms may synthesize benzoyl-derivatives in large quantities (akaegbobi et al., 2000). aromatic compounds found in crude oils and ancient sediments are believed to have been derived from modification through degradation of biologically produced compounds such as steroids and terpenoids (hall et al., 1984; strachan et al., 1988). they also observed that steroids give rise to substituted phenanthrenes, and terpenoids produce alkylnaphthalenes. other potential sources of alkylnaphthalene include the thermal degradation of spores, coals, and sporopollenin (allan and larter, 1981), and of kerogens and cyclic sesquiterpenoids present in resinous components (anderson et al., 1992). for the alkylated phenanthrenes, the terpenoids derived from higher plant constituents such as resins and waxes seem to be one major source. radke et al. (1982a) suggested that direct methylation of phenanthrene during catagenesis could be a substantial source of alkyl phenanthrene. furthermore, the conversion of higher-plant triterpenoids in sediments into aromatic hydrocarbons is through the release of oxygen functional groups and subsequent aromatization of the rings (johns, 1986; radke, 1987; rullkotter et al., 1994; akaegbobi et al., 2000). the ultimate products of such processes are therefore tetracyclic and pentacyclic aromatic hydrocarbons. through cleavages of bonds within the rings coupled with photochemical and acid catalyzed processes during diagenesis/ catagenesis, these tetracyclic and pentacyclic aromatic hydrocarbons could be broken into methyl-substituted naphthalenes and phenanthrenes, respectively. therefore, relative variations in the concentrations of akylnaphthalenes and alkylphenanthrenes in the sediments could supply information on the maturity of organic matter, organic facies input and depositional environment (seifert and moldowan, 1978, 1986; radke et al., 1982a, b; radke, 1988). the aim of this paper is to examine the ratios of aromatic compounds in coal and shale extracts from the maastrichtian mamu formation in the anambra basin, southeastern nigeria, and to assess the organic matter input and thermal maturity for the source rocks. geological setting the anambra basin, constituting the southwestern portion of the benue trough (fig. 1), is a roughly triangular-shaped sedimentary depression covering an area of approximately 40,000 km2 (akaegbobi et al., 2000). the present study area lies within latitudes 6º20/ to 7º55/ n and longitudes 7º10/ to 8º00/ e (fig. 2). the anambra basin was initiated in the early cretaceous with the formation of the benue-abakaliki trough as a failed arm of a rift triple junction associated with the separation of the african and south american continents and the subsequent opening of the south atlantic ocean (murat, 1972).the ne-sw trending benue-abakaliki trough is believed to be the result of a pre-albian rifting of the african shield prior to the opening of the south atlantic ocean (uzuakpunwa, 1974). the movements were reactivated by further plate activity in the lower tertiary soon after the intermittent upper cretaceous rifting (petters, 1978). the anambra basin contains over 6 km thick cretaceous/tertiary sediments and provides a structural link between the cretaceous benue trough and the tertiary niger delta (whiteman, 1982; onuoha, 1986; mohammed, 2005). figure 1: geologic map of the anambra basin and adjoining areas in southeastern nigeria showing sample locations (modified from akande et al., 2007). figure 2: location of study area with sampling sites. 53using aromatic biological markers as a tool for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation, southeastern nigeria figure 3: stratigraphic subdivision of southern benue trough. the stratigraphic succession ranges from late santonian to eocene (fig. 3), with sediments of fluvial, deltaic, pro-delta, and shelf facies origins. accumulation of sediments in the anambra basin commenced with the campano-maastrichtian marine and paralic shales of the enugu and nkporo formations, along with the deltaic owelli sandstone equivalent. these basal units are overlain successively by the coal measures of the mamu formation (lower coal measures), the ajali sandstone (middle coal measures) and the nsukka formations (upper coal measures).the nsukka formation, which overlies the ajali sandstone, begins with coarseto medium-grained sandstones and passes upward into well-bedded blue clays, fine-grained sandstones, and carbonaceous shales with thin bands of limestone. the marine shales of the imo formation were deposited in the paleocene, and overlain by the ameki formation (lateral equivalents of the tidal nanka sandstones) of eocene age which constitutes the tertiary succession (petters and ekweozor, 1982)(fig. 3). figure 4a: m/z 156, m/z 170 and m/z 184 mass chromatograms showing the distributions of naphthalenes and its alkyl derivatives of representative coal sample (og/1267/03) from the anambra basin figure 4b: m/z 156, m/z 170 and m/z 184 mass chromatograms showing the distributions of naphthalenes and its alkyl derivatives of representative shale sample (eng/1001/02) from the anambra basin materials and methods seven core samples consisting of coal (4) and shale (3) having total organic carbon (toc) values ranging from 3.35 to 55.07 wt.% were carefully selected at various depth intervals (75-221 m) from five boreholes penetrating the mamu formation in the anambra basin in three coal fields at enugu, ezimo and ogboyoga (fig. 2). pulverized samples were exhaustively extracted by soxhlet extraction using dichloromethane as solvent. all samples were prepared according to standard organic geochemical sample preparation techniques (radke et al., 1980, 1984). aliquots of the extracts were separated into saturated and aromatic hydrocarbons and heterocyclic compounds by liquid chromatography using a dual column system and a back-flushing technique as a preparatory step for biomarker analysis. the asphaltene fraction of the sample was recovered by precipitation in pentane overnight at room temperature. the pentane soluble portion of the sample (i.e., the maltenes) was further separated by liquid chromatography into saturated, aromatic and resin fractions using a large column containing silica gel with fractions quantified by gravimetric analysis. aromatic hydrocarbons were eluted using 40 ml of 90:10 (v/v) pentane/dichloromethane mixtures. the aromatic fractions were diluted with 10 µl of carbon disulfide (cs2) per mg of fraction and analyzed by gas chromatography-mass spectrometry using a hewlett packard 6890 gc/5973 msd system (agilent technologies, california, u.s.a.) equipped with a highresolution column (60 m, db-1 phase, 0.2 mm i.d., 0.2 µm film). a constant flow of hydrogen carrier gas was used during the entire gas chromatographic run.the gc was temperature programmed from 80 °c (2 minutes hold) to 3200c at 3.5 0c/minute with a final hold time of 20 minutes. there was a 7.5-minute solvent delay on acquisition, and selected ion monitoring (sim) was the standard method for analysis. results biomarker chemistry the mass chromatograms of naphthalenes, phenanthrenes and akyl derivatives of representative coal and shale extracts from the maastrichtian mamu formation in the anambra basin are shown in figures 4a-b and 5a-b, respectively. the identification of the peaks labeled in figures 4a-b and 5a-b are reported in tables 1 and 2. two methylnaphthalenes (mn), two ethylnaphthalenes (en), nine dimethylnaphthalenes (dmn), ten trimethylnaphthalenes (tmn), eleven tetramethylnaphthalenes (temn) and two ethyl-methyl-naphthalenes (emn) isomers were identified in the aromatic fractions of the coal and shale extracts (figs. 4a-b). among the tricyclic aromatic hydrocarbons, four methylphenanthrenes (mp), and fourteen dimethylphenanthrene isomers were identified (figs. 5a-b). dimethylnaphthalene was the most abundant among the naphthalene family (fig. 6). all samples have similar distributions of phenanthrenes (m/z 178). 54 jude e. ogala & mike i. akaegbobi figure 5a: m/z 178, m/z 192 and m/z 206 mass chromatograms showing the distributions of phenanthrene and its alkyl derivatives of coal sample (og/1267/03) from the anambra basin. figure 5b: m/z 178, m/z 192 and m/z 206 mass chromatograms showing the distributions of phenanthrene and its alkyl derivatives of shale sample (eng/1001/02) from the anambra basin table 1: peak identification in the m/z 156, 170, 184, 198, 154, 168, 178 and 192 mass chromatograms ion peak label compound name 128 naph naphthalene 142 2mn 2-methylnaphthalene 142 1mn 1-methylnaphthalene 156 2en 2-ethylnaphthalene 156 1en 1-ethylnaphthalene 156 26dmn 2,6-dimethylnaphthalene 156 27dmn 2,7-dimethylnaphthalene 156 13+17dmn 1,3 & 1,7-dimethylnaphthalenes 156 16dmn 1,6-dimethylnaphthalene 156 23dmn 2,3-dimethylnaphthalene 156 14dmn 1,4-dimethylnaphthalene 156 15dmn 1,5-dimethylnaphthalene 156 12dmn 1,2-dimethylnaphthalene 170 bb_emn ethyl-methyl-naphthalene 170 ab_emn ethyl-methyl-naphthalene 170 137tmn 1,3,7-trimethylnaphthalene 170 136tmn 1,3,6-trimethylnaphthalene 170 146+135t (1,4,6+1,3,5)-trimethylnaphthalenes 170 236tmn 2,3,6-trimethylnaphthalene 170 127tmn 1,2,7-trimethylnaphthalene 170 167+126t (1,6,7+1,2,6)-trimethylnaphthalenes 170 124tmn 1,2,4-trimethylnaphthalene 170 125tmn 1,2,5-trimethylnaphthalene 184 1357 1,3,5,7-tetramethylnaphthalene 184 1367 1,3,6,7-tetramethylnaphthalene 184 1247 (1,2,4,7+1,2,4,6+1,4,6,7)-tetramns 184 1257 1,2,5,7-tetramethylnaphthalene 184 2367 2,3,6,7-tetramethylnaphthalene 184 1267 1,2,6,7-tetramethylnaphthalene 184 1237 1,2,3,7-tetramethylnaphthalene 184 1236 1,2,3,6-tetramethylnaphthalene 184 1256 1,2,5,6-tetramethylnaphthalene 184 dbt dibenzothiophene 198 4mdbt 4 methyl dibenzothiophene 198 2+3mdbt 2 & 3 methyl dibenzothiophenes 198 1mdbt 1 methyl dibenzothiophene 154 bp biphenyl 168 2mbp 2-methylbiphenyl 168 dpm diphenylmethane 168 3mbp 3-methylbiphenyl 168 4mbp 4-methylbiphenyl 168 dbf dibenzofuran 178 phen phenanthrene 192 3mp 3-methylphenanthrene 192 2mp 2-methylphenanthrene 192 9mp 9-methylphenanthrene 192 1mp 1-methylphenanthrene 55using aromatic biological markers as a tool for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation, southeastern nigeria figure 6: relative distribution (concentration) of naphthalene homologues in rock extracts from the anambra basin. see table 3 caption for abbreviations. figure 7: relative distribution (concentration) of phenanthrene homologues in rock extracts from the anambra basin. see table 3 caption for abbreviations. dimethylphenanthrene was the least abundant of the phenanthrene homologues (fig. 7). the distribution of total naphthalenes and phenanthrenes in the coal and shale extracts from the mamu formation indicates that the total sum of the phenanthrenes and their isomers was greater than that of the naphthalenes in all the samples analyzed (fig.8). the distribution of these aromatic hydrocarbons and their akyl derivatives was strongly controlled by a selective expulsion mechanism and thermal maturation of organic matter (akaegbobi et al., 2000). in the present study, the relative abundance of the phenanthrenes and their akyl derivatives was greater than that of the naphthalenes. radke et al. (1982a) suggested that a substantial source of alkyl phenanthrene could be the direct methylation of phenanthrene during catagenesis. the results obtained in this study corroborate with those reported by akinlua et al. (2007).these isomers were applied in the calculation of temperature-sensitive maturity parameters such as methylphenanthrene indices (mpi-1 and mpi-2), methyldibenzothiophene ratio (mdr), dimethylnaphthalene ratios (dnr-1 and dnr-2) and trimethylnaphthalene ratio (tnr-1). 0 50000 100000 150000 200000 250000 300000 350000 400000 t-naph t-phen s1 s2 s3 s4 s5 s6 s7 figure 8: relative concentrations of the sum total of naphthalenes (t-naph) and phenanthrenes (t-phen) in rock extracts from the anambra basin. see table 3 caption for abbreviations. table 2: peak identification in the m/z 206, m/z 191, m/z 231 and m/z 253 mass chromatograms ion peak label compound name 206 36dmp 3,6-dimethylphenanthrene 206 26dmp 2,6-dimethylphenanthrene 206 27dmp 2,7-dimethylphenanthrene 206 39dmp (3,9+3,10+2,10+1,3)-dimethylphenanthrenes 206 29dmp (2,9+1,6)-dimethylphenanthrenes 206 17dmp 1,7-dimethylphenanthrene 206 23dmp 2,3-dimethylphenanthrene 206 19dmp 1,9-dimethylphenanthrene 206 18dmp 1,8-dimethylphenanthrene 206 12dmp 1,2-dimethylphenanthrene 191 bh32 c32 benzohopane 191 bh33 c33 benzohopane 191 bh34 c34 benzohopane 191 bh35 c35 benzohopane 231 1ta20 c20 triaromatic steroid 231 2ta21 c21 triaromatic 231 3ta26 c26 20s triaromatic 231 4ta26+27 c27 20s & c26 20r triaromatic 231 5ta28 c28 20s triaromatic 231 6ta27 c27 20r triaromatic 231 7ta28 c28 20r triaromatic 231 8ta29 c29 20r triaromatic (24 n-propyl) 253 1ma21 c21 ring-c monoaromatic steroid 253 2ma22 c22 monoaromatic steroid 253 3ma27 c27 reg 5b(h),10b(ch3) 20s 253 4ma27 c27 dia 10b(h),5b(ch3) 20s 253 5ma27 c27 dia10bh,5bch3 20r+reg5bh,10bch3 20r 253 6ma27 c27 reg 5a(h),10b(ch3) 20s 253 7ma28 c28 dia 10ah,5ach3 20s+reg5bh,10bch3 20s 253 8ma27 c27 reg 5a(h),10b(ch3) 20r 253 9ma28 c28 reg 5a(h),10b(ch3) 20s 253 10ma28 c28 dia 10ah,5ach3 20r+reg5bh,10bch3 20r 253 11ma29 c29 dia 10bh,5bch3 20s+reg5bh,10bch3 20s 253 12ma29 c29 reg 5a(h),10b(ch3) 20s 253 13ma28 c28 reg 5a(h),10b(ch3) 20r 253 14ma29 c29 dia 10bh,5bch3 20r+reg5bh,10bch3 20r 253 15ma29 c29 reg 5a(h),10b(ch3) 20r 56 jude e. ogala & mike i. akaegbobi figure 9: relative distribution (concentration) of aromatic hydrocarbons in rock extracts from the anambra basin. see table 3 caption for abbreviations. discussion thermal maturity the concentration of aromatic compounds, such as naphthalenes, phenanthrenes, dibenzothiophenes and their structural isomers are at present attracting increasing attention as maturity indicators for coals and shale source rocks. these indicators rely either on an increase in maturity with the degree of alkylation of a given parent compound, or a shift in the isomer distribution of alkylaromatic homologues increasing towards isomers with greater thermal stability (radke, 1988). the aromatic isomer ratios and indexes (dnr-1, dnr-2, tnr-1, tde-1, tde-2, mpi-1, mpi-2, and mdr) used in this study are reported in table 5. the total concentration of naphthalene and its isomers is greater than that of the phenanthrenes. the higher abundances of naphthalenes relative to phenanthrenes can be attributed to selective expulsion mechanisms and direct methylation during diagenesis/catagenesis (akaegbobi et al., 2000; radke et al., 1982a). the increase in the concentrations of naphthalenes along with an increase in the number of alkyl substitutes indicates that the oils are mature. this increase in the number of methyl substituents of naphthalene in mature oils occurs because of the thermal rearrangement of methylnaphthalene with table 3: abundance of aromatic hydrocarbons in rock extracts (concentrations are in ppm) s a m p l e number lithology n mn en dmn emn tmn temn p mp dmp t-naph t-phen og/1267/03 shale nd 16.70 9.40 532.90 65.60 1077.30 499.80 293865.20 52542.10 15581.00 2201.70 361988.30 eng/1001/02 shale nd 759.50 153.10 8496.00 292.60 3016.60 583.70 5554.50 2403.90 2799.50 13301.50 10757.90 eng/1001/12 coal 358.50 505.00 20.90 859.60 27.40 301.00 357.30 1280.10 1157.40 1275.00 2429.70 3712.50 eng/1001/16 coal nd 14.80 11.10 858.20 86.80 1703.10 1053.80 14007.80 8875.70 3444.00 3727.80 26327.50 eng/1002/06 shale nd 21.30 22.5.00 1479.40 230.90 2852.50 657.60 33841.70 3946.90 1240.40 5264.20 39029.00 eng/1008/13 coal 39.50 402.20 25.00 1175.80 49.30 581.90 287.20 2397.80 686.10 889.70 2560.90 3973.60 ez/1219/08 coal 372.50 793.00 62.20 1991.50 76.30 1157.80 526.30 63560.40 7983.40 2703.20 4979.60 74247.00 abbreviations: n – naphthalene, mn – methylnaphthalene, en – ethylnaphthalene, dmn – dimethylnaphthalene, emn – ethylmethylnaphthalene, tmn – trimethylnaphthalene, temn – tetramethylnaphthalene, p – phenanthrene, mp – methylphenenthrene, dmp dimethylphenenthrene, t-naph – total naphthalene, t-phen – total phenanthrene, nd – not detected table 4: abundance of isomers of trimethylnaphthalenes and methylphenanthrenes in rock extracts (concentrations are in ppm) sample number depth (m) lithology 2-mp 3-mp 1-mp 9-mp 125tmn 127tmn og/1267/03 75-76 shale 19203.20 13913.70 10818.90 8606.30 173.50 44.20 eng/1001/02 157 shale 519.90 369.80 595.40 918.80 389.90 159.40 eng/1001/12 186-187 coal 144.90 184.80 179.00 648.70 34.90 16.50 eng/1001/16 205 coal 1797.30 1065.80 2640.50 3372.10 418.20 144.80 eng/1002/06 167-178 shale 1446.70 1079.20 701.20 719.80 325.90 176.50 eng/1008/13 220-221 coal 144.50 120.40 108.90 312.30 62.50 30.60 ez/1219/08 80-82 coal 2567.10 1784.30 2113.60 1518.40 166.80 50.20 abbreviations: 1-mp – 1-methylphenanthrene, 2-mp – 2-methylphenanthrene, 3-mp – 3methylphenanthrene, 9-mp – 9methylphenanthrene, 125tmn – 1, 2, 5-trimethylnaphthalene, 127tmn – 1, 2, 7-trimethylnaphthalene 57using aromatic biological markers as a tool for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation, southeastern nigeria increasing thermal stress (akaegbobi et al., 2000). furthermore, the coal and shale extracts contain enhanced concentrations of 1, 2, 5-trimethylnaphthalene relative to 1, 2, 7-trimethylnaphthalene (table 4). this is typical of mature oils (strachan et al., 1988; akinlua et al., 2007). however, in the present study, the relative abundance of the total of phenanthrene and its isomers was greater than that of naphthalenes (table 3, fig.9). the abundance of phenanthrenes relative to naphthalenes suggests that the rock extracts are mature (akinlua et al., 2007). figure 10a: m/z 231 and 253 mass chromatograms showing the distributions of triaromatic steranes and monoaromatic steranes in representative shale sample og/1267/03 from the anambra basin. figure 10b: m/z 231 and 253 mass chromatograms showing the distributions of triaromatic steranes and monoaromatic steranes in representative coal sample eng/1001/12 from the anambra basin. table 5: thermal maturity parameters computed from the aromatic biomarkers and aromatic sulphur compounds in the coal and shale extracts interpretative ratios og/1267/03 eng/1001/02 eng/1001/12 eng/1001/16 eng/1002/06 eng/1008/13 ez/1219/08 naphthalenes dnr-1=(2,6+2,7)/(1,5)dmn 2.08 3.93 4.13 2.04 2.71 6.03 4.93 dnr-2=(2,6+2,7)/(1,4+2,3)dmn 1.21 2.52 3.37 1.35 1.72 2.93 1.91 tnr1=(2,3,6)/(1,4,6+1,3,5)tmn 1.12 0.86 0.73 0.75 0.95 0.87 1.22 tde-1=1,2,5-tmn/1,2,4-tmn 10.43 10.81 4.41 15.21 7.22 7.22 11.51 tde-2=1,2,7-tmn/1,2,6-tmn 0.20 0.30 0.33 0.53 0.33 0.31 0.20 1,2,7-tmn/1,3,7-tmn 0.45 0.46 0.44 1.07 0.54 0.39 0.41 phenanthrenes mpi-2=3(2mp)/(phen+1mp+9mp) 0.18 0.22 0.21 0.27 0.12 0.15 0.11 mpi-1=1.5(2mp+3mp)/(phen+1mp+9mp) 0.16 0.19 0.23 0.21 0.11 0.14 0.10 rc(a)=0.6(mpi-1)+0.37 (for ro<1.3) 0.47 0.48 0.51 0.50 0.43 0.45 0.43 rc(b)=0.6(mpi-1)+2.3 (for ro>1.3) 2.40 2.41 2.44 2.43 2.36 2.38 2.36 mdr=4mdbt/1mdbt 6.56 1.69 0.40 0.72 3.13 0.88 0.62 rm=0.40+0.30(mdr)-0.094(mdr)2+0.011(mdr)3 1.43 0.69 0.51 0.57 0.76 0.60 0.55 mdr23=23mdbt/dbt 0.42 0.31 0.65 0.32 0.21 0.75 0.06 mdr1=1mdbt/dbt 0.07 0.23 0.66 0.75 0.13 0.67 0.14 dbt/phenanthrene 0.06 0.02 0.04 0.01 0.01 0.02 0.01 triaromatic steroids tas c20+c21/all tas (20r and 20s) 0.52 0.13 0.06 0.11 0.21 0.09 0.13 tas c20/c20+c27 0.70 0.22 0.12 0.19 0.31 0.20 0.20 tas c20/c20+c28 (20r) 0.72 0.26 0.08 0.20 0.42 0.17 0.35 tas c21/c21+c28 0.15 0.08 0.04 0.07 0.13 0.05 0.10 tas c28/c26 (20s) 3.62 3.89 12.71 4.04 2.09 17.11 5.6 tas c28/c27 (20r) 2.12 2.16 3.96 2.92 2.27 2.87 1.33 % c26 tas 5.30 5.70 2.30 6.70 8.20 2.30 4.70 % c27 tas 36.70 35.90 23.70 32.20 40.5 23.9 35.80 % c28 tas 52.80 50.70 68.90 57.70 42.70 70.10 43.70 % c29 tas 5.20 7.60 5.10 3.40 8.60 3.60 15.70 monoaromatic steroids mas dia/reg c27 0.00 0.69 0.00 1.21 0.91 1.17 2.19 mas c21+c22/all mas 0.05 0.04 0.04 0.04 0.05 0.04 0.06 % c27 ma 50.10 43.00 11.00 26.50 46.30 14.20 11.60 % c28 ma 29.30 21.70 14.50 21.80 21.00 20.90 22.60 % c29 ma 20.60 35.20 74.50 51.70 32.70 65.00 65.70 tas/(tas+mas) 0.22 0.38 0.59 0.46 0.28 0.57 0.23 tas c28/(ta c28+ma c29) 0.58 0.79 0.87 0.87 0.69 0.88 0.39 58 jude e. ogala & mike i. akaegbobi a number of aromatic maturity parameters based on the distribution of naphthalene, phenanthrene and alkyl isomers were calculated for the study samples after radke et al. (1982a), radke et al. (1986) and alexander et al. (1985, 1986). radke (1988) observed that aromatic hydrocarbons do change in a regular fashion with increasing maturity and thus developed maturity parameters based on the distributions of alkylnaphthalene and alkylphenanthrene.the maturity parameters computed from the aromatic biomarker distributions in the coal and shale extracts are listed in table 5. the maturity parameters computed from naphthalene and its isomers, the dimethylnaphthalene ratios (dnr-1 and dnr-2) and trimethylnaphthalene ratio (tnr-1) ranged from 2.04 to 6.03, 1.21 to 3.37 and 0.73 to 1.22, respectively (table 5). the methylphenanthrene index (mpi-1) is one of the most widely used maturity parameters based on aromatic hydrocarbons (radke et al., 1982a, 1982b; radke, 1988). the parameter relies on a shift with maturity in the methylphenanthrene distribution towards a preponderance of β-type isomers. the methylphenanthrene indices (mpi-1 and mpi-2) were computed based on the relative concentrations of the 1-mp, 2-mp, 3-mp, 9-mp and phenanthrene (p). this was based on the assumption that the 2-mp and 3-mp are derived from the rearrangement of the 1-mp and 9-mp, and from phenanthrene (p) through methylation reactions. the relative abundances of the alkyl homologs of the aromatic hydrocarbons (phenanthrenes) were used to calculate the vitrinite reflectance (% rc) after the method of radke et al., (1982a, b) and radke (1988). the mpi-1 is often used for estimating the equivalent vitrinite reflectance values (% rc) for crude oils and source rocks. the mpi-1 and % rc values calculated for the coal and shale extracts range from 0.10 to 0.23 and 0.43 to 0.51, respectively (table 5). mpi-1 calculated reflectance (%rc) was carried out because of insufficient and good reflecting vitrinite in the samples. maturity parameters based on the isomers of the methyldibenzothiophene (mdbt) were calculated (radke et al., 1986). this ratio (mdr) generally increases with increasing maturity and correlates well with vitrinite reflectance and the rock-eval tmax of source rock (radke, 1988). the ratio is also useful in calculating mean vitrinite reflectance (% rm). the methyldibenzothiophene ratio (mdr) and mean vitrinite reflectance (%rm) for the coal and shale extracts ranged from 0.40 to 6.56 and 0.51 to 1.43, respectively (table 5). the high mdr exhibited by some of the samples may be a result of the effect of heating rates or thermal stress (akaegbobi et al., 2000). the % rm values show that the coal and shale extracts range from immature to mature (table 5). monoaromatic and triaromatic steroids were detected in all samples in the m/z 253 and m/z 231 fragmentograms (figs. 10a-b). both 20r and 20s homologues of c26-c28, c20 and c21 triaromatic compounds are present (figs. 10a-b). figure 12a: m/z 184 and 198 mass chromatograms showing the distributions of aromatic sulfur compounds of representative shale sample in well eng/1001/02 from the anambra basin. figure 12b: mass chromatograms showing the distributions of aromatic sulfur compounds of representative coal sample in well eng/1001/12 from the anambra basin. peters and moldowan (1993) and el-gayar (2005) suggested that the extent of cracking in the side chains of triaromatic steroids can be used to provide information about petroleum maturity. the triaromatic steroid maturity parameters c20/c20+c28 (20r) and (c20+c21)/ (c20+c21+c26 to c28 20r and 20s) were used to evaluate the thermal maturity of the coal and shale extracts in this study (peters and moldowan, 1993; peters et al., 2005). the maturity parameters computed from the peak areas of the m/z 231 ranged from 0.08 to 0.72 and 0.09 to 0.52, respectively (table 5). these values indicate an immature to marginally mature status of the source rock. the monoaromatic steroid ratios (mas dia/reg c27 and mas c21+c22/ all mas) were measured from the m/z 253 fragmentograms (table 5) and used to evaluate the source-rock depositional environment and thermal maturity (peters and moldowan, 1993). both of these ratios increase with higher thermal maturity (mackenzi et al., 1981; moldowan and fago, 1986). the mas dia/ reg c27 sterane ratio ranges from 0.69 to 2.19 (table 5). the coal and shale extracts in the mamu formation contain significant amounts of diasteranes, which appear to form through the interaction of steranes with clay mineral surfaces in source rocks (rubinstein et al., 1975; sieskind et al., 1979). thus, the presence of significant amounts of diasteranes relative to regular steranes in oils has been used as evidence of petroleum generation from argillaceous source rocks, whereas low concentrations are considered an indication of a source rock lacking in clay minerals (hughes, 1984; mello et al., 1988a, 1988b; hill et al., 2007). the calculated monoaromatic maturity ratio (mas c21+c22/ all mas) ranges from 0.04 to 0.05 (table 5). thermal maturity was also evaluated on the basis of monoaromatic and triaromatic steroids because they are more resistant to the effects of biodegradation than alkane-type biological markers (volkman et al., 1983). according to tissot and welte (1984), aromatization of c-ring steroid hydrocarbons occurs in mature oils. with an increase in thermal maturation, there is a conversion of c-ring monoaromatic steroid hydrocarbons to abcring triaromatic steroid hydrocarbons (akinlua and ajayi, 2009). the tas/ (tas+mas) and tas 28/ (tas 28+mas 29) ratios ranged from 0.22 to 0.59 and 0.39 to 0.88, respectively (table 5), indicating an immature to marginally mature status of the rock extracts. source of organic matter the trimethylnaphthalene is dominated by 1, 2, 5tmn in all the samples analyzed (figs. 4a-b and table 4). the high abundance of the naphthalenes, especially the dimethyland trimethyl-isomers, indicates organic matter derived 59using aromatic biological markers as a tool for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation, southeastern nigeria mainly from higher plants (strachan et al., 1988). high concentrations of 1, 2, 5and 1, 2, 7trimethylnaphthalene (tmn) in oils have been attributed to aromatization and cleavage of the ring-c of β-amyrin in angiosperms (strachan et al., 1988). 1, 2, 5and 1, 2, 7trimethylnaphthalenes can form as diagenetic products of oleanane-type triterpenoids (chaffee and johns, 1983; chaffee et al., 1984; sonibare et al., 2008). the presence of 1, 2, 5-tmn in the studied samples could be linked to aromatization of oleanane, as indicated by the absence of appreciable amounts of oleanane in the m/z 191 fragmentograms. this compound can also be formed from gymnosperm resins (van aarssen et al., 1999) or from hopanoid precursors (villar et al., 1988). among the tetramethylnaphthalene family, 1, 2, 5, 6tetramethylnaphthalene is the most abundant (figs. 4a-b). strachan et al. (1986) used two related ratios (tde-1 and tde-2) of trimethylnaphthalenes to differentiate coal swamps from marine, lacustrine, and deltaic environments. 1, 2, 7trimethylnaphthalene (tmn) has been used as a marker of angiosperm input. 1, 2, 5-trimethylnaphthalene and 1, 2, 7trimethylnaphthalene can form as diagenetic products of oleanane-type triterpenoids (chaffee and johns, 1983; chaffee et al., 1984). strachan et al. (1988) compared the relative abundance of trimethylnaphthalene isomers in southeast asian samples ranging in age from permian to tertiary. they found that the relative concentrations of the 1, 2, 7-trimethylnaphthalenes were appreciably higher in most oils derived from post-cretaceous source rocks dominated by the input of higher-plants, and in extracts from cretaceous and younger samples, than in samples of older age or with little or no terrigenous plant input. oils of this type yielded ratios of 1, 2, 7-/1, 3, 7-trimethylnaphthalenes from 0.46 to 1.36, while oils from older source rocks containing higher-plant material yielded ratios from 0.34 to 0.16, and oils from marine source rocks yielded ratios from 0.15 to 0.32. in this study, the rock extracts yielded ratios of 1, 2, 7-/1, 3, 7-trimethylnaphthalenes ranging from 0.40 to 1.07 (table 5). the computed tde-1 and tde-2 ratios for the study samples range from 4.41 15.21 and 0.20 0.53, respectively (table 5). the presence of 1, 2, 5and 1, 2, 7-tmn in the rock extracts (figs. 4a-b; table 4) indicate both angiosperm and gymnosperm material contribution to the organic matter that formed the coals and shales (killops and killops, 2005; adedosu et al., 2012). furthermore, the occurrence of appreciable amounts of 1, 6and 2, 6-dmns in all the samples indicate terrestrial organic matter input (achari et al., 1973). additionally, the occurrence of 1, 7-dmp in the analyzed samples (figs. 5a-b) is attributed to terrestrial organic matter (simoneit et al., 1986), while plots of c27,c28 and c29 monoaromatic steranes (fig. 11) indicate that the kerogens were derived from mixed terrestrial and marine organic matter. figure 11: ternary diagram showing the relative contributions of c27, c28 and c29 monoaromatic steranes for rock extracts from the anambra basin. the distribution of the aromatic sulfur compounds including dibenzothiophenes (dbts) and their methyl homologues in the coal and shale extracts are reported in figures 12a-b. the dibenzothiophene/phenanthrene (dbt/phen) ratio together with the pr/ph ratio can be used to infer source rock depositional environments and lithologies (hughes et al., 1995). high dbt/ phen ratios are found in oils from marine carbonates, and low ratios are found in oils from marine shales and most lacustrine rocks (petersen et al., 2007). the coal and shale extracts (table 5) have low dbt/phen ratios (0.01 – 0.06), indicating a strong input of terrigenous type iii organic matter. however, dbt/phen (table 5) and pr/ph ratios only reflect the eh and ph irrespective of depositional environment, and a cross-plot of the two ratios by itself cannot be used to distinguish lacustrine from marine oils (petersen et al., 2007). the presence of organic sulfur compounds and sulfur-containing aromatics in rock extracts and crude oils indicates an anoxic depositional environment (sinninghe damste et al., 1988, 1989a, 1989b; hughes et al., 1995). 1-methyldibenzothiophene (1-mdbt), 2+3-methyldibenzothiophene (2+3-mdbt) and 4-methyldibenzothiophene (4mdbt) were detected in all the samples analyzed (figs. 12a-b; table 6). the high relative abundance of dibenzothiophenes (dbts) in all of the samples is indicative of probable oxic/anoxic conditions prevailing during deposition (table 6). the distribution of dbts has been proposed as an indicator of crude oils derived from carbonate versus siliciclastic sources (hughes, 1984; hegazi and el-gayar, 2009). the presence of a definite v pattern (4-methyl > 2+3-methyl < 1-methyl) in the methyldibenzothiophenes is generally associated with oils from predominantly carbonate source rocks, while a stair-step pattern (4-methyl > 2+3-methyl > 1-methyl) is associated with predominantly siliciclastic source rocks, or oils of advanced maturity (lateto post-oil window) from carbonate sources (hegazi and el-gayar, 2009). furthermore, oils from carbonate sources are characterized by an abundance of benzothiophenes (bts) and a fairly equal distribution of substituted dbts, while siliciclastic oils exhibit low concentrations of bts and decreasing amounts of dimethyldibenzothiophenes and trimethyldibenzothiophenes relative to methyldibenzothiophenes (hughes, 1984; hegazi and el-gayar, 2009). the results obtained in this study indicate that the 4-methyldibenzothiophenes (24.0%) are more abundant than the 2+3-methyldibenzothiophenes (20.4%) and 1-methyldibenzothiophenes (4.4%) (table 6), indicating a siliciclastic source rock input for the organic matter (hegazi and el-gayar, 2009). (a) (b) figure 13a-b: m/z 191 mass chromatogram showing the distributions of benzohopanes in representative shale (a) og/1267/03 and coal (b) eng/1001/12 from the anambra basin. 60 jude e. ogala & mike i. akaegbobi the mass chromatograms (m/z 184 and m/z 198) of the extracts representing the dbts in this study show a high concentration of dbts (51.2%) and decreasing amounts of dimethyldibenzothiophenes (24.0%) and trimethyldibenzothiophene (20.4%) relative to methyldibenzothiophenes (4.4%) (figs. 12a-b and table 6). these results are consistent with previous analyses that indicate that the coal and shale extracts in this study were generated by a siliciclastic source rock (hegazi and el-gayar, 2009). benzohopanes are formed by the cyclization of extended hopanoid side chains followed by aromatization (hussler et al., 1984a, 1984b). benzohopanes range in carbon number from c32 to c35. he and lu (1990) observed that oils and bitumens from evaporitic and carbonate source rocks have the highest concentrations of benzohopanes, which occur in trace amounts in most source rocks and crude oils. benzohopanes were identified in all the rock extracts, but concentrations are low (figs. 13a-b). all samples have similar distributions of benzohopanes (m/z 191). the distribution of benzohopanes (figs. 13a-b) shows that c32 benzohopane is the most abundant member of the bezohopane family. the relative concentrations of the various members of the aromatic hopane series seem to follow that of the regular hopanes in the saturated hydrocarbon fraction, and like the saturated hopanes, they most likely have a bacterial origin (petersen et al., 2007). conclusion the results of this study indicated that the aromatic hydrocarbons contained in the coal and shale samples from the campano-maastrichtian mamu formation of the anambra basin are marginally mature. the kerogens were formed from organic matter of mixed origin (terrestrial and marine). the distributions/concentrations of naphthalenes and phenanthrenes show that trimethylnaphthalene is the most abundant member of the naphthalene family and methylphenanthrene is the most abundant phenanthrene family member. all samples have similar distributions of phenanthrenes. the relative abundance of the total of phenanthrene and its isomers was greater than that of the naphthalenes. thus, the distributions of aromatic hydrocarbons in the rock extracts are influenced by methylation reactions at elevated maturity. the slightly higher concentrations of c29 monoaromatic steranes (49%), compared with c28 (22%) and c27 (29%) monoaromatic steranes, suggests a mixed input of terrestrial and marine organic matter. acknowledgement we are grateful to daniel m. jarvie and the staff of humble geochemical laboratory, texas, usa, for the assistance rendered during the geochemical analyses. we gratefully acknowledge the nigerian geological survey agency (ngsa) for providing the core samples used in this study, and the anonymous reviewers for their helpful comments. table 6: abundance of dibenzothiophenes and their alkyl homologues in rock extracts (concentrations are in ppm) sample number lithology dbt 4mdbt 2+3mdbt 1mdbt og/1267/03 shale 17410.70 8454.20 7230.60 1287.90 eng/1001/02 shale 106.20 40.80 32.60 24.10 eng/1001/12 coal 48.80 13.10 32.00 32.40 eng/1001/16 coal 137.00 74.40 43.80 103.40 eng/1002/06 shale 281.80 116.50 59.00 37.20 eng/1008/13 coal 49.00 28.80 37.00 32.80 ez/1219/08 coal 726.40 63.50 41.60 102.90 abbreviations: dbt – dibenzothiophene, 4mdbt – 4-methyldibenzothiophene, 2+3mdbt – 2, 3methyldibenzothiophene, 1 methyldibenzothiophene references achari, r.g., shaw, g., and holleyhead, r., 1973. identification of ionene and other carotenoid degradation products from the pyrolysis of sporopollenins derived from some pollen exines, a spore coal and the green river shale. chemical geology, volume 12, pp. 229-234. adedosu, t.a., sonibare, o.o., tuo, j., and ekundayo, o., 2012. biomarkers, carbon isotopic composition and source rock potentials of awgu coals, middle benue trough, nigeria. journal of african earth sciences, volumes 66-67, pp. 13-21. akaegbobi, i.m., nwachukwu, j.i., and schmitt, m., 2000. aromatic hydrocarbon distribution and calculation of oil and gas volumes in post-santonian shale and coal, anambra basin, nigeria. in: mello, m.r., and katz, b.j., (eds.), petroleum systems of south atlantic margins. americanassociation of petroleum geologists memoir, volume 73, p.233-245. akande, s.o., ogunmoyero, i.b., petersen, h.i., and nytoft, h.p., 2007. source rock evaluation of coals from the lower maastrichtian mamu foundation, se nigeria. journal of petroleum geology, volume 30, number 4, pp. 303-324. akinlua, a., torto., n., and ajayi., t.r., 2007. oils in the northwestern niger delta: aromatic hydrocarbons content and infrared spectroscopic characterization. journal of petroleum geology, volume 30, number 1, pp. 91-100. akinlua, a., and ajayi, t.r., 2009. chemical characterization of central niger delta oils. journal of petroleum geology, volume 32, number 4, pp. 373-382. alexander, r., kagi, r.i., roland, s.j., sheppard, p.n., and chirila, t.v., 1985. the effects of thermal maturity on distributions of dimethylnaphthalenes and trimethylnaphthalenes in some ancient sediments and petroleum. geochimica cosmochimica acta, 49, pp. 385-395. alexander, r., cumbers, a.m., and kagi, r.i., 1986. alkylbiphenyls in ancient sediments and petroleums. organic geochemistry, 10, pp. 841-845. allan, j., and larter, s.r., 1981. aromatic structures in coal maceral extracts and kerogens. in: bjoroy, m. et al., (editors), advances in organic geochemistry, john wiley and sons ltd., new york, pp. 534-545. anderson, k.b., winans, r.e., and botto, r.e., 1992. the nature and fate of natural resins in the geosphereii: identification, classification and nomenclature of resinites. organic geochemistry, volume 18, pp. 829-841. chaffee, a.l., and johns, r.b., 1983. polycyclic aromatic hydrocarbons in australian coals: i. angularly fused pentacyclic triand tetraaromatic components of victorian brown coal. geochimica et cosmochimica acta, 47, pp. 2141-2155. chaffee, a.l., strachan, m.g., and johns, r.b., 1984. polycyclic aromatic hydrocarbons in australian coal: ii. novel tetracyclic components from from victorian brown coal. geochimica et cosmochimica acta, 48, pp. 2037-2043. 61using aromatic biological markers as a tool for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation, southeastern nigeria cumbers, k.m., alexander, r., and kagi, r.i., 1986. methylbiphenyl, ethylbiphenyl and dimethylbiphenyl isomer distributions in some sediments and crude oils. geochimica cosmochimica acta, volume 51, pp. 3105-3112. el-gayar, m.sh., 2005. aromatic steroids in mideastern crude oils: identification and geochemical application. petroleum science technology, volume 23, pp. 971-990. garrigues, p., de sury, r., angelin, m.l., bellocq, j., oudin, j.l., and ewald, m., 1988. relation of methylated aromatic hydrocarbon distribution pattern to the maturity of organic matter in ancient sediments from the mahakam delta. geochimica cosmochimica acta, volume 52, pp. 375-384. garrigues, p., oudin, j.l., parlanti, e., monin, j.c., robcis, s., and bellocq, j., 1990. akylated phenanthrene distribution in artificially matured kerogens from kimmeridge clay and brent formation (north sea). organic geochemistry, volume 16, pp. 167-173. hall, p.b., schou, l., and bjoroy, m., 1984. aromatic hydrocarbon variations in north sea wells. in: thomas, b.m. et al., (editors) petroleum geochemistry in exploration of norwegian shelf, norwegian petroleum society, graham and trotman ltd., london, pp. 293-301. hegazi, a.h., and el-gayar, m.sh., 2009. geochemical characterization of a biodegraded crudeoil, assran field, central gulf of suez. journal of petroleum geology, volume 32, number 4, pp. 343-355. hill, r.j., jarvie, d.m., zumberge, j., henry, m., and pollastro, r.m., 2007. oil and gas geochemistry and petroleum systems of the fort worth basin. american association of petroleum geologist bulletin, volume 91, number 4, pp. 445-473. hughes, w.b., holba, a.g., and dzou, l.i.p., 1995. the ratios of dibenzothiophene to phenanthrene and pristane to phytane as indicators of depositional environment and lithology of petroleum source rocks. geochimica cosmochimica acta, volume 59, pp. 3581-3598. johns, r.b., 1986. biological markers in the sedimentary records. elsevier, new york, 460p. killops, s.d., and killops, v.j., 2005. introduction to organic geochemistry. 2nd edition, blackwell science publishing limited. mackenzie, a.s., hoffmann, c.f., and maxwell, j.r., 1981. molecular parameters of maturation in the toarcian shales, paris basin, franceiii: changes in aromatic steroid hydrocarbons. geochimica cosmochimica acta, volume 45, pp. 1345-1355. mello, m.r., gaglianone, p.c., brassel, s.c., and maxwell, j.r., 1988a. geochemical and biological marker assessment of depositional environments using brazilian offshore oils. marine and petroleum geology, volume 5, pp. 205-223. mello, m.r., telnaes, n., gaglianone, p.c., chicarelli, m.i., brassel, s.c., and maxwell, j.r., 1988b. organic geochemical characterization of depositional environments in brazilian marginal basins. organic geochemistry, volume 13, pp. 31-46. mohammed, y., 2005. predictive petroleum system model of prospective anambra basin, nigeria.nigerian association of petroleum explorationists/american association of petroleum geologists abstracts (nape/aapg) publication, pp.70. moldowan, j.m., and fago, f.j., 1986. structure and significance of a novel rearranged monoaromatic steroid hydrocarbon in petroleum. geochimica cosmochimica acta, volume 50, pp. 343-351. murat, r.c., 1972. stratigraphy and palaeogeography of the cretaceous and lower tertiary in southern nigeria. in: t.f.j. dessauvagie and a.j. whiteman (editors), african geology. university of ibadan, nigeria, pp. 251-266. onuoha, m.k., 1986. basin subsidence, sediment decomposition and burial history modeling techniques: applicability to the anambra basin. nigerian association ofpetroleum explorationists bulletin, volume 2, number 1, pp. 6-17. peters, k.e., and moldowan, j.m., 1993. the biomarker guide: interpreting molecular fossils in petroleum and ancient sediments. englewood cliffs, new jersey, prentice hall, 363p. peters, k.e., walters, c.c., and moldowan, j.m., 2005. the biomarker guide, 2nd edition. cambridge university press, new york, 1155p. petersen, h.i., nytoft, h.p., ratanasthien, b., and foopattthanakamol, a., 2007. oils from cenozoic rift-basins in central and northern thailand: source and thermal maturity. journal petroleum geology, volume 30, number 1, pp. 59-78. petters, s.w., 1978. stratigraphic evolution of the benue trough and its implications for upper cretaceous paleogeography of west africa. journal of geology, volume 86, pp. 311322. petters, s.w., and ekweozor, c.m., 1982. petroleum geology of benue trough and southeastern chad basin, nigeria. american association of petroleum geologists bulletin, volume 66, pp. 1141-1149. puttman, w., and villar, h., 1987. occurrence and geochemical significance of 1, 2, 5, 6tetramethylnaphthalene.geochimica cosmochimica acta, volume 51, pp. 3023-3029. radke, m., 1987. organic geochemistry of aromatic hydrocarbons. in: j. brooks and d. welte, (editors) advances in petroleum geochemistry, academic press, new york, pp. 141-207. radke, m., 1988. application of aromatic compounds as maturity indicators in source rocks and crude oils. marine and petroleum geology, volume 5, pp. 224-236. radke, m., and welte, d.h., 1983. the methylphenanthrene index (mpi): a maturity parameterbased on aromatic hydrocarbons. in: m. bjoroy, c. albrecht, c. cornford, et al., (editors), advances in organic geochemistry, john wiley and sons, new york, pp. 504 512. radke, m., willsch, h., and welte, d.h., 1980. preparative hydrocarbon group type determination by automated medium pressure liquid chromatography. analytical chemistry, volume 52, pp. 406-411. radke, m., and welte, d.h., and willsch, h., 1982a. geochemical study on a well in the western canada basin: relation of the aromatic distribution pattern to maturity of organic matter. geochimica cosmochimica acta, volume 46, pp.1-10. radke, m., willsch, h., leythaeuser, d., and teichmuller, m., 1982b. aromatic components of coals: relation of distribution pattern to rank. geochimica cosmochimica acta, volume 46, pp.1831-1848. radke, m., willsch, h., and welte, d.h., 1984. class separation of aromatic compounds in rock extracts and fossil fuels by liquid chromatography. analytical chemistry, volume 56, pp. 2538-2546. radke, m., and welte, d.h., and willsch, h., 1986. maturity parameters based on aromatic hydrocarbons: influence of the organic matter type. organic geochemistry, volume 10, pp. 51-63. reyment, r.a., 1965. aspects of geology of nigeria. university of ibadan press. 145p. rubenstein, i., sieskind, o., and albrecht, p., 1975. rearranged steranes in a shale: occurrence and simulated formation. journal of chemical society, perkin transactions i, pp. 18331836. rullkotter, j., peakman, t.m., and ten haven, h.l., 1994. early diagenesis of terrigenous triterpenoids and its implications for petroleum geochemistry. organic geochemistry, volume 21, pp. 215-233. seifert, w.k., and moldowan, j.m., 1978. applications of steranes, terpanes, and monoaromatics to the maturation, migration, and source of crude oils. geochimica et cosmochimica acta, volume 43, pp. 77-95. sieskind, o., joly, g., and albrecht, p., 1979. simulation of the geochemical transformation of sterols: superacid effects of clay minerals. geochimica cosmochimica acta, volume 43, pp. 1695-1659. simoneit, b.r.t., 1986. cyclic terpenoids of the geosphere. in: r.b. johns (editors), biological markers in the sedimentary records, elsevier science publishers, amsterdam, pp. 43-99. sinninghe damste, j.s., rijpstra, i.c., de leeuw, j.w., and schenck, p.a., 1988. origin of organic sulfur compounds and sulfur-containing high molecular weight substances in sediments and immature crude oils. organic geochemistry, volume 13, pp. 593-606. 62 jude e. ogala & mike i. akaegbobi sinninghe damste, j.s., van koert, e.r., kock-van dalen, a.c., de leeuw, j.w., and schenck, p.a, 1989a. characterisation of highly branched isoprenoid thiophenes occurring in sediments and immature crude oils. organic geochemistry, volume 14, pp. 555-567. sinninghe damste, j.s., rijpstra, i.c., de leeuw, j.w., and schenck, p.a., 1989b. the occurrence and identification of series of organic sulfur compounds in oils and sediment extracts ii: their presence in samples from hypersaline and non-hypersalinepaleoenvironments and possible application as source, paleoenvironmental and maturity indicators. geochimica cosmochimica acta, volume 53, pp. 1323-1341. sonibare, o.o., adedosu, t., ekundayo, a.o., and jarvie, d.m., 2008. hydrocarbon potential and organic geochemistry of coals from benue trough, nigeria. journal of applied sciences research, volume 4, number11, pp. 1511-1520. strachan, m.g., alexander, r., and kagi, r.i., 1986. trimethylnaphthalenes as depositional environmental indicators. presented at the 192nd annual american chemical society meeting, anaheim, california. strachan, m.g., alexander, r., and kagi, r.i., 1988. trimethylnaphthalenes in crude oils and sediments: effects of source and maturity. geochimica cosmochimica acta, volume 52, pp. 1255-1264. tissot, b.p., and welte, d.h., 1984. petroleum formation and occurrence. 2nd edition, springer verlag, berlin, 699p. uzuakpunwa, a.b., 1974. the abakaliki pyroclastics eastern nigeria: new age and tectonic implications. geological magazine, volume 111, pp. 65-70. van aarssen, b.g.k., fisher, s.j., bastow, t.p., and alexander, r., 1999. distribution of methylated naphthalenes in crude oils: indicators of maturity, biodegradation and mixing organic geochemistry, volume 30, pp. 1011-1021. villar, j.h., puttmann, w., and wolf, m., 1988. organic geochemistry and petrography of tertiary coals and carbonaceous shales from argentina. organic geochemistry, volume 13, pp. 1011-1021. volkman, j.k., alexander, r., kagi, r.i., and woodhouse, g.w., 1983. demethylated hopanes in crude oils and their applications in petroleum geochemistry. geochimica et cosmochimica acta, volume 47, pp. 785-794. whiteman, a., 1982. nigeria: its petroleum geology, resources and potentials, volumes i and ii. grahamand trotman, 394p. palabras clave: carretera de minas; detección avanzada; análisis de polarización; imágenes finas. how to cite item wang, b., liu, s., jin, b., & qiu, w. fine imaging by using advanced detection of reflected waves in underground coal mine. earth sciences research journal, 23(1), 93-99. doi: https://doi.org/10.15446/esrj.v23n1.78750 safety of the mine roadway constructions is controlled by geological disasters such as faults, goaves and so on. the advanced prediction has become an in-demand topic, and advanced detection method of the reflected wave is a crucial technology for advanced prediction of geological anomalies. however, due to the influence of the complex near-source seismic wavefield in the coal mine roadway, the result of wavefield separation and migration imaging is not accurate, which lead to the fact that the fine imaging of geological anomaly is difficult. a comparative analysis of wavefield separation method of kinematics and dynamics is carried out in this paper to solve this problem. a factor of principal polarization direction is introduced according to “the orthogonal difference between the propagation direction of p-wave and s-wave and the vibration direction of particle” starting from the real-time polarization analysis from three-component seismic signals. by the factor, a modified function is constructed and integrated into pre-stack diffraction migration, which put forward the polarization migration method that incorporates wavefield separation with migration imaging. the results of physical simulation and field survey in xiangyuan coal mine in western china are as follows: the fine imaging by using advanced detection of the reflected waves in underground coal mine can be achieved by using polarization migration in linear observation system and the polarization migration has the effect of enhancing the spatial resolution. abstract keywords: magma; mine roadway; advanced detection; polarization analysis; fine imaging. fine imaging by using advanced detection of reflected waves in underground coal mine imágenes finas mediante el uso de detección avanzada de ondas reflejadas en una mina subterránea de carbón issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n1.78750 la seguridad de las construcciones de carreteras de mina está controlada por desastres geológicos, como fallas, obstrucciones, etc. la predicción avanzada se ha convertido en un tema de demanda, y el método de detección avanzada de la onda reflejada es una tecnología crucial para la predicción avanzada de anomalías geológicas. sin embargo, debido a la influencia del complejo campo de ondas sísmicas de fuente cercana en la carretera de la mina de carbón, el resultado de la separación de los campos de onda y la imagen de migración no es preciso, lo que lleva al hecho de que es difícil obtener imágenes finas de anomalías geológicas. en este trabajo se realiza un análisis comparativo del método de cinemática y dinámica de separación de campos de onda para resolver este problema. un factor de la dirección de polarización principal se introduce según "la diferencia ortogonal entre la dirección de propagación de la onda p y la onda s y la dirección de vibración de la partícula" a partir del análisis de polarización en tiempo real desde señales sísmicas de tres componentes. por el factor, una función modificada se construye e integra en la migración de difracción de pila previa, lo que presenta el método de migración de polarización que incorpora la separación del campo de onda con la imagen de migración. los resultados de la simulación física y el estudio de campo en la mina de carbón xiangyuan en el oeste de china son los siguientes: la imagen fina mediante el uso de detección avanzada de las ondas reflejadas en una mina de carbón subterránea se puede lograr mediante el uso de la migración de polarización en un sistema de observación lineal y la migración de polarización tiene el efecto de mejorar la resolución espacial. resumen record manuscript received: 30/05/2018 accepted for publication: 01/12/2018 earth sciences research journal earth sci. res. j. vol. 23, no. 1 (march, 2019): 93-99 bo wang, shengdong liu*, biao jin, wanyong qiu state key laboratory of deep geomechanics & underground engineering and school of resource and earth science, china university of mining and technology, xuzhou, 221116, china * corresponding author: wbsyes@126.com g eo te c h n ic a l en g in ee r in g 94 bo wang, shengdong liu*, biao jin, wanyong qiu introduction in recent years, as the coal mine is gradually entering the deep mining stage, higher requirements of advanced detection in tunnel face are put forward due to geological environment changes such as high geostress, high temperature, high karsts water pressure, mining strong disturbance and so on (liu, liu & yue, 2014; cheng, li, peng, & sun 2014). in chinese coal mine disaster accidents, according to statistics, more than 90% of the accidents occur in the process of roadway constructions, so it is very necessary to carry out the effective advanced prediction of the geological anomalies (zhang, & hu 2015; liu & wang, 2016). the advanced detection method of the reflected waves is used to predict the geological conditions of the advanced tunneling zone of the roadway by using the reflections generated from seismic waves’ propagation in inhomogeneous geological structures (lüth, buske, giese, & goertz, 2005; bohlen et al., 2007). the advanced detection methods of the reflected waves in mine roadway are as follows: shen (2006) proposed seismic prediction method of laneways, which is to achieve the imaging according to the superposition principle of diffraction and the principle of mirror imaging. liu, guo & zhang (2006); zhang & wu (2006), y wang et al. (2016) proposed mine seismic prediction method, which adopts the roadway observation system with high covering number to acquire seismic data. these data are processed by the methods of energy synthesis and decomposition and migration superposition to predict a variety of geological structures within 120m ahead of the tunnel face. he (2008) developed mine multi-component seismic data processing system to provide the support for the seismic advanced detection with multiwave. liang & song (2009) put forward the multi-component advanced detection method with multiwave, which can detect the situations such as rock broken zone etc, within 100m ahead of the tunnel face (reyes-mendoza et al., 2016). in addition, the advanced detection methods of reflected waves in traffic tunnel are sometimes used in underground coal mine roadway, mainly including seismic negative apparent velocity method (zeng, 1994), land sonar method (zhong, 2012), tunnel seismic prediction method (dickmann & sander, 1996), and the japanese oyo company (inazaki, 1999), the german gfz company (borm, giese, & otto, 2001), american nsa company (otto, buron, bretterebner, & schwab, 2002) also developed seismic prediction system based on the reflected waves method in the tunnel. the advantages of these seismic reflected waves method include long detection range, high resolution, abundant information and being sensitive to geological anomaly. due to the limitation of narrow observation space of the roadway and the interference of the near-source seismic wave field (cheng et al., 2014), the result of wavefield separation and migration imaging is not accurate, which lead to the fact that the fine imaging of geological anomaly is difficult (wang et al., 2016). based on this problem, the polarization migration that integrates wavefield separation with migration imaging in full space condition is put forward in this paper and the fine imaging is achieved by using the advanced detection method of reflected waves through the physical simulation and the field survey. wavefield separation the advanced detection method of reflected wave is carried out with sources and receivers in rock or coal seam. the method have the characters of broadband excitation and three components reception, which provides the possibility for seismic exploration with high resolution and multiwave attributes, while due to the limited offset, the complex seismic information makes the data processing be difficult. in multiwave seismic exploration, the data contains p-wave, s-wave, converted wave, surface wave, acoustic wave and so on. effective wavefield separation is the key to data processing. the common wavefield separation methods include: linear radon transform filtering, frequency wavenumber domain filtering, polarization filtering, etc.  − p filtering  − p filtering is to transform seismic records from time-spatial  −( )x domain into  − p domain and separate the effective wave energy in  − p domain, and then transform it into time-spatial  −( )x domain again. the integral formulas of radon transformations can be defined as: ψ τ ϕ, ,p t px x dx( ) = = +( ) −∞ +∞ ∫ τ (1) ϕ τ ψ, ,x t px p dx( ) = = −( ) −∞ +∞ ∫ τ (2) f k− filtering f k− filtering is the method to carry out wave field separation by using the difference of spectrum and apparent velocity of effective wave and interference wave. the effective waves of frequency band f f1 2− (frequency filtering) and apparent velocity v v1 2− (apparent velocity filtering) are reserved, and the interference wave are suppressed when v v v v f f< > <1 2 1, , , and f f> 2. two-dimensional fourier transformations are as follows: g f k t x e dtdxi fi kx, ,( ) = ( ) −∞ +∞ −∞ +∞ − +( )∫∫ ϕ π2 (3) ϕ t x g f k e dtdxi fi kx, ,( ) = ( ) −∞ +∞ −∞ +∞ − +( )∫∫ 2π (4) polarization filtering polarization filtering is also known as spatial directional filtering which is used to filter pand s-waves by using the difference of polarization characteristics. pand s-waves are represented as linear orthogonal polarization, while the polarization direction of p-wave is parallel with the propagation direction of wave, and polarization directions of s-wave are perpendicular to the propagation direction of waves. comparative analysis kinematic separation method of pand s-waves is based on the relatively clear phase axes of pand s-waves in the time domain. however, the actual data, especially underground roadway seismic records are complex. various types of waves are intertwined, including information of positive apparent velocity features such as roof, floor, rear and so on, thus phase axes of pand s-waves with clear negative apparent velocity are rarely, so it is difficult to separate pwaves and s-waves simply by using  − p filtering. a combination of kinematics apparent velocity features and dynamics frequency features is used by f k− filtering to separate pand s-waves. on the one hand, because of the lower frequency of surface waves and s-waves, they could not be separated effectively in the separation process because the frequency between surface wave and s-wave is close. on the other hand, when p-wave is obtained through high-pass filtering, it easily leads to the fact that some effective s-wave is lost. because the polarization filtering is not the band-pass filtering of frequency domain, the bandwidth of effective waves is not affected, and the polarization filtering is to use mathematical statistical method to carry out wavefield separation on basis of the polarization difference of pand s-waves, which makes the polarization filtering method become the commonly used separation method for pand s-waves. however, in the process of calculating eigenvalues and eigenvectors and setting filtering function, corresponding relations between xyz coordinate axes and p, sh, sv directions are usually artificially, without considering the corresponding relationship between the principal polarization direction of real-time particle in each receiver and propagation direction of seismic wave, which makes the results of unclean separation of pand s-waves in the roadway. therefore, through the real-time computation of principal polarization directions of particles in each receiver, starting from the difference between real-time particle vibration directions and pand s-waves propagation direction, 95fine imaging by using advanced detection of reflected waves in underground coal mine setting pand s-waves separation filter should be an effective method to extract p-wave or s-wave. migration migration imaging method can be divided into ray and wave equation migration. the ray migration is mainly as the prestack diffraction migration, which is based on huygens principle. wave equation migration mainly includes kirchhoff integral migration method, reverse-time migration method and so on. however, wave equation migration like reverse-time migration, etc., has lower calculation efficiency, so the prestack diffraction migration is mainly adopted for migration imaging of mine roadway seismic exploration. according to huygens principle, each reflection point g in the grid can be regarded as a secondary source, while carrying out diffraction scan migration, any discrete grid point should be regarded as a reflection point, and travel time of reflected wave or diffraction wave is as follows: t l v l h x x h x xi j i j i j r g g si j, , ,,= = + −( ) + + −( )2 2 2 2 (5) where i =1, 2, 3,...... m, and m is the recording trace for superposition calculation; v is the seismic velocity, h is the vertical depth of the scanning point gi j, , while ti j, is the travel time of diffraction wave of the j th shot in the i th receiver at the scanning point gi j, . the amplitude value ai of signal in different traces in the time of ti are carried out with superposition to get the amplitude value of gi j, point. dynamic polarization migration imaging principle as shown in figure 1, x direction denotes tunnel face, y direction denotes the left side and z direction denotes the roof. the grid division is carried out with the advanced space of the roadway, and the black mark is a hypothetical reflection point in the spatial grid; the same side behind tunnel face is set up with several sources and receiver. assuming that the propagation time is represented as t, and the propagation path is shown as the ray direction in figure; in the spatial coordinate system, the incident path is as l1 and the propagation time is as t1; the reflection path is l2 and the propagation time is as t2; the ray path of l1 and l2 in homogeneous space can be calculated by the spatial coordinates among source, the reflection point and the receiver. according to the spatial position of source, reflection point and receiver and spatial velocity parameters, based on ray tracing technology, it can calculate the total time t = t1 + t2. based on three-component amplitude at t time, particle spatial vibration direction li at t time of a receiver can be calculated by polarization analysis method. the angle  can be calculated by using particle spatial vibration direction li and the ray direction of reflection path l2 (as shown in fig. 2), and then based on the difference of angle  between p-wave and s-wave, the filter is set up in real time to realize dynamic polarization filtering. the dynamic filtering computation is integrated into prestack migration. polarization analysis hilbert transform polarization analysis method is a kind of complex number field polarization analysis method based on hilbert transform to construct the complex covariance matrix and then to obtain principal polarization parameters. among them, taking the two-dimensional space as an example, the calculation steps of the principal polarization direction li at the t time are as follows: hilbert transform is calculated: hx t x t ju x t( ) = ( ) + ( )( ) ⋅⋅ (6) hz t z t ju z t( ) = ( ) + ( )( ) ⋅⋅ (7) where x(t) and z(t) are seismic records of components x and z respectively, while the symbol u represents hilbert transform, j = 1. hermitian matrix is constructed: c t m t m t( ) = ∗ ( ) ⋅ ( ) (8) where, m t hx t hz t( ) = ( ) ( )  , the symbol * represents complex conjugate transpose of matrix. the maximum eigenvalue max of the matrix and its normalized corresponding eigenvector x zmax max,( ) are obtained. the principal polarization angle  is calculated: α=arctan re re max max z x ( ) ( )         (9) where re maxx( ) and re maxz( ) are respectively as the real parts of xmax and zmax,  indicates the main polarization direction li. polarization migration prestack depth migration is often expressed as a diffraction integral in a scalar wave field. on the basis of integral migration, a modified function receiver source advanced direction tunnel face reflection point ... y l2 l1 z x figure 1. polarization migration imaging particle trajectory particle vibration direction ray directiony l2 l1 z x  figure 2. particle spatial vibration direction 96 bo wang, shengdong liu*, biao jin, wanyong qiu is introduced into polarization migration, and the migration formula can be defined as: u f t a t dx n = ( ) ( )∫1 (10) in the above equation, f t q( ) = cos , f t( ) is a modified function of amplitude value for any reflection point during p-wave imaging (sinq  for s-wave imaging), and a t( ) is the amplitude. the acute angle  represents the angle between the ray direction from any reflection point and the principal polarization direction. the q is valued in a range of 1 to 10. physical simulation in order to simulate the seismic advanced detection field with the size of 160m * 160m * 80m (as shown in figure 3), the experimental physical model is designed according to scaling factors in table 1 with the size of 1.6m * 1.6m * 0.8m. the model velocity parameters are consistent with the actual detection field, and the sampling frequency is 1mhz, namely that the sampling interval is 0.001ms, while receiver line layout is also scaled down by 1: 100. figure 3. physical model and detection diagram yzx 0 5 10 15 20 25 30 35 40 45 trace(no.) 60 ti m e(  s. ) 45 0 30 0 15 0 0 figure 4. seismic data the advanced linear observation system is shown in figure 3-b. the excitation transducer slides on the receiver line to simulate excitation with multiple shots according to the moving step of 0.01m, and the position of threecomponent receiving transducer is fixed. in the process of the experiment, the offset trace is 0.059m with the total excitation of 21 times, and the sampling frequency is 1mhz, namely that the sampling interval is 0.001ms. the threecomponent ultrasonic data with 21 shot are collected in the experiment. the components of x, y and z are respectively of 21 channels. the interference of reflected waves from the sink sides and bottom is avoided effectively in the process of designing model, so the original signals are mainly direct waves and reflected waves from abnormal interfaces. figure 4 is obtained through  − p transform and f k− transform with the combination of filtering. from the signals of component x in figure 4, four groups of reflected waves can be observed, respectively as a, b, c and d, which are all performed with negative apparent velocity characteristics. from the analysis of the propagation time, a and b are respectively as reflected p-wave, whose time length of the signals between 1st to 21st shot is within 100-200 and 150-250s, while c and d are respectively as reflected s-wave, whose time length is within 97fine imaging by using advanced detection of reflected waves in underground coal mine y z x 0 0 0 280 280 280 -280 amplitude (mv) figure 5. polarization trajectory 200-380 and 260-440s. among them, signals of c and d in wave transducers of y and z are more obvious. the three-component signal of the 21st shot of a reflected p-wave is selected to make the polarization diagram (figure 5). the angle between projection of the principal polarization direction of a reflected p-wave in the xz plane and the x axis is 10° (it should be 0 degree in theory and there is a 10 degree error). the principal polarization direction in xz plane is parallel with the wave propagating direction (wave propagating direction is represented by a yellow arrow), while the principal polarization direction in xy plane is parallel -1 7. 0 0.0 r1 distance(cm) r2 30.0 40.0 50.0 60.0 0. 00 17 .0 20.0 figure 6. polarization migration profile with the x-axis, and the principal polarization direction in yz plane is nearly parallel with the z-axis. thus, it can be deduced that the principal energy is in the xz plane, which is parallel with the receiver line (along the x axis). those features conform to the characteristics of p-p wave. imaging by using polarization migration is shown in figure 6. it is analyzed from the figure that there are two groups of appear abnormal interfaces around in 0.2m and 0.32m, respectively as r1 and r2, where 0.32m is the designed model interface between the concrete and saturated fine sand, while the 0.2m is cracking zone of the model with developed fracture. similarly, three-dimensional polarization migration is carried out in order to describe the spatial morphology of the interface r2 after filtering energy anomaly signal of r1. figure 7 is the spatial profile of r2 energy anomaly interface. field survey the multiwave seismic instrument was adopted to record seismic data. a three-component geophone arranged in left side of open-off cut roadway of 1063 working face. there are 29 sources in left side and the spacing of two sources is 1m. figure 8 is the site survey layout. the sampling length is 2048 points and the sampling interval is 0.08ms, finally the 29 shots seismic data were recorded. spectrum analysis shows that the main frequency of the data is within 100-400hz. the velocity of p-waves was obtained as 2200m/s based on the first break picking. the advanced reflected waves were acquired through  − p and f k− transform. two groups of negative apparent velocity characteristics can be analyzed from the figure, which are respectively located at 30-50ms, 70-90ms. figure 9 was obtained by using polarization migration. it can be seen clearly from figure 15 that there is an abnormal energy interface r1 at 47m ahead of tunnel face, while the energy near the tunnel face is obviously abnormal. combine with field exploration data, there is a water tank near the tunnel face, so the interpretation of r1 interface is the goaf boundary. figure 10 indicates the result of prestack diffraction migration, and there exist energy anomaly of 0-30m, 40-60m, 75m in the front. pand s-waves could not be effectively separated without the polarization migration, thus the imaging resolution is low, which can not effectively determine location of goaf in the front. verified by field drilling, it is confirmed that the location of advanced goaf boundary is located at about 49m, which is close to the detection result and the error is 2m. 98 bo wang, shengdong liu*, biao jin, wanyong qiu figure 8. layout of field survey interface 50 r2 40 60 30 20 10 10 0 -10 -10 -20 -30 x z y 0 53684.2 41052.6 28421.1 15789.5 3157.89 -9473.68 -22105.3 -34736.8 -47368.4 -60000 figure 7. three-dimensional polarization migration result figure 9. two-dimensional polarization migration 99fine imaging by using advanced detection of reflected waves in underground coal mine conclusion on the basis of the three-component seismic data, the real-time polarization analysis was carried out. based on the difference of polarization parameters, extraction of advanced seismic reflection wavefield were carried out with the combination of  − p and f k− filtering. dynamic polarization migration imaging method was proposed in this paper, and wavefield separation of effective wave was completed in the migration imaging process, which solves the problem of multiple abnormal interfaces when detecting the complex wave field of the roadway. the method effectively improves the spatial resolution and detection precision. acknowledgements this research is supported by “the fundamental research funds for the central universities” (2017qnb11). references bohlen, t., lorang, u., rabbel, w., müller, c., giese, r., lüth, s., & jetschny, s. (2007). rayleigh-to-shear wave conversion at the tunnel face from 3d-fd. geophysics, 72(6), 67-79. borm, g., giese, r., & otto, p. (2001). integrated seismic imaging system for geological prediction ahead in underground construction. proceedings-rapid excavation and tunneling conference. cheng, j. l., li, f., peng, s. p., & sun, x. y. (2014). research progress and development direction on advanced detection in mine roadway working face using geophysical methods. journal of china coal society, 39(8), 1742-1750. dickmann, t. &, sander, b. k. (1996). drivage concurrent tunnel seismic prediction (tsp): result from vereina north tunnel mega-project and piora pilot gallery. feisbau, 1996(6), 406-411. he, z. y. (2008). design and application of data processing system for multi component seismic exploration in coal mine. beijing: china university of mining and technology (beijing). inazaki, t. (1999). stepwise application of horizontal seismic profiling for tunnel prediction ahead of the face. the leading edge, 18(12), 1429-1430. liang, q. h. &, song, j. (2009). advanced detection theory and experimental research of multi-component seismic exploration in mine. journal of central south university (science and technology), 40(5), 1392-1398. liu, s. d., guo, l. q., & zhang, p. s. (2006). experiment and application study of fore detecting by msp method for geological structure in laneway. chinese journal of engineering geophysics, 3(6), 437-442. liu, s. d., liu, j., & yue, j. h. (2014). development status and key problems of chinese mining geophysical technology. coal science and technology, 39(1), 19-25. liu, s. d. &, wang, b. (2016). mine seismic method and technology. xuzhou: china university of mining and technology press. lüth, s., buske, s., giese, r., & goertz, a. (2005). fresnel volume migration of multicomponent data. geophysics, 70(6), 121-129. otto, r., buron, e., bretterebner, h., & schwab, p. (2002). the application of trt at the unterwald tunnel. felsbau, 20(2), 51-56. reyes-mendoza, o., marino-tapia, i., herrera-silveira, j., ruiz-martinez, g., enriquez, c., & largier, j. l., (2016). the effects of wind on upwelling off cabo catoche. journal of coastal research, 32(3), 638-650. shen, h. y. (2006). reflected-wave tunnel seismic prediction. xian: changan university. wang, b., liu, s. d., zhou, f. b., lu, t., huang, l. y., & gao, l. j. (2016). polarization migration of three-component reflected waves under small migration aperture condition. acta geodynamica et geomaterialia, 13(1), 1-12. zeng, z. h. (1994). prediction ahead of the tunnel face by the seismic reflection methods. chinese journal of geophysics, 37(2), 268-271. zhang, p. s. &, hu, x. w. (2015). research status on technology of advanced detection by electromagnetic methods in mine laneway. coal science and technology, 43(1), 112-115, 119. zhang, p. s. &, wu, j. s. (2006). research and analysis of forward prediction technology using seismic reflection wave in tunnel and laneway in china. advances in earth science, 21(10), 1033-1038. zhong, s. h., sun, h. z., & wang, r. (2012). land sonar method. beijing: science and technology of china press. figure 10. two-dimensional prestack diffraction migration keywords: casing damage; shear deformation; numerical simulation; critical strain; early warning. palabras clave: tubería de revestimiento; deformación por cizallamiento; simulación numérica; tensión crítica; alerta temprana. how to cite item wang, s., wang, x., dong, k., jiang, m. chen, m., xie, w., zhang, z., & yang, l. (2020). shear damage mechanism and early warning study of shale standard layer casing. earth sciences research journal, 24(2), 225-229. doi: https:// doi.org/10.15446/esrj.v24n2.87932 aiming at the problem of shear-induced casing damage caused by standard layer shale soaking water, the theory of rock mechanics is used to establish the fluid-solid coupling calculation model of the formation-cement ring-casing of the standard interval shale interbed. it allows obtaining the strain and deformation changes of casing shear damage, and it puts forward the casing logarithmic strain exacerbated area should prompt adoption of reasonable measures to slow down the casing damage. three-level orthogonal numerical simulation experiments were carried out for the five factors affecting casing damage. the simulation results show the critical strain data of the casing damage and the true strain data of the casing under the influence of five factors. according to the statistical principle, the predictive mathematical model of the critical strain and true strain prediction of the casing is obtained by the multiple regression method. the prediction of the strain state of the casing under actual working conditions is realized, and then the casing damage early warning was also realized. abstract shear damage mechanism and early warning study of shale standard layer casing mecanismo de daño por cizallamiento y estudio de alerta temprana de revestimiento de capa de lutita estándar issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87932 con el fin de abordar el problema del daño de la tubería de revestimiento inducido por el corte causado por el agua de remojo en capas de lutita estándar, la teoría de la mecánica de rocas se utiliza para establecer el modelo de cálculo de acoplamiento fluido-sólido de la envoltura de anillo de cemento de formación del intervalo estándar de esquisto intercalado. esta permite obtener los cambios de tensión y de deformación del daño por cizalladura de la tubería de revestimiento, y presenta el área exacerbada por deformación logarítmica del revestimiento. también, debe impulsar la adopción de medidas razonables para reducir el daño de la tubería de revestimiento. en este trabajo se llevaron a cabo experimentos de simulación numérica ortogonal de tres niveles para los cinco factores que afectan el daño de la tubería. los resultados de la simulación muestran los datos críticos de tensión del daño de la tubería y los datos reales de tensión de esta bajo la influencia de los cinco factores. de acuerdo con el principio estadístico, el modelo matemático predictivo de la deformación crítica y la predicción de deformación verdadera de la tubería se obtiene mediante el método de regresión múltiple. en este estudio se realiza la predicción del estado de deformación de la tubería de revestimiento en condiciones de trabajo reales y luego se analiza la alerta temprana en daños de la tubería. resumen record manuscript received: 12/03/2019 accepted for publication: 17/03/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 225-229 suling wang1, xuefei wang1,2, minzhan chen1, kangxing dong1, minzheng jiang1*, lei yang3 1school of mechanical science and engineering, northeast petroleum university, daqing 163318, china 2school of mechatronics engineering, qiqihar university, qiqihar 161001, china 3ningde times new energy technology co. ltd. ningde 352000, china * corresponding author: jmz1964@126.com pe t r o l e u m e n g in e e r in g https://doi.org/10.15446/esrj.v24n2.87932 https://doi.org/10.15446/esrj.v24n2.87932 https://doi.org/10.15446/esrj.v24n2.87932 226 suling wang, xuefei wang, minzhan chen, kangxing dong, minzheng jiang, lei yang introduction by the end of 2016, the cumulative number of wells drilled in a certain area of daqing oilfield was 20,349, and the casing was damaged in 5,715 wells, accounting for 28.1% of the number of wells. among them, the number of casing damage wells in a standard layer is 2325, accounting for 40.6% of the total number of casing damage wells. the casing damage in the non-oil layer is the most harmful to oilfield development. data shows that the two peaks of the casing damage are concentrated in the standard layer shale parts. the casing damage wells continue to increase and have caused millions or even billions of economic losses to the oilfield. therefore, prevention of casing damage is imminent. the qra method and lrfd method for reliability evaluation the strength of casing was proposed abroad in the early 1990s (miller, 1998; lewis, 1995). the method of risk and reliability evaluation is an important tool to solve the problem of uncertain factors in foreign drilling projects (cunha et al., 2005). there are limited researches on domestic oil and water well casing damage warning (hualin et al., 2010). encheng wu and tie yan broke through the viewpoint that casing damage is caused by oil shale swelling with water, and concluded that after the fossil layer enters the water, under the action of fluid pressure and ground stress, when the shear stress exceeds the shear strength of the weak surface of the fossil layer, the formation becomes unstable and the sheet casing damage occurs (encheng & tie, 2007). lianping li pointed out that the standard layer casing damage is the result of mechanical shearing. shear damage is the main form of casing damage and there is a casing damage shear zone (lianping, 2007). chi ai et al analyzed the influence of casting steel grade and wall thickness on the critical condition of casing damage. he analyzed the main factors influencing the critical slip of casing from the angle of delaying the shear damage of casing (chi et al., 2015). yanjun zhou et al proposed a dynamic prediction method for causing damage based on the integration of rough sets and support vector machines (zhou et al., 2010). guorong wang et al obtained the prediction model of casing reliability life by qualitative and quantitative analysis of casing fault tree, and deduced the calculation formula of the upper and lower limit of casing damage in a period of time after completion (guorong et al., 2002). in order to solve the problem of large number of casing damage in the standard layer of daqing oilfield, this study applied on-site oil well test data and numerical simulation data for comparison, the data are in good agreement that the simulation model is available. the key factors affecting casing damage are formation elastic modulus, the formation thickness, the formation dip angle, casing strength and water injection pressure. three orthogonal tests were carried out on five factors. numerical simulations were carried out to obtain critical strain data and casing true strain data. the multivariate regression method is used to obtain the prediction equation of the critical strain of the casing and the true strain of the casing, and the prediction method of casing damage based on the parameters of the well and the production parameters is obtained. establishment of standard layer formation-cement ring-casing numerical model through the analysis of the lithological characteristics of the bottom of a standard layer in daqing oilfield, rock formation sampling is shown in fig. 1. the rock mass at the bottom of this section belongs to typical jointed rock mass, and there are abundant bedding joints and micro-cracks. this structure facilitates rapid immersion of injected water, forming an expanding flooded area, when the submerged area is squeezed by external force, the rock mass will easily slip along the structural plane of the rock mass. taking the standard layer as the research object, establish a threedimensional fluid-solid coupling mechanical model of the casing-cement ring-formation. formation thickness and formation inclination are determined according to well conditions, the reservoir rock mass is subjected to the pressure of the overlying rock and the weight of the rock itself and the pressure of the production fluid in the casing (wan mahari et al., 2018). the initial osmotic pressure is applied to the upper, lower and circumferential surfaces of the formation, a fixed constraint is applied on the bottom surface, displacement constraints in the x and y directions were applied to the outer surface to simulate the lateral constraint of surrounding formation rocks on the model, the mechanical model is shown in fig. 2. figure 2. standard layer casing damage analysis mechanical model considering the non-uniformity of the formation and the casing, the numerical model is used to calculate the above mechanical model, and the deformation and strain law of the casing under different formation conditions are obtained. the finite element model is shown in figure 3. figure 3. standard layer casing damage analysis finite element model according to the test data of a rock layer in daqing oilfield, set the rock formation of mudstone shale mudstone interbed structure, the upper and lower mudstones is set as creep materials, shale set into joint material (cheng et al., 2019). analysis under the condition of water immersion, due to the difference in mechanical properties between the layers, the deformation of the joint of the cementation position is uncoordinated and the casing deformation is damaged. figure 1. a sampling of a standard layer shale in daqing oilfield 227shear damage mechanism and early warning study of shale standard layer casing by simulation, as shown in fig. 5, casing strain value detected by the oil well and the casing strain value calculated by simulation. it can be observed in the figure that the oil well test strain data and simulation results are consistent, indicating the correctness of the calculation model and method. the outer load of the casing is 20 mpa, 30m pa, 40 mpa, 50 mpa respectively. the strain and deformation of the simulated casing change with time, as shown in the casing strain cloud diagram of fig. 6, the bending deformation position of the casing is the concentration region of plastic strain. joint materials include joint system water invasion and failure behavior, as well as the failure mechanism of matrix materials. the plastic behavior of the matrix material follows the linear drucker-prager failure criterion, and the joint material model uses the stress-based fracture opening criterion. when the crack is open, the model follows:   an ps el an el ( ) ≤ (1) in the formula, an el is the normal elastic strain component across the crack, is calculated from the plane stress assumption to calculate the normal elastic strain across the crack. ε − σ σ an ps el a a v e( ) = +( )1 2 (2) in the formula, e is the elastic modulus of the material, v is the poisson’s ratio, a1, a2 is the radial stress on the joint surface. the sliding failure surface of the joint system a is defined as follows: f p da a a a a= =τ − β −tan 0 (3) in the formula, a is the internal friction angle of the joint system a, da is the viscous force of the joint system a, a is the shear stress on the joint surface, pa is the vertical stress acting on the joint surface. when  fa < 0, the joint system a does not slip. when fa = 0, the joint system from sliding. in the calculation process, the casing deformation is caused by the deformation of the standard layer. in order to accurately simulate the deformation state of the standard layer after water inflow, the simulation of the deformation of the casing is performed in two steps (giwa et al., 2016). in the first step, the internal stress of the casing and the stress field of the formation caused by the cementing process are applied to the calculation model, and the initial geostress field of the casing-cement ring-formation model is simulated, in the second step, the water pressure boundary conditions are applied to the calculation model to simulate the shale water immersion process, and the casing strain and deformation law under different influent strengths are simulated to provide a basis for the casing damage determination. strain and deformation of the casing this study selected a well in daqing oilfield for testing, the shale layer is located 923 meters underground and the shale thickness is 2.6 meters, the principal stresses in the three directions of x, y and z are 23.32 mpa, 22.25 mpa and 25.16 mpa the dip angle of the formation is about 3 degrees, the casing diameter is 51/2”, and the yield limit of the casing is 552 mpa. in the field logging process, the well detection uses continuous monitoring and stores test data, and continuously monitors for 9 months, as shown in fig. 4. according to the above theory and data, the strain of the casing is obtained figure 4. oilfield oil well test true strain 0.0005 0 -0.0005 -0.001 -0.0015 -0.002 -0.0025 -0.003 time/month 15 20 strain obtained by simulation strain from oil well testing figure 5. comparison of monitoring and calculation of logarithmic strain outside the casing (a)elastic strain ee (b)plastic strain pe (c)logarithmic strain le figure 6. casing strain cloud 228 suling wang, xuefei wang, minzhan chen, kangxing dong, minzheng jiang, lei yang the true strain is defined as the natural logarithm of the ratio of the current length of the initial length, so it is also called logarithmic strain, which is a combination of elastic strain and plastic strain. the elastic strain, plastic strain and true strain in the bending position of the casing is extracted, figure 7 shows the variation of the elastic strain and plastic strain of the casing with time under different loads: (a) elastic strain (b) plastic strain figure 7. variation of elastic strain of casing with time under different loads when the water injection pressure outside the casing is faster or the pressure outside the casing is larger, the critical elastic strain of the casing gradually decreases with the increase of the external pressure of the casing. but the reduction is small, and the external load is 40 mpa compared with 50 mpa when the plastic deformation occurs. the difference of the elastic strain of the casing is 0.00003, which is about 3.5%, after the casing reaches the plastic deformation, the elastic strain of the casing increases slowly, and the rate of elastic strain of the casing under different external pressures is the same. with the extension of time, the elastic deformation of the casing is basically the same. the larger the external load is, the earlier the deformation will be. the plastic strain rate of the casing is not affected by external pressure. since field detection shows the comprehensive strain of elastic strain and plastic strain, the variation law of logarithmic strain is extracted, as shown in figure 8. figure 8. horizontal logarithmic strains of casing under different loads figure 9. horizontal deformations of the casing under different loads figure 8 shows the horizontal logarithmic strain of casing under different loads. the strain growth curve can be divided into the logarithmic strain slow growth phase, the accelerated growth phase and the rapid growth phase. the casing deformation law corresponding to fig. 9 is the slow deformation phase and the deformation acceleration phase. the slow growth phase of logarithmic strain mainly occurs in the initial stage of elastic deformation of casing, and the acceleration phase of logarithmic strain is the stage of elastic instability of casing, and the stage of rapid growth is the stage of plastic strain of casing. in the elastic instability stage and the plastic deformation stage, the casing deformation is in the acceleration zone, which is the instability zone of the casing deformation, and the casing damage is intensified at this stage. when the elastic strain of the casing accumulates to a certain extent, the plastic strain increases sharply and the casing enters the plastic strain phase. from the point of view of the management of casing damage, in the accelerated growth phase of casing logarithmic strain, timely take appropriate measures to slow the casing damage. this study proposes a method to predict the critical strain and the true strain of the casing. when the true strain of the casing exceeds the critical strain of the casing, the casing will enter the plastic deformation stage and an early warning to reduce casing damage. standard layers casing loss prediction the main factors affecting the casing include: formation elastic modulus, formation thickness, formation inclination, casing strength and water injection pressure. this study conducts 3 level orthogonal tests on 5 parameters, according to the numerical simulation method established above, the true strain value of the casing and the critical strain value when the casing is damaged are obtained, the calculation results are shown in fig. 10. figure 10. orthogonal test numerical simulation results data scatter plot it is concluded from figure 10 that the three levels of the reservoir elastic modulus, reservoir thickness, reservoir angle, casing strength and casing water injection pressure are combined to obtain 37 sets of parameters and corresponding casing damage critical strain values and true strain value of the casing. the data analysis software was used to obtain the binary multiple regression equations which characterize the relationship between the critical strain and true strain and the above five factors, and the variance analysis method was used to verify the rationality of the equation. the equation was used to predict the casing damage. then get the following mathematical model. ε −3.3 −3. −2.8 −3. − − − 3 − lj x x x= ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅8 4 10 10 10 10 7 1 5 2 4 6. 99 − − −⋅ ⋅ + ⋅ ⋅ + ⋅ ⋅x x x4 6 4 6 5 310 3 9 10 3 05 10. . (4) ε −82 − − − −−zs x x x x x x= + ⋅ ⋅ + ⋅ ⋅ ⋅ ⋅62 37 30 3 5 10 2 7 7 31 2 3 4 2 5 1 2. . . + ⋅ + + ⋅ ⋅ + ⋅ ⋅ ⋅ ⋅ + 22 6 8 10 2 3 10 2 4 4 2 3 2 4 2 5 5 2 2 1 2x x x x x x. . . − − − .. . . 1 10 2 4 10 15 10 1 1 1 3 2 1 4 4 1 5 1 2 ⋅ ⋅ ⋅ + ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ x x x x x x x − − −− − ⋅⋅ ⋅ ⋅ ⋅ + ⋅ ⋅ ⋅ ⋅ ⋅ ⋅ x x x x x x x x 3 2 4 2 2 5 3 4 32 9 10 1 9 2 8 10 3 9 − − − − −. . . . 33 5 1 4 5 410 3 4 10⋅ ⋅ ⋅ ⋅ ⋅x x x− −− . (5) 229shear damage mechanism and early warning study of shale standard layer casing among them, lj is the casing critical strain; zx is the casing true strain; x1 is the reservoir elastic modulus, gpa; x2 is the reservoir thickness, m; x3 is the reservoir inclination, degree; x4 is the casing strength, mpa; and x5 is the outer casing pressure, mpa. comparing the measured critical strain and true strain curve with the multiple regression fitting theoretical curve by comparing fig. 11 with fig. 12, it can be seen that the formula of multiple regression theory can fit the sample data well. figure 11. comparison of sampling critical strain data and multiple regression critical strain curves figure 12. comparison of on-site sampling data with theoretical curves according to the prediction equations of critical strain and true strain of the casing obtained by binary regression, the critical strain value of the casing and the true strain of the casing can be predicted under conditions of known oil well parameters and production parameters. the equation is  zs lj≥ (6) if the equation (6) is satisfied, the casing begins to undergo unstable deformation and enters the stage of plastic deformation, and measures need to be taken in time, otherwise, the casing is stable and no instability damage occurs. the prediction equation is suitable for the standard layer of the second section of daqing oilfield. other oilfields can use the above method to determine the prediction equation and to determine the damage of the casing. conclusion (1) the main cause of casing shear damage is the slippage of the standard shale layer. according to the theory of rock mechanics, creeping materials are utilized to the upper and lower mudstones, and joint materials are used for shale. the fluid-solid coupling calculation model of the formation-cement ringcasing of the standard interval mudstone shale interbed is established, and the deformation low of casing shear damage is obtained. the simulation results are compared with actual logging results of daqing oilfield, the rationality of the numerical model is verified and a simulation method of standard casing damage is proposed. (2) in the shear damage process, the transition from elastic strain to plastic strain is the initial stage of casing damage. from the perspective of casing damage prevention, steps taken to reduce casing damage should be taken during the casing strain acceleration phase. built on the above theory, this study proposes a method for predicting the true strain and critical strain. (3) the five factors affecting casing damage are casing strength, rock inclination, rock thickness, rock elastic modulus, and external load. the five factors were divided into three levels for orthogonal numerical simulation experiments, and data on the correlation between the five factors and the critical strain and true strain of the casing were extracted. according to the statistical principle, the multivariate regression method is used to obtain the mathematical model of the critical strain and true strain prediction of the casing. when the true strain exceeds the critical strain, the casing deformation enters the plastic phase and the casing damage occurs. the prediction of the strain state of the casing under actual working conditions is realized according to the formation parameters and logging data, which provide a new method for preventing casing damage. (4) the theoretical formula for the shear damage of the standard layer casing of the nen 2 section of daqing oilfield has certain practicability. acknowledgments this research was supported by cooperation project between heilongjiang province and academicians (2019sy0304) and china postdoctoral science foundation(2019m661249). the authors would like to thank daqing oilfield for providing data on samplingstress and strain. references cheng y. w., ng, k. h., lam, s. s., lim, j. w., wongsakulphasatch, s., witoon, t., & cheng, c. k. (2019). syngas from catalytic steam reforming of palm oil mill effluent: an optimization study. international journal of hydrogen energy, 44(18), 9220-9236. chi, a., chaoyang, h., & yueming, c. (2015). casing optimization for delaying standard layer damage of daqing oilfield. petroleum drilling techniques, 43(06), 7-12. cunha, j. c., demirdal, b., & gui, p.. (2005). use of quantitative risk analysis for uncertainty quantification on drilling operations-review and lessons learned. society of petroleum engineers, 94980. encheng, w., & tie, y. (2007). analysis of the mechanism of casing damage caused by the standard layer of fossil layer in the second layer of daqing oilfield. journal of daqing petroleum institute, (02), 3841+125. giwa, s. o., adama, k. o., nwaokocha, c. n., & solana, o. i. (2016). characterization and flow behaviour of sandbox (hura crepitans linn) seed oil and its methyl esters. international energy journal, 16(2), 65-72. guorong, w., qingyou, l., & xia, h. (2002). prediction of casing reliability life. natural gas industry, 22(5), 53-55 hualin, l., zhichuan, g., zhenlai, y., & guangjun, m. (2010). a risk assessment method for casing failure based on reliability theory. acta petrolei sinica, 31(01), 161-164+172 lewis, d. b. (1995). load and resistance factor design for oil country tubular goods. offshore technology conference, 7936. lianping, l., hongqing, k., guizhen, j., qin, l., & qing, liu. (2007). a new understanding of the formation mechanism of casing damage in xingbei development zone and the comprehensive prevention and control technology of casing damage. daqing petroleum geology and development, (01), 83-87. miller, r. a. (1998). real world implementation of qra methods in casing design. society of petroleum engineers, 48325. wan mahari, w. a., chong, c. t., cheng, c. k., lee, c. l., hendrata, k., yek, p. n. y., ma, n. l., & lam, s. s. (2018). production of value-added liquid fuel via microwave co-pyrolysis of used frying oil and plastic waste. energy, 162, 309-317. zhou, y., jia, j., & li, r. (2010). a dynamic prediction method for casing damage based on rough set theory and support vector machine. journal of china university of petroleum (natural science edition), 34(06), 71-75. keywords: dingshan area; longmaxi formation; tectonic fracture; binary method; quantitative prediction. palabras clave: área dingshan; formación longmaxi; fracturas tectónicas; método binario; predicción cuantitativa. how to cite item xie, j., qin, q., fan, c., hu, d., & li, h. (2019). the quantitative prediction of tectonic fractures of longmaxi formation in dingshan area, southeastern sichuan basin. earth sciences research journal, 23(2), 157-162. doi: https:// doi.org/10.15446/esrj.v23n2.79957 tectonic fractures are a significant factor for the production of the longmaxi formation in dingshan area. the analysis of the distribution and development degree of tectonic fractures is the main research method in present-day. based on the data of field outcrops and borehole core, the new method-binary method is put forward by rock fracture and energy value to decide tectonic fracture development. according to the method, the fitting formula is offered between the relation of the value of rock fracture, energy, and fracture density. compared to the fracture density predictive value and the observed value of 15 wells, all wells relative error of less than 25%. the prediction results are accurate. according to the fitting formula of fracture development and distribution in the study area are obtained. the results show that the fracture density is high as a whole of the longmaxi formation in dingshan area. whole shows by southeast to gradually reduce the trend of the northwest. supported the future of the oil and gas exploration to provide geological basis and guidance in dingshan area. abstract the quantitative prediction of tectonic fractures of longmaxi formation in dingshan area, southeastern sichuan basin predicción cuantitativa de fracturas tectónicas en la formación longmaxi del área dingshan, en el suroriente de la cuenca sichuan issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n2.79957 las fracturas tectónicas son un factor determinante en la producción en la formación longmaxi, en el área dingsahn. el análisis de la distribución y grado de desarrollo de las fracturas tectónicas es uno de los principales métodos de investigación en nuestros días. con base en la información de los afloramientos y el núcleo del pozo, el nuevo método binario se presenta como valor de las fracturas rocosas y de energía para decidir el desarrollo de las fracturas tectónicas. de acuerdo con este método, se ofrece una formula de ajuste sobre la relación del valor de la fractura de la roca, la energía y la densidad de la fractura. al comparar el valor predictivo de la densidad de fractura y el valor observado en 15 pozos, todos los pozos tienen un error relativo menor al 25 %. la predicción de resultados es exacta. gracias a la formula de ajuste se obtuvo el desarrollo de la fractura y la distribución en el área de estudio. los resultados muestran que la densidad de la fractura es tan alta como la de la formación longmaxi en el área de dingshan. los resultados muestran que el sudeste gradualmente reduce la tendencia del noroeste. estos hallazgos respaldan el futuro de la exploración de gas y petróleo que aporten a las bases geológicas para el área de dingshan. resumen record manuscript received: 24/07/2018 accepted for publication: 22/04/2018 earth sciences research journal earth sci. res. j. vol. 23, no. 2 (june, 2019): 157-162 jiatong xie1,2, qirong qin1,2*, cunhui fan1,2, dongfeng hu3, hu li1,2,4 1state key laboratory of oil and gas reservoir geology and exploitation, chengdu, 610500, china; 2school of science and technology, southwest petroleum university, chengdu, 610500, china; 3exploration company, sinopec, chengdu, 610041, china; 4tarim oilfield company, petrochina, korla , 841000, china * corresponging author: 986336692@qq.com te c to n ic s 158 jiatong xie, qirong qin, cunhui fan, dongfeng hu, hu li introduction with a heightened level of global oil and gas exploration, fractured hydrocarbon reservoirs have become an important new field for exploration (he & wei, 2007; pu & qin, 2008) the identification, accurate description, and semi-quantitative-qualitative study of the tectonic fractures in reservoir strata are the keys to their effective development at a later stage (pu et al., 2008). consequently, the quantitative prediction of fractures has become an effective research method at present. researchers have undertaken numerous studies in connection with fracture prediction methods. dai & wang (2003) and pu et al. (2008) have summarized fracture prediction methods, covering the early conventional logging method and seismic method as well as the presently commonly used type theory method and the stress field simulation method. wang et al. (2016); yu et al. (2016) and gao et al. (2015) used the finite element method for stress field simulation to make predictions on fractures. wang, dai, sheng, qiao & li (2009) and wang (2007) applied the fractal dimension method to carry out predictions. these methods have all made major contributions to fracture prediction, but there are still deficiencies in terms of the quantitative prediction of fractures. the fracture development zones in tight reservoir strata are important targets for oil and gas exploration. tectonic fractures are the main reservoir spaces and seepage channels, and the extent of fracture development has a major effect on the quality and oil and gas production capacity of reservoir strata. therefore, on the basis of existing research, a “binary method” that combines the rock breaking criteria with the energy method is used in this paper to carry out a study on the quantitative prediction of fractures for the longmaxi formation in the dingshan area. this study aims to predict the development and distribution situation of the tectonic fractures in the reservoir strata of the study area and to provide a new geological reference and basis for oil and gas exploration and development in the area. geological background of the study area the dingshan area is located in qiongjiang city at the juncture between xishui county of guizhou province and chongqing and is affiliated with shizhen, chongqing county, qiongjiang city. tectonically, the dingshan area is located in the down warping fold region of southeast sichuan, with the southern portion crossing over the juncture between the eastern edge of the waning folded belt and the sloping belt in southern sichuan (figure 1), and a triassic system is exposed on the surface. dingshan is tectonically subjected to the comprehensive influence of the huayingshan fault, the qiyueshan fault, and the zunyi fault. the tectonic shape is complex, with more developed folds and faults composed of a series of high-dipping anticline belts and fault zones in trough-like folds in the northeastern and north-northeastern directions (sun et al., 2010). the tectonic structure is now located near the intersection of the north-south and northeastern tectonic belts, with a tectonic belt in the northeastern direction formed under the action of the jiangnan uplift during the late cretaceous epoch and advanced superpositioning of the tectonic structure near the north-south direction. the overall shape of the dingshan tectonic plane is in the shape of a nose with broken anticlines (huang, wei & wang, 2017; mei, liu, tang, shen & fan, 2010; yuan, 2008) . the strata of the study area have comparatively complete development. the proterozoic sinian system, the early paleozoic and late paleozoic permian system, and the mesozoic erathem developed from bottom to top. the shale thickness of the longmaxi formation is approximately 158-334 m, is generally 250 m, and can be divided into two sections—upper and lower—with the lower section being graptolite shale facies and the upper section interbedded with gray-dark gray calcareous shale and calcareous mudstone with thin gray layers of argillaceous limestone or marl. since 2013, major breakthroughs in oil and gas exploration have been attained in the field of marine shale gas exploration in the dingshan area, as daily productions of 3. 4×104 m3 and 10. 5×104 m3 in industrial gas flows were obtained at the dy1 and dy2 wells, respectively, reflecting the good exploration prospects for shale gas in this area. fracture development characteristics fracture dip in accordance with the classification criteria of fracture dips, fracture dips less than 15° are horizontal seams, fracture dips of 15°-45° are low-angle seams, fracture dips of 45°-75° are high-angle seams, and fracture dips greater than 75° are vertical seams. a statistical analysis of the field observation data shows that the surface development of seams in the dingshan area are mainly centered on high-angle seams as vertical seams with fracture dips greater than 75° account for 63. 72%, high-angle seams with fracture dips of 45°-75° account for 34. 76%, and low-angle or horizontal seams did not develop as they only account for 1. 52% of all observed fractures (figure 2). extent of fracture filling according to core observations and a statistical analysis of image logging data, among the fractures in the reservoir strata of the longmaxi formation in the study area, the proportion of unfilled fractures account for approximately 20%, the proportion of partially filled fractures account for approximately 20%, and the proportion of fully filled fractures account for approximately 60%; partially filled fractures and unfilled fractures account for a total of approximately 40%, and these partially filled fractures and unfilled fractures play a comparatively important role in oil and gas seepage (figure 3). fracture density fracture density is an important parameter for quantitatively characterizing the extent of fracture development. there are mainly three common ways to express fracture density: linear density, fracture density by area, and fracture density by volume. in this study, linear density is used to describe fracture density. linear density (lfd) refers to the ratio of the number of fractures nf intersecting with a straight line (perpendicular to the direction of flow or referring to the centerline of the rock) and the length of this straight line lb (formula 1), which is also called the fracture frequency or linear frequency. figure 1. tectonic division of southeast sichuan and the location of dingshan area fracture dip 45°-75°15°-15°0°-15° 75°-90° 1.52 34.76 63.7270 60 50 40 30 20 10 0 pe rc en ta ge % fig. 2 fracture dip direction histogram of longmaxi formation 159the quantitative prediction of tectonic fractures of longmaxi formation in dingshan area, southeastern sichuan basin of those, the fractures in the northwest-west direction (290°±10°) and the northeast direction (40°±10°) are the most developed, followed by near southnorth (0°±5°), and the extent of fracture development in the north-northwest direction (345°±5°) is relatively poor. fracture distribution and quantitative prediction prediction methods (1) principal curvature method: researchers defined the principal curvature method as a mathematical method that portrays the distribution of tectonic fractures in the reservoir strata developed on the basis of the tectonic fracture formation mechanism, which predicts the extent of fracture development in accordance with the extent of the curvature of the layered rock. huang, peng, & wang, thought that the greater the tectonic stress on the strata, the stronger the strata bending action and the higher the principal curvature value, which can further reflect the enhancement of the strata breaking effect and the increase in the extent of tectonic fracture development. therefore, the principal curvature value can become a characteristic parameter for predicting the interwell distribution of tectonic fractures. in usual situations, the tectonic fracture prediction results obtained through the principal curvature method are often only applicable to longitudinal fractures derived from the bending and deformation of rock layers under force. for most of the oil and gas fields, the tectonic fractures in the reservoir strata are mainly shear fractures formed under the action of tectonic compression, and the proportion of longitudinal fractures is very limited. therefore, the advantages of using the principal curvature method to carry out qualitative or semi-quantitative predictions on the tectonic fractures are more obvious, but there is a certain level of difficulty in carrying out quantitative predictions. (2) rock breaking method: according to the findings in the study on the strength theory of rock mechanics by zhiying xu, shuai yin, et al. , the stressstrain function is generally used to characterize rock failure conditions—that is, the breaking strength criteria—of which there are generally five categories based on the criteria proposed by mohr (1900): the maximum positive strain theory, mohr’s strength theory, the shear strain energy intensity theory, the octahedral stress theory, and griffith’s strength theory. each criterion category has different conditions and different criteria, and the breaking criteria used also differ as environmental conditions differ. the shale gas reservoir strata are generally buried deeper with a greater confining pressure, and the filling degree no fillinghalf fillingfull filling 2021 59 70 60 50 40 30 20 10 0 pe rc en ta ge % fig. 3 fracture filling degree of histogram the expression for counting the centerline density of the core is the ratio of the number of observed core fractures and the length of the core, the units are bar(s)/m. l = n lfd f b (1) according to the description of core observations and analysis of the logging data and based on the statistical analysis of the core observation results, a statistical analysis was carried out on the core fracture density of the area to describe and analyze the fracture development situation in the study area. it can be seen from the histogram on fracture development density that the density values obtained from the core statistics generally tend to be small, and the fracture densities are mainly concentrated in 0. 5-1 bar/m, 1-1. 5 bars/m, and <0. 5 bar/m. fractures with densities between 0. 5 and 1 bar/m and fractures with densities between 1 and 1. 5 bars/m account for approximately 32% and 23% of all fracture development sections, respectively. fractures with densities <0. 5 bar/m and between 2. 5 and 3 bars/m account for 21% and 19% of all fracture development sections, respectively (figure 4). in summary, the vast majority of the fracture densities in the area are less than 1. 5 bars/m; the density is generally small. in analyzing the reason, the first may be related to the proportion of high-angle seams occupying this area as the existence of high-angle seams reduces the drilling ratio; the second may be related to the tectonic site where the observation well is located. fracture orientation figure 5 is the rose diagram of the overall fracture orientations for all wells in the longmaxi formation produced based on the fracture interpretation results of image logging. overall, the fracture orientations of the study area are mainly in four bearings: northwest-west (290°±10°), near south-north (0°±5°), north-northwest (345°±5°), and northeast (40°±10°). fracture dip 45°-75°15°-45°0°-15° 75°-90° 21,15 23,08 1,92 30 25 20 15 10 5 0 pe rc en ta ge % 32,69 1,92 19,24 35 figure 4. core fracture density distribution frequency diagram of longmaxi formation 36 n ew s 27 18 9 9 18 27 36 36 27 18 9 9 18 27 36 4 23 1 figure 5. fracture trend rose diagram 160 jiatong xie, qirong qin, cunhui fan, dongfeng hu, hu li mohr-coulomb criterion can usually be used as the basis for determining rock breakage. according to mohr’s theory, the expression for the rock failure proximity coefficient is η=f (σ)/ k (κ), and the meanings expressed by its parameters are shown in figure 6. (4) binary method: the rock breaking value represents the possibility of tectonic fracture development, and the energy value represents the size of the ability for tectonic fractures to develop. the use of the rock breaking value alone cannot determine the extent of tectonic fracture development after a rock breaks. similarly, when considering only the energy value, the extent of tectonic fracture development also cannot be accurately judged. the simulation results of these two methods also often complement one another, and both must be combined to enable accurate judgment of the extent of fracture development in the rock strata. therefore, when predicting the tectonic fracture distribution in this study, the research outcomes of ding (1998) and xingui zhou et al. were used to comprehensively consider these two methods—the rock breaking method and the energy method—and a “binary method” was proposed (the extent of tectonic fracture development is jointly determined by two aspects—the rock breaking rate and the strain energy density); the rock breaking rate represents the possibility of tectonic fracture development, and the strain energy density represents the size of the ability for tectonic fractures to develop. when the rock breaking rate is comparatively low, not reaching the critical value that causes macroscopically obvious shear breakage to occur to the rock, the extent of tectonic fracture development is primarily determined by the rock breaking rate; after the rock breaking rate reaches the critical value, macroscopically obvious shear breakage has already occurred in the rock, and the extent of tectonic fracture development is primarily determined by the strain energy density at this time. in actual application, the rock breaking rate and the strain energy density are used together with the linear density of tectonic fractures in carrying out fitting calculations, and the fitting relationship formula is then used to carry out predictions on the distribution pattern for the tectonic fracture density of the reservoir strata. according to the guiding ideology of this method, the rock breaking rate η and the strain energy density w together with the linear density of tectonic fractures d are fitted in accordance with the following equation to establish a mathematical model for the quantitative prediction of tectonic fractures based on the rock breaking value and the energy value: d a b w c d w e= η η2 2+ + + + (4) in the formula, d is the predicted value for the linear density of the tectonic fracture in bar(s)/m, η is the dimensionless rock breaking coefficient, w is the energy value, and the others are the fitting coefficients. distribution prediction results through the rock breaking criteria and the energy formula, the rock breaking values and the energy values of the longmaxi formation in the dingshan area were calculated. the rock breaking values of the longmaxi k f c σ1 σ2 σn τ ϕ figure 6. based on the failure approach index of the mohr coulomb model schematic diagram the specific expression is η= = σ −σ ϕ σ −σ ϕ 1 1 f k c tan sin 2 2 2 − (2) in the formula, σ1 and σ2 are the maximum and minimum principal stresses, respectively, c is the cohesive force of the rock, and φ is the internal friction angle of the rock. when η1, the rock begins to break and generate a fracture; the greater the value of η, the stronger the breakage of the rock and the more developed the fracture. when η<1, the rock is stable, no breakage is generated, and a fracture does not develop. the values of c and φ reflect the state of molecular cohesion within the rock and indicate the brittleness of the rock, which is an important factor when studying the extent of rock fragmentation. in principle, it is believed that the greater the comprehensive breaking value of the rock, the more developed the tectonic fracture system, wherein the state of local stress near the fracture is as shown in figure 7, in addition to the maximum and minimum principal stresses near the fracture, the internal friction angle also has an effect, which is the main factor for the generation of fractures. (3) energy method: for fracture prediction, ju w. and huizhen song et al. proposed a rock energy method that uses the specific energy of strain to carry out predictions on fractures. to compensate for the deficiency in the inability of η to independently determine the extent of fracture development after the rock breaks, the deformation energy value is used as a supplementary criterion of the breakage criteria to jointly study the extent of tectonic fracture development. when breakages occur in rocks, those with large energy values may develop more fractures. the specific energy of strain per unit volume in the strata (that is, the deformation energy value) is given by the following formula: w e " e = σ −σ σ −σ µ1 1 2 3 2 2 2 2 − (3) wherein e is the elastic modulus, and μ is poisson’s ratio the specific energy of strain can become the supplementary criterion of the breaking criteria to compensate for the deficiency in the inability of the η value to independently determine the extent of fracture development after breakage, reflecting the high or low extent of tectonic fracture development. σ1 σx σ3 τxy βx1σy σx σ1 τxy τxy σ3 τxy σy y x figure 7. the local stress state near the fracture 161the quantitative prediction of tectonic fractures of longmaxi formation in dingshan area, southeastern sichuan basin formation were greater near the ds1 well, the dy4-s2 well, and the dy3 well, with a maximum of up to 2. 1 j/m (figure 8); the energy values were greater near the areas of the ds1 well, the dy1 well, and the dy1-5 well, with a maximum of up to 1. 03 j/m (figure 9). as a whole, the overall trend of the rock breaking values and energy values for the longmaxi formation in the dingshan area was a gradual decrease from a southeast to a northwest direction, and the fractures in the southeastern portion of the study area were more developed. there are certain limitations to carrying out quantitative predictions using any one of the breaking values or energy values alone as it is impossible to completely include the possibility of fracture development and the size of fracture development from a quantitative perspective. the comprehensive the accuracy of the prediction results was verified by using relative error (f) to reflect the degree of accuracy of the tectonic fracture predictions. it is generally thought that if the relative error results do not exceed 25%, then the prediction model is considered good and can be established, and the prediction results obtained for the fracture distribution are also credible. figure 8. late yanshanian-middle himalayan rock rupture value distribution of the longmaxi formation in dingshan area figure 9. late yanshanian-middle himalayan rock energy value distribution of the longmaxi formation in dingshan area figure 10. late yanshanian-middle himalayan structural fracture density predictive distribution of the longmaxi formation in dingshan area table 1. rock crack coefficient and the strain ratio value well number dy4-3 dy4-s2 dy4 dy4-s1 dy3 dy5 dy2 dy1-5 dy5-2 dy5-6 dy2-1 dy3-c1 dy3-3 ds1 dy1 crack coefficient /η 1. 7 2 1. 2 1. 9 2. 1 1 1. 3 1. 6 1. 3 1. 2 0. 9 1. 5 1. 2 2. 1 1. 8 strain ratio/w 0. 51 0. 28 0. 21 0. 2 0. 23 0. 13 0. 25 0. 6 0. 14 0. 07 0. 09 0. 26 0. 24 1 0. 8 breaking value and energy value can complement each other in the simulation calculations. therefore, through the statistics on the rock breaking coefficient and specific energy of strain for every well (table 1), analysis and fitting were carried out using actual measurements of tectonic fracture density with the rock breaking coefficient and energy value—that is, the specific energy of strain—of the corresponding well through formula 6 to obtain the mathematical model for predicting the tectonic fractures of the longmaxi formation in the dingshan area in the late yanshan-middle himalayan period. d . w . w=−0.224−1.42η η−1.2022 2 20 966 1 575+ + = r 0.89 (5) in the formula, d represents the tectonic facture density of the longmaxi formation, w represents the strain energy, η represents the rock breaking value, and r represents the fitting correlation coefficient. based on the tectonic fracture prediction model, the development and distribution situation of the tectonic fractures for the longmaxi formation in the dingshan area can be obtained (figure 10). it can be seen through the prediction results that the fracture densities are higher near the ds1 well, the dy3 well, and the dy4-s2 well, being up to 2. 22 m-1, and the overall trend shows a gradual decrease from a southeast direction to a northwest direction. 162 jiatong xie, qirong qin, cunhui fan, dongfeng hu, hu li the relative error formula is f x %= − predicted measured predicted 100 (6) through an analysis and comparison of the actual measured values of the fracture densities and the predicted densities of the longmaxi formation in the dingshan area, the relative errors for all wells in the dingshan area were found to be all within 25% (table 2). the actual measured values basically coincided with the observed values, and the overall conformity effect was very good. the prediction results were satisfactory, and the fracture distribution prediction results were credible. conclusions (1) the tectonic fractures of the longmaxi formation in the dingshan area were found to be centered on high-angle and vertical seams, the main developments are fractures in the northwest-west direction and the northeast direction, and the fracture filling was found to be comparatively severe as most were invalid fractures. (2) on the basis of a survey of the tectonic fracture methods, a binary method that comprehensively utilized the rock breaking method and the energy method was used to carry out fracture formula fitting and prediction on the strata of the longmaxi formation in the dingshan area and to establish a mathematical model for the quantitative prediction of fracture distribution. the prediction results showed that the density errors were all within 25% and that the prediction results were satisfactory. (3) on the basis of the mathematical modeling results of quantitative prediction, the distribution pattern of the longmaxi formation in the dingshan area was studied. the study showed that the overall fracture density of the longmaxi formation in the study area is relatively high and that the fracture densities near the ds1 well, the dy3 well, and the dy4-s2 well are up to 2. 22 bars/m, and the overall trend showed a gradual decrease from a southeast direction to a northwest direction. references dai, j., & wang, b. (2003). integrated approach of identifying and predicting fracture in reservoir. petroleum geology and recovery efficiency, 01, 1-5. ding, z., qian, x., & huo hong. (1998). a new method for quantitative prediction of tectonic fractures two factor method. oil & gas geology, 01, 3-9. gao, s., zeng, l., & ma, s. (2015). quantitative prediction of fractures with different directions in tight sandstone reservoirs. natural gas geoscience, 26(03), 427-434. he, y., & wei, c. (2007). the present situation and research direction of evaluation methods in fracture type reservoir. progress in geophysics, 22(2): 537-543. huang, g., lu, s., & yang, f. (2003). application of curvature method to fissure forecast of the yingcheng formation stratum of xujiaweizi fault depression. journal of northeast petroleum university, 27(4), 9-11. huang, r., wei, x., & wang, q. (2017). key factors of shale gas accumulation in dingshan area of southeastern sichuan basin. marine origin petroleum geology, 22(02): 25-30. jiang, p. (1983). the failure criteria of the rock. journal of waterway transportation scientific research, (01), 82-110. ju, w., hou, g., & feng, s. (2014). quantitative prediction of the yanchang formation chang 63 reservoir tectonic fracture in the qingchengheshui area, ordos basin. earth science frontiers, 21(06), 310-320. li, z., zhang, y., & dou, y. (2013). fracture prediction in tight sandstone reservoir with curvature method. journal of southwest petroleum university (science & technology edition), 35(6), 57-63. mei, l., liu, z., & tang, j. (2010). mesozoic intra-continental progressive deformation in western hunan-hubei-eastern sichuan provinces of china: evidence from apatite fission track and balanced crosssection. editorial committee of earth science-journal of china university of geosciences, 35(2), 161-174. peng, h., xiong, y., & sun, l. (2005). research on main curvature method for fracture prediction in jialingjiang formation of maliuchang. special oil & gas reservoirs, 12(4), 21-23. pu, j., & qin, q. (2008). an overview of fracture prediction methods for oil and gas reservoirs. acta petrolei sinica, (03), 9-13. qin, q., zhang, l., & deng, h. (2004). determination of magnitude of palaeotectonic stress and application to tectonic geological modeling. chinese journal of rock mechanics and engineering, 23(23), 3979-3983. qin, q., zhang, l., & liu, l. (2004). numerical evaluation techniques of fracture porosity. natural gas industry, 02, 47-51. sun, w., liu, s., & wang, g. (2010). petroleum formed condition and process research for sinian to low paleozoic at dingshan structure in southeast of sichuan basin. geological science and technology information, 29(1), 49-55. wang, b., dai, j., & sheng, x. (2009). reservoir fracture forecast of midshasan member in niu 35 fault block. journal of china university of petroleum (edition of natural science), 33(03): 18-21. wang, b. (2007). description and quantitative predication of reservoir tectoclase. china university of petroleum. wang, k., dai, j., & wang, j. (2016). distribution of reservoir structural fractures and quantitative prediction of keshen-2 gas field, tarim basin. geotectonica et metallogenia, 40(06), 1123-1135. wang, k., zhang, h., & zhang, r. (2015). comprehensive assessment of reservoir structural fracture with multiple methods in keshen-2 gas field, tarim basin. acta petrolei sinica, 36(06), 673-687. yuan, h. (2008). the mechanism of hydrocarbon accumulation, sinian-lower palaeozoic, sichuan basin. chengdu university of technology, 2008. yu, x., hou, g., & li, y. (2016). quantitative prediction of tectonic fractures of lower jurassic ahe formation sandstones in dibei gasfield. earth science frontiers, 23(01), 240-252. zhou, x., & deng, h. (2003). the methods for quantitative prediction and evaluation of structural fissures in reservoirs. acta geoscientia sinica, 24(2), 175-180. zhou, x., zhang, l., & fan, k. (2007). methods for quantitative prediction of tectonic fractures in compact reservoirs in petroliferous basins and a case study. natural gas geoscience, 03, 328-333. zhou, x., zhang, l., & qu, x. (2009). characteristics and quantitative prediction of distribution laws of tectonic fractures of low-permeability reservoirs in yanhewan area. acta petrolei sinica, 30(02), 195-200. table 2. analysis of the fracture predictive relative error of the longmaxi formation in dingshan area well number dy4-3 dy4-s2 dy4 dy4-s1 dy3 dy5 dy2 dy1-5 dy5-2 dy5-6 dy2-1 dy3-c1 dy3-3 ds1 dy1 practical density (bar/m) 1. 5 2. 1 1. 1 1. 7 2. 5 1 1. 7 1. 8 1. 4 1. 6 0. 8 1. 4 1. 4 2. 3 1. 6 predictive density (bar/m) 1. 68 2. 1 1. 25 2. 05 2. 23 1. 07 1. 34 1. 56 1. 43 1. 38 0. 98 1. 57 1. 23 2. 22 1. 81 relative error/% 12 0 14 21 11 7 21 13 2 14 23 12 12 3 13 carlosvargaset al.indd 99 new geological and geophysical contributions in the section ibagué – armenia, central cordillera colombia carlos a. vargas1, andreas kammer1, mireya valdes2, carlos e. rodríguez1, alexander caneva3, jhon j. sánchez4, eliana arias5, carlos a. cortes6, héctor mora4 1 departamento de geociencias, universidad nacional de colombia – sede bogotá , colombia email: cavargasj@unal.edu.co 2 instituto colombiano del petroleo –icp, bucaramanga, colombia 3 universidad antonio nariño, dirección nacional de investigaciones, bogotá, colombia 4 red sismológica nacional de colombia, ingeominas, bogotá, colombia 5 incorporated research institutions for seismology – iris, washington dc, usa 6 universidad de manizales, facultad de ingeniería, manizales -colombia resumen ha sido interpretada la sección geológica entre las ciudades de ibagué y armenia basada en observaciones geofísicas y estructurales en busca de establecer el estilo estructural dominante y la configuración del basamento en profundidad. inicialmente se realizaron mediciones gravimétricas y magnetométricas levantadas cada 500 m entre estas ciudades, las cuales permitieron soportar el desarrollo del experimento sísmico “armenia-quake”. el alcance del experimento sísmico permitió detectar reflectores hasta aproximadamente 17 km de profundidad en un área de influencia de aproximadamente 25 km. la interpretación integral de la información geológica y geofísica permite concluir que el complejo cajamarca define una estructura sinclinal amplia cerca al municipio de cajamarca. partiendo de dicha información se pudo reconocer la presencia de un basamento compuesto por cuñas de rocas basálticas y sedimentarias asociadas a las formaciones quebradagrande y arquía. igualmente se reconoce la existencia de una gran flexión de la paleo-placa oceánica relacionada a la formación barroso, la cual se encuentra limitada al occidente por la falla cauca-almaguer. la estructura en flexión y la evolución compresiva del estado de esfuerzos en esta región de américa del sur podrían ser las responsables del depósito y posterior plegamiento de la formación cinta de piedra que constituye en gran medida la serranía de santa bárbara. palabras clave: armenia-quake, experimento de sísmica activa, modelamiento geofísico, serpentinitas y terremotos. manuscript received august 2005 paper accepted september 2005 earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 99-109 100 carlos a. vargas, andreas kammer, mireya valdes, carlos e. rodríguez alexander caneva, jhon j. sánchez, eliana arias, carlos a. cortes, héctor mora abstract based on geophysical and structural observations a geologic section between the cities of ibagué and armenia (colombia) has been interpreted to establish the dominant structural style and the basement configuration at depth. gravimetric and magnetic measurement were initially taken each 500 m along these cities, which helped in the development of the “armenian-quake” seismic experiment. the experiment illuminated reflectors down to approximately 17 km deep in an influence area of approximately 25 km. the complete interpretation of the geological and geophysical information allows concluding that the cajamarca complex defines a wide synclinal near to cajamarca town. with this information initially it was possible to recognize the presence of a basement compound for wedges of basaltic and sedimentary rocks associated to the quebradagrande and arquía formations. then was recognized the existence of a great flexion of the oceanic paleo-plate related to the barroso formation, which is limited at west by the cauca-almaguer fault. the flexion structure and the evolution of the stress in this region of south america could be responsible for the deposit and later folding of the “cinta de piedra formation” that constitutes the santa bárbara range. key words: armenia-quake, seismic active experiment, geophysical modelling, serpentinites and earthquakes. © 2005 esrj unibiblos. as the town of montenegro. this section represents one of the best outcrops in the central cordillera of colombia where rocks of diverse composition and age can be observed. generally, the stratigraphic base displays metamorphic rocks from the cajamarca complex varying between amphibolic green schists, graphitic schists, phyllites and locally marble whose protolith age has been assigned to the early palaeozoic (figure 2). the pericos fault brought these rocks into contact to the east with the ibagué batholith, an intrusive body of granodiorites of jurassic age, and at to the west the san jerónimo fault brought into contact the cajamarca complex with meta-sedimentary rocks of marine origin that are interspersed with metabasalts from the quebradagrande complex of cretaceous age (maya and gonzález, 1995). the quebradagrande complex formed a tectonic wedge limited by the silvia-pijao fault to the west. to the western side of the silvia-pijao fault some meta-igneous rocks of continental origin and formally denominated the arquía complex become visible (gonzález and núñez, 1991), which in turn are limited to the west by the cauca-almaguer fault. again to the west of the latter appear meta-sedimentary and meta-basalt rocks of oceanic origin of cretaceous age and denominated barroso formation (maya and gonzález, 1995). the barroso formation and the quebradagrande complex are parts of the geological province associated to the same subduction event. introduction the tectonic configuration in the northern andes has been interpreted as the product of at least three orogenic events resulting from the accretion of oceanic crust in the western margin of nw south america (duke-caro, 1990; freymueller et al., 1993). petrological and geochemical evidences of this process have been reported along the section that runs along the road between the cities of armenia and ibagué (figure 1). according to these authors, the protolith that constitutes the lands in the central cordillera of colombia was originated from intra-oceanic rocks and sediments of continental margins that have undergone processes of metamorphism. all these rocks evolved into an orogenic regime associated to intrusive and extrusive bodies that emerged throughout fractures affecting the crust in the region. base on geophysical and structural measurements we defined the relation at depth and the current seismo-tectonic activity that could have triggered the january 25, 1999, mw=6.2 armenia earthquake (monsalve and vargas, 2002). geological framework new observations have allowed to complement the geologic mapping by nelson (1962) and restrepopace (1992) between the city of ibagué and the town of calarcá,and to extend the mapping as far 101 new geological and geophysical contributions in the section ibagué armenia, central cordillera colombia geological and geophysical observations data initially, we performed a lithological and structural survey in the section ibagué calarcá – montenegro. measurements every 500 m were taken and allowed the design of magnetometric and gravimetric profiles supporting the controlled source seismic experiment that we called “armenia quake”. this experiment was developed during june, 2004, detonating four explosive charges of indugel ap+ at the fayat quarry near calarcá. the charges ranked between 75 and 117 kg. in this seismic experiment 36 seismographs reftek-125 and triaxial geophones (f n = 4.5 hz) from the iris-passcal consortium were placed in 95 recording locations. additionally, roughly 2000 hypocentral solutions reported by the seismological observatory of quindío –osqwere used to support the geological and geophysical interpretations. figure 1. geophysical profiles between ibagué and armenia cities along the main road (black line). figure 2. geologic map of the study area. continuous red lines are mapped faults. dashed red lines are main lineaments. volcaniclastic unites (quaternary) cajamarca complex quebradagrande complex ibagué batholith sodimentary formations (terciary) barroso formation arquia complex armenia earthquake (mw=6.2;25-01-99) 102 carlos a. vargas, andreas kammer, mireya valdes, carlos e. rodríguez alexander caneva, jhon j. sánchez, eliana arias, carlos a. cortes, héctor mora method preliminary interpretation of lithological and structural information suggests geological continuity at depth for some structures that are present in the region. further processing using the talwani method (talwani, 1959), of gravimetric and magnetometric data along an iterative adjust allowed to find a geological model that agrees with field observations. although the seismic experiment was designed to map a 45 km-deep section, the location of recording stations in urban areas with high noise levels made difficult quality results beyond 17 km depth. the selected seismic records were processed with a fourth order bandwidth filter (5 – 40 hz), properly edited and stacked to assemble a section, and then corrected by linear move out (lmo). the high topographical contrasts, the lateral variations of velocity and the complex geometry of the geological structures impeded an efficient application of normal move out (nmo) and migration techniques. the section was used to improve the geologic pattern and the corresponding gravimetric and magnetometric models until a model that agreed with the observations was found. results the adjustments of the geologic model with the geophysical information allowed the detection of a subduction structure probably affected by the prevalent orogenic processes from the mesozoic age in the nw andes of south america (figure 3a). it is possible to appreciate that the gravimetric and magnetometric information based on the physical characteristics of the rocks in this area (table 1), is well adjusted to the proposed geologic model. table 1. density and magnetic susceptibility values assigned to the rocks of the different geological units in the study area. geological formation lithology density (g/cm3) magnetic susceptibility (emu) 1. ibagué batholith granodiorite rocks 2.65 0.00030 2. cajamarca complex amphibolic green schists, graphitic schists, phyllites and marbles 2.75 0.00010 3. barroso formation basalts and tuffes 2.95 0.00250 4. subduction complex (arquía compex) serpentinites 3.00 0.00440 5. subduction complex (arquía compex) amphibolites 3.00 0.00011 6. quebradagrande complex basalts and tuffes 2.90 0.00900 7. oceanic paleo-plate peridotites 3.15 0.01000 8. continental crust amphibolites 2.80 0.00550 in this scenario, the cajamarca complex appears to be related with a synclinal structure limited by the pericos and san jerónimo faults, and immersed in a fractured continental crust. the structures associated to the paleo-subduction process have been transformed into wedges to form the quebradagrande complex and the barroso formation. probably the lithologies and structures located between the san jerónimo and silvia-pijao faults correspond to the limit area where an old subduction process happened. indeed, the presence of serpentine belts evidences the exhumation in the area where the old oceanic plate associated to the barroso formation penetrated. the seismic section (figures 3b and 3c) highlights the reflectors that define the wedge of the quebradagrande complex to depths beyond 10 km. on the other hand, the distribution of aftershocks related to armenia earthquake and reported by the seismological observatory of quindío (osq) suggests that the rupture process for this earthquake could be associated to weak planes in the old subduction zone (figure 4). therefore, the silvia – pijao fault may play a relevant role for the seismotectonic regime in region. the rupture process in the area, however, could be related to the same mechanisms responsible for the presence of serpentinite rocks with weak planes developed along the fibrous structure and parallel to the silvia pijao fault. 103 new geological and geophysical contributions in the section ibagué armenia, central cordillera colombia (a) (b) (c) figure 3. geophysical interpretation of the ibagué – armenia section. (a) density and magnetic susceptibility models. (b) seismic profile processed with lmo. distance in meters from calarcá town. (c) superposition of seismic profile onto geological model. 1. granodiorites: ibagué batholithe 2. metamorphites: (cajamarca complex) 3. basalts and tuffes (barroso formation) 4. serpentinites (subduction complex arquia complex) 5. amphibolites (subduction complex rquiacomplex) 6. basalts and tuffes (quebradagrande complex) 7. peridotites (oceanic paleo-plate) 8. amphibolites (continental crust) 104 carlos a. vargas, andreas kammer, mireya valdes, carlos e. rodríguez alexander caneva, jhon j. sánchez, eliana arias, carlos a. cortes, héctor mora figure 4. aftershocks (crosses) of the armenia earthquake (white star) superposed on the geologic section. the hypocentral solutions were reported by the osq. interpretation of the geologic section ibagué armenia this section exposes in short distances lithologies of different tectonic blocks that became the basement of the central cordillera (from east to west). this basement is composed of a metamorphic complex of low pressure temperature (p/t) conditions formed form rocks of the cajamarca complex, and rocks of high p/t ratio associated to gneiss and amphibolites of tierradentro (mccourt, 1984). both complexes are separated by the pericos fault. to the east of this fault the metamorphic basement of high p/t shows remnants of a palaeozoic cover (metasedimentary rocks of santa teresa). the ibagué batholith intruded along this contact during a magmatic event during the triassic – jurassic limit. these two rocky units form a landscape characterized by a moderate relief and a dendritic drainage with a trend nw-se (perpendicular to the direction of these units) that stands out. the cajamarca complex defines a wide synclinal near to the town of cajamarca, in this place the rocks are heavily metamorphosed and folded. our profile, however, does not have enough resolution to show the folds that are well documented in other works (nelson, 1962; restrepo-pace, 1992). near to the “la línea” sector (figure 1) this metamorphic sequence dips moderatelly toward east, defining a wide monoclinal (figure 3a). in this place appears the “la línea gneiss”, an intrusive body that constitutes a high to medium p/t ratio transition between basements. the landscape conformed by the cajamarca complex is dissected and is characterized by a rough topography. the drainage is segmented, with several families of alignments standing out. again in this zone there are notable perpendicular (nw-se) and parallel (sw-ne) alignments to the general direction. other alignments show trends n-s and e-w. as a very specific feature, this tectonic block is affected by an anomalously silicic volcanism that is best represented by cerro machín volcano. this volcano is formed for a complex of domes surrounded by thick sequences of piroclastic flow deposits. the crater has a diameter of 2.4 km and is occupied by dacitic domes with fumarolic activity (cepeda et al., 1995; cepeda and cortés, 1999). this volcano shows an activity restricted to the holocene (about 10.000 years) that is highly explosive of plinian type. the eruptions generated important piroclastic flows, which constitute continuous deposits in a diameter of 15 km from the volcanic amphitheater. between 105 new geological and geophysical contributions in the section ibagué armenia, central cordillera colombia the city of ibagué and the high parts of the toche, amaine and bermejón rivers seven piroclastic flows inserted in layers of ash whose ages vary among 45.000 to 800 years a.p. have been identified (méndez, 2001). near to “la línea”, the san jerónimo fault (also termed the aranzazu-manizales fault) is the limit between the metamorphic basement of the central cordillera and a tectonic complex that becomes a big flake roughly 5 km thick, composed of basaltic and piroclastic rocks of mid-cretaceous age belonging to the quebradagrande complex. these rocks are on the metamorphic basement (cajamarca complex) and are characterized by a marked dipping toward east and by multiple fractures. the intense fragile deformation, the inclination of the beds and the marked topographical relief of the western flank of the central cordillera makes this unit susceptible to the erosion, rock falls and big landslides. to the west of the quebradagrande complex appears the arquía complex, which constitutes the rock basement of the city of armenia and the town of calarcá. because of the strong inclination toward the east, the arquía complex belongs structurally to the quebradagrande complex. it is composed of basement flakes with thickness of less than 2 km that are inserted with serpentines. the basement is composed predominantly of hornblendic schists. to the east the quebradagrande complex is intruded by an elongated granodiorite body (the córdoba stock). the contact of this body with the quebradagrande complex is the silvia-pijao fault, with an area of deformation of several hundreds of meters wide. in the western area, the rocky basement is composed by basaltic rocks of the amaime formation that show a bigger flexion to the west of the cauca-almaguer fault and are dipping east. in the western foothill of santa bárbara range the rocks are affected by the quebradanueva fault that folded a paleogene sedimentary sequence called the “cinta de piedra” formation (mccourt et al., 1984). this sedimentary unit formed a barrier for the drainage that runs from the central cordillera and its existence explains how the volcanoclastic flows came from machín and tolima volcanoes and deposited its materials on the actual quindío glacis. “la línea” structures to the east of the san jerónimo fault the rocks of the central cordillera are dipping consistently toward the east. the dip of the sedimentary rocks of abejorral formation (gonzález and núñez, 1991) is affected by a set of normal faults that causes a sequence to be repeated (the “fault system effect”). figure 5 shows these fringes east of the campanario fault (a local name for the san jerónimo fault). the “fault system effect” is observed in a structural section some kilometres south from “la linea” cutting the cental cordillera in a domino-like fashion (figure 6). the spacing between faults varies between 1 to 2 km, and their continuity suggests a cut toward “la línea tunnel”. the “la linea” tunnel promises to be the most important civil work to improve the transportation system in the central region of colombia, thus considerations about the fault system are relevant. by extrapolating these faults as far as the bermejón river valley makes evident that at least four (five?) of these faults are cutting the route planned for this important ground transportation structure (figure 7). the prolongation of the geologic profile 15 kilometers west from our lithologic, structural, and geophysical observations, suggests that the cauca-almaguer fault could be the structure that facilitated the elastic energy dissipation related to the flexion process of the oceanic paleo-plate that was originally subducted (figure 8). it is possible that the continental rocks of tertiary age (“cinta de piedra” formation) allowed the deposition of sedimentary basins with n-s orientation that are visible today along the fault. 106 carlos a. vargas, andreas kammer, mireya valdes, carlos e. rodríguez alexander caneva, jhon j. sánchez, eliana arias, carlos a. cortes, héctor mora figure 5. geologic map of the region south of armenia city. figure 6. geologic section at south of the town of calarcá, where it is possible to see a set of antitetic faults that cut the central cordillera. legend alluvial fans upper eocene to recent continental deposits paleogene to neogene fluvial deposits middle eocene intrusive and protusive complex córdoba stocks (quarte diorites, gabbros) chlorite-actinolite schists, serpentinites quebradagrande group meta basalts and pyroclastics rocks with sedimentary layers basal non-volcanic sediments amaime group basalts, diabases, pyroclastics rocks, sedimentary layers abejorral formation siltstones, shales, pyroclastics rocks cajamarca group reverse fault anticline syncline 107 new geological and geophysical contributions in the section ibagué armenia, central cordillera colombia figure 7. topographic map of the region near “la línea” tunnel (line with open circles). at least four (five?)normal faults are cutting the route designed for this transport structure. figure 8. geologic section (interpreted and extended) west of the study area. 108 carlos a. vargas, andreas kammer, mireya valdes, carlos e. rodríguez alexander caneva, jhon j. sánchez, eliana arias, carlos a. cortes, héctor mora references • barrero, d. & vesga, c. j., 1976. mapa geológico del cuadrángulo k-9 armero y mitad sur del cuadrángulo j-9 la dorada. escala 1:100.000. ingeominas. bogotá. • cepeda, h., méndez, r., murcia, l. a. núñez, a. & monsalve, m.l., 1995. volcán cerro machín, departamento del tolima, colombia: pasado, presente y futuro. ingeominas, informe interno. • cepeda, h. & cortés, g. p. (1999). el cerro machín, un volcán holocénico explosivo en el centro de colombia. iv reunión técnica de la comisión de geofísica, departamento de geodesia y topografía, tucumánargentina. • duque-caro, h. (1990). the choco block in the northwestern corner of south america: structural, tectonostratigraphic, and paleogeographic implications. j. sout am. earth sci., 3, 71-84 p. • freymueller, j.t., kellog, j.n. & vega, v. (1993). plate motions in the north andean region. j. geophys. res. 98: 21853-21863. • gonzález, h. & núñez, a. (1991). mapa geológico generalizado del departamento del quindío, escala 1:100.000, memoria explicativa, 42 p., ingeominas, bogotá. • kammer, a., 1990. estructuras y deformaciones de los sedimentos cretáceos de la formación abejorral, departamento de antioquia, cordillera central. boletín de geología 34: 33-44, bucaramanga. • maya, m. & gonzález, h. (1995). unidades litodémicas de la cordillera central de colombia. ingeominas. boletín geológico. vol 35, 2 –3, 43-57. • mccourt, w. (1984). the geology of the central cordillera in the department of valle del cauca, quindío and nw tolima (sheets 243, 261, 280 and 300) ingeominas, cali, 49 p., inédito. • mc court, w.j., aspden, j.a. & brook, m. (1984). new geological and geochronological data from the colombian andes: continental growth by multiple accretion. j. geol. soc. london, 141 (5), 831 – 845. • mendez, r. a. (2001). informe sobre la geología y estratigrafía de flujos piroclásticos asociados al vcm. ingeominas, informe interno, inédito. manizales. 35 p. • monsalve, h. & vargas, c. a. (2002). el sismo de armenia, colombia (mw=6.2) del 25 de enero de 1999. un análisis telesísmico de ondas de cuerpo, observaciones de campo y aspectos sismotectónicos. revista geofísica – ipgh, 57:21-57. conclusion geologic and geophysical data used in this study suggest that the cajamarca complex defines a wide synclinal near the town of cajamarca. its structural configuration allows the assemblage to the west of a basement composed by wedges of basaltic and sedimentary rocks of the quebradagrande and arquía formations. new structural evidence supported by gravimetric, magnetrometric and seismic data suggests the existence of a bigger flexion of the barroso formation west of the cauca-almaguer fault which is dipping to the east. the flexion could be a factor for the deposition and folding of the “cinta de piedra” formation in the “santa bárbara” range. this sedimentary unit formed a barrier for the drainages that run from the central cordillera and its existence explain how the volcanoclastic flows came down from machín and tolima volcanoes and deposited their materials on the recent quindío glacis. acknowledgments we wish to thank sincerely many people and institutions that made this work possible. thanks to professors l. h. ochoa, l. castillo, l.a. montes and l.a ordoñez and their students of the graduate programs of geophysics and geotechnics and the undergraduate programs of geology and civil engineering of universidad nacional de colombia. much field work efforts were advanced by students of civil and electronic engineering departments of universidad del quindío under the direction of m. gómez, j. p. durán and c. correa. many thanks to corporación autónoma regional del quindío (crq), colciencias, and alcaldía de armenia for the financial support and particularly to a. lopezreina, a. duque and c. alvarez for their logistic and administrative help. we recognize the invaluable help in the field by ingeominas (manizales) especially to c. carvajal and m. londoño. thanks to iris-passcal inc. (bob greschke), universidad antonio nariño (dirección nacional de investigaciones) and l. sanchez of the german institute for geodesic research. 109 new geological and geophysical contributions in the section ibagué armenia, central cordillera colombia • nelson, h.w. (1962). contribución al conocimiento de la cordillera central de colombia sección entre ibagué y armenia. informe no. 1000. bol. geol.. vol. x (1-3): 161-202. ingeominas, bogotá. • núñez, a. & gonzález, h. (1991). memorias mapa geológico generalizado del departamento del quindío. ingeominas, bogotá. • restrepo-pace, p. (1992). petrotectonic characterization of the central andean terrane, colombia. j. south am. earth sci., 5, 97-116 p. • talwani, m. (1959). rapid gravity computations for two dimensional bodies with application to the mendocino submarine fracture zone., journal geophysical research., p 49-59. keywords: earthquake; geological hazards; longitudinal; land subsidence; deformation monitoring; simulator; sensing optical cable. palabras clave: terremoto; peligros geológicos; longitudinal; hundimiento de la tierra; monitoreo de la deformación; simulador; sensor de cable óptico. how to cite item xu, a., azarkhosh, h., & wu, e. (2020). monitoring method of longitudinal land subsidence and deformation in seismic geological disasters. earth sciences research journal, 24(3), 259-266. doi: https://doi. org/10.15446/esrj.v24n3.90290 monitoring of longitudinal land subsidence and deformation in seismic and geological hazards plays an important role in preventing and curing land collapse, land subsidence, and ground cracks. in this paper, a distributed monitoring model experiment on vertical land subsidence and deformation of seismic and geological hazards is carried out by brillouin optical frequency-domain analysis technology (botda). by using the self-made indoor longitudinal land subsidence and deformation simulation box, different intensity seismic ground is simulated by air bag. distributed optical fibers are used to monitor the longitudinal land subsidence and deformation during different intensity seismic and geological disasters. according to different intensity seismic and geological disasters, distributed sensing optical fibers cooperate with the ground to compress or stretch longitudinally and obtain the data of longitudinal land subsidence and deformation. the correction coefficient is introduced to modify the monitoring data of confining pressure-sensing optical fiber and complete the precise monitoring of vertical land subsidence and deformation in seismic and geological hazards. the experimental results show that this method can monitor the vertical ground subsidence and deformation of seismic and geological hazards under different conditions, and the monitoring efficiency and cost are superior to gps and inertial monitoring methods, and the practical application value is high. abstract monitoring method of longitudinal land subsidence and deformation in seismic geological disasters método de monitoreo de hundimiento y deformación longitudinal de la tierra en desastres sísmicos geológicos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n3.90290 el monitoreo del hundimiento longitudinal de la tierra y la deformación en los riesgos sísmicos y geológicos juega un papel importante en la prevención y corrección del colapso, el hundimiento de la tierra y las grietas del suelo. en este documento, la tecnología de análisis de dominio de frecuencia óptica de brillouin (botda) sirve para llevar a cabo un experimento de modelo de monitoreo distribuido sobre hundimiento vertical de la tierra y deformación de riesgos sísmicos y geológicos. mediante el uso de la caja de simulación de deformación del terreno longitudinal interior de fabricación propia se simula un suelo sísmico de diferente intensidad mediante una bolsa de aire. las fibras ópticas distribuidas se utilizan para controlar el hundimiento y la deformación del terreno longitudinal durante desastres sísmicos y geológicos de diferente intensidad. de acuerdo con estos movimientos, las fibras ópticas de detección distribuida cooperan con el suelo para comprimir o estirar longitudinalmente y obtener los datos de hundimiento y deformación del terreno longitudinal. el coeficiente de corrección se introduce para modificar los datos de monitoreo de la fibra óptica de detección de presión de confinamiento, y completar el monitoreo preciso del hundimiento y deformación vertical de la tierra. los resultados experimentales muestran que este método puede monitorear el hundimiento vertical del suelo y la deformación de los riesgos sísmicos y geológicos en diferentes condiciones. la la eficiencia y el costo del monitoreo son superiores a los métodos de monitoreo inercial y gps, y el valor práctico de la aplicación es alto. resumen record manuscript received: 20/05/2019 accepted for publication: 03/02/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 3 (september, 2020): 259-266 aimei xu, hojatallah azarkhosh2, erjun wu2 1school of civil engineering and architecture, xinyu university, xinyu, 338000, china 2college of civil and transportation engineering, hohai university, nanjing, china *corresponding author: yanyy1209@126.com g e o l o g ic a l h a z a r d s mailto:yanyy1209@126.com 260 aimei xu, hojatallah azarkhosh, erjun wu introduction land subsidence and deformation is a geological phenomenon in which the surface rock and soil subsides downward under the action of natural or manmade factors and forms a subsidence pit on the ground. according to the causes of subsidence deformation, land subsidence and deformation can be divided into natural subsidence deformation and artificial subsidence deformation. natural subsidence and deformation occur under the influence of natural environmental factors, and the main inducements include earthquakes, rainfall, and so on. artificial subsidence deformation refers to the surface subsidence deformation caused by overexploitation of groundwater and drainage of mine pits. with the rapid development of the city, the underground space of the city has been continuously developed. the underground of most cities is densely covered with various pipelines, tunnels, and other underground “holes” (tan et al., 2016). these underground “voids” are prone to collapse and then cause land subsidence and deformation under the action of external factors such as ground load and disturbance of underground engineering construction. in view of the increasingly serious development trend of land subsidence and deformation, monitoring of vertical land subsidence and deformation has become a new research hotspot, and many scholars at home and abroad have done a lot of research on it. common monitoring technologies include insar, gps, ground-penetrating radar, and transient electromagnetic methods. however, land subsidence and deformation are a geological hazard phenomenon of vertical compression and horizontal deformation after the coordinated deformation of soil structure. it is difficult for traditional point monitoring technology to fully understand the deformation characteristics of the whole and local layers of the soil-structure system, the compatibility of deformation among different layers, and the law of development. it is also difficult to predict the deformation process of the soil in the ground subsidence. distributed fiber optic sensing (dfos), as a new monitoring technology, has the advantages of high spatial resolution, high precision, and high repeatability. it can effectively solve the problems of traditional monitoring technology (diao et al., 2016). at present, this technology has been widely used in the monitoring of slope, embankment, ground subsidence, and ground fissures. distributed optical fiber sensing technology can sense and measure one-dimensional deformations along the direction of optical fibers. in practical monitoring, various conversion methods or optical fiber layouts are often needed to achieve the purpose of monitoring the required deformations of the measured objects (bornyakov et al., 2016). after more than ten years of research and accumulation, the author’s research group has developed a variety of technology-based deformation monitoring methods (sayle et al., 2016). the following mainly introduces the distributed optical fiber sensing monitoring method of soil longitudinal subsidence and deformation. distributed optical fiber monitoring technology is based on botda (brillouin optical time-domain analysis) monitoring technology. botda is a time-domain analysis technology of brillouin scattering light based on stimulated brillouin scattering. its sensing principle is that a certain frequency of pulsed light is injected into one end of the fiber, and the pulsed light interacts with the acoustic phonon in the fiber, resulting in brillouin scattering. the light-receiving unit and signal acquisition unit of botda mainly analyze and process the back brillouin scattering light which returns to the injecting end of the pulsed light along the original optical fiber path. the optical signal is converted into an electrical signal by the photodiode, and the electrical signal is amplified by a broadband amplifier into the heterodyne receiver. the digital signal is averaged and the scattering spectrum of each sampling point along the optical fiber is obtained. brillouin scattering light is affected by strain and temperature, that is, when the fiber is stretched (compressed) along the axis or the temperature along with the fiber changes, the corresponding changes in the stretching (compressed) or temperature range of the fiber have a good correlation with brillouin frequency shift (xu et al., 2018). therefore, the brillouin frequency drift can accurately reflect the temperature and strain changes of optical fibers. in this paper, bofda technology is used to conduct a distributed monitoring model test of vertical land subsidence and deformation in seismic and geological hazards. using the self-made physical model of land subsidence and deformation (tehrani et al., 2016), the longitudinal deformation of different types and thicknesses of land in the process of subsidence and deformation is monitored by distributed optical fiber. the experimental results are compared and analyzed to verify the feasibility of distributed optical fiber sensing technology in land subsidence monitoring (kim et al., 2016). it also provides a reference for further research and application of distributed optical fiber sensing technology. materials and methods test material (1) sand the sand used in this experiment is taken from the jiangning area (song et al., 2016), nanjing. the grain size distribution curve of sand is shown in figure 1, and its physical and mechanical parameters are shown in table 1. figure 1. diameter distribution curve of sand table 1. the physical parameters of the sand proportion maximum dry density/ (g/cm3) minimum dry density/ (g/cm3) maximum porosity minimum porosity cohesion/ kpa internal friction angle/(º) 2.64 1.69 1.35 0.97 0.14 0 30.12 (2) clay the clay used in this experiment is xiashu soil (li et al., 2016), which is widely distributed in the nanjing area. the natural state is yellowish-brown, mainly plastic to rigid plastic, with compact structure (chen et al., 2017). its physical and mechanical parameters are shown in table 2. table 2. the physical parameters of the caly relative density bulk density liquid limit plastic limit plasticity index natural moisture content cohesion/ kpa internal friction angle/(º) 2.72 19.8 30.3 17.5 12.8 20-28 15.8 6.3 (3) gasbag in this experiment, gasbags are used to simulate seismic and geological hazards. the gasbag is made of black rubber, ellipsoid, with a long axis of 20 cm, a short axis of 15 cm, and a height of 15 cm. it is connected with an air pump and a pressure valve. this kind of gasbag has the characteristics of good air tightness, strong elasticity, high wear resistance, high acid resistance, and aging resistance (lu et al., 2018). it is flattened after natural ventilation, which meets the test requirements. 261monitoring method of longitudinal land subsidence and deformation in seismic geological disasters effectively and comprehensively study the mechanism and impact of ground subsidence. in the experiment, two main parameters are changed: the thickness of the upper soil layer and the content of clay. filling thickness (h) refers to the thickness of the soil on the top of the optical fiber. the specific parameters are 5 cm and 10 cm. clay content refers to the content of clay in the upper layer of optical fiber. the specific parameters are 0% and 5%. the monitoring variables of land subsidence deformation are vertical deformation of soil, including longitudinal compression and rebound. by laying distributed sensing optical fibers vertically in ground subsidence monitoring, appropriate backfilling is selected according to stratum lithology. with the increase of laying depth, the lateral earth pressure will increase continuously, and the deformation coupling between distributed sensing fiber and soil will be stronger (jiang & zhang, 2017). therefore, when the soil is compressed or rebounded, the distributed sensing fiber will co-occur axial compression or tension. the distributed sensing fiber data will be collected regularly to obtain the strain increment of distributed sensing fiber (shi et al., 2017), and the product will be used. the deformations of distributed sensing optical fibers can be obtained separately, as shown in formula 1. longitudinal land subsidence and deformation information can be obtained according to the integral results. ∆l l dl i i d i l l d l l = ( ) = ( ) + +( ) = ∑∫ ε ε − 1 2 1 2 1 1 2 (1) where l is the longitudinal tension or contraction /m between the sensing cable l1 and l2; (l) is the longitudinal strain at the sensing cable lm; and d is the step length /m of data acquisition by sensing cable. test steps (1) laying cushion: dry sand with a thickness of 5 cm is laid on the bottom of the model box with sand rain normal direction and leveling the surface (li et al., 2018). (2) installation of gasbags: the three full-air gasbags in the model box are placed. from left to right, they are no. 1, no. 2 and no. 3, and the spacing of the gasbags is 40 cm. the air duct connecting the gasbag is led out of the model box to control the air pressure inside the gasbag during the test. (3) sand layer landfill: dry sand is laid around the gasbag by sand rain method until it stops 5 cm above the top of the gasbag and the surface is leveled. (4) embedding of optical fibers: the optical fibers are passed through the reserved holes and welded with jumpers. the fused optical fiber is connected with bofda optical fiber monitoring instrument. two heat sources are set at both ends of the model box, and the monitoring software is run to determine the specific location of the optical fiber test section. after the fixed point is completed, the optical fibers are fixed and pre-stretched on the outside of both ends of the model box (zhang et al., 2017). (5) test soil layer landfill: the upper test soil is filled into the model box. after the fill reaches the predetermined height, the soil surface is leveled and placed for 12 hours. (6) land subsidence simulation test: before the beginning of the test, the pre-stress applied on the optical fiber is removed, the initial data is recorded, the no. 2, no. 1, and no. 3 gasbags are filled in turn, and the filling interval is 30 minutes. the data before each filling is measured, and the last measurement is 30 minutes after the completion of the no. 3 gasbag filling. in the process of aeration and measurement, the change of soil surface is photographed, and the change of soil surface during the test is recorded accurately. (7) subsidence information record: after data acquisition, excavates along with the optical fiber until the bottom of the gasbag, obtain the longitudinal profile of the soil after deflation, and observe the shape of the cavern and record and describe it. test conditions tests are carried out under the following four working conditions, and the soil conditions under different working conditions are as shown in table 3. condition 1: gasbag 2 deflates naturally and collects data after 30 minutes. (4) in this experiment, two kinds of sensing optical cables are used: a. polyurethane strain sensing optical cable with a variable diameter (2mm cd), is a polyurethane optical cable with 2mm diameter, and 25 mm length quartz glass tube at 20 cm intervals. a concave-convex variable diameter sensing optical cable (dyshlyuk et al., 2017) is fabricated, as shown in figure 2; b. polyurethane strain sensing optical cable (2mm cd for short) is a conventional optical cable with a diameter of 2 mm, that is, a sensing optical cable without a quartz glass tube. figure 2. polyurethane sensing optical fiber with a changed diameter monitoring program the model used in this experiment is a self-made indoor simulation box for longitudinal land subsidence and deformation (foumelis et al., 2016), which length is 200 cm, width is 50 cm and height is 50 cm. three air bags are buried in the set position of the box (deverel et al., 2016), as shown in figure 3: figure 3. indoor longitudinal land subsidence deformation simulator according to the test requirements, layered backfill soil samples are buried with distributed sensing optical fibers in the upper soil of the gasbag during the backfill process. after backfilling is completed, the air pressure inside the gasbag is increased at a certain speed through the external air pressure switch, so that the soil around the gasbag will be deformed due to the excessive support at the lower part, and then land subsidence will occur on the surface of the gasbag (zhang et al., 2017). by changing different test variables, this experiment can simulate different seismic and geological engineering conditions, and then 262 aimei xu, hojatallah azarkhosh, erjun wu condition 2: gasbag 1 deflates naturally and collects data 30 minutes later. condition 3: gasbag 3 deflates naturally and collects data after 30 minutes. conditions 4: gasbag 1-3 deflates naturally and collects data after 60 minutes. table 3. detailed contents of different working conditions working condition thickness sand content/% clay content/% 1 5 100 0 2 10 100 0 3 5 95 5 4 10 95 5 data correction method for sensing optical fiber according to the indoor test results, when the sensing fiber is embedded in a certain depth of soil, the confining pressure of the soil will affect the testing effect of the sensing fiber. therefore, when analyzing the sensing optical fiber data in deep soil, it is necessary to introduce correction coefficient to modify the monitoring data of sensing optical fiber under different confining pressures (tao et al., 2019), as shown in formula 2 and 3: ∆ ∆l k ls s= (2) where, ls is the longitudinal tension or contraction of sensing optical fibers in soil/m, and ks is the correction factor of coupling between sensing optical fibers and soil deformation. k k k i s is i i = + +( ) + 2 1 1 ≤ < (3) ki is the drawing efficiency of sensing fiber and soil when confining pressure is impa. the modified coefficient ks is introduced to modify the monitoring data of sensing optical fiber, and the corrected monitoring data are accurate monitoring results of vertical land subsidence and deformation. results in the course of the test, the changes in the soil surface in the model box at each stage are fully recorded. it is found that the upper soil has a great influence on the surface morphology of the ground subsidence, and different test phenomena will appear. the vertical subsidence and deformation of the soil surface under four working conditions are described by figure 4 (a), figure 4(b), figure 4(c) and figure 4(d), respectively. it can be seen that different seismic actions have different effects on the vertical subsidence and deformation of the soil surface. the longitudinal subsidence and deformation of soils under different working conditions will be monitored by the proposed method as shown in table 4 (yang et al., 2020). table 4 shows the changes in soil longitudinal subsidence and deformation under different working conditions monitored by the proposed method. among them, the results of soil subsidence and deformation measured under working conditions 2 and 3 have less valid data. in order to verify the correctness of the results of the proposed method in monitoring the vertical land subsidence and deformation in seismic and geological hazards, the results of soil subsidence and deformation monitored by the proposed method under working conditions 1 and 4 are compared with the actually measured soil subsidence. the results of deformation are compared and described in figure 5 and figure 6 respectively. the average vertical subsidence of soil under working conditions 1 and 2 measured by two optical fibers in this paper is described in table 5. from the data in table 5, it can be seen that the method can accurately monitor the (a)longitudinal land subsidence deformation result of working condition 1 (b)longitudinal land subsidence deformation result of working condition 2 (c)longitudinal land subsidence deformation result of working condition 3 (d)longitudinal land subsidence deformation result of working condition 4 figure 4. longitudinal subsidence and deformation of soil surface under four working conditions 263monitoring method of longitudinal land subsidence and deformation in seismic geological disasters vertical subsidence and deformation of soil structure at different locations under different seismic and geological hazards (wu et al., 2020). table 5. average longitudinal subsidence deformation of soil layer measured by this method/mm working condition working condition 1 working condition 4 0 cm -20cm 4.5 5.05 21 cm -40 cm 5.6 25.5 41 cm -60 cm 4.1 8.3 61 cm -80 cm 13 6.75 81 cm -100 cm 17.7 33.7 101 cm -120 cm 12 9.45 121 cm -140 cm 6.6 6.25 141 cm -160 cm 3.85 22.8 161 cm -180 cm 0 12.55 181 cm -200 cm 0 6.25 from the curves of figure 5 and figure 6, it can be seen that the results of longitudinal land subsidence and deformation monitored by the proposed method are almost identical with those measured in practice, which indicates that the method has high accuracy in actual subsidence and deformation monitoring, and can be widely used in areas with a high incidence of seismic and geological hazards, so as to warn the displacement of vertical land subsidence and to take timely measures to reduce personnel and property losses (gao et al., 2018). based on the accuracy of monitoring the vertical land subsidence deformation in seismic geological hazards by the proposed method, the monitoring efficiency and economy need to be verified. a qualified monitoring method for vertical land subsidence deformation in seismic geological hazards needs both high precision and high-efficiency deformation monitoring results and low monitoring cost. the experiment is based on gps technology to monitor the vertical ground subsidence deformation in seismic geological hazards (cheng et al., 2020). as an experimental reference, the methods of vertical land subsidence and deformation monitoring based on inertial technology for seismic and geological hazards are compared in terms of time and cost of precise vertical land subsidence and deformation monitoring for seismic and geological hazards under work conditions 1 and 4. the results are shown in table 6. from the data in table 6, it can be seen that the time and cost of the proposed method for monitoring vertical land subsidence and deformation of seismic and geological hazards at different distances are lower than those of the other two methods. the time and cost of the proposed method for monitoring subsidence and deformation at different distances of working conditions 1 and 4 are respectively 0.91 million yuan and 0.41 s and 0.75 million yuan and 0.41 s. the time and cost of gps technology for monitoring subsidence and deformation at different distances of working conditions 1 and 4 are 0.91 million yuan and 0.41 s, respectively. the maximum time and cost of inertial technology monitoring for subsidence and deformation in table 4. longitudinal subsidence and deformation of soil under different conditions working condition 1 2 3 4 l1 optical fiber mean subsidence deformation/mm l2 optical fiber mean subsidence deformation/mm l1 optical fibermean subsidence deformation/mm l2 optical fibermean subsidence deformation/mm l1 optical fibermean subsidence deformation/mm l2 optical fibermean subsidence deformation/mm l1 optical fibermean subsidence deformation/mm l2 optical fibermean subsidence deformation/mm 0 cm -20cm 5.6 3.4 0 0 0 0 4.5 5.6 21 cm -40 cm 6.6 4.6 0 0 0 0 23.4 27.6 41 cm -60 cm 4.8 3.4 18.8 6.7 0 0 8.9 7.7 61 cm -80 cm 15.7 10.3 0 0 0 0 7.8 5.7 81 cm -100 cm 20.7 14.7 0 0 8.8 6.5 32.7 34.7 101 cm -120 cm 11.5 12.5 0 0 11.7 12.9 10.2 8.7 121 cm -140 cm 8.7 4.5 0 0 0 0 5.8 6.7 141 cm -160 cm 4.3 3.4 0 0 0 0 21.8 23.8 161 cm -180 cm 0 0 0 0 0 0 11.3 13.8 181 cm -200 cm 0 0 0 0 0 0 6.7 5.8 figure 5. the method presented in this paper and the actual measured subsidence and deformation results of the soil under working condition 1 264 aimei xu, hojatallah azarkhosh, erjun wu working conditions 1 and 4 are 12.3 million yuan and 15.7 million and 15.4 million yuan and 1.34 s, respectively. the maximum time and cost of inertial technology monitoring for subsidence and deformation in working conditions 1 and 4 are 15.4 million yuan and 1.52 s and 22.4 million yuan and 0.95 s, respectively. the time and cost of three methods for monitoring subsidence and deformation in working conditions 1 and 4 are summarized as: the proposed method<inertial technology monitoring method < gps technology monitoring method. it has strong economic benefits and monitoring speed, high practical value, and can be widely used in seismic and geological hazard monitoring, with a strong economy. figure 6. the method presented in this paper and the actual measured subsidence and deformation results of the soil under working condition 4 table 6. monitoring results of three methods method working condition 1 working condition 2 the method in this paper monitoring method of gps technology inertial technology monitoring method the method in this paper monitoring method of gps technology inertial technology monitoring method cost/ten thousand yuan time/s cost/ ten thousand yuan time/s cost/ ten thousand yuan time/s cost/ ten thousand yuan time/s cost/ ten thousand yuan time/s cost/ ten thousand yuan time/s 0 cm -20cm 0.65 0.35 1.55 1.26 1.23 0.59 0.73 0.26 1.28 1.02 2.26 0.65 21 cm -40 cm 0.84 0.25 1.26 1.35 1.5 0.68 0.62 0.32 1.54 1.13 2.15 0.54 41 cm -60 cm 0.91 0.31 1.54 1.45 1.02 0.78 0.55 0.36 1.26 1.24 1.59 0.95 61 cm -80 cm 0.65 0.41 1.58 1.35 1.35 1.52 0.64 0.35 1.14 1.16 2.24 0.85 81 cm -100 cm 0.65 0.26 0.64 1.21 1.36 1.05 0.57 0.24 1.05 1.15 1.68 0.86 101 cm -120 cm 0.68 0.34 1.59 1.57 1.24 0.65 0.65 0.26 0.62 1.25 1.54 0.87 121 cm -140 cm 0.64 0.35 1.88 1.24 1.52 0.54 0.59 0.35 0.54 1.34 2.14 0.83 141 cm -160 cm 0.67 0.31 1.54 1.23 1.36 0.64 0.64 0.34 0.69 1.06 2.15 0.87 161 cm -180 cm 0.68 0.26 2.01 1.45 1.54 0.85 0.75 0.37 0.85 1.04 1.64 0.64 181 cm -200 cm 0.64 0.28 2.23 1.33 1.36 1.03 0.62 0.41 1.32 1.26 1.82 0.87 total 7.01 3.12 15.82 13.44 13.48 8.33 6.36 3.26 10.29 11.65 19.21 7.93 discussion in this paper, botda is used to monitor the longitudinal ground subsidence and deformation of the simulated seismic and geological hazards. the soil model box is used to simulate the surface environment. three airfilled gasbags are installed in the soil of the model box to simulate the seismic and geological hazards with different intensities. distributed sensing optical fibers are installed in the soil to obtain the longitudinal ground subsidence and deformation. the results of longitudinal land subsidence and deformation under four different working conditions show that the proposed method can monitor the subsidence and deformation at different locations of monitoring objects. the results are compared with the measured values. from figure 5 and figure 6, it can be seen that the average longitudinal land subsidence and deformation of soil at different distances from working condition 1 and 4 are almost identical with the measured values, which shows that the proposed method can monitor the subsidence and deformation of soil at different distances from working condition 1 and 4. in addition, by comparing the proposed method with inertial technology monitoring method and gps technology monitoring method in terms of the monitoring efficiency and cost, it can be seen that the proposed method can not only obtain high-precision subsidence deformation results but also improve monitoring efficiency, reduce the cost of the monitoring process and enhance economic benefits. the results of subsidence and deformation monitoring of indoor model box simulation tests show that reasonable monitoring methods can accurately detect soil deformation by sensing optical cable, and the monitoring accuracy is better than the existing conventional monitoring technology. how to apply distributed optical fiber sensing technology to the monitoring of geological hazards such as land subsidence and ground fissures more reasonably and efficiently and develop it into a mature and popular monitoring method is the main content of the next research. according to the monitoring situation in this paper, the development direction of optical fiber sensing monitoring technology for vertical land subsidence and ground fissures is summarized as follows: (1) introducing fiber bragg grating sensing technology fiber bragg grating (fbg) sensing technology has very high accuracy, but also has the unique advantages of anti-interference and corrosion resistance. 265monitoring method of longitudinal land subsidence and deformation in seismic geological disasters in the monitoring of geological hazards such as land subsidence, the distributed optical fiber sensing technology, and fiber bragg grating sensing technology are combined to achieve the complementary advantages of the two technologies and achieve the purpose of efficient monitoring. (2) developing vertical buried monitoring methods the vertical buried monitoring method can realize the distributed monitoring of soil compression and deformation. how to solve the problem of shallow surface temperature compensation is the focus of the next step of the proposed method. if the target monitoring layer is deeply buried, the temperature compensation can be avoided. the construction technology of this monitoring method is basically mature. the next step is to select the location with subsidence monitoring significance for drilling construction, lay the sensing cable through the main subsidence compression layer, and realize the vertically distributed monitoring of soil compression deformation. at the same time, the monitoring data are combined with the geological environment information to carry out the research on the mechanism and development law of land subsidence based on the vertical deformation distributed measured data. (3) improving and optimizing the fixed-point monitoring party the fixed-point monitoring method can realize the monitoring of soil micro-deformation in-ground fissure area. in terms of the effective monitoring length of monitoring points, the dislocation deformation of soil mass on both sides of the crack area can be accurately detected, and the monitoring accuracy can reach the fixed point distance. the monitoring effect is better than the direct buried and pipe-type monitoring methods used at the same time. therefore, in the next step of distributed optical fiber ground fissure monitoring, we should focus on the fixed-point monitoring method, and further optimize and improve the monitoring method. the research on the proposed method can be carried out from indoor tests and field monitoring. the main work can be carried out around the following points: a. the monitoring effect of this method under different modes of dislocation deformation of ground fissures is studied through laboratory tests, and the design and laying methods of anchorage components are improved to realize directional and quantitative monitoring of soil movement. b. the monitoring accuracy of this method is further improved by testing various sensing optical cables and fixing modes. c. suitable monitoring sites are chosen to conduct a wide range of distributed monitoring of ground fissures with different spacing. on the basis of monitoring existing fissures, pre-warning, and monitoring of safety in fractured areas are carried out. conclusions ground, as the main transportation place, plays an important role in the development of the national economy and society. ground safety is more directly related to people’s lives and property security. the ground is usually used for a long time, often affected by various loads and external seismic and geological disasters. the structure has different degrees of deformation. the vertical land subsidence deformation in the early stage of seismic and geological disasters is very weak, and the location is relatively hidden. the traditional detection methods are difficult to achieve the task of precise monitoring of the vertical land subsidence deformation under seismic and geological disasters. distributed optical fiber sensing technology based on brillouin optical frequency domain analysis can make up for the shortcomings of traditional methods, and can accurately measure the average vertical subsidence deformation in different distances of the ground to be monitored. it has good application prospects in the monitoring of vertical land subsidence deformation and improves the accuracy of vertical land subsidence deformation. compared with gps technology and inertial technology, the proposed method is more effective in monitoring vertical land subsidence deformation. the rate is higher and the cost of monitoring is lower. reference bornyakov, s. a., miroshnichenko, a. i., & salko, d. v. (2016). diagnostics of the preseismogenic state of heterogeneous media according to deformation monitoring data. doklady earth sciences, 468(1), 481-484. chen, z., shao, x., he, x., wu, j., xu, x., & zhang, j. (2017). noninvasive, three-dimensional full-field body sensor for surface deformation monitoring of human body in vivo. journal of biomedical optics, 22(9), 1-10. cheng, h. y., & liu, y. n. (2020). an improved rsu-based authentication scheme for vanet. journal of internet technology, 21(4). deverel, s. j., ingrum, t., & leighton, d. (2016). present-day oxidative subsidence of organic soils and mitigation in the sacramento-san joaquin delta, california, usa. hydrogeology journal, 24(3), 569586. diao, x., wu, k., hu, d., li, l. & zhou, d. (2016). combining differential sar interferometry and the probability integral method for threedimensional deformation monitoring of mining areas. international journal of remote sensing, 37(21), 5196-5212. dyshlyuk, a. v., makarova, n. v., vitrik, o. b., kulchin, y. n., & babin, s. a. (2017). features of monitoring deformation processes in ferroconcrete designs with application of the reflectometer method of recording signals of fiber bragg gratings. measurement techniques, 60(7), 701-705. foumelis, m., papageorgiou, e., & stamatopoulos, c. (2016). episodic ground deformation signals in thessaly plain (greece) revealed by data mining of sar interferometry time series. international journal of remote sensing, 37(16), 3696-3711. gao, n. s., cheng, b. z., hou, h., & zhang, r. h. (2018). mesophase pitch based carbon foams as sound absorbers. materials letters, 212, 243246. doi:10.1016/j.matlet.2017.10.074 jiang, l. f., & zhang, z. w. (2017). analysis and design of a full-bridge cll resonant converter with double resonant tanks. journal of power supply, 15(1), 92-98. kim, d., kim, h., shin, k. j., & seo, h. (2016). the effect of concrete deformation on displacement of an axially loaded drilled shaft. marine georesources & geotechnology, 34(2), 116-126. li, g. w., nguyen, t. n., amenuvor, a. c. (2016). subsidence prediction of surcharge preloaded low embankment on soft ground subjected to cyclic loading. marine georesources & geotechnology, 34(2), 154161. li, w. j. wang, y. c. xu, b., zhang, z., yu, h., & wu, q. (2018). the research on network management and online monitoring for communication network for smart distribution and consumption network. automation & instrumentation, (4), 27-30. lu, y., ke, c. q., zhou, x., wang, m., lin, h., chen, d., & jiang, h. (2018). monitoring land deformation in changzhou city (china) with multiband insar data sets from 2006 to 2012. international journal of remote sensing, 39(4), 1151-1174. sayle, k. l., hamilton, w. d., cook, g. t., ascough, p. l., gestsdóttir, h., & mcgovern, t. h. (2016). deciphering diet and monitoring movement: multiple stable isotope analysis of the viking age subsidence at hofstaðir, lake mývatn, iceland. american journal of physical anthropology, 160(1), 126–136. shi, l. p., liu, d., li, m. h., zong, z. d. (2017). distributed dc power system based on parallel intelligent battery components. chinese journal of power sources, 41(4), 612-615. song, s. g., li, s. c., li, l. p., zhang, q. q., wang, k., zhou, y., & liu, h. l. (2016). study on longitudinal deformation profile of rock mass in a subsea tunnel. marine georesources & geotechnology, 34(4), 376-383. tan, j., cui, z. d., & yuan, l. (2016). study on the long-term subsidence of subway tunnel in soft soil area. marine georesources & geotechnology, 34(5), 486-492. tao, z., wang, y., zhu, c., xu, h., li, g., & he, m. (2019). mechanical evolution of constant resistance and large deformation anchor cables and their application in landslide monitoring. bulletin of engineering geology and the environment, (3), 1-17. 10.1016/j.matlet 266 aimei xu, hojatallah azarkhosh, erjun wu tehrani, j. n., mcewan, a., & wang, j. (2016). th-cd-207a-05: lung surface deformation vector fields prediction by monitoring respiratory surrogate signals. medical physics, 43(6), 3880-3880. wu, z., liu, y. n., & jia, x. x. (2020). a novel hierarchical secret image sharing scheme with multi-group joint management. mathematics, 8(3), 12. doi:10.3390/math8030448. xu, c., chen, j., zhu, h., liu, h., & lin, y. (2018). experimental research on seafloor mapping and vertical deformation monitoring for gas hydrate zone using nine-axis mems sensor tapes. ieee journal of oceanic engineering, pp(99), 1-12. yang, f., yang, f., wang, g. y., kong, t., wang, h., & zhang, c. s. (2020). effects of water temperature on tissue depletion of florfenicol and its metabolite florfenicol amine in crucian carp (carassius auratus gibelio) following multiple oral doses. aquaculture, 515, 9. doi:10.1016/j. aquaculture.2019.734542 zhang, y. x., guo, a. t., & wen, y. m. (2017). effects of different forms of iron and manganese on generation of h2o2 and degradation of sdbs in natural water biofilms under illumination. journal of jilin university (science edition), 55(2), 458-464. zhang, y. zhao, x. tu, h., lu, b., & li, y. (2017). a queue management mechanism to improve 3d video transmission in wlan. journal of china academy of electronics and information technology, 12(3), 246-250. 10.3390/math 10.1016/j.aquaculture 10.1016/j.aquaculture earth sciences research journal earth sci. res. sj. vol. 17, no. 1 (june, 2013): 53 60 biogeochemical and engineering characteristics of soils and groundwater around a dumpsite odukoya am1, oresanya o2 and abimbola, a.f3 1 department of geosciences, university of lagos, nigeria. email: sesanbiiodun@yahoo.com 2 lagos state waste management authority, lagos, nigeria 3 department of geology, university of ibadan, nigeria geochemestry an active dumpsite in lagos south-western nigeria was monitored to test natural attenuation efficiency in reducing contaminants by determining soil and groundwater biogeochemical and engineering properties within and around the site. from the casagrande plasticity chart, the soil could be classified as clay or silt having intermediate to high plasticity; unsaturated zone thickness was 10-20m and permeability was low (1.96x10-5 to 41.8x10-5 m/s). appropriate microorganisms, such as bacteria, viruses and fungi which are required for natural attenuation, were naturally present. nutrients such as calcium, copper, magnesium, manganese, potassium, sodium and zinc, as well as electron donor (organic carbon) and electron acceptor (oxygen, nitrate, sulphate and iron), were also present in various amounts for effective natural attenuation. all the parameters analysed in water samples came within environmental protection agency standards, except fe, na, cl, no3, al, ba, ni, total bacteria and total fungi. phenol and total viral count were not detected in the two boreholes studied, but they showed values as high as 10,000 mpn/100ml and 230 cfu/100ml, respectively, in leachate. the concentration of contaminants in the soil was very high and inversely proportional to depth. un basurero activo en lagos, al suroccidente de nigeria, fue monitoreado para medir la eficiencia de atenuación natural en contaminantes por la determinación del suelo, las características biogeoquímicas de las aguas subterráneas y las propiedades de ingeniería adentro y afuera del vertedero de desechos. según la escala de plasticidad de casagrande, el suelo se podría clasificar como arcilloso o cienoso, con plasticidad de intermedia a alta; el espesor de la zona sin saturar fue de 10-20 m y la permeabilidad fue baja (1.96x10-5 to 41.8x10-5 m/s). están presentes los microorganismos apropiados para la atenuación natural como las bacterias, virus y hongos. nutrientes como el calcio, el cobre, magnesio, manganeso, potasio, sodio y zinc, al igual que donantes de electrones (carbón orgánico) y receptores de electrones (oxigeno, nitrato, sulfato e hierro) se encuentran presentes en diferentes cantidades para una efectiva atenuación natural. los parámetros analizados en las muestras de agua están dentro de los estándares de la agencia de protección ambiental, excepto fe, na, cl, no3, al, ba, ni, el total de las bacterias y el total de hongos. el fenol y el total viral no fueron detectados en los dos hoyos estudiados, pero estos mostraron valores de 10,000 mpn/100ml y 230 cfu/100ml en los lixiviados. la concentración de contaminantes en la tierra es muy alta e inversamente proporcional a la profundidad. key words: leachate, natural attenuation, nutrient, contaminant, electron donor. palabras claves: lixiviados, atenuación natural, nutrientes, contaminantes, donantes de electrones record manuscript received: 26/01/2013 accepted for publication: 04/04/2013 abstract resumen introduction leachate production as a by-product of organic and inorganic decomposition in landfills poses a serious threat if it becomes released to the environment (kimmel and braids, 1974; baedecker and back, 1979; arneth et al., 1989; jankowski, 1997). the waste disposal sites and landfills in the study area were neither properly designed nor constructed. after some years, a dumpsite undergoes biologically-, chemically-, geologicallyand hydro-geologically-mediated changes, resulting in weathering; consequently, it becomes a source for any aquiferous units close to them becoming polluted (altindag and yigit, 2005; adeniyi et al., 2008; arienzo, 2005; whang and zhuo, 2005; awofolu et al., 2005). many studies of urban activity-derived heavy metal contamination have been undertaken for soils, plants, stream waters and stream sediments (zvinowanda et al., 2009; kim et al., 1998; abimbola and odukoya, 2006; odukoya and abimbola, 2010). odukoya am, oresanya o and abimbola, a.f if a landfill is to be regarded as sustainable, there must be no residual risk of pollution for the environment 30 to 50 years after disposal has ceased. contamination of groundwater by landfill leachate is particularly difficult to clean up and is often associated with high costs and insoluble technical problems. similar situations, related to the anthropological impact associated with rapid development in industrial and urban regions, have been reported in other parts of the world (zhang et al., 1999; thuy et al., 2000; ansari et al., 2000; kim et al., 1998; wilcke et al., 1998 in morton-bermea et al., 2002). this has led to research into in-situ bioremediation involving naturally-occurring microorganisms which can degrade harmful groundwater contaminants; such approach is known as natural attenuation. the environmental protection agency (epa) has defined natural attenuation as, “a variety of physical, chemical, or biological processes that, under favorable conditions, act without human intervention to reduce the mass, toxicity, mobility, volume, or concentration of contaminants in soil or groundwater.” a variety of other terms have also been used to describe natural attenuation, including natural restoration, intrinsic remediation, intrinsic bioremediation, passive bioremediation, spontaneous bioremediation and bioattenuation (national research council, 1993; wiedemeier et al., 1996). natural attenuation focuses on the destruction rather than transfer of contaminants (national research council, 1993); however, the use of natural attenuation (i.e. chlorinated solvents) to remediate is a growing area (renner, 1998). natural attenuation is potentially applicable for some inorganic compounds, including metals and non-metals. the natural attenuation technique can also be applied to pesticides, but less effectively (federal remediation technologies roundtable, 1997). these natural processes include biodegradation, dispersion, dilution, sorption, volatilisation, radioactive decay, and chemical or biological stabilisation, transformation or destruction of contaminants [epa, 1997]. certain site parameters and characteristics are significant regarding natural attenuation; these would include groundwater flow, geological conditions, the nature and extent of contamination, microbial population, presence of carbonate minerals, nutrient supply, electron donors and acceptors. this research was aimed at determining the efficiency of natural attenuation for certain leachate compounds associated with the olusosun dumpsite and ascertaining whether this would reduce the environmental risks involved with groundwater to acceptable levels. methodology four surface soil samples were collected at depths ranging from 0-20cm using a stainless-steel hand auger within the dumpsite to determine engineering properties, in accordance with british standards (anon, 1990) which included moisture content, grain size distribution, porosity, permeability, plasticity and atterberg limits. another eight subsurface samples (0.5g) were collected vertically at 2 m intervals (2-10 m) for geochemical and microbiological analysis within the dumpsite (fig 4). subsurface samples were stored in sample bags and later air-dried in the laboratory at room temperature for two weeks. the already air-dried samples were disaggregated in a porcelain mortal and later sieved through a 2.0mm polyethylene sieve to obtain a <60µm clay fraction for geochemical analysis. the analysis took place at actlabs, ontario, canada, using an aqua regia digestion and inductively-coupled plasma mass spectrophotometer (icpms) method for trace and major elements analysis in soil. the groundwater and leachate samples collected within the dumpsite were also analysed for microbial population, nutrient supply, electron donors and acceptors, as well as other physical-chemical properties. water samples of about 125 ml were collected for multi-element analysis and acidified on site with 3 ml analytical grade hno3 to bring the water acid solution to ph~2. trace elements and cations in water were analysed by using inductively-coupled plasma-optical emission spectrometry (icpoes) while un-acidified water samples were analysed for anion concentrations using dionex dx-120 ion chromatography techniques. water samples were also analysed for microorganisms such as virus, fungi, bacteria and coliforms. results and discussion engineering properties / geological condition the geological barrier in the study area was studied to see whether it met the following conditions; (1) low permeability, (2) low (effective) porosity, (3) large thickness and (4) high natural retention capacity for hazardous substances. the casagrande plasticity chart (table 1) showed that the soils could be classified as clay or silts having intermediate to high plasticity or compressibility (mi-mh) while a few samples were classified as sand having intermediate to high compressibility, especially those at the surface within a 1 m depth interval. a high percentage of clay content in the soil in the study area could have contributed considerably to the removal of pollution by acting as an effective medium through cation exchange. unsaturated zone thickness was 10-20 m. permeability tests showed low permeability, ranging from 1.96x10-5 to 41.8x10-5 m/s (table 1). the soil could thus have provided long-term protection for the groundwater by natural attenuation as a result of unsaturated zone thickness, high clay content and low permeability. a summary of some of the soil properties is shown in table 2. moisture content ranged from 5.68% to 33.3% (table 1). these values were slightly low as maximum oxygen uptake by micro-organisms in refuse usually occurs at 40%-50% moisture content (usepa, 1998). moisture is important for micro-organism and nutrient distribution; it also flushes away degradation products, apart from being the major factor in leachate formation. water analysis results major and trace elements as well as microorganisms such as ca2+, fe2+, mg2+, k+, na+, cl-, no3 -, so4 2-, hco3 -, al, ba, cu, ni, sr, cd, cr, pb, total bacteria, faecal coliform, total fungi, total viral and phenol were ascertained in water to monitor natural attenuation effectiveness. the water samples included two boreholes, ten shallow wells and leachate samples. both leachate and monitor boreholes were located within the dumpsite while the other water samples were taken very close to the dumpsite. a summary of the results, as well as epa drinking-water standards (usepa, 2009), are presented in tables 2 and 3. ph ranged from 4.86-7.58, higher values being found in the leachate compared to the groundwater. high ph reduces metals’ solubility in leachate and increases cation sorption in soil (clay), hence inhibiting these metals’ migration. this was probably why most contaminants had a higher concentration in leachate which became reduced significantly in borehole samples. tds ranged between 26–453 mg/l in groundwater, which was within the recommended standard, but leachate had values as high as 3580 mg/l, thus being higher than the recommended value. trace elements like cd, cr and pb were only detected in leachate with the following values 1,100, 2,200 and 6,200 ug/l, respectively, and were far above the recommended values. cu, zn and sr were within the standards for groundwater but higher in leachate while al, ba, and ni ranged from 0.1-0.8, bdl-520 and bdl-1000 ug/l, respectively, most samples being higher than the recommended standard. all major elements were within usepa 2009 standard in water samples, except fe, na, cl and no3 (values as high as 0.47, 243, 406 and 250.3 mg/l, respectively) (table2). there was a general trend to become decreased in metals along the groundwater flow paths, except for some cations (fig 2). decreased concentration of metals and some anions as well as a stable or retreating plume along the groundwater flow confirmed natural attenuation, like biodegradation, sorption and dilution. microbiological analysis was carried out for the leachate and two borehole samples within the dumpsite, showing total bacterial count to be 280, 550 and 23000 cfu/ml for the two boreholes and leachate samples, respectively. neither faecal coliforms nor viruses were detected in the two boreholes while their value in leachate was as high as 10,000 cfu/100ml and 230 mpn/ml, respectively. total fungi had 21, 56 and 1,500 cfu/ ml for the two borehole samples and leachate, respectively. phenol was not detected in the boreholes but was recorded at 0.08 in leachate. however, these micro-organisms are required in certain quantities for natural attenuation but may also be hazardous for human health at certain concentrations. coliform count was above the recommended standards for the two boreholes. the absence of faecal coliforms and fungi in the two boreholes and their high concentration in the leachate could have been related to biodegradation preventing their migration to the water table (table 3). all the parameters analysed had their highest concentration in leachate and were above usepa 2009 recommended standards for drinking water. soil analysis results soil samples taken at regular depth intervals within the dumpsite were submitted to geochemical analysis. major and trace elements as well as other biological parameters like phenol and microorganisms were analysed in the soil samples to determine whether natural attenuation was occurring in the study area (table 4). the results showed that the parameters were inversely proportional to depth, which also confirmed natural attenuation (figs 5 & 6). a greater percentage of these contaminants could have been attenuated before reaching the water table (second aquifer) and this was also confirmed by the borehole samples where most contaminants came within the recommended standards. total bacterial count ranged between 14.9 and 32.5 cfu/ml105, faecal coliform ranged between 27.4 and 63.3 cfu/100ml103 while total fungi values ranged from 10.2 to 5.34 mpn/ml102 (table 4). all the values showed that the soil samples were highly polluted but became reduced as depth increased (figs 3&5), thereby confirming natural attenuation. heavy metals like as, cd, pb, mo and la were only present in soil but absent in water. the concentration of trace elements and microorganisms was highest in soil samples and this may have been due to clay’s retention ability and its high sorption ability. conclusion the dumpsite at olusosun in lagos, south-western nigeria was monitored to determine natural attenuation efficiency in reducing contaminants regarding soil and groundwater within and around the site. most parameters needed for natural attenuation to take place such as water (moisture content), microbial population (bacteria, fungi, coliforms), nutrients (nitrogen, phosphorus, other cations), electron donors (total organic carbon) and acceptors (oxygen, nitrate, iron) were present in both the soil and water within and around the dumpsite. the casagrande plasticity chart showed that the soil could be classified as clay or silt having intermediate to high plasticity or compressibility (mimh) while a few could be classified as sand having intermediate to high compressibility. clay is a geological barrier in the study area and it met the following conditions; low permeability, low (effective) porosity, large thickness and high natural retention capacity for hazardous substances. a high ph value in leachate reduced metals’ solubility and increased their sorption in the soil thus inhibiting the migration of these metals from leachate to the water table. major elements and trace elements were within usepa 2009 standards in the water samples, except for fe, na, cl, no3, al, ba and ni. the absence of phenol and total viral count in the boreholes and their presence in leachate could also have been related to natural attenuation. leachate also showed highest concentrations for all the parameters, having values above usepa 2009 standards. the soil analysis showed that all the parameters were inversely proportional to depth which also confirmed that natural attenuation was occurring. acknowledgements the author wishes to thank mr. ola oresanya, the managing director, lagos state waste management authority, lagos, nigeria, for singlehandedly sponsoring this research and for his support and encouragement. table 1: summary of soil engineering and index properties *ll liquid limit *pl plasticity limit *pi plasticity index depth (m) natural moisture content (%) atterberg limits (%) grain size distribution (%) permeability (k) coefficient x 10-5 rating ll pl pi clay silt sand gravel l1 0-1 9.17 18.69 28.13 52.92 0.28 18.6 low 1-2 9.07 22.23 34.71 42.92 0.14 8.46 low 2-3 5.68 33.27 20.99 12.28 56.03 22.05 22.93 0.00 2.61 low 3-4 10.03 13.65 23.98 59.73 2.65 41.8 low l2 0-1 17.99 30.39 15.08 15.31 56.57 20.70 22.74 0.00 2.61 low 1-2 17.07 30.59 19.83 10.76 54.56 22.11 22.73 0.61 3.35 low l3 0-1 33.33 44.29 23.36 20.93 26.57 33.40 30.68 9.36 9.09 low 1-2 23.58 46.70 25.49 21.21 32.66 26.60 31.59 6.75 5.61 low l4 0-1 11.71 37.16 22.98 14.18 44.11 32.55 23.24 0.00 7.84 low 1-2 20.01 33.65 20.79 12.86 60.72 18.91 20.21 0.17 1.96 low biogeochemical and engineering characteristics of soils and groundwater around a dumpsite table 2: summary of physical-chemical analysis result of water samples table 4: summary of soil samples’ chemical and microanalysis parameters table 3: summary of phenol and microbiological parameters in water samples nd: not detected major elements –mg/l trace elements –µg/l parameter range n=12 mean usepa (2009) leachate ph 4.86-7.58 5.21 6.5-8.5 7.58 tds 26-453 184.80 1000 3,580 al 0.1-0.6 0.29 87 0.8 ba bdl-100 61.82 1000 520 cu 4-28 16.93 1000 1,300 ni bdl-600 122.0 20 1,000 zn 30-1100 449.60 5000 2,065 sr 20-250 48 260 ca 16-128 24.55 200 407.89 fe bdl-0.47 0.13 0.3 0.79 k 0.4-194 19.05 200.28 na 4.9-243 47.88 200 254.75 mg 0.6-63.1 9.40 50 225.6 p 0.04-3.64 1.20 4.44 si 2.4-10.3 5.44 15.4 cl 11.8-406 54.08 230 657.24 no 3 0.02-250.3 70.11 10 450.23 hco 3 0-78 12.6 132.22 so 4 2.81-187 35.59 400 56.98 cd nd 3 1,100 cr nd 50 2,200 pb nd 10 6,200 parameters borehole within dumpsite borehole near the dumpsite leachate sample who (2004) epa (2005) microbiological analysis t. bacterial (cfu/100ml) 280 550 23,000 0 0 faecal col. (cfu/100ml) nd nd 10,000 0 0 t. fungi (mpn/ml) 21 56 1,500 0 0 t viral nd nd 230 0 0 phenol nd nd 0.08 s/n parameter l1 l2 l3 l4 l5 l6 major elements 01 ca 4.36 3.52 2.14 1.58 1.61 1.96 02 fe 28.14 27.97 22.63 18.78 23.04 24.75 03 mg 4.82 3.95 2.73 2.75 2.34 1.99 04 k 6.15 3.46 4.92 2.78 3.01 2.58 05 na 5.74 4.11 5.18 3.15 2.89 3.11 06 p 19.55 20.32 67.38 20 32.98 18.47 07 sulphate 600 525 600 240 480 500 trace elements 08 cu 70 55 42 30 26 16 09 as 16 14 11 5 4 -2 10 cd 1.8 0.12 0.43 0.21 0.12 0.10 11 pb 379 223 126 162 88 50 12 cr 223.34 196 75 185.10 103.47 154.86 13 zn 887.76 758.94 836.78 1,087.11 945.84 402.54 14 mo 4 3 2 3 2 1 15 ni 32 30 26 28 21 14 16 sr 53 42 33 24 20 8 la 35 25 24 23 19 16 mn 1439 620 513 436 384 128 17 nitrogen (organic) 18.33 16.48 20.34 25.12 18.34 10.34 18 toc 6.096 10.230 10.494 14.432 11.316 12.588 19 phenol 223 196 75 185.10 103.47 154.86 microanalysis 20 total bacteria count cfu/ ml105 32.5 24.8 18.9 20.1 18.7 14.9 21 faecal coliforms cfu/100ml103 63.3 56.7 62.5 50.2 59.4 27.4 22 total fungi count mpn/ ml102 10.2 7.44 6.81 9.05 8.35 5.34 odukoya am, oresanya o and abimbola, a.f fig 1: location map showing dump site and sample points in ojota biogeochemical and engineering characteristics of soils and groundwater around a dumpsite cations -10 0 10 20 30 40 50 60 70 80 90 w1 le ac ha te w2 w3 w4 w5 w6 w7 w8 w9 w10 w11 w12 w13 w14 co ntr ol water samples co nc en tra tio n ( mg /l) ca fe k mg na p si groundwater flow path groundwater flow path anions 0 50 100 150 200 250 w1 le ac ha te w2 w3 w4 w5 w6 w7 w8 w9 w10 w11 w12 w13 w14 co ntr ol sample locations c on ce nt ra tio n (m g/ l) cl no3 so4 hco3 fig 2: concentration of metals in the groundwater flow 0 5 10 15 20 25 30 c o n c e n tr a ti o n ( m ) 0-1m 2m 4m 6m 8m 10m depth (m ) ca fe mg k na 0 10 20 30 40 50 60 70 c 0n ce n tr at io n ( p p m ) 0-1m 2m 4m 6m 8m 10m depth (m ) nitrogen phosphorus 0 1 2 3 4 5 6 co nc en tr at io n (p pm ) 0-1m 2m 4m 6m 8m 10m depth (m ) cd pb odukoya am, oresanya o and abimbola, a.f fig 4: sample location map 0 200 400 600 800 1000 1200 co nc en tr at io n (p pm ) 0-1m 2m 4m 6m 8m 10m depth (m ) zn cu cr 0 50 100 150 200 250 co n ce n tr at io n 0-1m 2m 4m 6m 8m 10m depth (m ) phenol 0 500 1000 1500 2000 2500 3000 3500 t o ta l b a c te ri a l (c f u /m l1 0 3 0-1m 2m 4m 6m 8m 10m depth (m ) total bactrial fig 3: concentration of cations, nutrients, trace elements, total bacteria and phenol in soil samples at different depth intervals fig 5: concentration of toc, faecal coliform/ total fungi in soil samples at various depth intervals 0 2 4 6 8 10 12 14 16 to ta l o rg an ic ca rb on (% ) 0-1m 2m 4m 6m 8m 10m depth (m ) toc 0 100 200 300 400 500 600 700 c on ce nt ra tio n 0-1m 2m 4m 6m 8m 10m depth (m) feacal coliform total fungi biogeochemical and engineering characteristics of soils and groundwater around a dumpsite references abimbola a.f and odukoya a.m: geochemical characterization of soils, sediments and waters around active and abandoned dumpsites in lagos 25, supplement 1 (2006) pp. 131-132, doi 10.1007/ bf02839992. adeniyi aa, okedeyi oo (2004); assessing the speciation pattern of lead and zinc in surface water collected from abegede creek, ijora, lagos. pak j sci indus res; 47:430-4. altindag a, yigit s (2005): assessment of heavy metal concentrations in the food web of lake beysehir, turkey. chemosphere; 60: 552-6. anon (1990): methods of tests of soil for civil engineering purposes. british standards institution, london. ansari a.a, singh i.b, tobschall h.j (2002): status of anthrogenical induced metal pollution in the kanpur-unnao industrial region of the ganga plain india. environ geol 38;25-33. arienzo m, adamo p, bianco mr, violante p (2001): impact of land use and urban run-off on the contamination of the samo river basin in southwestern italy. water air soil pollut ; 131: 349-66. arneth j.d, milde g., kerndorff, h and schleyer, r. 1989: waste deposit influences on groundwater quality as a tool for waste type and site selection for final storage quality. in: baccini p (ed) the landfill, vol 20, (lecture notes in earth sciences). springer. awofolu o.r, mbolekwa z, mtshemla v, fatoki o.s (2005): levels of trace metals in water and sediment from tyume river and its effects on an irrigated farmland. water south afr ; 31: 87-94. baedecker, m.j. and back,w., 1979: hydrogeological processes and chemical reactions at a landfill. ground water. 17(5):429–437 environmental protection agency. 1997: use of monitored natural attenuation at superfund, rcra corrective action, and underground storage tank sites, directive 9200.4-17p, 32 pp., epa, office of solid waste and emergency response, washington, d.c. jankowski, j., acworth r.i. 1997: development of a contaminant plume from a municipal landfill: redox reactions and plume variability. in: chilton j, hiscock k, morris b, nash h, tellam j, hennings s (eds) proceedings of the xxvii congress of the international association of hydrogeologists: groundwater in the urban environment, nottingham, uk, 21–27 september 1997. balkema, rotterdam, pp 439–444. kim, k.w, myung j.k, ahn j.s, chon, h.t(1998): heavy metal contamination in dusts and stream sediments in the teajon area, korea. j. geochem explor 64; 409-419. kimmel, g.e and braids, o.c. 1974: leachate plumes in a highly permeable aquifer. groundwater. 12(6):388–392. morton-bermea, o hernandez alvarez, e, gaso, i and segovia n (2002): heavy metal concentrations in surface soils from mexico city. bull. environ. contam. toxicol. 68; 383-388. springer verlag new york inc. doi: 10 1007/s00128-001-0265x. odukoya, a. m. and abimbola, a. f., (2010); contamination assessment of surface and groundwater within and around two dumpsites. int. j. environ. sci. tech., 7 (2), 367-376. thuy h.t.t, tobschall h.j, an p.v(2000); distribution of heavy metals in urban soils –a case study of danang-hoian area (vietnam) environ geol. 39: 603-610. united states environmental protection agency (us epa) (1998): epa report to congress, solid waste disposal in the united states volume ii. epa office of solid waste and emergency response, epa/530-sw88-011b. washington, d.c., usa, pp. 4-58. usepa.national drinking water standards (2009), available from http://www.epa.gov/safewater/ wang x, zhuo q (2005): ecotoxicological effects of cadmium on three ornamental plants. chemosphere 2005; 60: 16-21. wilcke w, muller s, kanchanakool n, zech w(1998): urban soil contamination in bangkok; heavy metal and aluminum partitioning in topsoils. geoderma 86, 211-228. zhug h, ma, d, xie., chen x (1999): an approach in studying heavy metal pollution caused by modern city development in nanjing, china. environ geol. 38, 223-224. zvinowanda, c. m.; okonkwo, j. o.; shabalala, p. n.; agyei, n.m., (2009). a novel adsorbent for heavy metal remediationin aqueous environments. int. j. environ. sci. tech., 6 (3), 425-434 (11 pages). odukoya am, oresanya o and abimbola, a.f keywords: semi-coal rock; geological evolution; non-coal pillar; distressed blind-zone; gas extraction. palabras clave: roca de semicarbón; evolución geológica; minería sin pilar; peligros de la zona ciega; extracción de gas. how to cite item xie, x., fang, x., liang, m., & ai, d. (2020). geological evolution process, mechanism, and application of protective layer in semi-coal and rock face. earth sciences research journal, 24(4), 499-506. doi: https://doi.org/10.15446/ esrj.v24n4.91378 under the influence of geological structure, the surrounding rock of semi-coal seam is easy to be loose and fractured, and the repairing rate is high. therefore, it is of great theoretical and practical significance to study semi-coal seam mining's mechanism and control technology. in this paper, the geological evolution principle of the semi-coal seam is analyzed theoretically. the mechanic's relationship between the mining height of the semi-coal face and the lower coal seam's pressure relief is interpreted. the mechanism of eliminating bad blind areas in non-pillar mining of upper protective seam is analyzed. through numerical simulation analysis, it is concluded that the expansion deformation rate of a protective layer increases linearly with the increase of mining thickness of the protective layer, and the width of undesirable blind zone of coal seam increases linearly with the rise of protective layer thickness. this paper puts forward the technology of non-pillar mining in the protective seam's semi-coal and rock face. field industrial test results show that the technology achieves the overall pressure relief and reflection reduction of low permeability and high gas coal seams. the underlying 3 + 4 protective layer achieves the general pressure relief and reflection reduction of low permeability and high gas coal seams. coalbed methane can be pre-drained up to 18 m3/min with a concentration of 90%. abstract geological evolution process, mechanism, and application of protective layer in semi-coal and rock face proceso de evolución geológica, mecanismo y aplicación de capa de protección en minería de semicarbón y en pared de roca issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.91378 bajo la influencia de la estructura geológica, la roca que rodea la veta de semicarbón puede soltarse y fracturarse, y el índice de reparación es alto. es de gran importancia teórica y práctica estudiar el mecanismo de minería de la veta de semicarbón y la tecnología de control. en este artículo el principio de la evolución geológica de la veta de semicarbón se analiza teóricamente. además, se interpreta la relación mecánica entre el peso de la minería en la pared y el alivio en la presión de la parte baja de la veta. también se analiza el mecanismo de eliminar las áreas ciegas en la minería sin pilares de la veta superior protectora. a través del análisis de simulación numérica se concluye que el índice de deformación por expansión de una capa de protección se incrementa linealmente con el aumento del grosor de la capa de protección, y el ancho de la indeseable zona ciega en la veta de carbón se incrementa linealmente también con el incremento en el grosor de la capa de protección. este trabajo ofrece una tecnología en minería sin pilar de la capa de protección en la veta de semicarbón y en la pared de la roca. los resultados de evaluaciones de campo industriales muestran que esta tecnología logra un alivio general en la presión y en la reducción del reflejo en vetas de baja permeabilidad y de carbón con alto índice de gas. la capa de protección 3 + 4 logra el alivio en la presión general y en la reducción del reflejo en vetas de baja permeabilidad y de carbón con alto índice de gas. el metano en capas de carbón puede ser drenado previamente en hasta 18 m3/min con una concentración del 90 por ciento. resumen record manuscript received: 12/03/2019 accepted for publication: 20/09/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 499-506 xiaoping xie1,2, xinqiu fang1*, minfu liang1,3, dechun ai2 1state key laboratory of coal resources and safe mining, china university of mining and technology, xuzhou 221116, china 2college of mining and civil engineering, liupanshui normal university, liupanshui 553004, china 3school of management, china university of mining and technology, xuzhou 221116, china” *corresponding author: 264580414@qq.com m in in g e n g in e e r in g 500 xiaoping xie, xinqiu fang, minfu liang, dechun ai introduction when the roadway excavation section has both rock and coal seam, and when the coal seam occupies 1/5 to 4/5 of the area of the excavation face, it is called semi-coal-rock mining working seam, which generally exists in thin coal seam conditions. taking china as an example, the reserves of unexploited semicoal and rock mining working strata are about 6.150 billion tons, accounting for 20.4% of the total recoverable reserves. in some areas, the proportion of semicoal seam reserves is larger, such as sichuan province accounts for about 60%, shandong province for about 54%, heilongjiang province for about 51%, guizhou province for about 37% (yuan, 2008a; zhou et al., 2016). other coal-producing provinces, such as henan, shanxi, inner mongolia, hebei and jilin, also have rich thin seam resources. with the continuous reduction of coal resources, semi-coal and rock mining will be a problem faced by china and even the world coal mining industry. semi-coal-rock working face has small space of movement, difficult worker operation, complex mining method, easy to occur safety accidents, low economic benefits. semi-coal-rock working face is mostly semi-coalrock roadway, which has low tunneling rate, high tunneling cost, tight mining replacement and low mining efficiency. it was shown as figure 1. in this paper, the principle of geological evolution of semi-coal seam is analyzed theoretically, and the technical measures of linear increase of expansion deformation rate of protective seam with the increase of mining thickness of protective seam are put forward. seam semi-coal-rock mechanism structural analysis semi-coal and rock transportation roadway are arranged in mud-stone and coal seam. the lithology of two sides, roof and floor is relatively fragile, the deformation parameters are relatively large, and the tensile and shear failure strength is relatively small (wu et al., 2004). under the influence of mining, the surrounding rock of two sides, roof and floor is seriously damaged, the scope of loosening zone is relatively large, and the bearing capacity of surrounding rock is relatively low. according to the analysis of consolidation theory, the length of roof bolt and side bolt is insufficient in the original supporting scheme, the anchorage section of bolt cannot be anchored in the original rock stress zone, and the anchorage force is insufficient, which leads to the insufficient strength of roof and two sides of roadway. the diameters of top anchor and side anchor are small, the tensile strength and prestressing force are not enough, the extrusion strength of anchor end to destroyed surrounding rock, coal seam and mud-stone is small, the secondary bearing area cannot be formed, and the mechanical properties of anchor solid cannot be improved. the deformation of surrounding rock cannot be controlled in time, and the roof pressure can be transferred to both sides, resulting in larger stress, horizontal stress and vertical stress in both sides. the force is almost vertical, so serious deformation, roof subsidence and floor heave appear on both sides of the roadway. in order to study the maximum damage range of roof strata and side wall of semi-coal-rock roadway, theoretical analysis is carried out according to that all roadways are coal seams (yuan, 2008b). because the thickness of carbonaceous mud-stone layer on the roof of roadway varies greatly, its caving shape is arch. according to the theory of pressure-free arch, as shown in figure 2. semi-coal and rock transportation roadway is a typical soft rock mining roadway, and its surrounding rock deformation characteristics are as follows: (a) the surrounding rock deformation is intense and large, the convergence rate of roadway section is as high as 38% during excavation, and the deformation has rheological characteristics and lasts a long time; (b) the degree of surrounding rock fragmentation gradually weakens from shallow to deep, and its damage scope exceeds the anchorage range of the anchor, resulting in the failure of the anchor; (c) the surrounding rock loosening and breaking; the rapid development and serious destruction of the bad scope result in the reduction of anchorage effect, failure of anchorage system and failure of supporting bearing structure, which further aggravates the development of figure 1. semi-coal-rock working face 501geological evolution process, mechanism, and application of protective layer in semi-coal and rock face in the formula: γ is the coal bed bulk density, n/m3; h is the depth of coal seam; δk is the maximum concentrated stress increment coefficient, δk=k-1; δ is the lead abutment pressure attenuation coefficient; x0 is the width in the plastic zone, m; x is the distance from the coal seam,. from the formulas 1, 2, and 3 it can be known that the plastic zone width x0 in the semi-coal rock working face is proportional to the coal mining height (h+δh) and the width of the plastic zone of the extremely thin coal seam is smaller. figure 3 shows the distribution rules of abutment pressure of working face. the curves in the figure meet the distribution formula of abutment pressure in plastic zone (1), and curves satisfy distribution formula of abutment pressure in elastic zone (2). 0 4 8 12 16 20 4 8 12 16 δx0 distance from the coal wall（m） w or ki ng fa ce ab ut m en t pr es su re ( m pa ) mining height h mining height h+δh x0① ② ③ figure 3. the distribution rules of abutment pressure of extremely thin coal seam during work according to the formula (3), the increase of width in plastic zone of semi-coal rock working face can be calculated as: ∆ ∆ x h f ln k h rc 0 2 =    ξ γ (4) according to the relationship between the lead abutment pressure of the working face and the destruction of the underlying coal and rock seams, a structure of the plastic destruction zone of the underlying coal and rock along the advanced direction of the coal seam is established. x0 a b c d e o h max f r0 r θ π/4-φ/2 π/4+φ/2 figure 4. section of plastic destruction zone of overlying coal strata according to the geometrical dimensions of the ultimate plastic destruction zone shown in figure 4, the maximum destruction depth of the floor destruction zone under the action of the extreme lead abutment pressure can be preliminarily determined : h x emax / tancos cos / = +( ) +( )0 4 2 2 4 2 φ π / φ π/ φ φ (5) in this formula: hmax is the maximum destruction depth of the coal seam floor; x0 is the width of the plastic support area in front of the coal wall; φ is the internal friction angle. surrounding rock loosening and failure to its due support effect. fourthly, the phenomenon of two sides of roadway is widespread, and the phenomenon of arch roof and arch waist separation layer falling off is widespread, which leads to the failure of external anchorage point of some bolts, which also leads to the failure of external anchorage point of some bolts. it is an important reason for the reduction or failure of anchorage effect. in conclusion, the failure mode of surrounding rock of semi-coal and rock transportation roadway belongs to the typical large deformation failure mode of soft rock rheology. the failure mechanism of support is as follows: under the condition of high in-situ stress, the scope of loosening and failure of weak surrounding rock expands rapidly, and its rheological action causes the surrounding rock of roadway to break and the deformation continues to increase, resulting in the failure of bolt-mesh combined support structure, which leads to the loss of bearing capacity of surrounding rock-support bearing structure. mathematical model in view of the high-efficiency mining that is hard to exploit the resource of extremely thin coal seams, the idea of small fully-mechanized mining in the semi-coal rock working face is proposed, and the mining height is increased by breaking the weak roof or floor of an extremely thin coal seam to increase mining height in order to install small fully-mechanized mining equipment and improve the pressure relief effect of the coal seam group at the below part. after the semi-coal rock working face is mined, the load of overlying strata in goaf is transferred to the front of coal wall forming an abutment pressure, and the coal wall is in a plastic or fracture state within a certain depth (zhou et al., 2015; cheng et al., 2012). usually, the front of coal wall is divided into two zones in the abutment pressure: plastic zone and elastic zone. according to the limit equilibrium condition, the abutment pressure in the plastic zone is distributed: σ ξ y c f h h x r e= + 2 ∆ (1) in this formula, σy is the lead abutment pressure, mpa; rc is the residual strength of the coal wall, mpa; ƒ is the friction coefficient of the layer, in the ƒ=tanφ1, φ1 stands for the friction angle between the roof and floor of the coal seam and the coal seam, o; ξ=(1+sinφ)/( 1-sinφ), φ is the internal friction angle of the coal, o; h is the thickness of the extremely thin seam, m; δh is the thickness of the broken rock, m;the abutment pressure distribution in the elastic zone is a negative exponential curve,: σ γ −δ − y x xh ke= +    ( )1 0∆ (2) x h h f ln k h rc 0 2 = +     ∆ ξ γ (3) figure 2. calculating model of semi-coal and rock structure 502 xiaoping xie, xinqiu fang, minfu liang, dechun ai according to the formulas 3 and 5, it can be known that after the mining of the semi-coal rock working face, the maximum destruction depth of the floor is proportional to the width x0 of the support pressure in plastic zone in front of the coal wall, it is equal to the mining height of the semi-coal rock working face (h+δh). evolution of semi-coal seam semi-coal and rock from the beginning of tunneling to withstand two mining face impact and roadway retention stage roof rotation subsidence (li, 2014), long service time, surrounding rock presents large deformation characteristics, surrounding rock pre-deformation directly related to the later deformation (li et al., 2011; liu et al., 2009), showing a typical evolution characteristics. obtaining the displacement, strain and fracture evolution law of surrounding rock from tunneling to the end of secondary mining is of great significance to reveal the deformation mechanism. evolutionary simulator restricted by the field testing environment and testing level, there are some difficulties in measuring the deformation evolution law of surrounding rock of semi-coal and rock. laboratory physical simulation can well realize the observation of deformation of semi-coal and rock. the testing environment is good and the testing accuracy is high (liu et al., 2011). in this paper, the physical simulation of semi-coal and rock is carried out by using the selfdeveloped two-way and four-side loading test device, and the deformation of semi-coal and rock is observed. the commonly used plane experimental devices are mostly used to simulate stope advancing, observe roof caving pattern, and simulate vertical stress by adding counterweight (liu et al., 2013). there is no active load on both sides and floor, which results in great difference between boundary conditions and original rock stress field, and the credibility of simulation results is reduced (ma et al., 2007). in order to better study the surrounding rock deformation of gob-side retaining roadway, a two-way and four-sided loading test device is developed, which overcomes the problem that the three-dimensional simulation test device cannot advance step by step and continuously, and solves the problem that the common plane simulation cannot exert high stress on top, bottom, left and right. the design of simulation device is controlled by different oil system, which can realize unequal pressure loading on all sides. it can simulate the process of driving, advance effect of primary mining, advance effect of retaining roadway and secondary mining under the corresponding real in-situ stress state (chen et al., 2003). it is innovative in the experimental study of retaining roadway along gob at home and abroad. the overall size of the test device is 4.5 x 0.5 x 4.5 m, and the effective test size is 4.0 x 2.5 x 0.5 m. the parameters of the simulator are shown in figure 5 and table 1. the numerical simulation of mechanical parameters of the coal mine is shown on table 1. process recording the simulation evolution process is to simulate the advance effects of gob-side entry retaining, primary mining, retaining and secondary mining according to the way of “excavation roadway-right boundary-middle mining, filling body-left boundary-middle mining”. the experimental device is a plane simulation device. (a) under the original rock stress state, the surrounding rock stress is redistributed when the roadway is excavated according to the design size (he et al., 2017). the existence of roadway space makes the floor of roadway change from three-dimensional stress state to two-dimensional stress state, and pressure relief occurs. this simulation process is the same as the actual situation. (b) after the excavation of the right coal seam, the advance abutment pressure of the mining face acts on the stress superposition of the surrounding rock of the roadway and the surrounding rock during the excavation, which increases the failure depth of the surrounding rock, the peak stress concentration factor and the influence range. the stress state of the surrounding rock on the right side of the roadway is the same as that in the field (jiang et al., 2013). the stress of the surrounding rock on the left side of the roadway is the superposition of the mining stress and the lateral pressure of the surrounding rock, slightly different from that in the field. different. (c) the filling body is placed immediately after the mining of the right working face is completed, and the simulation enters the stage of retaining roadway. in production, the most severe deformation stage of surrounding rock in gob-side retaining roadway is the rotary subsidence of the basic roof above the filling body. in simulation, after filling body is placed, the roof will not immediately rotate subsidence (khademalhoseiny et al., 2017). with the generation, development, expansion and penetration of cracks, the stress will seek to rebalance, the basic roof will rotate subsidence, and the overlying strata load will act on the floor through the filling body. the process can well simulate the deformation characteristics of surrounding rock during the stable period of retaining roadway. (d) after the mining of the left side face, the advance abutment pressure and the surrounding rock stress field after the stability of the retained roadway are superimposed (rosa et al., 2017). the process of the left side of the roadway is the same as that of the field, and the stress field of the right side of the roadway is slightly different from the actual situation of the field. the feasibility of the simulation scheme is discussed in stages below which shown as figure 6. evolution simulation result in order to study the relationship between the pressure relief effect of udder protected layer and the different heights of semi-coal rock working face, four simulation schemes were designed: mining height 0.5 m, mining height 1.5 m, mining height 2.5 m, and mining height 3.5 m. in the numerical calculation figure 5. semi-coal-rock working face table 1. mechanics parameters of the coal-rock joints lithology normal stiffness/gpa tangential stiffness/gpa cohesive force/mpa friction angle /° tensile strength/mpa coal seam 0.8 0.4 0 4 0 medium sandstone 0.5 0.3 0 6 0 fine-sandstone 0.4 0.2 0 7 0 mudstone 0.6 0.3 0 5 0 siltstone 0.5 0.5 0 7 0 sandy mudstone 0.7 0.4 0 6 0 503geological evolution process, mechanism, and application of protective layer in semi-coal and rock face model, the observation line is arranged in the lowest no. 5 protected coal seam, and the starting point of model excavation is at 50 m. when the working face is mined 150 m in an inclined direction, the expansion deformation rate change features of no. 5 are shown in figure 7. engineering application overall design the engineering background is as follows: the coal mine uses +400 m north wing main roadway as the preparing roadway arrangement for inclined long-wall working face of small fully mechanized mining and 22201 working face is the first working face of the protective layer in the no.2 extremely thin coal seam of north second mining area. the slope length of designing the working face is 150 m, the length of the propulsion is about 1538 m, and the elevation of the working surface is +396 m~+486 m. the designed semi-coal rock working face is 1.5 m in height, adopting long-wall retreat mining method. the working face is mined without coal pillar of gob-side entry retaining, and after the 22202 working face roadway is mined, a y-type ventilation system is formed: the 22201 auxiliary haulage roadway and the machine-rail unification roadway enter the wind, and the dirty wind passes through the 22201 gob-side entry retaining, 22202 open-off cut and rail roadway, it’s as shown in figure 8. there are mainly the following problems in tunneling of semi-coal and rock face: first, poor shaping and surrounding of working face rock failure and over-excavation are serious, especially in soft and broken strata (cheng et al., 2003). it causes a lot of waste of supporting materials and poor safety conditions figure 6. the feasibility of the simulation (a) mining height 0.5 m (c) mining height 2.5 m -7 -5 -3 -1 1 3 5 7 0 34 68 103 137 171 205 239 ex pa ns io n ra te o f t he pr ot ec tiv e la ye r ( ‰ ) horizontal position（m） -7 -5 -3 -1 1 3 5 7 0 34 68 103 137 171 205 239 ex pa ns io n ra te o f t he pr ot ec tiv e la ye r ( ‰ ) horizontal position（m） . -7 -5 -3 -1 1 3 5 7 0 34 68 103 137 171 205 239 ex pa ns io n ra te o f t he pr ot ec tiv e la ye r ( ‰ ) horizontal position（m） . -7 -5 -3 -1 1 3 5 7 0 34 68 103 137 171 205 239 ex pa ns io n ra te o f t he pr ot ec tiv e la ye r ( ‰ ) horizontal position（m） . (d)mining height 3.5 m (b)mining height 1.5 m figure 7. expansion deformation rate curve 22201working face 22201 auxiliary haulage roadway 22201 machine-rail unification roadway 22202 rail roadway o pen-off cut protective coal pillar line n orth w ing track tunnel n orth w ing tape tunnel n orth w ing return air tunnel 24208 return air roadway 24208 belt roadway n o.1 coal bunker gob-side entry retaining fresh wind dirty wind working face advancing direction 22201working face goaf figure 8. layout of the 22201 working face of semi-coal rock 504 xiaoping xie, xinqiu fang, minfu liang, dechun ai for head-on construction. second, the half-coal and rock excavation speed is slow, which seriously affects the construction and arrangement of followup projects. thirdly, rock loading and transportation are difficult because of inclination downhill; fourthly, water inrush control in working face also needs to take up a certain cycle time, which affects the driving speed of working face. at the same time, the rock stratum tends to be unstable due to the softening effect of water on rock, and the roadway is difficult to support. secondly, coal and rock are transported separately, the regular circulation organization is complex, and the degree of parallel operation is low. slow excavation speed has seriously affected the mining replacement relationship of mining face. the main reasons are as follows: first, the existing blasting technology is backward, which is not conducive to the formation of roadways; second, the support technology and support parameters are inadequate; third, the working line is not compatible with the site construction conditions; fourth, the construction technology, the layout of working face and the selection of labor organization forms, as well as the inadequate management system (abubakary et al., 2015). application method on the site, the 22201 semi-coal rock and small fully mechanized working face of the typical coal mine is shown in figure 9. during the observation period from january 6 to july 22, the working face was promoted to 602.6 m and a total of 1,812,020 tons of coal was mined. from january 6 to february 27, the progress of the initial mining phase on the working face was slow. in 50 days, the working face completed a total advance of 113.2 m, and the average daily advance was 2.26 m, and the mining efficiency was low; from february 27th to july 22, the speed of advancement has improved, with an average of 3.05 m per day. from the actual coal mining results, we can see that the equipment arranged in the working face and in the mining lanes meet the production requirements, and the semi-coal rock working face basically achieves safe and efficient mining. experimental results and analysis experimental results (a) overall pressure relief analysis of non-coal pillar in protective layer: coal seam deformation and stress change in the protected layer are the key to investigate the pressure relief effect of coal and rock below the remaining coal pillars and non-coal pillar (bustin & clarkson, 1998). the observation lines are arranged in the no.3+4 protected layer under the numerical calculation model. when working height of 22201 and 22202 working face is 1.5 m, the starting point of model excavation is at the coordinate of 40 m. after the length of two working faces is continuously mined in the direction of inclination of 150 m, non-coal pillars are left between the two working face interfaces and the width of the pillar of the section is left separately. for 20 m, 30 m, and 40 m, the expansion rate of the protected coal seam is shown in figure 10 and figure 11. -12 -9 -6 -3 0 3 6 9 12 0 34 68 103 137 171 205 239 273 308 342 376 ehtfo etar noisnapxe pr ot ec tiv e la ye r (‰ ) horizontal position（m） -15 -12 -9 -6 -3 0 3 6 9 12 15 0 34 68 103 137 171 205 239 273 308 342 376 ex pa ns io n ra te o f t he pr ot ec tiv e la ye r (‰ ) horizontal position（m） -15 -12 -9 -6 -3 0 3 6 9 12 15 0 34 68 103 137 171 205 239 273 308 342 376 ehtfo etar noisnapxe pr ot ec tiv e la ye r (‰ ) horizontal position（m） -15 -12 -9 -6 -3 0 3 6 9 12 15 0 34 68 103 137 171 205 239 273 308 342 376 ehtfo etar noisnapxe pr ot ec tiv e la ye r (‰ ) horizontal position（m） (a) pillar width 0 m (b) pillar width 20 m (c) pillar width 30 m (d) pillar width 40 m figure 10. the expansion deformation rate curve figure 9. the 22201 working face on-site picture 505geological evolution process, mechanism, and application of protective layer in semi-coal and rock face analysis from the analysis of figure 9, it can be seen that as the width of the upper coal pillar decreases, the expansion rate of the protected coal seam below the pillar that is less than 3‰ increasingly decreasing, which is the range of the distressed blind-zone. when the width of the section coal pillars is 40 m, 30 m, 20 m, and 0 m respectively, the inclination width of the no.3+4 coal seam below the coal pillar is less than 3‰, which is 64.3 m, 52.2 m, 44.2 m, and 0 m, respectively. when non-coal pillars are used, the distressed blind-zone in the protected layer below the pillars are eliminated. by establishing a rectangular coordinate system, the four simulation schemes with different widths of the coal pillars correspond to the width of the distressed blind-zone of the protected layer make four coordinate points, and then make the fitting straight line of four coordinate points. the analysis shows that the fitting linear equation is y= 17.09x-10.05, and the fitting r2= 0.8968, that is, the width of the blind zone of the pressure relief of protected coal increases with the increase of the pillar width of the protective layer, which is roughly in line with the linear growth law of the fitted line (chen et al., 2016). from the analysis of figure 11, it can be known that when the upper coal pillar is left in the upper protective layer, stress concentration occurs in the protected layer is under the coal pillar, and with the decrease of the width of the coal pillar in the upper protective layer, the stress concentration range of the protected coal seam under the pillar is decreasing; when the upper protective layer is mined without coal pillars, there is no stress concentration area in the protected (a) pillar width 0 m (b) pillar width 0 m (c) pillar width 0 m (d) pillar width 0 m figure 11. vertical stress cloud picture of the working face of semi-coal rock with different pillar widths y = 17.09x 10.05 r2 = 0.8968 0 10 20 30 40 50 60 0 10 20 30 40 t he b lin d ar ea w id th o f t he pr ot ec te d co al s ea m （ m ） different pillar width the blind area width 线性 (the blind area width)fitting a straight line figure 12. the relationship between the width of different pillars and the width of the distressed blind-zone in the protected layer 506 xiaoping xie, xinqiu fang, minfu liang, dechun ai layer below the original coal pillars. when the width of coal pillars is 0 m, 20 m, 30 m, and 40 m, the maximum vertical stress values of udder-protected layer below the coal pillars are 11 mp, 15 mp, 19 mp, and 21 mp, which are 0.84, 1.15, 1.45, and 1.6 times respectively for the original rock stress, that is, the protected layer under the original coal pillar will produce a pressure relief effect when non-coal pillar is used, which eliminates the distressed blind-zone. conclusions (a) by using similar material simulation and numerical simulation, the deformation evolution law of surrounding rock of semi-coal-rock working bed is obtained. during the excavation period, the deformation of surrounding rock is symmetrical distribution, and in the first mining advance influence stage, the support point of the side roof of the working face sinks due to the advance support pressure of the working face, the roof deviates, the deformation of the side coal wall of the working face is serious, the damage depth of the surrounding rock and the development of cracks increase. in the retaining roadway stage, the roof is further rotated and subsided due to the deformation of the empty roof and filling body, and the high-strength filling body is embedded in the floor. the floor heave is serious, the deviation is obvious, and the deformation of coal side under high abutment pressure is serious in the advanced stage of secondary mining. (b) the mechanism of filling body in semi-coal seam is analyzed, and the mechanical calculation model of filling body is established. the stress components of filling body and its internal stress are obtained. it shows that the horizontal stress gradually decreases from the side of goaf to the roadway, and the upper and lower parts of filling body bear higher vertical stress and shear stress, which is the key point of support. (c) on-site engineering practice shows that relative to the south wing of the mine is not protected by protective layer mining, the pressure of the gas in the protected coal seam is relieved and the permeability is increased after adopting the non-coal pillar mining technology for semi-coal rock of protective layer in the north wing of the mine. the gas extraction scalar is up to 18m3/min, and the maximum concentration is about 90%. the gas extraction scalar and concentration at each drilling field undergoes smoothness, ascent, descent, and recovery in turn. the stage shows that the overall pressure relief effect of the non-coal pillar mining technology for semi-coal rock of protective layer is significant. acknowledgements this research was supported by the national natural science foundation of china (no.51874276, no.52004273) and china postdoctoral science foundation (no.2019m661992). references abubakary, s., micah, n., & jenny, g. (2015). evaluation of the impact of commodity price change on mine plan of underground mining. international journal of mining science and technology, 25, 375-382. bustin, r. m., & clarkson, c. r. (1998). geological controls on coalbed methane reservoir capacity and gas content. international journal of coal geology, 38, 3-26. chen, y., bai, j. b., wang, x. y., ma, s. q., & xu, y. (2003). support technology research and application inside roadway of gob-side entry retaining. journal of china coal society, 37, 903-910. chen, d., pan, z., shi, j. q., si, g., ye, z., & zhang, j. (2016). a novel approach for modelling coal permeability during transition from elastic to postfailure state using a modified logistic growth function. international journal of coal geology, 163, 132-139. cheng, y. p., yu, q. x., & yuan, l. (2003). gas extraction techniques and movement properties of long distance and pressure relief rock mass upon exploited coal seam. journal of liaoning technical university, 22, 483-486. cheng, y. h., jiang, f. x., & lin, j. k. (2012). experimental study on gob-side entry retaining by roadside flexible packing under hard roof. journal of mining and safety engineering, 29, 757-761. he, m. c., chen, s. y., guo, z. b., yang, j., gao, y. b. (2017). control of surrounding rock structure for gob-side entry retaing by cutting roof to release pressure and its engineering application. journal of china university of mining and technology, 46, 959-969. jiang, j. p., dai, j., li, q. h. (2013). failure law and application of complex structure thin coal seam mining face. journal of china coal society, 38, 1912-1916. khademalhoseiny, m. s., nadoushan, m. a., & radnezhad, h. (2017). site selection for landfill gas extraction plant by fuzzy analytic hierarchy process and fuzzy analytic network process in the city of najafabad, iran. energy & environment, 28(7), 763-774. li, d. q. (2014). underground hydraulic mining of thin sub-layer as protective coal seam in coal mines. international journal of rock mechanics and mining sciences, 67, 145-154. li, l., bai, j. b., xu, y., xiao, t. q., wang, x. y., & zhang, k. x. (2011). research on rock control of roadway with complex roof driven along goaf. journal of mining and safety engineering, 28(3), 376-383. liu, h. b., cheng, y. p., song, j. c., shang, z. j., & wang, l. (2009). pressure relief, gas drainage and deformation effects on an overlying coal seam induced by drilling an extra-thin protective coal seam. international journal of mining science and technology, 19, 724-729. liu, y. k., zhou, f. b., liu, y., liu, c., & hu, s. y. (2011). experimental and numerical investigation on the deformation of overlying coal seams above double-seam extraction for controlling coal mine methane emissions. international journal of coal geology, 87, 139-149. liu, h. y., cheng, y. p., chen, h. d., mou, j. h., & kong, s. l. (2013). characteristics of mining gas channel expansion in the remote overlying strata and its control of gas flow. international journal of mining science and technology, 23, 481-487. ma, l. q., zhang, d. s., chen, t., & fang, g. w. (2007). study on packing body supporting resistance of enter-in packing for in-situ gob-side entry retaining in fully-mechanized top-coal caving mining face. chinese journal of mechanical engineering, 26, 544-550. rosa, m., barboza, e. g., menegon, b. s., frota, p. c., & anoni, r. (2016). geological evolution and inundation hazard: an analysis through geotechnologies. journal of coastal research, (752), 1227-1231. wu, c. f., zeng, y., & qin, y. (2004). present situation, application, and development of simultaneous extraction of coal and gas. journal of china university of mining and technology, 33, 137-140. yuan, l. (2008a). gas distribution of the mined-out side and extraction technology of first mined key seam relief-mining in gassy multi-seams of low permeability. journal of mining and safety engineering, 3, 1-7. yuan, l. (2008b). the technique of coal mining and gas extraction by roadway retaining and borehole drilling. journal of china coal society, 33, 898902. zhou, h., zhang, r., cheng, y., dai, h., ge, c., & chen, j. (2015). methane and coal exploitation strategy of highly outburst-prone coal seam configurations. journal of natural gas science and engineering, 23, 63-69. zhou, f., xia, t., wang, x., zhang, y., sun, y., & liu, j. (2016). recent developments in coal mine methane extraction and utilization in china: a review. journal of natural gas science and engineering, 31, 437-458. keywords: irrigation amount; alhagi sparsifolia; desertification; control; regression analysis. palabras clave: cantidad de riego; alhagi sparsifolia; desertificación; análisis de regresión. how to cite item li, j., liu, w, & zhang, w. (2020). evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts. earth sciences research journal, 24(4), 449-457. doi: https://doi.org/10.15446/esrj.v24n4.91626 desertification control is an important issue that must be considered in modern society. in order to effectively improve the accuracy and practicability of the evaluation algorithm of desertification control effect, the alhagi sparsifolia index under different irrigation amount was taken as the research object, and the evaluation algorithm of desertification control effect was proposed. in the “vegetation-sandstorm-soil” index system, a number of indexes were selected according to the core environmental parameters of alhagi sparsifolia and grassland desertification. and the analytic hierarchy process, remote sensing, geographic information system, and landscape technology were used to assign index weights of desertification control capacity, which were calculated by multiple discriminant matrices. finally, the data regression analysis was performed based on remote sensing and computer image information screening and processing to determine the final evaluation results. the experimental data show that the true positive rate of the algorithm in this paper is between 160 and 200, which is within a large range of advantages, indicating that the overall evaluation accuracy of the algorithm is high and the evaluation effect is perfect. abstract evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts algoritmo de evaluación para el control de la desertificación del alhagi sparsifolia bajo diferentes cantidades de riego issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.91626 el control de la desertificación es un tema importante que debe considerarse en la sociedad moderna. con el fin de mejorar efectivamente la precisión y la viabilidad de un algoritmo de evaluación del efecto de control de la desertificación se tomó como objeto de investigación el índice del alhagi sparsifolia con diferentes cantidades de riego. en el sistema de índices “vegetación-tormenta de arena-suelo” se seleccionaron varios índices de acuerdo con los parámetros ambientales básicos del alhagi sparsifolia y la desertificación de los pastizales. se utilizó el proceso de jerarquía analítica, la teledetección, el sistema de información geográfica y la tecnología del paisaje para asignar los índices de ponderación de la capacidad de control de la desertificación, los cuales se calcularon mediante matrices discriminantes múltiples. finalmente, el análisis de regresión de datos se realizó con base en la detección y el procesamiento de información de imágenes de computadora y detección remota para determinar los resultados finales de la evaluación. los datos experimentales muestran que la tasa de verdaderos positivos del algoritmo en este artículo está entre 160 y 200, lo que se encuentra dentro de una amplia gama de ventajas y que indica que la precisión de evaluación general del algoritmo es alta y el efecto de evaluación es perfecto. resumen record manuscript received: 18/11/2019 accepted for publication: 10/08/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 449-457 juan li1*, wei liu2, xinxin zhang3 1inner mongolia chemical vocational college, computer and information engineering, hohhot 010010, china 2ministry of technology, beijing rongzhilian science and technology co., ltd., beijing 100015, china 3ministry of environmental protection, shenhua jungle energy co., ltd., erdos 010300, china *corresponding author: 2485397887@qq.com e n v ir o n m e n ta l g e o l o g y https://doi.org/10.15446/esrj.v24n4.91626 https://doi.org/10.15446/esrj.v24n4.91626 mailto:2485397887@qq.com 450 juan li, wei liu, xinxin zhang introduction in 1977, the united nations conference on desertification defined desertification as a situation in which the biological potential of the land is reduced or destructed and ultimately leads to a desert-like landscape. this definition has been widely accepted and applied over the past two decades (lan et al., 2018). in 1991, aiming at the definition in 1997, the united nations environment program did not mention the possible areas of desertification and their causes. according to the latest research results, desertification was redefined as a land degradation process caused by human adverse effects in arid, semi-arid, dry, and semi-humid areas. throughout the definition and research of desertification in foreign countries, the concept of desertification is extensive (qiang et al., 2018), including vegetation degradation, water erosion, salinization, reduction of soil organic matter, soil crusts, and tightening, and accumulation of toxic substances. these are the concepts of generalized desertification, which is essentially land degradation. from the reality of our country, the concept of desertification has also been proposed in china, only referring to the process of wind erosion in generalized desertification. there are three main definitions. the first is the definition proposed by wu zheng: in arid, semi-arid, and partially semi-humid areas, due to natural factors or human activities, the fragile balance of the natural ecosystem is destroyed. in this non-desert region, there are environmental changes in desert-like landscapes characterized by sandstorms, and desert environmental conditions in desert areas are strengthened and expanded. the second is the definition of zhu zhenda: on a certain sandy material basis and under the conditions of drought and strong winds, an environment or land degradation process with windstorm as the main symbol due to excessive human activities inconsistent with resource and environment. this definition of desertification aims at the period of human history. it mainly appears in arid, semi-arid and partially semihumid areas and excessive economic activities are the main factors in the cause. based on the concept of desertification in the united nations, the third definition was proposed by dong yuxiang: a land degradation process characterized by windstorms in arid, semi-arid and semi-humid areas, and it is caused by climate change and human activities. this definition takes into account the human and natural factors (zhao et al., 2019; kang; 2018). the most serious desertification in land resources is grassland. grassland desertification is a special type of grassland degradation, which refers to land degradation caused by unbalanced human-land relationships in arid, semiarid, and semi-humid grassland areas. it is one of the extreme manifestations of grassland ecosystem degradation. grassland desertification, vegetation degradation, and surface exposure will intensify wind erosion and wind accumulation, which will cause continuous deterioration of the environment and a series of socio-economic problems. therefore, the restoration and reconstruction of degraded grassland ecosystems have received increasing attention. most of the grasslands in northern china are in arid and semi-arid climates, and the ecological environment has a very limited anti-interference ability. in the mid-1970s, the area of degraded grasslands in the country accounted for about 15% of the total grassland area. by the mid-1980s, it had increased to more than 30%. in the mid-1990s, it reached more than 50%, which had increased to 90% by the beginning of the 21st century. grassland productivity is declining and the grassland ecological environment is rapidly deteriorating (peng et al., 2018). therefore, the continuous increase of desertification of grassland has seriously affected the ecological environment construction, social and economic development of grassland areas in china, and caused great harm. in order to effectively curb the trend of grassland desertification, the state has launched a series of grassland desertification control projects, aiming at increasing the vegetation coverage, preventing sand and stopping desertification, and improving the grassland environment. especially since 2000, a series of large-scale grassland vegetation restoration and reconstruction ecological projects and scientific research (wei, 2017) has been launched. alhagi sparsifolia belongs to leguminosae sp, alhagi gagneb. it is a perennial semi-herb plant, mainly distributed in xinjiang, gansu, inner mongolia, china, and the former soviet union, iran, afghanistan, pakistan, and india. alhagi sparsifolia is one of the representative shrub vegetation species in the arid area of xinjiang, china, with a total area of 1.7 million hectares. it is important for sand-fixing vegetation and plays an important role in maintaining the local ecological environment. at present, grassland desertification control through alhagi sparsifolia is a common method in the world. in order to further study the effect of desert control, it is necessary to combine the natural characteristics of alhagi sparsifolia and construct an evaluation system of desertification control effect according to its natural growth ability under different irrigation amounts. material and methods evaluation algorithm for desertification control of alhagi sparsifolia under different irrigation amount in this paper, the index system of “alhagi sparsifolia vegetation-sandstorm activityalhagi sparsifolia soil” was adopted. according to the difference of irrigation amount, the vegetation coverage, aboveground biomass, grass height, plant species, annual plant proportion, fixed sand proportion, fixed sand fragmentation, mobile sand proportion, mobile sand fragmentation, soil erosion modulus, soil organic matter, soil mechanical composition, soil crust and other indicators were selected with the analytic hierarchy process to construct the ecological benefit evaluation index system of grassland desertification control project (meng et al., 2018). through the ground surveys, data on alhagi sparsifolia status and soil were obtained, and gps handheld terminals were used as an aid to locate and measure ground survey points (qiang et al., 2017). according to the meteorological data, the administrative zone map, the grassland type map, the topographic map of the study area, the knowledge of the alhagi sparsifolia, and the images obtained by the landsat 5 sensor and the landsat 7 sensor, the vegetation coverage, the aboveground biomass, the vegetation height, the change of plant species number of the alhagi sparsifolia are obtained inside and outside the project. in addition, the proportion of fixed and mobile sand, the change of fragmentation degree of fixed and mobile sand landscape before and after the implementation of the project can be analyzed to realize a quick and comprehensive evaluation of the ecological benefits of a grassland desertification control project. the overall process is as follows: the ecological benefit evaluation system of alhagi sparsifolia desertification control the grassland desertification control project mainly refers to the various measures in favor of the growth of alhagi sparsifolia or adjusting the land use structure. through the project, the vegetation on the desertification grassland gradually recovers, the sandstorm gradually disappears, the surface is stable, the crust is formed, and the soil is reconstructed after that. all these make great ecological and economic benefits. the ecological benefits of the grassland desertification control project refer to the protection and improvement of the natural environment, human life, and production environment after the implementation of the ecological environment construction projects. the ecoefficiency index is a kind of indicator that reflects the improvement of the environment, including indexes of improving the soil, preventing wind and sand, controlling soil erosion, regulating climate, conserving water sources, reducing disasters, and improving ecological environment conditions under the influence of alhagi sparsifolia. in this paper, the irrigation amount was taken as the core variable, so the vegetation coverage growth rate, grassland productivity, plant species, desertification land area, and soil organic matter content changed with it. the basis of ecological benefit evaluation under the influence of different irrigation amounts is the evaluation index and its system. even the national conditions, industry, and engineering construction properties are the determinants of the construction evaluation index system and evaluation index. at present, there is no uniform scope and standard at home and abroad. therefore, it is necessary to establish a practical, reasonable, and operable ecological benefit evaluation index system for the grassland desertification control project (yang et al., 2017). principles of index selection (1) principle of geographical regional differences grassland desertification in china is mainly distributed in sub-humid arid, semi-arid, and arid regions. there are significant differences in vegetation 451evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts (3) principle of specific indicator operability the ecological benefit index should reflect the environmental benefits brought by the grassland desertification control project in a comprehensive and systematic way. the indexes to be evaluated should be easy to obtain and to be processed. the original data should be easy to obtain through surveys, statistics, remote sensing, and other means, and be easy to be processed through quantitative calculation. they should also have clear realistic meaning. (4) the irreplaceable principle of a specific index there are many indexes for evaluating the ecological benefits of the grassland desertification control project. most of the indexes have a certain correlation and mutual transmission. therefore, when determining specific indexes, it is necessary to avoid the related indexes caused by the dominant factors. the determined indexes can make a scientific and reasonable evaluation of the construction effect of the project, and the indexes are irreplaceable (zhang et al., 2018). (5) cross-scale principle the development and application of remote sensing, geographic information systems, and gps technology have greatly broadened the research horizons of people; on the other hand, it also poses certain challenges to conventional work programs based on ground surveys. since there are differences between the ground survey and remote sensing, geographic information system, and gps in workflow and requirements, it is likely that different research conclusions will occur. researchers in the same area have found conflicting results only because different research methods are not expected. therefore, when selecting evaluation indexes, cross-scale indexes that can be compatible with different research methods to avoid the above problems should be selected. evaluation indexes and their meanings the grassland desertification control project can be similarly regarded as the reverse process of grassland degradation. the process is first reflected in the change of the composition of alhagi sparsifolia vegetation, and then the change of soil properties. alhagi sparsifolia vegetation restoration and soil restoration are interaction processes that reflect the nature of overall environment restoration. these two processes are sometimes consistent in time, and sometimes there is a certain hysteresis. that is, vegetation restoration and soil restoration occur simultaneously or vegetation is restored first, and soil restoration is only manifested after a certain period of time. an important factor associated with grassland degradation or restoration is sandstorms. according to the above principle of index selection, the ecological benefit evaluation of the grassland desertification control project is mainline. the three indicators, vegetation, soil, and sandstorms, that reflect the ecological benefits of the grassland desertification control project are selected as the criterion layers. the influencing factors related to project construction and ecological benefits are selected to form an index system. a comprehensive analysis is performed on the index system, compared with other ecological benefit evaluation index systems, and some indexes were removed according to the scientific, practical, and measurable requirements, and the ecological benefits they will bring. the ecological benefit evaluation index system of the grassland desertification control project was initially determined (zhang et al., 2018). the framework of the ecological benefit evaluation index system is shown in figure 2. the role of alhagi sparsifolia vegetation is the most important. alhagi sparsifolia vegetation is not only a comprehensive reflection of the quality of the project land, but also the most ideal protector of the soil, and also the main body of the ecosystem. the vegetation index mainly refers to the ones describing the shape and productivity of alhagi sparsifolia, such as the community composition, life form, biomass, and its coverage. undoubtedly, most camel thorn vegetation indexes can be obtained through ground surveys. however, some indexes that characterize vegetation productivity, such as alhagi sparsifolia coverage and biomass, have been quantitatively extracted by remote sensing data. although there are some differences in the data obtained at different scales, these indexes are clearly cross-scale. this paper selects four vegetation indexes that are operable and can directly reflect the ecological benefits of alhagi sparsifolia in the desertification control project. they are the types and structural characteristics in different regions. due to the influence of the natural environment, climate, and natural vegetation growth, the project measures adopted have significant differences. there are also significant differences in vegetation restoration ability and natural vegetation coverage in different regions. therefore, in order to reasonably evaluate the grassland desertification control project, the ecological benefit index should be determined according to the actual situation of the study area. (2) principle of dominant index the purpose of the alhagi sparsifolia grassland desertification control project is to control or reduce grassland desertification. because there are many factors affecting the ecological benefit index of the grassland desertification control project, when determining the specific index, it is necessary to combine the actual situation and select the core factors that can express the degree of grassland desertification control as the dominant index. the so-called core factors refer to the ones that control the wind erosion on the grassland, the formation of sandstorms, and the surface exposure. according to the research results and actual application, the dominant factor is vegetation change. figure 1. process of control evaluation 452 juan li, wei liu, xinxin zhang figure 2. index system coverage of the alhagi sparsifolia, the aboveground biomass, the height of the grass layer, and the proportion of annual plants (guan et al., 2018). (1) alhagi sparsifolia vegetation coverage refers to the ratio percentage of the vertical projection area to the sampling area of the total alhagi sparsifolia community. coverage is the most critical factor in vegetation, and it reflects the degree of vegetation and the size of the photosynthesis area. it is also the main factor affecting soil erosion, which is of great significance for environmental change and monitoring research in the control area. (2) aboveground biomass: the total weight (fresh or dry weight) of one aboveground plant (or all plants) per unit area, which is the quantitative index of forage growth. aboveground biomass indicates the growth rate and regenerative capacity of pasture. in general, for the same plant community, the amount of above-ground biomass can indicate the size of the coverage and leaf area, and can also indicate the ability of photosynthesis. (3) grass height: the grass height in nature can reflect the production status of the pasture, and has a great relationship with the utilization mode. the natural state of the grassland plants was maintained during the measurement, and the average height of the grass layer was measured (ajaj et al., 2017). (4) proportion of annual plants: the percentage of annual plant coverage of total plant coverage. the proportion of annual plants can reflect the stability of plant communities. when the proportion of annual plants is small, the stability of the community is good; otherwise, the community is unstable. wind erosion is the loss process of surface materials under the action of wind. it is the beginning of the sandstorm and desertification process. the sub-processes in the desertification process are directly or indirectly generated by wind erosion. wind erosion is one of the most important causes of grassland desertification. after desertification, the grassland is dominated by bare sand, and the floating sand, semi-fixed sand, and fixed sand are randomly distributed among them. therefore, the proportion of fixed sand, mobile sand, and the size of wind erosion modulus are significant indicators for evaluating the effects of grassland desertification control projects. in addition, this paper introduces the landscape fragmentation, a landscape analysis index, to quantitatively study the dynamic characteristics of mobile sand and fixed sand in the study area. (1) proportion of fixed sand: the percentage of fixed sand area in the total project land area. the larger the proportion of fixed sand, the better the recovery of the project area. (2) proportion of mobile sand land: the percentage of mobile sand area in the total project land area. the smaller the proportion of mobile sand, the better the recovery of the project area. (3) soil wind erosion modulus: the amount of wind erosion per square kilometer per year, and its common unit is t/km2*a. the soil wind erosion modulus value visually reflects the degree of wind erosion in the soil. (4) fragmentation: landscape fragmentation refers to the process of landscape plaques from simple to complex due to disturbances of natural or human factors, that is, the process of the landscape from single homogeneous to 453evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts complex heterogeneity. the total number of plaques, the number of plaques per unit area, the plaque area per unit area, and the boundary density are commonly used indicators to reflect the characteristics of landscape fragmentation. landscape fragmentation indicates that the landscape shape is fragmented, enhances the edge effect of the landscape, and reduces the total area of a certain type of landscape and the core area of each plaque. the soil is composed of four parts: minerals, organic matter, soil moisture, and soil air (tang et al. 2018). soil indexes are mainly indexes that describe the physical, chemical, and biological characteristics of the soil, such as soil thickness, soil texture, mechanical composition, structure, nutritional status, organic matter content, profile characteristics, salt content, salt type, soil microbial status. these indexes are mostly ground surveys or analytical assay indexes. although quantitative data on soil indexes through microwave remote sensing can be achieved in scientific research, it is costly and far from reaching the application stage. soil is one of the necessary conditions for vegetation growth. three main indexes that reflect the environmental characteristics of vegetation were selected to characterize the effect of the grassland desertification control project to restore the soil. they are soil organic matter, soil mechanical composition, and soil crust. (1) soil organic matter: soil organic matter content is one of the most important indicators of soil properties. it indicates various nutrients contents in the soil, especially nitrogen and phosphorus, and represents the potential capacity and stability of soil fertility. it is the key factor in the health of the soil. the soil sample was taken back to the laboratory for analysis to determine the organic matter content (g/kg) in the soil. the content of soil organic matter is an important index of soil fertility. organic matter also plays an important role in the physical and chemical properties of the soil. for example, soil organic matter can improve the physical structure and chemical properties of the soil, which is conducive to the formation of soil aggregate structure, promotion of nutrient absorption, and growth of plants. (2) soil mechanical composition: soil is composed of different sizes of soil particles in different proportions. these different grain sizes are mixed together to characterize the soil thickness, which is called soil mechanical composition or soil texture. it affects the soil moisture, air, and heat movement, and also affects the conversion of nutrients and the type of soil structure. the mechanical component is a stable natural property of soil, which determines the behavior and use of soil to a large extent, and is one of the basic materials necessary for studying soil wind erosion. the aluminum box was used to take the surface sand sample for indoor analysis, and the fine sand proportion is one evaluation index. (3) soil crust: the crust is a thin layer formed along with the fixation of sandy land, the accumulation of organic matter on the surface of the soil, and the growth of lower plants. soil crust is an important indicator of surface wind erosion and stationary status (dai et al., 2017). a clear indicator of the restoration of sandy vegetation communities is the degree of surface fixation. the better the restoration of sandy vegetation, the higher the degree of sandy land fixation. the visual index reflecting the degree of surface fixation is the thickness of the surface crust of sandy land. according to research, the crust material is mainly a mixture of fine particles (fine sand and ultrafine sand accounted for 93), salts, and amorphous silicon gel. it has strong wind erosion resistance and can resist wind with a speed of 30 m/s. however, if the crust is too thick, it will also inhibit the growth of sandy plants. for some regular fixed sands, the crust is too thick, affecting water infiltration, and the soil moisture condition is deteriorated, which will lead to poor plant regeneration. index weight calculation the weight determination of each evaluation index plays an important role in the comprehensive evaluation of the control effect of alhagi sparsifolia. the weight of the evaluation is very important for the final evaluation result, which reflects the relative importance of each index. in the vertical direction, it reflects the effect of index variety on the change of comprehensive ecological benefit. in the horizontal direction, it indicates the position of the index in the same evaluation index level. whether the weight is determined reasonably will directly affect the judgment results (sherratt & synodinos, 2017). the analytic hierarchy process was used to construct an analytic model and determine the index weights. as a basic evaluation method, the analytic hierarchy process (ahp) plays an extremely important role in multi-objective decision-making problems. it also has extremely high using frequency. to analyze problems with ahp, the problems should be organized and leveled first to construct a structural model of hierarchical analysis. the hierarchical analysis structure model usually consists of three layers. there is only one element in the highest layer, which is generally the target or ideal result of the analysis problem, so it is also called the target layer. the middle layer includes the intermediate links to achieve the goal. it can be composed of several layers, including the rule and sub-rule that need to be considered, so it is also referred to as the rule layer. the index layer is below the rule layer (rosen et al., 2019). table 1. the ecological benefit evaluation system of alhagi sparsifolia in the desertification control project a target layer rule layer b index layer c ecological benefits vegetation vegetation coverage aboveground biomass grass layer height growth rate sandstorm activities fixed sand scale fixed sand landscape broken moving sand landscape broken soil soil wind erosion modulus soil organic matter the proportion of fine sand soil crust according to the current data in table 1, a hierarchical judgment matrix was constructed. the judgment matrix indicates the relative importance of the elements in this hierarchy to elements in the previous one. a comparison judgment matrix was constructed by comparing the relative importance of the element in two neighboring hierarchies. based on the scale of 1-9 and its reciprocivity as the measurement scale, saaty proposed the corresponding judgment matrix. the scale is shown in table 2. table 2. the scale of judgment matrix factor scale meaning 1 meaning that the two factors are equally important 3 represents two factors, the former is slightly more important than the latter 5 compared with the two factors, the former is significantly more important than the latter 7 indicating that the former is more important than the latter in terms of the two factors 9 represents two factors, the former is more important than the latter 2468 represents the middle digit of the adjacent judgment of 1, 3, 5, 7 and 9 the derivative of the above value indicates the degree to which two factors are more important than the latter 454 juan li, wei liu, xinxin zhang according to the scale standard in table 2, the corresponding matrix is formed, and the judgment matrix of the ecological benefit evaluation factor of the grassland desertification control project is compiled as follows: a b =                   1 3 5 1 3 1 3 1 5 1 3 1 (1) b c1 1 2 3 2 2 1 2 1 2 2 2 1 3 1 3 1 1 3 1 3 1 2 1 2 2 1 1 1 2 1 2 2 1 1  =                               (2) b c2 1 3 2 3 3 1 3 1 1 2 1 1 1 2 1 2 1 1 2 1 2 1 3 1 2 1 1 1 1 2 2 1 1  =                               (3) b c3 1 3 2 1 3 1 1 2 1 2 2 1  =                   (4) there are three kinds of judgment matrix calculation methods: arithmetic mean method, geometric mean method, and maximum eigenvalue method. the geometric mean method was used in this paper. for example, the relative importance coefficient for problem a is estimated based on the a-b judgment matrix. calculate the square root of the product of all elements in each row of the judgment matrix. w ni j n ij= =1   (5) wi can be normalized according to formula (5) to obtain normalized weight. because the judgment matrix needs to be fuzzy quantized based on empirical data, it is impossible to achieve complete consistency. a concept of random consistency ratio was proposed based on this, and its unit is cr. when the cr is less than 0.1, the consistency is considered to be satisfied; otherwise, the matrix needs to be re-adjusted. the cr calculating formula is as follows: cr ci ri = (6) where c indicates the consistency index, ri is the proportional coefficient, which is related to the order of the judgment matrix n. in this situation, ci can be expressed as: ci n n = ( )λ − − max 1 (7) where max represents the maximum eigenvalue of the matrix, and the judgment results are as follows: table 4. a-b evaluation matrix a b1 b2 b3 wi wi b1 1 3 5 2.466 0.673 b2 1 3 1 3 1 0.258 b3 1 5 1 3 1 0.405 0.105  1 table 5. b1-c judgment matrix b1 c1 c2 c3 c4 c5 wi wi c1 1 2 3 2 2 1.888 0.345 c2 1 2 1 3 2 2 1.458 0.258 c3 1 3 1 3 1 1 2 1 2 0.858 0.087 c4 1 3 1 3 1 1 2 1 2 0.455 0.157 c5 1 2 1 2 2 1 1 0.870 0.157  1 table 6. b2-c judgment matrix b2 c6 c7 c8 c9 c10 wi wi c6 1 2 3 2 2 1.888 0.345 c7 1 3 1 3 2 2 1.458 0.258 c8 1 2 1 3 1 1 2 2 0.858 0.087 c9 1 2 1 3 1 2 0.455 0.157 c10 1 2 2 1 0.870 0.157 1 using the results of a single ranking of all hierarchies, the relative importance weights of all elements in this hierarchy were calculated compared to the previous one, which is the total ranking of hierarchies. the calculation needs to be performed hierarchy by hierarchy from top to bottom. for the second hierarchy below the highest one, the hierarchical single order is the total order. according to the consistency test, the cr values of the hierarchical single order and the hierarchical total order are both less than 0.10. table 3. random consistency proportion coefficient order number 1 2 3 4 5 6 7 8 9 the values 0 0 0.58 0.90 1.12 1.24 1.32 1.41 1.45 455evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts camelthorn index extraction and treatment evaluation analysis of information source data the ecological benefit evaluation system of alhagi sparsifolia desertification control and calculation results of its index weights were processed through the arcgis9.0 software. the software made a project distribution map by introducing surface attributes to the inflection point coordinates of the project and defining projections. the processing of remote sensing data was mainly performed in erdas image8.7 remote sensing image processing software, which realized image preprocessing, including the geometric correction and excavation of remote sensing images. the software can also realize enhanced processing of remote sensing information, extraction of point attributes, classification, and classification accuracy verification. the field gps data were spatialized and projected by arcgis 9.0 software to obtain a shapefile vector file. the correlation between ground survey data and remote sensing data was analyzed by mathematical-statistical analysis software spss to establish a regression model. remote sensing image processing during the imaging process of satellite remote sensing, radiation errors and geometric distortions may occur due to the attenuation of the sensor function, the influence of the atmosphere, and the flight mode of the satellite. therefore, the image should be subjected to radiation correction and geometric correction before information can be extracted. the remote sensing image used in this paper had been subjected to standard atmospheric radiation correction and scatter correction at the satellite ground station before purchasing, and the geometric correction of the system had been carried out according to the parameters of the sensor, which eliminated the geometric distortion caused by the earth rotation, satellite mode changes, and sensor aging. the real positions of a series of pixels that were distributed uniformly in the remote sensing image were obtained by gps. these positions are usually called ground control points (gcp). because the research area is mainly sandy land and transportation there is inconvenient, gcp cannot be obtained in some areas. therefore, the two remote sensing images were combined, and the previously corrected tm remote sensing image was taken as a correction reference. finally, the standard error of the correction model (root means squared error, rmse) was controlled within 0.5 pixels. evaluation of control effect according to the control weights obtained by the above analytic hierarchy process, combined with the experimental data under different irrigation amounts, the vegetation index was obtained. the general process of extracting vegetation index using erdas remote sensing image processing software is below: step1: open the remote sensing image using erdas remote sensing processing software. step2: according to the vegetation index formula, the band calculation was performed on different bands of the image in the modeling tool. step3: obtain image files of vegetation index under different irrigation amounts. the regression statistical analysis table based on the current index is as follows: table 8. the regression statistical analysis table sources of variance sum of squares degrees of freedom the mean square statistic return to the 72133.845 1 72133.852 62.179 residual 25525.38 23 1107.287 the sum of the 67957.602 24 the correlation coefficient 0.859 fitting 0.738 the slope 1035.5 intercept 21.352 it can be seen from table 8 that the correlation coefficient between the index and the aboveground biomass is 0.859. after the regression significance test, the regression effect reaches a significant level. the vegetation index and the model of aboveground biomass were obtained: y x= +1035 31 278. (8) where y is the biomass of alhagi sparsifolia under different irrigation amounts, x is the vegetation index. the grassland estimation model was established by the model module in the erdas software, and the biomass of each pixel on the image can be obtained to form a spatial distribution map of the aboveground biomass. then the aboveground biomass data in different project areas were summarized to get the final evaluation results. results the vegetation data obtained from the ground surveys and remote sensing image processing are taken as the basis. in the same period, the adjacent area, the enclosed development area, and the free grazing area outside the control project area were taken as the horizontal comparison, and the condition before the implementation of the project was taken as the vertical comparison. a comprehensive analysis was performed on the evaluation indexes of each vegetation index to see whether each index develops in a favorable direction, and thus draws the overall evaluation conclusion. the correctness of the conclusion is tested through experiment. in order to determine whether the evaluation algorithm of the control effect in this paper can effectively evaluate the control data and improve the evaluation accuracy, evaluation indexes usually select the cross-data samples, which are commonly used in the effect evaluation model. the area under the characteristic curve of alhagi sparsifolia was used to evaluate the comprehensive performance of the current evaluation method. the abscissa is the false positive rate (fpr) of the roc curve and the ordinate is the true positive rate (tpr). the area under the curve was used to determine whether the evaluation method has a fitting phenomenon and a larger area indicates a better evaluation effect. it can be seen from the data in figure 3 that the area under the overall characteristic line of this state evaluation method is significantly higher than the conventional method used for comparison. its true positive rate is between 160 and 200, which is within a large advantage range, indicating that the overall evaluation is accurate and its effect is perfect. this is because the algorithm in this paper first constructed a “vegetation-sandstorm-soil” index system, selected a number of indexes, and calculated their index weights. according to the calculation results, the analytic hierarchy process was used, with the help of remote sensing, geographic information system, and landscape technology, to assign weights to the grassland control indexes. finally, remote sensing, computer image information screening, and processing were used to perform data regression analysis and obtain the final evaluation results, which effectively improved the accuracy of the evaluation results. table 7. b3-c judgment matrix b3 c11 c12 c13 wi wi c11 1 3 2 1.888 0.345 c12 1 3 1 2 1.458 0.258 c13 1 2 2 2 0.858 0.087  2 0.858 0.087 456 juan li, wei liu, xinxin zhang discussion desertification is one of the most important environmental and socioeconomic problems in modern society, threatening the survival and development of mankind. china is one of the countries seriously affected by desertification in the world. desertification has attracted much attention because of its development speed and impact on the social economy. according to statistics, in terms of its development rate, in the 1950s-1960s, desertification land in china expanded to 1,560 km2. the desertification land in the 1970s-1980s expanded to 2,100 km2 and reached 3,600 km2 in the 1990s. the area of desertification land has increased by 46.74 million km2 from 1987 to 2000 and reached 385,700 km2 by the year 2000. compared with the monitoring results in 1987, the area of desertification land has decreased after more than ten years of control, and the desertification degree has been reduced. however, the overall development of desertification is greater than the reversal. especially in the past ten years, there has been a significant increase. under this background, this paper proposes an evaluation algorithm for the effect of alhagi sparsifolia desertification control under different irrigation amounts. the experimental results show that compared with the traditional methods, the algorithm in this paper has higher prediction accuracy and better prediction effect, which fully verified its effectiveness and feasibility. on the basis of obtained accurate assessment results, it helps relevant departments to propose environmental management programs in a targeted manner, which is conducive to the improvement of environmental problems and has practical significance. conclusion by reviewing the literature, summarizing the previous research and evaluation theory of desertification control and consulting experts, this paper adopted the alhagi sparsifolia vegetation, sandstorm, soil, and other indexes as a system, and selected the vegetation coverage, aboveground biomass, grass height, plant species, annual plant proportion, fixed sand proportion, fixed sand fragmentation, mobile sand proportion, mobile sand fragmentation, soil erosion modulus, soil organic matter, soil mechanical composition, soil crust and other indexes. these indexes almost cover the main ecological and environmental impact factors brought by the alhagi sparsifolia project and accurately and comprehensively reflect the ecological benefits of the project under different irrigation amounts. in addition, the paper uses the analytic hierarchy process to determine the index weights. the outstanding advantage of the analytic hierarchy process is that it can minimize the influence of subjective factors, as well as express and treat human subjective judgments in quantitative form. moreover, the mathematical principle of the ahp is rigorous, simple, and easy to implement, and the factors of complex problems are organized by dividing the order of interconnection. ahp also gives objective judgment and quantitative expression of the relative importance of each evaluation index in each hierarchy, which provides an effective technical method to solve the weight distribution in the evaluation of ecological benefits. according to the experiment data, it can be known that the evaluation algorithm in this paper has numerous reasonable results and application advantages. references ajaj, q. m., pradhan, b., noori, a. m., & jebur, m. n. (2017). spatial monitoring of desertification extent in western iraq using landsat images and gis. land degradation & development, 28(85):7-8. dai, q., peng, x., zhi, y., & zhao, l. (2017). runoff and erosion processes on bare slopes in the karst rocky desertification area. catena, 152(17):218-226. guan, p., zhang, x., yu, j., cheng, y., li, q., andriuzzi, w. s., & liang, w. (2018). soil microbial food web channels associated with biological soil crusts in desertification restoration: the carbon flow from microbes to nematodes. soil biology & biochemistry, 116(10):82-90. kang, s. (2018). prevention and control mode of desertification in china’s state energy group shendong mining area with annual coal output of 200 million tons. meteorological & environmental research, 9(02):27-33 lan, s., zhang, q., he, q., yangm h., & hu, c. (2018). resource utilization of microalgae from biological soil crusts: biodiesel production associated with desertification control. biomass & bioenergy, 116(3):189-197. meng, z., dang, x., yong, g., ren, x., ding, y., & wang, m. (2018). interactive effects of wind speed, vegetation coverage and soil moisture in controlling wind erosion in a temperate desert steppe, inner mongolia of china. journal of arid land, 10(4):1-14. peng, x., dai, q., li, c., & zhao, l. (2018). role of underground fissure flow in near-surface rainfall-runoff process on a rock mantled slope in the karst rocky desertification area. engineering geology, 243(58):10-17. qiang, c., zeng, y., huang, c., su, j., sun, z., zhang, q., & liu, y. (2018). sophora davidii: a native shrub used for rocky desertification control and development. meteorological & environmental research, 9(2):24-26. the flow of the test evaluation algorithm: table 9. the analysis process of test evaluation algorithm data input: evaluation information data point set x,g×g grid, (g+1) × (g+1) grid output: sample for each data point process step1 enter each data point xi step2 distribution point xi go to the corresponding indicator network step3 define an empty set of governance metrics and collect cluster examples step4 take each datagiput it into the corresponding grid step5 construct similarity matrix s1and s2 put them into the corresponding matrix step6 merge the sample to perform the third ap step7 for the current no matrix, construct a similar matrixs step8 the above formula is used to calculate the similarity step9 compute the governance data preference step10 distribution of x; corresponding example step11 end organized experimental data is as follows: figure 3. comparison of evaluation results of different methods 457evaluation algorithm of alhagi sparsifolia desertification control under different irrigation amounts qiang, l., pu, j., ni, h., du, h. m., qi, x. k., wang, l., & yang, h. (2017). a research approach for ecological, environmental and geological differentiation of rocky desertification and its driving mechanism in karst graben basin. advances in earth science, 32(9).19-85 rosen, a. m., hart, t. c., farquhar, j., schneider, j. s., & yadmaa, t. (2019). holocene vegetation cycles, land-use, and human adaptations to desertification in the gobi desert of mongolia. vegetation history and archaeobotany, 5(10):95-8. sherratt, j, & synodinos, a. (2017) vegetation patterns and desertification waves in semi-arid environments: mathematical models based on local facilitation in plants. discrete and continuous dynamical systems series b, (dcds-b) 17(8):2815-2827. tang, z., an, h., zhu, g., |& shangguan, z. (2018). beta diversity diminishes in a chronosequence of desertification in a desert steppe. land degradation & development, 29(3):35-9. wei, x. x. (2017). soil nutrient content and the activities of soil enzymes during the desertification restoration process to the east of qinghai lake. pratacultural science, 21(14):13. yang, w., tao, w., wei, f., wei, l., & yanli, l. (2017) the theory and progress of sand control with low coverage. journal of desert research, 3 (9).109-805 zhang, c. l., li, q., shen, y. p., zhou, n., wang, x. s., li, j., & jia, w. r. (2018). monitoring of aeolian desertification on the qinghai-tibet plateau from the 1970s to 2015 using landsat images. science of the total environment, 2(17):619-620. zhang, w., yu, x., li, m., lang, d., zhang, x., & xie, z. (2018). silicon promotes growth and root yield of glycyrrhiza uralensis under salt and drought stresses through enhancing osmotic adjustment and regulating antioxidant metabolism. crop protection, 107(3):1-11. zhao, y., ding, g., gao, g.., & le, p. (2019). regionalization for aeolian desertification control in the mu us sandy land region, china. journal of desert research, 9(15):4-6. earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 5 letter from editor more than two decades ago we have been working steadily to provide readers and scientific community with articles on a broad range of geosciences field around the world. this achievement would not be possible without the support of the universidad nacional de colombia, mainly, and partners who help bring these high-quality papers to light. one of those partners is colciencias, a colombian government-owned agency that support research in our country. currently, colciencias is carrying out a national strategic plan for science, technology and innovation in geosciences aimed to promote and develop megaprojects in the area, provide a management tool, and create a fund with resources coming from contracts of exploration and exploitation of petroleum. this 2015-2018 plan should result in the creation of new research groups, improvement of investigation centers, recognition of researchers and the double of resources in funding for projects. similar strategies are carried out by the centro de geociencias, at the national autonomous university of mexico. their primary objective is to gather the experience of their academicians to produce fundamental and applied investigation under the high-quality requirements. in venezuela, the non-government organization fundación geoparques de venezuela, under the motto "geosciences for human development", promotes research and divulgation of earth sciences with communities as the head axis. just these examples but in several countries of latin america government or non-government entities work for the development of geosciences. our call is for regional collaboration optimizing resources, promotion of open journal publications, no isolated academic communities, etcetera. on the other hand, find available our newest volume with a selection of ten works of the best quality in different subjects of the earth sciences. the first of them is a complete map of the seismotectonic and geologic hazards in iran evaluated using seismicity records, structural trends, tectonic settings, fault ruptures and neotectonics activities in the country. from colombia, the second manuscript is an approaching to the skarn and formation of the ti-clinohumite from its occurrence and petrologic significances. from china, but also in petrology subject, a paper presents geochemistry and petrography of the amphiboles and plagioclase from fenghuangshan granodiorite. combining with the barometry and thermometry analysis, the author estimates the conditions of the fenghuangshan magmatic rocks and explains the origin of the magma. moreover, evaluate the deeper exploration of copper mineralization prospecting potential. in applied geophysics, a paper coming from jordan provides information about integration of data obtained from geophysical methods, gis and the previous works and literature review of geology in southwestern jordan. about geomorphology, it is presented an altimetric and hypsometric analysis, with a longitudinal profile and geomorphological studies, for soil and water conservation based in the basins of anjani and jhiri rivers. as part of the strategies to warrant water resources, the method presented in next work is accomplished to be more efficient and cheaper to estimate runoff in a case based in tamilnadu district, in india. related to climatic change, one of the authors investigated the correlation of some climatic indicators with precipitation and discharge of madarsoo (iran) watershed. it is a meaningful study in the context of global climate change. the last block is dedicated to three works in the field of geotechnics, in areas of roads, construction of concrete piles and evaluation of polypropylene fibers for mining and civil engineering applications. we hope this articles will be useful for everyone of you. carlos vargas editor in chief, earth sciences research journal geociencias 15-1 julio 2011.vp communication regarding the geology and vertebrate palaeontology of the adamantina formation (bauru group, upper cretaceous) in campina verde, minas gerais state, brazil emerson ferreira de oliveira, caio césar rangel, diego sullivan de jesus alves, filipi silva limonta, adelino carvalho, carlos roberto a. candeiro laboratório de geologia da universidade federal de uberlândia campus pontal, ituiutaba-mg, brasil. e-mail author: oliveira_geoscience@yahoo.com.br a b s t r a c t several geological events occurred on the surface of the earth with the division of the megacontinent gondwana, including volcanic activity involving extensive lava flows, thus locally and regionally altering the surface’s geology. lava flows in the serra geral formation predominantly consist of volcanic lava fissure-derived basalt. sedimentation upon extrusive igneous rocks occurred after this volcanic event, leading to the deposition of sediments from three formations (adamantina, uberaba, and marília) collectively comprising the upper cretaceous bauru group, beginning about 83 million years ago. this was a period when large animals inhabited the planet, including the study area; the adamantina formation is the largest and oldest. the city of campina verde is located there and is where most studies were carried out; the region is known as fossil field. r e s u m e n varios eventos geológicos ocurren sobre la superficie de la tierra con la división del megacontinente de gondwana, incluyendo actividades volcánicas con extensos flujos lava, tanto local como regionalmente, que alteraron la geología de superficie. los flujos de lava en la formación serra geral consisten en su mayoría de lava volcánica fisurada derivada del basalto. la sedimentación sobre las rocas ígneas extrusivas, ocurrió después de este evento volcánico, depositando los sedimentos de tres formaciones (adamantina, uberaba, y marília) que comprenden el grupo bauru del cretáceo superior, y que inicio hace 83 millones de años aproximadamente. en este periodo habitaban grandes animales en el planeta, incluyendo el área de estudio; la formación adamantina es la más grande y antigua. la ciudad de campina verde, localizada en esta zona, es donde la mayoría de los estudios se llevaron a cabo y es conocida como un campo de fósiles. introduction the bauru basin in southern brazil and northeastern paraguay is related to tectonic and volcanic activity as part of gondwana during the cretaceous period. the basin covers about 370,000 km² in southern and central brazil. fernandes and coimbra (1996) identified the caiuá group (goio êre, rio paraná and santo anastácio formations) and the bauru group (admantina, uberaba and marília formations). the bauru group’s three sedimentary formations were deposited by wind and water, beginning 83 million years ago. these formations crop out in the states of goiás, mato grosso, mato grosso do sul, minas gerais (mg), paraná and são paulo and north-eastern paraguay, covering an area of 370,000 km²; sedimentation occurred in a hot, semi-arid climate, the main event occurring during the upper cretaceous period (brito, 2001). the bauru group’s three formations (in ascending order) are the adamantina, the uberaba and the marília formations. the uberaba formation is only found in the area surrounding the city of uberaba, mg, while the other two are found in every occurrence of the bauru group. this report will emphasise the adamantina formation which was studied in the triângulo mineiro region, specifically near the city of campina verde, mg. the bauru group’s rich fossil locations from the upper cretaceous period are found in the west of mg in central brazil; this is the earliest geological period represented in this area. triângulo mineiro palaeontological knowledge has been developing at pontal do triângulo mineiro (in the campina verde and monte alegre de minas area) since late 1930 and more recently in carneirinho, gurinhatã, iturama and prata. earth sciences research journal earth sci. res. s j. vol. 15, no. 1 (july, 2011): 39-43research groupin geophysics universidad nacional de colombia keywords: bauru group, stratification, fossil, sediment, campina verde. palabras clave: grupo bauru, estratificación, fósil, sedimento, campina verde record manuscript received: 30/11/2010 accepted for publication: 25/05/2011 recent discoveries have filled in gaps in our knowledge, yet much work still remains to be done. von huene (1934) recovered fragmentary fossil sauropod dinosaur remains while conducting palaeontological reconnaissance in campina verde in 1910; barcelos (1984) recognised the deposits in which they were found as being adamantina formation (fernandes and coimbra). researchers from faculdade de ciências integradas do pontal palaeontological group (facip) visited the field area and collected a number of isolated invertebrate and vertebrate fossil specimens. only a single paper describing dinosaur remains came from these studies (candeiro, 2007). methodology the studies were conducted in campina verde during 2009 by a group of researchers from the universidade federal de uberlândia’s geology laboratory (labgeol) in pontal with the support of the municipal government. the methodology involved using several research publications on work in campina verde regarding the region’s geology and palaeontology, as well as direct observation of specimens housed in the labgeol collection. the arrangements proposed by fernandes and coimbra (1996) and dias brito et al., (2001) were followed for defining the bauru group’s stratigraphic units. four strategic fossil localities were chosen during the fieldwork based on information from local farmers to facilitate the labgeol paleontological group’s excavations and research; the first two localities (p1 and p 2, figure 1) are located where road 364 is being paved at a great rate of advance. the third locality (p 3, figure 1) is at the crossroads of road 364 with mgt 484, mgt, leading to gurinhatã, mg; 364 continues towards the states of goiás and mato grosso (it is also called the são paulo – cuiabá road). the fourth locality (p 4, figure 1) is at the margin of road 364 on a farm, where fossil remains of a crocodile were taken to the universidade de são paulo (usp) in ribeirão preto in early 2008. excavation materials, such as geologic hammers, tips, spatulas and brushes, were used at these locations. paper boxes and plastic bags were used for the fossil remains to preserve specimens; another field method involved using hydrochloric acid on the rocks to test whether calcite existed and whether there could have been a fossil in the rock in the case of reaction. a pen was also used to record the collections selected and define gps coordinates in the camp book. geological setting barcelos (1984), described the adamantina formation near the city of uberlândia in triângulo mineiro and goiás state. the first outcrops appear in the city of monte alegre de minas and continue in the direction of the city of caçu, goiás state, always beingon top of hills at around 50 to 100 m above the base formed by the sandstone in this formation. between monte alegre de minas and são simão (goiás state), passing through ituiutaba, barcelos (1984) described the adamantina formation sediments as variegated bimodal coarse grained and relatively well-rounded sandstone,with a sparse matrix of fine sandy silt and clay. there are also carbonate nodules and layers. the sandstone has a massive aspect but it is rich in sedimentary structures; these are represented by layers having sandy communication regarding the geology and vertebrate palaeontology of the adamantina formation (bauru group, upper cretaceous) in campina verde, minas gerais state, brazil 40 figure1: geographic map highlighting the locations of study. prepared by oliveira, e. f.; rangel, c. c. stratification across small, deformed and broken layers of sandstone and coarser sandstone carbonate pseudo-nodules. sediments dominate towards the extreme west of the triângulo mineiro, beyond the area around veríssimo which is in close contact with the uberaba formation, extending towards prata, monte alegre de minas, campina verde, iturama and santa vitória. the adamantina formation represents the bauru group’s greatest extension, having a catchment area greater than that of the other formations (uberaba and marília); the adamantina formation covers a large area of the triângulo mineiro. according to fernandes and coimbra (1996), the adamantina formation consists of a combination of very fine-grained sandstone and fine pink to brown silty mudstones, with brown intercalation; this is usual in massive stratification or plane parallel bedding, alternating with medium and small cross-bedding. the unit often displays other hydrodynamic structures such as ripple marks, cross-lamination thrusting, features cutting and filling gaps with intraformal conglomerates with argillaceous silt intraclast grading features at the top. the adamantina formation consists of thin and very thin sand and silty mudstone beds with cross stratification and clay lenses (interbeds). these beds range from massive to plane-parallel strata alternating with cross-stratification layers containing small to medium pebbles (fernandes and coimbra, 1996). very red sediments have been found in the adamantina formation in the campina verde region but there are also dark grey fluvial sediments. concerning sedimentary structures, the adamantina formation has horizontal stratification with fine sediment, coarse sandstone and conglomerate inter beds, with layers of clay and pebbles (figure 2). the adamantina formation’s rocks have not changed much since deposition, especially their reddish, sometimes brown tones, so it is easy to differentiate them from other kinds of rocks. extensive bioturbation in this formation demonstrates that this region had rich invertebrate fauna living in extensive communities. another factor making campina verde an important site for paleontological research is the frequent discovery of fossil remains, such as a recently discovered reptilian, a predecessor of a crocodyliform (candeiro et al., 2010). vertebrate palaeontology palaeontontological fieldwork carried out near campina verde in 2009 resulted in the discovery of isolated elements representing several invertebrate 41 emerson ferreira de oliveira, caio césar rangel, diego sullivan de jesus alves, filipi silva limonta, adelino carvalho, carlos roberto a. candeiro figure 2: geological map of the triângulo mineiro region, minas gerais state, brazil (modified from fernandes and coimbra, 1996) fossil and vertebrate taxa (figure 3). although generic level identification is impossible, due to the specimens’ fragmentary nature, the faunal list includes crocodyliforms (crocodylomorpha indet.), dinosaurs (sauropoda, theropoda indet., abelisauroidea), other vertebrates (vertebrata indet.) and possible molluscs (gastropoda indet.). the theropods are represented by abelisauroidea teeth which have labial and lingual faces which are more convex in cross-section near the anterior edge and flatter posteriorly; the lingual face is slightly concave proximodistally and the cross-sections are exactly as in the abelisauroidea. crocodilian eggs, previously unreported from campina verde, are represented by several elements. numerous isolated mesoeucrocodilian teeth and bones indicate the presence of mesoeucrocodylia indet. adamantina formation palaeontology in the triângulo mineiro region is rich in fossil materials and marks by cretaceous period animals. the adamantina formation has a remarkable faunal record of essentially vertebrate taxa; dinosaur remains are the most representative ones recorded to date (candeiro, , 2010). according to dias-brito et al., (2001) the adamantina formation’s fossil sediments are primarily represented by vertebrates (remains of crocodylomorphs, dinosaurs, turtles, frogs, fish, snakes), plants (charophyta) and molluscs (conchostraceans and ostracods). the adamantina and marília formations account for most fossil records related to the bauru group and they are where dinosaur fossil records are most extensive for this period in brazil. saurischian dinosaur records in the states of goiás, mato grosso, minas gerais and são paulo include sauropods (long-necked herbivorous dinosaurs) and theropods (carnivorous dinosaurs). this region, located in extreme west of mg, represents a promising area and certainly much more research can still be carried out in the campina verde region and, with the collaboration of the labgeol palaeontological group’s services contracted by the municipality, a new point of palaeontological or archaeological research can be created in the triângulo mineiro (i.e. another palaeontologyarea which has been studied since 1940 is already situated in the uberaba region, called peirópolis). concluding remarks the bauru group is composed of sedimentary rocks containing abundant fossils; these rocks differin their modes of origin there by allowing the group to be divided into three formations: the adamantina, marília and uberaba. each of these units has its own peculiarities and characteristic colour and all have been deposited by water or wind. the adamantina formation has reddish coloured sandstone so it is easy to differentiate these rocks from other units; however, colour cannot always be assigned as a method for differentiating formations, as red-coloured rocks can also be founding the marília formation. other classification is required, such as grain-size thickness. calcite was observed in the rocks in the sediments at the study sites and they had bioturbation marks on them. the fossil fragments found were difficult to identify because they were damaged; however this is a good indication that the campina verde area promises to produce important new paleontological localities. this evidence already shows on going activity in that particular environment there by increasing interest in new, more detailed studies to improve scientific results achieved to date. although there are reports regarding some fossils discovered in campina verde, studies of the region are still few; new and more detailed research is necessary to allow better descriptions of geological and paleontological aspects. acknowledgments we would like to thank prof. gerli carrotti (new york) for reviewing earlier drafts of the manuscript and its grammar. the manuscript has been substantially improved after taking the comments made by two anonymous referee into account. this work was part of an undergraduate research project supported by a fellow ship from conselho nacional de desenvolvimento científico e tecnológico (cnpq) awarded to e. oliveira and produtividade em pesquisa/cnpq to r. candeiro. references barboza, e. m., maizatto, j. r. and castro p. t. a. (1993).evidências de tectonismo ativo na sedimentação de rochas do grupo bauru, cretáceo da bacia do paraná, no triângulo mineiro, mg, acta geológica leopoldensia. 16, n°. 38, 97-112. barcelos, j. u. (1993). geologia regional e estratigrafia cretácica do triângulomineiro, revista sociedade e natureza. 5, 9-24. barcelos, j. u. (1984). reconstrução paleogeográfica da sedimentação do grupo bauru baseadanasuaredefiniçãoestratigráficaparcialemterritóriopaulista e no estudo preliminar fora do estado de são paulo, thesis of doctor, instituto de geociências e ciênciasexatas, universidade federal paulista, rio claro, brasil. brito, i. m. (2001).geologia histórica, uberlândia edufu, 413. candeiro, c. r. a. (2007). cretaceous biota of the triângulo mineiro region (brazil): a review of recent finds, estudos geológicos. 63, nº. 1, 65-73. candeiro, c. r. a.(2010) record of the late cretaceous south american genus aeolosaurus (sauropoda, titanosauria): paleogeographical implications. estudios geológicos (madri), 66 243-253. candeiro, c. r. a., abranches, c. t., abrantes, e. a., avilla, l. s., martins, v. c., moreira, a. l., torres, s. r. and bergqvist, l. p. (2004). dinosaurs remains from western são paulo state, brazil (bauru basin, adamantina formation, upper cretaceus), journal of south american earth sciences. 18, 01-10. communication regarding the geology and vertebrate palaeontology of the adamantina formation (bauru group, upper cretaceous) in campina verde, minas gerais state, brazil 42 figure 3: late cretaceous fossils of adamantina formation a – tooth abelisauroidea; b – vertebra of the crocodilomorpha; c, indeterminate bone of crocodilomorpha. candeiro, c. r. a., carvalho, a. a., oliveira, e. f., pereira, c. t., alves, d. s. j., rangel, c. c., alves, e. j. f., alves, m. c., limonta, f. s., ramalho, f. l. and almeida, l. h. c. (2010). new records of vertebrate fossils in the adamantina formation (bauru basin, upper cretaceous) at campina verde, state of minas gerais, brazil, 10th congreso argentino de paleontología y bioestratigrafía 7th congreso latinoamericano de paleontologia, la plata, argentina, septiembre, 142-143. candeiro, c. r. a., marinho, t. s. and oliveira, e. s. (2004). distribuição geográfica dos dinossauros da bacia bauru (cretáceo superior), revista sociedade e natureza.16, nº. 30, 33-55. dias-brito, d., musacchio, e. a., castro,j. c., maranhão, m. s. a. s., suárez, j. m. and rodrigues, r. (2001). grupo bauru: uma unidade continental do cretáceo no brasil – concepções baseadas em dados micropaleontológicos, isotrópicos e estratigráficos, revue paleobio. 20, nº. 1, 245 – 304. etchebehere, m. l c., saad, a. r. and resende, a. c. (1993). reavaliação do potencial do grupo bauru para evaporitos e salmoras continentais, geociências. 12, nº. 2, 333-352. fernandes, l. a. and coimbra, a. m. (1996). a bacia bauru (cretáceo superior brasil), anais da academia brasileira de ciências. 68, nº. 2, 195-205. fernandes, l. m. and coimbra, a. m. (1996). estratigrafia y ambientes deposicionales de la cuenca bauru (cretáceo superior, brasil), acta geológica hispânica. 30, nº. 4, 11-30. oliveira, l. a. and campos, j. e. g. (2003). seqüência conglomerática do membro araguari – grupo bauru – norte do triângulo mineiro, geociências. 22, nº. 1, 43-51. 43 emerson ferreira de oliveira, caio césar rangel, diego sullivan de jesus alves, filipi silva limonta, adelino carvalho, carlos roberto a. candeiro issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n4.91059 earth sciences research journal earth sci. res. j. vol. 26, no. 1 (march, 2022): 13 22 g e o p h y si c s an experimental test for detecting effective reflector height with gps snr data nursu tunalioglu, cemali altuntas yildiz technical university, turkey corresponding author: ntunali@yildiz.edu.tr how to cite item: tunalioglu, n., & altuntas, c. (2022). an experimental test for detecting effective reflector height with gps snr data. earth sciences research journal, 26(1), 13-22. https://doi.org/10.15446/esrj.v26n1.95405 record manuscript received: 05/05/2020 accepted for publication: 30/03/2022 a b st r act: this study aims to estimate effective reflector heights and height differences using the basic geometrical principle of multipath theory by controlling the signal quality for estimations. the geometry of the reflecting signal allows computing the effective reflector height, which is extracted from where the signal reflects on the ground and arrives at the gps antenna phase center. to achieve that, an experimental case with two stations was conducted in the snow-free environment and gps receivers were mounted on reflectors, which allowed to measure daily in-situ reflector heights and artificial decrement variations. the reflections from the roof surface were tracked with the first-fresnel zones. to validate the estimated reflector heights in a controlled test environment, twelve different combinations within four simulated scenarios as a combination of decrement values have been implemented and accuracy analysis was performed. here, a vertical shift procedure on reflectors was applied. meanwhile, the vertical shift amount was tracked in each computation to determine, which reflected signal could be able to use for assigning reflector height as effective. comparisons of the estimated heights and in-situ measurements show congruency with ±1.2 cm to ±8 cm accuracy. the best overall accuracy of the model among the four scenarios is computed as ±2.2 cm. when the vertical shift decrements are considered, the rmse values are estimated within ±2.92 cm to ±3.96 cm. although the rmses of the differences show a good agreement with estimated reflector heights, it is found that some reflector height estimations are statistically insignificant. keywords: gps interferometric reflectometry (gps-ir); signal to noise ratio (snr); multipath theory; effective reflector height. ensayo experimental para detectar la altura de reflexión efectiva con información de la relación señal/ruido en gps r esu m en: este estudio se enfoca en calcular la altura de reflexión efectiva y las diferencias de altura a través de principios geométricos de la teoría de multitrayectos al controlar la calidad de la señal para realizar los cálculos. la geometría de la señal reflejada permite computar la altura de reflexión efectiva, la cual se toma entre la señal que se refleja en el piso y la que llega a la antena gps. para lograr esto se realizó un ensayo experimental con dos estaciones en un ambiente sin nieve y con los receptores gps instalados en los puntos de reflexión, lo que permite medir diariamente las alturas reflejadas in-situ y las variaciones en la reducción artificial. las reflexiones tomadas en la superficie del techo se monitorizaron con las primeras zonas de fresnel. para validar las alturas de reflexión estimadas en un ambiente controlado se implementaron doce combinaciones en cuatro escenarios simulados como valores de reducción y se realizaron análisis de exactitud. en este punto, se aplicó un procedimiento de migración vertical en los puntos de reflexión. además, la dimensión de la migración vertical se monitorizó en cada cómputo para determinar cual señal reflejada es efectiva para ser asignada como altura de reflexión. la comparación de las alturas estimadas y las medidas in-situ muestra congruencia con un rango de exactitud de ±1.2 cm a ±8. el mejor promedio de exactitud del modelo entre los cuatro escenarios se computó con un margen de ±2.2 cm. cuando se considera la reducción en la migración vertical, los valores rmse se estiman en el rango de ±2.92 cm a ±3.96 cm. a pesar de las diferencias de los valores rmse muestran coincidencia con la altura de reflexión estimada se encontró que algunas estimaciones de alturas reflejadas son estadísticamente insignificantes. palabras clave: interferometría y reflectometría de gps; relación señal/ruido; teoría multitrayectos; altura de reflexión efectiva. https://doi.org/10.15446/esrj.v25n1.74167 https://doi.org/10.15446/esrj.v26n1.92419 14 nursu tunalioglu and cemali altuntas introduction today, global navigation satellite systems (gnss) play a major role to extract high-accurate and four-dimensional (4d; x, y, z+t) geo-referenced positioning information. among these systems, global positioning system (gps), owned and operated by united states (us) has still dominant specifications by ongoing modernization facilities. until now, gps has been used for a wide-range geodetic/geomatics studies, such as deformation monitoring, precise navigation, geodetic networks, cadastral applications, hydrographic surveys, geographical information systems (gis), photogrammetric studies based on kinematic gps, terrestrial-mobile-airborne lidar applications, remote sensing, unmanned aerial vehicle (uav) systems, early warning system configurations for detecting natural disasters such as earthquakes, landslides, meteorological studies, which require high precision positioning information (ocalan et al., 2016). as commonly known, gps enables to estimate threedimensional precise positioning, navigation and timing facilities (hofmannwellenhof et al., 2008; gao, 2018; erdogan et al., 2018; zhou et al., 2019). the improvement in accurate positioning is an ongoing issue that mostly deals with modeling the error sources. besides that, the researches about environmental issues have become more significant than before because of the climate change that affects the whole world. especially accurate estimations of the underground water level, snow package volumes, tracking the vegetation grow, soil moisture have become crucial issues. recently, many scientific researches have been conducted to detect the contributions of these items to the climate system with a novel methodology called, gps interferometric reflectometry (gps-ir) method, which provides significant and effective outcomes for these studies (ozeki & heki, 2011; gutmann et al., 2012; larson & nievinski, 2013; larson et al., 2009a; chew et al., 2016; larson et al., 2009b; roussel et al., 2015; roussel et al., 2016; nievinski & larson, 2014a). precise point positioning in gps needs to eliminate or model the error sources. although one of the major error sources so called multipath, affects the positioning accuracy, it has become an effective tool to extract environmental features. signals reflected from the surface following one or more paths near a gps receiver interfere with the direct signals while a composite signal is formed as the sum of them. the direct and reflected signals can simultaneously be recorded at the antenna of the gps receiver. this interference creates significant oscillations, especially at low elevation angles. if the direct component of the signal is removed, the remaining part will be the reflected component of the signal. the initial procedure of using reflected signals instead of direct signals was introduced by martin-neira (1993) to measure the ocean surface (larson & small, 2016). besides, many studies have been conducted to retrieve surface specifications among the reflected signals. larson et al. (2009a) show that variations in snow depth estimated by gps-ir method agree very well with the in-situ measurements, which means that the method could be used successfully and high-correlated with the in-situ measurements for retrieving the effective reflector height (mccreight et al., 2014; nievinski & larson, 2014a; nievinski & larson, 2014b). ozeki and heki (2011) represent a geometry-free linear combination for the feasibility of analyzing multipath pattern in gps data series and demonstrate that it could be also used for snow depth retrieving facilities. gutmann et al. (2012) indicated the necessity of determination of the effective reflector height when the snow depth computation is indispensable for climatological studies in hydrological ecosystem and conducted a detailed study with nine-month period including winter season following gps-ir methodology by validating the results both laser scanning data and in-situ measurements from snow poles, and found accurate results within 9 cm to 13 cm interval. chen et al. (2014) used cost-effective dipole antenna instead of the geodetic antenna by analyzing l2c signals collected and showed an improvement in results when compared with the geodetic receiver for snow depth estimations. hefty and gerhátová (2014) examined snr data of l1 and l2 carrier phases and the geometry-free linear combination of carrier phases, l4 for two permanent gps stations data. the validation was performed with the regular manual snow depth measurements and it was demonstrated that better than 5 cm accuracy was mostly reached when comparing with the estimated reflector heights. bilich et al. (2011) define the ground specifications of a good gps-ir station as flat or slightly titled surface as lower than 5° ground slope, and previous studies show that implementing gps-ir method to the horizontal, planar surface performs good results. however, a new strategy proposed by wei et al. (2019), so-called tilted surface strategy (tss) exposed the slope effect of the topography to retrieve the snow depth and they concluded that following the tss for receivers located on slopes works better. in this paper, first a brief description of the multipath theory and the processing of snr data following larson and nievinski (2013) are introduced. then, the experimental site performed for accuracy assessment of the model by vertical shifting is applied and the results are presented. a snow-free environmental is demonstrated to retrieve the effective reflector height for the experimental test to mitigate the penetration effect of the gps signals into the surface and the estimations with the rmses of vertical shifting values are computed. multipath effect and snr observations hofmann-wellenhof et al. (2008) give the basic description of the multipath effect on carrier phases as receiving the satellite-emitted signal at the receiver by more than one path, caused by reflecting surfaces near the receiver in particular (figure 1). recently, multipath, as a major error source for gnss precise point positioning, has become a key element for retrieving information from reflecting surface. even though the antenna configuration of the gps is designed to suppress this error source, that can not be removed totally (larson et al., 2009b). the reflected signal contribution can be seen at low elevation angles, i.e. lower than 30°, due to antenna gain pattern (larson & small, 2016). several studies have been exposed to the use of multipath data for estimating the parameters of reflecting surface. besides that, the geometry of the reflecting signal allows computing the effective reflector height, which is extracted from where the signal reflects on the ground and arrives at the gps antenna phase center (apc). figure 1. multipath effect due to reflecting surface and converted vertical height from reflecting surface the multipath effect means that the satellite signals don’t reach the antenna center directly in one way, but some are reflected once or several times in advance on objects in the receiving environment or on the ground. the multipath effect is occurred both from the relation of the geometry of the reflecting surface and geodetic receiver antenna and the dielectric constant of the reflecting surface (larson et al., 2009b). the reflected signals have always a longer transit time than the direct signal. that leads to an extended pseudorange determination between the antennas on the ground and the satellite. since the phase of a wave is a function of the transit time, a longer transit time also affects the carrier phase measurement. because objects on the surface of the earth mainly affect satellite signals with low elevations, signals from low-flying satellites are more affected by multipath effects than signals sent by satellites with high elevation. the amount of the error, which is caused by multipath effects, depends on the multipath length (the resulting delay of the signal travel time), the phase difference between direct and reflected waves and their power ratio. the achievable accuracy of positioning using gnss measurements depends on the detection of the stations (variations of the antenna phase center, multipath effect, receiver noise, receiver clock and receiver signal errors) and satellite-specific factors (satellite geometry, clock and orbit errors of the satellites, including gnss code and system bias, phase wind-up) the behavior of the signals of the gnss satellites, which acts as high-flying fixed points at 15an experimental test for detecting effective reflector height with gps snr data an average altitude of 20200 km, is influenced during the propagation in the various layers of the atmosphere (delay in the signals in the ionosphere and troposphere). most of the error components that affect the accuracy of gps are largely eliminated by the differentiation method by data processing. a good example of such an approach is rtk (real-time kinematik) and networkrtk, where error components of gnss are calculated using data from one (rtk) or several reference stations (network-rtk) in the surroundings of the rover and by a data transmission medium either as correction of coordinates or as values for corrections from the rover received satellite data made available in real-time. the rover uses these corrections to increase the accuracy of the positioning, this requires a service for the calculation and transmission of the corrections to the rover as well as for the permanent maintenance of reference stations within the network. from the mid-1990s, cors (continuously operated reference station) networks were established in europe and in certain regions of the world to provide the rtk and network rtk services. multipath effect, the cause of which lies only in the immediate surroundings of the antenna, cannot be therefore eliminated using the differential method, because multipath effects are uncorrelated even from a distance of a few meters between two antennas. the computation of multipath effects at reference stations are therefore very important with respect to the precise positioning using gnss data, since a coordinate error at reference station also influences the determination of the correction data and thus falsifies the calculated coordinates of the rover. therefore, the cors of a network must be free of the multipath effect to reach the high quality for positioning. since any general model can be extracted for the multipath effect, the influence can be detected by code and carrier phase measurements (hofmannwellenhof et al., 2008). however, the signal power of the composite signal formed by interfering with the reflected and direct signals can be affected as well. multipath effects are also visible in the signal to noise ratio (snr) data constantly collected by gnss receivers (roussel et al., 2015). signal strengths are also stored in gps receivers beside carrier phases. the signal strength corresponds to the carrier-to-noise density ratio (c/no), which is a ratio of signal power to noise power density (larson & nievinski, 2013). the snr, computed from this ratio, is usually used for determination of signal quality and noise characteristics of classical gps observations (qian & jin, 2016; tabibi, geremia-nievinski, & van dam, 2017). since the snr data are in the logarithmic unit in db, they are converted to a linear scale in watt per watt (volt per volt) by snr ( (voltsvolts =10snr(db-hz/20), assuming a 1-hz bandwidth (larson & nievinski, 2013). the snr value recorded at a receiver is assumed to be equal to the amplitude of the received signal, then the equation equals the same quantity between snr and amplitude of the composite signal. during the satellite movements across the sky, the phase difference between direct and reflected signals is formed, which is also represented as an interference pattern (löfgren et al., 2014; drosinos, 2016). the interference between these signals is directly related to the satellite elevation angle. since, the satellite elevation angle increases, the multipath effect on snr data decreases (larson et al., 2008). as the signal transmitted from a satellite received at the antenna as a composite signal, interfere of direct and indirect signals, the relation between the signal to noise ratio can be expressed as equation 1 (larson et al., 2008; larson & nievinski, 2013; nievinski & larson, 2014a); snr2≈a2d +a 2 m +2ad amcosψ (1) here, ad and am are the amplitudes of direct and reflected signals (in volt unit), respectively and is the phase difference (in radian unit) between these signals. assuming ad>> am, the magnitude of the snr is mainly caused by direct signal, a quadratic polynomial function can be used to remove the dominant part effect (larson & small, 2016). after removing this part, the remaining snr value, as commonly expressed by detrended snr (dsnr), can be described by equation 2; dsnr=a cos(4πhλ-1 sinε+ψ) (2) where a is the amplitude, ψ is the phase offset, ε is the satellite elevation angle, λ is the carrier wavelength, h is the distance between the antenna phase center (apc) and the reflecting surface (generally called as reflector height). as the multipath pattern can be expressed as a function of the sine of the elevation angle, that causes an unevenly spaced sampling intervals, the dominant frequency of the dsnr will be constant and equal to the 2h⁄λ. once the dominant frequency of the dsnr values is extracted, it can be converted to the reflector height, which is the distance between the apc and reflecting surface. lomb scargle periodogram (lsp) is used to estimate the dominant frequency (larson & small, 2016). since lsp spectrum analysis represents the power of the frequencies for the snr series, the snr data used are grouped into varied elevation angle ranges including noisy frequencies to control the effective reflector heights estimated with in-situ reflector measurements. materials and methods study area the study aims to control the snr data fluctuations on retrieving effective reflector heights by checking the vertical shift amount from them. for that, two stations, so-called elk1 and elk2 were established on the roof of the faculty of electrical and electronics engineering at the davutpaşa campus of yıldız technical university in istanbul. the two-dimensional metrics of the roof are approximately 12 m and 75 m. the roof surface can be accepted as quite flat avoiding the surface roughness, with the exception of the slightly elevated ground as wall border near the elk1 station. also, there are no high buildings near the roof that could obstruct the open sky view and make disturbing reflections. the representation of the locations of the stations on the roof is given in figure 2 and the geographical coordinates are given in table 1. figure 2. study environment (a) locations of the stations on the roof (b) elk2 station-south direction (c) elk1 station-east direction table 1. geographical coordinates of the stations (wgs84) station φ(°) λ(°) h (m) elk1 41° 01’ 43.345647’’ 28° 53’ 24.047295’’ 130.2373 elk2 41° 01’ 44.521411’’ 28° 53’ 24.973158’’ 129.9659 data acquisition gps observations were performed in an experimental field site with two observation stages. the first one was carried out on day of years (doy) 203, 204, 205, 206, 207 and 210; and the latter on doy 211, 212, 213, 214, 215 and 217. the starting and ending times and dates for observations are given in table 2. the receivers used in the first part were topcon hiper pro at elk1 and topcon gr-5 at elk2, and in the second part were chc i70 at elk1 and chc i80 at elk2, respectively. moreover, the technical specifications of the used receivers are given in table 3. table 3 also covers the snr evaluations of the collected data that are done for l1 carrier frequencies. 16 nursu tunalioglu and cemali altuntas table 2. detailed information about the gps observations performed on sites at 1 hz rate date doy site receiver start (utc +3h) end (utc +3h) 22.07.2019 22.07.2019 203 203 elk1 elk2 topcon hiper pro topcon gr-5 10:05 10:16 16:34 16:35 23.07.2019 23.07.2019 204 204 elk1 elk2 topcon hiper pro topcon gr-5 09:16 09:20 16:30 16:32 24.07.2019 24.07.2019 205 205 elk1 elk2 topcon hiper pro topcon gr-5 09:22 09:24 16:37 16:37 25.07.2019 25.07.2019 206 206 elk1 elk2 topcon hiper pro topcon gr-5 09:07 09:09 17:06 17:06 26.07.2019 26.07.2019 207 207 elk1 elk2 topcon hiper pro topcon gr-5 10:09 10:13 16:35 16:35 29.07.2019 29.07.2019 210 210 elk1 elk2 topcon hiper pro topcon gr-5 09:18 09:21 16:37 16:40 30.07.2019 30.07.2019 211 211 elk1 elk2 chc i70 chc i80 09:25 09:30 16:32 16:33 31.07.2019 31.07.2019 212 212 elk1 elk2 chc i70 chc i80 09:23 09:26 16:32 16:33 01.08.2019 01.08.2019 213 213 elk1 elk2 chc i70 chc i80 09:22 09:24 16:34 16:35 02.08.2019 02.08.2019 214 214 elk1 elk2 chc i70 chc i80 09:21 09:24 16:32 16:33 03.08.2019 03.08.2019 215 215 elk1 elk2 chc i70 chc i80 09:29 09:31 16:57 16:58 05.08.2019 05.08.2019 217 217 elk1 elk2 chc i70 chc i80 09:37 09:39 16:39 16:40 table 3. technical specifications of gnss receivers and snr data collected (ε, satellite elevation angle) topcon hiper pro topcon gr-5 chc i70 chc i80 satellite systems gps, glo gps, glo gps, glo, gal, bds, qzss gps, glo, gal, bds, qzss, navic satellite signals l1, l2 l1, l2 l1, l2, l5 l1, l2, l5 l1 min-max-mean snr (ε>25) 40.0-52.0-47.4 db 42.0-51.0-47.8 db 40.3-51.5-47.3 db 34.2-51.3-47.3 db l1 min-max-mean snr (ε<25) 33.0-46.0-41.3 db 31.0-47.0-41.9 db 26.6-46.2-40.5 db 26.6-45.5-40.3 db l1 min-max-mean snr (overall) 33.0-52.0-43.7 db 31.0-51.0-44.6 db 26.6-51.5-43.3 db 26.6-51.3-43.2 db gps data were collected with a sampling interval of 1 sec. to detect the effective reflector heights on the site using gps snr data, a decrement as 10 cm assuming a vertical displacement occurred as an effect of changing the vertical distance was applied to both reflectors established on the stations. to do that, the observations were planned first to start from 2 m for the elk1 station and 1.70 m for the elk2 station, and then the reflector height was reduced by 10 cm in every two days by collecting daily data for at least 6 hours. the decrement is applied to simulate snow depth variations. in this study, only gps observations were used. the data collected during the site surveys are split as ascending and descending according to variations of the satellite elevation angle. all the data were recorded by the gps receivers, but the data for low elevation angles (5°-25°), where the major multipath effect was intense, were analyzed. in the analysis, observations, where the difference between the minimum and maximum elevation angles of each observed satellite during any satellite arc track is less than 10° are not taken into account. this 10° condition benefits from getting enough epoch numbers for each track. in the evaluations, the azimuth angle range of 30°-210°, where the roof surface is flat, was selected, and data from other directions were excluded due to the obstructs on the roof. here, the azimuth angles except the selected range were not proper to analyze while considering the direction and width of the sensed areas on the roof. the sensed area, i.e., first fresnel zone (ffz), varies depending on the change of satellite elevation angle and reflector height (larson & nievinski, 2013). due to that different reflector heights have been used to track the variation of the sensed area from ffzs. figure 3 shows the ffzs for observations to the satellites at elevation angles 5° and 10° for reflector heights of 1.70 and 2 m in doy 203. only snr values of l1 signals were analyzed and used in this study. the statement given in larson and nievinski (2013) is used to identify the signals as a strong reflection, where the peak value on data series is higher than four times of background noise, for each ascending and descending arcs of the satellite tracks. 17an experimental test for detecting effective reflector height with gps snr data figure 3. ffzs at elk1 and elk2 in doy:203 (a) for 5° elevation angle, 2.00 m reflector height (b) for 10° elevation angle, 2.00 m reflector height (c) for 5° elevation angle, 1.70 m reflector height (d) for 10° elevation angle, 1.70 m reflector height. red ellipses are within the desired azimuth range while blue ellipses are outside that range. the radii of the nested circles indicating the distance from the point were drawn in increments of 10 m. the signal that does not meet this condition is called weak reflection in this study. in this way, it has been found that 139 of the 371 signals have a strong reflection. maximum critical reflector height converted from the lsp was taken as 10 m, which was important to estimate the strong reflection statement that can change the arithmetical mean value of ai.. 18 nursu tunalioglu and cemali altuntas figure 4. four examples of dsnr values in doy:211 (a) strong reflection at elk1 (b) weak reflection at elk1 (c) strong reflection at elk2 (d) weak reflection at elk2. dashed lines are polynomial fits to the snr data. the black lines show the snr data, the blue dsnr in linear scale, the orange snr in logarithmic scale and an example for (e) total c/no for both direct and multipath signals with satellite elevation angle 19an experimental test for detecting effective reflector height with gps snr data results and discussion the examples of weak and strong reflections of dsnr values obtained from the analyses are represented in figure 4. the examples of strong reflections shown in figures 4a and 4c present sinusoidal trend for oscillations especially at low elevation angles. since the multipath effect decreases with increasing the satellite elevation angle, the oscillations in the resulting signal become smaller. thus, the oscillations become stable when the elevation angle increase. for the weak reflection examples, no strong periodic oscillation could be tracked (figure 4b, 4d). moreover, although the reflections between 5°-15° elevation angles did not show periodic oscillations, this situation has changed when the elevation angle interval is drawn for 15°-20° (see, figure 4b and 4d). this can be caused by the location of the reflected signal that falls to the roof surface when the elevation angle increases. the lsp was used to convert the snr data to the effective reflector heights. spectral amplitudes of each gps track computed by lsp were evaluated by following larson and nievinski (2013). then, the dominant frequencies are converted into effective reflector heights. figure 5 shows the 6-day amplitudereflector height values of strong signals for elk1 and elk2 stations. figure 5. some of the strong reflections for 6 days (a) chc i70 at elk1 (b) chc i80 at elk2 the correlation coefficients (ρ) between estimations and in-situ measurements for reflector heights are shown in figure 6. moreover, the detailed information about the correlation coefficients calculated from estimated reflector heights and in-situ measurements calculated separately from receivers is given in table 4. the highest correlation coefficient is computed at elk2 station as 0.9346 for 46 estimations. although the maximum estimation number, 49 is obtained at elk2, the correlation coefficient is calculated as 0.8587, which is the lowest agreement with the in-situ measurements. this could be caused by the gps instrument used (see, table 3). figure 6. correlation coefficients (a) topcon hiper pro at elk1 (b) chc i70 at elk1 (c) topcon gr-5 at elk2 (d) chc i80 at elk2 table 4. number of estimations and correlation coefficients for receivers receiver number of estimations correlation coefficient topcon hiper pro (at elk1) 23 0.8798 topcon gr-5 (at elk2) 49 0.8587 chc i70 (at elk1) 21 0.8786 chc i80 (at elk2) 46 0.9346 the daily mean values of estimated reflector heights and root mean square errors (rmse) with respect to doys are given in table 5. 20 nursu tunalioglu and cemali altuntas table 5. daily mean values of estimations and rmses elk1 topcon hiper pro chc i70 doy num. of est. in situ (m) estimated height (m) rmse (cm) doy num. of est. in situ (m) estimated height (m) rmse (cm) 203 3 2.000 1.991 1.2 211 1 2.000 1.976 2.4 204 4 2.000 2.010 3.6 212 4 2.000 2.004 1.4 205 4 1.900 1.920 4.2 213 4 1.900 1.904 3.6 206 5 1.900 1.942 6.1 214 3 1.900 1.920 3.8 207 4 1.800 1.843 5.1 215 5 1.800 1.846 5.7 210 3 1.800 1.817 3.3 217 4 1.800 1.839 5.7 elk2 topcon gr-5 chc i80 doy num. of est. in situ (m) estimated height (m) rmse (cm) doy num. of est. in situ (m) estimated height (m) rmse (cm) 203 7 1.700 1.762 8.0 211 5 1.700 1.717 3.0 204 8 1.700 1.738 6.0 212 6 1.700 1.728 4.0 205 10 1.600 1.627 4.8 213 8 1.600 1.576 3.3 206 10 1.600 1.624 6.7 214 7 1.600 1.572 3.6 207 7 1.500 1.522 6.2 215 10 1.500 1.488 4.1 210 7 1.500 1.499 6.5 217 10 1.500 1.479 4.8 here, the model efficiencies were estimated to figure out how the model fit the site observations considering the location of the points separately. then, the accuracy of the model for elk1 with topcon hiper pro receiver is computed as 2.7 cm, for elk2 with topcon gr-5 receiver is 3.4 cm, for elk1 with chc i70 receiver is 2.8 cm, for elk2 with chc i80, receiver is 2.2 cm. to verify the estimations with the vertical shift amount, assuming as constant in each couple of days, 12 different combinations were created for decrements of 10 cm or 20 cm. the rmses of estimated differences are computed from subtracting the in-situ differences from the differences in the mean of daily estimated heights. table 6 represents the computed rmses of 12 combinations for four simulated scenarios. the overall rmses of elk1 for scenarios 1 and 2 are calculated as ±2.92 cm and ±3.96 cm, respectively. meanwhile, the overall rmses of elk2 for scenarios 3 and 4 are calculated as ±3.13 cm and ±3.68 cm, respectively. these results agree with the rmses of the daily mean values of estimations. however, according to the rmse computations of estimated differences (see table 6, column 5 & 7), the rmse values of some of the vertical shifts (the underlined cell) are greater than vertical shift values. this situation can be interpreted as they are statistically insignificant. moreover, according to the results given in table 5 and 6, there are slight differences between the rmse values of estimations, which are arisen due to usage of different receivers. conclusions in multipath theory, a signal transmitted from a satellite uses more than one path to arrive at the gnss antenna. here, the signal may be reflected from the surface around where the receiver is set. the reflected and direct signals, which interfere with each other at the phase center of the gnss antenna, are recorded simultaneously. the multipath modulation can be observed in gps observations as snr values, which enable to estimate the vertical distance between the reflected surface of the signal and phase center of the gnss antenna. several studies have been conducted to estimate the reflector height to extract environmental variables, such as snow depth, vegetation height, etc. in this study, an experimental site was prepared to extract the effective reflector height with its accuracy assessment. the study involves gps l1 snr data due to provide strong signal strengths than the l2 and/or l5. to achieve varying reflector heights on the site, the reflector height is displaced manually in daily observations to detect artificial reflector height variations. the results indicate that the reflector heights and height differences between days given as 10 cm in each observation day agree with the daily in-situ measurements showing a high correlation with 0.9346 correlation coefficient at elk2 station. here, the results are supported by the in-situ measurements to present the accuracy of the estimations. when the 6-day estimations are considered to figure out how the model fits the site observations, the best accuracy of the model is computed as 2.2 cm for the same station as well. although the reflector height estimation has become a routine and well-known method, in this study the accurate estimation of vertical differences is examined to obtain precisions from the reflector heights. the absolute vertical differences as 10 cm and 20 cm are considered as the true value and rmses of vertical shifting amounts are computed. according to the results, the overall accuracy is obtained between 2.92 cm and 3.96 cm. however, this also shows that some of the reflector height estimations are insignificant. this situation brings a necessity to examine the differences as well. with the help of statistical considerations, more accurate and valid solutions in reflector height retrieving can be provided. acknowledgment the figure 2 was plotted by using generic mapping tools (wessel & smith, 1998). 21an experimental test for detecting effective reflector height with gps snr data table 6. vertical shift validation of estimations with rmses elk1 combinations of doys via scenarios topcon hiper pro chc i70 # (1) (2) in-situ difference (cm) estimated difference with rmse (cm) overall rmse (cm) estimated difference with rmse (cm) overall rmse (cm) 1 203-205 211-213 10.00 7.10 ± 4.37 ± 2.92 7.20 ± 4.33 ± 3.96 2 203-206 211-214 10.00 4.90 ± 6.22 5.60 ± 4.49 3 203-207 211-215 20.00 14.80 ± 5.24 13.00 ± 6.18 4 203-210 211-217 20.00 17.40 ± 3.51 13.70 ± 6.18 5 204-205 212-213 10.00 9.00 ± 5.53 10.00 ± 3.86 6 204-206 212-214 10.00 6.80 ± 7.08 8.40 ± 4.05 7 204-207 212-215 20.00 16.70 ± 6.24 15.80 ± 5.87 8 204-210 212-217 20.00 19.30 ± 4.88 16.50 ± 5.87 9 205-207 213-215 10.00 7.70 ± 6.61 5.80 ± 6.74 10 205-210 213-217 10.00 10.30 ± 5.34 6.50 ± 6.74 11 206-207 214-215 10.00 9.90 ± 7.95 7.40 ± 6.85 12 206-210 214-217 10.00 12.50 ± 6.94 8.10 ± 6.85 elk2 combinations of doys via scenarios topcon gr-5 chc i80 # (3) (4) in-situ difference (cm) estimated difference with rmse (cm) overall rmse (cm) estimated difference with rmse (cm) overall rmse (cm) 1 203-205 211-213 10.00 13.50 ± 9.33 ± 3.13 14.10 ± 4.46 ± 3.68 2 203-206 211-214 10.00 13.80 ± 10.44 14.50 ± 4.69 3 203-207 211-215 20.00 24.00 ± 10.12 22.90 ± 5.08 4 203-210 211-217 20.00 26.30 ± 10.31 23.80 ± 5.66 5 204-205 212-213 10.00 11.10 ± 7.68 15.20 ± 5.19 6 204-206 212-214 10.00 11.40 ± 8.99 15.60 ± 5.38 7 204-207 212-215 20.00 21.60 ± 8.63 24.00 ± 5.73 8 204-210 212-217 20.00 23.90 ± 8.85 24.90 ± 6.25 9 205-207 213-215 10.00 10.50 ± 7.84 8.80±5.26 10 205-210 213-217 10.00 12.80 ± 8.08 9.70±5.82 11 206-207 214-215 10.00 10.20 ± 9.13 8.40±5.46 12 206-210 214-217 10.00 12.50 ± 9.33 9.30±6.00 references bilich, a. l., slater, a.g. & larson, k.m. (2011). snow depth with gps: case study from minnesota 2010-2011. agu 2011 fall meeting. chen, q., won, d., & akos, d. m. (2014). snow depth sensing using the gps l2c signal with a dipole antenna. eurasip journal on advances in signal processing, 2014(1), 106. https://doi.org/10.1186/1687-6180-2014-106. chew, c., small, e. e., & larson, k. m. (2016). an algorithm for soil moisture estimation using gps-interferometric reflectometry for bare and vegetated soil. gps solutions, 20(3), 525-537. https://doi.org/10.1007/ s10291-015-0462-4. drosinos, e. c. (2016). gnss signal reflections off sea ice, master’s thesis in wireless photonics and space engineering, department of earth and space sciences, chalmers university of technology, gothenburg, sweden. erdogan, b., karlitepe, f., ocalan, t., & tunalioglu, n. (2018). performance analysis of real time ppp for transit of mercury.  measurement,  129, 358-367. https://doi.org/10.1016/j.measurement.2018.07.050. gao, y. (2018). precise gnss positioning for mass-market applications, fig general assembly, embracing our smart world where the continents connect: enhancing the geospatial maturity of societies, 6-11 may, istanbul, turkey. 22 nursu tunalioglu and cemali altuntas gutmann, e. d., larson, k. m., williams, m. w., nievinski, f. g., & zavorotny, v. (2012). snow measurement by gps interferometric reflectometry: an evaluation at niwot ridge, colorado. hydrological processes, 26(19), 2951-2961. https://doi.org/10.1002/hyp.8329. hefty, j., & gerhatova, l. u. (2014). using gps multipath for snow depth sensing-first experience with data from permanent stations in slovakia. acta geodynamica et geomaterialia, 11(1), 53-63. https://doi. org/10.13168/agg.2013.0055. hofmann-wellenhof, b., lichtenegger, h., & wasle, e. (2008). gnss–global navigation satellite systems: gps, glonass, galileo, and more. springer science & business media. https://doi.org/10.1007/978-3-211-73017-1 larson, k. m., small, e. e., gutmann, e., bilich, a., axelrad, p., & braun, j. (2008). using gps multipath to measure soil moisture fluctuations: initial results.  gps solutions,  12(3), 173-177. https://doi.org/10.1007/ s10291-007-0076-6. larson, k. m., braun, j. j., small, e. e., zavorotny, v. u., gutmann, e. d., & bilich, a. l. (2009a). gps multipath and its relation to near-surface soil moisture content. ieee journal of selected topics in applied earth observations and remote sensing, 3(1), 91-99. https://doi.org/10.1109/jstars.2009.2033612. larson, k. m., gutmann, e. d., zavorotny, v. u., braun, j. j., williams, m. w., & nievinski, f. g. (2009b). can we measure snow depth with gps receivers?  geophysical research letters,  36(17). https://doi.org/10.1029/ 2009gl039430. larson, k. m., & nievinski, f. g. (2013). gps snow sensing: results from the earthscope plate boundary observatory.  gps solutions,  17(1), 41-52. https://doi.org/10.1007/s10291-012-0259-7. larson, k. m., & small, e. e. (2016). estimation of snow depth using l1 gps signal-to-noise ratio data.  ieee journal of selected topics in applied earth observations and remote sensing,  9(10), 4802-4808. https://doi. org/10.1109/jstars.2015.2508673. löfgren, j. s., haas, r., & scherneck, h. g. (2014). sea level time series and ocean tide analysis from multipath signals at five gps sites in different parts of the world. journal of geodynamics, 80, 66-80. https://doi.org/10.1016/j. jog.2014.02.012. martin-neira, m. (1993). a passive reflectometry and interferometry system (paris): application to ocean altimetry. esa journal, 17(4), 331-355. mccreight, j. l., small, e. e., & larson, k. m. (2014). snow depth, density, and swe estimates derived from gps reflection data: validation in the western us.  water resources research,  50(8), 6892-6909. https://doi. org/10.1002/2014wr015561. nievinski, f. g., & larson, k. m. (2014a). inverse modeling of gps multipath for snow depth estimation—part i: formulation and simulations. ieee transactions on geoscience and remote sensing, 52(10), 6555-6563. https://doi.org/10.1109/tgrs.2013.2297681. nievinski, f. g., & larson, k. m. (2014b). inverse modeling of gps multipath for snow depth estimation—part ii: application and validation. ieee transactions on geoscience and remote sensing, 52(10), 6564-6573. https:// doi.org/10.1109/tgrs.2013.2297688. ocalan, t., erdogan, b., tunalioglu, n. & durdag, u.m. (2016). accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment. earth sci. res. j., 20(4):d1-d9. https://doi.org/10.15446/esrj.v20n4.59496 ozeki, m., & heki, k. (2012). gps snow depth meter with geometry-free linear combinations of carrier phases. journal of geodesy, 86(3), 209-219. https://doi.org/10.1007/s00190-011-0511-x. qian, x., & jin, s. (2016). estimation of snow depth from glonass snr and phase-based multipath reflectometry. ieee journal of selected topics in applied earth observations and remote sensing,  9(10), 4817-4823. https://doi.org/10.1109/jstars.2016.2560763. roussel, n., frappart, f., ramillien, g., darrozes, j., baup, f., & ha, c. (2015). detection of soil moisture content changes by using a single geodetic antenna: the case of an agricultural plot. in  2015 ieee international geoscience and remote sensing symposium (igarss) (pp. 2008-2011). https://doi.org/10.1109/igarss.2015.7326192. roussel, n., frappart, f., ramillien, g., darrozes, j., baup, f., lestarquit, l., & ha, m. c. (2016). detection of soil moisture variations using gps and glonass snr data for elevation angles ranging from 2 to 70. ieee journal of selected topics in applied earth observations and remote sensing, 9(10), 4781-4794. https://doi.org/10.1109/jstars.2016.2537847. tabibi, s., geremia-nievinski, f., & van dam, t. (2017). statistical comparison and combination of gps, glonass, and multi-gnss multipath reflectometry applied to snow depth retrieval.  ieee transactions on geoscience and remote sensing,  55(7), 3773-3785. https://doi.org/10.1109/ tgrs.2017.2679899. wei, h., he, x., feng, y., jin, s., & shen, f. (2019). snow depth estimation on slopes using gps-interferometric reflectometry. sensors, 19-4994. https://doi.org/10.3390/s19224994. wessel, p., & smith, w.h.f. (1998). new improved version of generic mapping tools released, eos trans. agu 79 (47)579. zhou, w., liu, l., huang, l., yao, y., chen, j., & li, s. (2019). a new gps snr-based combination approach for land surface snow depth monitoring.  scientific reports,  9(1), 3814. https://doi.org/10.1038/s41598019-40456-2. http://dx.doi.org/10.15446/esrj.v20n4.59496 _goback keywords: remote sensing technology; china’s northern region; precipitation period; wavelet analysis; change rule; oscillation period. palabras clave: tecnología de teledetección; región del norte de china; período de precipitación; análisis de ondículas; cambio de frecuencias; período de oscilación. how to cite item zhang, y. & liang, c. (2021). application of remote sensing technology in the analysis of periodic precipitation change in china’s northern semi-arid area. earth sciences research journal, 25(1), 85-91. doi: https:// doi.org/10.15446/esrj.v25n1.93925 the remote sensing technology is used to analyze the periodic changes of precipitation in the semi-arid areas of northern china and provide a reference basis for precipitation prediction. authors were based on gps observation data and corresponding ground meteorological parameters to obtain the study area’s comprehensive water vapor content value and use precipitation and time to establish a linear regression equation based on the content value to express the precipitation trend change. the periodic structure and abnormal changes of the periodic precipitation series on different time scales were obtained through wavelet analysis, combined with anomaly calculations to analyze the increase and decrease precipitation characteristics. the morlet continuous wavelet transform map was also used to reflect the periodic oscillations of precipitation on different time scales. the wavelet variance map was used to reflect the distribution of regional precipitation with scale years. the authors analyzed the cycle change characteristics of the comprehensive precipitation series in harbin, changchun, shenyang, and dalian. they concluded that the annual precipitation in the four regions showed a decreasing trend in 50 years. and through the wavelet cycle frequencydomain relationship change law, the change characteristics of precipitation in the northern semi-arid area in the entire time domain were obtained. abstract application of remote sensing technology in the analysis of periodic precipitation change in china’s northern semi-arid area aplicación de la tecnología de teledetección en el análisis del cambio periódico de las precipitaciones en el norte semiárido de china issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.93925 con el fin de proporcionar una base de referencia para la predicción de la precipitación, este estudio utilizó la tecnología de teledetección para analizar los cambios periódicos de la precipitación en las zonas semiáridas del norte de china. se usaron los datos de observación gps y los parámetros meteorológicos terrestres correspondientes para obtener el valor integral del contenido de vapor de agua del área de estudio, y se usó la precipitación y el tiempo para establecer una ecuación de regresión lineal basada en el valor del contenido para expresar el cambio de tendencia de la precipitación en el área. a través del análisis de ondículas se obtuvo la estructura periódica y los cambios anormales de las series periódicas de precipitación en diferentes escalas de tiempo, combinadas con cálculos de anomalías para analizar las características de aumento y disminución de la precipitación. el mapa de transformación de ondas continuas de morlet se utilizó para reflejar las oscilaciones periódicas de la precipitación en diferentes escalas de tiempo. mientras que el mapa de varianza de ondículas se utilizó para reflejar la distribución de la precipitación regional a escala de años. se analizaron las características del cambio de ciclo de la serie integral de precipitación en harbin, changchun, shenyang y dalian, y se concluyó que la precipitación anual en las cuatro regiones mostró una tendencia decreciente en 50 años. y a través de la ley de cambio de la relación del dominio de la frecuencia del ciclo de las ondas, se obtuvieron las características de cambio de la precipitación en el área semiárida del norte en todo el período analizado. resumen record manuscript received: 13/11/2019 accepted for publication: 04/11/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 85-91 yannan zhang, chuan liang* college of water resource and hydropower, sichuan university, chengdu 610065, china *corresponding author: lc3006930@2980.com m e t e o r o l o g y https://doi.org/10.15446/esrj.v25n1.93925 https://doi.org/10.15446/esrj.v25n1.93925 https://doi.org/10.15446/esrj.v25n1.93925 mailto:lc3006930@2980.com 86 yan lv, laijun lu, mengxue cao introduction global warming has led to global and regional climate change, frequent occurrence of high temperature, drought, flood, and other disastrous weather, causing ecological and environmental deterioration, seriously affecting agricultural production, social economy, and sustainable development (bi et al. 2018). especially in the context of global climate change, the spatial pattern of precipitation in china is directly related to agricultural production safety. the northern part of china is located in the northernmost end of the east asian monsoon, which belongs to the temperate continental monsoon climate (guo et al., 2019; huang et al., 2018). it is the transitional zone between the humid eastern monsoon region and china’s arid inland region (colliander et al., 2017). it is hot and rainy in summer, cold and dry in winter, and the continental climate is becoming stronger from east to west. the seasonal change of its environment is closely connected with the general circulation of east asia (hall et al., 2016). the climate and its evolution are pretty different (dai et al., 2018; ren et al., 2019; zeng et al., 2018). it is a typical “vulnerable area” and one of the most sensitive areas affected by climate warming. it is also the largest commodity grain production area and an essential heavy industry and energy base in china (ferrari dos santos et al., 2017). therefore, it is of great significance to study the characteristics of the periodic changes of precipitation in the north of china in the past hundred years for understanding the impact of climate change on rainfall and for short-term, medium-term, and long-term precipitation prediction. in recent years, many scholars have studied these characteristics. lan yongchao et al. studied the impact of climate change on the water resources system in the upper reaches of the yellow river; xu yueqing et al. studied the law of precipitation change in hebei plain based on wavelet analysis; lin zhenyao et al. studied the temporal and spatial characteristics of temperature and precipitation change in the qinghai tibet plateau; dai xingang et al. studied the multi-scale characteristics of precipitation in flood season and the inter-generational decay characteristics of summer monsoon in north china levy. in the past 50 years, under the background of global change, the changing trend of precipitation and the difference in the rule of time scale structure in northern china have attracted much attention (wang et al., 2018). since bevis and businger put forward the basic principle and method for remote sensing of atmospheric precipitable water by gps technology in the 1990s, many experts, scholars, and scientific research institutions in various countries have invested in this research work. the more famous experiments include gps/storm, wwave, clam. (wang et al., 2017). china has also done many experiments in the research of remote sensing of atmospheric precipitable vapor by gps, such as gps/storm experiment in shanghai, gps rainfall observation experiment in south china, gps/met experiment in beijing, and has also achieved good results. at present, the technology of gps remote sensing of atmospheric integrated water vapor content has been mature, and the recognized accuracy is 1-2 mm (zhao et al., 2018). at the same time, because gps remote sensing technology has a series of advantages, such as uniform global distribution, high resolution, low cost, easy maintenance and so on, it fully shows the potential application value of gps technology in meteorological, hydrological, atmospheric monitoring and so on (arrizabalaga & li, 2018; megat & hassan 2018). in order to study the change law of the average precipitation series in northern china in the past 50 years under the background of global change, this paper selects the inland cities harbin, changchun, shenyang and coastal city dalian, which have some differences in response to the regional precipitation change, and analyzes the periodic change law of precipitation in the northern china through the remote sensing technology using the annual precipitation from 1955 to 2005 in four regions (li et al., 2018). in this paper, the annual precipitation data of the above four typical cities are obtained by gps remote sensing technology, and the periodic change of precipitation in the four regions is analyzed by the trend analysis process and wavelet analysis process. material and methods data sources and methods based on the remote sensing data of precipitation time series from 1955 to 2005 in harbin, changchun, shenyang, and dalian stations, the periodic variation of precipitation in the four stations is analyzed. harbin, changchun, and shenyang are located in the northeast of china, the northernmost of the east asian monsoon, which belongs to the temperate continental monsoon climate. dalian is a coastal city, which is affected by the continental monsoon and marine climate. therefore, it is of great significance to analyze the 50 year precipitation law of four stations for understanding the precipitation cycle change law in the semi-arid area of northern china under the background of global change (rauber et al., 2018). retrieval of atmospheric precipitable water in four stations by gps data as the remote sensing technology of atmospheric precipitable water by gps technology is mature, the introduction of basic principles (mora et al., 2017) is omitted here, and the process of remote sensing of atmospheric precipitable water by gps technology is only illustrated in figure 1. according to the flow chart in figure 1, we can use the gps observation data and the corresponding ground meteorological parameters to obtain the final comprehensive water vapor content of harbin and the other three stations in 1955-2005. figure 1. gps processes for remote sensing atmospheric precipitation in the calculation process, the total tropopause delay is calculated by gamit software developed by the massachusetts institute of technology. saastamoinen model is used for tropopause dry delay, and its expression is as follows: bcm l m d n n = ±( ) ( ) 2 2668 0 0023. . ,µ (1) where, m d dn nµ µ, . cos .( )= − ( )−1 0 00255 2 0 00027 ;  is the geographic latitude; dn is the altitude of the station (m); ln is the ground pressure (hpa); bcm is the zenith dry delay (mm). when converting from wet delay to precipitable water content, the air’s weighted average temperature ej in the conversion factor used is the bevis model, namely: e ej n= +70 1 0 71. . (2) where en is the station temperature (k). analysis of periodic change of precipitation according to the gps remote sensing technology of the atmospheric precipitable water in the four stations in the last section, through trend analysis and wavelet analysis process, the periodic change of precipitation in the four stations is further analyzed (xing & wang, 2016). trend analysis the trend change is generally expressed by a linear regression equation based on precipitation and time. 87tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm f a bi= + (3) where, f is the fitting value of precipitation; a is the annual change rate; a × 10 is the climate tendency rate, which means the change rate of precipitation per 10 a; i = 1, 2… n years. wavelet analysis in this paper, the wavelet analysis method is based on fourier transform, which introduces the window function to decompose the time series into the contribution of time and frequency. it is very effective for obtaining the adjustment rule of a complex time series, diagnosing the hierarchical structure of climate change, and distinguishing the evolution characteristics of time series on different scales. through wavelet analysis, we can get the periodic structure and abnormal change rule of the studied object sequence in different time scales, which provides a scientific basis for short-term climate prediction (rao et al., 2016). combining with the anomaly calculation, the increase and decrease characteristics of precipitation in four stations are analyzed, and the periodic oscillation of precipitation in different time scales can be reflected by using morlet continuous wavelet transform chart. the strength of the signal is expressed by the size of the wavelet coefficient. if the isoline is positive, it means more precipitation; if the isoline is negative, it means less precipitation; if the wavelet coefficient is zero, it means the mutation point. the wavelet variance map reflects the distribution of energy with scale year, which can determine the relative intensity of various scale disturbances in a time series. the scale at the corresponding peak is called the main time scale of the series, which is used to reflect the main period of the time series (castro et al., 2017). wavelet analysis takes time and frequency as independent variables, expands one-dimensional signals in time and frequency, so as to clearly understand the frequency characteristics of different time-domains of time series and the time distribution characteristics of different frequencies. wavelet function is the key to wavelet analysis. it refers to a kind of function with oscillation and rapid attenuation to zero (luo et al., 2017; li et al., 2019). wavelet function meets the following requirements: ∫ ( ) = −∞ +∞ δ i ic 0 (4) in the formula, (i) is the base wavelet function, which can form a group of function systems through scaling and translation on the time axis: δ δb d i b i d b b d g b, / , ,( )= −      ∈ ≠ −1 2 0 (5) in the formula, b and d (i) are subwavelets; b is the scale factor, reflecting the cycle length of wavelet; d is a translation factor, reflecting the translation in time. for a given energy limited signal, its continuous wavelet transform is: h b d b f i i d b il g ( , ) ( )= −      − ∫½ cδ (6) in the formula, hl( b, d ) is the wavelet transform coefficient; l (i) is a signal or square-integrable function; b is the scaling scale; d is the translation parameter; δ i d b −      is the complex conjugate function of δ i d b −     . for discrete-time series data, assuming that there is a function l (ei) (e = 1, 2, ..., i; i) is the sampling interval), the discrete wavelet transform form is: h b d b l e i e i d b il e i ( , ) = ∆ ∆( ) ∆ −      − = ∑½ i cδ 1 (7) results periodic change of precipitation series the wave characteristics of the real part of the wavelet transform coefficients can reflect the alternation of more and less precipitation, and the alternation of positive and negative positions can be observed. the wave changes of more and less precipitation in the calculation time domain can be observed obviously. it can be seen from figure 2 that the precipitation structure corresponding to different time scales is different, and the number of smallscale changes shows a more complex number of structures nested on a larger scale. based on the average situation of precipitation in harbin, changchun, shenyang, and dalian stations, the precipitation situation in the northern semiarid area is estimated. there are five kinds of periodic changes in the average situation of precipitation in the four stations: 3-9, 9-18, 18-30, 30-41a, and 4163a. among them, the periodic oscillation is very significant on the scale of 4163a, and the annual precipitation has experienced the cycle of less-more-lessmore-less until the isoline of 2005 is not closed, which indicates that the annual precipitation will continue to be less in the future; the 10a and 50a cycles have more obvious oscillation periods after 1955. the modulus of morlet wavelet coefficients reflects the distribution of energy density in the time domain corresponding to the period of change of different time scales. the larger the modulus of coefficients is, the stronger the periodicity of the corresponding period or scale is. in the annual precipitation of 4 stations, the 50a time scale model value is the largest, which indicates that the cycle change of this time scale is the most obvious. the 10a and 20a cycles are relatively large after 1955, and the cycle change is more obvious, and the cycle change of other time scales is small (figure 3). figure 2. real-part time-frequency distribution of morlet wavelet transform coefficients figure 3. morlet wavelet transform is a digital modular square timefrequency distribution 88 yan lv, laijun lu, mengxue cao relationship between precipitation anomaly and wavelet periodic frequency domain of four stations in the north of china in recent 50 years in order to understand the variation law of precipitation in the northern semi-arid area in the past 50 years, the variation law of precipitation in harbin, changchun, shenyang, and dalian stations in the past 50 years is analyzed. figure 4(a) shows the precipitation distance map of harbin in the recent 50 years. from figure 4(a), in addition to the relationship between the annual precipitation and the 50-year average precipitation, it can be seen that the precipitation in harbin experienced two cycles of less to more in 1955-2005, and the precipitation anomaly from 1955 to 1982 was negative, indicating that the precipitation in this period was less than that in the whole year; the precipitation anomaly from 1983 to 2005 was positive, indicating that the precipitation in this period was more than that in the whole year. the average annual precipitation of harbin in 50 years is about 550 mm/a. by linear fitting of annual precipitation series in harbin, it is found that the precipitation tendency rate is 7.1 mm/10a, and the cumulative effect of such a high precipitation tendency rate is relatively large, which will lead to a gap of tens of millimeters between the current annual precipitation of harbin and that of 50 years ago. (a) (b) figure 4. annual precipitation anomaly (a) and morlet wavelet transform (b) of harbin from 1955 to 2005 in addition to the above analysis, it is difficult to see the change information of the shorter period from figure 4(a). in this regard, figure 4(b) gives the corresponding morlet wavelet transform graph. figure 4(b) clearly shows the periodic oscillation and abrupt point characteristics of annual precipitation in harbin over the past 50 years on different time scales. the intensity of signal oscillation in the figure is represented by the size of the gray-scale. the larger the gray-scale is, the smaller the annual precipitation is; the smaller the gray-scale is, the larger the annual precipitation is. it can be seen from the comparison between figure 4(b) and figure 4(a) that more abundant information of the change period scale can be obtained from figure 4(b). it can be seen from figure 4(b) that there is a phenomenon of nested complex structure on the scale of multiple time periods. there are obviously short-term oscillation periods of 2-3a and 5-6a, the oscillation period of 10a, the medium-term oscillation period of 20a, and the long-term oscillation period of 50 after 1955. from the change law in the figure, after 2005, harbin is in the period of 2-3a, 5-6a short-term oscillation cycle with more precipitation; that in the oscillation period of 10a is less precipitation; in the medium-term oscillation period of 20a has experienced two cycles of more-less alternation, and the isoline of 2005 oscillation cycle is far from closed, which shows that after 2005, the precipitation is in the period of less precipitation; and the isoline in the longterm oscillation period of 50a is still not closed, indicating that the precipitation will continue to maintain a slightly more trend in this period after 2005. to sum up, the precipitation of harbin in recent 50 years is controlled by multiple periodic laws, and there are obviously short-term oscillation periods of 2-3a and 5-6a, the oscillation period of 10a, the medium-term oscillation period of 20a, and the long-term oscillation period of 50a after 1950. from the point of view of the intensity of the oscillation, the medium-term change of 20a is the main control period of annual precipitation, but the short-term period of 5-6a and the long-term period change of 50a also have a great influence on the annual precipitation. figure 5(a) shows the precipitation distance map of changchun in the recent 50 years. from figure 5(a), it can be seen that the annual precipitation change law of changchun from 1955 to 2005 is similar to that of harbin, and it has also gone through two cycles, but the cycle law is just the opposite, that is, there are two cycles of more to less, the precipitation anomaly from 1955 to 1982 is positive, which indicates that the precipitation is more than that of the whole year in this period; the precipitation anomaly from 1983 to 2005 is negative, which indicates that the precipitation is less than that of the whole year in this period. the linear fitting of changchun’s annual precipitation series shows that the precipitation tendency rate is 12.7 mm/10a, which is higher than that of harbin. this shows that the annual precipitation in changchun is decreasing, and the decreasing range is larger than that in harbin. the accumulated precipitation in 50 years will cause the annual precipitation in (a) (b) figure 5. precipitation anomaly (a) and wavelet transform (b) of changchun from 1955 to 2005 89tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm changchun to decrease by 126.8 mm more than that at the beginning of the 20th century. considering that the annual average precipitation in changchun is about 596 mm, this reduction is quite amazing. in order to analyze the multi-periodicity of precipitation in changchun in recent 50 years more clearly, the corresponding morlet wavelet transforms chart is given in figure 5(b), which clearly shows the periodic oscillation and mutation point characteristics of precipitation in changchun in recent 50 years on different time scales. it can be seen from the comparison between figure 5(b) and figure 5(a), more abundant scale information of the change period can also be obtained from figure 5(b). it can be seen from the comparison between figure 4(b) and figure 5(b) that the wavelet transform-law of annual precipitation in changchun and harbin is very similar. there are not only nested complex structure phenomena on multiple time period scales but also obvious short-term oscillation periods of 2-3a, 5-6a, 10-15a, 20a, and 50a after 1955. this may be the reason why the two cities are in the same climate background in northeast china. the difference is that the long-term oscillation period of about 50 years is exactly the same as that of the less-more cycle in harbin, that is, the more-less cycle. according to the change law in the figure, changchun is in the short-term oscillation periods of 2-3a and 5-6a with more precipitation after 2005; the oscillation period of 10a is the period of less precipitation; the medium-term oscillation period of 20a has experienced the cycle of more-less alternation, and the isoline of oscillation cycle in 2005 is far from closed, which shows that the precipitation is in the period of less precipitation after 2005; the isoline of the long-term oscillation period of 50a is still not closed, which indicates that the precipitation will continue to be less in this period after 2005. in conclusion, the precipitation of changchun in recent 50 years is controlled by multiple periodic laws, and there are obviously short-term oscillation periods of 2-3a and 10a, medium-term oscillation periods of about 20a, and long-term oscillation periods of about 50a. from the point of view of the intensity of the oscillation, the medium-term period variation of 20a is the main control period of the annual precipitation, but the short-term period of 2-3a and the long-term period variation of 50a also have a great influence on the annual precipitation. figure 6(a) shows the precipitation distance map of shenyang in the recent 50 years. compared with harbin (figure 4(a)) and changchun (figure 5(a)) in the past 50 years, it is found that the precipitation decrease trend of shenyang in the past 50 years is not obvious, and the precipitation anomaly is close to zero. the linear fitting of the annual precipitation series in shenyang shows that the precipitation tendency rate is 1.3 mm/10a, and the precipitation slightly increases. in order to analyze more clearly the multi-periodicity of precipitation variation in recent 50 years in shenyang, the corresponding morlet wavelet transform chart is given in figure 3(b), which clearly shows the periodic oscillation and mutation point characteristics of precipitation in recent 50 years in shenyang on different time scales. it can be seen from the comparison between figure 6(b) and figure 6(a) that more abundant information of the change cycle scale can also be obtained from figure 6(b). compared with the wavelet transform of morlet in harbin (figure 4(b)) and changchun (figure 5 (b)), it can be seen that shenyang also has an obvious short period of 2-3a and 10a and a long period of 50a, but the main control period of annual precipitation of 20a does not appear. according to the change rule in the figure, shenyang is in the period of 2-3a, 10a, and 50a with more precipitation after 2005. in conclusion, the precipitation in the recent 50 years in shenyang is controlled by multiple periodic laws, and there are obviously short-term oscillation periods of 2-3a and 10a and long-term oscillation periods of about 50a. from the point of view of the intensity of the oscillation, the mediumterm period variation of about 10a is the main control period of the annual precipitation, but the short-term period of 2-3a and the long-term period variation of 50a also have a great influence on the annual precipitation. figure 7(a) shows the precipitation anomaly of dalian in the recent 50 years. it can be seen that the annual precipitation variation law of dalian from 1955 to 2005 is similar to that of changchun, and it has also gone through two-cycle periods, i.e. more-less cycle period, i.e. the precipitation anomaly from 1955 to 1980 is positive, which indicates that the precipitation in this period is more than that in the whole year; the precipitation anomaly from 1981 to 2005 is negative, which indicates that the precipitation in this period is less than that in the whole year. the precipitation tendency rate of dalian is 2.7 mm/10a, which is smaller than that of harbin and changchun. this may be related to the influence of marine climate in dalian. in order to analyze the multi periodicity of precipitation variation in dalian in recent 50 years more clearly, the corresponding morlet wavelet transform chart is given in figure 7(b), which clearly shows the periodic oscillation and mutation point characteristics of precipitation in dalian in recent 50 years on different time scales. it can be seen from the comparison between figure 7(b) and figure 7(a) that richer scale information of the change period can also be obtained from figure 7(b). compared with the wavelet transform of morlet in harbin (figure 4(b)), changchun (figure 5(b)) and shenyang (figure 6(b)), it can be seen that dalian also has obvious short-term periods of 2-3a and 10a and long-term period of 50a, but the main control period of annual precipitation in 20a does not appear which is the same as that in shenyang. it also shows that shenyang and dalian are close in geography and regional climate. from the change in the figure, dalian and shenyang are also in the precipitation period of 2-3a, 10a, and 50a after 2005. in conclusion, the precipitation of dalian in recent 50 years is controlled by multiple periodic laws, and there are obviously short-term oscillation periods of 2-3a and 10a and long-term oscillation periods of 50a. from the point of view of the intensity of the oscillation, the medium-term period change of 10a is the main control period of the annual precipitation, but the short-term period of 2-3a and the long-term period change of 50a also have a great influence on the annual precipitation. the main period of precipitation series the wavelet variance map can reflect the distribution of the fluctuation energy of the precipitation time series with the scale and can be used to (a) (b) figure 6. shenyang annual precipitation anomaly (a) and morlet wavelet transform (b) from 1955 to 2005 90 yan lv, laijun lu, mengxue cao determine the main period in the annual precipitation. there are five obvious peaks in the wavelet variance map of precipitation in four stations, which correspond to the time scales of 50a, 37a, 20a, 14a, and 6a in turn. among them, the maximum peak corresponds to the time scale of 50a, indicating that the cycle oscillation of about 50a is the strongest, which is the first main cycle of precipitation change in four areas; the 20a time scale corresponds to the second peak, which is the second main cycle of annual precipitation change; the third peak corresponds to the time scale of 14a, which is the third main cycle; the fourth peak corresponds to the time scale of 37a, which is the fourth main cycle; the fifth peak corresponds to the time scale of 6a, which is the fifth main period. the fluctuation of the above five periods controls the variation characteristics of the annual precipitation in the whole time domain of the four regions, as shown in figure 8. according to figure 8, the main periods of precipitation in four stations of harbin, changchun, shenyang, and dalian are 6a, 14a, 20a, 37a, and 50a respectively, of which the peak value of 50a precipitation is the most obvious. therefore, the periodic changes of precipitation in the four stations can be derived from the periodic changes of precipitation in the northern semi-arid areas. discussion the northern part of china consists of three parts: the northwest, the north, and the northeast. the area is large and the terrain is complex, the influence system is various, and the characteristics of precipitation change in different areas are quite different. especially in recent decades, under the background of global warming, there are signs of transition from drought to wetness in north china, but north china and the south of northeast china are still in a continuous semi-arid period. this paper analyzes the precipitation regularity in the north-east semi-arid area by taking the 50-year periodic precipitation variation regularity of harbin, changchun, shenyang, and dalian as an example. the research on the change rules of precipitation time series and the characteristics of multi-time scales in the semi-arid area of north china is the basis of the research on how to mine the water production potential and development degree of grain crops, which is of great theoretical significance for the rational development of grain production in the commodity grain production area of north china. based on the analysis of the 50-year precipitation series from 1955 to 2005 in harbin, changchun, shenyang, and dalian, the multitime scale complex structure of the annual precipitation change in the northern semi-arid area is studied. the main conclusions are as follows: (1) there are five kinds of periodic changes of precipitation in the north of china: 3-9, 9-18, 18-30, 30-41a, and 41-63a. the period of 9-18a and 18-30a appeared more obvious after 2005 and 1955 respectively, and the period of 4163a is more significant. (2) in the past 50 years, the annual precipitation in the northern region has shown a decreasing trend, with a decrease of 5.2 mm/10a in the whole northeast region; 12.7 mm/10a in changchun; 7.1 mm/10a in harbin; 2.7 mm/10a in dalian; and 1.3 mm/10a in shenyang. (3) the annual precipitation in the northern region has the periodic change characteristics under the regional multiple time scales, and the long-term oscillation periods of 2-3a, 5-6a, 10a, and 50a have the global characteristics. from the point of view of the intensity of the oscillation, the mid-period change of about 20 years is the main control period of the annual precipitation in changchun and harbin, but the short period of 5-6 years and the long period change of 50 years also have a great influence on the annual precipitation. (4) from the study of the periodicity of the wavelet transform, it is found that after 2005, harbin is in the period of 2-3a, 5-6a short-term oscillation cycle with more precipitation; the oscillation period of 10a and 20a is the period of less precipitation, and the long-term oscillation period of 50a will continue to tend to be more than one. changchun is in the period of 2-3a, 5-6a short-term oscillation with more precipitation; the oscillation period of 10a and 20a is less precipitation, and the long-term oscillation period of about 50a will continue to be less. shenyang is in a more precipitation period of 2-3a, 10a, and 50a. dalian is in a more precipitation period of 2-3a, 10a, and 50a. however, the whole northern region is in the period of more precipitation in 2-3a and 10a, less precipitation in 20a, and more precipitation in 50a. (5) there are five obvious main periods of precipitation in the four stations, which correspond to the time scales of 50a, 37a, 20a, 14a, and 6a, among which 50a is the highest peak, and the fluctuation of five periods controls the variation characteristics of annual precipitation in the whole time domain. based on the above analysis, the periodic variation of precipitation in the northern semi-arid area can be deduced. under the background of global warming, the characteristics of precipitation change in different regions in the north are different, which may be caused by the influence of different circulation characteristic factors in different regions and scales. therefore, to understand the causes of precipitation change (a) (b) figure 7. annual precipitation anomaly (a) and morlet wavelet transform (b) of dalian from 1955 to 2005 figure 8. the variance of wavelet transform 91tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm in different regions in the north, it is necessary to further combine the factors of zonal circulation, meridional wave train, east asian monsoon, and sst to make further research. conclusions in this paper, the precipitation regularity of the four stations in harbin, changchun, shenyang, and dalian from 1955 to 2005 is analyzed. the precipitation data of the four stations in the northern semi-arid area in the past 50 years are obtained by gps remote sensing technology. according to the precipitation data, the regression equation of trend change and the wavelet analysis method are used to describe the trend change of precipitation in four stations and the regular pattern of the periodic structure and the abnormal change of precipitation periodic series in different time scales. among them, in the process of wavelet analysis, the 50 year precipitation anomaly and the frequency domain relationship of wavelet period in four regions are analyzed, and the precipitation tendency rate of each region and the precipitation in the short, medium and long-term oscillation periods of wavelet are obtained. according to the precipitation time series fluctuation reflected by wavelet variance chart, the main period in the comprehensive annual precipitation of four regions is determined, and the annual precipitation in the whole time domain of four regions is studied based on the characteristics of internal variation, and the regularity of periodic variation of precipitation in the northern semi-arid area can be inferred. the results show that based on gps remote sensing technology and wavelet analysis, the proposed method can effectively analyze the periodic change of precipitation in the northern semi-arid area. acknowledgments the national natural science foundation of china (grant no. 61006403). references arrizabalaga, s., & li, z. (2018). factors influencing the assessment of albacore tuna resources in the indian ocean. ccamlr science, 25(2), 107-119. bi, s., qu, y., bi, s., wu, w., & jiang, t. (2018). multi-scale impacts of the pacific sst and pdo on the summer precipitation of north-central china from 1870 to 2002. theoretical & applied climatology, 132, 953-963. castro, s., souto, a., schiel, n., & biondi, l. (2017). techniques used by bearded capuchin monkeys (sapajus libidinosus) to access water in a semi-arid environment of north-eastern brazil. folia primatologica, 88, 267-273. colliander, a., cosh, m. h., misra, s., jackson, t. j., crow, w. t., chan, s., bindlish, r., chae, c., collins, c. h., & yueh, s. h. (2017). validation and scaling of soil moisture in a semi-arid environment: smap validation experiment 2015 (smapvex15). remote sensing of environment, 196, 101-112. dai, x., yin, h., & bai, y. (2018). spatial and temporal variation characteristics of marine wetland ecological environment based on gis. ccamlr science, 25(3), 143-150. ferrari dos santos, l., freitas guimaraes, m., oliveira de souza, g., gomes da silva, i. w., rodrigues santos, j., santos azevedo, s. s., bahia labruna, m., heinemann, m. b., & horta, m. c. (2017). seroepidemiological survey on leptospira spp. infection in wild and domestic mammals in two distinct areas of the semi-arid region of northeastern brazil. tropical animal health & production, 49, 1-8. guo, j., fan, h,. wang, x., zhang, l., ren, l., he, y., ma, f., meng, w., bai, x., & yin, j. (2019). integrating geochemical anomaly and remote sensing methods to predict oiland gas-bearing areas in the yanchang oil field, ordos basin, china. earth sciences research journal, 23, 79-86. hall, l. k., larsen, r. t., westover, m. d., day, c. c., knight, r. n., & mcmilan, b. r. (2016). influence of exotic horses on the use of water by communities of native wildlife in a semi-arid environment. journal of arid environments, 127, 100-105. huang, s., li, j., li, q., & zhang, h. (2018). optimization of big data fusion scheduling in maritime communication based on fuzzy c-means clustering algorithm. ccamlr science, 25(3):229-236. li, d., xue, h. f., & zhang, y. (2019). combined forecasting model of precipitation based on empirical mode decomposition. computer simulation, 36, 464-469. li, x. d., guo, d., zhang, c. p., niu, d., fu, h., & wan, c. (2018). contribution of root respiration to total soil respiration in a semi-arid grassland on the loess plateau, china. science of the total environment, 627, 12091217. luo, y. q., zhao, x. y., li, y. q., & wang, t. (2017). effects of foliage litter of a pioneer shrub (artemisia halodendron) on germination from the soil seedbank in a semi-arid sandy grassland in china. journal of plant research, 130, 1-9. megat, a., & hassan, m. (2018). the optimal irrigation level to maximize the vegetative growth of kenaf species. ccamlr science, 25(3), 178-182. mora, j. l., herrero, j., & weindorf, c. (2017). multivariate analysis of soil salination-desalination in a semi-arid irrigated district of spain. geoderma, 291, 1-10. rao, z. g., jia, g. d., li, y. x., chen j., xu, q., & chen, f. (2016). asynchronous evolution of the isotopic composition and amount of precipitation in north china during the holocene revealed by a record of compoundspecific carbon and hydrogen isotopes of long-chain n-alkanes from an alpine lake. earth & planetary science letters, 446, 68-76. rauber, r. b., demaria, m. r., jobbagy, e. g., arroyo, d. n., & poggio, s. l. (2018). weed communities in semiarid rainfed croplands of central argentina: comparison between corn (zea mays) and soybean (glycine max) crops. weed science, 66, 1-11. ren, j., yang, w., yang, x., deng, x., zhao, h., wang, f., & wang, l. (2019). optimization of fusion method for gf-2 satellite remote sensing images based on the classification effect. earth sciences research journal, 23, 163-169. wang, b., waters, c., orgill, s., gray, j., cowie, a., clark, a., & liu, d. l. (2018). high resolution mapping of soil organic carbon stocks using remote sensing variables in the semi-arid rangelands of eastern australia. science of the total environment, 630, 367-378. wang, j., fu, b. j., lu, n., & zhang, l. (2017). seasonal variation in water uptake patterns of three plant species based on stable isotopes in the semi-arid loess plateau. science of the total environment, 609, 27-37. xing, w., & wang, b. (2016). predictability and prediction of summer rainfall in the arid and semi-arid regions of china. climate dynamics, 49, 1-13. zeng, r., wang, y., & wang, w. (2018). a co-occurrence region based bayesian network stepwise remote sensing image retrieval algorithm. earth sciences research journal, 22, 29-35. zhao, y. c., wang, x. y., & vazquez-jimenez, r. (2018). evaluating the performance of remote sensed rain-use efficiency as an indicator of ecosystem functioning in semi-arid ecosystems. international journal of remote sensing, 39, 3344-3362. _1fob9te _97fs561w4vea issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n3.96209 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (september, 2022): 239 254 g e o l o g y geological education scenario in india and role of open educational resources in the light of covid-19 pandemic omkar vermaa*, manthena prashantha, roberto grecob, ashu khoslac, kulwant singhd a. geology discipline group, school of sciences, indira gandhi national open university, new delhi, india. e-mail: omkarverma@ignou.ac.in b instituto de geociências, universidade estadual de campinas, rua carlos gomes, 250, campinas, sp, brazil. c. department of geology, panjab university, chandigarh, india. d. department of geology, ggm science college, jammu tawi, india. abstract geological education plays a major role in the social, economic, and cultural growth and development of any country. india is a vast diversified country with wide-ranging geological features, consisting of rocks of all geological ages with well-developed physiographic divisions. the indian lithospheric plate is a storehouse of vast georesources and also serves as a natural field laboratory for developing and testing numerous geological principles. the exploration of indian georesources began in the past; with the organized exploration began way back in the 1830s. the number of institutions offering geological education is, however, comparatively low. thus, geology as a science discipline is less popular in comparison with other science subjects such as physics, chemistry, zoology, botany, and environmental science, based on enrolment data of the learners enrolled in academic sessions from 2011-2012 to 2018-2019 of the master’s degree programs. analysis indicates that teaching-learning resources of foreign origin are dominantly being adopted as learning resources in the teaching-learning pedagogy of degree programs in geology, and are over-riding those of indian origin. the covid-19 pandemic is affecting geological education in two ways namely, the disruption of conventional (face-to-face) teaching together with not allowing learners of geology programs to complete their compulsory out-door geological fieldwork assignments. the analysis of data gathered by an online questionnaire survey shows that the use of indigenous open educational resources in geology appears as a plausible solution to overcome the learning barriers created by the covid-19 pandemic, and also to enrich teaching-learning resources of indian origin. this article provides an up-to-date detailed account of the necessity and evolution of geological education, the current popularity of geology as a science discipline, the nature of available geological teaching-learning resources, the impact of the covid-19 on geological education, and the role of open educational resources in providing quality and equitable geological education, whilst removing educational barriers created by covid-19 in india. keywords: geological education; enrolment data; higher education; oer; covid-19; geological fieldwork; india. el escenario de la educación geológica en india y el rol de los recursos pedagógicos abiertos a las luces de la pandemia del covid-19 resumen la educación geológica juega un papel importante en el desarrollo y en el crecimiento social, económico y cultural de cualquier país. la india es un país vasto y diverso y con un amplio rango de características geológicas que consiste en rocas de todas las edades geológicas y divisiones fisiográficas bien desarrolladas. la placa litosférica india es un depósito de grandes georecursos y también sirve como un laboratorio de campo para el desarrollo y evaluación de numerosos principios geológicos. la exploración organizada de los georrecursos indios comenzó en la década de 1830. el número de instituciones que ofrecen educación geológica es, sin embargo, comparativamente bajo. de este modo, la geología es una ciencia disciplinar menos popular en comparación con otras materias de estudio como la física, la química, la zoología, la botánica y la ciencia ambiental, con base en la información de inscripciones de los estudiantes en los programas de maestría para los períodos de 2011-2012 hasta 2018-2019. el análisis indica que los recursos foráneos de enseñanza-aprendizaje son adoptados dominantemente como los recursos de aprendizaje en la pedagogía de los programas de pregrado en geología y superan aquellas de origen indio. la pandemia del covid-19 está afectando la educación geológica en dos vías especificamente: la disrupción de la enseñanza convencional (presencial) junto con no permitir a los estudiantes de los programas de geología completar sus tareas obligatorias de trabajo de campo. el análisis de la información recopilada por una encuesta digital muestra que el uso de recursos educativos nativos abiertos en geología es una solución posible para superar las barreras de aprendizaje creadas por la pandemia, y también enriquecen los recursos de enseñanza-aprendizaje de origen indio. este artículo provee un recuento detallado y actualizado de la necesidad y evolución de la educación geológica, la actual popularidad de la geología como una ciencia disciplinar, la naturaleza de los recursos disponibles de enseñanza-aprendizaje, el impacto de la pandemia del covid-19 en la educación geológica, y el rol de los recursos pedagógicos para proveer educación geológica equitativa y de calidad mientras se remueven las barreras educativas creadas por la pandemia en india. palabras clave: educación geológica; datos de inscripciones; educación superior; recursos educativos abiertos; covid-19; trabajo de campo geológico; india. record manuscript received: 28/05/2021 accepted for publication: 17/08/2022 how to cite item: verma, o., prashanth, m., greco, r., khosla, a., & singh, k. (2022). geological education scenario in india and role of open educational resources in the light of covid-19 pandemic. earth sciences research journal, 26(3), 239-254. https://doi. org/10.15446/esrj.v26n3.96209 https://doi.org/10.15446/esrj.v26n3.96209 mailto:omkarverma@ignou.ac.in https://doi.org/10.15446/esrj.v26n3.96209 https://doi.org/10.15446/esrj.v26n3.96209 240 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh 1. introduction the usage of the word “geology” in the scientific literature can be traced to the beginning of the eighteenth century (howarth, 2020). in the late nineteenth century, the scope of the discipline of geology was massively broadened in the western world (king, 2008). as a result, different terminologies for the discipline of geology came into existence. the new terms “earth sciences” emerged in 1852; “geoscience” appeared in 1929 and later, “geological sciences” in 1938, in the geological literature (wilson, 1883; adams, 1938; field, 1941; howarth, 2020). all these words are synonymous with each other and broadly, involved in the study of the whole earth and other planetary systems (manduca & kastens, 2012). geology has been described as a historical and descriptive science, where processes and conditions of earth, atmosphere, and oceans are investigated to understand earth for the well-being of the society (frodeman, 1995; manduca & kastens, 2012). it is a scientific discipline of societal relevance in meeting many of its needs notably, natural resources, environmental quality, and resilience to natural hazards (king, 2019). geoscientists use current natural phenomenon data to understand earth’s past data to evaluate earth’s present conditions and together, predict how earth will likely behave in the future. thus, geological education is the backbone of the socio-economic development of any country, and the dissemination of its knowledge to the different individuals (learners, public, and policymakers), is crucial in the decision-making processes, and for the long-term survival of humanity on the mother earth. geology plays a significant role in the overall growth and development of society. in support of geological education, an international geoscience education organization (igeo) was formally established in 2000 (king, 2019). the main objectives of the igeo are to promote geoscience education at the global level, enhance its quality and equity, and increase its awareness in all sectors of society at all levels (http://www.igeoscied.org/aboutthe-igeo/). in this context, eight international conferences on geoscience education have been organized worldwide so far by the igeo since 1993 (http://www.igeoscied.org/activities/geoscied/). the igeo is also promoting open online teaching geological resources encompassing earth learning ideas, complete teaching units, and geoscience textbooks for the use of the global teaching-learning community in various international languages (king, 2008; http://www.igeoscied.org/teaching-resources/). apart from this, the igeo has been conducting international earth science olympiad since 2007 for popularizing geoscience education among school kids (http://www.igeoscied.org/activities/ieso-2/). both, geological education and research are essential in india, because the indian subcontinent is endowed with significant amounts of georesources, whilst research on its geological settings has proved to be valuable in unraveling various mechanisms of the earth system and generation of the georesources of the subcontinent (wadia, 1964; tandon et al., 1991; naqvi, 2005; radhakrishna, 2005; santosh, 2012; valdiya, 2016; searle & treloar, 2019; kumari et al., 2020). geological research in india began in 1836 (oldham, 1856), but, geology as an independent educational science discipline first appeared in a few indian universities in the 1890s (murty, 1982). an analysis of the enrolment data for the master’s degree program in geology shows that geology is still a less popular science discipline in india. it also indicates that a majority of teaching-learning resources (tlr) consisting of reading material used in higher geological education, are more of foreign origin, and may or may not be accessible to the teaching-learning community in india. two types of negative impacts are visualized in the light of the current pandemic (i.e. covid-19) on geological education, firstly, temporary suspension of classroom/face-toface teaching, and secondly, the creation of an unfavorable environment for completing geological field-based compulsory assignments for bachelor’s and master’s degree programs. recently, open education resources (oer) are becoming popular among science disciplines and the global teaching-learning community has been using them extensively in day-to-day teaching-learning practices (smith, 2009; henderson & ostashewski, 2018; y verma, 2021a). oer are also gaining importance in india and currently, there are a few oer in geology, which are available on online platforms of the government of india (goi), for example, the e-pgpathshala portal of the university grant commission (ugc) and swayam of the ministry of human resources development (mhrd) (https:// epgp.inflibnet.ac.in/; https://swayam.gov.in/), and a few available on social media platforms like youtube. it has been shown that course-based-specific oer (including virtual labs, virtual field experiences, and google earth-based laboratory/field manuals) for bachelor’s and master’s degree programs in geology are the need of the hour (verma, 2021a). the use of oer in teachinglearning practice offers an opportunity to narrow down the gap of the tlr of indian origin. oer provide off-campus specific-course-based learning during any pandemic like covid-19, and also can, to an extent, facilitate the learners to complete their geological field-based assignments using google earth-based virtual field manuals. research focusing on the past, present, and future of geological education is a major concern and is being carried out in the western hemisphere particularly, in the us and uk (king, 2008, 2018; gsa position statement, 2016; dolphin et al., 2019). as a consequence, a journal has been launched, the “journal of geoscience education” for publishing geological education research. to date, substantial research on geological education has not taken place in india, and probably, this article may be first of its kind in india. the primary objectives of this paper are threefold: first, to evaluate why geological education is necessary for the country and review its historical development; second, to assess the current popularity of geology as a science discipline, the nature of available geological tlr, and impact of covid-19 on higher geological education; and third, to evaluate oer role in providing high quality and equitable geological education, and in removing some of the educational barriers created by covid-19. 2. methodology two methodological approaches were adopted for this study. the first approach involved the collection and analysis of literature to evaluate the importance of imparting geological education, trace the development of geological education from 1851 to the present in india, evaluate the status of geological education in comparison with other sciences at the level of master’s degree programs, and finally, know the nature of geological tlr in india. it was found that almost all research dealing with the historical development of geological education in india was undertaken by historians. to date, no researcher with a geological background had investigated evolutionary development and the future of geological education in india. the data dealing with the enrolment and successful completion of master’s degree programs in geology, physics, chemistry, botany, zoology, and environmental science for academic sessions from 2011-2012 to 20182019 was obtained from the annual reports of the ministry of human resources and development (now renamed as the ministry of education), goi within the ‘all india survey on higher education’ available on its website (http://aishe. nic.in/aishe/reports); this allowed comparative analysis between geology and other science disciplines. the data dealing with learners’ enrolment and their successful completion in the bachelor’s degree programs are not available; therefore, bachelor’s degree programs were not included in this study. we downloaded two curricula of bachelor’s degree programs in geology from the website of the university grant commission and the curricula of master’s degree programs from 25 universities (including central universities and state/union territory universities) from their websites to discover the nature of existing tlr. the number of tlr of foreign and indian origin recommended in the above-mentioned curricula were counted and analyzed. we had not disclosed the names of 25 universities and presented them as anonymous because the purpose of the research was to evaluate the nature of tlr available or used in india. the second method followed was a structured online survey. a google-based questionnaire dealing with the current use of the oer for geological teaching-learning practices in india, their role in overcoming the learning restrictions imposed by the prevailing covid-19 pandemic, and enriching teaching-learning resources of indian origin were prepared and sent by email to numerous academicians and researchers covering almost all parts of the country. the responses received were evaluated and interpreted. geological education: why it is important in india? indian geological education has many strategies for meeting the varied needs of society, addressing the economic growth of the country using natural resources, and refining basic and applied principles of geology (sharma, 1998; radhakrishna, 2005; sheth, 2007; jain et al., 2016; mandal, 2016; mukherjee et al., 2016; mazumder et al., 2019; parsons et al., 2020). since geological 241 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic education includes the whole earth system, geoscientists study where we get most of the things that we use in everyday life, such as minerals, soils, oil and natural gas, energy, air, water, and food (radhakrishna, 2005; press, 2008). geoprofessionals also play a critical role in numerous other societal services like the management of waste (sewage, water pollution, landfills, non-biodegradable, nuclear, and recycling), site selection for the construction of mega-projects (dams, tunnels, reservoirs, roads, rail tracks, and buildings) together with providing remedial measures for various environment-related problems. geoscientists study natural hazards (earthquakes, tsunamis, volcanoes, landslides, subsidence, floods, heat waves, droughts, natural contaminations, cyclones, and severe storms) to determine the magnitude of their risks and to develop resilient societies where life and property losses are minimized and preparedness, mitigation and recovery are initiated (shalley, 1971; chakraborti et al., 2011; mora, 2013; mukherjee, 2019; valdiya, 2016; fareeduddin et al., 2020; ali et al., 2021; verma, 2021b; prashanth & verma, 2022). in today’s world, due to the rapid increase in population and unsustainable practices in the use of natural resources, there is tremendous stress on earth’s resources, causing many problems such as climate change, global warming, fossil fuels scarcity, natural resources depletion, biodiversity loss, extreme weather events, water contamination, and air pollution. thus, nowadays, geo-professionals have been working to resolve various earth and environment-related issues whilst ensuring sustainability in the use of natural resources, and their protection for future generations (häggquist & söderholm, 2015; valdiya, 2016). additionally, geological education has important societal implications in improving human health standards and the environment by suggesting key remedial measures to reduce the impact of geogenic elements and anthropogenic interventions in the earth’s natural environment. the geogenic contaminants of groundwater, especially fluoride and arsenic, pose a serious threat to human health in india (ali et al., 2016; ali et al., 2021). the endemic fluorosis caused due to high concentration of fluoride in groundwater was recognized in 1937 and arsenic contaminated health effects in the year 1976. however, groundwater was not properly investigated for these contaminants until the late 1990s, resulting in the poor health of millions of people in india (chakraborti et al., 2011; raju, 2017; ali et al., 2019; saha et al., 2021). the awareness of geological time, the evolution of life, and earth’s past environments through the deep time is critically useful for society in view of the alternative perspectives of myths and religions that help us to understand how to live in harmony with the earth (ayala, 2008; kölbl-ebert, 2009). overall, geological education, information, and guidance play key roles in the decision-making process in numerous societal services such as sustainable natural resources management, mega-structures construction, natural hazards mitigation and management, local and regional planning, and environmental impact assessments (häggquist & söderholm, 2015). geological education is also critical in meeting some of the sustainable development goals of the united nations (un) such as zero hunger, good health and well-being, clean water and environment, infrastructure, sustainable cities, climate action, and sustainable marine resources (un, 2015; stewart & gill, 2017). apart from education, geological research on the indian lithospheric plate provides a platform for generating new geological ideas and offers many opportunities to study the geodynamics of the earth on the global scale (tandon et al., 1991; radhakrishna, 2005; naqvi, 2005; jain et al., 2016; santosh, 2012; mandal, 2016; mukherjee et al., 2016; valdiya, 2016; searle & treloar, 2019; parsons et al., 2020). the indian subcontinent has a distinctive topography with the indian ocean lying to the south and the snow-covered himalayan mountains belt to the north. it includes three main physiographic divisions (fig. 1): peninsular india to the south, extra-peninsular india (i.e., himalayan division) to the north, and the indo-gangetic plains sandwiched between them (shah, 2018). the indian lithospheric plate preserves the vast history of the earth from archean to holocene (4.6 billion years to the present) in its rock record (naqvi, 2005; santosh, 2012; mohanty, 2015; valdiya, 2016; shah, 2018). the peninsular region of india is dominantly composed of rocks of the precambrian age and is an amalgamation of archean cratons (dharwar, bastar, singhbhum, bundelkhand, aravalli, and meghalaya), and proterozoic mobile belts (eastern ghat, pandyan, satpura, and aravalli-delhi) and sedimentary basins (vindhyan, kurnool-cuddapah, chhattisgarh, gwalior, pakhal, marwar, bhima, and kaladgi) (fig. 1; valdiya, 2016; shah, 2018). the cratons and mobile belts suffered various phases of archean tectonic activity and as a consequence, they are a rich storehouse of varied economic minerals, building and industrial materials, gemstones, and decorative stones, and are excellent sites for investigating precambrian tectonism. the proterozoic sedimentary successions have yielded much biotic evidence, which is useful in studying the early evolution of earth’s life and atmosphere (naqvi, 2005; kumar, 2010; santosh, 2012; mohanty, 2015; valdiya, 2016; shah, 2018; mazumder et al., 2019; fareeduddin et al., 2020). figure 1. map of india showing the main physiographic divisions and geological units (modified after kumar, 2010; valdiya, 2016; shah, 2018). dc stands for dharwar craton, bc bastar craton, sc singhbhum craton, bkc bundelkhand craton, ac aravalli craton, mc meghalaya craton; pmb pandyan mobile belt, egmb eastern ghat mobile belt, smb satpura mobile belt, admb aravalli-delhi mobile belt; ckb cuddapahkurnool basin, cgb chhattisgarh basin, kb kaladgi basin, bb bhima basin, pb pakhal basin, vb vindhyan basin, gb gwalior basin, mb marwar basin; pgv pranhita-godavari valley, mv mahanadi valley, dv damodar valley, sv satpura valley; dt deccan traps, and rt rajmahal traps. prominent rift valleys such as pranhita-godavari, satpura, mahanadi, and damodar (fig. 1) probably developed during proterozoic times, where thick sedimentary covers developed within these valleys that are termed gondwana basins containing huge deposits of coal and a wealth of animal and plant fossils over upper palaeozoic to mesozoic time intervals (naqvi, 2005; valdiya, 2016; fareeduddin et al., 2020). the fossils of gondwana deposits have provided data enabling the reconstruction of land ecosystems of upper palaeozoic to mesozoic of peninsular india, together with their interrelations with ecosystems of continents that were adjacent at the time (valdiya, 2016). the prominent cretaceous volcanic traps (rajmahal and deccan) offer opportunities to study the interior of the earth, the role of mantle plumes in the rifting and drifting of the former gondwanan continents, the movement rate of the indian plate, and the environmental impacts of volcanic eruption on contemporary biota (beane et al., 1986; courtillot, 1990; verma, 2015; verma et al., 2016, mukherjee et al., 2016; verma & khosla, 2019; kania et al., 2022; khosla et al., 2022). the coastal regions of india contain a considerable cover of sedimentary successions such as bombay high (maharashtra), cambay basin (gujarat), assam shelf (assam), and krishna-cauvery basins (south india) largely formed in response to marine transgression and regression during cenozoic time, and have provided valuable sources of lignite, coal, oil and natural gas (biswas, 1982, 1987; nagendra et al., 2011; shah, 2018). these allow us to study the phenomenon of the past sea level changes concerning changes in paleoclimate (naqvi, 2005; kundal & humane, 2007; valdiya, 2016; fareeduddin et al., 2020 y verma, 2021b). 242 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh the himalayan mountains belt, which is a product of the india-asia collision that took place in eocene times, created a unique monsoon system, and provided numerous natural resources, together with increased risks of natural hazards across the indian subcontinent (sharma, 1998; valdiya, 2016; searle & treloar, 2019; parsons et al., 2020). the monsoon system is a boon for the agro-economic sector of the subcontinent. there are opportunities to study the monsoon system to understand its mechanisms and the influence of the el niño-southern oscillation cycle on it (das, 2015; dimri et al., 2016). the complex geological framework of the himalayan makes it as an astonishing natural laboratory to investigate pre-, post-, and syn-collisional and depositional events associated with geodynamics of the indian lithospheric plates and their influence on the environment. the rich fossiliferous palaeozoic-mesozoic successions of the north-western himalaya provide excellent rock outcrops for investigating the biological evolution of invertebrates (corals, mollusks, brachiopods, echinoderms). the himalayan cenozoic successions contain well-preserved vertebrate fossils (particularly of mammals), which are useful for reconstructing mammalian diversity and their bearing on paleoecology, paleoenvironment, and paleobiogeography of the indian subcontinent (naqvi, 2005; valdiya, 2016). the himalayan contains large regions of glaciers, snow, and ice that serve as a significant freshwater reserve in asia and provide uninterrupted water supply to the nine river systems of asia, and these give a lifeline for nearly one-third of humanity. climate change is seriously impacting the snow and ice cover of the himalayan ecosystem, which needs to be investigated (bisht et al. 2019; sabin et al., 2020). the himalaya is divided into five tectonic belts from north to south: trans, tethys, greater, lesser, and outer himalaya (fig. 1). each belt is bounded by prominent tectonic features such as tethyan, main central, main boundary, and himalayan frontal thrusts (naqvi, 2005; valdiya, 2016). these thrusts are seismically active zones and any tectonic movement along these thrusts can cause serious natural hazards such as slope instability, landslides, and earthquakes (rajendran et al., 2017; prashanth et al., 2021). a vast majority, around 75%, of total global landslides occur in the asian continent alone, with a significant number of landslide events occurring along the himalaya mountain belts (froude & petley, 2018). thus, the himalayan region provides unique opportunities for studying plate movements, plate boundaries, seismicity, neotectonics, slope stabilization, soil erosion proneness, climate change, glacial melting, harnessing geological resources, and mountain building processes (jain & sinha, 2005; rajendran et al., 2017; prashanth et al., 2021, 2022; kouser et al., 2022; tandon, 2022). the vast indo-gangetic plains of northern india comprise recent sediments, which were brought down from the rising himalayan by its various rivers (tandon et al., 1991; tandon, 2022). these plains contain enormous water resources and their sediments have made the soil cover of the plains very fertile and so suitable for agriculture and industrial activities. as a result, nearly 50% of the total food grain production to feed 40% population of india comes from these plains (pal et al., 2009). these plains are currently facing alarming amounts of soil degradation and water contamination (mythili & goedecke, 2016). the great thar desert of western india is an active area for exploring mineral resources and for harnessing renewable energy such as solar and wind. thus, geoscientific research plays an influential role in studying the availability and spatial distribution of natural resources, regions of natural hazards, environment-related issues, and sustainable natural resources management, which are of direct concern to the society. historically, the importance of the geological richness of the country was understood in earlier times by the mughal empire, and as a result, they ruled the subcontinent for 332 years from 1526 to 1858 (babu & nautiyal, 2013). meanwhile, britishers (i.e. east india company) invaded the country in 1757, and later, expanded their imperialism, and ruled the country from 1858 to 1947 (wagh, 2018). they understood mineral resource richness, especially the coal resources of the country, and extensively mined several mineral resources during their rule (fox, 1947). the country still has many georesources. some world-famous diamonds namely, kohinoor, great moghul, and regent were discovered from gravels of the krishna valley that were derived from the krishna lamproite (ultramafic igneous rock) field, near the north-eastern margin of the cuddapah basin, southern india (gsi, 2011). recently, geoheritage emerged as an important field of public and scientific interest that promotes geoscience education, research, geotourism, and sustainable development. geoheritage is an essential part of the natural (geocultural) heritage and encompasses “geosites of significant educational, cultural, scientific, and aesthetic value” (https://www.americasgeoheritage.com/geoheritage/). india has numerous rich and diversified geocultural heritage sites and thus, provides ample opportunities to promote education, research, and tourism related to geoheritage (shekhar et al., 2019; biswas & chauhan, 2021; chauhan et al., 2021). the gsi has recognized 26 geological sites as national geological monuments of india that are located in various geologically rich states such as andhra pradesh, telangana, chhattisgarh, gujarat, himachal pradesh, karnataka, kerala, odisha, rajasthan, maharashtra, and tamil nadu (intach, 2016). however, there are numerous geoheritage sites are in india, which are yet to be recognized (fig. 2). 2a 2b figure 2. photographs showing potential geoheritage sites in india. a. huge granite batholith, situated nearly 50 km from hyderabad along the hyderabad-warangal national highway at bhongir town, telangana, south india. this batholith is a part of the eastern dharwar craton and around 2.7 to 2.5 billion years old. b. quaternary fluvio-lacustrine deposits at lamayuru, union territory of ladakh, northern india. these deposits are 105 m thick, dominantly composed of clay and sand layers, which develop a lunar landscape at lamayuru. these deposits are popularly known as “moonland” of the lamayuru. both these geosites are places of tourist’s attraction (courtesy: naveen kumar, geologist, geological survey of india for b). 243 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic given this wide geological background, geological education is crucial for india to refine existing geological principles, establish geoheritage sites, explore georesources and ensure their sustainability, and resolve geoenvironment-related problems across the country. development of indian higher geological education the study of geological phenomena in india had been recorded from time immemorial when indians started to observe and record the behavior of earth’s natural processes. the bronze age and iron age, which were associated with the growth and development of human civilizations and their presence in the country, clearly indicate the exploration and utilization of mineral resources by ancient indians. there are several references for using basic concepts of geology and metallurgy in the ancient literature of the country. however, reliable evidence of the early existence of geological information and its utilization for the development of society comes from an old treatise “arthashastra”, composed by an ancient indian statesman and philosopher, kautilya, around 321-296 bc (chakraborty, 1970; murty, 1979). this treatise provides an account of geological knowledge, which is used to discover and extract metals from various ores (goi, 2012). the growth of geology as an independent science discipline emerged in the eighteenth century (chakraborty, 1971; rajendran, 2019). the initial phase of realization of the importance of geology in india was seen with the establishment of the survey of india in 1767 and the asiatic society of bengal in 1784 by the east india company. the main objective of the survey of india was to map the country and produce maps at different scales for discovering natural resources and society had as its objective “enquiry into the history and antiquities, arts, science and literature of asia” (chakraborty, 1971). the results of the survey and society together later played a profound role in the growth of geological studies in india. the opening of a museum in calcutta (now kolkata) in 1840 by the east india company was considered a step towards the beginning of geological education (stubblefield, 1970). indeed, the east india company understood the importance of the geological richness of india, especially the coal resources before 1835. as a consequence, they established a coal committee in 1836 for discovering the coal and mineral resources of the country (kumar, 1982). based on the rich and varied geology of the subcontinent, with its huge resource potential, the east india company formally established the world’s second-largest organization, the geological survey of india (gsi) in 1851, to undertake geological mapping and exploration of mineral resources of the country (oldham, 1856). in 1858, the east india company developed into the british raj (i.e. government of colonial india). but before the 1850s, the colleges of the east india company taught geology and mineralogy, but there was no separate department of geology (larwood, 1958; tolman, 2016). the opening of the gsi marked the informal beginning of geological higher education in india (kumar, 1982; murty, 1982; palit, 2013). shortly after its establishment, henry b. medlicott, an irish geologist, came to india in 1854 to become a professor of geology at the thomson college of civil engineering, roorkee (now indian institute of technology, roorkee) founded in 1847. he was probably the first teacher of geology in india. for a long period, british geologists undertook all activities of the gsi and they were of the view that indians were not capable of doing any kind of sophisticated teaching and research in geology, and no indian university was properly equipped to impart higher geological education. in fact, they were not much interested to open geological education in the higher learning institutes of india (kumar, 1982; tolman, 2016). as a consequence, geological education did not become part of higher education in indian universities until the 1870s (kumar, 1982; tolman, 2016). over time, the british raj found difficulties in recruiting geologists from europe because they rejected the offer of british raj to join gsi due to limited financial benefits and serious health problems in india; particularly, unfavorable climatic conditions, which often caused their death within a few years of joining the gsi (grout, 1995; kumar, 1995). this led to a shortage of geologists in the gsi and to fill this gap, the british raj decided to recruit indian geologists. in 1886, the british raj constituted two positions exclusively for indian geologists within the gsi and also felt the need to establish a sound education system to teach geology (kumar, 1995). to fill the shortage gaps of geologists and associated skilled manpower needed in the gsi and through continuous struggle by the indian educationists, a more systematic effort was made by the british raj and gsi for opening the geology departments at the presidency colleges of kolkata and madras (now chennai), and bombay university in the 1890s (murty, 1982). geology as an optional subject was introduced at bombay university in 1879. in 1892, a separate department of geology was opened at the presidency college (now presidency university), kolkata. another presidency college (now madras university) at chennai already had geology as an auxiliary subject, but an independent department of geology was started in 1910 (murty, 1982). meanwhile, the spread of geological education also reached northern india with the visit of the prince of wales to jammu in 1906. to commemorate his visit, his highness, the maharaja (i.e. ruler) sir pratap singh of jammu and kashmir, built the prince of wales college (now government gandhi memorial science college and is a part of the cluster university of jammu) on the banks of the river tawi. in this college, geology education commenced in 1907, under the stewardship of the renowned indian geologist, dr. d. n. wadia (stubblefield, 1970). over time, many geology departments were opened in other universities namely, banaras hindu university, varanasi in 1923; indian school of mines (now indian institute of technology), dhanbad in 1926; andhra university, visakhapatnam in 1941; lucknow university, lucknow in 1942; aligarh muslim university, aligarh in 1946; patna university, patna in 1946, and rashtrasant tukadoji maharaj university, nagpur in 1946 (stubblefield, 1970; bhattacharya, 2016). thus, the development of geological education was seen as a by-product of the gsi growth and rich georesources of the country. figure 3. map of india showing the distribution of higher learning institutes offering geological education in india. there were around 14 universities in the country that offered geological education before1952 (stubblefield, 1970). following one and a half decades of independence, out of a total of 35 universities, around 20 had departments of geology (wadia, 1964). later, geological education witnessed a substantial expansion, thus, many geology departments were opened in various colleges and universities across the country. at present, there are more than 153 higher learning institutes that host geology departments imparting geological education in india up to a master’s degree level (excluding colleges). these institutes include various universities (state/union territory, central and private), indian institutes of technology, indian institute of science, indian statistical institute, and indian institutes of science education and research (fig. 3). apart from conventional learning systems, geological education is also offered through open and distance modes of learning in india, with the opening of departments at dr. b. r. ambedkar open university in 1991, and indira gandhi national open university in 2010. currently, higher education in geology is being offered at three-degree levels across the country viz., 244 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh bachelor’s degree programs of three-year duration, master’s degree programs of two-year duration, and integrated bachelor’s-master’s degree programs of five-year duration. apart from this, ph.d. in geology is a common research degree program in indian universities. the need for geology education at the school level in india was first highlighted by wadia in 1964 (stubblefield, 1970). so far, the picture of geology education offered at the school level is very sparse; it is largely neglected in comparison with other science disciplines: physics, chemistry, mathematics, biology, engineering, and computer science (tandon, 2005). from kindergarten to 10th grade of high school level (pupils with age range 4 to 16 years), geology education is taught as a part of either general studies or social sciences (paliwal, 2005). in 11th and 12th grade higher school education (pupils with age range 16 to 18 years), there are very few school boards of indian states/union territories offering geology as an independent subject in their science or arts streams. the states/union territories that do so are kerala, rajasthan, jammu and kashmir, jharkhand, karnataka, maharashtra, and manipur (paliwal, 2005; padmanabhan, 2018). status of geology education among other sciences: a comparative analysis numerous public and private sectors including non-governmental organizations in india offer job opportunities to those who had qualified with a master’s degree in geology. thus, an analysis was carried out to evaluate the performance of master’s degree programs in geology, in terms of the successful completion of the program by learners and to make comparisons between the master’s degree programs in geology and other science disciplines to evaluate the percentage of their success rate and to ascertain awareness of geology among the learners. for this analysis, five science disciplines comprising physics, chemistry, botany, zoology, and environmental science were selected. the data about the learners’ enrolment and their successful pass rates (i.e. completion of programs) for the academic sessions from 2011-2012 to 2018-2019 of master’s degree programs in geology, physics, chemistry, botany, zoology, and environmental science was obtained from reports of the mhrd, goi on “all india survey on higher education” (table 1). analysis of the data shows that there was an increasing trend in enrolment numbers of learners in the master’s degree programs of all six science disciplines in the academic sessions 2011-2012 to 2018-2019 (table 1). nearly 40.86% of learners were enrolled in the master’s degree program in chemistry whilst a minimum of learners, at 2.16%, were enrolled in the master’s degree program in geology (fig. 4a). surprisingly that the importance of geology in india was realized in 1851 with the establishment of the gsi, but the enrolment percentage of the master’s degree program in geology indicates that geology had the lowest popularity as a science subject among these six science disciplines (fig. 4a). table 1. learner’s enrolment and successful completion data of master’s degree programs in geology, physics, chemistry, botany, zoology, and environmental science (complied after mhrd reports, 2014, 2015a, b, 2016a, b, 2017, 2018, 2019). academic sessions master’s programs in geology physics chemistry botany zoology environmental science enrolment qualified enrolment qualified enrolment qualified enrolment qualified enrolment qualified enrolment qualified 2011-2012 3564 1801 36954 16585 73984 34588 22850 10288 27581 12606 4520 3335 2012-2013 4028 1962 42898 17202 77858 35495 24896 11140 31965 14008 5100 2549 2013-2014 4749 1786 52496 18829 90101 35142 29680 12276 36559 15184 5395 3129 2014-2015 5179 2008 56824 20558 95295 37142 32619 12140 38850 15703 5797 2528 2015-2016 5597 2076 60889 20759 99888 34639 36736 12473 41025 13899 5393 2462 2016-2017 6039 2133 69754 24211 114328 38216 42542 14625 48774 17453 6252 2484 2017-2018 6281 2469 75114 26915 125805 44027 43967 17131 51672 19134 6766 2553 2018-2019 7134 2843 72797 28737 124526 46925 43036 19208 50904 20903 7281 3243 total 42571 17078 467726 173796 801785 306174 276326 109281 327330 128890 46504 22283 average completion percentage 40.11 37.15 38.18 39.54 39.37 47.91 figure 4. a. pie-chart showing relative enrolment data percentage for master’s degree programs in six science disciplines of academic sessions from 2011-2012 to 20182019. b. bar-chart showing average successful completion percentage of learners in master’s programs in six science disciplines of academic sessions from 2011-2012 to 2018-2019. refer to table 1 for enrolment data. env sci stands for environmental science. 40.86% 23.84% 16.68% 14.09% 2.37% 2.16% chemistry physics zoology botany envi sci geology 4b env sci geology botany zoology chemistry physics 47.91 40.11 39.54 39.37 38.18 31.15 successful comple�on percentage 245 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic further data analysis of learner’s enrolment and their successful completion in master’s programs in all six science disciplines from 20112012 to 2018-2019 had given quite interesting results (fig. 4b). the results indicate that the master’s degree program in geology had a very good learner’s average completion rate i.e. 40.11%, and stands second, after average program completion rate of environmental science, which is 47.91% (fig. 4b). the average successful program completion rate for botany, zoology, and chemistry was 39.59%, 39.37%, and 38.18%, respectively. physics had the lowest average completion rate, which is 31.15% (fig. 4b). this may indicate that better quality education was available to the qualified learners of the master’s program in geology, and alternatively, reflects a higher quality of qualified postgraduates of geology. the union public service commission (upsc) of the goi recruits candidates who had master’s degrees in geology as geologists for the gsi almost every year. the applicants from all over the country appear for the upsc written examination to obtain geologist positions in the gsi after completing their master’s degree in geology. thus, the performance of applicants in the upsc geologists’ written examination can provide a means of checking the quality of masters-level geological education in india. in the past few years of the upsc recruitment process for geologists had shown that the subject-specific knowledge of the applicants was a little low and applicants were securing average marks in the written examination, i.e. 30 to 40% (saha, 2016, 2018). nature of teaching-learning resources in geology over the past few years, several reforms in the country’s higher educational system had been made by the ugc and department of higher education of the mhrd of the goi. recently, the ugc implemented a minimum course curriculum under, the choice-based credit system (cbcs) and learning outcomes-based curriculum framework (locf) for bachelor’s degree programs in india to provide high-quality and equitable education (ugc, 2015a, 2019). under the cbcs and locf, the ugc issued a national curriculum in geology for implementation in bachelor’s (both general and honors) degree programs for all higher education institutes in india (ugc, 2015b, c, 2019). the mhrd had taken several remarkable steps in enhancing the quality and providing equity with extensive use of information and communication technology (ict) in higher education. nowadays, online higher education is also being offered to learners through massive open online courses (moocs) through various online platforms (http://www.sakshat.ac.in). in addition, tlr in the form of the oer in various disciplines for the use of teachers and learners have also been provided by the goi (e.g., https://epgp.inflibnet.ac.in/). to know the nature of the tlr in geology for its bachelor’s and master’s degree programs, the tlr recommended for all the courses of these programs were evaluated. for bachelor’s degree programs, the tlr mentioned in ugccbcs and locf in geology, in the form of recommended books/texts, were evaluated (table 2). for master’s degree programs, the master’s curricula of 25 universities (including five universities each from northern, western, southern, eastern, and central regions of india) were downloaded from their respective websites, and the tlr were recommended for all courses of programs were evaluated (table 3). table 2. statistics of the tlr recommended for bachelor’s program in geology by the ugc. curricula types of the ugc total tlr recommended tlr of foreign origin tlr of indian origin no. % no. % cbcs curriculum 239 189 79.08 50 20.92 locf curriculum 234 169 72.23 65 27.77 total tlrs recommended 473 358 75.69 115 24.31 the analysis of recommended tlr by the ugc for bachelor’s degree program in geology under its cbcs and locf schemes shows that 75.69% of the tlr listed in the syllabi are of foreign origin, and only 24.31% are of indian origin (table 2; fig. 5a). similarly, the analysis of recommended tlr in the syllabi of master’s degree programs in geology of 25 universities of five indian regions namely, northern, western, southern, eastern, and central shows that these regions contain 81.36%, 81.05%, 78.30%, 80.20%, 76.29%, and 18.64%, 18.95%, 21.70%, 19.80%, 23.71% trl of foreign and indian origin, respectively (table 3; fig. 5b). as a whole, the recommended trl of these 25 universities include 79.68% of foreign origin, and only 20.32% of indian origin (table 3; fig. 5b). a few of the moocs in geology such as physical and structural geology, stratigraphy and palaeontology, crystallography and mineralogy, photogeology and remote sensing, metamorphic petrology and thermodynamics, metamorphic petrology, and environmental geology are on offer through the ugc-moocs portal (http://ugcmoocs.inflibnet.ac.in/ ugcmoocs/moocs_courses.php). in addition, some modules related to geological courses namely, igneous and metamorphic petrology, sedimentology and petroleum geology, crystallography and mineralogy, economic geology and mineral resources of india, hydrogeology and engineering geology, remote sensing and geographic information system are available as open tlr at e-pgpathshala portal of the ugc (https://epgp.inflibnet.ac.in/home/ viewsubject?catid=448). cbcs locf cbcs-locf 79.08 72.23 75.69 20.92 27.77 24.31 trl percentage of foreign origin trl percentage of indian origin 5a northen western southern eastern central total trl 81.36 81.05 78.3 80.2 76.29 79.68 18.64 18.95 21.7 19.8 23.71 20.32 trl percentage of foreign origin trl percentage of indian origin 5b figure 5. a. bar-chart showing percentage of the tlr recommended in the curricula issued by the ugc under its cbcs and locf schemes for bachelor’s degree programs in geology. b. bar-chart showing the percentage of tlr recommended in the curricula of 25 universities across india for their master’s degree programs in geology. refer to table 2 for the tlr data. 246 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh overall, data indicate that a majority of the tlr for higher education programs in geology in india are of foreign origin. these resources are costly for learners and may or may not be sufficient to meet all content requirements of the various courses in the programs. geology is a distinctive and field-based science discipline so a majority of its courses are based on the geological setting of the country. therefore, the teaching-learning practice of such courses usually demands tlr of indian origin (wadia, 1916; krishnan, 1949; kumar, 2010; mishra, 2016; shah, 2018). the need to develop tlr of indian origin taking the indian geological setting into account has been highlighted by various workers (wadia, 1919; mishra, 2016; kumar, 2010; verma, 2017, 2018, 2019, 2021; shah, 2018). the recent development of tlr in the form of moocs and modules of e-pgpathshala of the ugc are progressive steps taken to enrich tlr of the indian origin for some geology courses, but these resources are still meagre and need to be added extensively. table 3. statistics of the tlr recommended for master’s program in geology covering all parts of india. the letters a to y have been sequentially assigned to these universities within parentheses for identification purposes as their names have been kept anonymous. s. no. university selected for accessing curricula / syllabi total trl recommended tlr of foreign origin tlr of indian origin region of india no. % no. % 1. union territory university (a) 274 218 79.56 56 20.44 northern 2. union territory university (b) 151 128 84.76 23 15.24 3. state university (c) 185 139 75.14 46 24.86 4. state university (d) 240 197 82.10 43 17.90 5. central university (e) 202 174 86.14 28 13.86 subtotal of trl 1052 856 81.36 196 18.64 6. state university (f) 222 191 86.10 31 13.90 western 7. state university (g) 275 221 80.36 54 19.64 8. state university (h) 123 97 78.86 26 21.14 9. state university (i) 200 164 82.00 36 18.00 10. state university (j) 394 311 78.94 83 21.06 subtotal of trl 1214 984 81.05 230 18.95 11. central university (k) 150 122 81.34 28 18.66 southern 12. state university (l) 187 150 80.21 37 19.79 13. state university (m) 287 210 73.17 77 26.83 14. state university (n) 156 131 83.97 25 16.03 15. state university (o) 192 148 77.08 44 22.92 subtotal of trl 972 761 78.30 211 21.70 16. state university (p) 148 114 77.03 34 22.97 eastern 17. central university (q) 182 149 81.87 33 18.13 18. central university (r) 267 210 78.65 57 21.35 19. central university (s) 249 206 82.73 43 17.27 20. state university (t) 169 135 79.88 34 20.12 subtotal of trl 1015 814 80.20 201 19.80 21. state university (u) 121 96 79.34 25 20.66 central 22. central university (v) 172 139 80.82 33 19.18 23. state university (w) 145 114 78.62 31 21.38 24. central university (x) 167 109 65.27 58 34.73 25. state university (y) 171 134 78.36 37 21.64 subtotal of trl 776 592 76.29 184 23.71 total of trl 5029 4007 79.68 1022 20.32 covid-19 pandemic and geological higher education: challenges and opportunities the corona virus disease 2019 (covid-19) was declared a global pandemic by the world health organisation (who) in march 2020 (who, 2020) and recognized as a health, economic, and social crisis. it can be viewed as both a challenge and opportunity as far as geological education and research are concerned. due to the covid-19 outbreak, all educational institutions including schools, colleges, and universities in india were closed from 24th march 2020 due to lockdown implementation across the country by the goi, which opened later on in a phased manner (https://www.mohfw.gov.in/). the months-long closure of educational institutions in 2020-2021 had stopped classroom teaching and thus, badly hampered teaching-learning processes whilst 247 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic disengaging learners from active learning. it impacted the mental health of both learners and teachers. in order to provide uninterrupted learning to learners, the ugc and mhrd advised higher learning institutes to make extensive use of the ict to deliver online higher education to learners (ugc, 2020). it also advised learners to use the oer available on various online platforms such as swayam online courses, e-pgpathshala, etc. to continue their learning. covid-19 has had at least two impacts on higher geological education, firstly, the suspension of regular classroom teaching, and secondly, not allowing learners to undertake geological fieldwork in out-door environments because it involves moving learners between sites and use of public transport in groups. the loss of classroom teaching can be substituted by the use of online tlr, although the value of online tlr for particular courses or programs needs to be considered. however, geological fieldwork is one of the distinctive and compulsory components of both bachelor’s and master’s degree programs in geology. it involves out-door visits of learners where they study rock outcrops exposed in their natural settings, for example, in quarries or mines, road cuttings, hilly and mountainous terrains, and along the river valleys. in the field, the learners carefully observe, record, and measure various geological features. they learn to identify various rock-types, fossils, structural features (fold, faults, joints, unconformities), and techniques for collecting samples in the field (lisle et al., 2011). in addition, they also learn how to measure dip, strike, and thickness of rock units and structural features, and how to document field data in field diaries. the overall purpose of geological fieldwork is to develop field-based geological skills and understanding in the learners (lisle et al., 2011). as geology is a field-based science, a degree without having field-related skills is of no value to learners. therefore, learners of degree programs in geology in one region of the country normally visit other regions to undertake geological fieldwork and develop sound field-based geological skills, often traveling distances of hundreds of kilometers. to summarize, the major challenges posed by covid-19 to the geological teaching community are: (a) to evaluate whether the currently available online tlr are sufficient to meet the current and future demands or whether more tlrs are required, and (b) if, it becomes impossible to arrange geological fieldwork in the coming two years due to travel restrictions because of the extended covid-19 infection, what would be the best possible solution to the problem might be. oer and their role in geological higher education the global teaching-learning community has come to understand the real value of the oer at a time when covid-19 has led to the sudden closure of educational institutions (huang et al., 2020). in order to meet educational challenges posed by the covid-19 outbreak, many countries, including india, advised institutions to adopt a policy of undisrupted learning by offering online, remote, and distance teaching (huang et al., 2020; ugc, 2020). in this context, oer emerged as ideal educational resources because they are in digital format, freely accessible at anytime, anywhere, and easy to incorporate into the existing learning system of any educational institution (smith, 2009). the oer had been a part of the education system since the advent of computer and internet technology around the 1990s. oer are “teaching, learning, and research materials in any medium – digital or otherwise – that reside in the public domain or have been released under an open license that permits no-cost access, use, adaptation, and redistribution by others with no or limited restrictions” (https:// hewlett.org/strategy/open-educational-resources/). there are numerous types of oer comprising course materials, full courses, modules, encyclopedias, text-books, games, lesson plans, syllabi, quizzes, instructional modules, model tests, assignments, practical modules, educational audios/videos, virtual field trips, virtual field experiences, and virtual science laboratories (smith, 2009; waldrop, 2013). in addition, the oer also contain rich teaching resources, including methods and tools of teaching that led to improved pedagogical skills, growth, and professional development of teachers. there are several higher education organizations/societies around the world that have been developing oer in geological education and offering them free online in various formats like pdf, html, videos, etc. for the benefit of the global teaching-learning community (table 4). table 4. name and web link of educational institutions and societies offering geological oer s.no. name educational institution /society/ organization web link of geological oer 1. humboldt state university https://libguides.humboldt.edu/openedu/geol 2. university of houston-victoria https://library.uhv.edu/oer/geology 3. washtenaw community college https://libguides.wccnet.edu/oer-subjects/geology 4. mt san antonio college https://mtsac.libguides.com/oer/earth-science 5. university of kentucky https://libguides.uky.edu/c.php?g=222941&p=6722485 6. austin community college https://researchguides.austincc.edu/c.php?g=434670&p=5818520 7. east tennessee state university https://libraries.etsu.edu/research/guides/geosciences/oer 8. salt lake community college https://opengeology.org/textbook/ 9. san bernardino valley college https://www.valleycollege.edu/open-education-resources/faculty/geology.php, 10. creative commons https://www.oercommons.org/browse?f.search=geology 11. the open university https://www.open.edu/openlearn/science-maths-technology/science/geology 12. university of edinburgh https://open.ed.ac.uk/tag/geoscience/ 13. international geoscience education organization http://www.igeoscied.org/teaching-resources/ 14. geological society https://www.geolsoc.org.uk/heresources 15. british geological survey https://www.bgs.ac.uk/discoveringgeology/newsandevents/links.html 16. us geological survey https://www.usgs.gov/science-support/osqi/yes/resources-teachers 17. society of vertebrate palaeontology http://vertpaleo.org/what-is-vertebrate-paleontology/resources-foreducators-and-students.aspx 248 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh many higher learning institutes in india had both typed and handwritten specific-course-based study material uploaded on their websites to provide continuity of learning to their learners during the covid-19 crisis. the goi has previously noted the potential of oer to offer high-quality and equitable education across the country (goi, 2009). the covid-19 crisis prompted the geology teaching-learning community to appreciate the potential of oer in higher education. as a consequence, a few oer in geology are available at the e-pgpathshala portal of the ugc (https://epgp.inflibnet.ac.in/). laboratory experiments and fieldwork are very important components in the successful completion of any degree program of higher geological education. in this regard, oer consisting of laboratory experiments, virtual geological laboratories, virtual field experiences, and geological field modules can offer hands-on training to learners in geological education. for example, google earth images have been extensively used as resources for teaching geological principles as well as for mapping, identifying, and interpreting geological features since 2005 (lisle, 2006; monet & greene, 2012). the indian lithospheric plate contains numerous geological features notably; geomorphic, tectonic, structural, stratigraphic, and petrologic that can be examined by utilizing freely downloadable high-resolution satellite images from the google earth program (http://earth.google.com). oer consisting of geological virtual fieldwork and modules of field-specific problems based on indian outcrops using images of google earth were being developed, through which learners of bachelor’s and master’s degree programs in geology may complete their compulsory fieldwork-based assignments in the light of the covid-19 outbreak. oer usefulness in higher geological education: data collection and interpretation a structured questionnaire was prepared to investigate the value of oer in higher geological education in india and their role in the covid-19 response efforts. the questionnaire was divided into three parts: personal data, educational resources in geological education, and, the use and need of oer in the higher indian geological education system. all questions of the questionnaire were simple, easy to understand, and consisted mainly of multiple-choice options with a few short answer types. part a: personal data provided information regarding age, gender, educational qualification, profession, affiliation, teaching experience, and mode of teaching before the covid-19 pandemic of the respondents. part b: use of educational resources in geological education was designed to get information from respondents about the tlr of indian and foreign origin that they had been using before and during the covid-19. part c: oer contained questions related to familiarity of the oer to the faculty, changes in their use before and during the covid-19, and their potential to meet the requirement of compulsory components (fieldwork and laboratory) of higher degree programs. the consolidated list of all indian universities was obtained from the website of the ugc (https://www.ugc.ac.in/#) and one by one, the websites of all the universities were visited to find whether or not the university had a department of geology/applied geology/earth science/geological sciences, and further e-mail addresses of the existing teaching faculty of the departments were collected. the questionnaire was prepared using the google forms service of google docs (https://www.google.com/forms/about/) and sent online to the collected e-mail addresses of almost all universities faculty of geology in the entire country. additionally, the questionnaire was also sent to the geological teaching faculty of indian institutes of technology, indian institutes of science education and research, a few geologists of the gsi, and the indian statistical institute. the questionnaire was sent to a total of 790 people, from which, 63 responses were received covering almost all parts of the country (fig. 6). the data was collected between july and october 2020. a majority of responses were complete and usable for interpretation. we followed braun & clarke’s (2006) six-step approach for undertaking thematic analysis of questionnaires received that involves becoming familiar with data, generating initial codes, searching themes, reviewing themes, defining themes, and, finally, undertaking interpretation. figure 6. map of india showing the distribution of places and response numbers in parenthesis from where the completed questionnaires were received. 3. descriptive statistics analysis part a: personal data the analyzed sample shows that out of the total sample number (n = 63), the majority (i.e. 81%) of respondents hold doctorate degrees (ph.d.) and a small group of just 17% holds master’s degrees. 83% (52) of respondents were males and only 17% (11) were females. the mean age of respondents was 44 years. they had a wide range of teaching experience and had been working at various positions viz., assistant professors (51%), professors (19%), associate professors (6%), researchers (5%), and geoscientists (5%) in descending order. a majority of respondents were at the beginning or in the middle of their teaching career, as is apparent from their teaching experience statistics: 0-5 years (28%), 5-10 years (23%), 10-15 years (15%) and 15-20 years (15%). two (3%) respondents were not involved in teaching and the remaining 61 (97%) were directly involved in bachelor’s and/or master’s level teaching. part b: educational resources in geological education, use of tlr of indian and foreign origin before and during the covid-19 before the covid-19 pandemic, a majority of respondents (84%) were involved in face-to-face teaching and a minor proportion was also involved in online teaching (16%). it is inferred that out of the 61 respondents who responded to this item, the institutes/universities of almost one-third of the respondents (27%) did not provide any support and training in shifting from face-to-face to online modes of teaching. however, the institutes/universities of the remaining respondents, who were involved in face-to-face teaching before the pandemic, did provide support and training in moving to online modes of teaching. such institutions had the advantage of being able to provide undisrupted learning during the pandemic. 249 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic the respondents, in assessing the quality of the tlr, showed more positivity towards the foreign origin resources (62%) in terms of cost, quality and availability, in comparison to the indian origin resources (38%). however, they were of the view that the tlr of indian origin are slightly more comprehensible than that of foreign origin. they thought that more preference should be given to the tlr of indian origin (75%) in comparison to those of foreign origin (67%). moreover, the majority (77%) declared that they found some shortage in the tlr of indian origin and almost all the respondents (94%) thought that more courses-based tlr incorporating indian geological context are needed. almost one-third (28%) of the respondents commented that the tlr of indian origin are not up-to-date. the questionnaire tested the change in the usage of tlr for teaching before and during covid-19 (table 5). the data showed a significant decrease in the use of hand-written content, followed by real field trips and real science laboratory work. however, typed learning content continued to be used almost in the same way as it was used before and during covid-19. during the pandemic, the use of online platforms increased significantly from 26% to 66%. table 5. kind and percentage of tlr used before and after the covid-19 kind of tlr before covid-19 during covid-19 hand written contents 61% 29% typed contents 65% 66% typed containing all learning requirements and plagiarism free with no copyright issue 47% 42% real science laboratory activities 55% 6% real field trips 69% 3% online platforms like swayam, e-pgpathshala, and other online platforms 26% 66% part c: oer use before and during the covid-19 the respondents reported that they were quite well informed about oer. around 81% had heard about oer before the survey, and nearly 61% had previously used them in the teaching-learning process. they (79%) had informed their students about the existence of the oer, and more than half (65%) were familiar with the appropriate geological oer, while 63% were acquainted with the oer of indian origin. a majority of respondents reported that they were aware of indian oer offering online platforms such as nptel (https://nptel.ac.in/), swayam (https://swayam.gov.in/), e-pgpathshala (https:// epgp.inflibnet.ac.in/), and others. additionally, they were also aware of several foreign geological oer offering online platforms. the data also revealed that respondents were using various types of oer such as open textbooks, educational audios and videos, lesson plans, modules, and practical modules both before and during the covid-19 pandemic. the majority of respondents were of the view that oer could work well for learners pursuing bachelor’s (70%) and even more, for master’s (78%) degree programs in geology. even before the covid-19 pandemic, there was a quite wide variety of oer that had been used in teaching-learning practice; the most commonly used were open textbooks and videos available online (table 6). there was a slight decrease in the use of open textbooks during covid-19 and an increase in the use of online audio and videos, especially of indian origin increasing from 12% to 28% (table 6). the use of online lesson plans and assignments for students remained almost the same. it was anticipated that there might have been an increase in the use of virtual field trips or virtual science lab oer, but this is not the case, at least not according to the data so far collected. this may reflect the non-availability of virtual field trips or virtual science lab oer of indian origin. table 6. trends in use of indian and foreign oer before and during the covid-19 oer types indian oer foreign oer before covid-19 during covid-19 before covid-19 during covid-19 open textbooks 79% 56% 66% 48% online videos 67% 77% 69% 83% online audios 12% 28% 26% 31% online lesson plans 34% 30% 19% 26% modules 24% 28% 24% 19% complete courses 24% 16% 17% 12% moocs 29% 37% 22% 24% practical modules 24% 16% 12% 17% virtual science laboratory 16% 7% 12% 14% virtual field trips 16% 8% 10% 10% assignment for students 38% 38% 25% 21% thematic analysis indicates that the respondents thought that virtual open laboratory and field-based manuals could enhance the practical and field skills of learners to a reasonable extent. in fact, they (35%) were also of the view that virtual field trips based on google earth programs and, identification and measurement of various geological features by using high-resolution images, was a good alternative for completing compulsory geological fieldwork assignment for learners during the covid-19 pandemic; 48% believed this at least to some extent. as far as practical components of the curriculum of programs are concerned, 35% of respondents realized that open laboratory manuals and virtual open labs were a good alternative, whilst nearly 48% thought this alternative was useful to a certain extent only. a small group consisting of 11% of respondents were quite negative about the potential use of virtual open laboratory, open field-based manuals, and open virtual field trips based on google earth. overall, 20% of respondents were enthusiastic about exploring all the possibilities of the oer related to field and laboratory components. meanwhile, more than one-third of the respondents (37%) were positive about the availability of indian origin oer for teaching geology. the majority of respondents (75%) thought that it is quite important to use oer of either indian or foreign origin, where content is enriched with indian geological examples. almost two-thirds (65%) feel that learners would benefit through the use of oer and that they have great potential to contribute to equitable and high-quality geological education across the country. when they were asked through the questionnaire what they had planned for the completion of compulsory geological fieldwork assignment during the covid-19 pandemic, almost one-third (31%) hoped to be able to carry out the fieldwork 250 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh after the end of the covid-19 pandemic, around 20% had no plan, and 10% would like to wait until the end of the pandemic, and in the meantime, they will use online resources. the remaining one-third (39%) planned to use the online training and virtual field geology programs available from professional organizations and webinars for this purpose. to complete the laboratory component of the curriculum during the covid-19 pandemic, around one-third (34%) plan to complete it after the end of the covid-19 pandemic, almost another third (24%) plan to complete the laboratory component of the curriculum by maintaining social distancing, using precautionary measures, and calling students in small groups to the laboratory, 15% had no plan, and the remaining 27% will think about the use of oers and online video demonstrations for completing the activity. finally, in the comment section of the questionnaire, respondents were of the view that geology is a field-based science and geological fieldwork is therefore vital for imparting real field skills to the learners. however, oer, particularly course-specific, open lab manuals, open field-based manuals, virtual field experiences, and virtual field trips based on the google earth program related to the indian geological context, should be developed in almost all branches of geology. overall, the questionnaire data indicates that oer will enhance understanding of geology and increase the skill of the learners, which will reduce the paucity of tlr. further, they will help in providing equitable and high-quality higher geological education and make geological education more flexible by providing a new teaching-learning experience with the potential to tackle any future covid-19-like situation. 4. conclusions geological education and research are the basic building blocks of the socio-economic growth and development of any country. geologically, india is a unique country, because of the huge amount of georesources it contains. the mineral resource richness of the country was first fully understood by the british east india company in 1835. they established the world’s secondlargest organization, the gsi, in 1851 in india, to prospect for mineral resources. after a long time gap, geology emerged as an independent science education discipline in the 1890s as a by-product of the gsi development. in the early nineteenth century, only a few universities offered higher education degree programs in geology. then, the number of institutes offering higher education programs in geology substantially expanded in the late nineteenth to the early twentieth century. however, geology as a science discipline within higher education is still less popular than other science disciplines in india. further, indian higher education in geology still relies on costly and less accessible teaching-learning resources. covid-19 has hit geological higher education badly by disrupting faceto-face teaching-learning processes and preventing learners from undertaking compulsory geological fieldwork assignments in their programs. in this context, open educational resources, which are course-based or chapter-specific free digital resources available in various downloadable formats (like pdf), have provided a suitable option for providing undisrupted, online, and remote learning, in overcoming many of the educational challenges posed by the covid-19 pandemic. open educational resources policies are valued by the government of india and many educational institutions are implementing them. they can play a critical role in offering high-quality and equitable education. the availability of these resources in higher geological education is poor and not sufficient to meet the current demands of the teaching-learning community of the country. the questionnaire data show that there is an urgent need for open educational resources in all branches of geology especially focused on course-specific modules, open lab manuals, open field-based manuals, virtual field experiences, and virtual field trips based on the google earth program. the latter should be based on indian geological contexts to provide effective teaching-learning practices in geological education. therefore, it is critical to design and develops more open educational resources in geology to meet the current demand of the country in terms of indigenous learning as well as to face any future covid-19-like challenges. acknowledgments the authors are highly thankful to the geological professionals (teachers and scientists) for their participation in the online survey and for offering their valuable views. we are highly thankful to late prof. c. king for offering initial comments on the manuscript. the authors also sincerely thank five anonymous reviewers, prof. m.a. malik, and prof. alexander caneva (editor-in-chief) for constructive criticism. dedication the authors are pleased to dedicate this paper to late prof. chris king. he was a professor and director of earth science education unit, keele university, united kingdom. prof. king was deeply involved in the establishment and development of the international geoscience organization and in organizing the first international conference in geoscience education. he was a great promoter of geological education across the globe and immensely contributed to the field of geological education by designing and developing several teaching-learning resources till his last breath. references adam, d. (2020). special report: the simulations driving the world’s response to covid-19. nature, 580, 316–318. https://doi.org/10.1038/d41586020-01003-6 adams, f. d. (1938). the birth and development of the geological sciences. williams and wilkins, baltimore. ali, s., fakhri, y., golbini, m., thakur, s. k., alinejad, a., parseh, i., shekhar, s., & bhattacharya, p. (2009). concentration of fluoride in groundwater of india: a systematic review, meta-analysis and risk assessment. groundwater for sustainable development, 9, 100224. https://doi.org/10.1016/j. gsd.2019.100224 ali, s., shekhar, s., chandrasekhar, t., yadav, a. k., arora, n. k., kashyap, c. a., bhattacharya, p., rai, s. p., pande, p., & chandrasekharam, d. (2021). influence of the water–sediment interaction on the major ions chemistry and fluoride pollution in groundwater of the older alluvial plains of delhi, india. journal of earth system science, 130(2), 1–16. https://doi. org/10.1007/s12040-021-01585-3 ali, s., thakur, s. k., sarkar, a., & shekhar, s. (2016). worldwide contamination of water by fluoride.  environmental chemistry letters,  14(3), 291–315. https://doi.org/10.1007/s10311-016-0563-5 ayala, f. j. (2008) science, evolution, and creationism. proceedings of the national academy of sciences of the united states of america, 105(1), 3–4. https:// doi.org/10.1073/pnas.0711608105 babu, m. s. u., & nautiyal, s. (2015). historical issues and perspectives of land resource management in india: a review. the institute for social and economic change, bangalore. http://www.isec.ac.in/wp%20309%20-%20 umesh%20babu%20and%20sunil%20 nautiyal_final.pdf (last accessed april 2021) beane, j. e., turner, c. a., hooper, p. r., subbarao, k. v., & walsh, j. n. (1986). stratigraphy, composition and form of the deccan basalts, western ghats, india. bulletin of volcanology, 48, 61–83. https://doi.org/10.1007/ bf01073513 bhattacharya, b. b. (2016). geoscience education – experiences and future perspective. in: k. sain & a.s.s.s.r.s. prasad (editors). geoscience for sustainability. indian geophysical union, national geophysical research institute, hyderabad, india, 14–15. bisht, h., rani, m., kumar, k., sah, s., & arya, p. c. (2019). retreating rate of chaturangi glacier, garhwal himalaya, india derived from kinematic gps survey and satellite data. current science, 1169(2), 304–311. doi: 10.18520/cs/v116/i2/304-311 biswas, s. k. (1982). rift basins in western margin of india and their hydrocarbon prospects with special reference to kutch basin. american association of petroleum geologists bulletin, 66, 1497–1513. https://doi. org/10.1306/2f919c1a-16ce-11d7-8645000102c1865d biswas, s. k. (1987). regional tectonic framework, structure and evolution of the western marginal basins of india. tectonophysics, 135, 307–327. https:// doi.org/10.1016/0040-1951(87)90115-6 https://doi.org/10.1038/d41586-020-01003-6 https://doi.org/10.1038/d41586-020-01003-6 https://doi.org/10.1016/j.gsd.2019.100224 https://doi.org/10.1016/j.gsd.2019.100224 https://doi.org/10.1007/s12040-021-01585-3 https://doi.org/10.1007/s12040-021-01585-3 https://doi.org/10.1007/s10311-016-0563-5 https://doi.org/10.1073/pnas.0711608105 https://doi.org/10.1073/pnas.0711608105 https://doi.org/10.1007/bf01073513 https://doi.org/10.1007/bf01073513 https://doi.org/10.1306/2f919c1a-16ce-11d7-8645000102c1865d https://doi.org/10.1306/2f919c1a-16ce-11d7-8645000102c1865d https://doi.org/10.1016/0040-1951(87)90115-6 https://doi.org/10.1016/0040-1951(87)90115-6 251 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic biswas, s. k., & chauhan, g. (2021). the kutch rift basin–potential location for a national geopark of india. a monograph on potential geoparks of india. indian national trust for art and cultural heritage (intach). new delhi, 1–36. braun, v., & clarke, v. (2006). using thematic analysis in psychology. qualitative research in psychology, 3, 77–101. http://dx.doi.org/10.1191/ 1478088706qp063oa chakraborty, a. r. (1970). scatter of indian geological documents in indian and foreign journals. annals of library science and documentation, 17(1–2), 47–53. chakraborty, a. r. (1971). early attempts of documentation of geological literature in indiaa review. annals of library science and documentation, 18, 132–140. chakraborti, d., das, b., & murrill, m. t. (2011). examining india’s groundwater quality management. environmental science & technology, 45, 27–33. https://doi.org/10.1021/es101695d chauhan, g., biswas, s. k., thakkar, m. g., & page, k. n. (2021). the unique geoheritage of the kachchh (kutch) basin, western india, and its conservation.  geoheritage,  13(1), 1–34. https://doi.org/10.1007/s12371021-00535-1 courtillot, v.  (1990). deccan volcanism at the cretaceous-tertiary boundary: past climatic crises as a key to the future? palaeogeography, palaeoclimatology, palaeoecology, 89(3), 291–299. https://doi.org/10.1016/09218181(90)90025-8 das, p. k. (2015). the monsoon. national book trust, new delhi, india. dimri, a. p., yasunari, t. y., kotlia, b. s., mohanty, u. c., & sikka, d. r. (2016). indian winter monsoon: present and past. earth-science review, 163, 297–322. https://doi.org/10.1016/j.earscirev.2016.10.008 dolphin, g., dutchak, a., karchewski, b., & cooper, j. (2019). virtual field experiences in introductory geology: addressing a capacity problem, but finding a pedagogical one. journal of geoscience education, 67(2), 114– 130. https://doi.org/10.1080/10899995.2018.1547034 fareeduddin, pant, n. c., gupta, s., chakraborty, p., sensarma, s., jain, a. k., prasad, g. v. r., srivastava, p., rajan, s., & tiwari, v. m., (2020). the geodynamic evolution of the indian subcontinentan introduction. episodes, 43(1), 7–18. https://doi.org/10.18814/epiiugs/2020/020001 field, r. m. (1941). geophysics and world affairs: a plea for geoscience. eos, 22, 225–234. fox, c. s. (1947). the geological survey of india, 1846 to 1947. nature, 4078, 889–891. frodeman, r. (1995). geological reasoning: geology as an interpretive and historical science. geological society of america bulletin, 107(8), 960–968. froude, m. j., & petley, d. n. (2018). global fatal landslide occurrence from 2004 to 2016. natural hazards and earth system sciences, 18, 2161–2181. https://doi.org/10.5194/nhess-18-2161-2018 government of india (2009). national knowledge commission: report to the nation 2006-2009. national knowledge commission, government of india, new delhi. government of india (2012). report of the committee for review and restructuring of the functions and role of the indian bureau of mines. ministry of mines, government of india. https://ibm.gov.in/index.php?c=pages&m=index&id=366 (last accessed april 2021) gsa position statement (2016). expanding and improving geoscience in higher education. https://www.geosociety.org/documents/gsa/positions/ pos18_highered.pdf (last accessed april 2021) geological survey of india (2011). detailed information dossier on diamond in india. geological survey of india, kolkata, india. https://employee.gsi. gov.in/cs/groups/public/documents/document/b3zp/mdyy/~edisp/ dcport1gsigovi062248.pdf (last accessed april 2021) grout, a. (1995). geology and india, 1770-1851. a study in the methods and motivations of a colonial science. phd dissertation, london university, london. häggquist, e., & söderholm, p. (2015). the economic value of geological information: synthesis and directions for future research. resources policy, 43, 91–100. https://doi.org/10.1016/j.resourpol.2014.11.001 henderson, s., & ostashewski, n. (2018). barriers, incentives, and benefits of the open educational resources (oer) movement: an exploration into instructor perspectives.  first monday,  23(12). https://doi.org/10.5210/ fm.v23i12.9172 howarth, r. j. (2020). etymology in the earth sciences: from ‘geologia’ to ‘geoscience’. earth sciences history, 39(1), 1–27. doi: 10.17704/1944-618739.1.1 huang, r., liu, d., tlili, a., knyazeva, s., chang, t. w., zhang, x., burgos, d., jemni, m., zhang, m., zhuang, r., & holotescu, c. (2020). guidance on open educational practices during school closures: utilizing oer under covid-19 pandemic in line with unesco oer recommendation. smart learning institute of beijing normal university, beijing.https:// iite.unesco.org/wp-content/uploads/2020/05/guidance-on-open-educational-practices-during-school-closures-english-version-v1_0.pdf (last accessed april 2021) indian national trust for art and cultural heritage. (2016). a monograph on national geoheritage monuments of india. indian national trust for art and cultural heritage, natural heritage division, new delhi. http://naturalheritage.intach.org/geoheritage-publications/ jain, a. k., dasgupta, s., bhargava, o. n., israil, m., perumal, r. j., patel, r. c., mukul, m., parcha, s. k., adlakha, v., agarwal, k. k., singh, p., bhattacharyya, k., pant, n. c., & banerjee, d. m. (2016). tectonics and evolution of the himalaya. proceedings of the indian national science academy, 82(3), 581–604. jain, v., & sinha, r. (2005). response of active tectonics on the alluvial baghmati river, himalayan foreland basin, eastern india. geomorphology, 70(3–4), 339–356. https://doi.org/10.1016/j.geomorph.2005.02.012 kania, s., khosla, a., verma, o., & prashanth, m. (2022). charophyte assemblage in the cretaceous–palaeogene boundary from chhindwara district (madhya pradesh), central india. himalayan geology, 43(2), 383–396. khosla, a., lucas, s. g., kania, s., & verma, o. (2022). charophytes from the cretaceous–palaeogene transition in the jhilmili intertrappean beds of central india. geological journal, 1-27. https://doi.org/10.1002/gj.4528 king, c. (2008). geoscience education: an overview. studies in science education, 44(2), 187–222. https://doi.org/10.1080/03057260802264289 king, c. (2018). earth science education across the globe. m.l.m., hernandez (editor). geoscience in primary and secondary education. united nations educational, scientific and cultural organization, france, 16–33. king, c. (2019). exploring geoscience across the globe. international geoscience education organization. http://www.igeoscied.org/wp-content/ uploads/2019/12/geotextbook_dec_2019.pdf (last accessed april 2021) kölbl-ebert, m. (2009). geology and religion: a history of harmony and hostility. geological society london special publication, 310(1), 1–357. kouser, b., bala, a., verma, o., prashanth, m., khosla, a., & pir, r. a. (2022). hydrochemistry for  the  assessment of  groundwater quality in  the  kathua region, jammu and kashmir, india. applied water science, 12:143. https://doi.org/10.1007/s13201-022-01673-9 krishnan, m. s. (1949). geology of india and burma. the madras law journal office, madras, india. kumar, d. (1982). economic compulsions and the geological survey of india. indian journal of history of science, 17(2), 289–300. kumar, d. (1995). science and the raj, 1857-1905. oxford university press, new delhi. kumar, r. (2010). fundamentals of historical geology and stratigraphy of india. new age international publishers, new delhi. http://dx.doi.org/10.1191/1478088706qp063oa http://dx.doi.org/10.1191/1478088706qp063oa https://doi.org/10.1021/es101695d https://doi.org/10.1007/s12371-021-00535-1 https://doi.org/10.1007/s12371-021-00535-1 https://doi.org/10.1016/0921-8181(90)90025-8 https://doi.org/10.1016/0921-8181(90)90025-8 https://doi.org/10.1016/j.earscirev.2016.10.008 https://doi.org/10.1080/10899995.2018.1547034 https://doi.org/10.18814/epiiugs/2020/020001 https://doi.org/10.5194/nhess-18-2161-2018 https://doi.org/10.5194/nhess-18-2161-2018 https://doi.org/10.1016/j.resourpol.2014.11.001 https://doi.org/10.5210/fm.v23i12.9172 https://doi.org/10.5210/fm.v23i12.9172 https://doi.org/10.1016/j.geomorph.2005.02.012 https://doi.org/10.1002/gj.4528 https://doi.org/10.1080/03057260802264289 https://doi.org/10.1007/s13201-022-01673-9 252 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh kumari, v., tandon, s. k., kumar, n., & ghatak, a. 2020. epicontinental permian-cretaceous seaways in central india: the debate for the narmada versus godavari rifts for the cretaceous-tertiary incursion. earth-science reviews, 211, 103284. https://doi.org/10.1016/j.earscirev.2020.103284 kundal, p., & humane, s. k. (2007). stratigraphic, paleobiogeograhic and paleoenvironmental significance of mesophyllum, a nongeniculate coralline alga from western kachchh (middle eocene to oligocene), india. in: d. k. sinha (editor). micropaleontology: application in stratigraphy and paleoceanography. narosa publishing house pvt. ltd., new delhi, 145–154. larwood, h. j. g. (1958). science in india before 1850. british journal of educational studies, 7(1), 36–49. lisle, r. j. (2006). google earth: a new geological resource. geology today 22(1), 29–32. https://doi.org/10.1111/j.1365-2451.2006.00546.x lisle, r. j., brabham, p., & barnes, j. w. (2011). basic geological mapping. john wiley & sons, oxford. mandal, n. (2016). current trends of geodynamic research in india: a review of theoretical and experimental studies. proceedings of the indian national science academy, 82(3), 425–434. manduca, c. a., & kastens, k. a. (2012). geoscience and geoscientists: uniquely equipped to study earth. geological society of america special paper, 486, 1–12. https://doi.org/10.1130/2012.2486(01) mazumder, r., de, s., & raju, p. v. s. (2019). archean-paleoproterozoic transition: the indian perspective. in: r. mazumder & p. eriksson (eds.). archean earth processes. earth science-review, 188, 427–440. https://doi.org/10.1016/j.earscirev.2018.11.013 ministry of human resource development report. (2014). all india survey on higher education (2011-2012). department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/ aishe2011-12.pdf (last accessed april 2021) ministry of human resource development report. (2015a). all india survey on higher education (2012-2013). department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/ aishe2012-13.pdf (last accessed april 2021) ministry of human resource development report. (2015b). all india survey on higher education (2013-2014). department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/ aishe2013-14.pdf (last accessed april 2021) ministry of human resource development report. (2016a). all india survey on higher education (2014-2015). department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/ aishe2014-15.pdf (last accessed april 2021) ministry of human resource development report. (2016b). all india survey on higher education (2015-2016). department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/ aishe2015-16.pdf(last accessed april 2021) ministry of human resource development report. (2017). all india survey on higher education (2016-2017). department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/ aishe2016-17.pdf(last accessed april 2021) ministry of human resource development report. (2018). all india survey on higher education 2017-2018. department of higher education, ministry of human resource development, government of india, new delhi. https://mhrd.gov.in/sites/upload_files/mhrd/files/statistics-new/aishe2017-18.pdf (last accessed april 2021) ministry of human resource development report. (2019). all india survey on higher education 2018-2019. department of higher education, ministry of human resource development, government of india, new delhi. http://aishe.nic.in/aishe/viewdocument.action?documentid=262 (last accessed april 2021) mishra, v. p. (2016). geoscience education in india: framework and structure in a digital age. in: k. sain & a.s.s.s.r.s. prasad (eds.). geoscience for sustainability. indian geophysical union, national geophysical research institute, hyderabad, india, 11-13. mohanty, s. (2015). precambrian continent assembly and dispersal events of south indian and east antarctic shields. international geology review, 57(6), 1992–2027. https://doi.org/10.1080/00206814.2015.1048751 monet, j., & greene, t. (2012). using google earth and satellite imagery to foster place-based teaching in an introductory physical geology course. journal of geoscience education, 60, 10–20. https://doi.org/10.5408/10-203.1 mora, g. (2013). the need for geologists in sustainable development. gsa today, 23(12), 36–37. doi: 10.1130/gsatg185gw.1 mukherjee, s. (2019). tectonics and structural geology: indian context. springer nature, switzerland. mukherjee, s., misra, a. a., calves, g., & nemcok, m. (2016). tectonics of the deccan large igneous province. geological society london special publication, 445, 1–363. murty, k. s. (1979). history of geoscience information in india. in: a. p. harvey & j. a. diment (eds.). geoscience information: a state-of-the art review. proceedings of the 1st international conference on geological information, london, 51–60. murty, k. s. (1982). geological sciences in india in the 18th-19th century. indian journal of history of science, 17(1), 164–178. mythili, g., & goedecke, j. (2016). economics of land degradation in india. in: e. nkonya, a. mirzabaec, & j. von braun. (eds.). economics of land degradation and improvement – a global assessment for sustainable development, springer, cham, 431–469. nagendra, r., kamalak kannan, b. v., sen, g., gilbert, h., reddy, a. n., & jaiprakash, b. c. (2011). sequence surfaces and paleobathymetric trends in albian to maastrichtian sediments of ariyalur area, cauvery basin, india. marine and petroleum  geology, 28(4), 895–905. https://doi.org/10.1016/j.marpetgeo.2010.04.002 naqvi, s. m. (2005). geology and evolution of the indian plate (from hadean to holocene – 4 ga to 4 ka). capital publishing company, new delhi. oldham, t. (1856). memoir of the geological survey of india. bengal military orphan press, calcutta, india, 1, 1–309. padmanabhan, g. (2018). why no geology at plus two? the hindu news paper (online). https://www.thehindu.com/education/why-no-geology-inplus-two/article23934680.ece (last accessed april 2021) pal, d. k., bhattacharyya, t., srivastava, p., chandran, p., & ray, s. k. (2009). soils of the indo-gangetic plains: their historical perspective and management. current science, 96, 1193–1202. palit, c. (2013). symbiotic relation between geology and botany – pramatha nath bose and girish chandra bose. indian journal  of  history  of  science, 48(3), 477–484. paliwal, b. s. (2005). geology introduced as an optional subject in schools in rajasthan. journal of the geological society of india, 66, 658. parsons, a. j., hosseini, k., palin, r. m., & sigloch, k. (2020). geological, geophysical and plate kinematic constraints for models of the india-asia collision and the post-triassic central tethys oceans. earth-science review, 208(103084), 1–63. https://doi.org/10.1016/j.earscirev.2020.103084 prashanth, m., kumar, a., dhar, s., verma, o., & sharma, s. (2021). morphometric characterization and prioritization of sub-watersheds for assessing soil erosion susceptibility in the dehar watershed (himachal himalaya), northern india. himalayan geology, 42(2), 345–358. https://doi.org/10.1016/j.earscirev.2020.103284 https://doi.org/10.1016/j.earscirev.2020.103284 https://doi.org/10.1111/j.1365-2451.2006.00546.x https://doi.org/10.1130/2012.2486(01) https://www.sciencedirect.com/science/article/pii/s0012825218301272 https://www.sciencedirect.com/science/article/pii/s0012825218301272 https://doi.org/10.1016/j.earscirev.2018.11.013 https://doi.org/10.1016/j.earscirev.2018.11.013 https://doi.org/10.1080/00206814.2015.1048751 https://doi.org/10.1016/j.marpetgeo.2010.04.002 https://doi.org/10.1016/j.marpetgeo.2010.04.002 https://doi.org/10.1016/j.earscirev.2020.103084 253 geological education scenario in india and role of open educational resources in the light of covid-19 pandemic prashanth, m., kumar, a., dhar, s., verma, o., & gogoi, k. (2022). hypsometric analysis for determining erosion proneness of dehar watershed, himachal himalaya, north india. journal of geosciences research, 7(1), 86–94. prashanth, m., & verma, o. (2022). medical geology: an interdisciplinary approach intended to unfold the issues of natural environment on public health. journal of geosciences research, 7(2), 139–144. press, f. (2008). earth science and society. nature, 451, 301–303. rajendran, c. p. (2019). shifting paradigms: why history matters in geological sciences. current science, 117(6), 927–931. doi: 10.18520/cs/v117/ i6/927-931 rajendran, k., parameswaran, r. m., & rajendran, c. p. (2017). seismotectonic perspectives on the himalayan arc and contiguous areas: inferences from past and recent earthquakes. earth-science review, 173, 1–30. https://doi.org/10.1016/j.earscirev.2017.08.003 raju, n. j. (2017). prevalence of fluorosis in the fluoride enriched groundwater in semi-arid parts of eastern india: geochemistry and health implications. quaternary  international, 443, 265–278. https://doi.org/10.1016/j. quaint.2016.05.028 radhakrishna, b. p. (2005). relevance of geology. journal of the geological society of india, 65, 531–536. sabin, t. p., krishnan, r, vellore, r., priya, p., borgonkar, h. p., singh, b. b., & sagar, a. (2020). climate change over the himalayas. in: r. krishnan, j. sanjay, c. gnanaseelan, m. mujumdar, & a. kulkarni (eds.). assessment of climate change over the indian region. springer, singapore, 207–222. saha, d. (2016). course curriculum in geology and quality of geoscience education in india. in: k. sain, & a.s.s.s.r.s. prasad (eds.). geoscience for sustainability. indian geophysical union, national geophysical research institute, hyderabad, india, 8–10. saha, d. (2018). woes of geoscience education in india. current science, 115(4), 595–596. https://doi.org/10.18520/cs%2fv115%2fi4%2f595-596 saha, d., shekhar, s., ali, s., elango, l., & vittala, s. (2020). recent scientific perspectives on the indian hydrogeology. proceedings of the indian national science academy, 86(1), 459–478. santosh, m. (2012). india’s palaeoproterozoic legacy. in: r. mazumder & d. saha (eds.). palaeoproterozoic of india. geological society london special publication, 263–288. searle, m. p., & treloar, p. j. (2019). introduction to himalayan tectonics: a modern synthesis. geological society london special publication, 483, 1–17. https://doi.org/10.1144/sp483-2019-20 shah, s. k. (2018). historical geology of india. scientific publishers, jodhpur, india. shalley, m. j. (1971). the role of geology and geophysics in mineral exploration. exploration geophysics, 2(2), 38–44. shekhar, s., kumar, p., chauhan, g., & thakkar, m. g. (2019). conservation and sustainable development of geoheritage, geopark, and geotourism: a case study of cenozoic successions of western kutch, india. geoheritage, 11(4), 1475–1488. https://doi.org/10.1007/s12371-019-00362-5 sheth, h. c. (2007). ‘large igneous provinces (lips)’: definition, recommended terminology, and a hierarchical classification. earth-science review, 85(3–4), 117–124. https://doi.org/10.1016/j.earscirev.2007.07.005 sharma, k. k. (1998). geologic and tectonic evolution of the himalaya before and after the india-asia collision. proceedings of the indian academy of sciences (earth and planetary sciences), 107(4), 265–282. smith, m. s. (2009). opening education. science, 323, 89–93. stewart, i. s., & gill, j. c. (2017). social geology — integrating sustainability concepts into earth sciences. proceedings of the geologists’ association, 128(2), 165–172. https://doi.org/10.1016/j.pgeola.2017.01.002 stubblefield, c. j. (1970). darashaw nosherwan wadia 1883-1969. biographical memoirs of fellows of the royal society, 16, 542–562. tandon, s. k. (2005). the science of the earth, and earth science education in india. current science, 88(4), 546–547. tandon, s. k. (2022). the himalayan foreland basinstratigraphic records of continent-to-continent collisional processes. journal of the geological society of india, 98(7), 1015–1015. https://doi.org/10.1007/s12594-022-2110-2 tandon, s. k., pant, c. c., & casshyap, s. m. (1991). sedimentary basins of india: tectonic context. gnyanodaya prakashan, nainital, india. tolman, a. b. (2016). geologists and the british raj, 1870-1910. all graduate theses and dissertations. department of history, utah state university. https://digitalcommons.usu.edu/etd/4989 (last accessed april 2021) university grant commission. (2015a). instructional template for facilitating implementation of choice based credit system (cbcs) (online). https://www.ugc.ac.in/pdfnews/4426331_instructional-template.pdf (last accessed april 2021) university grant commission. (2015b). syllabus for b.sc. (hons.) geology six semester course under choice based credit system. https://www.ugc. ac.in/pdfnews/1853907_b.sc.-hons.-goelogy.pdf (last accessed april 2021) university grant commission. (2015c). proposed syllabus and scheme of examination for b.sc. with geology under choice based credit system. https://www.ugc.ac.in/pdfnews/7834290_b.sc.-with-geology.pdf (last accessed april 2021) university grant commission. (2019). learning outcomes based curriculum framework (locf) for undergraduate programme b.sc. (geology). https://www.ugc.ac.in/pdfnews/4480737_geology.pdf (last accessed april 2021) university grant commission. (2020). let covid 19 not stop you from learningict initiatives of mhrd and ugc. https://www.ugc.ac.in/pdfnews/1573010_on-line-learning---ict-initiatives-of-mhrd-andugc.pdf (last accessed april 2021) united nations (2015). sustainable development goals (online). http://www.un.org/sustainabledevelopment/sustainable-development-goals/. (last accessed april 2021) valdiya, k. s. (2016). the making of india: geodynamic evolution. springer international publishing, switzerland. verma, o. (2015). cretaceous vertebrate fauna of the cauvery basin, southern india: palaeodiversity and palaeobiogeographic implications. palaeogeography, palaeoclimatology, palaeoecology, 431, 53–67. verma, o. (2017). elements of palaeontology. current science, 113(10), 2037. verma, o. (2018) paleoecology: past, present and future. journal of the geological society of india, 91(1): 115–116. https://doi.org/10.1007/s12594018-0843-8 verma, o. (2019). historical geology of india. journal of the palaeontological society of india, 64(1), 43–145. verma, o. (2021). role of open educational resources to support higher geoscience education in india. journal of geosciences research, 6(1), 1–10. verma, o., & khosla, a. (2019) developments in the stratigraphy of the deccan volcanic province, peninsular india. comptes rendus geoscience, 351(7), 461–476. https://doi.org/10.1016/j.crte.2019.10.002 https://doi.org/10.1016/j.earscirev.2017.08.003 https://doi.org/10.1016/j.earscirev.2017.08.003 https://doi.org/10.1016/j.quaint.2016.05.028 https://doi.org/10.1016/j.quaint.2016.05.028 https://doi.org/10.18520/cs%2fv115%2fi4%2f595-596 https://doi.org/10.1144/sp483-2019-20 https://doi.org/10.1007/s12371-019-00362-5 https://doi.org/10.1016/j.earscirev.2007.07.005 https://doi.org/10.1016/j.pgeola.2017.01.002 https://doi.org/10.1007/s12594-022-2110-2 https://doi.org/10.1007/s12594-018-0843-8 https://doi.org/10.1007/s12594-018-0843-8 https://doi.org/10.1016/j.crte.2019.10.002 254 omkar verma, manthena prashanth, roberto greco, ashu khosla, kulwant singh verma, o., khosla, a., goin, f. j., & kaur, j. (2016). historical biogeography of the late cretaceous vertebrates of india: comparison of geophysical and paleontological data. new mexico museum of natural history and science bulletin, 71, 317–330. wadia, d. n. (1919). geology of india. macmillan and co., london. wadia, d. n. (1964). progress of geology. journal of the geological society of india, 5, 1–4. wagh, s. (2018). the british rule and its impact on india. atharva publications, maharashtra, india. waldrop, m. m. (2013). education online: the virtual lab. nature, 499, 268–270. https://doi.org/10.1038/499268a   world health organization (2020). coronavirus disease (covid-19) pandemic. https://www.who.int/emergencies/diseases/novel-coronavirus-2019 (last accessed april 2021) wilson, a. (1883). facts and fictions of zoology. the world’s cyclopedia of science. j.b. alden, new york. 1, 207–272. https://doi.org/10.1038/499268a bau0005 bau0010 bau0015 perez & marfurt.indd earth sci. res. j. vol. 10, no. 2 (december 2006): 147-155 improving vertical and lateral resolution by stretch-free, horizon-oriented imaging gabriel pérez1 and kurt marfurt2, agl, university of houston (1) gabriel pérez: geosciences department. university of houston. houston, texas 77204-5007. geophysics ph.d research assistant. gabriel.perez@mail.uh.edu (2) kurt marfurt: geosciences department. professor, director of agl. university of houston. houston, texas 77204-5007. kmarfurt@uh.edu abstract the pre-stack kirchhoff migration is implemented for delivering wavelet stretch-free imaged data, if the migration is (ideally) limited to the wavelet corresponding to a target horizon. avoiding wavelet stretch provides long-offset imaged data, far beyond what is reached in conventional migration and results in images from the target with improved vertical and lateral resolution and angular illumination. increasing the range of imaged offsets also increases the sensitivity to event-crossing, velocity errors and anisotropy. these issues must be addressed to fully achieve the greatest potential of this technique. these ideas are further illustrated with a land survey seismic data application in texas, u.s. keywords: pre-stack kirchoff migration, seimic data, wavelet stretch-free data, texas. 147 earth sciences research journal manuscript received november 2 2006. accepted for publication december 20 2006. introduction surface seismic data are a major source of information about the subsurface. roughly, this comes in two different ways: through seismic images that illuminate structure and stratigraphy, and from the study of data attributes such as amplitude, arrival time or frequency content that are sensitive to lithology and fluid content. exploration challenges steadily place increasing demands on all factors that impact the usefulness and quality of this information such as frequency content and availability of long-offset data. increasing frequency content is desired to achieve the level of lateral and vertical resolution required for even smaller and/or more elusive targets. increasing the usable range of offsets in the data is desired to support avo studies, anisotropy and velocity analysis, or simply to increase the fold of stacking. unfortunately, these two factors are sometimes counteracting in common seismic exploration practice. specifically, when it comes to imaging, either in conventional nmo or in pre-stack migration, loss of frequency content and wavelet distortion due to stretch is a major problem for far-offset data. typically, imaged data beyond a certain offset, roughly between once to twice the reflection depth, is severely distorted due to stretch that it needs to be discarded by harsh muting. while several methods have been proposed to improving vertical and lateral resolution by stretch-free, horizon-oriented imaging alleviate nmo stretch (dunkin and levin, 1973; rupert and chun, 1975; barnes, 1992), little attention has been devoted to stretch due to migration. in addition, recent approaches attack the problem during the stacking process, thereby resulting in an improved stacked image but not delivering stretch-free pre-stack traces (trickett, 2003). in that case, an increased fold of stack but none of the other desired benefits mentioned above is achieved. recently, wavelet stretch has been recognized as a major adverse factor in avo (swan, 1997; dong, 1999) and increasing attention has been given to develop methods to quantify and correct for its effects. though most approaches focus on the improved estimation of attributes such as avo intercept and gradient or 3-term avo/ava inversion, some directly or indirectly attempt to correct for stretch on the data itself (shatilo and aminzadeh, 2000; castoro et al, 2001; brouwer, 2002; lazaratos and finn, 2004). hilterman and vanschuyver (2003), working in long offset avo, introduced a horizon-oriented pre-stack migration implementation that generates stretch-free pre-stack imaged data. besides from the avo focus, the impact of such horizonoriented stretch-free prestack imaging and the availability of long-offset stretch-free imaged data on image quality and lateral and vertical seismic resolution will be explored, through data from a land survey application in texas, u.s. the impact of such imaging on a land seismic dataset is assessed through the use of multitrace geometric seismic attributes. the attributes computed from pre-stack seismic data may reveal subtle geologic features that are lost in conventional stacked images. in the interest of maximizing the science with a reasonable computational effort, pre-stack time migration has been used as the preferred imaging method. horizon-oriented stretch-free prestack imaging as discussed by levin (1998), wavelet stretch occurs in all pre-stack imaging methods. within the context of kirchhoff migration, this stretch is associated to the variation in the curvature of isochron surfaces for different times on an input seismic trace (figure 1). using the analogy of kirchhoff migration as a moving operation, if “migration paths were somehow parallel”, stretch would be avoided. as suggested in figure 2, it would be possible to “move” a block of input samples along a single path. in this way, relative separation within the samples in the block is not altered so no stretch is introduced. this idea can be actually implemented in practice by computing a traveltime operator that is exact for a single output position and then, migrating a window of the data centered on the computed arrival time, instead of migrating just the single sample matching that time. the central sample is migrated to the “exact” position in the output and, for time migration; surrounding samples are migrated to positions in which the difference in time with respect to the central migrated sample are preserved. in depth migration, a similar outcome can be achieved by scaling the time differences by a factor given by the local value of the migration velocity. this simply amounts to conversion from uniform sampling in time to uniform sampling in depth with scaling given by that factor (see yu et al., 2004 for an application of this idea to wavelet-based depth migration). notice that samples other than the central matching sample are not migrated properly, that is, to the position they would be migrated in conventional migration. in that regard, this approach is only accurate locally, at most in a relatively narrow window around the central matching sample, where the error introduced is small. in the other hand, the migration operation just described makes perfect sense if the matching sample corresponds to a reflector, and instead of just a block of samples operating over the wavelet corresponding to the reflector is considered. in that case, the wavelet is simply kept (with proper scaling to depth, in the case of depth migration) attached to the image of the reflector that carries it. the whole operation amounts to the migration of a broad-band impulse representing the reflector, followed by convolution with the wavelet. using a block of samples is simply a practical way of avoiding the added complication of determination of the wavelet; the size of the block has to be large enough to include the wavelet. regardless of which of the two points of view just discussed is preferred, it is clear that the imaging approach described is limited, be 148 figure. 1. kirchhoff migration schematics and its action on a band-limited wavelet. the data sample whose timing matches the total traveltime from source location to subsurface position and back to receiver location, with proper scaling, is the contribution from the input trace to the image at the given position. in time migration, if the timing within the samples in the wavelet is changed in the output, relative to the input, the wavelet will be distorted accordingly. in depth migration, variations in the separation between output samples, assuming uniform sampling in time in the input data, will also result on differential stretch or squeeze on the wavelet. figure. 2. instead of “moving” individual samples independently, a block of samples is moved along “parallel” paths so that relative separation between samples is not altered, hence no stretch is introduced. ideally the block of samples should include the wavelet associated to the event of interest. this is a modification of the operation shown in figure 1. pérez and marfurt., esrj vol. 10, no. 2. december 2006 149 figure. 3. comparison of imaged gathers pertaining to the same image location, for the: (a) conventional and (b) target-oriented migrations. notice that, by contrast with the conventional migration, the target-oriented migrated data is stretch-free. the event used as a target for the target-oriented migration is shown in blue in both panels. as depicted by the dotted blue line, the desired definition of this event in the longest offsets in the target-oriented migration is obscured by interference from crossing events such as the one marked in yellow. the inferred location of this event on the conventional migrated data is shown with a broken line. cause of either accuracy or scope, to a relatively narrow window around a previously chosen target horizon or target depth. because of this, the migration is limited to a narrow window of input trace samples (usually within 50ms to 200 ms thick) surrounding the originally selected time. similar methods have been developed elsewhere. wyatt et al. (1997) describe an application with emphasis on the efficiency in computation and human intervention provided by imaging limited to a narrow zone around a horizon. our interest arises from the work of hilterman (2004) on improving the frequency content of long-offset data for avo studies, which evolved into a related but different implementation (hilterman and vanschuyver, 2003). the next sections will describe the application of the horizon-oriented migration to the imaging of a field dataset from the fort worth basin, texas, usa. as previously stated, this paper focuses on the impact of the improved frequency content from far-offset imaged data, on image quality, and on lateral and vertical resolution. data processing data was processed twice, initially with a conventional pre-stack time migration and then, with the horizon-oriented stretch-free implementation. the conventional migration work-flow included a migration velocity estimation step. the same velocity field was used for the conventional migration and the horizon-oriented migration. as appropriate for time migration, velocities were computed using a “deregowski-loop” approach where an initial migration was performed with the same velocities used for nmo correction in earlier time processing. using this velocity field, nmo was restored on selected gathers and an additional step of conventional hyperbolic moveout analysis yielded updated velocities, used for the final migration. as a target for the horizon-oriented migration, a horizon interpreted on the stack of the conventional pre-stack migrated data was used as corresponding to the top of one 150 improving vertical and lateral resolution by stretch-free, horizon-oriented imaging figure. 4. comparison between images for a selected inline from the 3d volume for the: (a) conventional and (b) target oriented migrations. images stacked over the full range of offsets available to each migration. the blue arrows in each image point to the location of the event used as a target in the target-oriented migrations. the yellow arrow points to the location of a fault that is better resolved in (b). figure. 5. comparison between images from the conventional and the target oriented migration. images stacked in the far-offset range. once again the fault pointed-to by the yellow arrow is better resolved in the target-oriented image. 151 pérez and marfurt., esrj vol. 10, no. 2. december 2006 figure. 6. comparison of time slices on the full offset range for the: (a) conventional and (b) target oriented migrations. slice is taken at 1.2 seconds, approximately the time of the target horizon. faults trending to the ne can be identified between the yellow arrows in each image. figure. 7. comparison of time slices on the far-offset range for the: (a) conventional and (b) target oriented migrations. time of the slice is 1.2 seconds. 152 improving vertical and lateral resolution by stretch-free, horizon-oriented imaging of the major formations of interest in the area. as seen in figures 4 and 5 this horizon runs roughly between 1.1 and 1.2 seconds; the size of the data block window was 200 ms, centered on the time in the input trace corresponding to the horizon picked time. figure 3 compares the results of the two imaging approaches. in the conventional migration, stretch shows up at the longer offsets, and it is notably stronger at the shallower times (i.e., where reflection angle is relatively large). in contrast, the target-oriented migrated data is stretch-free with nicely aligned reflections for the full range of offsets. at the longer offsets, however, crossing events from earlier and later times interfere with proper imaging of the target events. vertical resolution for the analysis the data was divided in three offset ranges: 0-4000 ft, 4000 – 8000 ft and 8000 – 12500 ft, corresponding to near, middle and far offset ranges. the absolute maximum offset present in the data was around 25000 ft but on average the maximum offset was closer to 15000 ft. the depth of the target horizon lies in the 7000 to 8000 ft range. figures 4 and 5 compare stacked images from the full and far offset ranges for the two migrated datasets. it can be observed that for the fullrange stack the target-oriented migration image is better resolved than the conventional migration. in the far-offset range images in figure 5, the target horizon and the fault zone are more sharply defined by the target-oriented migration. this should be attributed to the combined effect of improved frequency content on the far offsets and increased fold of stack available to the target-oriented images. notice that because of the lower fold, the offset-limited images are of infefigure. 8. comparison between principal component coherence images in the far-offset range stack for the: (a) conventional and (b) target oriented migrations. time of the slice is 1.2 seconds. 153 pérez and marfurt., esrj vol. 10, no. 2. december 2006 154 rior quality compared to those with the full range of offsets. besides, the target-oriented migration images are slightly noisier and more impacted by migration artifacts than the conventional migration images. this is explained because for this dataset, the signal-to-noise ratio is poorer (figure 3) and fold distribution is less uniform in the extended offset range included in the faroffset target-oriented migration. in the very-long offsets (beyond 12500 ft) in the horizon-oriented migration, overcorrected events from the later times in the imaged data window interfere with properly aligned data from earlier times (figure 3). failure to image the data at the very-long offsets is a limitation of the hyperbolic moveout assumption implicit in conventional prestack time migration. inclusion of the overcorrected, not properly imaged events in the stack, results in a distorted and defocused stacked image. avoiding major distortion in the images is the main reason to limit the offsets from the horizon-oriented migrated data to a maximum of 12500 ft, as described above. reduction of the offset range is a drastic and suboptimal solution. a better approach should consider non-hyperbolic moveout in the computation of the operator for the horizon-oriented migration. for the conventional migration, imaging of long-offset data is simply not attempted and, as observed in figure 3, it is severely distorted by the migration. as usual, this data was muted prior to stacking to obtain stacked images from the conventionally migrated data. lateral resolution an improved lateral resolution is fully achieved from the target-oriented migration (figures 6 and 7). just as for the vertical images, the pickcorrected migration is comparatively much less resolved and is not shown. in those figures, the ne-trending faults, indicated by the arrows, are more sharply defined in the target-oriented images for both the far-offset and full offset ranges. though subtle, definition of features like these is potentially very important for the exploration goals in the survey area. a final comparison is made in figure 8 between the conventional and target-oriented migrations at the far-offset range, this time using principalcomponent coherence. principal-component coherence (gersztenkorn and marfurt, 1999) is a multitrace attribute that provides an image of lateral changes in waveform by computing the similarity between neighboring traces in a 3d data volume. multitrace attributes such as coherence may reveal subtle geologic features that are lost in conventional images. these attributes are used as a way to assess the lateral resolution in the migrated images generated in this paper. the ne trending faults and other features pointed to by the arrows, are once again better defined by the target-oriented migration (figure 8). conclusions a block-moveout, horizon-oriented implementation of kirchhoff pre-stack migration effectively delivers stretch-free pre-stack imaged data. imaged data is augmented by long offsets that are discarded by conventional imaging algorithms. this contributes to an improved image quality and larger vertical and lateral resolution. fully achieving the potential of this technique requires that non-hyperbolic moveout is properly accounted for in the migration. the possibility to apply it in a target-oriented fashion has been repeatedly mentioned as one of the strengths of kirchhoff migration. as shown by others before, and reinforced here, if the target is a horizon, kirchhoff migration can be implemented so that it is also stretch-free. this imaging approach provides an increased fold of imaged far-offset data with improved frequency content. this impacts image quality and resolution. in the application of this approach to a 3d dataset from the fort worth basin, good results have been achieved in imaging subtle fault features that are nevertheless important for the exploration targets in the area. finally, moving further in the long-offset range of what is conventionally available, opens up a whole can of worms of new issues that have been swept under the rug or discarded by harsh muting in the conventional approach. improving vertical and lateral resolution by stretch-free, horizon-oriented imaging acknowledgments this work was supported by the sponsors of agl and the state of texas. field data was provided by devon energy. references barnes, a. e. (1992). another look at nmo stretch. geophysics. 57. 749-751. brouwer, j. h. (2002). improved nmo correction with a specific application to shallow-seismic data. geophysical prospecting. 50. 225– 237. castoro, a., white, r. e., and thomas, r. (2001). thin-bed avo: compensating for the effects of nmo on reflectivity sequences. geophysics. 66. 1714-1720. dong, w. (1999). avo detectability against tuning and stretching artifacts. geophysics. 64. 494-503. dunkin, j. w. and levin, f. k. (1973). effect of normal moveout on a seismic pulse: geophysics. 28. 635–642. gersztenkorn, a. and marfurt, k. j. (1999). eigenstructure based coherence computations as an aid to 3d structural and stratigraphic mapping. geophysics. 64. 1468-1479. hilterman, f. (2004). personal communication. hilterman, f. and van schuyver, c. (2003). seismic wide-angle processing to avoid nmo stretch. 73rd annual international meeting: society of exploration geophysicists, expanded abstracts. 215-218. lazaratos, s. and finn, c. (2004). deterministic spectral balancing for high-fidelity avo. 74th annual international meeting, society of exploration geophysicists, expanded abstracts. 219–223. levin, s. a. (1998). resolution in seismic imaging: is it all a matter of perspective? geophysics. 63. 743-749. rupert, g. b. and chun, j. h. (1975). the block move sum normal moveout correction. geophysics. 40. 17–24. shatilo, a. and aminzadeh, f. (2000). constant normal-moveout (cnmo) correction. a technique and test results. geophysical prospecting. 48. 473-488. swan, h. w. (1997). removal of offset-dependent tuning in avo analysis. 67th annual international meeting, society of exploration geophysicists, expanded abstracts. 175–178. trickett, s. (2003). stretch-free stacking. 73rd annual international meeting, society of exploration geophysicists, expanded abstracts. 20082011 wyatt, k. d., valasek, p. a., wyatt, s. b., and heaton, r. m. (1997). velocity and illumination studies from horizon−based psdm, 67th annual international meeting: society of exploration geophysicists, expanded abstracts, 1801−1804. yu, z., mcmechan, g. a., anno, p. d., and ferguson, j. f. (2004). wavelet transform-based prestack multiscale kirchhoff migration. geophysics. 69. 1505-1512. 155 pérez and marfurt., esrj vol. 10, no. 2. december 2006 the purpose of this research is to assess the suitability of dune sands as construction materials, moreover, such a purpose is considered beneficial in assessing appropriate methods for soil stabilization or ground improvement and to assess the suitability of dune sands as subgrade layer for carrying roads and rail foundation. dune sand samples were collected from a region in baiji area in salah-aldeen governorate, north of iraq. a grey-colored densified silica fume (sf) and lime (l) are used. three percentages are used for lime (3%, 6% and 9%) and four percentages are used for silica fume (3%, 6%, 9% and 12%) and the optimum percentage of silica fume is mixed with the percentages of lime. unsoaked cbr on compacted dune sands treated dune sands with l-sf by mixing and cured for a period of 1 day. the increasing in cbr ranged between 443 – 707% at 2.54 mm penetration and 345 – 410% at 5.08 mm penetration. el propósito de esta investigación es evaluar el uso de arena de dunas como materiales de construcción. además, este objetivo permite determinar los métodos apropiados para la estabilización del suelo, el mejoramiento del terreno y la evaluación de pertinencia de la arena de dunas en capas subbase para carreteras y cimientos férreos. se recolectaron muestras de arena de dunas en el área de baiji, del comisionado salah-aldeen, al norte de irak. se utilizó vapor de óxido de silicio (sf, en inglés), grisáceo y densificado, y óxido de calcio (l). se utilizaron tres porcentajes para el óxido de calcio (3 %, 6 % y 9 %), y cuatro para el óxido de silicio (3 %, 6 %, 9% y 12%) y el máximo porcentaje del óxido de silicio se mezcló con las proporciones de óxido de calcio. se realizó en seco el ensayo de relación de soporte de california (del inglés california bearing ratio, cbr) en arena de dunas compactada y tratada con la mezcla l-sf curada durante un día. el incremento en el ensayo cbr osciló entre 443-707 % en la penetración de 2.5 mm y 345-410 % en la penetración de 5.08 mm. abstract resumen keywords: dune sand, lime, silica fume, stabilization, cbr, strength. palabras clave: arena de dunas, óxido de silicio, óxido de calcio, estabilización, ensayo de relación de soporte de california. compaction and collapse characteristics of dune sand stabilized with lime-silica fume mix record manuscript received: 20/05/2015 accepted for publication:13/05/2016 how to cite item fattah, m. y., joni, h. h., & al-dulaimy, a. s. a. (2016). compaction and collapse characteristics of dune sand stabilized with lime-silica fume mix. earth sciences research journal, 20(2), i1-i8. doi:http://dx.doi.org/10.15446/esrj.v20n2.50724 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.50724 mohammed y. fattah1 hasan h. joni2 and ahmed s. a. al-dulaimy3 1 professor, building and construction engineering department, university of technology, baghdad, iraq. 2 assistant professor, building and construction engineering department, university of technology, baghdad, iraq. 3 formerly graduate student, building and construction engineering department, university of technology, baghdad, iraq. email: myf_1968@yahoo.com earth sciences research journal earth sci. res. j. vol. 20, no. 2 ( june, 2016 ) : i1 i8. características de compactación y de colapso de arena de dunas estabilizada con una mezcla de óxido de calcio y óxido de silicio g e o t h e c n ic s i2 mohammed y. fattah, hasan h. joni, and ahmed s. a. al-dulaimy. introduction: ground improvement refers to any procedure undertaken to increase the shear strength, decrease the permeability and compressibility, or otherwise render the physical properties of soil more suitable for projected engineering use. the improvement may be accomplished by drainage, compaction, preloading, reinforcement, and grouting, electrical, chemical or thermal methods. among the various soil stabilization procedures, the most suitable one is selected depending upon the type of soil available, time, cost involved etc. (kumar, 2010). one of the most dramatic depositional landforms, which occupy large areas all over the world, is sand dune. sand dunes in various forms cover large areas of the world. extensive efforts have been made to understand their origin (shearzid, 1989) physical and mineralogical properties, environmental impacts. some studies have also been directed towards dune sand stabilization (al-soud, 2000). there is, however, scarcity in the literature concerning the geotechnical characteristics of these soils. a single dune can be defined as a mound or hill of sand, which rises to a single summit. they are accumulations of windblown sand, which change their position or their shape due to wind action as long as their surface consists of loose granular material of appropriate size. some of typical properties of sand dunes are as follows (das, 2007): 1. grain size distribution of the sands at any particular location is uniform. the uniformity can be attributed to the sorting action of the winds. 2. the general grain size percentages decrease with distance from the source, because the wind carries the small particles farther than the large ones. 3. the relative density of sand deposited on the windward side of the dunes is as high as (50-65) % and decreases to about (0-15) % on the leeward. collapse is a typical feature of unsaturated, loose and low plasticity soils, which are typical features of dune sand. the behavior of collapsible soil is often quantified by conducting a collapse potential test. however, these test interpret the soil behavior in terms of total stress, while it is well known that the behavior of soil is governed by its effective stress. also the collapse is defined as the significant volume reduction observed when wetting an unsaturated sample under load. the collapsing behavior of this type of soil generally occurs when the water is introduced into the soil. collapsing cause damage too many civil engineering structures such as: spread footing, buildings, roads, highways, and earth dams leading to high economic losses. there are many factors affecting collapse behavior of compacted and soils which are: initial dry unit weight, initial water content, percentage of fines, and the method used in compaction, (al-juari, 2009). to measure the collapsing potential of this type of soil, it may use an oedometer/consolidation ring to prepare an undisturbed sample (similar preparation as consolidation test). when water is introduced to this specimen, the sample may start to collapse, ie. change in height(dh/h) or change in initial void ratio (e/eo) may be used as the degree of collapse of the soil. janning and knight discussed the problem in 1957. then, in 1963, they proposed a test method by which they determined the collapse potential for one specimen under an applied vertical stress of 200kpa, after one day of soaking, (albusoda and salem, 2012 and mansour et al., 2008). abbeche et al., 2010 have studied the soil collapse and listed the factors contributing to the occurrence of this phenomenon. the main factors are summarized as follows: 1. an open unstable structure and partially saturated. 2. a high total stress is applied. 3. the presences of cementing agent in the soil, for example calcium carbonate (caco3), in unsaturated conditions contribute to stabilize the soil. a moistening the soil will reduce the suction and destroy the liaison officer. albusoda and salem (2012) used cement kiln dust (ckd) to stabilize affek dune sand the effectiveness of adding different percentages of (ckd) was studied. time of curing effect, was also considered. it was found that stabilization of collapsible soils with ckd can provide tremendous advantages. ading of 8% ckd, for example, increased the ultimate bearing capacity of shallow foundation to 250%. the variation in shear strength parameters became almost constant after fourteen days of curing. vegetation cover on sand dunes mainly depends on wind power (drift potential—dp) and precipitation. when this cover decreases below a minimal percentage, dunes will start moving. it is therefore necessary to study the effects of dp and precipitation on contemporary dune activity in order to predict likely future dune mobility in the coming decades. ashkenazy et al. (2012) concentrated on the future activity of the currently fixed dune fields of the kalahari and the australian deserts. these sand seas include the largest areas of stabilized dunes in the world, and changes in their mobility have significant economic implications. global maps of dp are introduced, based on real and reanalysis data. analyses of two global circulation models (gfdl and cgcm3.1) provide future predictions under the sres-a1b ipcc scenario, which is a moderate global warming scenario. according to the gfdl model, both the australian and kalahari basin dunes will apparently remain stable towards the end of the 21st century because the dp will stay small, while the rate of precipitation was expected to remain much above the minimal threshold necessary for the vegetative growth that leads to dune stabilization. the cgcm model predicts insignificant changes in dps and shows that the precipitation rate is above 500 mm/year for almost the entire kalahari basin. the central-northern part of australia is predicted to have larger dps and greater precipitation than the southern part. since the predicted changes in dp and precipitation are generally not drastic, both the australian desert and kalahari basin dunes are not likely to become active. still, the australian dunes are more likely to remobilize than the kalahari ones due to some decrease in precipitation and an increase in wind power. the research work carried out by bouziani et al. (2012) dealt with the effect of dune sand, available in huge quantities in these regions, on the properties of flowing sand-concrete (fsc) prepared with different proportions of dune and river sands. mini-cone slump test, v-funnel flow-time test and viscosity measurements were used to characterize the behaviour of fsc in fresh state. the 28-day compressive strength was also determined. test results showed that an optimal content of dune sand, which makes satisfied fresh and hardened properties of fsc, is obtained. moreover, the obtained flow index (constant b) calculated by the help of power-law viscosity model is successfully correlated to the experimental results of v-funnel flow time. jackson et al. (2014) stated that seven types of coastal dune are identified on kwazulu-natal coast: estuary barrier dunes; mainland beach foredunes; prograded beachridge dunes; headland bypass dunes; transverse dune fields; parabolic/migrating dunefield; and climbing dunes. the development and distribution of the various dune forms is more strongly controlled by the geological setting, than the aerodynamic regime. ciccarelli (2014) tried to assess the conservation status of coastal dune systems in tuscany (italy). emphasis was given to the presence and abundance of plant communities identified as habitat in accordance with the directive 92/43/eec. twenty transects perpendicular to the shoreline were randomly positioned on the whole coastal area (30 km in length) in order to sample the full spectrum of plant communities. vegetation zonation and relationships with the most frequent disturbance factors in the study area—beach cleaning, coastline erosion, presence of paths and roads, bathing settlements and trampling—were investigated through principal coordinate analysis and canonical correspondence analysis. natural factors, such as distance from the sea and total length, were also considered. differences in the conservation status of the sites were found, ranging from the total disappearance of the foredune habitats to the presence of the complete psammophilous (sand-loving) plant communities. erosion, trampling, and paths were found to be closely correlated with degradation and habitat loss. furthermore, the overall plant species diversity of dunes was measured with nhdune, a modified version of the shannon index; while the incidence of invasive taxa was calculated using n, a naturalness index. however, these diversity indices proved to be a weaker bioindicator of ecosystem integrity than habitat composition along transects. a possible strategy for the conservation and management of these coastal areas could be i3compaction and collapse characteristics of dune sand stabilized with lime-silica fume mix to protect the foredunes from erosion and limit trampling through the installation of footbridges or the use of appropriate fences. the purpose of this research is to assess the suitability of dune sands as construction materials. the tests include field density, moisture content, classification, compaction, relative density and cbr test. the effectiveness of adding different percentages of lime-silica fume mix was studied. experimental work: dune sand samples were obtained from baiji area in salah-aldeen governorate – north of iraq. three disturbed samples were taken from the surface at a depth of 0.4 m. the consistency and other physical properties of the dune sand were studied by a series of tests. lime-silica fume mix is suggested to overcome the problems of dune sands. the physical and chemical properties of these materials are determined. the specification of each material is as follows: in situ field density (rfield) was obtained by sand-cone test according to (astm d 1556-00). the moisture content (%w.c) for the soil studied was determined according to (astm d 2216-00). grain size distribution was determined by sieve analysis and hydrometer tests conducted on dune sands samples in accordance with (astm d 422-00). the grain size distribution of the soil used reveals that the soil is composed of 91% sand, 3% silt and 6% clay. table 1 summarizes the physical properties of dune sand. the minimum and maximum dry densities, rd.min and rd.max, were determined according to (astm d 4254-00) and (astm d 4253-00), respectively. table 1. summary of baiji dune sand properties. the lime material (l): calcium hydroxide (slaked lime) is most widely used for stabilization. calcium oxide (hydrated lime) may be more effective in some cases. chemical and physical tests were carried on the lime material in the national center for construction laboratories and research (ncclr) in iraq. the results of chemical and physical properties of the lime used are shown in table 2. table 2. results of hydrated lime tests. silica fume material (sf): a grey-colored densified silica fume is used. it is a pozzolanic material which has a high content of amorphous silicon dioxide and consists of very fine spherical particles. silica fume was used as an additive material to improve soil properties. the presence of this substance gives greatly improved internal cohesion, water retention and increased density when set. chemical and physical tests were carried on the silica fume material in the national center for construction laboratories and research (ncclr) in iraq. the results of the chemical analysis of this material are presented in table 3. table 3. results of silica fume tests. compaction test: moisture-density relationship was determined by means of the standard and modified proctor compaction test. the tests were performed in accordance with (astm d 698-00, method a) and (astm d 1557-00, method a) standards, respectively. the moisture density relationships are shown in figure 1. i4 mohammed y. fattah, hasan h. joni, and ahmed s. a. al-dulaimy. figure 1. standard and modified compaction curves for the dune sand. california bearing ratio (cbr) test: cbr-value is used as an index of soil strength and bearing capacity. a series of tests were conducted on each dune sand sample as unsoaked samples. the preparation of specimens and testing procedure were generally in accordance with astm d 1883-99 and astm d 1557-00. the specimens were prepared at 95% of maximum dry density of the modified compaction test, and compacted in five layers using 4.5 kg hammer dropped from a height of 45 cm. fifty six blows per layer were used for compacting the two specimens. california bearing ratio values were determined and calculated as follows (astm d 1883-99): (1) cbr value was measured after 1 day curing for the soil stabilized with l-sf. the test was performed on samples prepared at 95% of optimum moisture content and compacted to various densities. collapse test: collapse test of soil the soil studied was determined according to (astm-d5333-03). the magnitude of one-dimensional collapse that occurs when the soil inundated with water. the test method consists of placing a compacted soil specimen at the desired water content in an oedometer, applying a predetermined applied vertical stress to the specimen and inundating the specimen with distilled deionized water to induce the potential collapse in the soil specimen. the test method follows these steps. the one-dimensional response-to-wetting test, which is performed using conventional consolidation equipment represents the frequently used laboratory collapse test for determining the collapse potential of the soil. pressure was applied until the stress on the sample was equal to (or greater than) that expected in the field or up to 200 kpa as suggested by jennings and knight (1975) and as standardized by astm d-5333 (2003). according to astm d5333 (2003), collapse: indicates a decrease in the height of confined soil following wetting at a constant applied vertical stress. a standard oedometer ring of 75 mm diameter specimen was prepared in steel ring. after careful placing. the effective vertical stresses used were: (25, 50, 100, 200 and 400) kpa. the collapse potential (ic) denotes the percent-relative magnitude of collapse determined at any stress level as follows (astm d-5333-03): (2) where: dh= the change in specimen height resulting from wetting, mm, and ho= the initial specimen height, mm. equation (2) may be rewritten in terms of void ratio as follows: (3) where: eb, e1= the void ratio at the appropriate stress level before wetting, ei , e2 = the void ratio at the appropriate stress level after wetting, and eo= the initial void ratio. the collapse index (ie), which is the wetting induced strain measured at a reference stress level of 200 kpa. the classification of collapse index is listed in table 4, (astm d5333-03). table 4. classification of collapse index (ie), (astm d5333-03, table 1). results and discussion: table 5 shows the maximum density, minimum density and relative density values of the soil mixed with different percentages of lime-silica fume mix. the table shows the increase in relative density of soil with increasing of l-sf content due to the low values of the density of lime (0.60 gm/cm3) and silica fume (0.75 gm/ cm3). this indicates that the additives (lime and silica fume) are fine materials and these materials relatively filled the voids of dune sands about on executing maximum density test (astm d 4253), which led to the increase in maximum density for soil and thereby increase the relative density. this difference may be attributed to the round shape of sand particles tested, (lambe and whitmen, 1979). table 5. maximum and minimum dry densities and relative densities of soil mixed with different percentages of l-sf. results of compaction test: table 6 and figure 2 show the moisture-dry density relationship for stabilized soil. by examining the results, the following points could be noticed: 1. for small percentage of silica fume, (3%) and any percentage of lime, there is a decrease in the optimum water content with slight increase in maximum dry density. i5compaction and collapse characteristics of dune sand stabilized with lime-silica fume mix 2. the soils stabilized with (6%, 9% and 12%) of silica fume and any percentage of lime (which are fine materials) were compacted at lower maximum dry density and higher optimum water content. the increase in optimum moisture content due to addition of lime and silica fume may be caused by the absorption of water by l-sf mix. 3. the percentage (6% l + 3% sf) revealed the maximum dry density. 4. the shape of curves became more flat with increase of the l-sf percentage. similar finding was reported by parsons et al., (2004) and al-jobouri, (2013). table 6. results of proctor compaction tests of soil mixed with different percentages of l-sf. figure 2. variation of the dry density with moisture content of standard proctor compaction for soil mixed with different percentages of l-sf. results of california bearing ratio (cbr) tests: the values of stresses are plotted against depth of penetration in figure 3. stress and penetration curves for baiji dune sands consist of an approximately straight portion for about 3 mm penetration in which stress and penetration are directly proportional, then tend to become a horizontal at further penetration. figure 3. stress–penetration curves from cbr test for natural dune sand and selected soils stabilized with different percentages of l-sf by mixing. unsoaked cbr values at 2.54 and 5.08 mm penetration are plotted in figure 4 and summarized in table 7. the higher cbr values are obtained at 5.08 mm penetration for baiji dune sands, where the soil is sand and containing lower percentage of material passing the no. 200 sieve. table 7. results of cbr test a chosen soils stabilized by mixing with l-sf. the results of figure 4 show that when stabilizing the soil with l-sf after curing for 1 day, there is an increase in cbr due to the increase in the angle of internal friction (f) or cohesion (c) or both. the added materials of l-sf play the major role in stabilization. using this type of material is reflected in the improvements in terms of bearing capacity which is directly proportional to cbr. cbr-value increase with increasing silica fume but this increasing reduces with increasing lime. the increase in cbr ranged between (443–707)% at 2.54 mm penetration and (345–410) % at 5.08 mm penetration as illustrated in table 7 and this increase is calculated as follows: (4) figure 4. cbr-value of natural dune sand compacted and stabilized with different percentages of l-sf by mixing at different penetrations. results of collapse test: single oedometer test is selected for the collapse test for disturbed and undisturbed samples. this test is carried out according to the procedure recommended by astm d5333. collapse potential (ic) and collapse index (ie) magnitude are determined according astm d5333-03 by using the conventional oedometer device in a constant temperature and humidity environmental at different stress level. it shows high values when flooding the specimens by water during primary consolidation. it is concluded that there is an increases in collapsibility due to removal of bonding between particles upon wetting due to stress release and leaching of soil during i6 mohammed y. fattah, hasan h. joni, and ahmed s. a. al-dulaimy. a. natural soil b. treated soil water infiltration leading to softening due to rearrangement of soil particles. moreover, the collapse potential decreases with the increase in the relative density. this is due to the reduction in void ratio of the specimens with increasing the relative density. table 8 . results of collapse potential test a chosen soils stabilized by mixing with l-sf.£ plate 1. collapse of dune sands after the test for natural and treated soils. figure 5. single oedometercollapse test result for natural dune sands. figure 6. single oedometer collapse test result for dune sands stabilized by mixing of l-sf, (3%l + 12%sf). figure 7. single oedometercollapse test result for dune sands stabilized by mixing of l-sf, (6%l + 3%sf). figure 8. single oedometer collapse test result for dune sands stabilized by mixing of l-sf, (6%l + 12%sf). figure 9. single oedometer collapse test result for dune sands stabilized by mixing of l-sf, (9%l + 6%sf). figures 5 to 9 and table 8 show the collapse potential decreases slightly with increasing the percentage of l-sf mix for soil samples tested by single collapse. the collapse index and potential of treated soil are found less than that of natural soil and decrease with increasing stabilizer content. collapsing increased continuously with applied stresses. on the other hand, the added silica content to the treated soil has an effect of higher than lime in reducing the collapse index at different stresses. plate 1 shows the collapse of natural and treated dune sands after test. the decreasing in collapse index ranged between (17.2–36)% at 200 kpa as illustrated in table 8 and this decrease is calculated as follows: (5) i7compaction and collapse characteristics of dune sand stabilized with lime-silica fume mix the calculated collapse index (ie)gives an indication thatdegree of specimen collapse for soil. baiji dune sands undergo only small collapse. the degree of collapse of this natural soil is classified as “slight” under different soaking pressure. this result coincides with the results found by (al-taieet al., 2013). also degree of collapse for treated soils by mixing with l-sf mix are classified as “slight” as shown in table 8. table 8. results of collapse potential test a chosen soils stabilized by mixing with l-sf.£ figure 10 gives the collapse percent versus percent of lime and silica fume, diagrams obtained from one dimensional consolidation. it is clear that there is a slightly decrease in collapse with increasing stabilizers content. l-sf is effective in reducing the collapse potential of the soil. on the other hand, add l-sf decrease void ratio of soil. this indicates that the additives (lime and silica fume) are fine materials and these materials relatively filled he voids of dune sands about on executing compaction, which led to the increase in relative density for soil and thereby decrease the collapse index. figure 10. effect of additives (lime and silica fume) on collapse potential for natural dune sand and selected soils stabilized with different percentages of l-sf by mixing. conclusions: 1. the mix of l-sf allowed compaction of the soil at higher maximum dry density. the treatment with l-sf shows a general increase in the maximum dry density from (1.609 to 1.722) gm/cm3 of standard compaction, and from (1.609 to 1.834) gm/cm3 of modified compaction at (6% l + 3% sf). 2. the optimum water content increases from 13.6% to 15.5% at 12% sf of standard compaction, the increase in the optimum moisture content is due, in spite of the reduced surface area caused by flocculation and agglomeration, to the additional fine contents to the samples which requires more water in addition to the free lime and silica fume that needed more water for the pozzolanic reactions to take place. 3. unsoaked cbr on compacted dune sands treated with l-sf by mixing and cured for a period of 1 day, exhibited higher cbr values than compacted untreated dune sands. moreover, the 5.08 mm penetration cbr of the treated dune sands is higher than the 2.54 mm penetration cbr. the increasing in cbr ranged between 443 – 707% at 2.54 mm penetration and 345 – 410% at 5.08 mm penetration. 4. the degree of collapse of natural baiji dune sands and treated soils by mixing with l-sf mix are classified as “slight” under different soaking pressure. 5. the difference in collapse potential between some percent of l-sf becomes very little. this means that any additional amount of l-sf does not improve considerably the control of collapse. collapse potential and collapse index decreased with stabilizers content and drop to (ie= 1.06%) at (6%l + 12%sf). finally, l-sf treatment in dune sands generally reduces collapse. the decreasing in collapse index ranged between (17.2–36)% at 200 kpa for treated soils. references: abbech, e. k., laouar, m. s., & messaoud, f. (2010). prediction of collapsible soils by cone penetrometer and ultra-sonic tests. studia geotechnica et mechanica, vol. xxxii, no. 2. albusoda, b. s. & salem, l. a. (2012). stabilization of dune sand by using cement kiln dust (ckd). journal of earth sciences and geotechnical engineering, 2(1), 131-143. al-jobouri m. m. (2013). strength and compressibility characteristics of soft soil stabilized with lime silica fume mix. m.sc. thesis, civil engineering department, university of baghdad, iraq. al-juarikh a. (2009). volume change measurement of collapsible soil stabilized with lime and waste lime. civil engineering, tikrit journal of engineering sciences, 16(3), 38-54. al-soud, m. s. (2000). stabilization of baiji sand dunes by petroleum residues. m.sc. thesis, civil engineering department, university of baghdad, iraq. al-taie, a. j., al-shakarchi, y. j., & mohammed, a. a. (2013). investigation of geotechnical specifications of sand dune soil. a case study around baiji iraq. civil engineering, international journal of advanced research, 1(6), 208-215. ashkenazy, y., yizhaq, h., & tsoar, h. (2012). sand dune mobility under climate change in the kalahari and australian deserts. climatic change, 112, 901–923, doi: 10.1007/s10584-011-0264-9. astm d1883-99 (1999). standard test methods for cbr (california bearing ratio) of laboratory-compacted soils. american society for testing and materials. astm d5333-03 (2003). standard test methods for measurement of collapse potential of soils. american society for testing and materials. bouziani, t., bederina, m., hadjoudja, m. (2012). effect of dune sand on the properties of flowing sand-concrete (fsc). international journal of concrete structures and materials, 6(1), 59-64. doi 10.1007/s40069-012-0006-z. ciccarelli, d. (2014). mediterranean coastal sand dune vegetation: influence of natural and anthropogenic factors. environmental management, 54, 194–204. doi: 10.1007/s00267-014-0290-2. cooke, r. u., & warren, a. (1973). desert geomorphology. london: b. t. batsford ltd., 374 pp. 1st. edition. das, b. m. (2007). principles of foundation engineering. adapted international student edition, nelson a division of thomson canada limited. sixth edition. jackson, d. w. t., cooper, j. a. g., & green, a. n. (2014). a preliminary classification of sand dunes of the kwazulu-natal coast. in: green, a. n. & cooper, j. a. g. (eds.), proceedings 13th international coastal symposium (durban, south africa), journal of coastal research, special issue 70, 9, pp. 718-722, issn 0749-0208. doi: 10.2112/si70-121.1. i8 mohammed y. fattah, hasan h. joni, and ahmed s. a. al-dulaimy. jannings, j. e., & knight, k. (1975). a guide to truction on or with materials exhibiting additional settlement due to collapse of grain structure. proceedings, sixth regional conference for africa on soil mechanics and foundation engineering johannesburg, south affrica, pp. 99-105. kumar, s. (2010). a study on the engineering behaviour of grouted loose sandy soils. ph.d. thesis, division of civil engineering, cochin university of science and technology, india, 2010. lambe, t. w. & whitman, r. v. (1979). soil mechanics. john wiley and sons inc., new york. mansour, z., chik, z., & taha, m. r. (2008). on soil collapse potential evaluation. university kebangsaan malaysia, iccbt 2008 -e(03), pp. 21-32. parsons, r. l., kneebone, e., & milburn, j. p. (2004). use of cement kiln dust for subgrade stabilization. final report, kansas department of transportation, usa. shearzid, m. f. (1989). sediment wind and its impact on the design of human settlements. m.sc. thesis, department of architectural engineering, university of technology, baghdad, iraq. keywords: 3d location algorithm; mass and non-mass interference; geostratigraphy; surrounding rock; ree; trace; palabras clave: algoritmo de localización 3d; interferencia masiva y no masiva; geoestratigrafía; roca circundante; ree; rastro. how to cite item lv, y., lu, l., & cao, m. (2021). tracing method of rare earth elements in surrounding rock of geological formation based on threedimensional positioning algorithm. earth sciences research journal, 25(1), pending pages. doi: https://doi.org/10.15446/esrj. v25n1.93725 because of the interference of mass spectrum and non-mass spectrum, the tracing accuracy of rare earth elements in the surrounding rock of geological formation is low. pretreatment of test sample reagent, dissolution of test sample residue, characterization of rare earth element doped materials, analysis of mass spectrometry and non-mass spectrometry interference in rare earth element tracking, using three-dimensional positioning algorithm to track rare earth elements in geological strata surrounding rock. in the experiment, five samples of surrounding rock of geological strata are selected as experimental indexes. the experimental results show that the tracking accuracy of four rare earth elements is high. abstract tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm método de rastreo de elementos de tierras raras en roca circundante de formación geológica basado en un algoritmo de posicionamiento tridimensional issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.93725 debido a la interferencia del espectro de masas y del espectro no masivo, la precisión de la traza de elementos de tierras raras en la roca circundante de formación geológica es baja. en este trabajo se realizó el pretratamiento del reactivo de la muestra de prueba, disolución del residuo de la muestra de prueba, caracterización de materiales dopados con elementos de tierras raras, análisis de espectrometría de masas e interferencia sin espectrometría de masas en el seguimiento de elementos de tierras raras, con un algoritmo de posicionamiento tridimensional para rastrear elementos de tierras raras en los estratos geológicos que rodean la roca. en el experimento se seleccionaron cinco muestras de rocas circundantes de estratos geológicos como índices experimentales. los resultados experimentales muestran que la precisión de seguimiento de cuatro elementos de tierras raras es alta. resumen record manuscript received: 03/20/2020 accepted for publication: 25/11/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 21-28 yan lv1, laijun lu1, mengxue cao2* 1college of earth sciences, jilin university, changchun, 130061, china 2geomathematics key laboratory of sichuan province, chengdu university of technology, chengdu, 610059, china lvyanjlu_bj@163.com; lulj1956@163.com; caomengxue_sc@163.com g e o l o g y https://doi.org/10.15446/esrj.v25n1.93725 https://doi.org/10.15446/esrj.v25n1.93725 https://doi.org/10.15446/esrj.v25n1.93725 mailto:lvyanjlu_bj@163.com mailto:lulj1956@163.com mailto:caomengxue_sc@163.com 22 yan lv, laijun lu, mengxue cao introduction rare earth elements are the key elements in the development of high and new technology and have a very broad application prospect in the fields of materials, information, energy, ocean and space science and technology. rare earth elements have some special geochemical properties, such as stable chemical properties, a high degree of homogenization, not easy to be disturbed by metamorphism, and so on. once “recorded” in the terrane, it is easy to be preserved and so on, so it is a good geochemical indicator. as a good indicator and tracer, rare earth elements can be used to reverse the geochemical state and evolution process in the geological body. it is of great significance in judging the formation environment and conditions of sedimentary rocks, tracing the source of diagenetic materials of sedimentary rocks, and explaining the formation reasons of sedimentary rocks. at the same time, with the development of rare earth resources, rare earth elements inevitably enter the environment, food chain, and organisms in various ways. since the sixties of this century, with the development of analytical and testing technology, not only the total content of rare earth in geological bodies but also the content of individual rare earth elements can be measured, which makes rare earth elements play an important role in the research work of diagenesis, mineralization and celestial evolution. however, due to more or less different physical and chemical properties of different rare earth elements, especially in different geological conditions, it often leads to a different distribution. in modern industrial automation production, quality control requires more and more high sensitivity for the determination of rare earth elements, and the chemical properties of rare earth elements are very similar, and their separation has always been rare one of the problems and hot spots in soil analytical chemistry. as a result, the study of rare earth element separation and analysis is not only of great theoretical significance but also of practical application value. therefore, a tracing method of rare earth elements in the surrounding rock of geological formation based on a three-dimensional positioning algorithm is designed. material method test reagent before tracing the rare earth elements in the surrounding rock of the geological formation, in order to eliminate the interference of organic matter and suspended matter, oxidize the elements of various valence states into a single high valence state or transform them into inorganic compounds that are easy to be separated, so that the water sample can be concentrated, showing a clear, transparent and precipitation free state, the surrounding rock materials of the geological formation need to be digested (wen et al., 2019), the main applied instruments and reagents are as shown in table 1: table 1. instruments and reagents serial number instruments and reagents model 1 element 2 plasma mass spectrometer (thermo fisher, usa) 2 preparation of hydrofluoric acid and nitric acid by boiling distillation 3 element standard stock solution mixed standard solution of rare earth elements with scan and yttrium (national nonferrous metals and electronic materials 4 high purity water 18mω/ cm milli-q purified water 5 sodium hydroxide powder, ar, kemo, tianjin 6 ammonium chloride tianjin como 7 magnesium chloride tianjin miaoyou 8 triethanolamine xilong chemical 9 concentrated ammonia water jiangxi united chemical 10 hydrofluoric acid xilong chemical 11 hydrochloric acid xilong chemical co., ltd 12 nitric acid xilong chemical 13 standard stock solution standard solution of rare earth elements 14 water purifier cooperative uv preparation of standard solution (vatin-perignon et al., 2018): blank solution: 2mol l-1 hydrochloric acid; 1 g·ml-1 rare earth element mixed standard solution: accurately transfer 5ml of 100 g ml-1 mixed standard solution into a 50ml volumetric flask with a pipette, and fix the volume with 2mol l-1 hydrochloric acid (kwan et al., 2019), and then dilute the solution 10 times to a 100ml volumetric flask to obtain 1 g·ml-1 ree standard solution, which is ready for use. 5 g·ml-1 rare earth element mixed standard solution: pipette 5ml 10 g·ml-1 mixed standard solution to a 100ml volumetric flask, and use 2 mol·l-1 hydrochloric acid to fix the volume to obtain 5 g·ml-1ree, which is ready for use. after digestion, accurately weigh about 0.5g (accurate to 0.0001g) of sample gss-4 into microwave digestion tank, add 2ml of hcl, 2ml of hf, 5ml of hno3 for digestion under the three different conditions (li et al., 2018), then drive out the acid, and observe whether the digestion is complete. the digestion conditions are as shown table 2: table 2. sample digestion conditions instrument parameter digestion conditions step 1 step 2 condition 1 300w, heating time: 5 min, holding time: 10 min heating time: 10 min, holding time: 20 min condition 2 300w, heating time: 5 min, holding time: 10 min heating time: 10 min, holding time: 20 min condition 3 300w, heating time: 5 min, holding time: 10 min 1000w, heating time: 5 min, holding time: 30 min: after sample digestion, it is necessary to heat the sample and prepare rare-earth doped materials by microwave heating. the device is mainly composed of a multi-mode microwave resonator, magnetron, insulation system, turntable, and other parts. the magnetron is used to emit microwaves, and the resonator makes the microwaves resonate in the cavity, while the turntable is used to load the raw materials in the resonator at a uniform speed, so as to make the raw materials rotate at a uniform speed. the raw materials are evenly radiated by microwave. it is worth mentioning that in the whole process, due to the high temperature of the reactant, a large amount of heat generated by microwave will be dissipated in the resonant cavity in the form of thermal radiation (bidhendi & geitmann, 2018), heat conduction, and other forms, resulting in energy waste and damage to the resonant cavity and electronic components. therefore, it is necessary to insulate the reaction raw materials, as shown in figure 1: figure 1. heating device among them, 1 is a multi-mode microwave cavity, 2 is a magnetron, 3 is a thermal insulation device, 4 is the raw material to be heated, and 5 is a turntable. in addition, if the raw material is directly put into the microwave cavity for heating, according to the microwave heating principle, the microwave will rapidly increase the temperature of the reactant in the way of “body heating”, at this time, the surrounding environment temperature is relatively low, and the high-temperature reactant will exchange heat with the surrounding air through the way of heat conduction or heat radiation, which will reduce the surface temperature, resulting in the high inside and low outside of the reactant temperature gradient, resulting in uneven heating. 23tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm in order to avoid this kind of situation, the microwave heating and insulation device as shown in figure 2 came into being. the device is a combination of the buried powder insulation layer and basket insulation layer and realizes the insulation of reactants through multi-layer insulation structure. the outermost layer is an aluminosilicate (3al2o3-2sio2) container with a thickness of about 3cm, which is transparent to the microwave, so the microwave can penetrate almost without damage; the thermal conductivity (cemlyn et al., 2018) is small, which can play a very good role in heat preservation; moreover, it is resistant to high temperature, has a uniform expansion, and is still stable at 1700 °c. the innermost layer is a corundum crucible with reactant powder, good microwave permeability, and high-temperature resistance. outside the corundum, crucible is a layer of graphite crucible, which serves as a microwave sensor. because of the good coupling of graphite to microwave, this setting can not only improve the uniformity of heating but also preheat the reactant, which plays the role of indirect heating. the space between the aluminosilicate container and the graphite crucible, between the graphite crucible and the corundum crucible, is filled with quartz sand, so as to reduce the space between the insulation layer and the reactant as much as possible, achieve the purpose of further insulation, and minimize the temperature difference on the surface and inside of the reactant. figure 2. test sample insulation device in fact, the heat preservation device has three functions of microwave coupling (makridakis, 2018), heat preservation, and heat insulation. on the one hand, it can reduce the temperature gradient inside and outside the reactant, improve the heating uniformity, and at the same time, it can maintain the hightemperature thermal field environment inside the reactant for a period of time after the microwave radiation stops, so as to protect the reaction sample; on the other hand, due to the existence of the microwave coupling graphite, the microwave conventional mixed heating of the reaction sample is realized in a very simple way, which provides a uniform heat field for the sample, accelerates the heating speed and improves the heating environment; third, it can insulate the high temperature inside the insulation body and protect the microwave cavity and electronic components. pretreatment of test samples after crushing, the sample is sieved through 180 mesh and dried at 105 °c for standby. the sample pretreatment was carried out in the super clean laboratory of grade 100. the specific steps are as shown: a. accurately weigh 50 mg of the sample into a 25 ml polytetrafluoroethylene beaker, moisten with a small amount of water, and repeat three times; b. add 1 m lhf and 1 m lhno3 respectively, put them into a highpressure tank and heat them up to 190 °c, and keep them warm for 72 hours; c. take out the high-pressure tank, cool it, open the cover and evaporate it to dryness. when the solution is nearly dry, add 1m l100pg / ml rh internal standard solution, 1ml hno3 and 4-5ml of three times water, put it into the high-pressure tank, seal and heat it to 140 °c, keep it warm for 5 hours, so as to dissolve the residue and ensure complete extraction; d. after the solution is cooled, take 0.2ml to 10ml polytetrafluoroethylene test tube, volume it, shake it, and then test it on the machine. the acidity of the solution is 2% hno3. first separation: microwave acid decomposition method: in the digestion solution, naoh solution (about 5g, dissolved in 30ml water) is used to precipitate the digestion solution for the first time, making rare-earth ions precipitate in naoh alkaline condition, separating with matrix interference ions which do not precipitate in alkaline condition, and adding triethanolamine (1 + 1), making a large amount of fe, al, zr in the soil and so on, forming a complex with it, filtering and being separated and removed, then dissolving and settling the solution with hcl hot solution to obtain the solution (wen et al., 2018), and then selecting different precipitants for secondary precipitation separation to further remove the interference elements of the matrix in the soil (hebaz et al., 2018). muffle furnace alkali melting method: add triethanolamine (1 + 1) into the digestion solution to form a large number of fe, al, zr, and other complexes in the soil, filter them and remove them. after that, use hcl hot solution to dissolve and precipitate the solution, and then select different precipitants for secondary precipitation and separation to further remove matrix interference elements in the soil. the second separation adopts the two methods: first, add 2.5% (m / v) cupferron (su et al., 2018) into the solution, stir it continuously until there is no precipitation, filter it, take the filtrate with a beaker, then heat it on a hot plate to concentrate it to 3ml, and then fix the volume to 10ml colorimetric tube. to be tested; second, add 3g of nh4cl to the digestion solution, 50ml of water, and then add ammonia drop to produce a large amount of precipitation. add about 50ml of ammonia to the solution until the ph value is greater than or equal to 12, then heat and boil it, cool and filter it, discard the filtrate, dissolve the hot hcl (2mol / l) used for precipitation in a beaker, heat it on a hot plate to concentrate to 3ml, and fix the volume to 10ml in the colorimetric tube, to be tested. after heating, the whole sample treatment process is divided into the stages: first, the formation stage, which is the main physical and chemical reaction stage, in which various chemical and physical changes are carried out for each component in the matrix raw material, and most of the gas products are discharged; second, as the temperature continues to rise, the raw material becomes a melt, and the sintering products diffuse to each other and form a network. at the end of this stage, there are basically no matrix raw material particles in the melt; third, the homogenization stage of the melt, that is, under long-term high temperature preservation, the components in the melt gradually evenly distributed; fourth, the rapid cooling stage, this stage is mainly to make the products fragile and convenient for ball milling later; fifthly, in the phase of ball milling and mixing, this is the last stage of the whole sample treatment, which is mainly to grind the products of the large particles to make them mix with different materials according to different needs and make the required samples. test instruments all instruments used in the experiment are as shown in table 3: table 3. test instruments order number instrument model 1 elemental analyser vario el-iii germany 2 inductively coupled plasma emission spectrometer vista-mpx icp-oes valian, usa 3 uv-vis spectrophotometer uv-7502pc xinmao, shanghai 4 infrared spectrometer nicolet avatar 380ft-ir of the united states 5 x ray photoelectron spectrometer japanese ulvac-phi phi5000versa-probe type 6 thermal analyser perkin-elmer, pyris diamond tg-dta, perkin elmer, usa 7 automatic vertical pressure steam sterilizer yxq-ls-30sii, shanghai 24 yan lv, laijun lu, mengxue cao inductively coupled plasma mass spectrometer is mainly used for quantitative analysis of multi-element in surrounding rock materials. the main parameters are as as shown in table 4: table 4. parameters of icp ms measuring instrument serial number instrument parameters numerical value 1 rf power 1500 watts 2 sampling depth 8.0 mm 3 cooling air volume 15l / min 4 auxiliary air flow 1.0l / min 5 carrier gas flow 0.8l / min 6 compensated air flow 0.2l / min 7 peristaltic pump speed 0.1rps 8 sample extraction time 45 years 9 stable schedule 1960s 10 integration time 0.3 s 11 number of repetitions 3 times 12 oxide <1.2% 13 collection mode full spectrum mode 14 tuning mode no gas regulation mode on this basis, the icp-ms analysis instrument is very complex. it has many analysis functions and variable working parameters. among the working parameters of the instrument, some are given by the instrument, some can be determined by experiments. however, there are many factors that directly affect the sensitivity, precision and detection limit of icp-ms (bernardi et al., 2018). in addition to the working parameters given above, the working parameters of the ion mirror system in ms and the calibration of the peak position are more important. in multiple parameters of ion mirror (budko & martynova, 2019), as long as one of the parameters changes, other conditions also need to be adjusted accordingly to obtain the corresponding sensitivity and precision, and then the next measurement after peak position calibration can be carried out. and with the long-term use of sampling cone and cutting cone, the surface and cone hole will change accordingly. therefore, before each start-up test, it is necessary to do the instrument condition optimization experiment to determine the corresponding working parameters. there are 7-10 working parameters of the ion mirror system, and the conditions obtained after each optimization experiment are not the same. for the simultaneous determination of elements in the whole range from light mass number to heavy mass number, the optimization experiment of instrument conditions is particularly important. in this experiment, the solution of 1 g/ul containing the elements is used, and the isotopes are selected for tuning. determination method when the sample is tested, it must be separated and purified as much as possible, and at the same time, the concentration and medium of the sample and the standard should be consistent as much as possible. therefore, it is necessary to evaluate the recovery rate of each sample. only when the recovery rate is guaranteed, the ideal data can be obtained. the two methods are adopted for data quality control in this experiment. after all the samples are dissolved, take one solution before and after chemical separation to test the trace element one, so as to evaluate the overall recovery rate and matrix removal effect. according to the current evaluation, the overall recovery rate is up to, while the removal rate of the matrix, such as the main same heavy elements or molecules, ions, is basically close, so as to ensure the mass fractionation effect that may be brought about in the chemical pretreatment process and the matrix effect in the mass spectrometry determination process. in the chemical pretreatment process, the standard solution that has been calibrated and several parallel samples are added for synchronization learn to handle (varney & bousfield, 2018). generally, the relative comparison method is used to determine the content of elements in the sample, so is the three-dimensional positioning algorithm. it is also the most common method to configure a series of standard solutions or use a series of standard substances to make the working curve. in this experiment, a national level geochemical reference material was selected, which was prepared into a solution according to the sample decomposition procedure and used as the working curve of the standard solution. the element content of reference material is the standard value or recommended value provided by many analytical methods and laboratories after many times of analysis. with this method, the matrix effect is compensated to some extent because the operation is completely consistent with the sample analysis procedure. after the experiment, the selected sample can be completely decomposed, all the measured elements enter the solution, and the concentration does not exceed the linear range. the correction results and practicability are very satisfactory. at the same time, another 2-3 national standard samples are inserted into the sample (abbott, 2018) for further verification, and the analysis results are very reliable. the relevant isotopes selected in the analysis and the determination method are as shown in the table 5: table 5. selected isotopes and determination methods elemental determination preferred m / z abundance sensitivity (%) measurement method y 89 100 simulation rh 103 100 pulse la 139 100 simulation ce 140 100 simulation pr 141 100 pulse nd 146 17.2 pulse sm 147 15 pulse eu 153 52.2 pulse gd 157 15.7 pulse tb 159 100 pulse dy 163 24.9 pulse ho 165 100 pulse er 166 33.6 pulse tm 169 100 pulse yb 172 21.9 pulse lu 175 97.4 pulse blank and standard drift the analysis of trace sample is very important for the control of blank. in addition to the process blank of sample, the instrument blank will also affect the quality of sample data seriously. the instrument blank will drift in the determination process. generally speaking, the strength is high at the beginning of determination. therefore, the negative value often appears for the low content sample. at the same time, the drift degree of each element in the determination process may not be good as such, gaps and standards need to be monitored frequently. internal calibration is one of the quantitative calibration methods often used in many analytical techniques. the function of internal standard is: a. monitor and correct the short-term drift of the signal; b. monitor and correct the long-term drift of the signal; c. correct the second element; d. correction of general matrix effects. the internal standard correction method (cao & wen, 2019) is used in icp-ms, especially in the determination of large quantities of samples. it is better to add internal standard at the beginning of the experiment, especially for the experiment with separation and enrichment steps (it is required that the internal standard is consistent 25tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm with the recovery rate of the elements to be measured), so that the loss in the experiment process does not affect the results, and when transferring the solution, there is no need to clean and fix the volume, which saves a lot of time, because the ratio of the elements to the internal standard to be determined is unchanged. the selection principle of icp-ms internal standard elements is that the elements to be determined do not contain the selected internal standard elements, the interference of the internal standard elements is as little as possible, and the mass spectrum behavior is as consistent as possible with the elements to be measured, which is very difficult for the determination of geochemical samples. in addition, when the multi-element determination is carried out, the elements are grouped according to the mass spectrum behavior, and multiple elements are selected as the internal standard. the determination effect is more ideal, but the elements that can be selected as the internal standard are more limited. dissolution of test sample residue because hf is used to decompose samples, but hno3 can’t completely drive hf out (remon et al., 2018), some experimental sample precipitates remain in the residue. for surrounding rock samples, when opening dissolution, the result of ree is seriously low. only under a certain temperature and pressure, these precipitates will be dissolved. however, at 140 ° c, closed dissolution (du et al., 2018) for 3 hours, the results were basically consistent with the recommended values. table 6 is the experimental data table of the relationship between the dissolution amount and time of the national standard sample of surrounding rock. table 6. relationship between residue dissolution and time element open dissolution seal and dissolve for 3 hours seal and dissolve for 6 hours seal and dissolve for 9 hours recommended value la g/g 26.7 57.4 53.9 54.4 54±4 ce g/g 59.4 112 107 110 108±7 y g/g 39.0 64.0 62.7 63.0 62±5 eu g/g 0.75 1.11 0.98 1.06 0.85±0.07 lu g/g 0.86 1.17 1.02 1.10 1.15±0.09 mass and non mass interference in rare earth element tracing in plasma mass spectrometry, there are two kinds of interference: mass spectrum interference and non mass spectrum interference. mass spectrum interference includes isometric interference, background interference and polyatomic ion interference (oxide, hydroxide and bicharged ion generation are interference, among which matrix effect and polyatomic ion interference are the main factors restricting the accuracy and precision of analysis. non mass spectrum interference mainly comes from matrix effect, which refers to the enhancement or inhibition effect of matrix elements on the analysis signal strength. it is a kind of nonlinear interference, and its dynamic drift degree varies with the mass number. the so-called matrix refers to the collection of all components with their own properties. this “all components” should include analytes, but not additives. the combined effect of each component of the matrix on the analysis signal of the element to be measured is matrix effect, that is to say, it is the combined interference effect of all accessories except additives. in the early analysis of icp-ms, the behavior and interpretation of matrix effect are relatively scattered, but generally speaking, there are several common conclusions as shown: high concentration matrix can inhibit the analysis signal, and the degree of inhibition is related to the quality; the inhibition effect of heavy matrix is greater than that of heavy analyzed elements; the matrix effect is related to the lens system of the instrument. with the development of research, the mechanism of matrix effect is gradually clear, and space charge effect is considered as the main reason. the influences on the determination are: (1) reducing the taper hole; (2) consuming energy in ionization, affecting the ionization rate. for the same concentration of elements, the signal strength of the matrix with high concentration is low and the reduction is fast. when the geological samples are diluted more than 2000 times (0.5 mg / ml), the matrix effect is small. the more prominent problems in the determination of trace rare earth elements (habal, 2018) are the interference of polyatomic ions, especially the interference of oxides of light rare earth elements, the interference of hydroxides to heavy rare earth elements, and the interference of oxides formed by seven natural isotopes of barium and hydroxides to light rare earth elements. the yield of rare earth oxides and hydroxides (reo (h) + is affected by the power of the ionomer, the velocity of carrier gas, the depth of sampling, the diameter of the sampling cone and the diameter of the intercepting cone, while the high plasma power and the low velocity of carrier gas can effectively reduce the yield of oxides. many analytical scientists have done research in this field, including the influence of the aperture size of the sampling cone and the intercepting cone on the oxide signal. characterization of rare earth element doped materials in surrounding rock of geological formation by using a polycrystalline x-ray diffractometer to analyze the structure and phase of the anti-counterfeiting powder, the average grain size of the powder can be measured. according to the diffraction, the grain size can be calculated by scherrer equation (1): d k = λ θcos (1) in equation (1), cos  represents the diffraction angle of crystal surface, k is constant, and  represents the wavelength of the target source. the size distribution, morphology and agglomeration of particles were directly observed by philips xl-30tmp environmental scanning electron microscope (sem) and nova nanosem 430 field emission scanning electron microscope (fesem). the specific steps are as shown in figure 3: figure 3. micro characterization process in order to characterize the application of the materials in spectrum anticounterfeiting and laser protection, the reflection spectrum and fluorescence emission spectrum of the synthesized powders were measured after they were uniformly mixed into the coating to make the anti-counterfeiting and protective coating. when the light hits the ink surface, part of it is reflected, reflection coefficient: r i l u  / (2) 26 yan lv, laijun lu, mengxue cao in equation (2), u represents the incident light intensity, l represents the reflected light intensity, and i represents the change parameter of the reflection coefficient with the wavelength. trace of rare earth elements in surrounding rock of geological formation according to the above process, the three-dimensional positioning algorithm is used to trace the rare earth elements in the surrounding rock of geological formation, and the microwave is used for initial positioning before tracking. microwave, also known as ultra-high frequency wave, refers to electromagnetic waves with very high frequency. its wavelength lies between infrared light and radio waves. the microwave frequency is 2.45ghz and the wavelength is 12.2cm. microwave high temperature heat treatment is a kind of heating method which uses microwave radiation instead of traditional heat sources. when the dielectric is in the microwave field, the electric field component of the microwave will give a force to the charged particles inside the dielectric. if the charged particles can move freely in the medium, they will generate current in the medium; if the charged particles are bound, they can only keep the same phase with the external electric field, that is, the polarization of the medium. according to the different kinds of charged particles in the medium, the polarization of the medium can be divided into four parts: dipole turning polarization, electron polarization, atomic polarization and interface polarization (litasov, 2018), which is expressed as shown in equation (3): a a d r e w= + + +( ) (3) in equation (3), is the total dielectric polarization, is the dipole polarization caused by the permanent dipole in the medium, is the electronic polarization caused by the electrons rotating around the atomic nucleus, is the atomic polarization caused by the atomic nucleus, and is the interface polarization caused by the charges at the interface. on the basis of the above processing, a three-dimensional positioning algorithm is used to trace the rare earth elements in the surrounding rock of the geological formation. the anchor nodes in the three-dimensional positioning algorithm are used as the reference for the location of unknown nodes, figure 4. surrounding rock structure of geological strata which are usually uniformly or randomly distributed in the wireless sensor network, usually with high energy. the anchor node has the ability to control the transmitting power. in an ideal environment, the relationship between the transmitting power of the signal and the radius of the area covered by the signal can be described by the equation (4): v p a f = ( ) 2 (4) in equation (4), p is the signal power received by the sensor node, v is the transmit power, c is the distance between the receiver and the transmitter, a is the speed of light, f is the transmission frequency. from the above equations, it can be seen that when other parameters remain unchanged and the minimum receiving power of the receiving node is fixed, the stronger the signal transmitting power of the transmitting node is, the farther its propagation distance is, so as to track relevant information, realize self positioning (wu & zou, 2018), and generate information table. the information in the information table contains all the received effective beacon information broadcast by the anchor node. anchor nodes will broadcast beacon information at different levels of power repeatedly. therefore, an unknown node will listen to multiple identical beacon information. only when the number exceeds the beacon decision threshold can the beacon information be determined as true. but at this time, the beacon information has not been saved to the information table. after the anchor node broadcasts its own beacon information, the unknown node may decide that several beacon information from the same anchor node is true. at this time, the beacon information with the lowest transmitting power level is taken as the correct information from the anchor node, which is stored in the information table to complete the tracking of rare earth elements in the surrounding rock of the geological formation. experimental study in order to verify the feasibility of the three-dimensional positioning algorithm for tracking rare earth elements in the surrounding rock of geological strata, the traditional method is compared with the designed method, and the tracking accuracy of the two methods in different time periods is compared. it is as shown in figure 4. four samples of rare earth in the surrounding rock of geological strata were measured to determine the element values of these four samples. the accurate determination of the element values represents the trace effect of rare earth elements. which is as shown in table 7. table 7. determination value of rare earth elements in sample 1 element measured value (mg / kg) measured value by traditional method measured value of this design method la 55 45 52 ce 125 100 135 pr 10.3 8 10.2 nd 25 17 24 pm 4.0 2.0 4.3 sm 4.8 3 4.5 eu 0.78 1.5 0.75 gd 4.2 2.6 4.3 tb 0.85 0.4 1.0 dy 6.0 3 6.3 ho 1.34 3 1.35 er 3.7 2.3 3.9 tm 0.6 0.2 0.9 yb 4.2 2 5 lu 0.69 1.3 0.79 sc 0.81 0.15 0.90 y 33 42 35 27tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm the measured values of sample 2, traditional methods and the designed values are as shown in the table 8: table 8. determination value of rare earth elements in sample 2 element measured value (mg / kg) measured value by traditional method measured value of this design method la 50 40 52 ce 100 81 102 pr 9 17 10 nd 15 5 13 pm 6.0 4.3 7.2 sm 6.8 3.5 6.9 eu 1.2 2.6 1.5 gd 5.2 9.2 5.6 tb 0.95 2.69 1.23 dy 7.0 4.32 7.9 ho 2.4 0.9 2.6 er 5.2 9.8 5.6 tm 1.6 8.2 1.9 yb 4.5 6.8 6.8 lu 0.9 0.45 1.3 sc 0.7 2.3 1.9 y 22 45 23 the measured values of sample 3, traditional methods and the designed values are as shown in table 9: table 9. determination value of rare earth elements in sample 3 element measured value (mg / kg) measured value by traditional method measured value of this design method la 45 38 42 ce 142 125 143 pr 8.2 6.9 8.5 nd 18 6 19 pm 6.5 8.9 6.24 sm 5.9 9.6 4.99 eu 0.95 2.36 0.98 gd 5.8 9.5 5.6 tb 0.92 2.36 1.2 dy 8.0 13.2 7.5 ho 2.42 5.23 2.32 er 5.7 9.32 5.96 tm 1.6 3.6 1.6 yb 5.2 6.9 5.4 lu 7.3 10.3 7.9 sc 0.87 1.23 0.87 y 42 56 45 the measured values of sample 4, traditional methods and the designed values are as shown in table 10: table 10. determination value of rare earth elements in sample 4 element measured value (mg / kg) measured value by traditional method measured value of this design method la 78 62 79 ce 152 144 154 pr 17.2 12.3 18.2 nd 22 15 25 pm 7.8 3.2 7.9 sm 6.3 2.6 6.54 eu 1.3 3.5 2.2 gd 4.9 6.6 5.2 tb 1.23 3.34 1.54 dy 8 15 9 ho 2.5 6.5 2.6 er 5.2 8.3 5.8 tm 9.2 15.3 9.5 yb 7.1 9.6 7.9 lu 3.2 6.3 3.9 sc 2.1 8 2.9 y 52 70 49 from the above table 7, table 8, table 9 and table 10, it can be seen that the tracking value of the designed tracking method for rare earth elements in the surrounding rock of geological strata based on the three-dimensional positioning algorithm has a small difference from the measured value, and there is a timely error, which is also within a standard value. however, there is a big difference between the traditional method and the actual measurement value, so it can be proved that the designed method based on 3d positioning algorithm is better than the traditional method. according to the content of rare earth elements, the correlation curve was drawn. which is as shown in figure 5. scanning electron microscopy was used to characterize the morphology of surrounding rock. rare earth elements have good dispersion. the abscissa represents the spectral wavelength, and the ordinate represents the three primary color stimulus values needed to match the spectral colors. the chromatogram of rare earth element tracing detection is as shown in figure 6. point a and point b are in the green area. point b is darker than point a, and the higher the color saturation is, which is conducive to the trace detection of rare earth elements. in this paper, the interference of mass spectrometry and non mass spectrometry in rare earth element tracing is analyzed, and the threedimensional positioning algorithm is used to trace rare earth elements in the surrounding rock of geological strata, which has better tracking effect. conclusions in this paper, based on the three-dimensional positioning algorithm, the study of rare earth element tracking in the surrounding rock of geological formation is proposed, and the interference of mass spectrum and non mass spectrum in rare earth element tracking is analyzed. the three-dimensional positioning algorithm is used to track rare earth elements in the surrounding rock of geological formation. compared with the traditional method, it has better tracking effect. in the next step of research, we need to establish a more reasonable channel model to verify the effectiveness of the tracking method. we can use simulation software or build a more real network environment to experiment the method, so as to promote the application in the actual network environment. this is a problem that needs to be seriously considered. 28 yan lv, laijun lu, mengxue cao references abbott, r. n. (2018). trace element thermometry of garnet-clinopyroxene pairs, revisited. american mineralogist, 103(7), 1169-1171. bernardi, c., dib, s., girault, v.., hecht, f., & sayah, t.. (2018). finite element method for darcy’s problem coupled with the heat equation. numerische mathematik, 139(2), 315-348. bidhendi, a. j., & geitmann, a. (2018). finite element modeling of shape changes in plant cells. plant physiology, 176(1), 41-56. budko, d. f., & martynova, d. m. (2019). trace element uptake assessment in the planktonic biofiltration system. oceanology, 59(1), 27-36. cao, j., & wen, f. h. (2019). the impact of the cross-shareholding network on extreme price movements: evidence from china. journal of risk, 22(2), 79-102. cemlyn, b. r., susanto, h., henning, i. d., adams, m. j., & li, n. (2018) nonlinear dynamics of solitary and optically injected two-element laser arrays with four different waveguide structures: a numerical study. optics express, 26(4), 4751-4765. du, y., kuntzman, m., & chen, y. h. (2018). a revisit to the lumped element model of capacitive mems microphones. journal of the acoustical society of america, 143(3), 1776-1776. habal, m. b. (2018). re-engineering the cleft craft surgery: function is an essential element of the repair. journal of craniofacial surgery, 29(2), 257-258. hebaz, s. e., benmeddour, f., moulin, e., & assaad, j. (2018) semi-analytical discontinuous galerkin finite element method for the calculation of dispersion properties of guided waves in plates. journal of the acoustical society of america, 143(1), 460-469. kwan, j. s. k., sze, e. h. y., & lam, c. (2019). finite element analysis for rockfall and debris flow mitigation works canadian geotechnical journal, 56(9), 1225-1250. li, j. h., lu, x. s., farquharson, c. g., & hu, x. y. (2018). a finite-element time-domain forward solver for electromagnetic methods with complex-shaped loop sources. geophysics, 83(3), 1-63. litasov, k. d., ishikawa, a., bazhan, i. s., ponomarev, d. s., hirata, t., podgornykh, n. m., & pokhilenko, n. p. (2018). trace element composition and classification of the chinga iron meteorite. doklady earth sciences, 478(1), 62-66. makridakis, c. g. (2018). on the babuška–osborn approach to finite element analysis: l2 estimates for unstructured meshes. numerische mathematik, 139(4), 831-844. remon, l., siedlecki, d., cabeza-gil, i., & calvo, b. (2018). influence of material and haptic design on the mechanical stability of intraocular lenses by means of finite-element modeling. journal of biomedical optics, 23(3), 1-10. su, y., chen, y. p., cheng, y. f., wen, c., & zhou, y. m. (2018). effects of modified palygorskite supplementation on egg quality and mineral element content, and intestinal integrity and barrier function of laying hens. biological trace element research, 186(2), 1-9. varney, d. h., & bousfield, d. w. (2018). discrete element method to predict coating failure mechanisms. tappi journal, 17(1), 21-28. vatin-perignon, n., poupeau, g., oliver, r. a., la venu, a., labrin, f., keller, f., & bellot-gurlet, l. (2018). trace and rare-earth element characteristics of acidic tuffs from southern peru and northern bolivia and a fission-track age for the sillar of arequipa. journal of south american earth sciences, 9(1-2), 91-109. wen, f., zhao, y., zhang, m., & hu, c. (2019). forecasting realized volatility of crude oil futures with equity market uncertainty. applied economics, 51(59), 6411-6427. wen, s. r., wu, z. j., & lu, n. l. (2018). high-precision solution to the moving load problem using an improved spectral element method. acta mechanica sinica, 34(1), 68-81. wu, h., & zou, j. (2018). finite element method and its analysis for a nonlinear helmholtz equation with high wave numbers. siam journal on numerical analysis, 56(3), 1338-1359. figure 5. comparison of average content distribution curve of rare earth elements figure 6. cie chromatogram the superlarge dongfeng gold deposit is located in the potouqing faults-alteration belt of the eastern part of the ‘zhao-lai-gold ore belt’, which belongs to the northwestern part of the jiaodong area. tectonically, ore bodies are controlled by faults and gold mainly occurs in the pyrite and polymetallic sulfide-bearing quartz vein. in this paper, rb–sr isotopic analysis is carried out with the beresite, which formed by hydrothermal metasomatism, and the rb–sr isochron age is 125.5±6.7ma, indicating this deposit set up in the early cretaceous of the late yanshanian. based on the relationship between the dongfeng gold deposit and the mesozoic granite, it is suggested that the formation of the gold deposit is a complex geological process of gradual enrichment and precipitation of the ore-forming elements. the initial 87sr/86sr ratio of the beresite is 0.711502±0.000069, which indicates the ore-forming materials mainly come from the crust. combined with the complex mineralization process of the dongfeng gold deposit and the reported h-o isotopic data, it is suggested that the ore-forming materials are mainly derived from the crust with some mantle materials, while the oreforming fluids are originated primarily from magmatic hydrothermal and mantle with some precipitate water. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): i1 i7 abstract keywords: rb–sr isochron, (87sr/86sr) ratio, mineralization, dongfeng gold deposit, jiaodong gold-centralized area. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.55178 o r e d e p o si t s wang zongyong1*, lv guxian2, zhang xunyu1, han xin1, zhang yingchun1, fan xiao1, huo qinglong1, xu yaqing 1.school of earth sciences and resources, china university of geosciences, beijing 100083, china 2.institute of geomechanics, chinese academy of geological sciences, beijing 100081, china * e-mail: hawks_wzy@163.com rb-sr isotopic geochronology and geological implications of dongfeng gold deposit in jiaodong area record manuscript received: 15/01/2016 accepted for publication: 19/04/2016 how to cite item zongyong, w., xin, h., guxian, l. v., xunyu, z., yingchun, z., xiao, f., qinglong, h., & yaqing, x. (2016). rb-sr isotopic geochronology and geological implications of dongfeng gold deposit in jiaodong area. earth sciences research journal, 20(1),i1 -i7. doi: http://dx.doi.org/10.15446/esrj.v20n1.55178 1. introduction as one of the most significant gold concentration areas in china, jiaodong area has special metallization background and metallogenesis (goldfarb and santosh, 2013). the great linglong gold ore field is located in the northwest of the jiaodong area, the eastern part of the ‘zhao-lai-gold ore belt’, including linglong, jiuqu, dakaitou, dongfeng and dongshan ore blocks or fields. previous researchers have carried out many studies of all types of gold deposits in the jiaodong area. yiao et al. (1990), miao et al. (1999) and hu et al. (1998) suggested the gold mineralization was closely correlated with the mesozoic granites. based on the studies of the gold deposit geochronology, yang et al. (2000), wang et al. (2000) and chen et al. (2004) implied that gold mineralization occurred in the mesozoic and concentrated from 128 to 115 ma. regarding the ore-forming fluids, fan et al. (2005) suggested that ore-forming fluids of gold deposits were consistent throughout the whole jiaodong area under similar mineralizing temperature and pressure conditions and the fluids were characterized by h2o-co2-nacl±ch4, with low salinity and low-to-moderate temperature. moreover, mao et al. (2005) pointed out that mantle fluids were associated with the gold mineralization. sun et al. (2000), wan et al. (2000), zhai et al. (2002), zhai et al. (2004), lv et al. (2007), song (2008), song et al. (2013) and yang et al. (2014a) have comprehensively studied and reviewed the mineralizing setting and mineralization model of the gold deposits in the jiaodong area. since the dongfeng deposit has been found at the end of 1980’s, some studies have been implemented in this reservoir. xu et al. (2013) established the deposit database and ore grade distribution model for the dongfeng deposit based on the surpac software and applied the ordinary kriging to estimate the orebody resources. feng et al. (2009) studied the ore-forming conditions, ore body characteristics and wall rock alteration about the no.171 gold vein of dongfeng gold deposit. jiang et al. (2011) primarily discussed the metallogenic regularity of the dongfeng gold deposit. however, the research of the metallogenic geochronology and the source of minerogenic substances is relatively weak. in this paper, rb and sr isotopic analysis has been carried out with the beresite, which has a close relationship with the mineralization of the dongfeng gold deposit. the purpose of this study is to determine the mineralization age and discuss the source of oreforming materials of the dongfeng gold deposit. this study also put forward a new understanding of the mineralization for the dongfeng gold deposit. i2 wang zongyong, lv guxian, zhang xunyu, han xin, zhang yingchun, fan xiao, huo qinglong, xu yaqing 2. geological background and deposit geological characters the dongfeng gold deposit is located in the uplift of the northwestern part of jiaodong area, which belongs to north china craton. the regional strata in the studying area consist mainly of metamorphic rocks in jiaodong rock group, in archaean era and quaternary. the metamorphic rocks are composed of biotite-granulite and amphibolite. faults are well-developed in the dongfeng area. the faults are mainly neand nne-trending. the ne-trending faults are present at the potouqing fault. the ne-trending faults are represented by the luanjiahe fault. the intersection of two sets of faults controls the morphology and distribution of the dongfeng gold deposits (fig 1). the intrusions in the studying area mainly include the linglong biotite adamellite, guojialing granodiorite, and luanjiahe monzonitic granite. the linglong biotite adamellite occurs as batholiths or stocks, with shrimp zircon u-pb age of 160.45±0.83ma (wan, 2014). the guojialing granodiorite occurs as batholiths or stocks, with shrimp zircon u-pb age of 131.86±0.99ma (wan, 2014). the luanjiahe monzonitic granite also occurs as batholiths or stocks, with shrimp zircon u-pb age of 154~152ma (luo and miao, 2002). the recent data of geophysical prospecting, drilling project, and deep exploration reveals that there is a giant guojialing granodiorite batholith in the deep of the studying area, overlain by the linglong biotite adamellite (feng et al., 2009). figure 1. regional geological sketch map of dongfeng deposit in jiaodong area 1. quaternary; 2. jiaodong group; 3. linglong granite; 4. guojialing granite; 5. luanjiahe granite; 6. faults; 7. fractured belt; 8. gold body; 9. gold deposit range; f1. jiuqujiangjia faults; f2. potouqing faults; f3. luanjiahe faults. the main vein of the dongfeng gold deposits is no. 171 vein. the shape of the vein is simple, the vein strikes 60° 70°ne and dips se. (fig. 2). the ore minerals are mainly pyrite, galena, and sphalerite. the ores mainly comprise cataclastic, idiomorphic or hypidiomorphic granular, and gric textures. the structures of the ores can be disseminated, massive, vein, or veinlet (fig. 3). the alteration assemblage includes beresitization, potash feldspathization, chloritization, and carbonatization, occurring as bands in the ore-bearing faults and pervasive alteration of nearby wall rocks. the dominant alteration related to the au mineralization is beresitization. gold mainly exists in polymetallic sulfide. figure 2. no. 160 vertical cross section of the no.171 vein of dongfeng gold deposit. 3. sampling and analytical methods the five samples for the rb and sr isotopic analysis were beresite which formed by hydrothermal metasomatism and closely related to metallogenesis. the samples were collected at the -350m level in orebody no. 171 of the dongfeng gold deposits. the sampling location of each sample and detailed characteristics of the five samples are shown in the table 1 and figure 3. table 1. the sampling locations and the characteristics of five samples the crushing work of the samples was carried out in the analytical laboratory of cnnc beijing research institute of uranium geology. the rb and sr isotopic analysis of 5 samples were performed at the igge of chinese academy of geological sciences. the steps of the rb and sr isotopic analysis are as follows: 1) according to “the specification of testing quality management for geological laboratories” (dz/t01302006), the beresite samples were crushed up to 200 meshes without contamination. 2) the sample powder (ca. 0.1–0.2 g) was separated for the rb-sr analysis. samples were dissolved in teflon bombs with hf+hno3+hclo4 acid (24h). 3) after completely dissolved, the samples were followed by drying.4) then, the samples were treated with 6mol/l hcl and completely converted to chlorate, then dried. 5) samples were dissolved by 0.5mol/l hcl. following the centrifugal separation, the solution has been into cation exchange column and cleared by 1.75mol/l hcl with rb. then, the drying was carried out; the dissolve was cleared by 2.5mol/l hcl with sr; at last, the drying was carried out. the rb and sr isotopic analysis were performed on a phoenix isoprobe-t thermal ionization mass spectrometer. the relative i3rb-sr isotopic geochronology and geological implications of dongfeng gold deposit in jiaodong area figure 3. the characteristics of beresite samples of the dongfeng gold deposit acelandine green beresite; bmassive beresite ores, pyrite distributes in the quartz veins, with the fine vein and dissemination structure; cquartz closely coexists with sericite under the cross-polarized light; qtzquartz; srt-sericite; py-pyrite 4 analytical results the rb and sr isotopic compositions of the five beresite samples in the dongfeng deposit are presented in table 2. the isochron age was calculated with the isoplot program (ludwing, 1998). the rb and sr concentrations of 5 beresite samples range from 71 to 171 ug/g and 114 to 1269 ug/g, respectively. the isotope ratios of 87rb/86sr range from 0.3891 to 3.4915, averaging 1.76702, and the isotope ratios of 87sr/86sr vary from 0.712201 to 0.718079, with an average of 0.7146836. five beresite samples yield a rb–sr isochron age of 125.5±6.7 ma, with an initial 87sr/86sr ratio of 0.711502±0.00006 and an mswd of 0.96 (fig. 4). these results indicate that the dongfeng gold deposit was formed at 125.5±6.7 ma. table 2. rb–sr isotopic analysis of beresite from the dongfeng gold deposit. figure 4. rb–sr isochron for the beresite from the dongfeng gold deposit. 5 discussion 5.1 mineralization age endogenic gold deposits were developed among hydrothermal metasomatic alteration, granulating relatively small alteration minerals which were in a fair preservation of rb and sr. moreover, rb-sr isochron method of altered minerals has been widely used in the study of the mineralization age of gold deposits (jager, 1979; andre and deutsch, 1986; wei et al., 1997; yang and zhou, 1999; xie and hu, 2000; wang et al., 2002). however, as we know, under the effect of the crystal chemical conditions, rb and sr occur as the isomorph in the crystal lattice of the potassium-bearing and calcium-bearing minerals, respectively. therefore, coinstantaneous enrichment of rb and sr in the same kind of mineral seems impossible. also, a part of the sr will be lost in the process of single mineral selection. meanwhile, the purity of the single mineral can notbe ensured because of the tedious separation work. therefore, the hydrothermal metasomatic system of total rocks, with good sealing ability, has been used in the rb-sr isotopic analysis. based on the above theoretical foundation, the beresite which formed by strong hydrothermal metasomatism was collected to be employed in the rb-sr isotopic analysis to obtain the formation of the mineralization age of the dongfeng gold deposit. humidity and temperature are 45% and 20cº respectively. during the analysis, reproducibility and accuracy of sr isotope running have been periodically checked by running the standard reference material nbs 987, with a measured 87sr/86sr ratio of 0.710250±7 (2 σ mean). both of the total procedure blanks for rb and sr were 2×10-10 g. i4 wang zongyong, lv guxian, zhang xunyu, han xin, zhang yingchun, fan xiao, huo qinglong, xu yaqing in recent studies, by using rb-sr, ar-ar, and shrimp u-pb isotope system of ore minerals, the ages of the representative deposits in the jiaodong gold-centralized area have been obtained (table 3). the statistical results show that the ages of representative deposits are similar and range from 128 to 115ma, indicating these deposits mainly formed from the late jurassic to early cretaceous, which is consistent with the previous conclusions of the time of the large scale gold metallogeny in the jiaodong area (fan et al., 2005; chen et al., 2004). the rb-sr isochron age of the beresite from the dongfeng gold deposit in this study is 125.5±6.7 ma. this result is consistent within the age range showed in table 3, indicating the mineralization age of the dongfeng gold deposit is similar to other gold deposits in the jiaodong area. this fact suggests that the gold deposits in the jiaodong area are formed in a relatively short time and under the same mineralization background, and related with the mesozoic tectonic transition. thus, this age (125.5±6.7ma) can be considered as the mineralization age of the dongfeng gold deposit. table 3. representative isotopic ages of the gold deposits in the jiaodong gold-centralized area since 2000 5.2 relationship between the mesozoic magmatism and the dongfeng gold metallization the mesozoic linglong biotite monzonitic granite, guojialing granodiorite, and luanjiahe monzodiorite are extensively distributed in the jiaodong gold concentration area, which was closely related to the gold deposit mineralization. therefore, research on isotope chronology for these mesozoic granites is introduced in great detail. based on the reported zircon shrimp dating results of granites (table 3), combined with the rb-sr isochron ages of the dongfeng gold deposit, this paper discussed the relationship between the mesozoic magmatism and the dongfeng gold mineralization. as shown in table 3, linglong biotite monzonitic granite, luanjiahe monzodiorite and guojialing granodiorite formed from 160 to 153 ma, 154 to 152 ma and 131 to 126 ma, respectively, which indicate that mesozoic granites are mainly formed in two phases: 160 to 152 ma and 131 to 126 ma. these results are a coincidence with the mineralization ages (128 to 115 ma) of the representative gold deposits of the jiaodong gold concentration area and the ore-forming age (125.5±6.7m) of the dongfeng gold deposit. furthermore, the mineralization of the dongfeng gold deposit is contemporaneous with the formation of the guojialing granodiorite, but later than the formation of the linglong biotite monzonitic granite and luanjiahe monzodiorite (fig. 5). there is 155±3ma inherited zircon in the guojialing granodiorite, which indicates that the guojialing granodiorite is considered to be the partial melting product of the jiaodong group and early granites. also, the δ34s values of the goldbearing pyrite in the dongfeng gold deposit and the guojialing granodiorite are consistent, while the pb isotopic values（208pb/206pb） show an increasing trend from linglong biotite monzonitic granite to guojialing granodiorite, then to gold deposits, which reveals an inheritance evolution trend (wan, 2014). therefore, it could be ascertained that the dongfeng gold deposit should go hand in hand with the mesozoic magmatism. this can be embodied in the following respects: (1) during the emplacement of the linglong granite during the late paleoproterozoic–neoproterozoic period, the rock mass not only carried a gold element of itself but also gave rise to the activation and migration of the ore-forming elements disseminated in the archean metamorphic rocks in jiaodong area. then these elements could be enriched in the favorable place for mineralization. (2) during the indosinian-early yanshanian period, the tectonicmagmatic activity was extensively active, the intense interaction between crust and mantle materials occurred, with the formation of the guojialing granodiorite. during the emplacement, the guojialing granodiorite further activated and extracted the gold elements disseminated in the early metamorphic basement and granites. also, the neand nnetrending brittle fracture activities were intense, which provided the migration pathway and ore-forming space for the ore-forming fluids, while the ore fluids continuously interacted with the wall rocks. under a favorable physicochemical condition, the ore-forming materials could precipitate and form the deposit in the convenient place. as a result, the dongfeng gold mineralization was eventually completed. i5rb-sr isotopic geochronology and geological implications of dongfeng gold deposit in jiaodong area table 4. the shrimp zircon u-pb ages of mesozoic granites in gold concentration area of jiaodong note: “——” means no data; data in brackets mean the number of analysis spots. figure 5. the collective age diagram of mesozoic granites diagenesis and dongfeng gold mineralization in jiaodong area. 5.3 sources of ore-forming materials and ore fluids the rb and sr isotopic analysis are an effective tool to discuss the source of ore-forming materials (fan et al., 2005; medford et al., 1983; bell et al., 1989; darbyshire et al., 1996). also, the isr value is considered as an important indicator to discriminate the properties of the source of ore-forming materials (tu et al., 1982; wang and zhou, 2002); the isr value is less than 0.705 for the mantle source, the isr value is more than 0.709 for the crustal source. while the isr value is between 0.705 and 0.709 for the mixture of the mantle and crust sources. however, the indicators of different areas are variable due to the different geological background. in this study, we apply this indicator to discuss the origin of ore-forming materials. the（87sr/86sr）i ratio of beresite in this paper is 0.711502±0.000069（fig. 4); it is thought to be reliable due to the reliability of the rb-sr isochron age of beresite. as mentioned above, the （87sr/86sr）i ratio of beresite is more than 0.709, appearing to indicate that the ore-forming materials are mainly from crust or dominant crust. according to the results of previous researchers, the(87sr/86sr)i ratios of the linglong granite were from 0.710498 to 0.712286 (deng et al., 2011), the (87sr/86sr)i ratios of the guojiaing granodiorite were from 0.71071 to 0.71172 (wang et al., 2014). by the above discriminant indicator, the linglong granite and the guojialing granodiorite were both from the crust. moreover, there was inherited zircon (155±3ma) in the guojialing granodiorite; the age of the inherited zircon was consistent with the diagenetic age of early granites. so we can infer that the guojialing granodiorite is the partial melting product of the early linglong granite, which indirectly supports the view that the two kinds of granites are both from crust sources. as discussed above, the formation of the gold deposit is a complex geological process of gradual enrichment and precipitation for the ore-forming elements. especially, during the indosinian-early yanshanian period, tectonicmagmatic activities caused the upwelling of the mantle materials, the mantle materials interacted with the previous crust materials (such as the rock series of archean jiaodong group and neoproterozoic linglong granite) existing in the lithosphere, then formed a crust-mantle mixed-source type of magma, i.e. guojialing granodiorite. based on the above evidence, it is inevitable that there will always be the mantle material involved in the ore-forming materials and ore fluids. therefore, this article suggests that the ore-forming materials of the dongfeng gold deposit are mainly from the crust. meanwhile, some mantle materials also participate in mineralization of gold deposit. the previous studies also confirmed this judgment (zhou et al., 2002; liu et al., 2003). the rb and sr isotopic analysis of the gold-bearing mineral (pyrite), the carbonate minerals in gold ores, the contemporary magmatic rocks with gold mineralization and basement metamorphic rocks from the typical gold deposits, suggested that the rb and sr isotopic compositions of the gold ore minerals were similar with that of the mantle-derived magmatic rocks, and had a certain i6 wang zongyong, lv guxian, zhang xunyu, han xin, zhang yingchun, fan xiao, huo qinglong, xu yaqing connection with that of the basement metamorphic rocks and the contemporary magmatic rocks with gold mineralization, which indicated that the ore-forming materials of the gold deposits were of multi-sources, including contemporary magmatic rocks, basement metamorphic rocks, and mantle-derived magma. also, based on the hydrogen and oxygen isotopic data of the fluid inclusions in quartz of the dongfeng deposits (wan, 2014), this article uses sheppard ‘s diagram to discriminate the sources of ore-forming fluids (fig. 5). in the hydrogen and oxygen isotopic diagram, the plotted points of the linglong granite and guojialing granodiorite samples are partly located in the usual range of magmatic water. however, another part of the plotted points is close to the range of water in the mantle, which implies that there may be some mantle materials participated in the process of granites diagenesis. moreover, the plotted points of the ore samples of dongfeng deposit are on the left of the range of magmatic water, and near the range of precipitation water, indicating that the ore-forming fluids are possibly mainly from magmatic hydrothermal and mantle with some precipitate water. figure 6. the hydrogen and oxygen isotopic diagram (base map from sheppard, 1977, and data from wan, 2014) 6 conclusions 1) the rb–sr isochron age of beresites for the dongfeng gold deposit in jiaodong gold concentration area is 125.5±6.7ma, indicating this deposit formed in the early cretaceous of the late yanshanian. 2) it could be ascertained that the dongfeng gold deposit should go hand in hand with the mesozoic magmatism. the formation of the gold deposit is a complex geological process of gradual enrichment and precipitation for the ore-forming elements. 3) the ore-forming materials of the dongfeng gold deposit are mainly from the crust. meanwhile, some mantle materials also participate in mineralization of gold deposit. the ore-forming materials of the gold deposits are of multi-sources, including contemporary magmatic rocks, metamorphic basement rocks, and mantle-derived magma. moreover, the ore-forming fluids are possibly mainly from magmatic hydrothermal and mantle with some precipitate water. acknowledgements sincerely give thanks to prof. guxian lv, dr. xin han, dr. tiegang li, dr. jingjing liu for their constructive comments in the process of writing the paper. we also thank xunyu zhang, yingchun zhang, xiao fan, qinglong huo, yaqing xu, jiancheng ren and zhifang liu for their help offered in the field work. this research was financially supported by the national natural science foundation of china (no. 40972061), and the research program of the shandong gold mining (ling long) co., ltd. references andre, l. and deutsch, s. (1986). magmatic 87sr-86sr relicts in hydrothermally altered quartz diorites (brabant massif, belgium) and the role of epidote as a sr filter. contributions to mineralogy and petrology, 92, 104–112. bell, k., anglin, c.d. and franklin, j.m. (1989). sm-nd and rb-sr isotope systematic of scheelites: possible implications for the age and genesis of vein-hosted gold deposits. geology, 17, 500–504. chen, g.j., sun, f.y., li, y.c. and liu, k. (2014). u--pb dating geochemical characteristics and geological significance of guojialing grandiorite in jiaodong peninsula (in chinese with english abstract). global geology 33, 39–47. chen, y.j, pirajno, f., lai, y. and li, c. (2004). metallogenic time and tectonic setting of jiaodong gold province, eastern china (in chinese with english abstract). acta petrologica sinica, 20(4), 907–922. darbyshire, d.p.f., pitfield, p.e.j. and campbell, s.d.g. (1996). late archean and early proterozoic gold-tungsten mineralization in the zimbabwe archean craton: rb-sr and sm-nd isotope constraints. geology, 24(1), 19–22. deng, j., chen, y.m., and liu, q. (2011). the exploration of gold metallogenic system and resources of jiaodong sanshandao fault belt (in chinese). beijing: geology press, 371. fan, h.r., hu, f.f., yang, j.h., shen, k. and zhai, m.g. (2005). fluid evolution and large-scale gold metallogeny during mesozoic tectonic transition in the eastern shandong province (in chinese with english abstract). acta petrologica sinica, 21, 1317–1328. feng, t., sun, z.f., li, w., liu, r.f. and cai, x.n. (2009). geological characteristics and deep resources forecast of the 171 gold veins in dongfeng mining area, linglong gold deposit, shandong province (in chinese with english abstract). gold science and technology, 17, 12–16. goldfarb, r.j. and santosh, m. (2013). the dilemma of the jiaodong gold deposits:are they unique? geoscience frontiers, 5(2), 139–153. doi: http://dx.doi.org/10.1016/j.gsf.2013.11.001 hu, f.f., fan, h.r., yang, j.h., wan, y.s., liu, d.y., zhai, m.g., and jin, c.w. (2004). mineralizing age of the rushan lode gold deposit in the jiaodong peninsula: shrimp u-pb dating on hydrothermal zircon. chinese science bulletin, 49(15), 1629– 1636. doi: 10.1007/bf03184134 hu, s.x., chen, f.j., xu, j.f., guo, k.h. and li, s.m. (1998). relations of gold abundance in the granites and strata to genesis of gold deposits (in chinese with english abstract). acta metallurgica sinica, 4(2), 121–126. jager, e. (1979). introduction to geochronology. in jager e & hunziker j.c. (ed.), lectures in isotopic geology (1–2). berlin, germany: springer. jiang, w.m., lv, g.x., zhang, x.y. and zhang, y.c. (2011). characteristics of the metallization enrichment belt in the dongfeng deposit, linglong ore field and the prediction of the next deep enrichment belt (in chinese with english abstract). acta mineralogica sinica s1,36–37. li, h.m., mao, j.w., shen, y.c., liu, t.b. and zhang, l.c. (2003). arar ages of k-feldspar and quartz from dongji gold deposit, northwest jiaodong, and their significance (in chinese with english abstract). mineral deposits, 22, 72–77. li, j.w., vasconcelos, p.m., zhang, j., zhou, m.f., zhang, x.j. and yang, f.h. (2003). 40ar/39ar constraints on a temporal link between gold mineralization, magmatism, and continental margin transtension in the jiaodong gold province, eastern china. the journal of geology 111(6), 741–751. doi: 10.1086/378486 lin, b.l. and li, b.l. (2013). geochemistry, u-pb dating, lu-hf isotopic analysis and geological significance of linglong granite in jiaodong peninsula (in chinese with english abstract). journal of chengdu university of technology (science & technology edition) 40, 147–159. liu, j.m., ye, j., xu, j.h., sun, j.g. and shen, k. (2003). c-o and srnd isotopic geochemistry of carbonate minerals from gold deposits in east shandong, china (in chinese with english abstract). acta petrologica sinica, 19, 775–784. i7rb-sr isotopic geochronology and geological implications of dongfeng gold deposit in jiaodong area ludwing, k.r. (1998). using isoplot/ex: age of chronological toolkit for microsoft excel, version 1.00. berkeley geochronnology center special publication, pp.1–4. luo, z.k. and miao, l.c. (2002). granite and gold deposit in the zhao-lai area, jiaodong area (in chinese). beijing: metallurgical industry press, 143. lv, g.x., guo, t., shu, b., shen, y.k., liu, d.j., zhou, g.f., ... ha, b.h. (2007). study on the multi-level controlling rule for tectonic system in jiaodong gold-centralized area (in chinese with english abstract). geotectonica et metallogenia 31, 193–204. mao, j.w., li, h.m., wang, y.t., zhang, c.q. and wang, r.t. (2005). the relationship between mantle-derived fluid and gold ore-formation in the eastern shandong peninsula: evidences from d-o-c-s isotopes (in chinese with english abstract). acta geologica sinica, 79(6), 839–857. medford, g.a., maxwell, r.j. and armstrong, r.l. (1983). 87sr/86sr ratio measurements on sulfides, carbonates and fluid inclusion from pine point, northwest territories, canada: an 87sr/86sr ratio increase accompanying the mineralizing process. economic geology, 78(7), 1375–1378. doi: 10.2113/gsecongeo.78.7.1375 miao, l.c., zhu, c.w., zhai, y.s., guan, k. and luo, z.k. (1999). discussion on the relationship between granitoid and gold mineralization in the zhaoye gold belt, shandong province (in chinese with english abstract). gold geology, 5(4), 7–11. song, m.c. (2008). the composing, setting and evolution of tectonic units in shandong province (in chinese with english abstract). geological survey and research 31, 165–175. song, m.c., yin, p.h., cui, s.x., xu, j.x., zhou, m.l., jiang, h.l., ... cao, c.g. (2013). thermal uplifting-extension ore-forming theory and its prospecting significance in jiaodong gold deposit (in chinese with english abstract). shandong land and resources 29, 1–12. sun, j.g., hu, s.x. and zhao, y.y. (2000). a preliminary discussion on the metallogenic model of gold deposits in jiaodong area (in chinese with english abstract). mineral deposits, 19, 26–34. tu, g.z., zhang, y.q. and zhao, z.h. (1982). preliminary studies on two alkalirich intrusive belts in south china. in xu, k.q. and tu, g.z. (eds.), geology of granites and their metallogenetic relations. beijing: science press, 33–54. wan, d. (2014). shandong jiaodong zhaoping fault belt prediction rules of gold mineralization and mineralization in north of the zhao-ping fault belt, jiaodong area, shandong province (in chinese with english abstract). changchun: jilin unicersity (ph. d thesis), 1–200. wan, t.f., teyssier, g., zeng, h.l., zhou, w.x. and tikoff, b. (2000). the emplacement mechanism of the linglong granites, shandong province (in chinese with english abstract). science china earth sciences, 30, 337–344. wang, d.z. and zhou, x.m. (2002). petrogenesis of the late mesozoic granitic volcanic-intrusions and crust evolution of the southeast of china (in chinese). beijing: science press, 297. wang, z.l., zhao, r.x., zhang, q., lu, h.w., li, j.l. and chen, y. (2014). magma mixing for the high ba-sr guojialing-type granitoids in northwest jiaodong peninsula: constraints from petrogeochemistry and sr-nd isotopes (in chinese with english abstract). acta petrologica sinica 30, 2595–2608. wang, y.w., zhu, f.s. and gong, r.t. (2002). study on the metallogenic chronology of gold deposits in jiaodong gold concentration zone (in chinese with english abstract). gold geology, 8, 48–55. wei, c.s., zheng, y.f. and tu, g.z. (1997). the geological significance of the estremely hydrothermal metasomatized auriferous ore rb-sr isochron (in chinese with english abstract). acta geoscientia sinica, 18, 290–298. xie, g.q. and hu, r.z. (2001). some progress in geochronology of gold ore deposits (in chinese with english abstract). geologygeochemistry, 29, 57–62. xu, x.l., gao, j.g., wang, f. and jia, f.j. (2013). research and application of the ore body three-dimensional mathematical model based on surpac: example from shandong dongfeng deposit 171 pulse (in chinese with english abstract). geological science and technology information 32, 202–206. yang, j.h. and zhou, x.h. (1999). study on dating of the gold mineralization (in chinese with english abstract). geological science and technology information, 18(1), 85–88. yang, j.h. and zhou, x.h. (2000). rb-sr isochron age and mineralogenetic epoch of ore and gold-carrying mineral of the linglong gold field in the eastern shandong province (in chinese with english abstract). chinese science bulletin, 45(14), 1547–1553. yang, j.h. and zhou, x.h. (2001). rb-sr, sm-nd and pb isotope systematics of pyrite: implications for the age and genesis of lode gold deposits. geology 29, 711–714. yang, j.h., zhou, x.h. and chen, l.h. (2000). dating of gold mineralization for super-large altered tectonite-type gold deposits in northwestern jiaodong peninsula and its implications for gold metallogeny (in chinese with english abstract). acta petrologica sinica 16(3), 454–458. yang, l.q., deng, j., goldfarb, r.j., zhang, j., gao, b.f. and wang, z.l. (2014b). 40ar/39ar geochronological constraints on the formation of the dayingezhuang gold deposit: new implications for timing and duration of hydrothermal activity in the jiaodong gold province, china. gondwana research 25(4), 1469–1483. yang, l.q., deng, j., wang, z.l., zhang, l., guo, l.n., song, m.c. and zheng, x.l. (2014a). mesozoic gold metallogenic system of the jiaodong gold province, eastern china (in chinese with english abstract). acta petrologica sinica 30(9), 2447–2467. yiao, f.l., liu, l.d., kong, q.c. and gong, r.t. (1990). the vein-type gold deposits in the northwest of the jiaodong area (in chinese). changchun: jilin science and technology press, 234. zhai, m.g., fan, h.r., yang, j.h. and miao, l.c. (2004). large-scale cluster of gold deposits in east shandong: anorogenic metallogenesis (in chinese with english abstract). earth science frontiers 11, 85–98. zhai, y.s., miao, l.c., xiang, y.c., deng, j. and wang, j.p. (2002). preliminary analyses of metallogenic systems about the greenstone belt gold deposits, north china craton. earth science, 27, 522-531. zhang, l.c., shen, y.c., liu, t.b., zeng, q.d., li, g.m. and li, h.m. (2003a). 40ar/39ar and rb-sr isochron dating of the gold deposits on northern margin of the jiaolai basin, shandong, china. science in china series d: earth sciences 469(7), 708–718. zhang, x.o., cawood, p.a., wilde, s.a., liu, r.q., song, h.l., li, w. and snee, l.w. (2003b). geology and timing of mineralization at the cangshang gold deposit, north-western jiaodong peninsula, china. mineralium deposita 38(2), 141–153. doi: 10.1007/s00126-002-0290-7 zhou, x.h., yang, j.h. and zhang, l.c. (2003). metallogenesis of superlarge gold deposits in jiaodong region and deep processes of subcontinental lithosphere beneath north china craton in mesozoic. science in china series d: earth sciences 46, supplement 1, 14–25. a modelling study by factorial design on gnss positioning veli ilci1, yasemin sisman1 department of geomatics engineering, faculty of engineering, ondokuz mayis university, samsun, turkey. corresponding author: veli.ilci@omu.edu.tr keywords: factorial design; regression analysis; gnss error sources; cors; gnss palabras clave: diseño factorial; análisis de regresión; fuentes de error gnss; estaciones de referencia en funcionamiento continuo; sistemas globales de navegación satelital issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n4.95060 earth sciences research journal earth sci. res. j. vol. 25, no. 4 (december, 2021): 391-396 g e o d e sy record manuscript received: 14/04/2021 accepted for publication: 03/11/2021 abstract although researchers have widely studied the analysis and modeling of error sources on global navigation satellite systems positioning, some of these errors have not been eliminated significantly. only some of the global navigation satellite system’s data are modeled. the present work was undertaken to determine the effect of different variables: season, the number of visible satellites, and dilution of precision on the efficiency of horizontal and vertical cors (continuously operating reference stations) positioning. the cors data was collected at 14 different test points during 600 epochs with 1-second intervals for this aim. factorial designs supply an efficient solution to understand the impact of several factors on a response variable. a full factorial design with three factors at two levels was applied for these purposes. the main and the interaction effects of factors were analyzed on the cors horizontal and vertical positioning. according to the full factorial design results, while all main and interaction effects of factors significantly affected the cors horizontal positioning error, some elements did not affect the cors vertical positioning error. also, the regression equations were obtained for all situations to investigate the other level of selected factors in the response variables. estudio de modelado por diseño factorial en posicionamiento de sistemas globales de navegación satelital resumen a pesar de que los investigadores han analizado y modelado ampliamente las fuentes de error de posicionamiento en los sistemas globales de navegación satelital (gnss, global navigation satellite system), algunos de estos errores no se han eliminado significativamente. y solo parte de esta información de los sistemas satelitales de navegación global ha sido modelada. este trabajo se realiza con el fin de determinar los efectos de diferentes variables: temporada, número de satélites visibles, e imprecisión en la eficiencia de posicionamiento horizontal y vertical cors (estaciones de referencia en funcionamiento continuo). la información cors se recolectó en 14 escenarios de prueba durante 600 períodos y con intervalos de un segundo para este objetivo. los diseños factoriales proveen una solución eficiente para entender el impacto de varios factores ante una respuesta variable. en este trabajo se aplicó un diseño factorial completo con tres factores en dos niveles. los efectos principales y de interacción de factores se analizaron con el posicionamiento cors horizontal y vertical. de acuerdo con los resultados del diseño factorial completo, mientras que los efectos principales y de interacción de factores afectaron significativamente el error de posicionamiento horizontal, algunos efectos principales y de interacción de factores no afectaron el posicionamiento vertical cors. también las ecuaciones de regresión se obtuvieron para todas las situaciones con el fin de investigar los otros niveles de factores seleccionados en la respuesta de variables. how to cite item ilci, v., & sisman, y. (2021). a modelling study by factorial design on gnss positioning. earth sciences research journal, 25(4), 391-396. https://doi. org/10.15446/esrj.v25n4.95060 mailto:veli.ilci@omu.edu.tr https://doi.org/10.15446/esrj.v25n1.74167 https://doi.org/10.15446/esrj.v25n4.95060 https://doi.org/10.15446/esrj.v25n4.95060 392 veli ilci, yasemin sisman introduction global navigation satellite systems (gnss) have been commonly used for positioning for decades. however, the positioning accuracy obtained by the gnss is prone to error sources, which can be primarily categorized as clock-related errors, signal propagation errors, system errors, and intentional errors (selective availability and signal jamming etc.) (karaim et al., 2018). the quality of the obtained positioning accuracy is expressed with two metrics. the first metric is the user equivalent range error (uere) that defines the total errors on the pseudorange. the range between satellites and the receiver is called the pseudorange due to the noise and errors. the second quality metric is the dilution of precision (dop) that specifies the effect of geometry on the relationship between positioning and measurement (chen et al., 2013; li et al., 2018; teng et al., 2015; teng & wang, 2016). accurate positioning can be obtained with the good spatial spread of the visible satellites that provide lower dop values (banerjee & bose, 1996; verma et al., 2019). dop is utilized for the optimum selection of visible satellites (teng & wang, 2014) and is categorized as the horizontal dilution of precision (hdop), vertical dilution of precision (vdop), and time dilution of precision (tdop). the combination of hdop and vdop is named position dilution of precision (pdop), and the variety of pdop and tdop is called the geometric dilution of precision (gdop). global positioning system (gps) is still the most reliable satellite-based positioning system designed to have at least six satellites at any part of the earth (busznyák et al., 2019; wang et al., 2011). however, in some environments such as urban areas, heavy tree cover, open-pit mines, etc., the desirable number of gps signals cannot reach the receiver due to the signal blockage by obstacles (alkan et al., 2015). in this case, a weak or no positioning solution can be obtained. in the last decades, multi-gnss solutions have been investigated to exceed this limitation and acquire more accurate and reliable positioning results (alkan et al., 2017). in the last years, multi-constellation solutions have been worked with gps/glonass, gps/beidou, gps/galileo, gps/ glonass/beidou, gps/glonass/galileo, and gps/glonass/ galileo/beidou in many studies. the multi-constellation positioning ensures improved positioning solutions using more visible satellites with better geometric distribution and more satellite availability (wang et al., 2019). the tropospheric and ionospheric errors are the two primary error sources related to signal propagation. the seasonal changes of atmospheric variables like temperature, water vapour, hydrostatic pressure, and humidity significantly affect these errors. the effect of seasonal variation on the gnss accuracy was analyzed by dogan et al. (2014). they concluded that the gnss positioning accuracy in summer is better than that in winter. additionally, saracoglu & sanli (2020) studied the effect of seasonal changes on gnss positioning in different world regions. they concluded that the seasonal impact on positioning accuracy changes according to climate zones. in zheng et al. (2018), zenith tropospheric delay accuracy was investigated on the gnss data. the results showed that the accuracy in winter is better than in the other seasons. many seasonal atmospheric factors affect gnss signals quality and positioning. therefore, the results of the seasonal effects on gnss positioning accuracy differ from the others. the analysis of the gnss measurements is critical in geodesy due to working conditions in the field. some experiments can be performed, and the results can be investigated to shorten the working time in the field. in the process of obtaining measurements, a lot of factors affect the gnss measurements. the experimental design methods can be used frequently to determine the significant factors on the response variable. the primary and interactive effects of dependent variables (x) can be determined using the experimental design process on the response variable (y). the statistical analysis and graphical presentation make the interpretation of the results easier. also, a suitable experimental design model will reduce the required data and spend time on the field (çoruh et al., 2012; seltman, 2018). there are different experimental designs such as full factorial, plackettburman, tagucci box–behnken design, and central composite design (gündoğdu et al., 2016). full factorial design (ffd) is a widely used procedure to determine dependent variables’ primary and interactive effect on the response variable in different levels, such as 2p, 2p-k, 3p (george et al., 2005; navidi, 2008; sisman, 2014a). the experimental design has been used in many engineering applications. geomatics engineering uses many types of application data from different sources for positioning. the size of application data has increased a lot in recent years, and therefore the analysis of application data has become more critical. experimental design is one of the analyzing methods of application data, but the experimental design is limited in geomatics. the main aim of this study is to investigate the factorial effects of the season, the number of satellites, and dop on the cors horizontal and vertical positioning errors using ffd. although there are some studies in geodesy that have investigated the factor effects on the gnss positioning (abad & suárez, 2004; ahmad, 2015; brenneman et al., 2010; cai & gao, 2007; catania et al., 2020; pirti, 2008; raghunath et al., 2011; stone & powell, 1998; svabensky & weigel, 2004; wielgosz et al., 2019; wing et al., 2008; yoshimura & hasegawa, 2003), there is no study in literature for network-real-time-kinematic (nrtk) using ffd. the factorial effects at two levels were studied using a 23 ffd in this study. the regression equation as y = f(x) was obtained from the nrtk horizontal and vertical positioning error results. the rest of this paper is organized as follows. in section 2, gnss positioning and the factorial design are explained in detail. the results and discussions are presented in section 3. finally, the conclusions are given in section 4. material and methods gnss positioning this study was conducted along a 280 km long route from samsun to the kırıkkale provinces of turkey. fourteen geodetic points were established with 20 km intervals through the course. the locations of geodetic points and the stations of the continuously operating reference stations network of turkey (called cors-tr or tusaga-aktif) around the geodetic points are shown in figure 1. figure 1. study area. tusaga-aktif delivers real-time gnss correction data to the receivers of nrtk systems 24/7 like the other countries. tusaga-aktif system was established with 146 reference stations using baselines ranging 70-100 km across turkey and northern cyprus (aykut et al., 2015; gülal et al., 2013). this system transmits the correction data with virtual reference stations (vrs), flachen korrectur parameter (fkp), and master auxiliary concept (mac) techniques in gsm, networked transport of rtcm via internet protocol (ntrip), and radio connections by radio technical commission for maritime services (rtcm) 3.0 and higher protocol (bakici et al., 2017). the obtained 3d coordinates are in the international terrestrial reference frame 1996 (itrf96) datum and 2005.0 epoch with cm-level accuracy (i̇lçi, 2019). a trimble r10 gnss receiver for gnss observations was used. the site surveys were conducted in two different periods (winter and summer) to determine the seasonal effect on gnss measurement accuracies. in winter, the site surveys were conducted on two consecutive days (december 1st and 2nd, 2017), and site surveys in summer were executed on june 2nd and 3rd, 2018. the gnss receiver was mounted on the geodetic points to determine the 393a modelling study by factorial design on gnss positioning reference position of the sites in the winter and summer periods. (soler et al., 2006) revealed that 2-hour static sessions are sufficient to obtain the desired accuracy for 280 km baseline length between the base and the rover stations. (firuzabadì & king, 2012) obtained mm level horizontal and vertical precisions where the baseline lengths were less than 200 km in 2 hours. in this study, taking into account the baseline lengths between 1 to 100 kilometres and the open sky area conditions, we determined the gnss static observation duration as 100 minutes. simultaneously, the receiver was connected to the tusagaaktif service and the cors data were collected for 10 minutes with 1-second intervals, and the elevation mask was set to 10 degrees. although the tusagaaktif service only transmits the correction data related to gps and glonass constellations, we have observed all gps, glonass, galileo, beidou, and qzss constellations at all geodetic points for further analyses. 2.2 factorial design the factorial design investigates the effect of two or more factor levels on the response variable using designed experiments. thus, the mathematical model can be obtained between the main and interactive effects of factors and response variables; also, the time, effort, and operational cost can be reduced (george et al., 2005; ismail et al., 2008; montgomery, 2001; navidi, 2008). if the factors have a main or interactive effect on the response variable in the experimental design, their results can be determined as significant. although there are several factorial design methods, the 2p ffd is the most preferred method. 2 and p show different levels and the number of factors, respectively (box et al., 2005; gygi et al., 2006; ismail et al., 2008; wu & hamada, 2009). the two aims were realized by ffd. one of them is mathematical model development between selected factors and response variables. this mathematical model is a regression equation, including main and interactive effects, given in equation (1). (1) here; b0, bi, and bij are the main and interactive effects’ coefficients; and are the factors; ε is the error of the mathematical model. the second aim of ffd is to test significance of factors on the obtained mathematical model. in this stage, the null and alternative hypotheses are established. the main and interactive effects of factors are tested according to the selected significance level using the analysis of variance (anova). more details of ffd can be found in (george et al., 2005; gygi et al., 2006; montgomery, 2001; navidi, 2008). results and discussion in this study, the season, number of satellites, and dop data were taken as factors; the cors’s horizontal and vertical positioning errors were taken as a response variable, and 23 ffd was established. the level of factors was taken as high (+1) and low (-1). the levels of the season, satellite number and dop were selected for summer as (high) and for winter as (low); >15 (high) and <15 (low); <1.4 (high) and >1.4 (low), respectively. the high (+1) and low (-1) levels of factors are shown in table 1. table 1. the levels of factors. factor low level (-1) high level (+1) season (x1) winter summer satellite number (x2) <15 >15 dop (x3) >1.4 <1.4 the design matrix of application data was taken from the data obtained from cors-tr, as explained in section 2.1 (table 2). minitab 16 statistical software was used for all the analyses of the experimental process (minitab, 2021). table 2. the design matrix of application data. run no. factor rms x1 x2 x3 horizontal errors (mm) vertical errors (mm) 1st trial 2nd trial 1st trial 2nd trial 1 winter <15 >1.4 15.2 15.3 20.6 20.8 2 summer <15 >1.4 10.1 10.2 16.5 16.7 3 winter >15 >1.4 9.11 9.6 14.3 14.6 4 summer >15 >1.4 10.4 11.0 17.0 17.9 5 winter <15 <1.4 10.1 10.3 15.3 15.7 6 summer <15 <1.4 10.3 10.5 15.8 17.5 7 winter >15 <1.4 10.5 11.1 16. 3 17.8 8 summer >15 <1.4 9.2 9.9 15.0 16.0 the descriptive statistic is a supply to understand the data by users using central tendency (mean, median, mode, quartiles) or variability (range variance, skewness, etc.) (sharma, 2019). there are several graphical methods to represent the descriptive statistic parameters (potter, 2006). the box plot analysis of application data was realized (table 3 and fig. 2). table 3. results of the descriptive statistical analysis of application data. application data mean (mm) st.dev. (mm) minimum (mm) median (mm) maximum (mm) horizontal errors 10.8 1.8 9.1 10.3 15.3 vertical errors 16.7 1.9 14.3 16.4 20.8 figure 2. the box plot of application data. the box plot represents that the min, max, q1 (%25), q2 (median), q3 (75%) and mean values of application data. firstly, the mathematical model was obtained for the application data (table 4). the significance of factors was determined using a hypothesis test. if the p-value of the factor is bigger than the significance value (selected as %5 in this study), it is decided that the factor is insignificant in the regression model. in table 4, coef. and p-values, s and r-sq represent the regression equation coefficient, the student-test value of factors, standard deviation and the ratio of explained variation to total variation, respectively. it was seen that all main and interactive factors except x1 and x1*x3 on the vertical positioning had a significant effect on the response variable. in this case, the insignificant term should be removed, and the regression coefficient determined with substantial factors. the regression equation of the response variable, for horizontal and vertical positioning, is given according to equations 2 and 3, respectively. st.dev 394 veli ilci, yasemin sisman (2) (3) the response variable can be increased or decreased according to the multiplication of coefficients and the levels of factors. since the corshor. and corsver. are supposed to decrease; this multiplication is desired to have a negative sign. in this study, the main effects should have negative signs in equation 2 and equation 3. therefore, the level factors can be taken as (+1) for equations 2 and 3 for the main effects. on the other hand, the interactive effects of factors should be considered. according to the magnitude of coefficients in equation 2, since the coefficients of x1*x2 and x2*x3 are bigger than x1 and x3 coefficients and the coefficient of x1*x2*x3 are the maxima the x1, x2, and x3 should be taken as (-1), (+1) and (-1). also, according to the magnitude of coefficients in equation 3, since the coefficients of x2*x3 is bigger than x2 and x3 coefficients and the coefficient of x1*x2*x3 is the maximum, the x1, x2, and x3 should be taken as (-1), (+1) and (-1). the r2 is the rate of explained variability and total variability and describes the goodness of fit for the model (mason et al., 2003; seltman, 2018; sisman, 2014b). in this study, r2 was equal to 98.56% and 92.91% for the cors horizontal and vertical positioning errors, respectively. this means that 98.56% and 92.91% of the application data can be explained with the obtained mathematical model. moreover, the anova test is realized in the ffd. the anova results of horizontal and vertical cors positioning errors were obtained for application data (table 5). it is decided that the effect has a significant impact on the response variable if the p-value is more powerful than the selected significance value (5%). in table 5, df, adj ss, adj ms, p-value represent the degrees of freedom, adjusted sums of squares, adjusted mean squares, the fisher-test value of factors, respectively. some graphical representations such as the normal plot of the standardized effects can also obtain the effects of factors on the response variable. the normal probability plots are used to estimate the significance of interaction effects in a factorial design (kavuri et al., 2009). the magnitude, direction, and importance of the effects can be determined using the normal probability plot of the effects. figure 3 illustrates the normal plot for application data. figure 3. normal plot for horizontal (left) and vertical (right) cors positioning. if the effects are close to the distribution fit line, they have no significant impact on the response variable. it was also seen that although all main and interactive results were significant on the cors horizontal positioning error, the x1 and x1*x3 effects were not significant on the cors vertical positioning error. the change effect of the factor level can be seen with the main effect plots. the main effect plots are given in figure 4. figure 4. main effects plot for cors horizontal (left) and vertical (right) positioning error. it was determined that the (+1) level of all factors had reduced the horizontal and vertical cors positioning errors. also, it was seen that the x1 (i.e. the seasonal effect) was not a significant effect on the response variable of the vertical component. since the effects have a high slope, it is decided that selecting factor levels is suitable for this study. the interactive effect of the factor level can be obtained from interaction plot. for this study, the interaction plots were given in figure 5. table 4. estimated effects and coefficients for cors horizontal (left) and vertical (right) positioning errors. term effect coef. p-value term effect coef. p-value constant 10.8110 0.000 constant 16.740 0.000 x1 -1.2030 -0.6015 0.000 x1 -0.392 -0.196 0.288 x2 -1.4069 -0.7035 0.000 x2 -1.259 -0.629 0.006 x3 -1.1158 -0.5579 0.000 x3 -1.140 -0.570 0.011 x1*x2 1.2480 0.6240 0.000 x1*x2 1.097 0.549 0.013 x1*x3 0.7005 0.3502 0.002 x1*x3 0.175 0.087 0.625 x2*x3 1.2840 0.6420 0.000 x2*x3 1.483 0.742 0.003 x1*x2*x3 -1.9609 -0.9804 0.000 x1*x2*x3 -2.436 -1.218 0.000 s=0.299 r-sq=98.56% s=0.688 r-sq=92.91% table 5. anova for horizontal (left) and vertical (right) cors positioning. source df adj ss adj ms p-value source df adj ss adj ms p-value x1 1 5.7893 5.7893 0.000 x1 1 0.6132 0.6132 0.288 x2 1 7.9177 7.9177 0.000 x2 1 6.3372 6.3372 0.006 x3 1 4.9800 4.9800 0.000 x3 1 5.1951 5.1951 0.011 x1*x2 1 6.2300 6.2300 0.000 x1*x2 1 4.8155 4.8155 0.013 x1*x3 1 1.9627 1.9627 0.002 x1*x3 1 0.1225 0.1225 0.625 x2*x3 1 6.5947 6.5947 0.000 x2*x3 1 8.7977 8.7977 0.003 x1*x2*x3 1 15.3799 15.3799 0.000 x1*x2*x3 1 23.7319 23.7319 0.000 395a modelling study by factorial design on gnss positioning it was seen that while only the x1*x3 interaction does not have a significant effect on the cors vertical positioning error, the other interaction effects were substantial in both the cors horizontal and vertical positioning errors. if the 2-effect factor levels have the same slope, it will not be meaningful on the result variable. the main results of the study can be summurazed as follows; • r2 values were satisfactory (98.56% for cors horizontal, 92.91% for cors vertical positioning errors). • all main and interactive effects of factors were significant on cors horizontal positioning errors at the 5% level. • season (x1) and season*dop (x1*x3) effects of factors were insignificant on cors vertical positioning errors at the 5% level. • the 3-way interaction effect season*satellite number*dop(x1*x2*x3) had the most significant coefficient on the regression models of cors horizontal and vertical positioning errors. conclusions this study aimed to investigate the effects of factors on cors positioning using statistical design experiments. for this, a 23 full-factorial design (three factors at two levels) was established to evaluate the main and interaction effects of season, the number of satellites, and dop on the cors positioning. the main conclusions for the study are shown below: • the minimum values of cors horizontal and vertical positioning errors were 9.1 mm and 14.3 mm in the case of the winter season, bigger than 15 satellite numbers, and bigger than 1.4 dop, respectively. • the maximum values of cors horizontal and vertical positioning errors were 15.3 mm and 20.8 mm, in the case of the winter season, lower than 15 satellite numbers, and bigger than 1.4 dop, respectively. • the factors levels should be taken as (+1) levels (summer, >15, and < 1.4) considering the main effects of factors. but it is seen that when the interactive effects were added to the regression model, the factors should be taken as (-1) levels for season and dop, (+1) level for the number of satellites. • the value of corshor. was calculated 9.6 mm with (+1) levels (summer, >15 and <1.4) of factors. but the value of corshor. was calculated 9.4 mm with (-1) levels (winter and >1.4) of seasons and dop and (+1) level (>15) of the number of satellites because of the interactive effects of factors on the response variable. • while the value of corsver. is calculated 15.6 mm (+1) levels (summer, >15 and <1.4) of factors. but the value of corshor. was calculated 14.1 mm with (-1) levels (winter and >1.4) of seasons and dop and (+1) level (>15) of the number of satellites. • in the light of these explanations, it can be said that the main and interactive effect of the number of satellites was always (+1) levels (>15); the other factor levels (seasons and dop) can be changed for application data. many different application data sources must be analyzed attentively in applied sciences such as geomatics engineering. it is seen that the experimental design is quite helpful and practical for positioning applications, as it demonstrates the factor effects on positioning error. the statistical analysis and graphical presentation of the experimental setup are easier to understand the application data. also, the regression equations derived from the experimental design allow investigating different levels of selected factors on the corshor. and corsver. easily. it is suggested that the experimental design studies must be carried out in other areas of geomatic engineering such as remote sensing, photogrammetry and surveying. references abad, p., & suárez, j. p. (2004). bi-factorial analysis for resolution of gps equations. journal of computational and applied mathematics, 164–165, 1–10. https://doi.org/10.1016/j.cam.2003.11.006 ahmad, k. a. (2015). reliability monitoring of gnss aided positioning for land vehicle applications in urban environments. ph.d. thesis, universite de toulouse, france. alkan, r. m., ilçi, v., ozulu, i. m. & saka, m. h. (2015). a comparative study for accuracy assessment of ppp technique using gps and glonass in urban areas. measurement, 69, 1-8. https://doi.org/10.1016/j. measurement.2015.03.012 alkan, r. m., saka, m. h., ozulu, i. m. & i̇lçi, v. (2017). kinematic precise point positioning using gps and glonass measurements in marine environments. measurement, 109, 36-43. https://doi.org/10.1016/j. measurement.2017.05.054 aykut, n. o., gülal, e. & akpinar, b. (2015). performance of single base rtk gnss method versus network rtk. earth sciences research journal, 19(2), 135–139. https://doi.org/10.15446/esrj.v19n2.51218 bakici, s., erkek, b., i̇lbey, a. & kulaksiz, e. (2017). businnes model of cors-tr (tusaga-aktif). 4th international workshop on geoinformation science: geoadvances 2017, safranbolu, karabuk, turkey, october, iv-4-w4(2017), 109-116. https://doi.org/10.5194/isprs-annals-iv-4-w4-109-2017 banerjee, p. & bose, a. (1996). evaluation of gps pdop from elevation and azimuth of satellites. indian journal of radio and space physics, 25, 110-113. box, g. e. p., hunter, j. s. & hunter, w. g. (2005). statistics of experimenters: design, innovation, and discovery. john wiley & sons. 2nd ed., wiley, hoboken, 672 pp. brenneman, m. t., morton, y. t. & zhou, q. (2010). gps multipath detection with anova for adaptive arrays. ieee transactions on aerospace and electronic systems, 46(3), 1171–1184. https://doi.org/10.1109/ taes.2010.5545181 busznyák, t., pálfi, g., & lakatos, i. (2019). on-board diagnostic-based positioning as an additional information source of driver assistant systems. acta polytechnica hungarica, 16(5), 217–234. https://doi. org/10.12700/aph.16.5.2019.5.12 cai, c. & gao, y. (2007). precise point positioning using combined gps and glonass observations. journal of global positioning systems, 6(1), 13–22. catania, p., comparetti, a., febo, p., morello, g., orlando, s., roma, e. & vallone, m. (2020). positioning accuracy comparison of gnss receivers used for mapping and guidance of agricultural machines. agronomy, 10(7), 924. https://doi.org/10.3390/agronomy10070924 chen, c. s., chiu, y. j., lee, c. t., & lin, j. m. (2013). calculation of weighted geometric dilution of precision. journal of applied mathematics, 953048, 1-10. https://doi.org/10.1155/2013/953048 çoruh, s., elevli, s. & geyikçi, f. (2012). statistical evaluation and optimization of factors affecting the leaching performance of copper flotation waste. the scientific world journal, 758719, 1–8. https://doi. org/10.1100/2012/758719 dogan, u., uludag, m. & demir, d. o. (2014). investigation of gps positioning accuracy during the seasonal variation. measurement, 53, 91-100. https://doi.org/10.1016/j.measurement.2014.03.034 firuzabadì, d., & king, r. w. (2012). gps precision as a function of session duration and reference frame using multi-point software. gps solutions, 16, 191–196. https://doi.org/10.1007/s10291-011-0218-8 george, m. l., rowlands, d., price, m., & maxey, j. (2005). the lean six sigma pocket toolbook. mcgraw hill, new york. gülal, e., erdoǧan, h., & tiryakioǧlu, i̇. (2013). research on the stability analysis of gnss reference stations network by time series analysis. digital signal processing, 23(6), 1945-1957. https://doi.org/10.1016/j. dsp.2013.06.014 https://doi.org/10.1016/j.cam.2003.11.006 https://doi.org/10.1016/j.measurement.2015.03.012 https://doi.org/10.1016/j.measurement.2015.03.012 https://doi.org/10.1016/j.measurement.2017.05.054 https://doi.org/10.1016/j.measurement.2017.05.054 https://doi.org/10.15446/esrj.v19n2.51218 https://doi.org/10.5194/isprs-annals-iv-4-w4-109-2017 https://doi.org/10.1109/taes.2010.5545181 https://doi.org/10.1109/taes.2010.5545181 https://doi.org/10.12700/aph.16.5.2019.5.12 https://doi.org/10.12700/aph.16.5.2019.5.12 https://doi.org/10.3390/agronomy10070924 https://doi.org/10.1155/2013/953048 https://doi.org/10.1100/2012/758719 https://doi.org/10.1100/2012/758719 https://doi.org/10.1016/j.measurement.2014.03.034 https://doi.org/10.1007/s10291-011-0218-8 https://doi.org/10.1016/j.dsp.2013.06.014 https://doi.org/10.1016/j.dsp.2013.06.014 396 veli ilci, yasemin sisman gündoğdu, t. k., deniz, i̇., çalişkan, g., şahin, e. s., & azbar, n. (2016). experimental design methods for bioengineering applications. critical reviews in biotechnology, 36(2), 368-388. https://doi.org/10.3109/073 88551.2014.973014 gygi, c., williams, b., & gustafson, t. (2006). six sigma workbook for dummies. wiley, indiana. 320 pp. i̇lçi, v. (2019). accuracy comparison of real-time gnss positioning solutions: case study of mid-north anatolia. measurement, 142, 40-47. https:// doi.org/10.1016/j.measurement.2019.04.067 ismail, a. a., el-midany, a. a., ibrahim, i. a., & matsunaga, h. (2008). heavy metal removal using sio2-tio2 binary oxide: experimental design approach. adsorption, 14(2008), 21–29. https://doi.org/10.1007/ s10450-007-9042-4 karaim, m., elsheikh, m., & noureldin, a. (2018). gnss error sources. chapter 4 in multifunctional operation and application of gps. intechopen, rijeka. https://doi.org/10.5772/intechopen.75493 kavuri, n. c., sahu, s., & kundu, m. (2009). bioleaching of zinc sulphide ore using thiobacillus ferrooxidans: screening of design parameters using statistical design of experiments. the iup journal of chemical engineering, 1(1), 39–53. https://ssrn.com/abstract=1487464 li, c., teng, y. & kang, r. (2018). some remarks on geometric dilution of precision (gdop) at user level in multi-gnss positioning. advances in space research, 62(11), 3048-3052. https://doi.org/10.1016/j. asr.2018.08.029 mason, r. l., gunst, r. f., & hess, j. l. (2003). statistical design and analysis of experiments: with applications to engineering and science. 2nd ed. john wiley & sons, hoboken, new jersey, usa. https://doi.org/10.1002/0471458503 minitab. (2021). statistical software. https://www.minitab.com/en-us/ (last accessed october 2021) montgomery, d. (2001). editorial: research industrial statistics-part i. in: quality and reliability engineering international. 17(6), 3-4, john wiley & sons, ltd. https://doi.org/10.1002/qre.449 navidi, w. (2008). statistics for engineers and scientists. 3rd ed. mcgraw-hill companies inc., new york, usa, 933 pp. pirti, a. (2008). accuracy analysis of gps positioning near the forest environment. croatian journal of forest engineering, 29(2), 189–199. potter, k., hagen, h., kerren, a., & dannenmann, p. (2006). methods for presenting statistical information: the box plot. visualization of large and unstructured data sets, 4, 97–106. raghunath, s., malleswari, d. b. l., & sridhar, k. (2011). analysis of gps errors during different times in a day. international journal of research in computer science, 2(1), 45–48. https://doi.org/10.7815/ ijorcs.21.2011.014 saracoglu, a., & sanli, d. u. (2020). effect of meteorological seasons on the accuracy of gps positioning. measurement, 152, 107301. https://doi. org/10.1016/j.measurement.2019.107301 seltman, h. j. (2018). experimental design and analysis. carnegie mellon univesrity, 428 pp. http://www.stat.cmu.edu/~hseltman/309/book/ book.pdf sharma, s. (2019). descriptive statistics and factorial design. ph.d. thesis, horizons university, paris, france. https://www.researchgate.net/ publication/333220406_descriptive_statistics#fulltextfilecontent sisman, a. (2014a). an experimental design approach on georeferencing. boletim de ciências geodésicas, 20(3), 548–561. https://doi. org/10.1590/s1982-21702014000300031 sisman, y. (2014b). a full-factorial design approach for coordinate transformation. arabian journal for science and engineering, 39, 227–235. https://doi.org/10.1007/s13369-013-0864-y soler, t., michalak, p., weston, n. d., snay, r. a., & foote, r. h. (2006). accuracy of opus solutions for 1to 4-h observing sessions. gps solutions, 10(1), 45–55. https://doi.org/10.1007/s10291-005-0007-3 stone, j. m., & powell, j. d. (1998). precise positioning with gps near obstructions by augmentation with pseudolites. ieee 1998 position location and navigation symposium, 562–569. https://doi.org/10.1109/ plans.1998.670213 svabensky, o., & weigel, j. (2004). optimized technology for gps height determination. fig working week 2004, athens, greece, 22-27 may, 1–8. https://www.fig.net/resources/proceedings/fig_proceedings/ athens/papers/ts07/ts07_6_svabensky_weigel.pdf teng, y., & wang, j. (2014). new characteristics of geometric dilution of precision (gdop) for multi-gnss constellations. journal of navigation, 67(6), 1018-1028. https://doi.org/10.1017/s037346331400040x teng, y., & wang, j. (2016). a closed-form formula to calculate geometric dilution of precision (gdop) for multi-gnss constellations. gps solutions, 20, 331-339. https://doi.org/10.1007/s10291-015-0440-x teng, y., wang, j., & huang, q. (2015). minimum of geometric dilution of precision (gdop) for five satellites with dual-gnss constellations. advances in space research, 56(2), 229-236. https://doi.org/10.1016/j.asr.2015.04.010 verma, p., hajra, k., banerjee, p., & bose, a. (2019). evaluating pdop in multi-gnss environment. iete journal of research, 1-8. https://doi. org/10.1080/03772063.2019.1666750 wang, y., huang, s., xiang, w. & pei, y. (2011). multipattern road traffic crashes and injuries: a case study of xi’an city. acta polytechnica hungarica, 8(4), 171–181. wang, y., zhao, x., pang, c., feng, b., tong, h., & zhang, l. (2019). bds and gps stand-alone and integrated attitude dilution of precision definition and comparison. advances in space research, 63(9), 2972-2981. https://doi.org/10.1016/j.asr.2017.11.032 wielgosz, p., hadaś, t., kłos, a., & paziewski, j. (2019). research on gnss positioning and applications in poland in 2015–2018. geodesy and cartography, 68(1), 87–119. https://doi.org/https://doi.org/10.24425/ gac.2019.126089 wing, m. g., eklund, a., john, a., & richard, k. (2008). horizontal measurement performance of five mapping-grade global positioning system receiver configurations in several forested settings. western journal of applied forestry, 23(3), 166–171. https://doi.org/10.1093/wjaf/23.3.166 wu, j. c. f., & hamada, m. s. (2009). experiments: planning, analysis, and optimization. 2nd. ed. a john wiley & sons. inc., hoboken, new jersey. 760 pp. yoshimura, t., & hasegawa, h. (2003). comparing the precision and accuracy of gps positioning in forested areas. journal of forest research, 8(3), 147–152. https://doi.org/10.1007/s10310-002-0020-0 zheng, f., lou, y., gu, s., gong, x., & shi, c. (2018). modeling tropospheric wet delays with national gnss reference network in china for beidou precise point positioning. journal of geodesy, 92, 545-560. https://doi. org/10.1007/s00190-017-1080-4 https://doi.org/10.3109/07388551.2014.973014 https://doi.org/10.3109/07388551.2014.973014 https://doi.org/10.1016/j.measurement.2019.04.067 https://doi.org/10.1016/j.measurement.2019.04.067 https://doi.org/10.1007/s10450-007-9042-4 https://doi.org/10.1007/s10450-007-9042-4 https://doi.org/10.5772/intechopen.75493 https://ssrn.com/abstract=1487464 https://doi.org/10.1016/j.asr.2018.08.029 https://doi.org/10.1016/j.asr.2018.08.029 https://doi.org/10.1002/0471458503 https://www.minitab.com/en-us/ https://doi.org/10.1002/qre.449 https://doi.org/10.7815/ijorcs.21.2011.014 https://doi.org/10.7815/ijorcs.21.2011.014 https://doi.org/10.1016/j.measurement.2019.107301 https://doi.org/10.1016/j.measurement.2019.107301 http://www.stat.cmu.edu/~hseltman/309/book/book.pdf http://www.stat.cmu.edu/~hseltman/309/book/book.pdf https://www.researchgate.net/publication/333220406_descriptive_statistics#fulltextfilecontent https://www.researchgate.net/publication/333220406_descriptive_statistics#fulltextfilecontent https://doi.org/10.1590/s1982-21702014000300031 https://doi.org/10.1590/s1982-21702014000300031 https://doi.org/10.1007/s13369-013-0864-y https://doi.org/10.1007/s10291-005-0007-3 https://doi.org/10.1109/plans.1998.670213 https://doi.org/10.1109/plans.1998.670213 https://www.fig.net/resources/proceedings/fig_proceedings/athens/papers/ts07/ts07_6_svabensky_weigel.pdf https://www.fig.net/resources/proceedings/fig_proceedings/athens/papers/ts07/ts07_6_svabensky_weigel.pdf https://doi.org/10.1017/s037346331400040x https://doi.org/10.1007/s10291-015-0440-x https://doi.org/10.1016/j.asr.2015.04.010 https://doi.org/10.1080/03772063.2019.1666750 https://doi.org/10.1080/03772063.2019.1666750 https://doi.org/10.1016/j.asr.2017.11.032 https://doi.org/https://doi.org/10.24425/gac.2019.126089 https://doi.org/https://doi.org/10.24425/gac.2019.126089 https://doi.org/10.1093/wjaf/23.3.166 https://doi.org/10.1007/s10310-002-0020-0 https://doi.org/10.1007/s00190-017-1080-4 https://doi.org/10.1007/s00190-017-1080-4 keywords: waste; management; construction and demolition waste; life cycle assessment; impact category. palabras clave: residuos; gestión; residuos de construcción y demolición; evaluación del ciclo de vida; categoría de impacto how to cite item suarez silgado, s. s., calderon valdiviezo, l. j., & mahecha vanegas, l. f. (2021). application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study. earth sciences research journal, 25(3), 341351. https://doi.org/10.15446/esrj.v25n3.82815 the construction industry not only consumes more raw materials and energy than any other economic activity, but also generates the largest fraction of waste, known as construction and demolition waste (cdw). this waste has major environmental implications, most notably in south american countries such as colombia, where it is handled inappropriately. in this study, the management processes that are currently used for fractions of construction and demolition waste (cdw) generated in ibagué (colombia) were evaluated and the environmental impacts of the management of 1 kg of cdw were calculated. other cdw management alternatives were evaluated, in which the percentage of the fraction of the waste and/or the treatment or management process that is used was modified to determine its environmental and economic viability. the information was obtained through telephone interviews and visits to recycling plants, construction companies, quarries, government entities, and inert landfills in the country. it was completed with secondary sources and the ecoinvent v.2.2 database. life cycle assessment (lca) methodology and simapro 8 software were used to calculate the environmental impacts. an economic study of each management process and each alternative was also carried out. a comparison of the alternatives revealed the current alternative contributes most to the environmental impacts in all categories. the results of this study indicate that the most beneficial alternative in environmental and economic terms in ibagué (colombia) is that in which 100% of the metals are recovered, 100% of excavated earth is reused, and 100% of the stone waste is recycled (alternative 3). when a sensitivity analysis was carried out with different distances (30 km and 50 km), alternative 3 continued to be the most favorable. abstract application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study aplicación de la metodología de análisis de ciclo de vida (acv) y evaluación económica para los residuos de construcción y demolición: un caso de estudio colombiano issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n3.82815 la industria de la construcción no solo consume más materias primas y energía que cualquier otra actividad económica sino que también genera la mayor fracción de residuos, conocidos como residuos de construcción y demolición (rcd). estos residuos presentan importantes implicaciones ambientales, especialmente en países de américa del sur como colombia, donde se manejan de manera inapropiada. en este estudio se evaluó el sistema de gestión actual de los residuos de construcción y demolición (rcd) en ibagué (colombia) y se calcularon los impactos ambientales de la gestión de 1 kg de rcd. se evaluaron otras alternativas de gestión de rcd para determinar su viabilidad ambiental y económica, en las cuales se modificó el porcentaje de la fracción de los residuos y el proceso de tratamiento o de gestión. para la consecución de la información se realizaron entrevistas telefónicas y visitas a plantas de reciclaje, empresas constructoras, canteras, entidades gubernamentales y vertederos inertes en el país. dicha información se completó con fuentes secundarias y con la base de datos ecoinvent v.2.2. para calcular los impactos ambientales se empleó la metodología de análisis de ciclo de vida (acv) y el software simapro 8. también se realizó un estudio económico de cada proceso de gestión y de cada alternativa. una comparación de las alternativas reveló que la alternativa de gestión actual contribuye en mayor medida a los impactos ambientales en todas las categorías. los resultados de este estudio indican que la alternativa más beneficiosa en términos ambientales y económicos en ibagué es aquella en la que se recupera el 100% de los metales, se reutiliza el 100% de la tierra excavada y se recicla el 100% de los residuos de piedra (alternativa 3). cuando se realizó un análisis de sensibilidad con diferentes distancias (30 km y 50 km), esta alternativa continuó siendo la más favorable. resumen record manuscript received: 12/10/2019 accepted for publication: 16/04/2021 earth sciences research journal earth sci. res. j. vol. 25, no. 3 (september, 2021): 341-351 sindy sofía suárez silgado1*, lucrecia janneth calderon valdiviezo2, leandro fernando mahecha vanegas1 1universidad antonio nariño, ibagué, colombia. 2escuela técnica de arquitectura, barcelona, españa. *corresponding author: sindysofia@uan.edu.co g e o t e c h n ic a l e n g in e e r in g https://doi.org/10.15446/esrj.v25n3.79170 https://doi.org/10.15446/esrj.v25n3.79170 mailto:sindysofia@uan.edu.co 342 sindy sofía suárez silgado, lucrecia janneth calderon valdiviezo, leandro fernando mahecha vanegas 1. introduction construction is one of the essential industrial activities for the development and progress of cities. however, it is also one of the sectors that contributes most to environmental impacts, due to the extraction of raw materials, energy use, and waste generation. it is considered that the construction sector consumes more raw material and energy than any other economic activity, and generates the largest fraction of waste. in europe alone, around 900 million t of construction and demolition waste are produced every year (bravo et al., 2015). according to ramesh et al. (2010), the term construction and demolition waste (cdw) refers to solid waste produced in the construction sector. more specifically, the term is defined as the waste that arises from construction, renovation and demolition activities. cdw includes a range of materials such as ceramic products, concrete waste and asphalt material, and to a lesser extent other components such as wood, glass and plastics (yuan and shen, 2011). the main components of this waste depend on the materials used, the construction practices, and the technological development of the sector. there are various management options for this waste, whose hierarchy depends on the environmental impacts they generate. the five levels of low to high environmental impact are: reduction, reuse, recycling, incineration and final disposal. some authors (yuan and shen, 2011) have grouped these six levels into four: waste reduction, reuse, recycling and disposal. however, the impacts of management systems vary. for example, inadequate disposal of construction and demolition waste (cdw) can generate negative environmental impacts such as soil degradation and erosion, destruction of vegetation, and loss of environmental services (mejía et al., 2015). the impacts of waste disposal in landfill are associated with the extraction or obtaining of raw material, and land occupation. if waste is sent to landfill, then new materials or products must be manufactured from original raw material (suárez silgado, 2017). faced with this situation, european countries have tried to find innovative alternatives for the recovery of waste from construction and demolition. yilmaz et al, (2018) have used construction and demolition waste as cemented paste backfill material for underground mine openings. also, numerous eu regulations have been drawn up on this topic. one is the european waste directive, which foresees that by 2020, 70% of cdw should be properly valued. the objective is to achieve much higher levels of recycling by minimizing the extraction of additional natural resources. thus, prevention and recycling are key elements of the new waste policy in europe (suárez-silgado, 2016). in this same line, several studies have used life cycle assessment for effective municipal waste management, because it helps in environmental evaluations of alternative waste management systems (koci and trecakova, 2011). according to iso 14040 (2006), lca is composed of an inventory of the relevant inputs and outputs of the system, the definition of the goal and the scope, an assessment of the potential environmental impacts associated with these inputs and outputs and, finally, an interpretation of the results of the inventory and impact phases in terms of the study objectives. this methodology has been used by several prominent authors in the field, including zabalza et al., 2011; monahan and powell, 2011 and tošić et al., 2015. mercante et al. (2012), coelho and de brito (2012), yeheyis et al. (2013), carpenter et al. (2013), dahlbo et al. (2015), guignot et al. (2015) and wang et al. (2018) have used the lca for the environmental assessment of cdw management systems in europe, north america and asia. the situation has led to the development and implementation of technologies in the international area. the cdw in china are usually randomly dumped or disposed in landfills and the average recycling rate of cdw is only about 5% (huang et al., 2018). in countries such as denmark, the netherlands, and belgium, the recycling of cdw for uses other than landfill is promoted, and recycling percentages of over 75% are achieved. this high level of recycling is mainly due to the shortage of natural aggregates and space for landfill sites (zabalza bribián et al., 2011). one factor that has increased recycling rates has been the increase in the cost of landfilling, or its prohibition in some cases, such as in denmark or the netherlands. however, in most south american countries this is not the situation. the management of cdw waste is carried out inadequately, and there is clearly a large gap between south american and other countries in terms of management and technology. in colombia, cdw is sometimes managed using a controlled discharge system, but usually its disposal is uncontrolled. the authorized sites for waste disposal are disseminated widely, and there are few alternatives for recovery, recycling or reuse. for this reason, only 5% of cdw is recycled in colombia (castaño et al., 2013). the generation of cdw has great environmental implications in this country, as it is sometimes disposed of in illegal landfills or thrown on public roads contributing to changes in the landscape and urban areas (aguilar et al, 2010; pinzón, 2014). the situation is particularly palpable in some municipalities in the country, such as ibagué, where around 488.000 t of cdw are generated per year (ibagué limpia, 2017), without counting the waste that is generated and disposed of clandestinely. the problem has increased in recent years, due to population growth, increased construction activity, and remodeling of buildings. however, an attempt has been made to advance in this area, and for this reason regulations and programs have been promulgated at district level that encourage the adequate disposal of cdw. examples are resolution nº1115/2012, which technically regulates the treatment and/or use of cdw in the capital district; and the municipal development plan “bogotá humana” (2012-2016) with the “zero waste-rubble zero” program. in ibagué, there have been new initiatives for cdw management, expressed in agreement no. 19 of 2013. the agreement implements environmental compare as a tool for citizens related to the proper management of solid waste in ibagué, and the integral solid waste management plan of ibagué (pgirs, 2015), which aims to promote the integral management of cdw, broaden the characterization of this waste, and design programs to take advantage of cdw through feasibility studies. in the same field, studies have been carried out at national level on the perspectives and limitations of cdw management (castaño et al., 2013; pinzón, 2014), and on the current waste situation in some municipalities (jiménez, 2013). technical diagnoses have been made of the use of cdw in the capital district (escandón, 2011; chávez et al., 2014), along with studies of quantification and characterization of cdw (sda, 2012; pinzón, 2014). finally, in some cases, pilot proposals for recycling plants have been made (chávez et al., 2014). however, despite these advances at national level, no studies have been undertaken to date on the environmental assessment of managing fractions of this waste. therefore, to help solve the problem of cdw generation in colombia, the objective of this study was to evaluate the current treatment or management of each fraction of cdw. subsequently, other alternatives were evaluated in which either the treatment that is used for fractions of waste or the percentage of waste in the treatments was modified. hence, the objective was to determine which is the most beneficial management alternative at environmental and economic level. the area chosen for the study was the department of tolima, specifically, its capital ibagué. this area was selected for two main reasons: the current problems associated with the generation of cdw because of population growth and an increasing number of buildings, and the new waste management initiatives described in agreement no. 19 of 2013 and the pgirs (2015), mentioned above. the research is novel because the environmental impacts of waste management systems in colombia were evaluated using life cycle assessment (lca), to obtain a more realistic view of the impacts in this country, specifically in ibagué. as life cycle analysis (lca) is a decision tool, the results could help to encourage or give greater impetus to the development of projects and programs that contribute to cdw recycling or reuse. the management scenarios were also evaluated from an economic perspective. in addition to applying lca methodology, a sensitivity analysis was carried out to determine how the final results are affected. the results obtained may lead to the creation of more demanding regulations in the field of waste management in colombia. 2. data and methods 2.1 study area this study was carried out in ibagué, a city located in the center-west of colombia on the central mountain range of the andes, at an altitude of 1.285 m.a.s.l. this city has an extension area of 4.605 hectares, according to the territorial ordinance plan (pot, 2015), and a population of 553.524 inhabitants (dane, 2015). the average annual generation of construction and 343application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study demolition waste is 488.000 t, according to data provided by ibagué limpia (2017). for this study, it is considered that waste is generated in the expansion zone, where the largest amount of waste is produced. 2.2 description of the construction and demolition waste in the study area in accordance with the regulations that apply in this country (resolution 1115/2012), construction and demolition waste are generated during the development of a construction project, and include excavation products, leveling and leftovers from site preparation; products used for foundations and pilings; stone waste (concrete, sand, gravel, pieces of bricks and blocks, ceramics, leftovers of mortar and concrete mix); and non-stone waste (glass, wood, plastics, metals, cardboard and gypsum). construction companies in the city were visited to gather information about the current management of cdw in the city of ibagué. direct contact was also established through telephone calls and email. according to the information obtained from the companies, over 80% of them are engaged in constructing new dwellings. therefore, the largest amount of waste in this sector comes from construction processes. of all the construction activity, 92.8% corresponds to dwellings (icer, 2016). figure 1 shows the composition of cdw in the study area. the highest percentage corresponds to excavated earth. the amount of excavated land that is generated is high compared to other countries. the reason is that in colombia construction companies (in which the earth movement phases are notable) predominate over demolition companies. in another similar study, a high percentage of this waste was found in ibagué (suárez-silgado et al, 2018). figure 1. composition of cdw in ibagué the management system currently applied for cdw in the city of ibagué (according to the information provided directly by construction companies in the city in 2017) consists basically of the generation of waste in construction, and its subsequent transport to the landfill or final disposal site, without any treatment or recovery. in some cases, around 50% of construction companies separate the metal and excavated earth from the rest of the waste. the metals that are separated are transported by managers to a metal classification or preparation plant for subsequent recycling in a smelting plant in another city. excavated earth is separated to be reused in the same construction or on a nearby site. in some construction projects, waste is not separated and is simply taken to landfill. this management system is due partly to the fact that ibagué does not have a recycling plant for cdw, and therefore the waste is mostly taken to the final disposal site. however, in the absence of sufficient control and monitoring in the management of these sites, waste is also transported to non-formal disposal sites in some cases, causing great damage to public roads, riverbeds, and vacant lots. exact data on this form of disposal are not known. 2.3 life cycle assessment (lca) methodology 2.3.1 goal and scope definition considering the current management of cdw in ibagué and the problems associated with waste generation in this city, the objective was to identify the potential environmental impacts associated with the management of construction and demolition waste fractions and compare them with new alternatives. the functional unit of the study was 1 kg of waste. based on the characterization of waste in ibagué (figure 1) and the information obtained from construction and other companies that were consulted directly, the three waste fraction considered were: stones (19.5%) (concrete, brick and other inerts), excavated earth (80%) (essentially sands and clays; the organic material content within the ground was considered negligible) and metals (0.5%) (steel). 2.3.2 alternatives three alternatives were evaluated in the life cycle assessment (table 1). they were chosen according to the quantity and characterization of the waste and the current waste management system in the study area; information that was provided by ibagué limpia (2017). the first alternative (a1) corresponds to the current management of waste in the city of ibagué. in this alternative, all stone waste is taken to landfill, metals are taken to the sorting and compaction plant and 50% of excavated earth is reused on a nearby construction site (5 km), while the other 50% is taken to landfill. in two alternatives (a2 and a3), the waste treatment and percentages were modified, to determine viability. in alternative a2, the stone waste management was modified. in this case, 50% of stone is recycled and the other 50% is taken to landfill. in alternative a3, 100% of all waste is recovered. all the stone is recycled and all the excavated earth is reused. figure 2. alternatives evaluated figure 3 shows the limits of the study systems. the generation and separation of waste fractions were not considered, as waste is separated on site. only the stages of transportation and waste management were included. once the waste has been generated and separated, it is transported for treatment or final disposal. thus, the type of management (recycling/recovery, reuse and disposal) varied depending on the waste fraction. finally, the loads of processes that are avoided were subtracted. figure 3. limits of study systems. 344 sindy sofía suárez silgado, lucrecia janneth calderon valdiviezo, leandro fernando mahecha vanegas 2.3.3 life cycle inventory the aim was to obtain all data directly through company visits and incorporate them into the inventory. data that could not be obtained in this way (above all, emissions data) were complemented with the ecoinvent database. in all cases, data were contextualised for the colombian context. information was obtained from primary sources through visits, telephone calls and emails to construction companies, cdw and metal recycling plants, inert landfill sites and quarries. government organizations such as ibagué limpia and cortolima were contacted. fifty-six construction companies registered in ibagué, the existing legal inert landfill, 4 metal recycling plants, and 6 quarries near the study area were also contacted. as the city of ibagué does not have a cdw recycling plant, data had to be obtained by visiting waste recycling plants that are currently in operation in colombia. data were also collected from other facilities at national level and from 2 landfills to complement and compare with data from landfills in the study area. the data correspond to different years (2015-2017) with a similar production capacity. the information was supplemented with secondary sources such as articles, journals, projects on the same subject, and the ecoinvent v3 database. to evaluate the environmental impacts of each of the management processes applied to the cdw fractions in ibagué, new processes were created in ecoinvent v3, taking as a starting point the information collected from companies, recycling plants and landfill, completed with the inventory of the ecoinvent v3 database. the electricity mix of colombia was used for the calculations. each of the treatment or management processes was evaluated per kg of waste to obtain the environmental impacts. to determine the environmental impacts of managing 1 kg of cdw generated in ibagué, a new process was created in which each treatment was incorporated, considering the percentage that it occupies in the total construction and demolition waste (figure 2). in this way, each of the alternatives was compared. next, each type of treatment or management was defined: earth reuse, metal recovery (classification and compaction), stone recycling, stone and earth landfill. • excavated earth reuse the process of reusing excavated earth was evaluated in another nearby construction site. it was assumed that the excavated earth was not contaminated and could therefore be reused without any other type of treatment. the external transportation to the other site is considered. according to the information obtained from the companies, 5 km was considered the maximum distance between the extraction site and the reuse site. the compaction of earth on the new construction site and the emissions (leached) due to its use (100 years) were also considered. solid waste was considered as output, which is transported to a sanitary landfill located 11 km from the site on which the waste is generated. the reuse of excavated earth on a nearby construction site avoids disposal of this material in landfill. therefore, the costs of transporting excavated earth to the landfill and its disposal were subtracted in this case. table 1 shows the input and output of the reuse of excavated earth and the process that is avoided. table 1. input and output of the reuse of excavation lands excavated earth reuse cant unit source input transport 5 km companies consulted, 2017 diesel (compaction) 0.001 mj companies consulted, 2017 output emissions to air table 6 ecoinvent v3 emissions to water (leached) table 5 lópez and lobo (2014) waste solid (transport-11 km) 1.00e02 kg companies consulted, 2017 avoided products cant unit source earth landfilling/ transport 1 kg ecoinvent v3; companies consulted, 2017 • metal (steel) recovery for the recovery of metal, the stage considered was transport to the treatment plant, where the waste is prepared (classified and compacted) before it is cast, since there is currently no foundry for this material in ibagué. it was assumed that waste comprised of steel is separated on the construction site and transported separately to the treatment site. the process in these plants basically consists of classifying the material and compacting it, for subsequent export and smelting. according to the companies consulted the steel waste is transported from the expansion area of ibagué to the classification and compaction plants located at an average distance of 9 km. the production capacity of the metal treatment plant was taken to be approximately 3500 t/year of waste, with an infrastructure of 1500 m2. as the recovery of this waste and its preparation for use as secondary raw material (scrap) replaces the production of iron ore, this production is classified as an avoided product, so the impacts due to this process can be subtracted. table 2 shows the inputs and outputs of the recovery of metal and the process that is avoided. table 2. inputs and outputs of the recovery of the metal metal (steel) recovery cant unit source input transport 9 km companies consulted, 2017 diesel 0.03 mj companies consulted, 2017 electricity 5.00e-03 kwh companies consulted, 2017 output emissions to air table 6 ecoinvent v3 waste solid (transport-26 km) 1.00e-02 kg companies consulted, 2017 avoided products cant unit source extraction of iron ore 1 kg ecoinvent v3 • stone recycling taking into account the information collected and based on the existing unit processes in the ecoinvent v.3 database, a process called stone recycling was identified. from the amount of cdw generated in the city of ibagué (2017), it was determined that stone waste made up of 45% brick and 55% concrete. for stone waste recycling, crushing of the aggregate was considered in a fixed recycling plant. as there is currently no recycling plant in the city of ibagué, a possible location was chosen, considering the map of the area and the territorial ordinance plan (pot, 2015). a viable site would be suburban land in this city, which would allow the location and development of an industry of this type, since it is inside the industrial zone. the site chosen for a possible location of the recycling plant is 7.5 km from where the waste is generated (ibagué expansion zone). this distance is well below the limit proposed in ulubeyli et al. (2017), which states that the distance between facilities should be no more than 50 km for them to be viable environmentally and economically. based on these data, the t-kilometers to be entered in the ecoinvent v3 database were calculated. it was found that 0.008 tkm is used to transport 1.02 kg of stone waste. it was assumed that ceramics and concrete waste had been separated from the rest of cdw in work. the output of the recycling process was considered to be air and water emissions, the product of the crushing process, and the steel recovered from the reinforced concrete that is included in concrete and brick stone waste. to calculate the amount of steel generated by mixed aggregate, it was assumed that 1 kg of concrete generates 0.02 kg of steel (nsr, 2010). as the mixed aggregate was composed of 55% concrete, according to data from ibagué limpia (2017), then 0.55 kg of concrete generated 0.01 kg of steel. this value was taken as the maximum amount of steel generated from the mixed aggregate. it was assumed that the generated steel waste could be transported to the treatment plants that are currently situated in ibagué, which classify and 345application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study compact the metal. therefore, in this study, the steel management system included only transport to these plants. in this case, an average distance of 16 km was selected, according to the maps (pot, 2015). the transport of aggregate extracted from the quarry was considered a process that was avoided. it was assumed that the recycling plant had a maximum production capacity of 100000 t/year. table 3 shows the inputs and outputs of stone recycling and the process that is avoided. table 3. inputs and outputs of stone recycling stone recycling cant unit source input transport 8 km companies consulted, 2017 diesel 0.05 mj companies consulted, 2017 electricity 8.00e-05 kwh companies consulted, 2017 water 6.00e-03 kg ecoinvent v3 output particulates <2.5 um 1.66e-05 kg ecoinvent v3 particulates >10 um 8.35e-05 kg ecoinvent v3 particulates >2.5 um and <10 um 6.34e-05 kg ecoinvent v3 water 1.91e-04 m3 ecoinvent v3 waste reinforcement steel (transport16 km) 1.00e-02 kg nsr (2010); ibagué limpia (2017) avoided products cant unit source extraction of aggregates/transport 1 kg ecoinvent; companies consulted, 2017 • excavated earth and stone landfilling the stages of landfill disposal are similar for stone and earth. this process considers landfill infrastructure, land occupation, fuel consumption for compaction and grading of the material, and the effect of landfilled waste (leachate). in this process, it was considered that waste from construction works is transported from the expansion zone to the inert waste landfill located at 13 km. factors that were considered included the landfill infrastructure (14 hectares), the use of land, and fuel (diesel) consumed in the compaction and grading of the material. table 4. inputs and outputs of earth and stone landfilling excavated earth and stone landfilling cant unit source input occupation, construction site 4.44e-05 m2a ecoinvent v3 occupation, dump site 0.000444 m2a ecoinvent v3 transformation, from dump site 4.44e-05 m2 ecoinvent v3 transformation, to dump site 4.44e-05 m2 ecoinvent v3 diesel 1.00e-03 mj companies consulted, 2017 electricity 2.70e-06 kwh companies consulted, 2017 transport 13 km companies consulted, 2017 output emissions to air table 6 ecoinvent v3 emissions to water (leached) table 5 lópez and lobo (2014) a life cycle inventory (lci) for inert landfill rarely includes leachate, since in general the waste material in this kind of landfill has a low pollutant content, and is chemically inert to a large extent. however, according to bovea et al. (2016), models for inert material landfills should take these emissions into account, since a small percentage of biodegradable materials may be disposed of at these sites. considering studies by butera et al. (2015) and colomer et al. (2017), the leachate emissions framework was considered to be 100 years. the leachate composition from cdw landfill according to lópez and lobo (2014), bovea et al. (2016), and ecoinvent v3 was taken into account. table 5 shows the composition of the leachates in the inert waste landfill. the landfill gas collection system was excluded, since this was not relevant in a landfill for cdw (butera, 2015; bovea et al., 2016). the table 4 shows the inputs and outputs of this process. table 5. primary data for leachate composition from inert landfill (lópez and lobo, 2014) composition quantity cod (mg/l) 1571 bod5 (mg/l) 227 nh4-n (mg/l) 401 sulfates (mg/l) 405 ca (mg/l) 150 na (mg/l) 495 cr (µg/l) 105 cd (µg/l) 27 cu (µg/l) 28 zn (µg/l) 276 pb (µg/l) 987 ni (µg/l) 59 as(µg/l) 233 hg (µg/l) 1.4 table 6. emissions to air (ecoinvent v3) emmissions to air cant unit acetaldehyde 1.50e-07 kg acrolein 5.83e-08 kg ammonia 5.18e-07 kg arsenic 3.66e-12 kg benzaldehyde 4.51e-08 kg benzene 2.30e-09 kg butane 4.94e-09 kg cadmium 3.18e-10 kg carbon dioxide 0.1147 kg carbon monoxide 0.00011 kg chromium 1.10e-09 kg chromium vi 2.20e-12 kg copper 7.76e-10 kg dinitrogen monoxide 6.14e-06 kg ethane 9.88e-10 kg formaldehyde 2.76e-07 kg heptane 9.88e-09 kg lead 1.90e-09 kg mercury 1.94e-10 kg methane 8.10e-08 kg m-xylene 3.23e-08 kg nickel 3.22e-10 kg (continued) 346 sindy sofía suárez silgado, lucrecia janneth calderon valdiviezo, leandro fernando mahecha vanegas emmissions to air cant unit nitrogen oxides 5.22e-05 kg nmvoc 2.67e-06 kg o-xylene 1.31e-08 kg pah 2.86e-09 kg particulates <2.5 4.94e-07 kg pentane 1.97e-09 kg propane 3.30e-09 kg selenium 3.66e-12 kg styrene 1.84e-08 kg sulfur dioxide 5.66e-07 kg toluene 3.29e-10 kg zinc 6.36e-08 kg 2.3.4 life cycle assessment for the study of the lca, the impact categories were chosen based on several studies in this field. these studies are shown in table 7. the impact assessment method selected for the study is impact 2002+, which has been used by several authors (hossain, 2016; hossain, 2017; and suárez et al., 2016). this method is one of the most feasible and widely used environmental impact assessment methodologies. it combines a midpoint and damage approach, and links all types of life cycle inventory results via several midpoint categories to several damage categories (hossain, 2017). the software used to perform the calculations was simapro 8. simapro is one of the most widely used and accepted lca tools. it helps to model various products and processes, and analyses the results to achieve sustainability goals. simapro includes many lci datasets, including the renowned ecoinvent v3 database, the new industry-specific agri-footprint database, and the elcd database (pré consultants, 2017). table 7. impact categories evaluated references c ri old po lo a e gw nre me bueno et. al. (2015) x x x x x x hoxha et. al. (2017) x x liu et. al. (2017) x x x x x x x ripa et al. (2017) x x x x x x x heinonen et al. (2016) x x x x x x x x x hossain (2016) x x x x x x x x x x bovea et al. (2016) x x x x x x x x butera et al. (2015) x x x x x x x x hossain (2017) x x jullien (2014) x suárez et al. (2016) x x x x x x x x x x milutinovic´ (2017) x x x x x x yay (2015) x x x x x x x x c= carcinogenic; ri= inorganic respiratory; old= depletion of the ozone layer; po= photochemical oxidation; lo= land occupation; a= acidification; e= eutrophication; gw= global warming; nre= non-renewable energy; me= mineral extraction. 2.3.5 interpretation the interpretation includes the presentation and evaluation of results, and a sensitivity analysis to check the reliability and robustness of the results by varying assumptions and methods. a sensitivity analysis with different distances was performed. 2.4 economic evaluation to determine the economic viability of each management process and, subsequently, the feasibility of each alternative, a study was carried out in which the cost of each treatment per kg of treated waste and the cost of each alternative was calculated, considering the type of treatment and the percentage of each fraction of waste within the total generated cdw. the economic data were obtained from the construction companies, recycling plants and entities visited in ibagué. 3. results and discussion 3.1 results of the management processes the results of the management processes that are currently applied in ibagué for each waste fraction are presented below. these results are given per kg of waste. 3.1.1 stone recycling on a relative scale of 0%–100%, where positive percentages represent impacts and negative percentages represent savings, figure 4 (a) shows that the avoided process of gravel extraction contributed to the greatest savings in all the categories that were evaluated (-100%). the other processes contributed to environmental impacts. diesel consumption had the greatest environmental impacts in all categories, apart from photochemical oxidation. the crushing process had major impacts on the category of respiratory effects (52%) due to the emission of particles. the transport process affected the soil occupation category (28%) as did the infrastructure of the recycling plant (14%). the environmental burdens of the rest of the processes were minimal. 3.1.2 excavated earth reuse figure 4 (b) shows that the avoided process of earth landfilling contributed to great environmental savings in all the categories (-100%). the other processes contributed to environmental impacts, but in a smaller proportion (less than 20%). transportation was the process that contributed most to the impacts on the categories, particularly global warming (17%), mineral extraction (13%) and photochemical oxidation (11%). diesel consumption and leachate emissions during land compaction did not have significant impacts in the process of earth reuse. 3.1.3 excavated earth and stone landfilling the process of earth and/or stone disposal does not avoid any processes, which is why it contributed to positive environmental loads. the process that contributed most to environmental impacts was the infrastructure of the landfill, especially in the categories carcinogenic (81%), acidification (77%), nonrenewable energy (77%), depletion of the ozone layer (77%) and eutrophication (74%). another process that contributed to environmental impacts was transport, particularly in the global warming category (44%). transformation and use of earth for landfill had an impact on the land occupation category (41%) (figure 4 [c]). leachate emissions are not relevant in the process of earth and stone disposal, as they are inert materials. 3.1.4 metal (steel) recovery metal recovery plays an important role as it avoids the process of extracting iron ore. this had environmental benefits in all the impact categories. the greatest savings were found in the inorganic respiratory (-100%), photochemical oxidation (-100%) and mineral extraction (-100%) categories. significant savings were also generated in the following categories: acidification (-80%), reduction of the ozone layer (-37%), non-renewable energy (-29%) and global warming (-27%). the infrastructure of the classification plant contributes to greater impacts in the evaluated categories. the greatest impacts were identified in the categories of eutrophication (98%), land occupation (97%) and carcinogenic effects (91%) (figure 4 [d]). 347application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study 3.2 comparison of the evaluated processes in figure 5, the processes that were evaluated are compared to determine the environmental impacts of each one per kg of waste. figure 5 shows that the metal recovery process contributed to impacts on the carcinogenic effects (100%), ozone depletion (67%), land occupation (100%), acidification (64%), eutrophication (100%), global warming (100%) and non-renewable energy categories (100%). however, there were also savings in the inorganic respiratory (-100%), photochemical oxidation (-63%) and mineral extraction (-100%) categories. these savings were due to avoidance of the processes of extracting and producing raw material. the results agree with those found by mercante et al. (2012). the stone recycling process led to savings in all categories. the biggest savings were in photochemical oxidation and acidification (-100%). this is mainly due to the avoidance of natural aggregate extraction loads. the lowest savings in the inorganic respiratory category (-1%) were due to the emission of particulate material during the stone crushing process. as in mercante et al. (2012), it was found that stone recycling led to savings in all categories. the results agree with those of mercante et al. (2012), coelho and de brito (2012), kucukvar et al. (2014) ,guignot et al. (2015) and wang et al. (2018), who also found that recycling cdw led to environmental savings. the process of earth reuse resulted in environmental savings in all categories, as the high loads in the earth landfilling process are avoided. the biggest savings were in the categories of ozone depletion (-89%) and non-renewable energy (-86%). significant savings were also identified in photochemical oxidation (-46%), acidification (-38%) and soil occupation (-38%) categories. the disposal of earth and/or stones had an impact on all categories. the greatest impacts were found in the category of ozone depletion (100%) and non-renewable energy (97%) as a result of energy consumption during the process. there were also impacts on land occupation due to the transformation and use of the earth. the impacts of leachate were not relevant in this process; therefore, eutrophication effects were not very high (4%). the results coincide with butera (2015), who found impacts due to waste disposal and no savings from this process. figure 5. comparison of the processes evaluated for each type of waste per kg 3.3 analysis of the different alternatives finally, the environmental impacts of the current management system were evaluated, considering the waste fraction and the management process applied to each type of waste that make up 1 kg of the cdw generated in ibagué (colombia). this management alternative (a1) was compared with two other alternatives in which the percentage of the waste fraction and/or the management process was modified (a2 and a3). the results in figure 6 show that the current management alternative (a1) for 1 kg of cdw generated in ibagué had an impact on all categories, except mineral extraction, in which environmental savings were found (-21%) due to the metal recovery process. the greatest impacts were in the eutrophication category (47%). this may be due to the fact that in this alternative, 100% of the stone and 50% of the excavated earth were dumped, which has a greater impact on this category due to the infrastructure of the landfill and transport of waste to the landfill. other impacts associated with this alternative were in the categories of global warming (34%), non-renewable energy (30%), land occupation (30%) and ozone depletion (29%). these impacts are due to fuel consumption and the use and transformation of the earth in the waste disposal process. in addition, borghi et al. (2018) evaluated the environmental performance of the regional management of construction and demolition waste in italy. the results showed that the induced environmental impacts are generally greater than the benefits arising from recycling activities. figure 6. comparison of the scenarios per kg of cdw alternative 2 had lower environmental impacts than a1 in all categories. in the categories of photochemical oxidation and acidification, it led to environmental savings, unlike a1, because it includes the process of recycling 50% of the stone waste, thus reducing the disposal of this material. a) b) c) d) figure 4. environmental impacts. a) stone recycling; b) excavated earth reuse; c) excavated earth and stone landfilling; d) metal recovery. 348 sindy sofía suárez silgado, lucrecia janneth calderon valdiviezo, leandro fernando mahecha vanegas alternative 3 led to environmental savings in all the categories (-100%). this is due to the fact that disposal of waste is eliminated and the percentage of stone waste recycling and reuse of earth increased. in other words, in a3 all generated cdw is recovered. since the impacts that are avoided when waste materials are recovered and used instead of virgin material are much greater than the impacts that are generated, the net result is that recovery or recycling contributes to savings. these results are consistent with those obtained by coelho and de brito (2012) who found better results for management alternatives in which 96% recycling was carried out, resulting in lower environmental impacts in the categories of climate change, acidification, summer smog, nitrification and heavy metals than the alternatives in which recycling was null or scarce. in addition, rosado et al. (2019) evaluated the environmental performance of construction and demolition waste management in brazil. the results highlighted the importance of the impacts that are avoided by recovering materials, particularly those related to steel, glass and plastics recycling. table 8 shows the environmental impacts of the alternatives evaluated per kg of cdw. these results coincide with those found by coelho and de brito (2013a) who in their study concluded that the impacts prevented through material recycling can be up to one order of magnitude greater than those generated. according to the world bank (2013) ibagué is classified as the second city with the most growth and strongest development perspective in colombia. this is why it is believed that this region could have good demand for this type of materials for infrastructure construction. table 8. environmental impacts of the different alternatives evaluated per kg of cdw impact category unit a1 a2 a3 c kgc2h3cl eq 1.83 e-5 4.82e-6 -6.36e-5 ri kg pm2.5 eq 8.57e-7 1.05e-7 -5.17e-6 old kg cfc-11 eq 4.15 e-10 1.53e-10 -1.41e-9 po kg c2h4 eq 8.79e-7 -1.11e-7 -3.73e-6 lo m2org.arable 0.000209 0.00011 -0.000709 a kg so2 eq 6.67e-6 -1.97e-6 -3.08e-5 e kg po4 p-lim 3.47e-7 1.75e-7 -7.33e-7 gw kg co2 eq 0.00136 0.00037 -0.00402 nre mj primary 0.0381 0.0146 -0.128 me mj surplus -3.65e-5 -7.19e-5 -0.000177 3.4 sensitivity analysis it was considered necessary to carry out a sensitivity analysis to determine how the final results varied. it was assessed how the results changed when the distance to the recycling plant was altered. previous studies were taken into account to select the distances. rodríguez et al. (2015) indicated that the maximum viable distance between the sources of residues and recycling facilities is 30 km. ulubeyli et al. (2017) proposed that the maximum limit for viability is 50 km between facilities. therefore, a sensitivity analysis was required, for which scenario 3 was taken as a basis and the environmental impacts were evaluated according to the distances of 30 km (a4) and 50 km (a5). figure 7 reveals that a3 continued to be the most favorable scenario, due to the greater savings in the evaluated categories. also when coelho and de brito (2013b) performed a sensitivity analysis, they found that in spite of the variations in overall primary energy and co2eq balances, the prevented impacts are always higher than the generated impacts. when the transport distance to the recycling plant was increased, savings were still made in the alternatives. the savings decreased as the distance to the recycling plant increased. these results agree with those found by butera (2015), who considered that the process of transporting cdw to the recycling plant contributes to environmental impacts. butera (2015) concluded that transportation contributed 30 per cent of carcinogenic human toxicity impacts, but only played a minor role with respect to non-carcinogenic human toxicity. blengini and garbarino (2010) reached a similar conclusion. in their case, it was concluded that the transportation distance of recycled aggregate should increase two to three times before the induced impacts outweighs the avoided impacts. figure 7. sensitivity analysis for different distances 3.5 economic evaluation the economic factor was also relevant when different alternatives were assessed. in this study, the economic evaluation was undertaken on the basis of the cost of each management process per ton, and considering the cost avoided by the management process (table 9). the net cost of each alternative (table 10) was obtained from: the values of the net cost of table 9 and the percentages of each process within the alternative (figure 2). table 9 shows that a3 was the most beneficial scenario, from the economic perspective, and that increasing the distance to which the recycling plant would be located (50 km, a5) would continue to lead to economic savings, compared to the current management alternative (a1). 3.6 study limitations the limitations of the study mainly refer to the fact that not all the inventory analysis data could be obtained through visits to companies. consequently, they were completed with the ecoinvent v.2.2 database, which was contextualized for the colombian context. table 9. economic cost of each process per kg waste management process process cost ($/kg) avoided cost ($/kg) net cost ($/kg) source stone recycling cra1+tc2=16+27=43 cna4+tc2=23+136=159 -116 1 recycling plants consulted 2017; 2 ibagué limpia 2017; 3 metal recovery plants 2017; 4 quarries consulted, 2017; 5 companies consulted, 2017 landfilling ct2+clr2=44+5=49 ----- 49 earth reuse ex situ tc2 + c= 17+20=37 tc2+cg2=36+49=85 -48 landfilling ct2+clr2=44+5=49 ----- 49 metals recovery tc2+ccc3=30+22=52 ciron=167 -115 cra= cost of recycled aggregate; tc= transport cost; clr= cost of landfill rate; lc= labour cost (separation); ccc= classification and compaction cost; cna= cost of natural aggregate; cg=cost of earth; sc= scrap cost 349application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study table 10. net economic cost of each alternative evaluated process a1 a2 a3 a4 a5 stone recycling -----11.3 -22.6 -7.99 5.07 metal recovery -0.58 -0.58 -0.58 -0.58 -0.58 earth reuse -19.2 -19.2 -38.4 -38.4 -38.4 stone landfilling 9.55 4.77 ------------earth landfilling 19.6 19.6 ------------net cost by scenarios ($) 9.38 -6.70 -61.6 -46.97 -33.91 4. conclusion in this study, management processes for construction and demolition waste fractions in ibagué (colombia) were evaluated. the environmental impacts of the management of 1 kg of generated cdw were calculated, considering the current management of each fraction of waste and the percentage of the fraction within the total cdw that is generated (a1). new management alternatives were evaluated (a2 and a3) in which the percentage of the waste fraction or the type of treatment applied to each one was varied, to determine viability. the environmental evaluation was performed using lca methodology. an economic evaluation of alternatives was also carried out to determine viability. the comparison of processes evaluated per kg of waste fraction revealed that metal recovery contributed to impacts in some categories and savings in others. likewise, the process of earth reuse resulted in environmental savings in all categories, due to the high loads of the process that was avoided (disposal of earth in landfill). the last process to be evaluated corresponded to earth and stone landfilling, which had an impact on all categories. the comparison of management alternatives evaluated per kg of cdw revealed that a1, which corresponds to the current management of cdw in ibagué, contributed to environmental impacts in all categories, except mineral extraction. a2 led to savings in photochemical oxidation, acidification and mineral extraction. a3 brought about great savings in all categories. thus, the results of this study allow us to verify that the most beneficial cdw management scenario from the environmental and economic point of view in ibagué is that in which 100% of the metals are recovered, 100% of the earth is reused, and 100% of the stone waste is recycled. the results could help to raise awareness among all the agents involved in the construction sector, who can clearly see the vital role they play in the proper management of cdw and, as a result, promote the implementation of mechanisms and infrastructure for waste recovery and recycling. likewise, the results could contribute to enhancing environmental awareness in the educational and professional field, leading to more sustainable production and consumption habits in colombia. finally, it is hoped to help solve current problems related to the inadequate management of cdw, and contribute to progress in this area, as has occurred at international level. 5. acknowledgments the authors thank antonio nariño university for the financial support. we also appreciate all the support provided by construction companies and organizations in ibagué, as well as recycling plants, governmental entities and inert landfills in colombia that provided information for the project. references aguilar, m., & ardila, f. (2010). study of waste management alternatives generated by the construction and demolition of buildings in the city of bogotá (estudio de alternativas de manejo de escombros generados por la construcción y demolición de edificaciones en la ciudad de bogotá). industrial university of santander. specialization monograph. bucaramanga, colombia. blengini, g. a., & garbarino, e. (2010). resources and waste management in turin (italy): the role of recycled aggregates in the sustainable supply mix. journal of cleaner production, 18(10-11), 1021-1030. doi:10.1016/j.jclepro.2010.01.027 bovea, m. d. & powell, j. c. (2016). developments in life cycle assessment applied to evaluate the environmental performance of construction and demolition wastes. waste management, 50, 151-172. https://doi. org/10.1016/j.wasman.2016.01.036 borghi, g., pantini, s., & rigamonti, l. (2018). life cycle assessment of nonhazardous construction and demolition waste (cdw) management in lombardy region (italy). journal of cleaner production, 184, 815825. https://doi.org/10.1016/j.jclepro.2018.02.287 bravo, m., de brito, j., pontes, j., & evangelista, l. (2015). durability performance of concrete with recycled aggregates from construction and demolition waste plants. construction and building materials, 77, 357–369. https://doi.org/10.1016/j.conbuildmat.2014.12.103 bueno, g., latasa, i. & lozano, p. j. (2015). comparative lca of two approaches with different emphasis on energy or material recovery for a municipal solid waste management system in gipuzkoa. renewable and sustainable energy reviews, 51, 449–459. https://doi. org/10.1016/j.rser.2015.06.021 butera, s., christensen, t. h., & astrup, t. f. (2015). life cycle assessment of construction and demolition waste management. waste management, 44, 196-205. https://doi.org/10.1016/j.wasman.2015.07.011 carpenter, a., jambeck, j. r., gardner, k. & weitz, k. (2013), life cycle assessment of end-of-life management options for construction and demolition debris. journal of industrial ecology, 17(3), 396-406. doi: 10.1111/j.1530-9290.2012.00568. castaño j. o., rodriguez, r. m., lasso, l. a., gomez cabrera, a., & ocampo, m. s. (2013). management of construction and demolition waste in bogotá: perspectives and limitations. tecnura, 17(38), 121-129. http:// www.scielo.org.co/pdf/tecn/v17n38/v17n38a10.pdf chávez porras, a., palacio león, o., & guarin cortes, n. l. (2014). logistic unit for the recovery of construction and demolition waste: case study bogotá c.d (unidad logística de recuperación de residuos de construcción y demolición: estudio de caso bogotá d.c). ciencia e ingeniería neogranadina, 23, 95-118. http://www.scielo.org.co/pdf/ cein/v23n2/v23n2a06.pdf coelho, a., & de brito, j. (2012). influence of construction and demolition waste management on the environmental impact of buildings. waste management, 32(3), 532–541. doi: http://dx.doi.org/10.1016/j. wasman.2011.11.011. coelho, a., & de brito, j. (2013a). environmental analysis of a construction and demolition waste recycling plant in portugal – part i: energy consumption and co2 emissions. waste management, 33(5), 12581267. doi: https://doi.org/10.1016/j.wasman.2013.01.025 coelho, a., & de brito, j. (2013b). environmental analysis of a construction and demolition waste recycling plant in portugal – part ii: environmental sensitivity analysis. waste management, 33, 147-161. https://doi. org/10.1016/j.wasman.2012.09.004 colomer, f., altabella, j. e. & izquierdo, a. g. (2017). application of inert wastes in the construction, operation and closure of landfills: calculation tool. waste management, 59, pp.276–285. https://doi. org/10.1016/j.wasman.2016.10.041 dahlbo, h., bacher, j., lähtinen, k., jouttijärvi, t., suoheimo, p., mattila, t., (…) saramäki, k. (2015). construction and demolition waste management – a holistic evaluation of environmental performance. journal of cleaner production, 107, 333-341. https://doi.org/10.1016/j. jclepro.2015.02.073 dane (2015). national administrative department of statistics (departamento administrativo nacional de estadística). available at: https:// colaboracion.dnp.gov.co/cdt/inversiones%20y%20finanzas%20 pblicas/tolima%2015.pdf, accessed: 27/07/2017. http://dx.doi.org/10.1016/j.jclepro.2010.01.027 https://doi.org/10.1016/j.wasman.2016.01.036 https://doi.org/10.1016/j.wasman.2016.01.036 https://doi.org/10.1016/j.jclepro.2018.02.287 https://doi.org/10.1016/j.conbuildmat.2014.12.103 https://doi.org/10.1016/j.rser.2015.06.021 https://doi.org/10.1016/j.rser.2015.06.021 https://doi.org/10.1016/j.wasman.2015.07.011 http://www.scielo.org.co/pdf/tecn/v17n38/v17n38a10.pdf http://www.scielo.org.co/pdf/tecn/v17n38/v17n38a10.pdf http://www.scielo.org.co/pdf/cein/v23n2/v23n2a06.pdf http://www.scielo.org.co/pdf/cein/v23n2/v23n2a06.pdf http://dx.doi.org/10.1016/j.wasman.2011.11.011 http://dx.doi.org/10.1016/j.wasman.2011.11.011 https://doi.org/10.1016/j.wasman.2013.01.025 https://doi.org/10.1016/j.wasman.2012.09.004 https://doi.org/10.1016/j.wasman.2012.09.004 https://doi.org/10.1016/j.wasman.2016.10.041 https://doi.org/10.1016/j.wasman.2016.10.041 https://doi.org/10.1016/j.jclepro.2015.02.073 https://doi.org/10.1016/j.jclepro.2015.02.073 https://colaboracion.dnp.gov.co/cdt/inversiones%20y%20finanzas%20pblicas/tolima%2015.pdf https://colaboracion.dnp.gov.co/cdt/inversiones%20y%20finanzas%20pblicas/tolima%2015.pdf https://colaboracion.dnp.gov.co/cdt/inversiones%20y%20finanzas%20pblicas/tolima%2015.pdf 350 sindy sofía suárez silgado, lucrecia janneth calderon valdiviezo, leandro fernando mahecha vanegas escandón, j. c. (2011). technical and economic diagnosis of the use of construction and demolition waste in buildings in the city of bogotá (diagnóstico técnico y económico del aprovechamiento de residuos de construcción y demolición en edificaciones en la ciudad de bogotá). final degree project. pontifical javeriana university. bogotá d.c. https://repository.javeriana.edu.co/bitstream/handle/10554/7516/ tesis603.pdf;jsessionid=f1b8a7a9adcb48759fb2ebf32e84fd33?sequence=1 guignot, s., touzé, s., von der weid, f., ménard, y., & villeneuve, j. (2015). recycling construction and demolition wastes as building materials: a life cycle assessment. journal of industrial ecology, 19(6), 1030–1043. https://doi.org/10.1111/jiec.12262 heinonen, j., säynäjoki, a., junnonen, j. m., pöyry, a., & junnila, s. (2016). pre-use phase lca of a multi-story residential building: can greenhouse gas emissions be used as a more general environmental performance indicator? building and environment, 95, pp.116–125. https://doi.org/10.1016/j.buildenv.2015.09.006 hossain, m. u., poon, c. s., lo, i. m. c., & cheng, j. c. p. (2016). comparative environmental evaluation of aggregate production from recycled waste materials and virgin sources by lca. resources, conservation and recycling, 109, 67–77. https://doi.org/10.1016/j. resconrec.2016.02.009 hossain, m. u., poon, c. s., lo, i. m. c., & cheng, j. c. p. (2017). comparative lca on using waste materials in the cement industry: a hong kong case study. resources, conservation and recycling, 120, 199–208. https://doi.org/10.1016/j.resconrec.2016.12.012 hoxha, e., habert, g., lasvaux, s., chevalier, j., & le roy, r. (2017). influence of construction material uncertainties on residential building lca reliability. journal of cleaner production, 144, pp.33–47. https://doi.org/10.1016/j.jclepro.2016.12.068 huang, b., wang, x., kua, h., geng, y., bleischwitz, r., & ren, j. (2018). construction and demolition waste management in china through the 3r principle. resources, conservation and recycling, 129, 36–44. https://doi.org/10.1016/j.resconrec.2017.09.029 icer. regional economic situation report (2016). tolima, ibagué. dane banco de la república. colombia. issn 1794-3582. jimenez, h. (2013). evaluation of the current situation of the management of construction and demolition waste (cdw) in the municipality of madrid (evaluación de la situación actual del manejo de residuos de construcción y demolición (rcd) en el municipio de madrid (cundinamarca) (colombia)). final career project. pontificia universidad javeriana. science faculty. bogotá (colombia), pp. 41. https://repository.javeriana.edu.co/handle/10554/11829 jullien, a., dauvergne, m. & cerezo, v. (2014). environmental assessment of road construction and maintenance policies using lca. transportation research part d: transport and environment, 29, 56-65. https://doi. org/10.1016/j.trd.2014.03.006 koci, v., & trecakova, t. (2011). mixed municipal waste management in the czech republic from the point of view of the lca method. the international journal of life cycle assessment volume, 16, 113-124. https://doi.org/10.1007/s11367-011-0251-4 kucukvar, m., egilmez, g., & tatari, o. (2014). evaluating environmental impacts of alternative construction waste management approaches using supply-chain-linked life-cycle analysis. waste management & research, 32(6), 500-508. doi: 10.1177/0734242x14536457 liu, g., hao, y., dong, l., yang, z., zhang, y., & ulgiati, s. (2017). an emergy-lca analysis of municipal solid waste management. resources, conservation and recycling, 120, pp.131-143. https://doi. org/10.1016/j.resconrec.2016.12.003 lópez, a. & lobo, a. (2014). emissions of c&d refuse in landfills: a european case. waste management, 34(8), 1446–1454. doi: 10.1016/j. wasman.2014.04.004 mejía, e., osorno bedoya, l., & osorio, vega, n. (2015). construction waste: an option for soil recovery (residuos de la construcción: una opción para la recuperación de suelos). revista eia, 12(2), 1-7. https:// revistas.eia.edu.co/index.php/reveia/article/view/706/661 mercante, i. t., bovea, m. d., ibáñez-forés, v., & arena, a. p. (2012). life cycle assessment of construction and demolition waste management systems: a spanish case study. the international journal of life cycle assessment, 17, 232-241. https://doi.org/10.1007/s11367-011-0350-2 monahan, j. & powell, j. (2011). an embodied carbon and energy analysis of modern methods of construction in housing: a case study using a lifecycle assessment framework. energy and buildings, 43 (1), 179– 188. https://doi.org/10.1016/j.enbuild.2010.09.005 milutinović, b., stefanović, g., dekić, p. s., mijailović. i., & tomić, m. (2017). environmental assessment of waste management scenarios with energy recovery using life cycle assessment and multi-criteria analysis. energy, 137, pp. 917-926. https://doi.org/10.1016/j. energy.2017.02.167 nsr-10. (2010). colombian standards for seismic resistant design and construction. title a general requirements for seismic resistant design and construction. reglamento colombiano de construcción sismo resistente. ministerio de ambiente, vivienda y desarrollo territorial. colombia. rosado, l. p., vitale, p., penteado c. s.g., arena, u. (2019). life cycle assessment of construction and demolition waste management in a large area of são paulo state, brazil. waste management, 15, 477-489. https://doi.org/10.1016/j.wasman.2019.01.011 pinzón, a. (2014). formulation of guidelines for the management of construction and demolition waste (cdw) in bogotá. (formulación de lineamientos para la gestión de residuos de construcción y demolición (rcd) en bogotá). final degree project. universidad militar nueva granada. bogotá, d.c. https://repository.unimilitar.edu. co/handle/10654/11004 pgirs. (2015). integral solid waste management plan of ibagué. alcaldía de ibagué (colombia). pot. (2015). territorial ordinance plan. ibagué (colombia). pré consultants. (2017). introduction to lca with simapro. pré consultants, the netherlands. ramesh, t., prakash, r., & shukla, k. k. (2010). life cycle energy analysis of buildings: an overview. energy and buildings, 42(10), 1592–1600. doi:10.1016/j.enbuild.2010.05.007 resolución 1115/2012. (2012). technical-environmental guidelines for the activities of exploitation and treatment of construction and demolition waste in the capital district (lineamientos técnico-ambientales para las actividades de aprovechamiento y tratamiento de los residuos de construcción y demolición en el distrito capital). district secretary of the environment. bogotá-colombia. restrepo, á., becerra, r., & tibaquirá, j., e. (2016). energetic and carbon footprint analysis in manufacturing process of bamboo boards in colombia. journal of cleaner production, 126, pp.563–571. https:// doi.org/10.1016/j.jclepro.2016.02.144 ripa, m., fiorentino, g., vacca, v., & ulgiati, s. (2017). the relevance of site-specific data in life cycle assessment (lca). the case of the municipal solid waste management in the metropolitan city of naples (italy). journal of cleaner production, 142(1), 445-460. https://doi. org/10.1016/j.jclepro.2016.09.149 rodríguez, g., medina, c., alegre, f. j., asensio, e., & (2015). assessment of construction and demolition waste plant management in spain: in pursuit of sustainability and eco-efficiency. journal of cleaner production, 90, 16-24. https://doi.org/10.1016/j.jclepro.2014.11.067 suárez-silgado, s. (2017). stone materials from construction waste. methodological proposal for environmental and economic evaluation (materiales pétreos a partir de residuos de construcción. propuesta metodológica para la evaluación ambiental y económica). alemania. editorial publicia isbn: 978-3-8416-8281-9, pág. 473. https://repository.javeriana.edu.co/bitstream/handle/10554/7516/tesis603.pdf;jsessionid=f1b8a7a9adcb48759fb2ebf32e84fd33?sequence=1 https://repository.javeriana.edu.co/bitstream/handle/10554/7516/tesis603.pdf;jsessionid=f1b8a7a9adcb48759fb2ebf32e84fd33?sequence=1 https://repository.javeriana.edu.co/bitstream/handle/10554/7516/tesis603.pdf;jsessionid=f1b8a7a9adcb48759fb2ebf32e84fd33?sequence=1 https://doi.org/10.1111/jiec.12262 https://doi.org/10.1016/j.buildenv.2015.09.006 https://doi.org/10.1016/j.resconrec.2016.02.009 https://doi.org/10.1016/j.resconrec.2016.02.009 https://doi.org/10.1016/j.resconrec.2016.12.012 https://doi.org/10.1016/j.jclepro.2016.12.068 https://doi.org/10.1016/j.resconrec.2017.09.029 https://repository.javeriana.edu.co/handle/10554/11829 https://doi.org/10.1016/j.trd.2014.03.006 https://doi.org/10.1016/j.trd.2014.03.006 https://doi.org/10.1007/s11367-011-0251-4 https://doi.org/10.1016/j.resconrec.2016.12.003 https://doi.org/10.1016/j.resconrec.2016.12.003 https://revistas.eia.edu.co/index.php/reveia/article/view/706/661 https://revistas.eia.edu.co/index.php/reveia/article/view/706/661 https://doi.org/10.1007/s11367-011-0350-2 https://doi.org/10.1016/j.enbuild.2010.09.005 https://doi.org/10.1016/j.energy.2017.02.167 https://doi.org/10.1016/j.energy.2017.02.167 https://doi.org/10.1016/j.wasman.2019.01.011 https://repository.unimilitar.edu.co/handle/10654/11004 https://repository.unimilitar.edu.co/handle/10654/11004 https://doi.org/10.1016/j.jclepro.2016.02.144 https://doi.org/10.1016/j.jclepro.2016.02.144 https://doi.org/10.1016/j.jclepro.2016.09.149 https://doi.org/10.1016/j.jclepro.2016.09.149 https://doi.org/10.1016/j.jclepro.2014.11.067 351application of life cycle assessment (lca) methodology and economic evaluation for construction and demolition waste: a colombian case study suárez-silgado, s., roca, x. & gasso, s. (2016). product-specific life cycle assessment of recycled gypsum as a replacement for natural gypsum in ordinary portland cement: application to the spanish context. journal of cleaner production, 117, 150–159. https://doi.org/10.1016/j. jclepro.2016.01.044 suárez-silgado, s., molina, j. d. a., mahecha, l., & calderón, l. (2018). diagnóstico y propuestas para la gestión de los residuos de construcción y demolición en la ciudad de ibagué (colombia). gestión y ambiente, 21(1), 9-21. https://doi.org/10.15446/ga.v21n1.69637 tošić, n., marinković, s., dasic, t., & stanić, m. (2015). multicriteria optimization of natural and recycled aggregate concrete for structural use. journal of cleaner production, 87, 766-776. https://doi. org/10.1016/j.jclepro.2014.10.070 ulubeyli, s., kazaz, a. & arslan, v. (2017). construction and demolition waste recycling plants revisited: management issues. procedia engineering, 172, 1190-1197. https://doi.org/10.1016/j. proeng.2017.02.139 wang, t., wang, j., wu, p., wang, j., he, q., & wang, x. (2018). estimating the environmental costs and benefits of demolition waste using life cycle assessment and willingness-to-pay: a case study in shenzhen. journal of cleaner production, 172, 14-26. https://doi.org/10.1016/j. jclepro.2017.10.168 world bank (2013). doing business colombia. available in: http://espanol. doingbusiness.org/content/dam/doingbusiness/media/subnationalreports/db13-colombia-spanish.pdf. yay, a. s. e. (2015). application of life cycle assessment (lca) for municipal solid waste management: a case study of sakarya. journal of cleaner production, 94, 284-293. doi:10.1016/j.jclepro.2015.01.089 yeheyis, m., hewage, k., alam, m. s., eskicioglu, c., & sadiq, r. (2013). an overview of construction and demolition waste management in canada: a lifecycle analysis approach to sustainability. clean technologies and environmental policy, 15(1), 81–91. https://doi. org/10.1007/s10098-012-0481-6 yılmaz, t., ercikdi, b., & deveci, h. (2018). utilisation of construction and demolition waste as cemented paste backfill material for underground mine openings. journal of environmental management, 222, 250-259. doi:10.1016/j.jenvman.2018.05.075 yuan, h., & shen, l. (2011). trend of the research on construction and demolition waste management. waste management, 31(4), 670-679. doi:10.1016/j.wasman.2010.10.030 zabalza bribián, i., valero capilla, a., & aranda uson, a. (2011). life cycle assessment of building materials: comparative analysis of energy and environmental impacts and evaluation of the eco-efficiency improvement potential. building and environment, 46(5), 1133-1140. https://doi.org/10.1016/j.buildenv.2010.12.002 https://doi.org/10.1016/j.jclepro.2016.01.044 https://doi.org/10.1016/j.jclepro.2016.01.044 https://doi.org/10.15446/ga.v21n1.69637 https://doi.org/10.1016/j.jclepro.2014.10.070 https://doi.org/10.1016/j.jclepro.2014.10.070 https://doi.org/10.1016/j.proeng.2017.02.139 https://doi.org/10.1016/j.proeng.2017.02.139 https://doi.org/10.1016/j.jclepro.2017.10.168 https://doi.org/10.1016/j.jclepro.2017.10.168 http://espanol.doingbusiness.org/content/dam/doingbusiness/media/subnational-reports/db13-colombia-spanish.pdf http://espanol.doingbusiness.org/content/dam/doingbusiness/media/subnational-reports/db13-colombia-spanish.pdf http://espanol.doingbusiness.org/content/dam/doingbusiness/media/subnational-reports/db13-colombia-spanish.pdf https://doi.org/10.1007/s10098-012-0481-6 https://doi.org/10.1007/s10098-012-0481-6 https://doi.org/10.1016/j.buildenv.2010.12.002 figure1 _ref505936252 _ref505936610 zabalza2011 keywords: geothermal characteristics; shallow layer; geothermal energy; resource development; potential evaluation; model construction. palabras clave: características geotérmicas; capa superficial; energía geotérmica; desarrollo de recursos; evaluación potencial; construcción de modelo. how to cite item hou, j., luo, x., & zhang, l. (2020). establishment of evaluation model for shallow geothermal energy resource development potential based on characteristic of geotemperature. earth sciences research journal, 24(3), 317-325. doi: https://doi. org/10.15446/esrj.v24n3.89513 the potential analysis of geothermal energy based on geothermal characteristics requires research of the regional shallow geothermal energy. then, an evaluation model can be constructed. taking weinan city as an example, this paper studied the depth and temperature characteristics of the constant zone of subsurface temperature, the vertical variation characteristic of geotemperature, and the horizontal distribution characteristic of geotemperature in weinan city. based on the analysis of geotemperature characteristics, the analytic hierarchy process was used to construct the zoning evaluation system of the suitability of ground source, groundwater source, and surface water source. then, the suitability of weinan city was comprehensively zoned by the zoning evaluation system. the heat capacity and heat-exchange power of the heat pump system at the groundwater source were calculated based on the zoning results, and the heat-exchange power of the ground-coupled heat pump was obtained. according to the evaluation index, the development potential of shallow geothermal energy resources was evaluated. it can be seen that the heat pump system has great potential for heating in winter and cooling in summer through the experimental analysis model applied in this paper. the area is 50.79 km2. the area of the potential middle region of the buried pipe heat pump is 135.74 km2, and the potential of the pump system is low. the possible area of the ground source heat pump is 82.14 km2. upon comparing the model results with the current results, it was found that the consistency between the two is high. abstract establishment of evaluation model for shallow geothermal energy resource development potential based on characteristic of geotemperature establecimiento de un modelo de evaluación para el desarrollo potencial de recursos de energía geotérmica poco profunda con base en las características de la geotemperatura issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n3.89513 con el fin de construir un modelo de evaluación del potencial de desarrollo de recursos de energía geotérmica poco profunda basado en características geotérmicas es necesario investigar la energía geotérmica superficial regional. este estudio define la ciudad de weinan como ejemplo para analizar las características de profundidad y temperatura constante del subsuelo de la zona, la característica de variación vertical de la geotemperatura y la característica de distribución horizontal de la geotemperatura. con base en el análisis de la característica de geotemperatura se utilizó el proceso de jerarquía analítica para construir el sistema de evaluación de zonificación de la idoneidad de la fuente subterránea, la fuente de agua subterránea y la fuente de agua superficial. luego, la idoneidad de la ciudad de weinan fue zonificada por el sistema de evaluación. la capacidad de calor y la potencia de intercambio de calor del sistema de bombeo subterráneo se calcularon con base en los resultados de zonificación y se obtuvo la potencia de intercambio de calor de la bomba acoplada a tierra. según el índice de evaluación, se evaluó el potencial de desarrollo de los recursos de energía geotérmica poco profunda. a través del análisis experimental, aplicando el modelo en este documento, se puede ver que el sistema de bomba de calor de fuente subterránea tiene un gran potencial para calentar en invierno y enfriar en verano. el área es de 50.79 km2. el área de la región media potencial de la bomba de calor de tubería enterrada es de 135.74 km2 y el potencial del sistema de bomba de calor de fuente subterránea es bajo. el área potencial de la bomba de calor de fuente terrestre es 82.14 km2. al comparar los resultados de la evaluación de este modelo con los resultados reales se encuentra que la consistencia entre los dos es alta. resumen record manuscript received: 12/03/2019 accepted for publication: 07/06/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 3 (september, 2020): 317-325 jikun hou1,2*, xingqi luo2, lei zhang3 1power china northwest engineering corporation limited, construction management company, xi’an 710100, china 2xi’an university of technology, institute of water resources and hydropower engineering, xi’an 710048, china 3xi’an electric power college, power engineering department, xi’an 710032, china *corresponding author: hjk19800529@163.com g e o t h e r m a l e n e r g y https://doi.org/10.15446/esrj.v24n3.89513 https://doi.org/10.15446/esrj.v24n3.89513 https://doi.org/10.15446/esrj.v24n3.89513 mailto:hjk19800529@163.com 318 jikun hou, xingqi luo, lei zhang introduction energy problems, environmental, and sustainable development issues always trouble all over the world. the extensive energy use consumes a large number of non-renewable energy sources such as coal and oil. but those resources are expensive. energy has become a bottleneck restricting sustainable development (liu & nasr, 2019). people have paid more and more attention to energy conservation, new energy development, environmental pollution reduction, and sustainable development. all countries are trying to find and develop environmental, renewable energy. china is also facing the same problem (wang et al., 2017). as a renewable, environmental, clean energy resource, the shallow geothermal energy resource has been gradually recognized, accepted and valued in recent years in china (neupane et al., 2016). the shallow geothermal energy is a new member of the geothermal renewable energy family. it is also another form of solar energy, widely in the surface layer of the earth (thermostat layer). it can be recovered and renewable, and it is inexhaustible low-temperature energy (zhang & hu, 2018). the shallow geothermal energy is a low-temperature geothermal resource in the rock and soil of the earth’s surface; it is also an essential part of thermal energy. it is a clean, renewable resource with abundant reserves, wide distribution, and stable temperature. its development and utilization play an essential role in energy saving, emission reduction, and environmental protection. its development and utilization prospects are broad (zhang et al., 2017). currently, the development of technology is mature, which has been widely applied in heating and cooling. it is a good alternative energy and clean energy. with the gradual popularization of ground source heat pumps, the utilization of shallow geothermal energy has attracted more and more attention, which has become a new growth point of geothermal energy utilization. due to the excessive growth of demand in the area where the ground source heat pumps are popularized, the application of this technology shows a blind and disorderly development trend, ignoring the hydrogeological condition of the application site to a large extent, leading to the waste of groundwater resource and the inefficiency of the heat pumps (mihoub et al., 2016). to solve these problems objectively and ensure the rational development and utilization of shallow geothermal energy, we not only normalize the heat pump market but also investigate and evaluate the shallow geothermal resource according to the hydrogeological conditions, regional development and energy demand. we should scientifically carry out the suitability zoning of ground source heat pumps’ application and make reasonable development, and utilization plans to determine favorable development regions and appropriate development and utilization methods. all this to achieve orderly development, rational use and scientific management of shallow geothermal energy resources. these can provide a scientific basis for the macro decision-making on unified planning, improving energy efficiency and ensuring energy security (michopoulos et al., 2016). therefore, it is necessary to establish a model for evaluating the adaptability of ground source heat pumps. the standardized and systematic evaluation of shallow geothermal energy is critical. it is essential to organize them with various hydrogeological conditions for unified management. after considering the factors influencing the spatial distribution of shallow lowtemperature energy, the standard way is challenging to achieve timely, effective management. therefore, an evaluation model for shallow geothermal resources development potential should be constructed (zhang et al., 2016). some relevant literature shows that many scholars have studied it. liu licai has divided the suitability of water source heat pumps in the planning area of beijing. wang yabin has partitioned the suitability of hydrogeological conditions of the water source heat pump system in tianjin. wang tao evaluated the suitability of shallow geothermal energy utilization in key cities along the yellow river economic belt in ningxia. generally, china is still in the early stage due to its short development time. the unbalanced development and some problems need to be studied and solved urgently. otherwise, it will directly affect the development and sustainable utilization of shallow geothermal energy resources. in conclusion, there are many complex methods for evaluating the suitability of shallow geothermal. there is no specific mature method theory. some scholars directly use the method of artificial qualitative zoning to make the zoning intervals for the values of each element, and then corresponding suitability zoning is carried out. this method is too subjective, which is not scientific and reasonable enough. some scholars use programming or software for the subarea. although this method is efficient, it is not flexible enough. meanwhile, the results are usually complex, and its practicability is too low. on this basis, this paper uses the analytic hierarchy process (ahp) to divide the suitability of development and utilization. after that, the evaluation model is used to evaluate the development potential of shallow geothermal resources based on the zoning (alkhasov et al., 2016). material and method geothermal characteristic depth and temperature of the constant cone of subsurface temperature because the geothermal characteristics of different regions are different, this paper takes weinan city as an example to research. in order to analyze the relationship between the depth and temperature of the layer of constant temperature in weinan, eight wells of different geomorphic units in this region (six of them are the engineering geological exploration holes, two of them are motor-pumped wells) are monitored numerically with the depth of water temperature (if the water temperature is close to the formation temperature, the measured water temperature is substitute for the ground temperature). in the thermal response test construction, a u-shaped pipe is embedded in the borehole, and water is injected to ensure the tightness. the u-tube can be used for future temperature tests, and the digital temperature measurement single cable multi-probe detection system is adopted. the actual construction proves that when the stratum temperature changes slowly, it is accurate in any depth (zhou et al., 2017). the geothermal change curve and the scatter plot of the relationship between geo-temperature and depth in weinan are obtained by corresponding technology. the geothermal change curve is shown in figure 1. the scatter plot corresponding to the relationship between ground temperature, and depth is shown in figure 2. figure 1. map of ground temperature variation from the variation curve of ground temperature, we can see that the ground temperature of each monitoring hole in the depth of 0m-15 m underground varies with the rise and fall of temperature in different seasons. the seasonal variation of ground temperature is evident. the ground temperature in the depth of 15 m-20 m underground is basically constant, and the ground temperature is 15°c-17°c. the ground temperature of each monitoring hole at the same depth does not change with the season when the depth is below 20 m. the ground temperature in different seasons is basically unchanged, and the temperature increases linearly with the increase of burial depth (bisht & rupenthal, 2016). geothermal variation characteristic at vertical direction according to the data of eight geothermal monitoring points in different geomorphic locations, the geothermal field within the region generally increases with the depth at the vertical direction, but the geothermal changes vary in 319establishment of evaluation model for shallow geothermal energy resource development potential based on characteristic of geotemperature m. it can be seen that the geothermal anomaly regions are zonally distributed in the second and third terraces of weihe river in the southern part of weinan, and the average geothermal gradients are mostly concentrated in the range between 3.0°c/100 m-3.35 °c/100 m. the basement faults in the southern part of weinan are mature (czycholl et al., 2017), mainly influenced by the faults in the south bank of the f1 weihe river. it is affected by the faults in front of the tableland of the f2 weihe river and f3 zero river faults. in addition, two ground fissures are developed in the southern part of weinan and the trend is east-west, which are roughly parallel to the faults. the fault structure in this area is conducive to heat conduction from deep to shallow. the direction of geothermal anomaly is consistent with that of underlying fault structure. at present, the exploitation of shallow geothermal energy in weinan is relatively low. no one wants to exploit and utilize the geothermal anomaly area. development and utilization suitability zoning principle and method of suitability zoning the suitability zoning of shallow geothermal energy development and utilization runs through the research. its purpose is to provide the basis for the selection of development and utilization mode, resource evaluation and development and utilization planning. (1) zoning principle is based on the geological and hydrogeological conditions. the geological conditions are the important basis for the occurrence of shallow geothermal energy, and the hydrogeological conditions are the main influencing factors for the occurrence of shallow geothermal energy. for the purpose of development and utilization of shallow geothermal energy, the reasonable development and utilization mode and the effective heat pump system are the important media. we need to focus on both economic benefits and environmental protection. under the current technological and economic conditions, regions and ways with relatively good economic, environmental and energy-saving effects should be selected for the development and utilization of shallow geothermal energy. the plane division and vertical control should be combined. the range of plane zoning areas is the long-term planning area of urban construction in 2020. the overall evaluation control depth is 200 m. (2) zoning method: it is necessary to closely combine with the conditions of shallow geothermal energy in this area. firstly, the necessary conditions of suitability are divided by key factor method, and then the suitability level is divided by analytic hierarchy process (ahp) and comprehensive index method respectively. the analytic hierarchy process (ahp) is a qualitative, systematic and hierarchical analysis method. it has practicality and effectiveness in processing complex decision-making problems. the comprehensive index method refers to the comprehensive evaluation index which uses the weight calculated by the analytic hierarchy process and the numerical value obtained by the fuzzy evaluation method to put forward the economic benefit index. the more influential factor on suitability division will be more important. by establishing the comparison matrix, we can check the consistency of comparison matrices and adjust the comparison matrix if necessary, so as to achieve acceptable consistency. according to the occurrence condition of shallow geothermal energy, it can be divided into three development and utilization modes: ground-source type (buried pipe), groundwater-source type and surface-water source type, and then we can divide them into different suitable areas. ground source types can be divided into five types: favorable area, suitable area, general area, undesirable area and unsuitable area. among them, suitable area can be divided into three sub-areas: loose rock subregion, semi-consolidated rock subregion and consolidated rock subregion (winters et al., 2016). the groundwater source can be divided into three types: favorable area, suitable area and unsuitable area. the surface water source can be divided into two types: suitable area and unsuitable area. suitability zoning evaluation system of ground source the geological and hydrogeological conditions, construction conditions, thermophysical properties, topography, geomorphology, environmental geological problems are chosen as the attribute-level indexes for the evaluation of suitability zoning of development and utilization of geo-source type. where, the thickness of the loose layer, the probability of geological hazards and the different depths. in the normal temperature zone (15 m-25 m), especially the near-surface area of 0 m-3 m, the ground temperature is affected by seasonal climate change, and the fluctuation is sensitive. moreover, the influence of climate change on the ground temperature between 3 m and 15 m decreases gradually. under the normal temperature zone of about 25 m, the ground temperature has not been influenced by local climate change. the temperature increases with depth. the ground temperature is19.8°c-22.0°c when the depth is 200 m. the average geothermal gradient is 1.18/100 m to 3.57/100 m. based on the monthly average water temperature at different depths of eight long-term observation wells, the variation curves of ground temperature in various depths and different periods (the water temperature below groundwater level is the same as the ground temperature) with depth are drawn to analyze the vertical variation characteristics of ground temperature in natural state. according to the curve, the vertical geothermal distribution feature in the research area belongs to the linear gradual rising type. under 20 m, the ground temperature of each observation hole at different depths does not change significantly with the seasons. with an increase of depth, the ground temperature increases linearly. the change of ground temperature in qianlu waterworks (wn178) is the largest, from 40m to 180m. as the ground temperature increases by about 5°c and the geothermal gradient is about 3.57°c/100 m. the changing trend of ground temperature in zhang yicun (wg001) is relatively slow, from 30m to 100m. the ground temperature increases by about 0.81°c, and the geothermal gradient is 1.18°c/100 m. in the stratum warming zone, the geothermal temperature is controlled by deep geothermal warming, and the geothermal temperature gradually increases with depth. due to the influence of geological structure, groundwater activity, and stratum rock characteristics, vertical geothermal warming varies from place to place. it is concluded that the geothermal temperature in the research area changes with depth. the features are mainly influenced by the geological structure and groundwater activity (ruelle et al., 2016). geothermal horizontal distribution characteristic generally, the ground temperature in 100 m buried depth is high in the south and low in the north. there is a geological fault in the front of the second terrace on the south bank of weihe river. the ground temperature in the 100m buried depth of the third terrace on the south of fault can reach 19.1°c. the main reason is that the basement faults are mature in the area, conducive to the heat conduction from deep to shallow. the distribution direction of the third terrace in the south of weihe river is consistent with the underlying fault structure. according to the contour maps of ground temperature and average geothermal gradient at 100 m depth, the geotemperature is abnormal if the geotemperature is higher than 17.5°c; and the average geothermal gradient is higher than 3°c/100 figure 2. scatter map corresponding to the relationship between ground temperature and depth 320 jikun hou, xingqi luo, lei zhang topography are the key factors. on the premise of determining the key factors, the method of analytic hierarchy process and the method of comprehensive index are used to divide the development and utilization suitability zoning of shallow geothermal energy of ground source type. as shown in figures 3 and figure 1: figure 3. structure diagram of suitability division for development and utilization of source type table 1. selection table of crucial factors of suitability suitability partition key factor suitable area an area with loose layer thickness greater than 100 m lower area semi-consolidated rock area general suitable area geological hazard (collapse, landslide, debris flow, karst ground collapse) low-prone area, and loose layer thickness of 30 m-50 m difference suitable area geological hazards (collapse, landslide, debris flow, karst ground collapse) in prone areas, middle hills and above geomorphologic units unconfined area geological hazard (collapse, landslide, debris flow, karst ground collapse) high-prone area from figure 3, we can see that the structure of suitability division of development and utilization of geo-source type is mainly based on the geothermal characteristic. by considering the hydrogeological condition, equation condition, thermal physical property, topography, geological hazards and other factors, we can divide the suitability of development and utilization of geo-source type. meanwhile, it can be seen that the suitability can be divided into five categories (yan & xu, 2016). subarea evaluation system for suitability of groundwater source the groundwater abundance and groundwater dynamic field conditions are selected as attribute-level indexes for the suitability evaluation of groundwater source. the six indexes such as single-well water inflow, effective thickness of aquifer, recharge capacity, water level burial depth, water level descent rate and recharge modulus are selected as element layers of the hierarchical analysis. where, the groundwater richness, the recharge capacity and environmental geology are the key factors (li et al., 2016). on the premise of determining the key factors, the method of analytic hierarchy process and the comprehensive index are used to divide the suitability of development and utilization of shallow geothermal energy of groundwater sources. figure 4. structure diagram of suitability division for development and utilization of underground water source type table 2. selection of key factors key evaluation factor suitability suitable area more suitable area unconfined area single well water inflow/(m3/d) >3000 1000-3000 <1000 groundwater irrigation and production ratio/% >80 50-80 <50 recharge capacity best secondary difference (1) the area with land subsidence > 500 mm is an unsuitable area of groundwater source type. (2) land subsidence 0 mm-500 mm area, take down a suitability level from figure 4, the division structure of suitability of development and utilization of groundwater source is mainly based on the geothermal characteristic. after considering the geological condition, hydrogeological condition and groundwater dynamic field condition, the suitability of groundwater source can be divided by above factors. meanwhile, the key evaluation factors of suitability division of groundwater source development and utilization mainly include the number of single well users, groundwater irrigation-production ratio and recharge capacity. according to above evaluation factors, the suitability evaluation of groundwater source development and utilization can be achieved. zoning assessment system for suitability of surface water source the development and utilization of surface water source shallow geothermal energy can be divided into two types: separated type and closed type. the evaluation indexes of suitability of surface water source type mainly include the recyclable amount of surface water, water depth, water temperature, water quality and distance from the project. 321establishment of evaluation model for shallow geothermal energy resource development potential based on characteristic of geotemperature table 3. classification of suitability for development and utilization of surface water source types utilization mode open closed volume during dry season/(×104m3) >20 >20 water quality meet the requirements of the specification meet the requirements of the specification water depth in dry season/m >3 >3 coldest monthly mean water temperature >10 >5 distance between utilization engineering and water source/km <0.5 <0.5 comprehensive evaluation standard of suitability it is appropriate to meet the five indicators from table 3, we can see that the evaluation indexes of development and utilization of surface water resource mainly include five evaluation indicators: volume, water quality and depth in dry season. the suitability of surface water resources development and utilization can be divided by the degree of conformity of development and utilization of surface water resources to these indicators (wen & fang, 2018). evaluation of tap potential of shallow geothermal energy resource the evaluation for development potential of shallow geothermal resources includes the evaluation of shallow geothermal capacity and heat-exchange power. based on the zoning results, the evaluation index of shallow geothermal resource development potential is selected at first, and then the development potential of shallow geothermal resource is evaluated by the evaluation index. heat capacity the shallow geothermal energy capacity is the unit temperature difference heat stored in shallow rock, soil and groundwater. the thermal storage method is adopted. the depth range is 200 m. the calculated area is the suitable area. q q q qr s w a= + + (1) in the formula, qr is the geothermal energy capacity. qs is the energy capacity in rock and soil. qw is the energy capacity of water contained in rock and soil (sowizdzal, 2016). qa is the energy capacity of air contained in rock and soil. the formula for calculating the energy capacity of rock and soil is as follows: q c mds s s= ( )ρ −φ1 (2) in the formula, s is the density of rock and soil. cs is the specific heat capacity.  is the porosity. m is the area. d is the thickness of rock and soil are considered (al-samari & ali, 2018). the formula for calculating the energy capacity of water in rock and soil are as follows: q c mdw w w= ρ ω (3) in the formula, w is the water density. cw is the specific heat capacity. is the water content of rock and soil. the formula for calculating the energy capacity of air in rock and soil are as follows: q c mda a a= ( )ρ φ−ω (4) in the formula, a is the air density. ca is the specific heat capacity of air. according to the main rock character of the aquifer, the depth of the quaternary base boundary and the depth of shallow groundwater, this region is divided into 148 sub regions. according to the difference of area conditions, the sub regions are merged into 18 evaluation units. according to the main rock character of aquifer, they are divided into three areas, and then we calculate them. heat-exchange power of groundwater source heat pump system the heat-exchange power of groundwater heat pump systems is calculated by groundwater demand conversion method. the calculation process is as follows: q q nq h= × ×  (5) in the formula, qq is the heat-exchange power of shallow geothermal energy in the evaluation area. qh is the heat-exchange power of shallow geothermal energy of a single well. n is the number of drillable holes in the calculation area.  is the land usage rate. the formula for calculating the heatexchange power of shallow geothermal energy of single well is as follows: q q t ch w w w= × ×∆ ρ −1 16 10 5. (6) in the formula, qh is the heat-exchange power for a single well. qwis the amount of groundwater recycling utilization. ∆t is the difference in temperature for groundwater utilization. w is the density of water, 1000 kg/m3. cw is the specific heat capacity of water, 4.18 kj/(kg·°c). heat-exchange power of buried tube heat pump system according to the heat-exchange power of the ground-coupled heat pump, the unit heat transfer method (fu, 2018) is adopted. the calculation process is as follows: q q mk k= ×10 3 (7) in the formula, qk is the heat-exchange power of shallow geothermal energy, and qk is the heat-exchange power of shallow geothermal energy per unit area. m is the calculation area. establishment of shallow geothermal energy potential evaluation model according to the heat-exchange power of ground source heat pump system and the indexes of heating in winter and cooling in summer, the area for heating and cooling of suitable and more suitable regions of groundwater ground source heat pump in winter and summer was further calculated, and then the resource potential of ground source heat pump system (zhang et al., 2018) is obtained. the detailed formulas are as follows: m q q q= 1000 (8) q m mzq = (9) in formulas, m is the heating/cooling area of groundwater/buried pipe ground source heat pump system. qq is the heat-exchange power in the suitable region and the more suitable region of groundwater/buried pipe ground source heat pump system. q is the heating index in winter and the refrigeration index in summer. qzq is the resource potential of groundwater/buried pipe ground source heat pump system. m is the heating/refrigerating area of groundwater/ buried pipe ground source heat pump system. m is the area of the calculation region. through the above formulas, the index table of potential evaluation zoning is obtained. 322 jikun hou, xingqi luo, lei zhang table 4. potential assessment division evaluation project divisional indicators unit area heating area/(×104 m2/km2) >32 20-32 <20 refrigerable area per unit area/(×104 m2/km2) >40 28-40 28 the land use coefficient is 100% of the heating area per unit area./(×104 m 2/km2) >162 <162 the land use coefficient is 100% refrigerable area per unit area/(×104 m2/km2) >204 <204 potential zoning high potential area central potential area low potential area on the basis of the above analysis, the development potential of shallow geothermal energy resources can be evaluated by the above table. results in order to verify the validity of the evaluation method, the proposed method is used to analyze the occurrence condition of shallow geothermal energy in weinan city and the characteristic of geotemperature in weinan city. it evaluates the suitability of development and utilization and the heating/ cooling potential of groundwater ground source heat pump system and buried pipe ground source heat pump system. the heat capacity of shallow geothermal energy is calculating. the environmental benefits and economic benefits of shallow geothermal energy development are discussed. the detailed analysis process is shown below. suitability zoning of groundwater ground source heat pump in order to research the suitability zoning effect of groundwater ground source heat pump of the proposed method, a practical experiment is carried out. the results of the practical adaptability zoning are shown in figure 5. the area and percentage of each region after the zoning are shown in table 5. figure 5. suitability zoning of groundwater ground source heat pump in weinan city from the charts, we can see that the suitability of groundwater source heat pumps is obviously different due to the difference of hydrogeological characteristics. the suitability can be divided as follows. the suitable region of groundwater source heat pump: it is distributed in the mid-weihe floodplain of the evaluation region. the rock character of aquifer is composed of gravel, gravel-bearing medium-coarse sand and silty clay. the stratum has good water conductivity, including double-layer phreatic water-confined water. the waterabundance is good. the unit water inflow is 10 m3/h·m-30 m3/h·m. therefore, it is suitable for groundwater pumping. the area of this region is 61.14 km2, accounting for 22.23% of the total area of the evaluation area. the suitable region of groundwater source heat pump: it is distributed in the first terrace on the north bank of weihe river, the terrace on the south bank of weihe river and the alluvial-proluvial fan region. the rock character of aquifer is quaternary holocene alluvion, including the gravelly sand, medium sand, coarse sand, silt and silty clay. the groundwater pumping condition in this area is relatively good, the water abundance is medium-weak, the unit inflow is 1 m3/h·m-10 m3/h·m. the area is 196.60 km2, accounting for 71.48% of the total area of the evaluation area. the unsuitable region of groundwater source heat pump: it is mainly distributed in the third terrace covered by loess in the south evaluation region. the stratum lithology is mainly pleistocene aeolian loess and silty clay with little thin sandy soil. the water abundance of groundwater in this region is weak, and the water conductivity is poor. the unit water inflow is 1 m3/h·m-5 m3/h·m, which is not suitable for the project of groundwater source heat pump. the unsuitable area of groundwater source heat pump is 10.93 km2, accounting for 3.98% of the total area of the evaluation area. the unevaluated area is weihe river, namely the existing river in the evaluation area, with an area of 6.37 km2, accounting for 2.32% of the total area of the evaluation region. heat exchange power and potential of groundwater ground source heat pump in order to study the validity of calculation and evaluation for the heat-exchange power and potential of groundwater ground source heat pumps in weinan city, it is necessary to carry out practical analysis. the analysis results are shown in table 6. from table 6, the total area of suitable and more suitable regions of groundwater source heat pump in this area is 145.22 km2, and the total heat transfer power of groundwater source heat pump system is 1.17×106 kw (winter)/ 3.58×106 kw (summer). when using the groundwater ground source heat pump system to develop and utilize the shallow geothermal energy, the average heating potential in winter is 1.48×105m2/km2, and the average refrigeration potential in summer is 2.13×105 m2/km2. heat exchange power and potential of buried tube ground source heat pump in order to study the validity of this method in calculating and evaluating the heat-exchange power and potential of ground source heat pump, it is necessary to take the average temperature of fluid in the buried pipe of weinan city in winter and summer as 6°c and 32°c respectively. the heat-exchange power potential of the ground source heat pump system in weinan city is calculated and evaluated by the proposed method. the results obtained by the proposed method are as follows. table 5. statistical table of suitability zoning and area of groundwater ground source heat pump suitability zoning real area /km2 percentage suitable area 36.49 11.69 more suitable area 153.61 49.24 general suitable area 95.48 30.61 difference suitable area 10.62 3.4 unconfined area 12.56 4.03 unevaluated area 3.21 1.03 total 311.97 100 323establishment of evaluation model for shallow geothermal energy resource development potential based on characteristic of geotemperature table 7 shows that the heat-exchange power in winter and summer is 1.04×107 kw and 1.26×107 kw respectively. the average heating potential in winter is 9.64×105 m2/km2 and the average refrigeration potential in summer is 1.18×106 m2/km2 when using the ground source heat pump system to develop and utilize shallow geothermal energy. analysis of resources development potential in order to verify the accuracy of the proposed method in evaluating the development potential of shallow geothermal energy, it is necessary to carry out actual experimental analysis. weinan city is divided into three regions. finally, the evaluation results as shown in table 8. from table 8, we can see that land use coefficient has a great influence on the potential of ground source heat pump systems. meanwhile, most of the floodplain areas in weinan are high potential areas of ground source heat pumps. the water level in the area is shallow and the effective aquifer is thick. the comprehensive heat transfer coefficient is high and single-hole heat exchange power is large. the ground source heat pump system has large heating/ cooling potential in winter and summer. the area is 50.79 km2, accounting for 18.47% of the total area. the first-order region of weihe river within the zoning area is the middle potential area of ground source heat pump, with an area of 135.74 km2, accounting for 49.35% of the total area. the built area of weinan has small land use coefficient and low potential of ground source heat pump system, which belongs to the low potential area of ground source heat pump. the area is 82.14 km2, accounting for 29.86% of the total area. through actual investigation and analysis, we can see that the evaluation results of the proposed method are in good agreement with the actual results. therefore, the evaluation accuracy of the proposed method is higher. discussions there are many ways to exploit the shallow geothermal energy resource. the groundwater ground source heat pump and buried pipe ground source heat pump are the two main ways. the groundwater source heat pump system consists of groundwater heat-exchange system, computer room system and terminal system. the working principle of the ground water source heat pump system is to pump groundwater and exchange heat through a ground source heat pump unit or plate heat exchanger. in winter, all the low-temperature water pumped from underground water after heat exchange through a ground source heat pump unit is recharged into the same aquifer, and the heat quantity is used for the building heating and domestic hot water. in summer, all the hightemperature water is recharged into the same aquifer after the heat exchange of pumped groundwater through a ground source heat pump unit or plate heat exchanger. at the same time, the high-temperature water can be exchanged through a heat pump unit to provide domestic hot water. the characteristics are as follows: at suitable sites for sinking a well, a certain number of pumping and irrigation wells are drilled, and heat is absorbed or discharged from groundwater by recycling, so that the number of wells and the area are few. the heat-exchange way of the underground heat-exchange system is mainly the table 6. heat transfer power and potential of groundwater ground source heat pump zoning and numbering zhao zhai-li jia wan longjiang town-hantai district-pu zhen liangshan town zhongsuo town-dahekan town sujiashan village total zonal area/km2 21.95 94.37 15.33 13.24 0.33 145.22 suitability more suitable suitable suitable suitable more suitable winter heat transfer power/kw 8.83×104 1.21×106 2.67×105 2.32×105 1.16×103 1.17×106 summer refrigeration power/kw 1.76×105 2.42×106 5.34×105 4.62×105 2.32×103 3.58×106 winter heating potential/ (m2/km2) 5.39×104 1.72×105 2.33×105 2.35×105 4.68×104 1.48×105 summer refrigeration potential/(m2/km2) 7.77×104 2.48×105 3.36×105 3.38×105 6.74×104 2.13×105 table 7. calculation table of heat transfer power meter potential of a ground source heat pump with buried pipe zoning and numbering praise canal longjiang town-qili town-paving town liangshan town-belt floodplain on the south bank of hanjiang river floodplain on the east bank of lianshui river floodplain east of dahekan town on the south bank of hanjiang river the first class terrace south of liangshan town huilong templeyuying village total zonal area/km2 3.36 112.69 6.04 1.26 2.36 9.29 9.94 144.94 suitability more suitable suitable suitable suitable suitable more suitable more suitable winter heat transfer power/kw 2.75×105 7.93×106 4.60×105 1.02×105 1.98×105 6.98×105 7.54×105 1.04×107 summer refrigeration power/kw 3.34×105 9.61×106 5.59×105 1.23×105 2.41×105 8.48×105 9.12×105 1.26×107 winter heating potential/(m2/km2) 1.09×106 9.44×105 1.01×106 1.09×106 1.12×106 1.03×106 1.02×106 9.64×105 summer refrigeration potential/(m2/km2) 9.59×105 8.24×105 8.92×105 9.47×105 9.83×105 8.83×105 8.89×105 1.18×106 324 jikun hou, xingqi luo, lei zhang thermal convection, and the temperature of groundwater is basically constant throughout the year. the initial cost is lower than that of a buried pipe ground source heat pump system. under appropriate groundwater depth, cop is higher and the operation cost is lower. the buried-pipe ground source heat pump system is similar to the groundwater ground source heat pump system. it consists of a heat exchange system with buried pipe, computer room system and terminal system. the working principle of the buried-pipe heat exchange system: the heat transfer medium circulates in the sealed vertical or horizontal buried pipe, and then the temperature difference among the heat transfer medium, the underground rock and soil and the groundwater is used to exchange heat, so that the purpose of utilizing the shallow geothermal energy can be achieved. meanwhile, the heating and cooling of buildings and domestic hot water through heat pump technology can be realized. the features are as follows: the project needs to drill many boreholes according to the cold load and heat load, and sets a closed circulating pipe with certain strength, corrosion resistance and good heat transfer performance, and then uses the circulating liquid to exchange heat, so as to connect all circulating pipes. finally, it enters the computer room and main engine. the buried-pipe ground source heat pump (gshp) system is based on the fact that the temperature of underground rock and soil and groundwater remains constant throughout the year. if the vertical buried tube heat exchanger is used to conduct and absorb heat between underground rock and soil and groundwater, the heat exchange efficiency is lower than that of groundwater ground source heat pump systems. compared with traditional air-conditioning systems and groundwater ground-source heat pump systems, the main disadvantage of buried-pipe ground-source heat pump systems is high initial investment of ground-source heat exchangers. the heat exchanger of a ground source heat pump system with a buried pipe occupies a larger area than that of the pumping well of an underground water source heat pump system. the ground source heat pump system does not draw water from the ground. theoretically, the ground source heat pump system with buried pipe has less impact on the environment of underground space than that of groundwater source heat pump, and its procedure is simpler. therefore, the ground source heat pump system with buried pipe has the characteristics of green environmental protection, high efficiency and energy saving, low operation cost, mature technology, wide application range (it is applicable to any stratum and building). in addition, it does not need to pump the groundwater. conclusions the evaluation of shallow geothermal energy development potential is an effective technology for rational planning of urban layout, protection of shallow geothermal resources and sustainable development and utilization of shallow geothermal resources. based on the analysis of the geotemperature characteristic in weinan, this article achieves the suitability zoning of comprehensive development of groundwater source heat pump, buried pipe ground source heat pump and shallow geothermal energy according to the requirements of analytic hierarchy process. meanwhile, this article uses the shallow geothermal capacity, heat-exchange power and resource potential to quantitatively evaluate the research region. the total area of suitable and more suitable regions of groundwater ground source heat pump in weinan is 145.22 km2. when using groundwater ground source heat pump systems to develop and utilize the shallow geothermal energy, the average heating potential in winter is 1.48×105 m2/km2, and the average refrigeration potential in summer is 2.13×105 m2/km2. the potential middle area of ground source heat pump in weinan is 135.74 km2, accounting for 49.35% of the total area, while the potential low area of ground-source heat pump in weinan is 82.14 km2, accounting for 29.86% of the total area. the result and the amount of available resources directly reflect the prospects and potential of shallow geothermal energy development and utilization in weinan. references alkhasov, a. b., alkhasova, d. a., ramazanov, a. s., & kasparova, m. a. (2016). prospects of development of highly mineralized hightemperature resources of the tarumovskoye geothermal field. thermal engineering, 63, 404-408. al-samari, a. s. h., & ali, s. d. (2018). experimental evaluation of surface geothermal energy for air conditioning applications in iraq. journal of mechanical engineering research and developments, 41(1), 38-43. bisht, r., & rupenthal, i. d. (2016). plga nanoparticles for intravitreal peptide delivery: optimization, characterization and toxicity evaluation. pharmaceutical development & technology, 20, 1-25. czycholl, i., kniese, c., schrader, l., & krieter, j. (2017). assessment of the multi-criteria evaluation system of the welfare quality® protocol for growing pigs. animal an international journal of animal bioscience, 11, 1-8. fu, b. w. (2018). study on fuzzy control of extraction temperature of traditional chinese medicine based on generalized prediction. automation & instrumentation, 14-17. li, y., xu, s. c., & xu, p. (2016). simulation of low temperature starting of proton exchange membrane fuel cell system. chinese journal of power sources, 40, 1202-1204. liu, h., & nasr, m. (2019). construction of deep thermal models based on integrated thermal properties used for geothermal risk management. mathematical geosciences, 51, 295-317. table 8. evaluation of development potential partition number i ii iii average zonal area/km2 143.02 114.76 10.89 268.67 suitability both land source and water source are suitable or more suitable both land source and water source are suitable or more suitable only the ground source is suitable or more suitable. considering the heating potential of land use coefficient in winter/(m2/km2) 2.68×105 2.04×105 1.54×105 2.36×105 consider the potential of land use coefficient for summer refrigeration/(m2/km2) 3.43×105 2.62×105 1.95×105 3.03×106 heating potential in winter when land use efficiency is 100%/(m2/km2) 1.37×106 1.48×106 1.55×106 1.43×106 summer refrigeration potential when land use efficiency is 100%/(m2/km2) 1.75×106 1.88×106 1.96×106 1.82×106 potential division central potential area low potential area low potential area 325establishment of evaluation model for shallow geothermal energy resource development potential based on characteristic of geotemperature michopoulos, a., voulgari, v., tsikaloudaki, a., & zachariadis, t. (2016). evaluation of ground source heat pump systems for residential buildings in warm mediterranean regions: the example of cyprus. energy efficiency, 9, 1-16. mihoub, r., chabour, n., & guermoui, m. (2016). modeling soil temperature based on gaussian process regression in a semi-arid-climate, case study ghardaia, algeria. geomechanics and geophysics for geoenergy and geo-resources, 2, 397-403. neupane, g., mattson, e. d., mcling, t. l., palmer, c. d., smith, r. w., wood, t. r., & podgorney, r. k. (2016). geothermometric evaluation of geothermal resources in southeastern idaho. geothermal energy science, 4, 11-22. ruelle, e., delaby, l., wallace, m., & shalloo, l. (2016). development and evaluation of the herd dynamic milk model with focus on the individual cow component. animal, 10, 1986-1997. sowizdzal, a. (2016). possibilities of petrogeothermal energy resources utilization in central part of poland. applied ecology and environmental research, 14(2), 555-574. wang, g. l., zhang w., liang, j. y., lin, w. j., liu, z. m., & wang, w. l. (2017). evaluation of geothermal resources potential in china. acta geoscientica sinica, 38, 449-459. wen, p., & fang, h. z. (2018). simulation modeling technology for life prediction of lithium ion battery. computer simulation, v.35,70-74. winters, a. a., bou-ghannam, s., thorp, h., hawayek, j. a., atkinson, d. l., bartlett, c. e., silva, f. j., hsu, e. w., moreno, a. p., grainger, d. a., & patel, a. n. (2016). evaluation of multiple biological therapies for ischemic cardiac disease. cell transplantation, 25, 1591-1607. yan, c., & xu, h. d. (2016). discussion on modulation strategy of t-type threelevel photovoltaic inverter adapted to low voltage crossing. journal of power supply, 14, 1-9. zhang, x., & hu, q. (2018). development of geothermal resources in china: a review. journal of earth science, 29, 452-467. zhang, x., tong, h., & yan, h. y. (2018). optical isomerization of valine catalyzed by hydrogen-oxygen root water cluster and damage mechanism induced by hydroxyl radical in water environment. journal of jilin university (science edition), 56, 260-270. zhang, y., yu, h., li, j., yu, z., & zhang, j. (2016). prediction models of deep hydrothemal geothermal potential areas based on gis: a case study of western anatolia, turkey. journal of jilin university, 46, 855-864. zhang, y., zhang, t., yin，r., zheng, j., liu, t., & xie, y. (2017). geothermal anomaly areas exploration and ground temperature prediction based on 2-meter temperature survey. journal of jilin university, 47, 189-196. zhou, j., qin, x., liu, l., & hu, y. (2017). a potential evaluation model for land consolidation in fragmental regions. ecological indicators, 74, 230-240. keywords: sgeological structure; ecological environment; destruction; restoration countermeasures. palabras clave: : roca de semicarbón; evolución geológica; minería sin pilar; peligros de la zona ciega; extracción de gas. how to cite item huang, y. (2020). destruction process and restoration countermeasures of the ecological environment of a comprehensive geological structure. earth sciences research journal, 24(4), 429-437. doi: https://doi.org/10.15446/ esrj.v24n4.92387 aiming at the destruction of the ecological environment of the comprehensive geological structure, the traditional restoration countermeasures have the problems of high input cost and low economic benefits. for this reason, the ecological environment destruction process and restoration countermeasures of the comprehensive geological structure were proposed. the common geological structure characteristics and the impact of activities on the ecological environment in the mining area were analyzed. based on the analysis results, an evaluation system for the degree of damage to the ecological environment by mining activities was constructed. according to the expert scoring method, a rating standard was developed to measure the degree of damage to the ecological environment. depending on the degree of damage, appropriate recovery measures were formulated. the experimental results show that compared with the traditional restoration countermeasures, considering the destruction of the ecological environment during the mining process of comprehensive geological structures, the proposed restoration countermeasures have the advantages of low cost and high economic benefits. abstract destruction process and restoration countermeasures of the ecological environment of a comprehensive geological structure proceso de destrucción y contramedidas de restauración del entorno ecológico de una estructura geológica integral issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.92387 con el objetivo de destruir el entorno ecológico de la estructura geológica integral, las contramedidas de restauración tradicionales tienen el problema de un alto costo de los insumos y bajos beneficios económicos. por esta razón, se propuso el proceso de destrucción del medio ambiente ecológico y las contramedidas de restauración de la estructura geológica integral. se analizaron las características de la estructura geológica común y el impacto de las actividades sobre el medio ambiente ecológico en una área minera. con base en los resultados del análisis, se construyó un sistema de evaluación del grado de daño al medio ambiente ecológico por las actividades mineras. de acuerdo con el método de calificación de expertos, se desarrolló un estándar de calificación para medir el grado de daño al medio ambiente. dependiendo del grado de daño, se formularon las medidas de recuperación adecuadas. los resultados experimentales muestran que en comparación con las contramedidas de restauración tradicionales, considerando la destrucción del ambiente ecológico durante el proceso de extracción de estructuras geológicas integrales, las contramedidas de restauración propuestas tienen las ventajas de bajo costo y altos beneficios económicos. resumen record manuscript received: 13/11/2019 accepted for publication: 08/10/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 429-437 yanyan huang school of management, heilongjiang university of science and technology, harbin, 150022, china *corresponding author: hyy123_75@163.com e n v ir o n m e n ta l g e o l o g y 430 yanyan huang introduction economic development and ecological environmental protection are major issues of common concern in countries around the world today. development is the main body of the times in all countries and regions today, and it has become the key to solving all other problems. various countries and regions are doing everything possible to promote their own development (staccione et al., 2019). the ecological environment is the prerequisite for economic development, the resources and environmental foundation required for human survival and development, and the indispensable material and energy guarantee for regional development. in the idea of sustainable development, sustainable economic development must be based on a virtuous cycle of the ecological environment. without the material flow and energy flow created by the ecological environment system, the economic system cannot operate, and the sustainable development of the economy cannot be talked about. therefore, there is a close interdependence and restriction between economic development and the ecological environment (song et al., 2019). in a sense, the course of economic development in human society is a dialectical unified process of human beings’ continuous recognition, transformation, respect, and obedience to the ecological environment. ecological protection work is to continuously enhance the capacity for sustainable development, improve the ecological environment, significantly improve the efficiency of resource utilization, promote harmony between man and nature, and promote the entire society to a civilized development path of productive development, wealthy life, and good ecology. as one of the goals of building a well-off society in an all-around way, it further clarified the relationship between humans and the ecological environment and its significance in achieving sustainable development (dochartaigh et al., 2019). geological disasters caused by the destruction of the ecological environment are disasters related to geological effects that endanger the safety of people’s lives and property caused by natural factors or human activities during the development and evolution of the earth. sudden geological disasters are the main activities that occur suddenly and complete disasters in a relatively short period of time. they mainly include volcanoes, earthquakes, collapses, landslides, mudslides, etc., which cause human life and property loss (sur & chauhan, 2019). china is one of the countries with the most severe geological disasters in the world. there are many types of disasters, high frequency of occurrence, wide geographical distribution, and large disaster losses. it severely restricts the development of the national economy in disaster-prone areas and threatens the safety of people’s lives and property in disaster prone areas (jurczak et al., 2018). over-exploitation of resources is one of the main factors leading to geological disasters. whether it is underground mining or open-pit mining, the impact on the geological environment mainly includes three aspects: the destruction of the subsidence to the soil environment, water resources and ecological environment, the destruction of the ecological environment by waste (grinding stone, slime, tailings water), and the destruction of greenhouse gases and mine wastewater to the atmosphere and water during mining (josepheph & humphries, 2018). therefore, the geological environment problems encountered in the mining process were analyzed and studied, and at the same time, the current status of environmental protection and rehabilitation of mine geology in china and abroad was used for reference, and targeted prevention measures were studied and discussed. from the implementation of measures and investment of funds, ecological environment restoration countermeasures were proposed. eventually, the problems of high input costs and low economic benefits of ecological restoration in traditional environmental restoration countermeasures were solved. materials and methods the ecological environment and mining resources are important material foundations on which human beings depend for survival and development. for a long time, due to the unreasonable development and utilization of resources, a variety of mining environmental geological problems have been raised (chen et al., 2019). according to the statistical analysis of national geological environmental survey data, environmental geological problems caused by mining activities mainly include geological disasters such as landslides, collapses, mudslides, ground subsidence, ground subsidence, and ground fissures, occupation and destruction of land resources, destruction of water resources, land desertification and soil erosion (webster et al., 2019; cui & wang, 2019). the specific content is shown in table 1. considering the geological structure of the mining area, from the perspective of the four problems shown in the table, the damage of mining to the ecological environment was analyzed, and then corresponding countermeasures for ecological environment restoration were proposed. common geological structural features according to the investigation of actual mining projects, common geological structures in mining projects include faults, subsidence columns, and fold structures. taking into account the characteristics of the above geological structures, the damage to the ecological environment was analyzed. generally, in the mining area, there are reverse faults and normal faults that are substantially parallel to the axial direction of the anticline, and faults develop along the two sides of the anticline. the normal fault is the fault that the upper disk descends and the lower disk rises, as shown in figure 1. the reverse fault is the fault that the upper disk raises and the lower disk falls, as shown in figure 2. table 1. major damage problems caused by mining to the ecological environment type division expression form affect result geological disaster natural stress-strain changes in rock and soil during mining, causing collapse, landslides, mudslides, ground subsidence (karst collapse), ground subsidence (goaf subsidence), ground fissures cause economic losses and casualties, complicate resource construction and mining, increase workload and reduce mining efficiency occupation or destruction of land, vegetation, and other resources occupying and destroying land, destroying surface vegetation, destroying geological sites, destroying landforms and landscapes, destroying buildings and transportation infrastructure decreased forest, grassland, and arable land, deteriorating habitat conditions for wild animals, reduced crop yields, and reduced forestry yields water damage damage to surface and groundwater systems near mining areas regional hydrological and hydrogeological conditions are destroyed, the original balance of groundwater and surface water is broken, and water resources are reduced water and soil pollution surface water pollution, groundwater pollution, soil pollution, etc. water bodies and soils are polluted by harmful substances and soil quality deteriorates 431destruction process and restoration countermeasures of the ecological environment of a comprehensive geological structure destroyed, and the geological ecological environment was severely damaged. a large amount of waste rock mined will occupy mountain slopes, and there will be serious hidden dangers of geological disasters. figure 1. normal fault figure 2. reverse fault in the case where the geological structure of the mining area is a fault, the roof near the fault plane is relatively fragile, and it is easy to delaminate and fall off, and cause a safety accident. at the same time, more solid waste is produced, which affects water resources and the atmospheric environment. in addition to faults, in some limestone formations, karst is formed due to long-term dissolution of flowing groundwater. under the action of gravity in the geological structure and the upper overlying rock layer, some karst caves have collapsed. because this collapse is a round or irregular oval column, it is called the collapse column (rosińska et al., 2019). the characteristics of the collapse column are shown in the figure below. the feature of the collapse column shown in the figure is that the contact surface between the collapse column and the surrounding normal rock formation is jagged. it is mainly composed of slumped rock blocks. the filling rock has the characteristics of large and small size, obvious edges and corners, different shapes, and mixed accumulation (buttler & wilson, 2018). if the collapse column is damaged, mining cannot be performed normally, which limits the effective use of machinery and seriously affects the safety of mining. the fold structure in the geological structure is divided into anticline and syncline. the geological characteristics of the two are shown below. in actual mining projects, most choose anticline. the anticline forms a valley, and the anticline is a mountain. it is easier to mine at the anticline, and the anticline is relatively loose in geology, and the topsoil layer is easy to be stripped. during the mining process, a large amount of forest vegetation was figure 3. the collapsed column in the complex formation figure 4. geological characteristics of anticlines and synclines 432 yanyan huang the transmission process of the impact of mining on the ecological environment underground resource exploitation causes changes in the stress of the original rock, leading to breakage and movement of the rock formation, further causing ground surface settlement and deformation. caving zones, fissure zones, and curved subsidence zones are usually formed above the mined-out area, which can be simplified into the process of rock breaking and surface settlement (yuan et al., 2019). due to the mining strength and the particularity of the overlying rock properties, thickness, and structure, the above process has its special performance. when the mining depth is small, there is only a fall zone and a fissure zone above the goaf, which shows that the ground surface is discontinuously deformed, and the surface cracks are directly connected to the goaf. in this case, due to different roof conditions and mining heights, and different speeds, roof cutting and sand breaking accidents will occur. the topcut accident is mainly determined by the thickness ratio of the bedrock to the load layer (referred to as the base-load ratio). relevant research shows that when the base-load ratio is less than 0.8, the roof sinks along with the coal wall steps. if the amount of step-like subsidence is large enough, a top-cut accident will occur. sand breaking accidents usually occur when the ratio of depth to thickness is less than 25. in the case of large mining depth, it still conforms to the formation law of the “three zones”, but if the mining depth is not large enough, the downward fractures in the curved zone and the upward fractures in the goaf may still be conductive (yaegashi et al., 2018). accompanied by the rock breaking process and the surface subsidence process is the process of surface animal and plant habitat change and landuse change, and the succession process of the mining area ecosystem at any time. the changes in the habitats of animals and plants after mining resources are manifested by changes in topographic conditions, changes in groundwater table depth, changes in hydrological conditions such as surface sinks, changes in site conditions such as soil, and changes in suitable biological populations. land-use change is a comprehensive reflection of changes in the surface environment. changes in land use structure have further caused changes in regional ecosystems (wu et al., 2018). ecosystem succession has a long-term nature. it is not only affected by natural factors, but also by many human activities, including resource extraction activities and ecological restoration activities after extraction. it is relatively simple to describe the changes of thesurface animal and plant habitats and the changes in land use. as long as the changes in the indicators and spatial distribution of relevant factors are listed, they can be clearly displayed. the description of ecosystem succession is relatively complicated. it involves the succession of ecosystem structure and function, which is a long-term and gradual process. the specific process is shown in the following. one of the salient features of surface damage caused by mining is the development of ground fissures. the treatment of ground fissures is not only the focus of ecological restoration but also an important factor affecting the safe production of resources and the production and life of local residents. according to the development period, there are two types of temporary cracks during mining and permanent cracks after surface stabilization. common ground cracks are shown in the figure below. groundwater is closely related to plant growth and conservation of fragile ecological environments. large-scale resource mining activities not only cause the ground surface to collapse and change the shape of the surface but also have a greater impact on the region’s own poor groundwater resources. the impact mode of underground mining on groundwater can be summarized as follows: in the preparation stage of mining, groundwater is mainly affected by artificial dredging and drainage measures and has nothing to do with the damage of rock formations, and the decline of groundwater level is extremely significant; in the mining stage, in addition to the effects of drainage measures, groundwater is also affected by changes in the formation of the stratum and the development of water-conducting fissures. groundwater movement is mainly based on vertical seepage to goaf; in the recovery phase, groundwater is mainly affected by the recharge and water use in the adjacent area, and the water level recovery process is slow. at this time, the distribution of diving space is significantly affected by the fluctuation of the bedrock. vegetation is the most important indicator of ecological health. the largescale ground cover has a high correlation with the monthly change in rainfall, and the correlation coefficient between vegetation cover and monthly rainfall can be as high as 0.99. rainfall is still the dominant factor influencing changes in mining vegetation cover. according to the mining subsidence theory, the surface subsidence deformation can be divided into the neutral zone, compression zone, and lazhong zone. due to the spatial heterogeneity of soil water in these three areas, the response changes of vegetation in the deformation area after mining figure 5. the transmission process of the impact of mining on the ecological environment figure 6. surface cracks 433destruction process and restoration countermeasures of the ecological environment of a comprehensive geological structure subsidence are also spatially different (robertson et al., 2018). compared with the compressed area, lazhong area, and neutral area, the vegetation is easier to improve, and even the vegetation is better than the vegetation cover before mining subsidence due to moderate disturbance. the reason is that the neutral zone is often located in the central area of the surface subsidence basin, and its lower terrain is beneficial to the collection of surface water, and the soil water content increases. the vegetation in the lazhong area is more susceptible to the negative impact of mining subsidence. increased soil porosity and the appearance of a large number of permanent cracks in the lazhong area will aggravate soil water evaporation and cause damage to plant roots. evaluation of the degree of damage to the ecological environment by mining based on the above analysis content, the quantitative and qualitative theory is used to evaluate the damage degree of mining to the ecological environment, and then corresponding recovery measures and enforcement measures are proposed according to different degrees. there are three main references for assessing the extent of damage to the ecological environment by mining: first, the extent of damage, that is, the scope, intensity, and duration of damage to development and construction projects, and whether the damage is related to the dominant factors of the ecosystem. the second is the sensitive nature of the ecological environment, that is, whether it affects sensitive targets, whether it exacerbates regional natural disasters, and whether the required level of protection is particularly high. the third is the nature of destruction, that is, the reversibility, recoverability, and compensability of destruction, one-time destruction or accumulation, and so on. based on the above, the evaluation index system for the damage caused by mining to the ecological environment is determined, as shown in the following table. table 2. the evaluation index system for the damage caused by mining to the ecological environment level i evaluation indicators level ii evaluation indicators level iii evaluation indicators geological environment background conditions in the mining area geomorphic conditions vegetation distribution land use population density average rainfall geological hazard types impact of mining activities geological disaster occurrences economic losses death toll sphere of influence occupation and destruction of land resources occupied damaged land area goaf area wastewater disposal annual emissions is it toxic solid waste accumulation accumulated stock is it toxic groundwater equilibrium damage drainage area according to the evaluation principles, various destructive factors are comprehensively analyzed and the evaluation indicators are classified. the scores corresponding to different levels are given to form the graded scoring criteria for the evaluation index. the evaluation index grading standard table is as follows: according to the evaluation index system determined above, the analytic hierarchy process and expert scoring method are used to evaluate the damage degree of the ecological environment in the mining area. the specific evaluation methods are as follows: suppose that the rating of the level ii evaluation index is expressed as w1, the evaluation factor score is expressed as ri, and the evaluation factor weight is expressed as i. the calculation formula is: w ri i= ×∑  (1) i i i r r = −1 (2) suppose that the total score of the i-level evaluation indicator is expressed by g, and the indicator weight is expressed by rj. then: g w rj= ×∑ (3) where ri indicates the weighting of the scoring factor of the i-th ii level indicator, and rj indicates the weighting of the scoring factor of the j-th i level indicator. according to the scoring results, the degree of damage to the ecological environment by mining activities is divided into three levels. when g < 3 1111, it was slightly damaged; when 3 ≤ g ≤ 5, it was severely damaged; when g ≤ 5, it was severely damaged. based on the above-mentioned level division, corresponding recovery measures are formulated. results experiment preparation the purpose of the experimental research is to compare the proposed restoration countermeasures with traditional ecological environment restoration countermeasures and to study which restoration countermeasures have low input costs and high economic benefits. considering that the actual ecological environment restoration requires a longer period, the experiment is verified by simulation. the impact of resource mining activities on the ecological environment is large. the direct and indirect values of ecological restoration are combined as a standard indicator for measuring the economic and economic benefits. the direct use value includes the value of forest resource productivity, scientific research and tourism value, and the indirect use value mainly includes the value of conserved water resources, the value of biodiversity and the value of degradation of pollutants. the calculation of the above value is shown below. s n q we ai i i i n = +( ) = ∑1 1 1 (4) l r ui i= × | (5) h n t yi i i n = +( ) = ∑1 1 1  (6) d c vi i= (7) j n mbi i i n = +( ) = ∑1 1 1  (8) 434 yanyan huang table 3. scoring criteria for rating indicators geomorphic conditions grading hilly aeolian landform loess floodplain plain alluvial fan lake score 6 8 8 8 10 10 4 vegetation conditions grading agricultural vegetation natural vegetation no vegetation lake score 10 8 6 4 land use grading arable land woodland grassland bare ground desert swamp water body score 9 8 7 6 2 2 4 population density grading <100 100-300 300-500 500-1000 >1000 score 2 4 6 8 10 average annual rainfall grading <100 100-300 300-700 700-1000 1000-4000 score 10 8 6 4 2 geological hazard types grading landslides, mudslides ground collapse soil erosion snowfall, freeze-thaw medium strong medium strong medium strong medium strong score 5 10 5 10 5 10 5 10 geological disaster score 10 8 6 4 2 frequency ≥10 5-9 3-4 1-2 0 economic losses >100 10-100 1-10 0.1-1 ≤0.1 casualties ≥10 5-9 3-4 1-2 0 sphere of influence >1 0.1-1 0.01-0.1 0-0.01 0 occupation and destruction of land resources score 10 8 6 4 2 occupied land area >1 0.5-1 0.1-0.5 0.01-0.1 ≤0.01 area of occupied land >0.5 0.1-0.5 0.01-0.1 0-0.01 0 waste water disposal score 10 8 6 4 2 annual emissions >1 0.1-1 0.01-0.1 0-0.01 0 solid waste discharge score 10 8 6 4 2 accumulated stock >10 1-10 0.1-1 0.01-0.1 ≤0.01 wastewater toxicity score 10 2 toxicity poisonous non-toxic groundwater equilibrium damage score 10 8 6 4 2 drainage area >1 0.1-1 0.01-0.1 0-0.01 0 435destruction process and restoration countermeasures of the ecological environment of a comprehensive geological structure where, s represents the productivity value of forest resources in the resource extraction area, n is constant, qi represents the area of the plant in i years, wi represents the increase in the unit area of forest resources, ei represents the price of net income per unit area of plant resources such as forests,  represents the price growth rate; l indicates the value of scientific research and tourism, ri indicates the benefits of scientific research and tourism brought by the unit forest, ui represents the forest area for ecological environment restoration; h represents the average annual storage value, ti represents the average price of building a reservoir per square meter, yi indicates the amount of water stored before the restoration of the ecological environment,  represents the price increase factor; d represents the value of biodiversity, i indicates the ecological value of the unit area of the ecological environment, ci indicates the area of ecological environment restoration; jis the value of degrading pollutants, miindicates the treatment cost per unit of pollutant, bi is the amount of pollutants degraded in the surrounding area after the restoration of the ecological environment.  is price growth rate. experimental results and analysis according to the information about the direct use value and indirect use value, the corresponding ecological environment data is simulated. in the simulation experiments, the cost input is gradually increased. the economic benefits obtained by different recovery countermeasures are compared and analyzed. after the ecological environment is destroyed, the economic growth caused by different recovery measures is shown below. observe the experimental results in the figure 7. experimental result 1 is the change of economic benefits after ecological restoration corresponding to the traditional restoration measures based on the environmental protection system. the results show that the economic benefits brought by the restoration of the ecological environment are lower than the input cost, and the difference between the two is at least 100,000 yuan. experimental result 2 is the economic benefit brought by the traditional recovery measures relying on natural recovery. the results show that when the input cost reaches one million yuan, the economic benefit exceeds the input cost, but with the increase of the input cost, the economic benefit increases less and less. when the input cost is about 1.5 million yuan, the input cost is greater than the economic benefit. experimental result 3 is the economic benefits brought by the proposed restoration countermeasures. the results show that with the increase of the input cost, the economic benefits increase significantly. when the investment is about 800,000 yuan, the economic benefit is greater than the investment cost, and it has been in a substantial growth trend. in summary, considering the ecological environment destruction process of the comprehensive geological structure, the proposed ecological environment restoration strategy has the advantages of low cost and high efficiency, and the restoration strategy is superior to the traditional restoration strategy. discussion establish a new resource mining mechanism in order to adapt to the development of the market economy under the new situation, a new type of resource development enterprise management and operation mechanism should be established. based on the current development of mining activities, a unified energy development enterprise joint system is established, and the overall development of regional resources is implemented to promote the formation and development of the overall economy. this is conducive to the industrial mix of energy-saving and emission reduction and circular economy and is more conducive to integrating funds and forces for comprehensive ecological environment improvement and pollution control (asghari et al., 2018). with the massive development of various resources, a large area of land subsided and formed lakes and swamps, which changed the nature of land use, broke the existing urban and rural layout, and caused major changes in the regional economic structure and productivity layout. it has been transformed from a single agricultural economy into a mining-type agricultural economy. therefore, to accelerate the planning and construction of towns in the mining area, the agricultural population will surely shift to towns, industrial and mining (a) experimental results 1 (b) experimental results 2 (c) experimental results 3 figure 7. economic benefits from different recovery strategies 436 yanyan huang (yang et al., 2020). the level of urbanization in the mining area needs to be improved, and the scale of urban construction needs to be expanded. gradually implement integrated and integrated construction to accept the transfer of agricultural population to urban areas and expand employment opportunities, and reduce the pressure on the social environment. the establishment of local central towns should be expanded, and auxiliary enterprises, residential areas, and service industries should be incorporated into local central town construction and planning. relying on the development of local central towns, new residential, and commercial areas should not be built separately (wu et al., 2020). the built-up residential areas should be gradually replaced and rebuilt and relocated to local towns. resettlement in mining areas should be closely integrated with urban planning. based on the principle of relative concentration, rationally arrange resettlement for new villages. according to the standards of the new socialist countryside, an ecological village in the form of a mining area will be set up so that farmers will not leave their homes, which will facilitate the cultivation and management of land. mining areas and cities should set up operating service companies to develop the tertiary industry and serve the resource extraction and urban areas in order to solve the employment problems of residents who are relocated in coal mine development and ensure social stability (cheng & liu, 2020). coordinate energy development and ecological protection with a scientific development concept the government is responsible for the quality of the ecological environment in the area under its jurisdiction, and it is necessary to strengthen the leadership of energy resource development and ecological environmental protection. carefully study the development strategy of energy resources in the area under its jurisdiction, work out an energy development model that is conducive to resource development and environmental protection, and sustainable economic development, and strive to create an environment-friendly energy development economy. the environmental protection executive head accountability system should continue to be seriously implemented, the environmental protection target responsibility system needs to be implemented, the first approval right and veto power of the environmental protection department should be implemented, and the environmental protection work in energy development in its own jurisdiction should be organized and led (gao et al, 2018). the development of energy development and ecological protection plans should be adapted to local conditions. after the work priorities and goals are determined, the outstanding ecological and environmental problems in the area shall be solved in a planned and step-by-step manner. the ecological environment in yulin is fragile and the carrying capacity is limited. the local farmers live in poverty, and the desire to get rid of poverty is more urgent. energy development companies have an inherent need to maximize economic benefits for their own development needs. most of them only care about development, regardless of the carrying capacity of the ecological environment, only the current development, but not sustainable development and they have the idea of development first, then governance. the scientific development concept must be used to coordinate energy development and ecological protection (gao et al., 2019). the grassroots government and corporate legal persons must adhere to the people-oriented principle, establish a comprehensive, coordinated, and sustainable development concept, handle the relationship between resource exploitation and ecological protection, and implement the policy of “developing during protection and protecting during development”. the interests of all parties must be balanced, a dynamic balance of the coordinated development of economic and social development, resource development and utilization, and ecological and environmental protection should be established. a resource-saving and ecologically-friendly mining area should be built to achieve harmony between humans and nature in the mining area. the scientific concept of development and the resource-saving society must be vigorously publicized through various means, and the legal knowledge of environmental protection and resource conservation needs to be vigorously popularized. let everyone know that protecting the environment means protecting their homes, themselves, and their children. the sound public participation mechanism needs to be established as soon as possible in order to implement the public’s right to know the quality of the ecological environment and strengthen supervision of illegal ecological environment enterprises and illegal acts. green gdp accounting system gdp is a measure of the economic performance of all countries and regions, and it reflects the state of wealth and the living standards of the people. however, the ecological environment is part of the comprehensive economy of a country or region. environmental pollution and ecological damage can also increase gdp. without incorporating ecological and environmental factors, the gdp kernel algorithm cannot fully reflect the real economic situation of a country or region. if only from the perspective of regional economic development, the local economy has increased, but it cannot truly reflect the ecological and environmental costs behind the economic growth in the area. therefore, in order to fully reflect the real economic situation, especially for areas such as resource extraction that are eager to pursue rapid economic development and easily overlook the ecological environment, it is necessary to establish a green gdp accounting system. through the establishment of a green gdp accounting system, the increase and consumption of ecological environmental resources are included in the national economic account in the form of currency to reflect the changes in the ecological environment over the years and its internal relationship with economic development. correctly estimating the actual state of economic development and future development potential can provide a basis for establishing an ecological environment compensation mechanism, especially for establishing horizontal fiscal transfer payments. taking the above three points into consideration with the degree of damage to the ecological environment, different levels of damage have different emphases in the implementation of restoration measures, and relevant measures should be implemented according to actual conditions. conclusions when the comprehensive geological structure was mined, the ecological environment was damaged. through the evaluation of the damage degree of the ecological environment, the corresponding restoration countermeasures are proposed, which effectively solves the problems of high cost and low efficiency in the traditional restoration countermeasures. it has certainly help to the destruction and restoration of the ecological environment caused by mining and promotes the sustainable development of the ecological environment. acknowledgement general project of philosophy and social science research of heilongjiang province research on the interactive development mode of new urbanization and ecological industrial cluster in heilongjiang province based on the perspective of ecological civilization (no. 19glb088). references asghari, m., noaparast, m., shafaie, s. z., ghassa, s., & chehreh chelgani, s. (2018). recovery of coal particles from a tailing dam for environmental protection and economical beneficiations. international journal of coal science & technology, 5(2), 253-263. buttler, c. j., & wilson, m. a. (2018). paleoecology of an upper ordovician submarine cave. dwelling bryozoan fauna and its exposed equivalents in northern kentucky, usa. journal of paleontology, 92(4), 568-576. chen, w. j., li, s. y., wang, j., sun, k., & si, y. (2019). metal and metaloxide nanozymes: bioenzymatic characteristics, catalytic mechanism, and eco-environmental applications. nanoscale, 11(34), 15783-15793. cheng, h. y., & liu, y. n. (2020). an improved rsu-based authentication scheme for vanet. journal of internet technology, 21(4). cui k., & wang y. x. (2019). structural styles and origin of the dabashan foreland arcuate belt and basin-mountain system in central china. journal of asian earth sciences, 176, 244-252. dochartaigh, b. é. ó., archer, n. a. l., peskett, l., macdonald, a. m., black, a. r., auton, c. a., merritt, j. e., gooddy, d. c., & bonell, m. (2019). geological structure as a control on floodplain groundwater dynamics. hydrogeology journal, 27(2), 703-716. 437destruction process and restoration countermeasures of the ecological environment of a comprehensive geological structure gao, n. s., cheng, b. z., hou, h., & zhang, r. h. (2018). mesophase pitch based carbon foams as sound absorbers. materials letters, 212, 243246. gao, n. s., guo, x. y., cheng, b. z., zhang, y. n., wei, z. y., & hou, h. (2019). elastic wave modulation in hollow metamaterial beam with acoustic black hole. ieee access, 7, 124141-124146. josephs, l. i., & humphries, a. t. (2018). identifying social factors that undermine support for nature-based coastal management. journal of environmental management, 212, 32-38. jurczak, t., wojtal-frankiewicz, a., kaczkowski, z., oleksinska, z., bednarek, a., & zalewski, m. (2018). restoration of a shady urban pond the pros and cons. journal of environmental management, 217, 919-918. robertson, l. w., weber, r., nakano, t., & johansson, n. (2018). pcbs risk evaluation, environmental protection, and management: 50-year research and counting for elimination by 2028. environmental science and pollution research, 25(17), 16269-16276. rosińska, j., romanowicz-brzozowska, w., kozak, a., & goldyn, r. (2019). zooplankton changes during bottom-up and top-down control due to sustainable restoration in a shallow urban lake. environmental science and pollution research international, 26(19), 19575-19587. song, d. q., che, a. l., zhu, r. j., & ge, x. (2019). natural frequency characteristics of rock masses containing a complex geological structure and their effects on the dynamic stability of slopes. rock mechanics and rock engineering, 52(11), 4457-4473. staccione, a., mysiak, j., ostoich, m., & marcomini, a. (2019). financial liability for environmental damage: insurance market in italy, focus on veneto region experience. environmental science and pollution research international, 26 (25), 25749-25761. sur, k., & chauhan, p. (2019). dynamic trend of land degradation/restoration along indira gandhi canal command area in jaisalmer district, rajasthan, india: a case study. environmental earth sciences, 78(15), 1-11. webster, e., gaudin, a. c. m., pulleman., m., siles, p., & fonte, s. j. (2019). improved pastures support early indicators of soil restoration in lowinput agroecosystems of nicaragua. environmental management, 64 (2), 201-212. wu, y. x., lyu, h. m., shen, j. s., & arulrajah, a. (2018). geological and hydrogeological environment in tianjin with potential geohazards and groundwater control during excavation. environmental earth sciences, 77(10), 1-17. wu, z., liu, y. n., & jia, x. x. (2020). a novel hierarchical secret image sharing scheme with multi-group joint management. mathematics, 8(3), 12. yaegashi, y., yoshioka, h., unami, k., & fujihara, m. (2018). a singular stochastic control model for sustainable population management of the fish-eating waterfowl phalacrocorax carbo. journal of environmental management, 219, 18-27. yang, f., yang, f., wang, g. y., kong, t., wang, h., & zhang, c. s. (2020). effects of water temperature on tissue depletion of florfenicol and its metabolite florfenicol amine in crucian carp (carassius auratus gibelio) following multiple oral doses. aquaculture, 515, 9. yuan, j. q., chen, w. z., tan, x. j., yang, d., & wang, s. (2019). countermeasures of water and mud inrush disaster in completely weathered granite tunnels: a case study. environmental earth sciences, 78(18), 1-16. keywords: low frequency; rock petrophysics; lithology; coalbed methane; enrichment. palabras clave: baja frecuencia; petrofísica de roca; litología; carbón metano; enriquecimiento. how to cite item he, y., wang, x., sun, h., xing, z., chong, s., xu, d., & feng, f. (2019). coal seam roof: lithology and influence on the enrichment of coalbed methane. earth sciences research journal, 23(4), 359-364. doi: https://doi. org/10.15446/esrj.v23n4.84394 to identify the lithology of coal seam roof and explore the influence of these roofs on the enrichment of coalbed methane, low-frequency rock petrophysics experiments, seismic analyses and gas-bearing trend analyses were performed. the results show that the sound wave propagation speed in rock at seismic frequencies was lower than that at ultrasound frequencies. additionally, the p-wave velocities of gritstone, fine sandstone, argillaceous siltstone and mudstone were 1,651 m/s, 2,840 m/s, 3,191 m/s and 4,214 m/s, respectively. the surface properties of the coal seam roofs were extracted through 3d seismic wave impedance inversion. the theoretical p-wave impedance was calculated after the tested p-wave velocity was determined. by matching the theoretical p-wave impedance of the four types of rocks with that of the coal seam roofs, we identified the lithology of the roofs. by analyzing known borehole data, we found that the identified lithology was consistent with that revealed by the data. by comparing and analyzing the coal seam roof lithology and the gas-bearing trends in the study area, we discovered that the coal seam roof lithology was related to the enrichment of coalbed methane. in the study area, areas with high gas contents mainly coincided with roof zones composed of mudstone and argillaceous siltstone, and those with low gas contents were mainly associated with fine sandstone roof areas. thus, highly compact areas of coal seam roof are favorable for the formation and preservation of coalbed methane. abstract coal seam roof: lithology and influence on the enrichment of coalbed methane cimas con gas de filones de carbón: litología e influencia en el enriquecimiento de metano en capas de carbón issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.84394 para identificar la litología de una cima con gas de filones de carbón y explorar la influencia de estas cimas en el enriquecimiento de metano en capas de carbón, se realizaron experimentos de petrofísica en rocas de baja frecuencia, análisis sísmicos y análisis de tendencias con gas. los resultados muestran que la velocidad de propagación de la onda de sonido en roca a frecuencias sísmicas fue menor que la de las frecuencias de ultrasonido. además, las velocidades de la onda p de arenisca, arenisca fina, limolita arcillosa y lutita fueron 1.651 m/s, 2.840 m/s, 3.191 m/s y 4.214 m/s, respectivamente. las propiedades de la superficie de los techos con gas de filones de carbón se extrajeron mediante inversión de impedancia de onda sísmica 3d. la impedancia teórica de la onda p se calculó después de determinar la velocidad de la onda p probada. al hacer coincidir la impedancia teórica de la onda p de los cuatro tipos de rocas con la de las cimas con gas de filones de carbón, se identificó la litología de las cimas. al analizar los datos conocidos del pozo, se encontró que la litología identificada era consistente con la revelada por los datos. al comparar y analizar la litología de la cima con gas de filones de carbón y las tendencias con gas en el área de estudio, descubrimos que la litología de esta cima estaba relacionada con el enriquecimiento del metano de las capas de carbón. en el área de estudio, las áreas con alto contenido de gas coincidieron principalmente con zonas de techo compuestas de lutita y limolita arcillosa, y aquellas con bajo contenido de gas se asociaron principalmente con áreas de techo de arenisca fina. por lo tanto, las áreas altamente compactas del techo con vetas de carbón son favorables para la formación y preservación del metano de las capas de carbón. resumen record manuscript received: 27/04/2019 accepted for publication: 11/07/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 359-364 yunlan he1, 2*, xikai wang3, hongjie sun3, zhenguo xing3, shan chong1, dongjing xu2, feisheng feng3 1state key laboratory of coal resources and safety mining, china university of mining & technology, beijing 100083, china 2shandong provincial key laboratory of depositional mineralization & sedimentary minerals, college of earth science & engineering, shandong university of science and technology, qingdao 266590, china 3college of geoscience and surveying engineering, china university of mining & technology, beijing 100083, china * corresponding author: 55069008@qq.com h y d r o c a r bo n s 360 yunlan he, xikai wang, hongjie sun, zhenguo xing, shan chong, dongjing xu, feisheng feng introduction coalbed methane exists in coal rock, and coalbed methane enrichment remains a difficult research topic. in addition to the coal seam, which influences methane enrichment (groshon, pashin, & mcintyre, 2009; kędzior, 2009; hemza, sivek, & jirásek, 2009; gao et al., 2012; moore, 2012), the coal seam roof also affects the enrichment of coalbed methane. thus, the role of the roof has attracted increased research attention (chen, cui, liu, & lang., 2006; zhou, 2013; song et al., 2013). the properties of coal seam roofs are difficult to determine based on sparse borehole data; however, 3d seismic data collected at low frequencies (5-100 hz) can greatly improve the density of control points and identification accuracy. the theories regarding the propagation velocity of sound waves in rock at seismic frequencies have not been supported by experimental data, although some have described the velocity and attenuation of sound waves in fluid-saturated rock at seismic frequencies (ba, 2010; white, 1975; biot, 1956). these theoretical models are mainly based on experimental data at ultrasonic frequencies (mhz). tutuncu, gregory, sharma, and podio (1998) measured the p-wave and s-wave velocities of a saline-saturated sandstone at frequencies of 10 hz-1 mhz, and the results showed that both the p-wave and s-wave velocities displayed upward-increasing trends. batzle, han, and hofmann (2006) tested and compared the propagation velocity of sound waves at seismic and ultrasonic frequencies using a low-frequency stress-strain method and an ultrasonic test. wang, zhao, harris, and quan (2012) measured the bulk modulus of different types of rocks at 600 hz using resonance spectroscopy and compared the results with those at ultrasonic frequencies. using a low-frequency stress-strain method, wei, wang, zhao, tang, and deng (2015a; 2015b) discussed the factors that influence the propagation velocity of sound waves in sandstone and studied the propagation velocity of sound waves and their dispersion characteristics under different conditions. in this paper, the propagation velocity of sound waves in four types of rocks was measured and studied using a low-frequency stress-strain method. by matching the tested velocity with 3d seismic data, we identified the coal seam roof lithology to explore the influence of the roofs on the enrichment of coalbed methane. we hope that this study will improve studies of the propagation velocity of sound waves at low frequencies and the enrichment trends of coalbed methane (hungerford, 2013; li & wu, 2016; liu, 2018). research methods low-frequency rock petrophysics experiments the samples used in the low-frequency rock petrophysics experiments included four representative types of rocks: gritstone, fine sandstone, argillaceous siltstone, and mudstone. the rocks were collected from the top of the #3 coal seam of the permian shanxi formation at the sihe coal mine in the southern qinshui basin, china. the gritstone was coarse-grained quartz sandstone with no primary pore development, some micro-fissures, and a porosity of less than 3%. the fine sandstone was fine quartz sandstone with primary pore development and a porosity of approximately 20%. the argillaceous siltstone was composed of fine particles and miscellaneous bases with poor separation, no primary pore development, and porosity of almost 0%. the mudstone was collected from the immediate coal seam roof, was composed of argillaceous rocks, and had a porosity of 0%. in this experiment, the physical system used to test the rock samples at low frequencies (fig. 1) was developed by the colorado school of mines, and a low-frequency stress-strain method was used as the experimental method. the tested rock samples were cylindrical, with a diameter of 38 mm and a length of 50-80 mm. each sample had cylindrical aluminum block bases and pand s-wave probes at both ends. additionally, longitudinal and transverse strain gauges were fixed on the aluminum block bases and rock samples, respectively. the physical testing system can produce acoustic waves with frequencies of 3-2,000 hz, and the vertical and transverse strains can be tested with fixed strain gauges on the aluminum block and rock samples. the young’s modulus of the aluminum block was 69.8 gpa. each rock sample and the aluminum block bases comprised a complete unit, so their stresses were the same during the test. notably, the stress on the aluminum block al( ) was equal to that on the rock sample s( ) (formula 1). σ =σal s (1) the stress on the aluminum block al( ) was as follows. σ ε −al al al ve= ⋅ (2) the young’s modulus of the rock sample (es) can be deduced from formulas 1 and 2. e e gpa s s s v al al v s v al v s v = = ⋅ = ⋅σ ε ε ε ε ε− − − − − 69 8. (3) the poisson ratio of a rock sample (v) can be obtained by formula 4. v s v s h = ε ε − − (4) because the young’s modulus and poisson ratio of a rock sample were given, the other parameters of the rock sample at different frequencies and conditions could be obtained by the formulas listed in table 1. figure 1. physical testing system of the rock at low frequencies 361coal seam roof: lithology and influence on the enrichment of coalbed methane results results of the low-frequency experiments the propagation velocity of sound waves in the four types of rocks was measured at 16 frequencies from 3-100 hz. the p-wave and s-wave velocities of the rocks at the 16 frequencies are shown in fig. 2. fig. 2a shows the relationship between the p-wave velocity and frequency. the p-wave velocity of the gritstone varies little at different frequencies, with an average of 1,651 m/s. the p-wave velocities of the fine sandstone, argillaceous siltstone and mudstone exhibited little variability, with average values of 2,840 m/s, 3,191 m/s and 4,214 m/s, respectively. fig. 2b shows the relationship between the s-wave velocity and frequency, and the resulting trends are similar to those for the p-wave velocity, but with smaller variations and average values of 1,233 m/s, 1,492 m/s, 1,778 m/s and 3,514 m/s for the four materials. by comparing the propagation velocity of sound waves in the four types of rocks, we found that both the p-wave and s-wave velocities successively increased at different frequencies. wave impedance properties of the coal seam roof because wave impedance inversion is important for identifying rock properties, it is a common method used to identify sandstone and mudstone based on the sound wave velocity and density log curves. figure 3 shows the p-wave impedance surface properties of the #3 coal seam roof in the study area. the depth of the roof was 10 m higher than the bottom boundary of the #3 coal seam, and the p-wave impedance varies between the roof and the boundary. according to the characteristics of the p-wave impedance of the roof, the impedance was divided into three types: yellow-red p-wave impedance (6,400 m/s·g/cc to 7,800 m/s·g/cc), blue p-wave impedance (7,800 m/s·g/cc-9,400 m/s·g/cc) and purple p-wave impedance (> 9,400 m/s·g/cc). the p-wave impedance, with a range of approximately 6,400 m/s·g/cc-10,000 m/s·g/cc, greatly differed and displayed obvious partitioning and zoning. gas content of the study area the gas properties in the study area were analyzed with a kriging interpolation method based on the gas contents of the boreholes in the #3 coal seam (fig. 4). the overall gas content in the area was high but considerably table 1. formulas for calculating the elastic parameters of the rock samples elastic parameter formula shear modulus g e v = ( )2 1 bulk modulus k e v = ( )3 1 2 longitudinal wave velocity v k g p = + ρ 4 3 transverse wave velocity v g s =  3d seismic data and gas content of the study area the 3d seismic data had a cmp mesh density of 5 m×10 m, with 16 stacked layers. first, the data underwent a series of processes, including static correction, amplitude preservation, denoising, deconvolution, and velocity analysis. then, post-stack wave impedance inversion was performed using the processed data. in terms of the stratum of the coal seam roof, the depth of the coal seam roof can be determined by observing the strong reflection interface formed by the coal seam and the surrounding rock. from the post-stack wave impedance inversion data, the p-wave impedance surface properties of the coal seam roof strata were extracted. to identify the coal seam roof lithology, the properties of the coal seam roof were correlated and matched with the product of the propagation velocity of the sound waves and the density of rock measured in the low-frequency experiments. using the gas contents measured in boreholes, the content difference and distribution were analyzed. the study area was 3.57 km long and 0.87 km wide, with an area of approximately 3.10 km2. the gas contents were collected from 20 wells, and measurements were obtained from cores via a direct analytic method for the #3 coal seam. when analyzing the gas content, a content distribution map was drawn using the kriging interpolation method according to the gas contents of the boreholes. figure 2. relationships between the propagation velocity of sound waves in the four types of rocks and the frequency 362 yunlan he, xikai wang, hongjie sun, zhenguo xing, shan chong, dongjing xu, feisheng feng varied. the study area (3.10 km2) exhibited a gas content ranging from approximately 19 m3/t to nearly 29 m3/t, a difference of nearly 10 m3/t. three sub-areas with gas contents greater than 27 m3/t were observed and are colored in red in figure 5. four sub-areas with gas contents lower than 20 m3/t were observed and are colored in yellow in figure 5. discussion wave impedance inversion can be used to analyze the properties of coal seam roof strata. wave impedance is the product of the propagation velocity of a sound wave in rock and the density of the rock, and it can be used to identify the lithology. however, sandstone and mudstone cannot be distinguished simply by wave impedance and can be influenced by many factors. moreover, the logging velocity was obtained at high frequencies in the process of wave impedance inversion. after well logging calibration was performed for the velocity and density log curves and 3d seismic data, wave impedance inversion was performed for data interpolation. low-frequency experiments were performed to obtain the propagation velocity of sound waves in four types of rock, and the velocity was used to calculate the theoretical wave impedance. the results did not match the wave impedance values obtained by 3d seismic inversion at higher frequencies. for example, murphy (1984) measured the propagation velocity of sound waves in sandstone at different frequencies using the resonance bar technique, and the results showed that the propagation velocity of sound waves in sandstone at 500 khz was 10-25% higher than that at 5 khz. tutuncu et al. (1998) measured the propagation velocity of sound waves in saline-saturated sandstone, and the p-wave velocity and s-wave velocity increased by 33% and 20%, respectively, as the frequency increased from 10 hz to 1 mhz. batzle et al. (2006) measured the p-wave velocity of high porosity and high permeability sandstone (with a porosity of 35% and a permeability of 8.7×10-12 m2) using a stress-strain method and ultrasonic measurement technique. the p-wave velocity of the 90% saturated sandstone at 0.8 mhz was 32% higher than that at 5-10 hz (batzle et al., 2006). these experiments confirmed that the propagation velocity of sound waves at seismic frequencies is lower than that at ultrasonic frequencies. according to the collected borehole data, the coal seam roof was mainly composed of gritstone, medium sandstone, fine sandstone, sandy mudstone, siltstone, and mudstone. in this study, the propagation velocity of sound waves was monitored at both low and ultrasonic frequencies (1 mhz). the p-wave velocity at low frequencies was lower than that at ultrasonic frequencies, but the p-wave velocities of rocks with different lithologies considerably differed. to better match the theoretical wave impedance calculated according to the experimental results, the p-wave velocity was increased by 20% to compensate for the difference in velocity due to the difference in frequency. the specific data are shown in table 2. figure 3. p-wave impedance properties of the #3 coal seam roof figure 4. gas content distribution in the study area table 2. the p-wave velocity and p-wave impedance of the rocks lithology density (g/cm3) vp-lf (m/s) vp-1mhz (m/s) vp-lf×(1+20%) (m/s) vp-lf×density (m/s·g/cc) (1.2×vp-lf)×density (m/s·g/cc) gritstone 2.30 1651 1805 1981 3797 4557 fine sandstone 1.95 2840 3854 3408 5538 6646 argillaceous siltstone 2.41 3191 3588 3829 7690 9228 mudstone 2.20 4214 5070 5057 9270 11125 *vp-lf is the vp value of the low-frequency condition, and vp-1mhz is the vp value of the 1 mhz frequency condition. 363coal seam roof: lithology and influence on the enrichment of coalbed methane the difference in wave impedance for different lithologies was essential for matching the theoretical wave impedance with the wave impedance of 3d seismic inversion. the p-wave velocity of the gritstone was low, possibly due to the existence of micro-fissures. the p-wave impedance did not match that based on 3d seismic inversion, and the weighted p-wave impedance of the gritstone was 4,557 m/s·g/cc. the p-wave impedance of fine sandstone was 6,646 m/s·g/cc, which fell within the scope of the yellow-red wave impedance (6,400 m/s·g/ cc-7,800 m/s·g/cc) when matching the wave impedance obtained by 3d seismic inversion. the p-wave impedance of argillaceous siltstone was 9,228 m/s·g/ cc, which is within the scope of blue wave impedance (7,800 m/s·g/cc-9,400 m/s*g/cc). additionally, the p-wave impedance of the mudstone exceeded 10,000 m/s·g/cc, which is within the scope of purple p-wave impedance (> 9,400 m/s·g/cc). the lithology of the identified coal seam roof was tested according to the known borehole data. the depth of the coal seam roof was 10 m higher than the bottom boundary of the coal seam, and 2000 m/s was selected as the average propagation velocity of the sound wave. fig. 5 shows the coal seam of the selected test well and the top and bottom lithology. the identified coal seam roof lithology was consistent with that revealed by the borehole data. in terms of lithological identification, sandstone could not be distinguished from fine sandstone, and sandy mudstone could not be distinguished from mudstone. however, the fine sandstone, siltstone and mudstone could be distinguished. in addition to coal seams, coal seam roof also influence the enrichment of coalbed methane. during the diagenetic evolution of coal, a ton of coal, from lignite to anthracite, can produce 268-393 m3 of gas (zhang & li, 1988), but only a fraction of this coalbed methane is eventually preserved. the coal seam roof directly contacts the coal seam and undergoes diagenetic evolution together with the coal seam, and the compactness of the coal seam roof has a direct influence on the residual coalbed methane content (li et al., 2003; liu, li, yang, liu, & yan, 2012). by comparing and analyzing the coal seam roof lithology and gas content, we found that the coal seam roof lithology is related to the enrichment of coalbed methane. the coal seam roof in the three areas with high gas contents was mainly mudstone and argillaceous siltstone, and the coal seam roof in the four areas with low gas contents was mainly fine sandstone. the mudstone and argillaceous siltstone, which are highly compact, were more conducive to the preservation of coalbed methane, and the fine sandstone, which is poorly compact, was not. conclusion in this paper, the variable propagation velocity of sound waves in four types of rock was measured at 3-100 hz using a low-frequency stress-strain method. the p-wave and s-wave velocities of the gritstone, fine sandstone, argillaceous siltstone, and mudstone successively increased at the same frequency. the propagation velocity of sound waves in the four types of rock at seismic frequencies was lower than that at higher frequencies. after the p-wave velocities obtained via testing were considered, the theoretical p-wave impedance of the four types of rocks was calculated, and the values were then matched with the wave impedance values of seismic inversion for the coal seam roof. the roof was mainly composed of mudstone, argillaceous siltstone, and fine sandstone, and the boreholes verified that the lithology. the lithology of the roof is related to the enrichment of coalbed methane. the roof in areas with high gas contents was mainly mudstone and argillaceous siltstone, and that in areas with low gas contents was mainly fine sandstone. compact areas of the coal seam roof were more conducive to the formation and preservation of coalbed methane. acknowledgments the project was financially supported by the subject of open fund of state key laboratory of coal resources and safe mining, china university of mining & technology, beijing (sklcrsm19zz01), the fundamental research funds for the central universities (2019qmo1) and the national natural science foundation of china (41602168). references ba, j. (2010). wave propagation theory in double-porosity medium and experimental analysis on seismic responses. scientia sinica (physica, mechanica & astronomica), 40(11), 1398-1409. batzle, m. l., han, d. h., & hofmann, r. (2006). fluid mobility and frequencydependent seismic velocity-direct measurements. geophysics, 71(1), n1-n9. biot, m. a. (1956). theory of propagation of elastic waves in a fluid-saturated porous solid. i. low-frequency range. the journal of the acoustical society of america, 28(2), 168-178. chen, t. j., cui, r. f., liu, e. r., & lang, y. q. (2006). prediction of coal seam methane enriched areas using seismic data. journal of china university of mining & technology (english edition), 16(4), 421-424. gao, l. j., tang, d. z., xu, h., meng, s. z., zhang, w. z., meng, y. j., & wang, j. j. (2012). geologically controlling factors on coal bed methane (cbm) productivity in liulin. journal of coal science and engineering (china), 18(4), 362-367. figure 5. the lithological column sections from the known wells 364 yunlan he, xikai wang, hongjie sun, zhenguo xing, shan chong, dongjing xu, feisheng feng groshong jr, r. h., pashin, j. c., & mcintyre, m. r. (2009). structural controls on fractured coal reservoirs in the southern appalachian black warrior foreland basin. journal of structural geology, 31, 874-886. hemza, p., sivek, m., & jirásek, j. (2009). factors influencing the methane content of coal beds of the czech part of the upper silesian coal basin, czech republic. international journal of coal geology, 79, 29-39. hungerford, f. (2013). in-seam directional drilling for improving safety and productivity in underground coal mines. journal of mines, metals and fuels, 61(7-8), 213-218. kędzior, s. (2009). accumulation of coal-bed methane in the south-west part of the upper silesian coal basin (southern poland). international journal of coal geology, 80, 20-34. li, t., & wu, c. (2016). influence of pressure, coal structure and permeability on cbm well productivity at various burial depths in south yanchuan block, china. earth sciences research journal, 20(3), d1-d9. doi: https://doi.org/10.15446/esrj.v20n3.58000 li, w. z., wang, y. b., cui, s. h., xian, b. a., chen, c. h., & wang, x. h. (2003). analysis of the generation conditions of coalbed gas reservoir, southern qinshui basin. coal geology & exploration, 31(2), 23-26. liu, y. h., li, m. x., yang, x., liu, c. c., & yan, l. (2012). laws of coalbed methane enrichment and high productivity in the fanzhuang block of the qinshui basin and development practices. natural gas industry, 32(4), 29-32. liu, z. (2018). economic analysis of methanol production from coal/biomass upgrading. energy sources part b-economics planning and policy, 13(1), 66-71. moore, t. a. (2012). coalbed methane: a review. international journal of coal geology, 101, 36-81. murphy, w. f. . (1984). acoustic measures of partial gas saturation in tight sandstones. journal of geophysical research, 89(b13), 11549-11559. song, y., liu, s. b., ju, y. w., hong, f., jiang, l., ma, x. z., & wei, m. m. (2013). coupling between gas content and permeability controlling enrichment zones of high abundance coal bed methane. acta petrolei sinica, 34(3), 417-426. tutuncu, a. n., gregory, a. r., sharma, m. m., & podio, a. l. (1998). nonlinear viscoelastic behavior of sedimentary rocks, part 1: effect of frequency and strain amplitude. geophysics, 63(1), 184-194. wang, s. x., zhao, j. g., harris, j. m., & quan, y. (2012). differential acoustic resonance spectroscopy for the acoustic measurement of small and irregular samples in the low frequency range. journal of geophysical research atmospheres, 117(b6), 6203. wei, x., wang, s. x., zhao, j. g., & deng, j. x. (2015a). laboratory investigation of influence factors on vp and vs in tight sandstone. geophysical prospecting for petroleum, 54(1), 9-16. wei, x., wang, s. x., zhao, j. g., tang, g.y., & deng, j. x. (2015b). laboratory study of velocity of the seismic wave in fluid-saturated sandstones. chinese journal of geophysics, 58(9), 330-3388. white. j. e. (1975). computed seismic speeds and attenuation in rocks with partial gas saturation. geophysics, 40(2), 224-232. zhang, w & li, x. (1988). analysis of the hydrocarbon-producing potential of main coal measures source rocks in china. coal geology & exploration, 16(3), 31-37. zhou, q. q. (2013). geological control law of coalbed methane accumulation of heshun block in qinshui basin. china coalbed methane, 10(2), 12-15. keywords: volcanic fault basins; wenxiang granite; la-icp-ms; isotopic dating; late early jurassic. palabras clave: cuencas de fallas volcánicas; granito wenxiang; la-icp-ms; datación isotópica; jurásico temprano tardío. how to cite item tang, j., & lou, f. (2020). la-icp-ms zircon u-pb dating of early jurassic granite basement in chagan area of the northeast guangdong and its geological implications. earth sciences research journal, 24(2), 133-140. doi: https:// doi.org/10.15446/esrj.v24n2.87364 the rencha basin is one of the typical mesozoic-cenozoic volcanic fault basins in northeastern guangdong. the borehole data indicate that the basement is mainly composed of paleozoic metamorphic rocks and wenxiang granite. at present, the basement granite has been lacking in precise years. then, to discuss its formation age and improve the understanding of the geological evolution of the area, la-icp-ms zircon u-pb isotopic dating of wenxiang granite cores from two boreholes in the chagan area of the basin is carried out. the two harmonic ages are 179+1 ma and 186 ma, respectively. the results show that the basement wenxiang granite in rencha basin was formed in late early jurassic, and it is the product of yanshan's first volcanic activity. this age not only deepens the understanding of the formation and evolution of rencha basin but also provides new chronological data for the lack of magmatic activity in the early jurassic in northeastern guangdong, china. abstract la-icp-ms zircon u-pb dating of early jurassic granite basement in chagan area of the northeast guangdong and its geological implications aplicación del método la-icp-ms y datación u-pb en granitos de basamento de principios del jurásico en el área de chagan, en el noreste de guangdong (china) y sus implicaciones geológicas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87364 la cuenca de rencha es una de las cuencas típicas de fallas volcánicas mesozoico-cenozoicas en el noreste de guangdong, china. los datos del pozo indican que el basamento está compuesto principalmente por rocas metamórficas paleozoicas y por granito wenxiang. en la actualidad, no se conocen datos específicos del granito del basamento. para discutir su edad de formación y mejorar la comprensión de la evolución geológica del área se aplicó el método laicp-ms y datación u-pb a dos muestras de granito wenxiang de dos pozos en el área de la cuenca de chagan. las dos edades armónicas son 179+1 ma y 186 ma, respectivamente. los resultados muestran que el granito wenxiang del basamento en la cuenca rencha se formó a finales del jurásico temprano, y que es el producto de la primera actividad volcánica de yanshan. esta edad no solo profundiza la comprensión de la formación y evolución de la cuenca rencha, sino que también proporciona nuevos datos cronológicos para la falta de actividad magmática en el jurásico temprano en el noreste de guangdong. resumen record manuscript received: 13/11/2019 accepted for publication: 03/02/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 133-140 jinhui tang1, 2*, feng lou2 1school of geophysics and measurement-control technology, east china university of technology, nanchang, 330013, china 2party 292, geological bureau for nuclear industry of guangdong, heyuan, 517001, china * corresponding author: amjiangbo@163.com g e o l o g y https://doi.org/10.15446/esrj.v24n2.87364 https://doi.org/10.15446/esrj.v24n2.87364 https://doi.org/10.15446/esrj.v24n2.87364 mailto:amjiangbo@163.com 134 jinhui tang, feng lou introduction the renzi basin is located at the intersection of the eastern end of the east-west nanling polymetallic metallogenic belt and the northwestern wuyi polymetallic metallogenic belt (fig. 1-b). the caprock is mainly composed of the late cretaceous yetang formation volcanic rocks, followed by the paleogene paleocene and eocene sedimentary rocks. the formation is relatively flat, forming an open syncline basin. according to a large number of field geological surveys and borehole control, the basement of the basin is mainly composed of cambrian, devonian-carboniferous metamorphic rocks and wenxiang granite, namely “double base”. the cap layer is not integrated on the basement, forming a pattern of the upper collapsed basin and the red basin of the brazier. at present, many uranium, molybdenum, lead, zinc, gold, silver and rare earth polymetallic deposits and mineralization have been found in renchao fault basin (liu et al., 2007; xu et al., 2010). according to dating data, the cover volcanic rocks in rencha basin were mainly formed at 90 main late cretaceous. the dashenba rock mass (135 ma) of early cretaceous, guikeng granite of indosinian-yanshanian period and eight-foot gneissic granite (165 +1 ma) were widely exposed on both sides of the basin (liu, et al., 2007). so far, no plutonic rocks corresponding to the caprock volcanic series have been found in the basin. in addition, the cover of the basin is thick and the granite is less exposed. at present, the granites exposed by the borehole are mainly composed of fine-grained zeolitic granites, which are in contact with the overlying paleozoic strata. at present, the igneous rocks of these basements have been lacking in precise dating. the formation of the la-icp-ms zircon u-pb age was studied by geological exploration of the bored area of the rencha basin. survey of research areas renchao basin is located in the caledonian fold system of south china, the late paleozoic depression belt of yongan-meizhou, and the northeastern part of the heyuan deep fault zone. it belongs to pingyuan county, meizhou city, guangdong province. it is about 15 km in length from north to south, 10-15 km in width from east to west, and has an area of about 200 km 2m2. the basin is a typical volcanic faulted basin, which consists of a set of rhyolitic and acidic rock-dominated complex formations. the lower part of the strata is breccia tuff and rhyolite, the middle part is pyroclastic rock and lava tuff, and the upper part is pyroclastic rock (song et al., 2010). the basin is nnetrending and belongs to mesozoic-cenozoic continental volcanic faulted basin (k2-e). the basement of the basin is composed of cambrian, devoniancarboniferous epimetamorphic rocks and early yanshanian wenxiang granites, which mainly distribute in the margin and outside of the basin. the caprock is mainly composed of cretaceous volcanic rocks, followed by late cretaceoustertiary volcanic-clastic sedimentary rocks. the basin structure is dominated by faults, mainly nne, followed by ew and nww. the nne on the west side of the basin to the zhegulong fault and the east side nne to the zhumaba basin is the basin control structure, and the near ew to malou fault constitutes the southern boundary of the basin. several nne-ne-trending secondary faults are also developed in the central part of the basin. in addition, the nwwtrending structures developed in the basin are filled with medium-basic dikes. the granite is mainly distributed in the east and west sides and the margin of the basin. the east side is the early dashenba rock mass in the yanshan mountains, and the west side is the indosin yanshan period guikeng rock mass. the lithology is medium-grained porphyritic biotite granite. in the western part of the basin, there are also medium-fine-grained mica granites and fine-grained biotite granites. in the south, there are sub-flow porphyries with narrow strips and other outputs (li et al., 1998). samples and analytical methods in this work, samples of granite cores from zk1702 (760 m) and zk1401 (610 m) drilling holes in chagan area, northern renchao basin are collected and zircons are separated for la-icp-ms zircon u-pb dating. the sample number collected in zk1702 is b-1 (north latitude 24°53’07′′; east longitude 115°57′33′′), and the sample number in zk1401 is b-2 (north latitude 24°53′26′′; east longitude 115°57′ 07′′). the core of the collected sample is preserved intact, and the rock is fresh and not weathered. the rock hand specimen is fine grained granitic granite (fig. 2-a~b). the sample is patchy, the matrix is microscopic, and the image symbiote is between 0.05 mm and 0.15 mm. there is a local mosaic structure (fig. 2-c~d). the size of automorphic quartz phenocrysts and potassium feldspar (diorite) phenocrysts is mostly 1-2mm, and the maximum is 5mm. firstly, the core samples weighing about 1 kg are cut up and crushed, 80-100 meshes are screened, the rock powder washed out. the obtained heavy sand is electromagnetically sorted, the non-magnetic part is separated by heavy liquid, and the heavy part is taken. finally, zircon crystals are manually selected. the zircon crystal is made into an epoxy sample target and polished. zircon transflective and cathodoluminescence (cl) photography is performed at the laboratory of geological mineralization and process center of sun yat-sen university. the target of zircon sample is observed by optical microscopy and scanning electron microscopy cathodoluminescence (cl). zircon crystals with good crystal shape, no cracks and undeveloped inclusions are selected for determination (figure 3). la-icp-ms zircon u-pb dating is completed online by icp-ms and laser denudation system at the state key laboratory of geological processes and mineral resources, china university of geosciences (wuhan). icp-ms is agilent 7500a produced by agilent company in the united states, and laser denudation system is geolaspro produced in the united states. the laser spot size of sample melted by laser is 18μm. icpmsdatacal software is used for data processing (liu et al., 2010; liu et al., 2008; gao et al., 2020). u-pb isotope dating uses standard zircon 91500 as external standard for isotope fractionation and u content correction. recommended values for u-th-pb isotope ratios for zircon standard 91500 according to wiedenbeck et al. (1995). using standard zircon plesovice as a monitoring sample, the tims measured at 206 pb / 238 u was 337.13 ± 0.37 ma (2σ) (slama et al., 2008). sample u-pb harmonic plotting and weighted average calculations are performed using isoplot/ex_ver3 (ludwig, 2003). normal pb correction is performed using the compbcorr #3.18 program (andersen, 2002). thirty zircons are tested in two samples. a group of figure 1. (a) simplified geological map of rencha volcanic basin area (b) tectonic location of the studying area 135la-icp-ms zircon u-pb dating of early jurassic granite basement in chagan area of the northeast guangdong and its geological implications the two samples is not obvious, considering the th/u ratio of zircon, the zircon selected in this test has typical magmatic zircon characteristics, and the u-pb age of the main group zircon can represent the age of magmatic crystallization. a total of 30 data of 30 particles were obtained from sample b-1. the weighted average age of 206pb/238u is186+1ma (mswd=0.084). one of the analysis points (point 9 has a harmonicity of 7%) has a low degree of harmonicity and is eliminated when recalculating the age. the remaining 28 analysis points have an age of 206pb/238u between 184ma and 187ma. the weighted mean age of the remaining 29 analysis points is calculated by isoplot software to be 186. ma (mswd = 0.082), which represents the crystalline age of b-1 fine grained granite (figures 4 and 5). a total of 2830 data are measured for sample b-2 (points 6 and 7 and points 11 and 12 belong to the same zircon). in the analysis point, the age of point 3 is 519 ma, which may be the early magmatic zircon captured in the process of rock formation. the ages of point 9 and 21 deviate from the mode obviously. therefore, the three analysis points do not participate in the calculation of weighted age. the weighted average value of the remaining 27 206pb/238u ages obtained by isoplot software is 179+1ma (mswd=0.18), which represents the crystallization age of fine grained granite from b-2 samples. the above la-icp-ms zircon u-pb dating data show that the mswd values of the basement fine-grained hieroglyphic granites in renchao basin are small and reliable. the two diagenetic ages are 179±1 ma and 186 ma, samples are determined for 60 seconds, in which the first 20 seconds turned off the laser and the last 40 seconds turned on the laser. because apatite contains more common lead, if the content of la or p is higher at the analytic point, it indicates that the laser melting point contains apatite inclusions, these analytic data points should be excluded (liang et al., 2006; harris et al., 2004). u-pb dating of zircon the results of la-icp-ms zircon u-pb isotope dating parameters of fine grained wenxiang granite in rencha basin are shown in tables 1 to 10, respectively. the results of optical microscopy show that the zircon grains in the samples are small, mostly self-shaped or semi-self-shaped columnar, with good crystal shape, light yellow, light rose or colorless. the polymorphism is composed of cylindrical (110) (100) and conical (111) (131). the lengthwise direction of the zircon grains is 80-200μm, and the aspect ratio is between 1 and 3. the th/u ratio of zircon can indicate metamorphism or magmatic origin to some extent. the th/u ratio of metamorphic zircon is usually less than 0.1, and that of magmatic zircon is generally greater than 0.4. th/u ratios at all points measured are greater than 0.4, averaging 0.76 (b-1) and 0.69 (b-2), with obvious magmatic zircon characteristics. although a small part of zircon oscillating zoning or fan-shaped zoning of fine-grained wenxiang granite in (a) (b) (c) (d) figure 2. (a) sample hand specimen and micrograph in chagan area. a, b-fine grain granitic granite hand specimen; c, d-text image granite microphotograph (orthogonal polarized light) 136 jinhui tang, feng lou table 1. la-icp-ms zircon (b-1) uth-pb dating results (01-06) point number 01 02 03 04 05 06 pb ppm 8.6 42.3 21.1 17.1 30.3 22.3 th ppm 119 962 323 312 545 600 u ppm 192 856 525 433 730 523 th/u 0.61 1.12 0.61 0.71 0.74 1.14 207pb / 206pb ratio 0.074 0.0661 0.0526 0.0524 0.0539 0.049 δ 0.004 0.0022 0.0014 0.0018 0.0014 0.0013 207pb/ 235u ratio 0.2905 0.276 0.2122 0.2131 0.2197 0.1996 δ 0.0125 0.0104 0.0055 0.0075 0.006 0.0052 206pb/ 238u ratio 0.0295 0.0294 0.0294 0.0293 0.0294 0.0295 δ 0.0006 0.0003 0.0003 0.0004 0.0003 0.0003 208pb/232th ratio 0.0115 0.009 0.0101 0.0082 0.01 0.0086 δ 0.0004 0.0001 0.0002 0.0002 0.0002 0.0002 207pb /235u age(ma) 258.92 247.5 195.38 196.14 201.67 184.82 δ 9.79 8.28 4.56 6.29 5.03 4.38 206pb/ 238u age(ma) 187.52 186.83 186.74 186.37 186.5 187.29 δ 3.46 2.12 1.92 2.25 1.98 2.17 208pb /232th age (ma) 232 180 203 165 201 174 δ 8 3 4 4 4 5 harmonious degree % 68 72 95 94 92 98 figure 3. cl images of analyzed zircon from graphic granite (b1, b2) basement in chagan area (the figures in the figure represent the point number and zircon age, which correspond to tables 1 and 2) table 2. la-icp-ms zircon (b-1) uth-pb dating results (07-11) point number 07 08 09 10 11 12 pb ppm 9.9 7.7 51.6 48.3 20.9 14 th ppm 162 125 938 897 322 217 u ppm 217 181 848 1096 526 335 th/u 0.74 0.69 1.1 0.81 0.61 0.64 207pb / 206pb ratio 0.0692 0.0639 0.1419 0.0642 0.0582 0.0596 δ 0.0027 0.0025 0.0111 0.0014 0.0016 0.0022 207pb/ 235u ratio 0.2722 0.2518 0.6446 0.263 0.2361 0.241 δ 0.0094 0.0091 0.0655 0.0061 0.0067 0.0089 206pb/ 238u ratio 0.0293 0.0294 0.0291 0.0294 0.0294 0.0293 δ 0.0005 0.0005 0.0009 0.0003 0.0004 0.0004 208pb/232th ratio 0.0109 0.0106 0.0155 0.0103 0.0102 0.0106 δ 0.0003 0.0003 0.0015 0.0002 0.0002 0.0003 207pb /235u age(ma) 244.44 228.04 505.13 237.09 215.19 219.22 δ 7.49 7.38 40.48 4.91 5.53 7.3 206pb/ 238u age(ma) 186.03 186.91 184.98 186.51 186.6 186.36 δ 3.4 3.09 5.5 1.99 2.47 2.55 208pb /232th age (ma) 220 214 310 207 206 213 δ 7 7 30 4 5 6 harmonious degree % 72 80 7 76 85 83 137la-icp-ms zircon u-pb dating of early jurassic granite basement in chagan area of the northeast guangdong and its geological implications table 3. la-icp-ms zircon (b-1) uth-pb dating results (13-18) point number 13 14 15 16 17 18 pb ppm 24.5 15.4 9.1 21.1 28.2 30.5 th ppm 379 212 155 392 493 539 u ppm 601 285 215 471 665 725 th/u 0.63 0.74 0.72 0.83 0.74 0.74 207pb / 206pb ratio 0.0537 0.097 0.0576 0.0553 0.0504 0.0511 δ 0.0012 0.0073 0.0019 0.0016 0.0011 0.0013 207pb/ 235u ratio 0.2158 0.4463 0.2293 0.2248 0.2047 0.2059 δ 0.0048 0.0414 0.0073 0.0068 0.0045 0.0051 206pb/ 238u ratio 0.0293 0.0295 0.0292 0.0293 0.0293 0.0292 δ 0.0004 0.0006 0.0004 0.0004 0.0003 0.0003 208pb/232th ratio 0.0099 0.0141 0.0092 0.0095 0.0089 0.0097 δ 0.0002 0.0009 0.0003 0.0002 0.0002 0.0002 207pb /235u age(ma) 198.44 374.68 209.66 205.85 189.08 190.06 δ 4.01 29.08 6.01 5.59 3.82 4.27 206pb/ 238u age(ma) 185.9 187.33 185.73 185.98 186.12 185.83 δ 2.35 3.48 2.47 2.23 1.71 2.17 208pb /232th age (ma) 198 283 185 191 180 195 δ 4 17 5 5 3 4 harmonious degree % 93 33 87 89 98 97 table 4. la-icp-ms zircon (b-1) uth-pb dating results (19-24) point number 19 20 21 22 23 24 pb ppm 44.7 12.3 25.1 14.8 10.1 39.6 th ppm 731 183 491 267 180 670 u ppm 1020 294 610 328 228 897 th/u 0.71 0.62 0.8 0.81 0.78 0.74 207pb / 206pb ratio 0.0569 0.0605 0.0608 0.0645 0.0528 0.0615 δ 0.0011 0.0022 0.0019 0.0021 0.0019 0.0015 207pb/ 235u ratio 0.2313 0.2449 0.25 0.2588 0.212 0.256 δ 0.0045 0.0087 0.0094 0.0078 0.0074 0.0085 206pb/ 238u ratio 0.0292 0.0293 0.0294 0.0294 0.0293 0.0295 δ 0.0003 0.0004 0.0005 0.0004 0.0004 0.0003 208pb/232th ratio 0.0104 0.0106 0.0098 0.0096 0.01 0.0103 δ 0.0002 0.0003 0.0003 0.0002 0.0003 0.0002 207pb /235u age(ma) 211.3 222.44 226.58 233.72 195.2 231.4 δ 3.67 7.13 7.66 6.28 6.2 6.88 206pb/ 238u age(ma) 185.73 186.21 186.8 186.53 186.13 187.15 δ 1.81 2.31 2.86 2.39 2.32 2.11 208pb /232th age (ma) 210 212 198 194 200 207 δ 4 6 6 5 6 5 harmonious degree % 87 82 80 77 95 78 table 5. la-icp-ms zircon (b-1) uth-pb dating results (25-30) point number 25 26 27 28 29 30 pb ppm 14.2 12.8 40.5 31.1 28.3 21.5 th ppm 231 223 737 355 456 403 u ppm 289 272 913 595 658 492 th/u 0.8 0.82 0.8 0.59 0.69 0.81 207pb / 206pb ratio 0.0829 0.0664 0.0645 0.1131 0.0653 0.0563 δ 0.004 0.0025 0.003 0.0086 0.0024 0.0015 207pb/ 235u ratio 0.3286 0.2637 0.2743 0.5127 0.2764 0.2277 δ 0.0148 0.0084 0.0158 0.0501 0.0129 0.0055 206pb/ 238u ratio 0.029 0.0295 0.0294 0.0292 0.0293 0.0292 δ 0.0004 0.0004 0.0004 0.0007 0.0004 0.0003 208pb/232th ratio 0.0116 0.0104 0.0096 0.018 0.0109 0.0087 δ 0.0003 0.0002 0.0002 0.0014 0.0003 0.0002 207pb /235u age(ma) 288.49 237.68 246.09 420.24 247.76 208.33 δ 11.32 6.73 12.6 33.67 10.25 4.57 206pb/ 238u age(ma) 184.1 187.3 186.49 185.4 186.09 185.8 δ 2.22 2.66 2.21 4.57 2.58 2.05 208pb /232th age (ma) 234 210 193 360 220 174 δ 7 5 4 28 7 4 harmonious degree % 55 76 72 22 71 88 table 6. la-icp-ms zircon (b-2) u-th-pbdating results (01-06) point number 01 02 03 04 05 06 pb ppm 11.8 20.6 24.7 11.3 13.2 8 th ppm 176 335 148 162 132 137 u ppm 285 429 209 290 219 204 th/u 0.62 0.78 0.71 0.56 0.6 0.67 207pb / 206pb ratio 0.073 0.0834 0.0669 0.0632 0.1616 0.0569 δ 0.0039 0.0049 0.002 0.005 0.0073 0.0025 207pb/ 235u ratio 0.281 0.318 0.77 0.236 0.6465 0.2215 δ 0.0134 0.0175 0.031 0.0113 0.0337 0.0096 206pb/ 238u ratio 0.028 0.028 0.0839 0.0283 0.0281 0.0283 δ 0.0004 0.0003 0.0028 0.0005 0.0005 0.0004 208pb/232th ratio 0.0113 0.0138 0.0279 0.0106 0.0205 0.0092 δ 0.0004 0.003 0.0009 0.0004 0.001 0.0003 207pb /235u age(ma) 251 280.35 579.74 215.11 506.32 203.12 δ 11 13.49 17.78 9.26 20.76 7.96 206pb/ 238u age(ma) 178.21 178.24 519.53 180 178.7 179.61 δ 3 2.06 16.43 3.32 3.31 2.49 208pb /232th age (ma) 226 278 556 213 411 186 δ 8 59 18 8 19 6 harmonious degree % 65 55 89 82 4 87 138 jinhui tang, feng lou table 7. la-icp-ms zircon (b-2) u-th-pbdating results (07-12) point number 07 08 09 10 11 12 pb ppm 8.3 8.1 21.9 13.9 13.4 9.3 th ppm 150 129 434 194 239 130 u ppm 217 202 589 326 348 245 th/u 0.69 0.64 0.73 0.59 0.68 0.53 207pb / 206pb ratio 0.0538 0.0635 0.0664 0.0763 0.055 0.0542 δ 0.0025 0.0035 0.0023 0.0038 0.0029 0.0027 207pb/ 235u ratio 0.2045 0.2477 0.2321 0.3083 0.2149 0.2056 δ 0.0089 0.0134 0.0078 0.0181 0.0109 0.009 206pb/ 238u ratio 0.028 0.0285 0.0252 0.0281 0.0284 0.0284 δ 0.0004 0.0005 0.0003 0.0004 0.0004 0.0006 208pb/232th ratio 0.0086 0.0113 0.009 0.0121 0.0088 0.0107 δ 0.0003 0.0005 0.0003 0.0005 0.0003 0.0004 207pb /235u age(ma) 188.92 224.75 211.94 272.87 197.68 189.82 δ 7.51 10.94 6.44 14.07 9.07 7.6 206pb/ 238u age(ma) 177.91 180.91 160.6 178.87 180.56 180.3 δ 2.66 2.99 1.98 2.6 2.63 3.73 208pb /232th age (ma) 174 226 181 243 178 216 δ 5 11 5 10 5 7 harmonious degree % 93 78 7 58 90 94 table 8. la-icp-ms zircon (b-2) u-th-pbdating results (13-18) point number 13 14 15 16 17 18 pb ppm 11.7 19.4 12.8 9.8 8 15.9 th ppm 179 405 188 170 125 262 u ppm 296 454 298 249 210 411 th/u 0.6 0.89 0.63 0.68 0.59 0.63 207pb / 206pb ratio 0.0578 0.0545 0.0793 0.058 0.0585 0.0577 δ 0.0032 0.0023 0.0026 0.0024 0.0028 0.0028 207pb/ 235u ratio 0.2292 0.2146 0.3069 0.2229 0.223 0.2195 δ 0.0124 0.0103 0.0097 0.0087 0.0091 0.0072 206pb/ 238u ratio 0.0284 0.0283 0.0282 0.0282 0.0282 0.0281 δ 0.0004 0.0004 0.0004 0.0004 0.0004 0.0003 208pb/232th ratio 0.0099 0.0095 0.012 0.0097 0.0097 0.0093 δ 0.0004 0.0002 0.0003 0.0003 0.0003 0.0002 207pb /235u age(ma) 209.55 197.39 271.81 204.31 204.37 201.52 δ 10.22 8.61 7.5 7.22 7.52 5.99 206pb/ 238u age(ma) 180.75 179.74 179.17 179.43 179.39 178.88 δ 2.61 2.3 2.39 2.75 2.64 2.1 208pb /232th age (ma) 199 192 242 195 196 187 δ 8 5 6 6 6 5 harmonious degree % 85 90 58 87 86 88 table 9. la-icp-ms zircon (b-2) u-th-pbdating results (19-24) point number 19 20 21 22 23 24 pb ppm 17.7 9.9 18.8 14 18 13.6 th ppm 294 171 700 195 306 260 u ppm 466 252 745 326 439 336 th/u 0.63 0.67 0.93 0.59 0.69 0.77 207pb / 206pb ratio 0.051 0.0552 0.0873 0.0836 0.079 0.0494 δ 0.0015 0.0021 0.0034 0.0044 0.0037 0.0017 207pb/ 235u ratio 0.1991 0.2114 0.1983 0.3455 0.3045 0.1914 δ 0.0057 0.0069 0.0087 0.0229 0.0146 0.0058 206pb/ 238u ratio 0.0282 0.0282 0.0163 0.0282 0.0283 0.0283 δ 0.0003 0.0004 0.0003 0.0005 0.0006 0.0004 208pb/232th ratio 0.0091 0.0092 0.0058 0.014 0.0131 0.009 δ 0.0002 0.0002 0.0002 0.0008 0.0006 0.0002 207pb /235u age(ma) 184.41 194.74 183.69 301.35 269.93 177.87 δ 4.84 5.81 7.36 17.3 11.33 4.92 206pb/ 238u age(ma) 179.43 179.52 104.45 179.46 179.94 179.83 δ 1.81 2.65 1.98 2.85 3.6 2.24 208pb /232th age (ma) 182 186 117 282 263 181 δ 4 5 4 16 12 5 harmonious degree % 97 91 44 49 59 98 table 10. la-icp-ms zircon (b-2) u-th-pbdating results (25-30) point number 25 26 27 28 29 30 pb ppm 14.9 10.7 17.8 23.8 15.4 11.7 th ppm 273 162 324 436 279 214 u ppm 351 249 439 550 386 279 th/u 0.77 0.65 0.73 0.79 0.72 0.76 207pb / 206pb ratio 0.0635 0.1233 0.0516 0.0635 0.0537 0.0574 δ 0.0025 0.065 0.0017 0.0022 0.0017 0.003 207pb/ 235u ratio 0.2468 0.5182 0.1996 0.2543 0.2048 0.2235 δ 0.0094 0.2958 0.0056 0.0104 0.0061 0.0113 206pb/ 238u ratio 0.0281 0.0282 0.0281 0.0283 0.0277 0.0282 δ 0.0004 0.0005 0.0003 0.0004 0.0003 0.0005 208pb/232th ratio 0.0105 0.0099 0.0091 0.0109 0.0093 0.0095 δ 0.0003 0.0003 0.0002 0.0003 0.0002 0.0003 207pb /235u age(ma) 223.95 423.93 184.78 230.07 189.16 204.8 δ 7.65 200.37 4.75 8.42 5.18 9.34 206pb/ 238u age(ma) 178.88 179.56 178.91 179.72 176.12 179.54 δ 2.43 2.93 1.86 2.67 2.17 2.9 208pb /232th age (ma) 212 199 183 220 187 192 δ 6 6 4 7 5 6 harmonious degree % 77 19 96 75 92 86 139la-icp-ms zircon u-pb dating of early jurassic granite basement in chagan area of the northeast guangdong and its geological implications respectively. the ages are late early jurassic. because of the age difference between magmatism and consolidation diagenesis, the basement granite in renchao basin should be the product of early jurassic magmatism. discussion the early yanshanian in southeastern china is generally considered to be a critical period for the transition from the paleo-tethys tectonic domain to the pacific tectonic domain (shu & zhou, 2002; chen et al., 2002). before the middle jurassic was affected by the tethys tectonic domain and the indosinian collision, the late jurassic has been mainly constrained by the combination of pacific tectonic and intracontinental deep structures. however, during the transition period of the two tectonic domains (ie, the early yanshanian), the south china provinces lacke the isotopic age values (200-180 ma) of the early jurassic igneous rocks (j1). therefore, it is generally believed that j1 is the intermittent period of magmatic activity (yu et al., 2009). although the geological work in southeastern china has been relatively high in the past, it mainly focuses on the late mesozoic igneous rocks and cenozoic volcanic rocks. the study on the geological characteristics and tectonic setting of early mesozoic igneous rocks is not in-depth, especially the isotopic age of early jurassic igneous rocks. chen et al. (2002) made rb-sr isochron dating of bimodal volcanic rocks in the xunwu-baimianshi basin and dongkeng-linjiang basin in southern jiangxi province. the values of 178±7ma for basalts in dongkeng basin, 173±6ma for basalts in baimianshi basin and 165±2ma for rhyolite are equivalent to those of late early jurassic to early middle jurassic, which changed the predecessors’opinions on the formation of igneous rocks in these areas in late jurassic. xu (1992)proposed for the first time that the volcanic rocks in fankeng formation in yongding, southwest fujian province were bimodal assemblages of early jurassic (rb-sr age 179 ma), formed in intracontinental extensional environment, which aroused scholars’ attention to the early jurassic igneous rocks in the southeastern continent. zhou and chen (2001) obtained the rb-sr isochron age of the rhyolite in the yongding hulei basin again. li and li (2007) have made shrimp zircon u-pb dating of the keshubei intrusion in southern jiangxi, and obtained an isotopic age of 189±3ma, which coincides with the age of 187±4 ma (zircon u-pb) measured by ding et al. (2005) of the xiangzikou intrusion in the middle part of nanling mountains, hunan province. xing et al. (2002) made rb-sr isochron dating of gabbro in meizhou xialan complex, and obtained 179 ma age. again, shrimp zircon u-pb dating of granite (1196±2ma)) and gabbro (1195±1ma) in the intrusion proves that magmatic activity also existed in the early jurassic in northern guangdong (yu et al., 2009), which is also the earliest granite age of yanshanian period up to now. it is believed that the pacific tectonic domain has begun to act on south china (chen et al., 1999) at least around 195ma. the yanshanian magmatic activity along the southeastern coast of china is a part of the tectonic-magmatic cycle in the western pacific ocean. compared figure 4. sample number b-1 weighted average age figure 5. sample number b-2 weighted average age 140 jinhui tang, feng lou with the unprecedented magnitude and intensity of magmatic activity in the late jurassic (~140 ma), the signs of early jurassic magmatic activity are not obvious. however, the above age data indicate that the pacific plate subducted to the eurasian plate in the early mesozoic (~200ma), which has a precise chronological record in the middle and eastern part of nanling mountains. the u-pb zircon ages of the two groups of fine grained wenxiang granite in the renchao basin are 179±1ma and 186 ma, which provide new information for the lack of reports of magmatic activity during the j1 period in southeastern china. the origin of the u-pb zircon may be related to plate subduction, mantle material upwelling or heating of crustal material and the introduction of some mantle-derived material. although the early yanshanian granites are mostly the products of magmatic emplacement under extensional tectonic environment, the alkaline rock-syenite-granite assemblages formed in the typical extensional tectonic environment of the same period have not been found yet because of the intense volcanic activity in the late yanshanian period of renchao basin, the thick coverage of volcanic caprock and the few outcrops on the surface. in addition, the study area is located at the junction of nanling and wuyi metallogenic structural belts, and its tectonic significance is particularly special (zhao et al., 2008). therefore, the geological work in this area needs to be further strengthened. conclusion the la-icp-ms zircon u-pb dating of the basal-image granite in the poor dry area of the renzi basin yielded the isotopic ages of 179±1ma and 186ma, respectively. it indicates that the formation of the basement lithology granite in the yanshen basin is in the late early jurassic period and is the product of the first magma activity of yanshan. this age provides new chronological data for reports of the lack of magmatic rock activity during the j1 period in northeastern guangdong. the subduction of the pacific plate to the asian plate in the early mesozoic (~200ma) resulted in the warming of the crust, the melting of crustal materials into magma, and the compression of the crust resulted in a series of ne-trending sag basins. the basement hieroglyphic granites in the renchao basin were formed at this stage. during this period, the dading iron ore back rock mass (176±4m a) in lianping and zijinshan rock mass in southwestern fujian were (168±4ma) (wu & zheng, 2004), both of which developed granite basement. although they were formed later, they should be the product of magmatic activity in the same period regionally. acknowledgments: central geological exploration fund project (2007441004) reference andersen, t. (2002). correction of common lead in u-pb analyses that do not report 204pb. chemical geology, 192, 59-79 chen, p., hua, r., zhang, b., lu, j., & fan, c. (2002). early post-orogenic granitoids in yanling, nanling: petrological constraints and geodynamic background. science in china series d: earth science, 32(4), 279-289. ding, x., chen, p., & chen, w. (2005). lazims u-pb dating of zircons from the lushan granite in hunan: diagenetic implications and significance. science in china d (earth science), 35(7), 606-616. gao, w., yel, g., baskonus, h. m. & cattani, c. (2020). complex solitons in the conformable (2+1)-dimensional ablowitz-kaup-newell-segur equation. aims mathematics, 5(1), 507–521. harris, c. a., allen, c. m., bryan, s. e., campbell, i. h., holcombe, r. j., & palin, j. m. (2004). ela-icp-ms u-pb zircon geochronology of regional volcanism hosting the bajo de la alumbrera cu-au deposit: implications for porphyry-related mineralization. miner deposit, 39, 46-67. liang, h. y., campbell, i. h., allen, c. m., sun, w., liu, c. w., yu, h. x., xie, y. w., & zhang, y. q. (2006). zircon ce4 +/ce3 + ratios and ages for yulong ore-bearing porphyries in eastern tibet. miner deposit, 41, 152-159. liu, y., gao, s., hu, z., gao, c., zong, k., & wang, d. (2010). continental and oceanic crust recycling-induced melt-peridotite interactions in the trans-north china orogen: u-pb dating, hf isotopes and trace elements in zircons of mantle xenoliths. journal of petrology, 51, 537571. liu, y., lu, y., & liang y. (2007). basic characteristics of volcanic rocks in renzi basin, northeast guangdong. journal of guilin university of technology, 27( 2), 159-164. liu, y. s., hu, z. c., gao, s., günther, d., xu, j., gao, c., & chen, h. (2008). in situ analysis of major and trace elements of anhydrous minerals by la-icp-ms without applying an internal standard. chemical geology, 257, 34-43. li, w., mao, j., & zhu, y. (1998). mesozoic igneous rocks and deposits in southeastern china. beijing: science press. li, z. x. & li, x. h. (2007). formation of the 130km-wide intra-continental orogen and post-orogenic magmatic province in mesozoic south china: a flat-slab subduction model. geology, 35, 179-182. ludwig, k. r. (2003). isoplot 3.00: a geochronological toolkit for microsoft excel. berkeley: berkeley geochronology center, california. shu, l. & zhou, x. (2002). the late mesozoic tectonic effect in southeastern china. geological review, 48(3), 249-260 sláma, j., kosler, j., condon, d. j., crowley, j. l., gerdes, a., hanchar, j. m., horstwood, m. s. a., morris, g. a., nasdala, l., norberg, n., schaltegger, u., schoene, b., tubrett, m. n., & whitehouse, m. j. (2008). plesovice zircon_a new natural reference material for u-pb and hf isotopic microanalysis. chemical geology, 249, 1235. song, s., zhao, w., & zheng, m. (2010). structural tectonic features and uranium mineralization of malou in renchang basin, northeastern guangdong province. journal of guilin university of technology, 26(5), 283-287. wang, d., & zhou, x. (2002). the genesis and crustal evolution of the late mesozoic granitic volcanic-intrusive complex in southeastern china. beijing: science press, 217-229. wiedenbeck, m., alle, p., corfu, f., griffin, w. l., meier, m., oberli, f., von quadt, a., roddick, j. c., & spiegel, w. (1995). three natural zircon standards for u-th-pb, lu-hf, trace element and ree analyses. geostandards and geoanalytical research, 19, 1-23. wu, y., & zheng, y. (2004). study on the mineralogy of zircon and its restriction on u-pb age interpretation. chinese science bulletin, 49(16), 15891604. xing, g., yang, z., mao, j., shu, l. s., shen, j. l., sun, q. h., chen, y., & tao, k. y. (2002). characteristics of early jurassic igneous rocks on the margin of the southeastern china and their tectonic significance. geological bulletin, 21 (7), 384-391. xu, m. (1992). early jurassic bimodal volcanic rocks and their tectonic setting in yongding area, fujian province. geology of fujian, (2), 115-125. xu, x., zhang, y., & jia, d. (2010). zircon la-icp-ms u-pb and muscovite 40ar/39ar chronology and its constraints to early yanshan incidents in southeastern china. data in brief, 29(3), 87-94. yu, x., di, y., & wu, y. z. (2009). the existence of the early jurassic magmatism in northern guangdong: evidence from the shrimp zircon u-pb chronology of the xiayan complex. science in china d (earth science), 39(6), 681693. zhao, x., mao, j., & chen, r. (2008). zircon shrimp dating of the zijinshan rock mass in southwestern guizhou and its geological significance. china geology, 35(4), 590-597. zhou, j., & chen, r. (2001). late mesozoic crustal slow-acting geochemistry in southeastern fujian. geochemistry, 30(6), 547-558. keywords: gypsum rock; dry-wet cycle; particle flow numerical simulation; water absorption; porosity. palabras clave: roca de yeso; ciclo seco-húmedo; simulación numérica de flujo de partículas; absorción de agua; porosidad. how to cite item wu, g., chen, w., & cui, k. the numerical simulation of the drying-wetting cycles' influence on the deterioration characteristics of gypsum rocks under particle flow code. earth sciences research journal, 24(1), 55-59. doi: https://doi.org/10.15446/esrj.v24n1.85486 in order to study the influence of dry-wet cycling on the deterioration characteristics of gypsum rocks and solve the problems encountered in engineering construction, in this study, gypsum rocks are taken as the research object. with the combination of laboratory test and theoretical analysis, the numerical simulation of particle flow is carried out, and the deterioration characteristics of physical and mechanical properties of gypsum rock under dry-wet cycling are studied. the results show that gypsum, quartz, zeolite and dolomite are the main components of gypsum rocks. gypsum occupies the most components in gypsum rocks, so the various characteristics of gypsum greatly affect the characteristics of gypsum rocks. the process of water absorption and loss of gypsum is similar, which shows that the rate of water absorption or loss of gypsum is faster in the early stage, and tends to be stable in the later stage. the curve of the whole process of water absorption and loss is fitted by negative exponential function, and the effect is better. the larger the porosity of gypsum rock is, the better its water absorption performance is. intergranular pore, dissolution pore and dissolution pore are the main pore types of gypsum rock. intergranular pore is the main water absorption channel of gypsum rock. the cumulative water absorption increases with the increase of wetting and drying cycles. the change of water absorption curve is mainly manifested in water absorption rate and time. the more the number of wet-dry cycles is, the higher the water absorption rate in the early stage of water absorption is, the closer the characteristic curve to the coordinate axis of water absorption is, and the shorter the water absorption time is. in contrast, the shape difference of water loss curve is very small. it can be seen from this that in the process of wetting and drying cycle, the hydrophysical and hydrochemical processes promote each other, which changes the crystal structure and pore structure of gypsum rocks, reduces the crystal strength and increases the porosity, thus leading to the deterioration of the mechanical properties of gypsum rocks. abstract the numerical simulation of the drying-wetting cycles' influence on the deterioration characteristics of gypsum rocks under particle flow code simulación numérica de la influencia de los ciclos de secado-humedecimiento en las características de deterioro de las rocas de yeso bajo el código de flujo de partículas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.85486 con el fin de estudiar la influencia del ciclo seco-húmedo en el deterioro de las rocas de yeso y resolver algunos problemas de la ingeniería de construcción, las rocas de yeso son el objeto de investigación de este estudio. se realizó la simulación numérica del flujo de partículas con la combinación de pruebas de laboratorio y análisis teóricos, y se estudiaron las características de deterioro de las propiedades físicas y mecánicas de la roca de yeso expuesta al ciclo seco-húmedo. los resultados muestran que el yeso, el cuarzo, la zeolita y la dolomita son los componentes principales de las rocas de yeso. el yeso ocupa la mayoría de los componentes en las rocas de yeso, por lo que las diversas características del yeso afectan en gran medida las características de las rocas de yeso. el proceso de absorción de agua y pérdida de yeso es similar, lo que demuestra que la tasa de absorción de agua o pérdida de yeso es más rápida en la etapa inicial y tiende a ser estable en la etapa posterior. el índice de porosidad es directamente proporcional con la absorción de agua. la curva de todo el proceso de absorción y pérdida de agua se ajusta mediante una función exponencial negativa, y el efecto es mejor. los poros intergranulares y los poros de disolución son los principales tipos de poros de la roca de yeso. el poro intergranular es el principal canal de absorción de agua. la absorción acumulada de agua aumenta con el aumento de los ciclos de humectación y secado. el cambio de la curva de absorción de agua se manifiesta principalmente en la velocidad y el tiempo de absorción de agua. cuanto mayor sea el número de ciclos húmedo-seco mayor será la tasa de absorción de agua en la etapa inicial, más cercana es la curva característica al eje coordinado y más corto es el tiempo de absorción de agua. en contraste, la diferencia de forma de la curva de pérdida de agua es muy pequeña. se evidencia que en el proceso del ciclo de humectación y secado los procesos hidrofísicos e hidroquímicos se promueven entre sí, lo que cambia la estructura cristalina y la estructura de los poros de las rocas de yeso, reduce la resistencia del cristal y aumenta la porosidad, lo que conduce a la deterioro de las propiedades mecánicas de las rocas de yeso. resumen record manuscript received: 12/03/2019 accepted for publication: 03/09/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 55-59 guopeng wu1,2,3, wenwu chen1,2, kai cui1,2,3 1key laboratory of mechanics on disaster and environment in western china, department of geological engineering, lanzhou university,lanzhou 730000, china 2school of civil engineering and mechanics, lanzhou university, lanzhou 730000, china 3key laboratory of disaster prevention and mitigation in civil engineering of gansu province, department of geotechnical engineering, lanzhou university of technology, lanzhou 730050, china * corresponding author: g e o l o g ic a l e n g in e e r in g 56 guopeng wu, wenwu chen, kai cui introduction from the deep part of the earth to the surface of the earth, water-rock interaction is ubiquitous, especially the wet-dry cycle (jian & niu, 2016). it is the most closely related to human activities, and is also one of the most common water-rock interactions. for example, the precipitation and evaporation of rainwater and the rise and fall of reservoir water level are all the inducements of dry-wet circulation. the deterioration of rock mass caused by dry-wet cycling is equivalent to a kind of “fatigue effect”, and this deterioration is usually stronger than that of long-term immersion (yao et al., 2015), which has an important impact on the stability and safety of rock mass engineering (jiang & shao, 2014). therefore, it is of great engineering significance to study the deterioration characteristics of mechanical properties of rock mass and rock mass under dry-wet cycling. in recent years, scholars at home and abroad have carried out extensive experimental studies on the changing law of rock mechanical properties under dry-wet cycling, and obtained many research results (li et al., 2015). numerical simulation is one of the important means to study the mechanical characteristics of rock (gaylarde et al., 2017). in recent years, the discrete element method (dem) has been widely used in the numerical simulation of rock mechanical properties because it does not involve complex macro-constitutive models and can analyze the numerical simulation results from macro and micro levels (ion et al., 2016). particle flow code (pfc) is a kind of dem, which constructs rock materials into aggregates of particles and simulates the macro-mechanical properties of rock materials through the interaction of particles (grandclerc et al., 2016). pfc has been widely used in rock fracture analysis. in pfc, the macro mechanical properties of rock cannot be given directly, but should be matched by setting the geometric parameters and contact properties of particles and choosing reasonable meso-parameters. trial-and-error method is the most commonly used method for calibrating meso-parameters at present (sijakova-ivanova et al., 2014). by comparing the macro-mechanical parameters of rock and numerical simulation results, the meso-parameters are adjusted continuously to achieve the similarity between them. if the relationship between macro and micro parameters is not understood and the micro parameters are randomly adjusted, it may lead to a large number of numerical experiments and increase the difficulty of modeling (de weerdt & justnes, 2015). based on the above background, in this experiment, gypsum rock is taken as the research object in order to study the deterioration degree of the physical and mechanical properties of gypsum rock caused by wet-dry cycling. combined with laboratory tests, theoretical analysis and numerical simulation of particle flow, water absorption and water loss tests are carried out on gypsum rocks under different drying and wetting cycles. water absorption and water loss characteristics, pore characteristics and mechanical characteristics of gypsum rocks are obtained. the deterioration effect of gypsum rock under drywet cycling is discussed in detail, which provides a new theoretical basis for future research. method research content and technical route in this study, through data acquisition and field investigation, combined with laboratory experiments and theoretical analysis, the basic physical characteristics of gypsum rocks are described. the main components, water absorption, water loss, fracture inclination and uniaxial compressive strength of gypsum rocks are studied, and the preparation for the subsequent dry-wet cycle experiments is made in advance. the effect of dry-wet cycling on the deterioration characteristics of gypsum rocks has been studied under the pfc. gypsum rocks are subjected to dry-wet cyclic deterioration tests, and gypsum rocks with different cycles are subjected to water absorption and loss tests as well as pore characteristics experiments. then, the effects of dry-wet cycle on water absorption loss, pore structure and mechanical properties of gypsum are discussed. on this basis, the deterioration mechanism of gypsum is also analyzed. finally, the particle flow law of gypsum in different drying and wetting cycles was determined under the background of pfc. the relationship between crack inclination and uniaxial compressive strength under different drying and wetting cycles is studied through bonding clearance and meso-parameters. the mechanism of meso-deterioration of gypsum rock under drying and wetting cycles is further discussed. the technical route of this research is shown in the following figure. data collection determination of basic physical properties and selection of samples dry-wet cycle deterioration test two-dimensional particle flow simulation degradation mechanism of gypsum rock under the action of dry-wet circulating water-rock research conclusion figure 1. technical roadmap of this paper main components of gypsum minerals gypsum samples are collected from the field. in the laboratory, its components are studied experimentally. the samples are analyzed by x-ray images and the diffraction patterns are obtained. then, according to the diffraction pattern, the main components and their respective contents of gypsum rocks are calculated. according to the different contents of main minerals, the samples of experimental rock blocks are divided into four groups, numbered a, b, c and d in turn. the final results are shown in the table below. according to the table, gypsum, quartz, zeolite and dolomite are the main components of gypsum rocks. gypsum occupies the most components in gypsum rocks, so the various characteristics of gypsum greatly affect the characteristics of gypsum rocks. table 1. mineral composition and content of gypsum rocks classification of rocks gypsum (%) quartz (%) zeolite (%) dolomite (%) a > 95 1 0 < 4 b 90 2 2 6 c 83 1 0 16 d 71 1 1 27 characteristics of water absorption curve of gypsum rock in order to obtain representative standard specimens from four groups of specimens, water absorption tests are carried out at room temperature of 16°c ± 2°c. after 10 days of continuous water absorption, the water absorption of 57the numerical simulation of the drying-wetting cycles' influence on the deterioration characteristics of gypsum rocks under particle flow code of pore will change the water absorption characteristics of gypsum rocks. in order to obtain more accurate results, four test samples with different porosity are selected to study the effect of porosity on water absorption characteristics. water absorption experiments are carried out under the same experimental conditions, so water absorption curves and parameters are obtained. the experimental results show that the higher the porosity of the sample is, the higher the water absorption is and the higher the water absorption rate is. it is found that intergranular pore, dissolution pore and dissolution pore are the main pore types of gypsum rocks. intergranular pore is the main pore type. intergranular pore refers to the pore between gypsum crystals. because of the close contact between crystals, the intergranular pore becomes limited and the diameter becomes smaller, but they are the main channel of water absorption. the dissolution pore is produced by the contact between gypsum rock and groundwater. the dissolution pore of the dissolution pore is small and the regularity is not strong, and the shape and size of the dissolution pore are also different. if the connectivity is good, the dissolution pore can form a pore network together with the intergranular pore. in addition, there are some independent dissolution holes in gypsum crystal, which is caused by incomplete dissolution of gypsum crystal. the dissolution pore is relatively small in volume and low in connectivity, and has little water absorption to rocks. particle flow model of gypsum rocks when establishing the particle flow model of gypsum rock, because the macro-mechanical parameters of the model are affected by the number and size of gypsum rock particles, it is necessary to take into account the effect of res (resolution) on the macro-mechanical parameters at the minimum scale of the model. some studies believe that the number and size of gypsum particles have the least influence on the particle flow model of gypsum rocks when res ≥ 10, while others believe that the number and size of gypsum particles have the least influence on the particle flow model of gypsum rocks when res ≥ 20. therefore, in this experiment, the simulation scale and calculation speed are considered comprehensively, and the particle flow model of gypsum rock is analyzed. the calculation formula of res is as follows: res l r r r= ( ) +( ) / / /min max min1 1 (2) in the formula above, rmin represents the diameter of the smallest particle. in this study, rmin .= 0 4, and r rmax min/ .= 1 66. l is expressed as the minimum scale of the model. rmax denotes the diameter of the largest particle. uniaxial compression is achieved by imposing loading rates of the same size and opposite direction on the walls at both ends of the sample. the loading rate must be small enough to keep the sample close to static equilibrium. the results show that the loading speed of 0.2m/s is small enough to simulate the particle flow of gypsum. results and discussion change of one-dimensional water absorption curve of gypsum rocks with different cycles the following figure shows the water absorption curve of gypsum rock samples after different times of dry-wet cycles. as can be seen from the figure below, the cumulative water absorption of gypsum rock samples increases with the increase of wet and dry cycles. moreover, with the increase of wetting and drying cycles, the water absorption curve of gypsum rock samples is closer to the water absorption coordinate axis. this phenomenon appears in the characteristics of water absorption curve, which indicates that the water absorption rate increases with the increase of the number of cycles. finally, with the increase of time, the water absorption of gypsum rock samples reaches saturation, and the duration of water absorption becomes shorter. as can be seen from the above figure, the cumulative water absorption of the sample increases with the increase of the number of wet-dry cycles, which indirectly indicates that the porosity of the gypsum sample increases. the exponential function is used to fit the water absorption curve of the each sample remains basically stable. the following figure shows the variation curve of water absorption of gypsum rocks with water absorption time. during the 10-day water absorption process, the accumulative water absorption of group a rock sample is the largest, reaching 5.78g. secondly, the accumulative water uptake is 2.54g and 1.98g in group c and d, respectively. the cumulative water absorption of group b is the smallest, which is 1.46g. the cumulative water absorption of each group of samples is mainly completed in the early stage of water absorption. one day after water absorption, the accumulative water absorption of group a is 61.3%, and the accumulative water absorption of other groups is more than 36%. four days after water absorption, the accumulative water absorption of group a is over 92%, and the accumulative water absorption of other groups is over 82%. there are also obvious differences in the water absorption rate of each stage, and the water absorption rate decreases gradually with the increase of time. figure 2. water absorption characteristic curves of gypsum rocks of different groups with the increase of water absorption time, the water absorption of gypsum rocks reaches a certain degree of saturation, and the water absorption decreases slowly, gradually tends to zero. then, the water absorption characteristic curves of each group of gypsum rocks are fitted by negative exponential function. the general fitting expressions are as follows: m m aes bt=   (1) in the above formula, t represents the time when gypsum rock absorbs water. ms indicates the saturation of gypsum rock when it is saturated by water absorption. a and b represent the fitting parameters. the following figure shows the fitting parameters of water absorption curve of gypsum rock. from the following table, it can be seen that the fitting effect is good and the fitting correlation coefficient is above 0.99. table 2. fitting parameters of the whole process curve of water absorption serial number fitting parameters ms a b r 2 a 5.725 5.472 0.038 0.992 b 1.503 1.412 0.019 0.995 c 2.684 2.589 0.023 0.997 d 1.975 2.015 0.015 0.994 effect of porosity of gypsum rocks on water absorption characteristics the water absorption characteristics of gypsum rocks are mainly influenced by porosity and pore structure, and the size, shape and connectivity 58 guopeng wu, wenwu chen, kai cui experimental sample. the fitting effect is good and the correlation coefficient is 0.982. the following table shows the water absorption parameters of different drying and wetting cycles. table 3. water absorption parameters of samples with different cycles of wetting and drying number of dry-wet cycles water absorption (g) 0.4 h average water absorption rate (g/h) 0 1.25 0.250 1 1.48 0.500 3 1.67 1.100 6 2.13 3.050 12 2.55 3.750 uniaxial compressive strength of complete and single crack specimens under different dry-wet cycles the following table shows the uniaxial compressive strength of complete specimens and single crack specimens under different wet and dry cycles. it can be seen from the table that the crack dip will affect the uniaxial compressive strength. the relationship between crack inclination and uni axial compressive strength under different wet-dry cycles is shown in the following figure. it can be seen from the figure that the number of dry-wet cycles and the inclination of cracks will affect the uniaxial compressive strength. the uniaxial compressive strength of single fracture specimen increases with the increase of crack inclination regardless of the number of dry-wet cycles. under the same fracture inclination, the uniaxial compressive strength of single fracture specimen decreases with the increase of the number of dry-wet cycles. with the same number of wet and dry cycles, the uniaxial compressive strength of single fracture specimen increases with the increase of fracture inclination. the uniaxial compressive strength of a single fracture specimen is the lowest when the fracture dip is equal to 30 degrees. after that, the uniaxial compressive strength of the single fracture specimen increases with the increase of the crack inclination. conclusion in this study, gypsum rock is taken as the research object for the deterioration of gypsum rock under the action of dry-wet cycle. in order to study the deterioration of the physical and mechanical properties of gypsum rocks caused by dry-wet cycling, combined with laboratory tests, theoretical analysis and numerical simulation of particle flow, gypsum rocks under different drywet cycles are tested for water absorption and water loss, and finally the water absorption and water loss characteristics, pore characteristics and mechanical characteristics of gypsum rocks are obtained. the deterioration effect of gypsum rock under dry-wet cycling is discussed in detail. experiments show that during the dry-wet cycle, the hydrophysical and hydrochemical processes promote each other, which changes the crystal structure and pore structure of gypsum rocks, reduces the crystal strength and increases the porosity, thus leading to the deterioration of the mechanical properties of gypsum rocks. due to the limited ability and time, there are still many deficiencies in the research content of this study. at present, the dry-wet cyclic deterioration tests are often aimed at intact rocks, without considering the influence of cracks on the deterioration of rock mechanical properties. in this study, the deterioration characteristics of single-fissure gypsum rocks under dry-wet cycling are preliminarily studied by means of the particle flow model of gypsum rocks. the results show that the occurrence of cracks has changed the deterioration law of rocks and has obvious influence. in practical engineering, more fractured rock mass is involved. the deterioration characteristics of fractured rock mass must be different from that of intact rock mass. therefore, in the future research, the deterioration mechanism of fractured rock mass under dry-wet cycling can be further studied. reference de weerdt, k., & justnes, h. (2015). the effect of sea water on the phase assemblage of hydrated cement paste. cement & concrete composites, 55, 215-222. gaylarde, c., baptista-neto, j. a., ogawa, a., kowalski, m., celikkol-aydin, s., & beech, i. (2017). epilithic and endolithic microorganisms and deterioration on stone church facades subject to urban pollution in a sub-tropical climate. biofouling, 33(2), 113-127. figure 3. one-dimensional water absorption characteristic curve of samples after different dry-wet cycles table 4. uniaxial compressive strength of complete and single crack specimens under different dry-wet cycles number of cycles crack dip complete sample 0 30 45 60 75 90 0 15.76 15.02 16.03 19.86 21.74 24.12 25.64 5 11.43 11.96 13.52 14.97 18.12 19.43 20.63 10 9.98 9.21 10.42 11.41 14.52 16.18 17.06 15 8.12 8.25 9.64 10.15 13.49 15.95 16.27 figure 4. the relationship between uniaxial compressive strength and crack inclination of single fracture specimen under different wet-dry cycles 59the numerical simulation of the drying-wetting cycles' influence on the deterioration characteristics of gypsum rocks under particle flow code grandclerc, a., gueguen-minerbe, m., nour, i., dangla, p., & chaussadent, t. (2016). accelerated biodeterioration test for the study of cementitious materials in sewer networks: experimental and modeling. key engineering materials, 711, 1069-1075. ion, r. m., teodorescu, s., ştirbescu, r. m., dulama, i. d., suica-bunghez, i. r., bucurica, i. a., fierascu, e. c., fierscu, i., & ion, m. l. (2016). effects of the restoration mortar on chalk stone buildings. iop conference, 133(1), 012038. jiang, l., & niu, d. (2016). study of deterioration of concrete exposed to different types of sulfate solutions under drying-wetting cycles. construction & building materials, 117, 88-98. jiang, y., & shao, m. (2014). effects of soil structural properties on saturated hydraulic conductivity under different land-use types. soil research, 52(4), 340. li, w., wang, m. y., liu, li. y., & yan, y. (2015). assessment of long-term evolution of groundwater hydrochemical characteristics using multiple approaches: a case study in cangzhou, northern china. water, 7(3), 1109-1128. sijakova-ivanova, t., eric, s., & saric, k. (2014). preliminary investigations of deterioration on monument to fallen soldiers in the second world war in stip, republic of macedonia. journal of chromatography a, 8(1), 2000. yao, q., li, x., zhou, j., ju, m., chong, z., & zhao, b. (2015). experimental study of strength characteristics of coal specimens after water intrusion. arabian journal of geosciences, 8(9), 6779-6789. earth sciences research journal earth sci. res. sj. vol. 17, no. 1 (june, 2013): 67 71 late permian brachiopoda fauna in north-western iran mostafa yousefirad1, touran khamooshi2 and rahim shaabanian2 1 department of geology, faculty of earth sciences, payame noor university,tehran, iran 2 department of geology, faculty of earth sciences, payame noor university,tabriz, iran radyousefi@yahoo.com mobile: 00989183638957 paleontology the late permian marine sequence in the northwest of iran, in eastern azerbaijan province in the zal stratigraphic section, was selected for studying brachiopods. samples were collected from the ali bashi and jolfa formation.s. twenty-seven species from 13 brachiopoda genera were recognised in this study. the recognised fossil community was compared to brachiopod communities in some regions of iran and the tethyan region, suggesting the late dzhulfian period as the age for the deposits being studied. recognized brachiopods belonge to the orders athyris, rhynconellida, productida and strophomenida. keywords: brachiopoda, late permian, iran. la secuencia marina del pérmico superior en el noroeste de irán y el este de azerbaiyán, sección estatigráfica de zal, fue seleccionada para este estudio de braquiópodos. las muestras fueron tomadas en las formaciones ali bashi y jolfa. se reconocieron 27 especies de 13 géneros de braquiópodos en este análisis. las muestras recolectadas fueron comparadas con sus similares de otras regiones de irán y de la región del tetis, lo que sugirió el período de tardío de dzhulfan como la proveniencia de estos depósitos. los braquiópodos reconocidos en este trabajo pertenecen a las familias de athyris, rhynconellida, productida y strophomenida. key words: brachiopoda, late permian, iran palabras clave: braquiópodos, pérmico superior, irán record manuscript received: 16/11/2012 accepted for publication: 04/04/2013 abstract resumen 1. introduction marine permian deposits are exposed in many parts of eastern and western azarbaijan, in north-western iran (shabanian and bagheri 2008) .the zal stratigraphic section, ( 40,45’ eastern longitude and 38,43’ latitude), is located in north -eastern azerbaijan (figure 1);. this section is part of the abade-jolfa deposit sequence, having a general northwest-southeast trend.. the permian deposit sequence in the region being studied consisteds of the vajnan formation (baghbani, 1997; shabanian and bagheri, 2008 ) (or doroud formation), the surmaq formation (iranian-japanese research group, 1981), the jolfa formation (parto azar, 1997) and ali bashi formation (parto azar, 1997; tiechert, 1973). the permian deposit in the zal section,( having igneous nonconformity), is located on an igneous rock series before the permian age and is itself covered by constant transition to clararia limestone, vermiculite and the elika formation,caused by middle and early trassiac aging. the 46-meter-thick ali bashi formation was examined in this research to enable a paleontological study of brachiopods. the permian sequence in southwest jolfa was comprehensively studied for the first time by stepanone et al., in 1969. they divided the permian and triassic beds in the ali bashi mountain section in to 8 separate stone units;. unitse a and b were from the guadalupian age, c-d from the dzulfian age, e transitional stages, f a paratirolites horizon with early triassic age and h and g introduced the elika formation in to the section being studied. tiechert et al., then examined the ali bashi section and made seminal changes to the divisions of stepanov et al., (1967);. the according to them, e and f belonged to the highest part of the permian division, i.e. the dorashamian stage and they called it the ali bashi formation. re-examining the section, parto azar made some changes in permian borders and lithostratigraphic unites names. according to him, the lower part of unit a was the surmaq formation, the upper part and all of unit b consisted of thes jolfa formation and the 52 meters thickness c,d,e, and f were introduced as the ali bashi formation . based upon brachiopod fossils, (cephalopods and condonts), the age of the jolfa formation has been classified as being late dzhulfian in terms of permian stages in the tethyans region (chen et al., 2005). several reports have been concerned with the palaeontology of foraminifera, corals, brachiopods, conodonts and cephalopods of the ali bashi formation and formations from the same stage in alborz and abade. first information to be published on iran’s brachiopods dates back to 1916, (stayonav). stayonav et al., (1916) released a comprehensive report on the permian and permo-triassic border in the ali bashi section mostafa yousefirad, touran khamooshi and rahim shaabanian in southern jolfa and studied its brachiopods. restudying the ali bashi section, tiechert et al., (1973) and baud (2008) introduced many of its macrofossils, including brachiopods. an iranian-japanese research group (iranian-japanese research group, 1981) introduced brachiopods from the hambast formation in abade; and research into late permian brachiopods was published by fantini-sistin, in1965, fantini-sistini & glaus in, 1966 and angiolini in, 2010. biostratigraphy and event stratigraphy in iran around the permian-triassic boundary (ptb) has been published by kozur (2007). figure 1: position of study area the present research was aimed at introducing late permian brachiopods in stratigraphic section so as to determine their age and stratigraphic correlation as well as stratigraphically examining the value of recognising brachiopod fossils (figure 2). methodology systematic sampling was used for recognising and studying permian brachiopods in the zal stratigraphic section,; 60 samples were collected, along with corals and bryozoans, 13 genera and 27 species being recognised in the 15-meter thicksection. an attempt was made to determine change in their age and size, based on brachiopods. discussion the ali bashi formation in the zal stratigraphic section is 50 meters thicks and includes thin layers of limestone, shale, marl and middle layers of limestone (khamooshi 2012). an accurate field study has revealsed brachiopod, separate rugose and colonial coral , bryozoan, crinoids and cephalopod macrofossil frequency;. 27 species f rom 13 genera regarding late permian brachiopod s were recognized in the present research work, theost significant brachiopod families, in terms of frequency and variety, included productida, strophomenida, rhynconellida and athyris. studies were based upon external shell features , including shell, length, thickness and width, examining features such as pedicle valve, brachial valve, hinge line, foramen development or non-development, delthyrium (notch) status , shell shape, umbo (posterior extremity) shape, commissure line shape, existence or non-existence of fold and sulcus , their development and type of decoration . according to brachiopod stratigraphic dispersion in the zal stratigraphic section and comparison to a recognized late permian fossil community in iran and other parts of the tethyan region , late dzhulfian to dorashamian has been introduced as the age of the ali bashi formation. this research resulted in 13 genera and 27 species belonging to 5 different families being recognized, as follows: spinomarginifera helica(abich), s.spinicilicata(arthaber), orthotetina cf. persica (schewien) orthotetina cf. persica (schewien), o.eusarkos, o. ex.gr eusarkos, araxilevis intermedius (abich) tschernyschewia typica (stoyanow), compresoproductus dzhulfensis (stoyanow), permophricodothyris ovata (pavlova), araxathyris protea (abich), a . quadrilobata (abich) a. lata (grunt), a. lata minor (grunt), a. araxensis (grunt), a. araxensis minor (grunt), a. aff. ogbinensis, , wellerella dorashamensis,( sokolskaja ), psedowellerella araxensis (sokolskaja) , haydenella sp. figure 2: a chart of brachiopod biozonation in the zal section and stratigraphic column for the zal section brachiopods, in terms of biostratigraphy value, can be categorized into two assemblage areas (figure 2): figure 2.stratigraphic column of the zal section the araxilevis – orthotetina assemblage biozone this biota association was 10 metrs thick. the zone began with the appearance of araxilevis, tschernyschewia, spinomarginifera and end with appearance of compresoproductus, orthotetina. the biozons had many huge brachiopods, among which productidae and arthotetins were remarkable. some of the most frequently recognized productidae are given below: spinomarginifera helica(abich , 1878), araxilevis intermedius (abich, 1878), tschernyschewia typica (stoyanow, 1910), haydenella sp., compressoproductus dzhulfensis (stoyanow,..). huge (araxilevis) and moderate productida, such as tschernyschewia, compresoproductus, and huge orthotetina such as orthotetina eusarkos (abich), along with the coral plentamplexus leptoconicus (abich 1878) and the crinoid erisocrinus araxensis (yakovlev and ivanov, 1965) represent significant fossils from this biota association. permophricodothyris – araxathyrisa assemblage biozone: this biota association was 12 metres thick. this biozone began with the appearance of compresoproductus dzhulfensis (stoyanow), orthotetina eusarkos (abich) other brachiopods from this biota association were: compresoproductus dzhulfensis (stoyanow), araxilevis intermedius (abich), tschernyschychewia typica (abich), permophricodothyris ovate (pavlova). spinomarginifera helica (abich, araxathyris araxensis minor (grunt), araxathyris cf. protea (abich) and araxathyris quadriloba (abich; coral type pleramplexus leptoconicus (abich) has also been found in this biota association(figure 3). figure 3.biozonation of the zal section based on brachiopod fauna systematic classification of brachiopods 13 genera and 27 species were recognized in this research in the zal section (plates 1-5). comparing the recognized fossil community with other brachiopod communities other parts of iran and tethyan region suggested late dzulfian as of the age of deposits being studied. differen late permian brachiopod genera were found in the zal section. some of the recognized species and genera were as followi (based on moore, 1965): phylum brachiopoda class articulata huxley , 1869 order spiriferida waagen .1883 suborder athyrididina boucot, johnson&staton,1964 superfamily athyridacea mcoy, 1844 subfamily spirigerellinae grunt, 1955 genus araxathyris quadriloba, abich,1878. description: shell shape was convex-convex, dimensions beng almost the same, its width a little being a little bit more than its length, the pedicle valve had a sulcus lying from the umbo to the front side, and the brachial valve had a huge, liftedup fold. the commissure line was uniplicate. the most sulcus depression was in the front 1/3 of the shell. it was 3.1 cm thick, 3.9 cm long and 3.2 cm wide. geographical spread: shahreza, caucus, alborz – balade. order strophomenida opik, 1934 suborder strophomenidina opik ,1934 superfamily davidsonicea king, 1850 family meekellidae stehli, 1954(lamp shell) subfamily meekellinae stehli, 1954 genus orthotetina eusarkos abich, 1878 description: shell shape was convex – convex, with direct hinge line. the delthyrium coud be seen. interarea was obvious. the commissure line was plain. decorations were thorny. it was 2.3 cm thick, 4 cm long and 5 cm wide. suborder productidina waagen, 1883 superfamily productacea gray, 1840 subfamily marginiferidae stehli, 1954 subfamily marginiferinae stehli, 1954 genus spinomarginifera helica abich,1878 pl1, ig 2a-2e. description: shell shape was concave – convex, its width beinggreater more than its length.raised umbo.pedicle was little and compressed and the ears not distinct. sulcus waas replaced, brachial valve depressed and edged at the front. rear decoration was thorny. the shell was 1 cm thick, 1.7 cm long and 1.8 cm wide. geographical spread: shahreza, jolfa, balade, zal suborder productidina waagen, 1883 superfamily productacea gray, 1840 subfamily marginiferidae stehli, 1954 subfamily marginiferinae stehli, 1954 genus araxilevis intermedius. abich, 1878 description: shell shape was convex – convex. pedicle valve had sulcus from umbo to the end of the shell. brachial valve was convex, hinge line curved, commissure line antiplicate, umbo raised, development line delicate and unicentric. the shell was 2.5 thick, 4.4 long and 4.2 wide. suborder productidina waagen, 1883 superfamily productacea gray, 1840 family productidae gray, 1840 genus: tschernyschewia typical stoyanow, 1910 late permian brachiopoda fauna in north-western iran description: the shell was convex – concave and large. umbo was convex and raised. hinge line was direct, the beak obvious, and decorations had radial lines and were thorny. the shell is 2 cm thick, 3.5 long and 4 cm wide. suborder productidina waagen , 1883 superfamily productacea gray , 1840 subfamily marginiferidae stehli , 1954 subfamily marginiferinae stehli , 1954 genus: araxilevis intermedius, (abich,1878) description: the shell was convex – convex. pedicle valve had sulcus from umbo to the end of the shell. brachial valve was convex, hinge line curved, commissure line antiplicate, umbo raised and , development line unicentric and delicate. the shell was 2.5 cm thick, 4.4 cm long and 4.2 cm wide. plate 1 cl bl al el dl c2 b2 a2 e2 d2 1: araxathyris quadrilobata( abich, 1878). 1a: dorsal, 1b: posterior, 1c: ventral, 1d: lateral, 1e: anterior. 2: araxathyris cf. protea (abich, 1878). 2a: ventral, 2b: posterior, 2c: dorsal, 2d: lateral, 2e: anterior. plate 2 1a 1b 1c 1d 1e 2a 2b 2c 2d 2e 1.araxathyris araxensis minor (grunt 1965).1a:ventral, 1b:dorsal, 1c:lateral, 1e:posterior. 2.orthotetina eusarkos (abich 1878). 2a:ventral, 2b:posterior, 2c:anterior, 2d:lateral, 2e:dorsal. plate 3 2a 2b 2c 2d 2e 1.: spinomarginifera helica (abich, 1878).1a: dorsal, 1b: lateral, 1c: ventral, 1d: anterior, 1e: posterior. 2.: araxilevis intermedius (abich, 1878). 2a: anterior, 2b: lateral, 2c: venteral,,2d: dorsal, 2e: posterior. plate 4 cl bl al el dl c2 b2 a2 1.: tschernyschewia typica (stoyanow,1910), 1a: anterior, 1b: posterior, 1c: lateral, 1d: ventral, 1e:dorsal. 2.: haydenella sp,. 2a: posterior, 2b: dorsal, 2c: lateral. plate 5 mostafa yousefirad, touran khamooshi and rahim shaabanian cl bl al cd b2 a2 2d 2e 1.: sarytchevinella diulfensis (or compresoproductus dahulfensis lamp shell) (stoyanow,1910). 1a: dorsal, 1b: venteral, 2c: posterior. 2.: permophricodothyris ovate (pavlova,1965). 2a: lateral, 2b: dorsal, 2c : posterior, 2d: anterior, 2e: ventral conclusion field study led to 27 species from 13 genera being examined, thereby revealsing that brachiopods havingglarge shells, belonging to the productidae and spiriferida families, likepermiansphaeroidothyris, araxilevis and orthotetina, were more frequent and the shells were bigger in the lower and middle parts of the sequence being studied. the shells become smaller towards the headof the sequence, that shapes having small shells belonging to the athyrisd and rhynconellida families (like werella and psedowerella) were more frequent during the late dzhulfian and dorashamian ages. the studies indicated that changes in brachiopod frequency and dimensions suggested their dependency upon deep environmental changes. based upon brachiopods, stratigraphic dispersion, the sequence being studied was determind to be late permian to dorashamian in terms of tethyans permian stage. acknowledgement the authors gratefully acknowledge the financial and other support of this research, provided by payame noor university,tehran, iran. authors would like to thank reviewers for their structured comments which improved the manuscript. references angiolini,l.,carabelli,l.,nicora,a.,crasquin-soleau,s.,marcoux,j. and rettori,r.2005, brachiopoda and other fossils from the permiantriassic boundary beds of the antalya nappes(sw taurus,turkey). geobios,40:715-729. baghbani, d. 1997. correlation chart of selected permian strata from iran. pemophiles, 30:24-25. baud, a. 2008. correlation of upper permian localities in the kuh-e-ali bashi area, nw iran: old collections, old and new data. permophiles 52,6–11. chen, z.,campia,m.,j.,shi,r.g.and kaiho ,k.,2005, post-extinction brachiopoda fauna from the late permian wuchiapingian coal series of south china ,acta paloentologica polanica, 50(2):343-363. fantini sestini ,n., 1965,the geology of the upper djaerud and lar valley (north iran); ii.paleontology;bryozoans , brachiopods and mollusks form ruteh limestone (permian). riv.ital .paleont .stratiger.,71(3):773-787. fantini sestini ,n.1966,milano-brachiopods from the upperpermian nessen formation(north iran)-v.72-n.4. fantini sestini,n.and glaus,m.,1966, brachiopods from the upper permian nessen formation (north iran):riv.ital.paleont stratigr.v.72,n4,p.887.923. iranian-japanes reserch group,1981, the permian and the lower triassic systems in abadeh region.central iran-mem.fac.sci.,kyoto uni v.,so .p.61-133. khamooshi, t,2012, biostratigaphy and chemostratigraphy of permotriassic sequence in the zal section in the south julfa,estherrn azarbaijan province,payame noor university,pp 182. kozur, h.w. 2007. biostratigraphy and event stratigraphy in iran around the permian-triassic boundary (ptb); implications for the causes of the ptb biotic crisis; environmental and biotic changes during the paleozoic-mesozoic transition. global and planetary change 55, 155–176. partoazar, h.,1997, permian system in iran.geological survey of iran,p.1-340. shabanian. r. and bagheri, m., 2008, permian in northwest of iran, permophiles, n.51, p 28-31. stoyanow,a.a.,1910,on the character of the boundary of paiaeozoic and mesozoic near djulfa:zapiski peterb.mineral. obshch.,ser.2,ch.47,p.61-135. stepanov ,l.d.,golshani, f.and stoklin ,j. 1967, upper permian and permian-triassic boundary in north iran.geological survey of iran ,report,12:72+xvplate. teichert,c.,kummel,b. and sweet, w., 1973, permian – triassic strata, kuhe-ali bashi , north western iran .bull.mus.zool.,145(8):359-472. late permian brachiopoda fauna in north-western iran in recent years, due to the increase in providers of orbit and clock corrections of satellites for data evaluation in real-time and post-processing the method of precise point positioning (ppp) using measurements of global navigation satellite system (gnss) and web-based online positioning services have become widespread. owing to some advantages, such as work-duration and cost-effectiveness, many of users have implemented ppp method instead of the traditional relative positioning method for several applications. on gnss applications the quality of satellite ephemerides products used for data evaluation is a significant factor that affects the results in post-processing solutions either applying relative or ppp methods on analyses. these products, classified as ultra-rapid, rapid and final orbits, are regularly provided by several national and international organizations to the users. in this paper, the accuracy of ppp method has been studied comparing the outcomes from various online web services using different software and satellite ephemerides products. for this purpose, three test points were established on a place with completely free satellite visibility (ac01) and on the other two places with partially (yc01) and vastly (kc01) prevention of satellite signals near and within a forest area at campus of davutpaşa of the yildiz technical university of istanbul. at these stations, static observations have been conducted with a time span of 6 hours on 4th may 2015. the dataset collected using topcon hiperpro receiver, a receiver for gps and glonass data, was evaluated manually by means of the bernese v5.2 (bsw) and gipsy-oasis v6.3 (gipsy) scientific software. moreover, the gnss data were also proceeded using six different web-based online services (auspos, opus, csrs-ppp, apps, gaps, trimble-rtx) with ultra-rapid, rapid and final satellite ephemerides products. for the station with free satellite visibility (ac01), the analyses of outcomes indicate a coordinate accuracy of 1 cm for web-based and manual data processing. that suggests relative good quality of orbit and clock corrections of satellites used by online data processing services for ppp. furthermore, for station with partially (yc01) prevention of satellite signals, manual solutions have approximately 1 cm coordinate accuracies for n, e and u components. moreover, when we compare the results among web-based services, auspos in relative solution and trimblertx in ppp solution provides the best results. here, one must emphasize that trimble-rtx is a web-based processing service for gnss data (gps, glonass, beidou, and qzss) and operates an own global network of approximately 100 stations around the world for this purpose. for station with vastly (kc01) prevention of satellite signals, the best solutions were provided by apps and csrs-ppp using ppp approach among the web-based services. here, trimble-rtx could not produce any solution. these results could demonstrate the benefit of multi constellation of gnss in the areas with limited satellite visibility because of increasing of the number of measurements. en los años recientes, debido al incremento de operadores satelitales de corrección de órbita y tiempo para la evaluación de información en tiempo real y de postproceso, se ha generalizado el método de posicionamiento preciso (ppp, del inglés precise point positioning) que utiliza medidas del sistema global de navegación por satélite (gnss, del inglés global navigation satellite system) y servicios web de posicionamiento en línea. debido a ventajas como el tiempo de ejecución y la relación costo-efectividad, muchos de los usuarios han implementado el método ppp en varias aplicaciones antes que abstract resumen keywords: global navigation satellite system (gnss), precise point positioning (ppp), relative positioning, forest environment, satellite ephemerides, web based positioning services palabras clave: sistema global de navegación por satélite, método de posicionamiento preciso, posicionamiento relativo, ambiente boscoso, efemérides satelitales, servicios web de posicionamiento accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n4.59496 comparación de precisión del método de posicionamiento preciso frente al posicionamiento relativo a través de diferentes productos de efemérides satelitales cerca y bajo un ambiente boscoso earth sciences research journal earth sci. res. j. vol. 20, no. 4 (december, 2016): d1 d9. taylan ocalan, bahattin erdogan, nursu tunalioglu, utkan mustafa durdag yildiz technical university, department of geomatic engineering, davutpaşa campus, 34220esenler, istanbul, turkey author for correspondence: nursu tunalioglu (e-mail: ntunali@yildiz.edu.tr). g e o d e sy d2 taylan ocalan, bahattin erdogan, nursu tunalioglu, utkan mustafa durdag 1. introduction global navigation satellite system (gnss) has been routinely used in several scientific and engineering applications including positioning, navigation and timing systems for many years. traditional relative positioning method has been handled for gnss-based positioning applications when high precision and accuracy are required. however, precise point positioning (ppp) method, a state space solution to the processing of gnss data acquired from a single receiver, has become an alternative to the relative positioning method by providing high absolute accuracy (gao and shen, 2001; rizos et al., 2012). traditionally, in all of the relative positioning methods depend on gnss technique, simultaneous observations should be made at one or more reference points. in other words, minimum two or more gnss receivers should be occupied at; one reference station, whose coordinates are known, and the other/s should be established to the point/s, whose coordinates will be estimated. however; in ppp technique to make observations on one point, whose coordinates will be estimated is enough for the estimation with one gnss receiver using absolute positioning approach. with regard to this, ppp technique allows a significant advantage to the users in terms of time and budget saving (ocalan, 2015). many factors such as preferred positioning method (relative or absolute), survey mode (static or kinematic), observation duration, equipment, signals and codes, data processing algorithms, reference receiver/receivers utility, satellitereceiver geometry, post-processing analyses or real-time applications provide different levels of performance for gnss positioning (rizos et al., 2012). several error sources affect positioning accuracy in gnss surveys. today relative techniques provide better solutions than ppp technique in terms of point positioning accuracy. the basic reason of this is that the effects of satellite orbit errors in relative techniques are lower than ppp technique. moreover, clock errors can be eliminated in relative techniques using double differencing of phase measurements. however, these effects are quite significant and dominant error sources in ppp technique. although relative positioning method removes some of these error sources either by applying phase differences techniques or by modeling of them, no modelled errors occurred in the forested areas, where especially satellite vision is limited, and orbital errors have significant effects on solutions (hoffmanwellenhof et al., 2001). this is a significant problem for ppp method. since precise satellite orbit and clock information are used, several error models are inevitable for ppp method. satellite visibility problem is the critical issue that should be taken into consideration for selection of the point location. it is required to ensure that there should not be any blockage or limitation of signals transmitted from satellite to receiver while establishing the points. this may cause problems at some of the engineering applications. at the applications where especially are performed around the forested areas, satellite vision has significant effects on positioning accuracy (naesset, 1999; pirti, 2005; tachiki et al., 2005; hasegawa and yoshimura, 2007; pirti, 2008). the other significant problem is orbital errors. several organizations such as igs, jpl, code and nrcan present several data and products freely to the users to be used for decreasing or eliminating the orbital errors. these products are classified in terms of their accuracies as broadcast, ultra-rapid, rapid and final. igs also presents some other standardized products such as satellite and station clocks, ionospheric and tropospheric parameters, besides satellite ephemerides. the principle of gnss method is based on resection problem using satellites, so satellite ephemerides are crucial on analyses. these orbit parameters have different levels of accuracy as seen in table 1. therefore, accuracies of the results have been altered according to the used product. el método tradicional de posicionamiento relativo. en las aplicaciones gnss, la calidad de los productos de efemérides satelitales utilizados para la evaluación de información es un factor significante que afecta los resultados en las soluciones postproceso sea con la aplicación de métodos relativos o ppp en los análisis. estos productos, clasificados como ultrarrápidos, rápidos y de órbita final, los proveen a los usuarios diferentes organizaciones nacionales e internacionales. en este artículo se estudia la exactitud del método ppp en comparación con los resultados de varios servicios web que utilizan diferentes productos de software y de efemérides satelitales. con este objetivo se establecieron tres puntos de evaluación: el primero en un lugar con completa visibilidad satelital (ac01) y los otros dos con parcial (yc01) y reducida (kc01) señal satelital, en los alrededores y al interior de una zona boscosa del campus de davutpaşa , en la universidad técnica yildiz de estambul. en las estaciones se llevaron a cabo observaciones estáticas con intervalos de seis horas el 4 de mayo de 2015. la información se recolectó con el programa topcon hiperpro, un receptor de gps y glonass (sistema gnss desarrollado por la unión soviética), y se evaluó manualmente con el software científico bernese v5.2 (bsw) y gipsy-oasis v6.3 (gipsy). además, la información también se procesón con seis servicios web en línea (auspos, opus, csrs-ppp, apps, gaps, trimble-rtx) con productos de efemérides ultrarrápidas, rápidas y finales. para la estación con completa visibilidad satelital (ac01), el análisis de los resultados indican una precisión coordinada de 1 cm en los procesamientos web y manual. esto sugiere una buena calidad de las correcciones satelitales de órbita y tiempo usadas por los servicios de procesamiento de información en línea con el método ppp. además, para la estación con señal parcial (yc01), las soluciones manuales tienen aproximadamente 1cm de coordinación en la precisión para los componentes norte y este. luego, cuando se comparan los resultados entre los servicios web, los mejores resultados fueron de auspos en soluciones relativas y trimble rtx en el método ppp. es importante recalcar que trimble rtx es un servicio de procesamiento web para información gnss (gps, glonass, beidou y qzss) y opera una red global propia de unas 100 estaciones en el mundo para este fin. para la estación con una reducida señal satelital (kc01), las mejores soluciones fueron ofrecidas por apps y csrs-ppp con el método ppp entre los servicios web. en este caso, trimble-rtx no puede producir ninguna solución. estos resultados demuestran los beneficios de la multiconstelación gnss en las áreas con visibilidad satelital limitada gracias al incremento del número de medidas. table 1. gps/glonass satellite ephemerides (url-1, url-4 and url-6) how to cite item ocalan, t., erdogan, b., tunalioglu, n., & durdag, u. m. (2016). accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/ under forest environment. earth sciences research journal, 20(4), d1-d9 doi:http://dx.doi.org/10.15446/esrj.v20n4.59496 record manuscript received: 09/08/2016 accepted for publication: 09/12/2016 d3accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment in this study, it is aimed to investigate the accuracy of ppp method versus relative positioning especially around or under the forested areas (limited satellite visibility) considering the developments of ppp method and webbased online services. for this purpose, gnss data collected at three points with long observational durations have been examined for static positioning and analyzed by different software and web-based online services by selecting different satellite orbit products (ultra-rapid, rapid and final). these dataset was first processed manually using gipsy-oasis v6.3 software in ppp mode with jpl products, and then analyzed by the various web-based online ppp services, namely csrs-ppp, apps, gaps and trimble-rtx. moreover, the related dataset was processed manually by bernese v5.2 software using code and igs products (dach et al., 2015), and then evaluated by web-based online services namely auspos and opus. hereby, we investigate the accuracy of ppp method versus relative positioning according to the outcomes of analysis, especially for the points established around the forested areas. methods mentioned in this study use different satellite ephemerides provided from various services. auspos, opus, gaps and csrs-ppp services use igs products; apps uses jpl products and trimble rtx uses its own orbit products. 2. ppp technique versus relative positioning gnss users prefer relative positioning method in surveying applications if high accuracy is needed. all gnss methods depending on relative positioning principle require simultaneous observations occupied at least one reference station whose coordinates are well known. therefore, minimum two receivers should be used on surveys: one is busy at the reference station and the other is occupied at the point whose coordinates will be determined. the primary factors for point positioning accuracy are the baseline length between two receivers and the observation duration. in this context, establishing continuously operating reference stations (cors) networks has a significant contribution to the relative positioning (abd-elazeem et al., 2011; rizos et al., 2012). relative positioning determines the point coordinates by using double differences on phase measurements as given below with eq.1 (hoffmanwellenhof et al., 2001): ∇δφ t =∇δr(t, t-τ) + ∇δds(t-τ) ∇δdiono + ∇δdtropo+∇δλν + ∇δε(φ) (1) where ∇δ is the double difference operator at the time of receiving data, φ is the phase measurement, (t) is the time of receiving data, (t-τ) is the satellite time, τ is the travel time from the satellite to the receiver, r(t, t-τ) is the true geometric range, ds is the orbital prediction error, diono and dtropo are the ionospheric and tropospheric errors, respectively, λ is the wavelength, ν is the integer phase ambiguity and ε is the noise components. scientific software is used for long baselines in post processing solutions of relative positioning, and commercial software is used for short baselines. the list of the most commonly known web-based online services can be seen in table 2. table 2. web based online services using relative solution approach (url-2, url-3) ppp method is the particular case of traditional absolute point positioning approach that became widespread after the establishment of global gnss networks with permanent stations. for a worldwide positioning in sub-meter level, an accurate determination of corrections of satellite orbits and clocks is possible using data of a global gnss network. data collection with a dual frequency receiver at the point, whose coordinates will be computed accurately, is enough in ppp method for determining high accurate position. so, by using code and carrier phase observations with a doublefrequency receiver that are utilized by the un-differenced and ionosphere-free combinations, decimeter to centimeter level point positioning accuracies may be achieved due to observation durations (zumberge, 1997; gao and shen, 2001; kouba and héroux, 2001; gao and shen, 2002; rizos et al., 2012). the accuracies of orbit and clock correction products and error models to be used in determination have significant importance on the point positioning accuracies to be achieved in ppp method. therefore, considering the errors and biases separately shown in table 3 for ppp depending on the quality of the application is the key factor to increase the position accuracy. table 3. biases and errors that need to be applied for in typical ppp and differential gnss positioning techniques (rizos et al. 2012) in typical ppp method, to reduce the effect of ionospheric errors, double frequency gnss observation models (code observations and ionosphere-free carrier phase observations) have been adopted. the first effective use of the functional model in ppp has been discussed and represented in the study of zumberge et al. (1997). kouba and heroux (2001), as given in equations 2 and 3, has presented mathematical observation models of typical ppp ℓp = ρ + c(dtdt) + tr + εp (2) ℓф = ρ + c(dt-dt) + tr + nλ + εф (3) here ℓp (p3) is the ionosphere-free combinations of p1 and p2 code observations, (p3)=(2.546p1-1.546p2), ℓф (l3) is the ionosphere-free combinations of l1 and l2 carrier phase observations, (l3)=(2.546 λ1 ф11.546 λ2 ф2), ρ is the geometric distance between satellite and receiver antennas in meters, c is the speed of light in meters per second, dt is the receiver clock offset according to gps time in seconds, dt is the satellite clock offset according to gps time in seconds, tr is the signal delay in atmosphere (prior tropospheric delay) in meters, n carrier-phase ambiguities on l1 and l2 frequencies in cycles, respectively. λ1, λ2, λ are the carrierphase wavelengths on l1, l2 frequencies, and combined l3 (10.7 cm) carrier phases, respectively, and εp, εф are the components of related survey noises including multipath and un-modeled errors. d4 taylan ocalan, bahattin erdogan, nursu tunalioglu, utkan mustafa durdag 3.2 data post-processing all data obtained from observations were converted to rinex format and quality check of data was done by bkg ntrip client (bnc) v2.11.1 software. later, gnss data were uploaded and processed on the web-based online services using ppp and relative approaches. ac01 has the best satellite visibility among three points, thus the positional dilution of precision (pdop) values are better than other points’ values (fig. 2 and 3). the pdop values and cycle slips for point yc01 are worse than the pdop values for ac01 (fig. 2 and 4). the worst pdop values and satellite visibility are belonged to the point kc01 (fig. 2 and 5). here, improvement of the satellite visibility spatial geometry brings the pdop values are decreased. in figure 4, the cycle slips in satellite visibility is seen in south-west direction for yc01 where the forest area (with approximately 10 m-15 m tree heights) are placed. moreover, as seen in figure 5, for kc01 that has the limited satellite visibility, the satellite visibility is only traced in north-west direction limitedly. despite the fact that traditional ppp method has had several advantages aforementioned, the most significant disadvantage of the method is to require long convergence time to solve the carrier phase integer ambiguities. to resolve an ambiguity float solution a convergence time up to 20 minutes is expected for centimeter-level positioning accuracy depending of accuracy of apriori coordinates of station and the number of received satellites, (rizos et al., 2012). therefore, another significant factor to increase the position accuracy is the growth of the observation duration. since the usage percentage of ppp method is growing due to supports and gnss products provided by several organizations such as igs, jpl, nrcan, bkg and code to the users, many software algorithms for ppp in real-time and post-process have been developed (i.e. bernese, gipsy-oasis, bnc, rtk-lib, panda, wappp). in addition to this, web-based ppp services, csrs-ppp, gaps, apps, magic-gnss, trimble centerpoint rtx, etcetera, listed in table 4, offer to the users the possibility of processing of uploaded data (el-mowafy, 2011; ocalan and alkan, 2013; ocalan et al., 2013; alkan and ocalan, 2013; alkan et al., 2013). similarly, these services constantly improve to provide a userfriendly service without any fees for usage. table 4. web-based services using ppp solution approach (url-4, url-5, url6, url-7 and url-8) 3. case study 3.1 test area and control points characteristics to investigate the accuracy of the relative and ppp methods by using different satellite ephemerides and software with different satellite visions around the forested areas, three test points were established at campus of davutpaşa, yildiz technical university, istanbul. points namely; ac01, yc01 and kc01 were installed in the open area, close to the forest and within the forest area, as shown in figure 1, respectively. the average height of the trees is approximately within 10 m-15 m in the forest area. topcon hiperpro gnss (gps+glonass) geodetic receiver was used for collecting the data with a time span of 6 hours from 08:30 am to 02:30 pm (utc) on the 4th may 2015 (day of year: 124) at static mode. during the view, the antenna slope-heights were measured carefully minimum from two sides, and then vertical antenna heights from antenna reference point (arp) were computed by using antenna parameters. figure 1. test area and the points figure 2. pdop plots for points ac01,yc01, kc01 figure 3. analysis sky plots of point ac01 d5accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment figure 4. analysis sky plots of point yc01 figure 5. analysis sky plots of point kc01 scientific software and web-based online services were used in data processing and analyzing stages. for manual solutions, bernese v5.2 and gipsy-oasis v6.3 scientific software were used by applying relative positioning approach and ppp method, respectively. for webbased data processing, auspos, opus (relative technique), csrsppp, apps, gaps and trimble-rtx (ppp technique) were preferred. in table 5, a brief description has been given according to used gnss satellite ephemerides for all computations of post processing web-based online services and manual solutions. table 5 types of gnss satellite ephemerides used in post processing the data processing strategy of the gaps v5.5.0 was applied to solve the gps observations by using ultra-rapid and rapid orbit datasets. during the analyzing process, the updated version of gaps v5.9.0 has been presented to the users, thus data were processed according to the final orbit datasets with updated version. datasets were uploaded to the web-based online services within 17 hours for ultra-rapid solutions, within 3 days for rapid solutions and approximately 14 days later for final solutions. to compare the solutions, the outcomes of the relative solution is considered as reference coordinates of stations (target coordinates) computed by bernese v5.2 software using final orbit products. stations bucu, crao, ankr, sofi and nico, involved in igs global network, were settled as datum stations for bernese v5.2 solutions. here, all computed coordinates were based on the igs realization of itrf 2008 reference frame at observation epoch. in the evaluation process, the offsets and correction values for the antenna phase variations of the antenna types available from the us national geodetic survey were used. in the first processing step, single point positioning has been performed by using ionosphere-free (l3) observable, the receiver clock errors have been eliminated, and then baselines in the network have been pre-determined. cycle slips and outliers have been checked by using triple differences. after this process, the outliers have been removed. later, ambiguity parameters have been solved. in order to solve the ambiguity parameters, quasi ionosphere free (qif) strategy has been applied. consequently, precise cartesian coordinates (x,y,z) of the points in itrf 2008 reference frame at related observation epoch have been computed. we processed our gps data using gipsy-oasis v6.3. it is an automated, fast, ultra-precise high precision gps data processing software package for ppp rather than relative positioning (zumberge et al., 1997). all of the cartesian coordinates (x,y,z) provided by the systems and determined manually in itrf2008 datum were then converted to the local topocentric coordinates (north, east and up) and differences for all components were calculated as reads by equation 4. difference (δ) = solution – reference bernese v5.2 solution (4) 3.3 results and analysis the point ac01 was established on the area with free satellite visibility. coordinate differences of ac01 estimated by using different orbit products for north (n), east (e), up (u) directions and their standard deviations are listed in tables 6 and 7 and illustrated in figures 6, 7 and 8, respectively. the horizontal-axis shows the solution methods and verticalaxis corresponds the coordinate differences and standard deviations derived by mentioned orbit ephemerides. the solutions estimated manually by scientific software indicate that the differences computed with bernese v5.2 using ultra-rapid and rapid orbit products are less than 1 mm in all coordinate directions. this proves the good quality of both orbit products. in contrary to this, the results obtained by gipsy-oasis v6.3 software are in the range of 1 and 4 mm for n and e directions and the difference amounts to maximum 8.6 mm for u direction of coordinates. these differences to the outcomes of the bernese v.5.2 mainly accumulate due to the different modelling and default values for the signal delay in the troposphere. when the results mentioned above compare to their standard deviations, only the coordinate differences of u component for rapid ephemerides are greater than the standard deviations. moreover, an increase has been occurred on coordinate differences when comparing the solutions of ultra-rapid orbit product and rapid orbit product. then, a decrease has been determined when comparing the results between rapid orbit solution and final orbit solution. these differences have been in limits of their standard deviation values. when the results of web-based service for the ac01 point are analyzed, it is found that relative methods give better solution than ppp method. the maximum difference between n and e directions for auspos is determined as 2.0 mm. on the u direction, the difference is increased to 12.7 mm. in auspos system, a decrease has been occurred on coordinate differences, when comparing with ultra-rapid orbit solution, rapid orbit solution and final orbit solution, respectively. when the opus system has been analyzed, maximum 17.1 mm difference has been determined on n and e directions. all values are lower than standard deviation values. in spite of this, differences on u direction are greater and all of these values are bigger than their standard deviations. according to the results, it is determined that coordinate differences obtained from ultra-rapid orbit products in opus system are closer to the reference coordinates. d6 taylan ocalan, bahattin erdogan, nursu tunalioglu, utkan mustafa durdag table 6. coordinate differences for point ac01 table 7. standard deviations of the coordinate differences for point ac0 figure 6. coordinate difference and standard deviation of point ac01 for n direction figure 7. coordinate difference and standard deviation of point ac01 for e direction figure 8. coordinate difference and standard deviation of point ac01 for u direction when web-based ppp service solutions have been examined for ac01 point, it is seen that maximum difference on n and e directions is as 13.2 mm for csrs-ppp, apps and trimble-rtx services. among them, especially apps service has the lowest differences. for u direction, all of the service solutions except trimble-rtx give the highest difference. trimble-rtx service gives approximately similar solutions for all type orbit products. however, although gaps service provides worse results for ultra-rapid and rapid orbit products, it produces better solutions especially for u direction than the other services while using final orbit products. the point, namely yc01 was selected on the relatively forest area. the observation duration was 6 hours. coordinate differences of n, e, u directions determined by using different orbit products and standard deviations are presented in tables 8 and 9, respectively. moreover, the graphical representation of coordinate differences and standard deviations are given in figures 9, 10 and 11 for point yc01. table 8. coordinate differences for point yc01 when the statistical results of yc01 point have been examined, it is seen that the standard deviations have been increased due to the decrease of satellite vision. when the results of web-based services have been examined, it is seen that there is a deviation from reference coordinates on u direction, especially when comparing with ultra-rapid and rapid orbit products. at online services, there is progress on results when comparing with ultra-rapid orbit products and final orbit products. when considering the coordinate differences of n and e directions and standard deviations of them, the best results have been provided by apps service. moreover, if n, e and u directions have been considered together, trimble-rtx service provides the best results. table 9. standard deviations of the coordinate differences for point yc01 figure 9. coordinate difference and standard deviation of point yc01 for n direction d7accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment figure 10. coordinate difference and standard deviation of point yc01 for e direction figure 11. coordinate difference and standard deviation of point yc01 for u direction as seen in figure 10, gaps service solutions for ultra-rapid ephemerides provide the worst solutions. moreover, csrs-ppp and gaps services solutions for u direction provide great coordinate differences. the point, namely kc01 was selected under the forested area. the observation duration is 6 hours. coordinate differences of n, e, u directions determined by using different orbit and standard deviations are given in tables 10 and 11, respectively. moreover, the graphical representation of coordinate differences and standard deviations are given in figures 12, 13 and 14 for point kc01. table 10. coordinate differences for point kc01 table 11. standard deviations of the coordinate differences for point kc01 the kc01 point has the worst satellite vision among the points. problems on solving the carrier phase integer ambiguity were occurred during the analyses. trimble-rtx could not produce any solution on point kc01. the most appropriate results for ultra-rapid and rapid orbit solutions were achieved by bernese v5.2 scientific software. when the differences and standard deviations of them determined from web-based services have been examined, it is found that the deviations are greater than the results of services, which use relative model. results obtained from apps and csrs-ppp services among the other web-based services give efficient solutions. figure 12. coordinate difference and standard deviation of point kc01 for n direction figure 13. coordinate difference and standard deviation of point kc01 for e direction figure 14. coordinate difference and standard deviation of point kc01 for u direction as seen in figures 12, 13 and 14, the gaps service for ultra-rapid ephemerides provides the worst solutions among the web-based online ppp services. in addition, opus service solutions for all directions provide great coordinate differences among the web-based online relative services. d8 taylan ocalan, bahattin erdogan, nursu tunalioglu, utkan mustafa durdag 4. conclusions and recommendations satellite vision in gnss survey has a significant importance for the accuracy of the estimated coordinates. this constraint has been encountered especially in surveys performed near the forested area. moreover, accuracy of the products used in studies affects the expected results. after the survey stage, due to the determination process, one of the products, ultra-rapid, rapid or final orbit products can be used. today, besides commercial and scientific software, web based positioning services have been used for analyzing gps/gnss data, which are free. due the development and common use of these services, ppp method for next generation positioning has come to the forefront. in this study, the accuracy of ppp method versus relative method has been investigated according to the different satellite orbit products where satellite vision is limited. for this purpose, analyses have been performed with six different webbased services, then performances of them have been compared. for manual solutions, the estimation was achieved by bernese v5.2 and gipsy-oasis v6.3 scientific software. post processing solutions for csrs-ppp and trimble-rtx web-based online services use gps+glonass observations. however, the other web-based online services and manual solutions in post processing use gps only observations. in survey areas, where the satellite visibility is limited such as forest regions as illustrated in this study, the use of gps+glonass observations in post processing could be an advantage due to the increase of number of observations. but, this is not the unique parameter to improve the point positioning accuracy in post processing solutions. the analyses of outcomes show that web-based services (except gaps and trimble-rtx) and manual solutions used in this study estimate accuracy of coordinate components of north and east as approximately 1 cm level (in all cases when ultra-rapid, rapid and final satellite orbit products were used) for the station with free satellite visibility (ac01). for ac01, considering the ephemerides used in post processing solutions, the significant improvement on coordinate components are provided by gaps service in respect to use of ultra-rapid, rapid and final products. for the point yc01 located near the forest area where the satellite vision decreases or partially disturbed, the positioning accuracy decreases. here, manual solutions processed by scientific software (bernese and gipsy-oasis) give the best results. among the web-based online services applying relative model approach, auspos offers the best solution for partially prevention of satellite signals. furthermore, the ppp-solution of the trimble-rtx delivers a better positioning accuracy than the outcomes of the other web-based services using ppp approach for yc01. this outcome for yc01 estimated by trimble-rtx, using ppp solution and its own satellite ephemerides products could be related to use of multi constellation of gnss in the areas with limited satellite visibility because of increasing of the number of measurements. in contrary to this, trimble-rtx could not produce any solution for point kc01 (vastly prevention of satellite signals). here, the best solution was provided by apps service using ppp approach. among the web-based online services applying relative model approach, auspos offers the best solution for kc01. consequently, using scientific software manually on the analyses provides advantages to the users to achieve high accuracy on results in all conditions. the fundamental reason of it is that the user has an option to interfere the estimation and analyzes stages during the process. in spite of this, the professional use of scientific software requires a long-term experience. however, web-based services represent basic internet based interface, which users of all level may easily use. they produce fast and easy solutions for post-processing stage at positioning applications. however, several factors affect the positioning accuracy under the different satellite visibilities. these factors can be listed as; used satellite observations (gps only, gps+glonass etc.), the solution algorithms (relative or ppp), satellite ephemerides product types (ultra-rapid, rapid and final) and the other correction models, that the users should take into consideration of them acknowledgements we are grateful to organizations, which present web-based online services (opus, auspos, csrs-ppp, gaps, apps, trimble-rtx) and thank to nasa’s jet propulsion laboratory for providing gipsy oasis ii software. we are grateful to anonymous referees for their helpful comments to the manuscript. references abd-elazeem, m., farah, a. & farrag, f. a. (2011). assessment study of using online (csrs) gps-ppp service for mapping applications in egypt. journal of geodetic science. 1(3): 233-239. doi: 10.2478/ v10156-011-0001-3 alkan, r. m. & ocalan, t. (2013). usability of the gps precise point positioning technique in marine applications. journal of navigation. 66(4): 579588. doi: http://dx.doi.org/10.1017/s0373463313000210 alkan, r.m., ocalan, t. & tunalıoğlu, n. (2013). kinematic ppp performance of low-cost gps receiver using online post-processing service. gnss precise point positioning workshop: reaching full potential, ottawa, canada, 12-14 june. dach, r., lutz, s., walser, p. & fridez, p. (2015). user manual of the bernese gps software version 5.2. astronomical institute, university of berne, november. el-mowafy, a. (2011). analysis of web-based gnss post-processing services for static and kinematic positioning using short data spans. survey review, 43(323): 535-549. doi:10.1179/00396261 1x13117748892074 gao, y. & shen, x. (2001). improving ambiguity convergence in carrier phase-based precise point positioning. proceedings of ion gps2001, salt lake city, usa, september 11-14. gao, y. & shen, x. (2002). a new method for carrier-phase–based precise point positioning, navigation. journal of the institute of navigation, 49(2): 109-116. doi: 10.1002/j.2161-4296.2002.tb00260.x hasegawa, h., & yoshimura, t. (2007). estimation of gps positional accuracy under different forest conditions using signal interruption probability. journal of forest research, 12: 1-7. doi: 10.1007/ s10310-006-0245-4 hoffman-wellenhof, b., lichtenegger, h. & collins, j. (2001). gps theory and practice. fifth revised edition. springer wien newyork, isbn: 3-211-83534-2. kouba, j. & héroux, p. (2001). precise point positioning using igs orbit and clock products. gps solutions. 5(2): 12-28. doi: 10.1007/ pl00012883 naesset, e. (1999). point accuracy of combined pseudo-range and carrier phase differential gps under forest canopy. canadian journal of forest research. 29: 547-553. doi: 10.1139/x99-021 ocalan, t. (2015). gps precise point positioning (gps-ppp) technique solutions i̇n gnss networks, phd. thesis, yildiz technical university, graduate school of natural and applied sciences, december, 2015. ocalan, t., erdogan, b. & tunalioglu, n. (2013). analysis of web-based online services for gps relative and precise point positioning techniques. boletim de ciencias geodésicas. 19(2): 191-207. doi: 10.1590/s1982-21702013000200003 ocalan, t. & alkan, r. m. (2013). performance analysis of web-based online precise point positioning (ppp) services for marine applications. journal of the arab institute of navigation. (29): 24-29. pirti, a. (2005). using gps system near the forest and quality control. survey review, 38 (298): 286-298. doi:10.1179/sre.2005.38.298.286 pirti, a. (2008). accuracy analysis of gps positioning near the forest environment. croatian journal of forest engineering. 29(2): 189-201. rizos, c., janssen, v., roberts, c. & grinter, t. (2012). precise point positioning: is the era of differential gnss positioning drawing to an end? fig working week 2012, rome, italy, 6-10 may. d9accuracy investigation of ppp method versus relative positioning using different satellite ephemerides products near/under forest environment tachiki, y., yoshimura, t., hasegawa, h., mita, t., sakai, t. & nakamura, f. (2005). effects of polyline simplification of dynamic gps data under forest canopy on area and perimeter estimations. journal of forest research, 10: 419-427. doi: 10.1007/s10310-005-0161-z zumberge, j.f., heflin, m.b., jefferson, d.c., watkins, m.m. & webb, f.h. (1997). precise point positioning for the efficient and robust analysis of gps data from large networks. journal of geophysical research. 102(b3): 5005-5017. url url-1: international gnss service (igs) web page, http://www.igs.org (last accessed february 2016) url-2: australian online gps processing service (auspos) web page, http://www.ga.gov.au/bin/gps.pl (last accessed february 2016) url-3: online positioning user service (opus) web page, http://www.ngs. noaa. gov/opus/ (last accessed february 2016) url-4: canadian spatial reference system-precise point positioning (csrs-ppp) web page, http://webapp.geod.nrcan.gc.ca/geod/toolsoutils/ppp.php (last accessed february 2016) url-5: gps analysis and positioning software (gaps) web page, http:// gaps. gge.unb.ca/ (last accessed february 2016) url-6: automatic precise positioning service (apps) web page, http:// apps.gdgps.net/ (last accessed february 2016) url-7: magicppp-precise point positioning solution (magicgnss) web page, http://magicgnss.gmv.com/ppp/ (last accessed february 2016) url-8: trimble centerpoint rtx post-processing service web page, http:// www.trimblertx.com/ (last accessed february 2016) keywords: tight sandstone; pore size structure; sem; porosimetry; diagenetic effects. palabras clave: areniscas compactas; estructura del tamaño de poros; microscopio electrónico de barrido; porosimetría; efectos diagenéticos. how to cite item wang, w., yu, c., zhao, l., xu, s., & gao, l. (2020). combing sem and mercury intrusion capillary pressure in characterization of porethroat distribution in tight sandstone and its modification by diagenesis: a case study in the yanchang formation, ordos basin, china. earth sciences research journal, 24(1), 19-28. doi: https://doi.org/10.15446/esrj.v24n1.84838 determining the characteristics of pore-throat structures, including the space types present and the pore size distribution, is important for the evaluation of reservoir quality in tight sandstones. in this study, the results of various testing methods, including scanning electron microscopy (sem), pressure-controlled porosimetry (pcp) and rate-controlled porosimetry (rcp), were compared and integrated to characterize the pore size distribution and the effects of diagenesis upon it in tight sandstones from the ordos basin, china. the results showed that reservoir spaces in tight sandstones can be classified into those with three types of origins (compaction, dissolution, and clay-related) and that the sizes and shapes of pore space differ depending on origin. considering the data obtained by mercury injection porosimetry and the overestimation of pore radii by pressure-controlled porosimetry, the full-range pore size distribution of tight sandstones can be determined by combining data from pcp with corrected rcp data. the pore-throat radii in tight sandstone vary from 36 nm to 200 μm, and the distribution curve is characterized by three peaks. the right peak remains similar across the sample set and corresponds to residual intergranular pores and dissolution pores. the middle and left peaks show variation between samples due to the heterogeneity and complexity of nano-scale throat bodies. the average micro-scale pore content is 33.49%, and nano-scale throats make up 66.54%. the nano-scale throat spaces thus dominate the reservoir space of the tight sandstones. compaction, dissolution, carbonate cementation and clay cementation have various effects on pore-throats. compaction and carbonate cementation decrease the pore body content, pore-bridging clay cementation decreases the throat space content, and simultaneously, pore-lining clay cementation preserves pore space. abstract combining sem and mercury intrusion capillary pressure in the characterization of pore-throat distribution in tight sandstone and its modification by diagenesis: a case study in the yanchang formation, ordos basin, china combinación de los métodos de microscopio electrónico de barrido e intrusión capilar de mercurio a presión en la caracterización de la distribución de gargantas de poro en areniscas compactas y su relación con la diagénesis: caso de estudio en la formación yanchang, cuenca ordos, china issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.84838 para evaluar la calidad de yacimiento en areniscas compactas es importante determinar las características de las gargantas de poros, entre ellas las clases de espacio presentes y la distribución de los tamaños de poros. en este trabajo se comparan e integran los resultados de varios métodos de prueba, entre ellos el microscopio electrónico de barrido (sem), porosimetría de presión controlada (pcp), y porosimetría de velocidad controlada (rcp), para caracterizar la distribución de los tamaños de poros y los efectos de la diagénesis en las areniscas compactas de la cuenca ordos, en china. los resultados muestran que los espacios de yacimiento en areniscas compactas se pueden clasificar de acuerdo con tres tipos de origen (compactación, disolución y arcilloso) y que los tamaños y formas del espacio de poros difiere de acuerdo con el origen. de acuerdo con la información obtenida por la porosimetría con inyección de mercurio y la sobreestimación de los radios de poros por la porosimetría de presión controlada, se puede determinar el rango completo de la distribución del tamaño de poros en areniscas compactas al combinar la información del método pcp con la información corregida del método rcp. el radio de la garganta de poro en areniscas compactas varía de 36 nm a 200 μm; la curva de distribución se caracteriza por tres picos. el pico de la derecha permanece similar a lo largo de todas las muestras y corresponde a los poros intergranulares residuales y a los poros de disolución. el pico central y el de la izquierda muestran la variación entre las muestras debido a la heterogeneidad y la complejidad de los cuerpos de garganta a nano-escala. el promedio del contenido poroso a micro-escala es de 33.49 %, y las gargantas a nano-escala ascienden al 66.54 %. esto significa que los espacios de gargantas a nano-escala dominan el espacio del yacimiento en areniscas compactas. la compactación, disolución, cementacion de carbonato y cementación arcillosa tienen varios efectos en las gargantas de poros. la compactación y la cementación de carbonato decrecen el contenido del cuerpo del poro. el puente arcilloso de poros reduce el contenido del espacio de las gargantas. mientras que la cementación arcillosa de revestimiento de poros preserva el espacio entre los poros. resumen record manuscript received: 30/01/2019 accepted for publication: 12/06/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 19-28 wei wang1,*, caili yu2, le zhao2, shuang xu2, lei gao2 1college of chemistry and chemical engineering, yulin university, yulin, pr china 2petrochina changqing oilfield company, xi’an, pr china * corresponding author: wangwei_1988@yulinu.edu.cn g e o l o g y 20 wei wang, caili yu, le zhao, shuang xu, lei gao introduction with the gradual decline of hydrocarbon reserves in conventional reservoirs and the increasing global energy demand, the great potential of unconventional reservoirs has made them the focus of global oil production (ghanizadeh et al., 2015). tight sandstones are characterized by extremely low porosity and permeability, the presence of nano-scale throat space, and having undergone strong diagenesis (rezaee et al., 2012; zou et al., 2013), all of which influence the characteristics of seepage from tight reservoirs (jarvie et al., 2007; nelson, 2009; xi et al., 2016). therefore, a comprehensive understanding of the pore-throat distribution in tight sandstone and how it is affected by diagenesis is necessary for quality-evaluation and successful production. the pore size distribution (psd) in tight sandstones can be characterized by many techniques, such as thin section analysis, scanning electron microscopy (sem), x-ray computed tomography (ct), pressure-controlled porosimetry (pcp), rate-controlled porosimetry (rcp), nuclear magnetic resonance (nmr) and n2 adsorption, but these techniques all have limitations that hinder determination of the characteristics of pore-throats of a wide range of sizes. sem (fe-sem) and thin sections provide direct information on the morphological characteristics of reservoir space and can be used to determine the genetic origins of pore-throats, but they fail to provide quantitative data (klaver et al., 2012; loucks and ruppel, 2007). pcp can be used to obtain quantitative data on pore-throat radius from samples and effectively identify nanopores but fails to measure large pores due to the shielding effect (gane et al., 2004). rcp can examine pore spaces and throats spaces separately according to the oscillation of intrusion pressures (yuan and swanson, 1989) but cannot reveal the distribution of tiny throats due to the limited maximum intrusion pressure (6.22 mpa) (wang et al., 2018; zhao et al., 2015). ct can be used to reconstruct 3d images of pore-throats, but its application is restricted by its high cost and the inevitable error that arises in calculating pore-throat radius (mayo et al., 2015). nmr can be employed to obtain the pore size distribution through a transformation of the value of transverse relaxation time (daigle and johnson, 2016; gao et al., 2019). however, the test results need to be corrected, and the test range is limited by surface relaxivity (xiao et al., 2017a). nitrogen adsorption can only obtain information on nanopores in tight sandstones (yao and liu, 2012) stated thus, one single technique can only partially reveal the pore-throat distribution of a tight sandstone, so integration of several techniques appears to be necessary for characterizing their overall pore-throat distribution. therefore, scholars have combined pcp, rcp, nitrogen adsorption, and nmr to study the pore-throat distribution of tight sands by considering the limitations of the different tests (xiao et al., 2017b; yan et al., 2017; yao et al., 2010; zhang et al., 2017; zhao et al., 2015) this approach has expanded the range of pore sizes that can be investigated and has improved the accuracy of pore size characterization. rcp and pcp are based on the same principle, so the results from these two methods can be well-integrated to reveal the pore size distribution of tight sandstones. rcp overestimates the size of pore bodies (xiao et al., 2017b), however, so the data derived by rcp must be corrected to obtain accurate pore distribution data. the evolution of pore-throats is influenced by several factors such as the original rock composition, sedimentary environment and diagenesis. sedimentation forms the basis of the pore-throats, and diagenesis modifies them (waples, 2002; zhu et al., 2014). many scholars have studied the evolution of pore-throats during diagenesis. some of the conclusions that have been reached are as follows: (1) compaction and cementation reduce the space between gains and are the main diagenetic processes that reduce the pore-throat space in tight sandstone (hillier, 2000). (2) chemical dissolution increases the relative content of secondary pores and improves the physical properties of the reservoir (jia et al., 2018). however, less work has been done separating the effects of compaction and cementation on pores and throats because data are not available on the distribution of pore-throats of a wide range of sizes, which is essential for the study of pore-throat characteristics. in this study, we first analyze pore-throat structures based on thin sections and sem observations. pore and throat types are divided into those with three origins: compaction, dissolution and clay-dominated. pore-throats with different origins are shown to have different morphological characteristics. secondly, the full-range psd is established through a combination of pcp and corrected-rcp results. finally, the influence of various diagenetic processes on the pore-throat distribution of tight sandstones is discussed. samples and experiments seven cylindrical core plug samples with a diameter of 2.48 cm and a length of 5.00 cm were prepared. oil was removed from the samples with a mixed solution of benzene and alcohol, and they were then dried in a bake oven at 115°c for 12 h. the samples were analyzed via porosity and permeability tests and were then were divided into parts for further experiments. samples all of the samples were chang 7 tight sandstones, which from part of the upper triassic yanchang formation in the ordos basin (fig. 1). the chang 7 sandstones were deposited in a semi-deep and deep lake sedimentary environment and are dominantly fine-grained sandstone. chang 7 developed in the period when lakes were at their deepest, leading to the formation of highquality source rocks within it. the crude oil generated from the source rocks in chang 7 migrated immediately to the adjacent chang 7 tight sandstones as the rocks were maturing (zhang et al., 2009), making the chang 7 tight sandstone a tight oil reservoir. experimental methods the morphological characteristics of pores and throats were measured with an fei quanta 400 feg sem at 30 kv and 1.2 nm. each sample was first polished and then gold-coated to better capture the morphological characteristics of the pore-throats. an autopore 9420 mercury porosimeter was used for pcp. the maximum injection pressure was 200 mpa, corresponding to the ability to detect a pore radius of 36 nm. rcp was carried out with an aspe730 mercury porosimeter, and the rate of mercury intrusion was kept constant at a value of 0.001 μl/s. the maximum intrusion pressure was 6.8 mpa (1000 psi), corresponding to a pore radius of 0.1 μm. results lithological characteristics of tight sandstone the mineralogical compositions of the seven samples are shown in table 1. the dominant minerals are quartz, feldspar, clay and carbonate. the quartz content ranges from 31% to 43%, with an average value of 37%. the feldspar content varies from 40% to 51%, with an average value of 46%. the carbonate content is from 2% to 18%, with an average value of 8.7%. the clay content ranges from 5% to 15%, with an average value of 8.2%. the porosity values of the samples vary from 9.54% to 12.26%, with the average value being 10.79%. their permeability varies from 0.06×10-3 μm2 to 0.32×10-3 μm2, with an average value of 0.19×10-3 μm2. the tight sandstone thus has extremely poor petrophysical properties. the pores in the samples can generally be classified into three types on the basis of the sem and thin section images: residual intergranular pores, mineral dissolution pores and intercrystalline pores. residual intergranular pores are the major pore type in the tight sandstone and formed through compaction, cementation and dissolution (fig. 2a and b). mineral dissolution pores are generated by dissolution and were usually found in feldspar and debris (fig. 2c and d), though dominantly in feldspar. intercrystalline pores were mainly found in illite and chlorite (fig. 2e and f). the size and shape of the three types of pores vary significantly. the residual intergranular pores take the form of triangles and regular polygons, and their edges are generally clear and straight. the radii of residual intergranular pores are relatively large, ranging from 50 μm to 200 μm, and the connectivity between pores is good. the feldspar dissolution pores are irregularly shaped, and their sizes are generally <100 μm. the connectivity between feldspar dissolution pores is worse than between residual intergranular pores. intercrystalline pores range in size from 1.5 μm to 5 μm, and their connectivity is the worst of the three types. the throats in all of the samples were classified into three main types on the basis of the sem images and thin section images, namely compacted throats, dissolution throats and intercrystalline throats. with intense compaction and cementation, the primary pore space in the samples is gradually squeezed, and compacted, cement-infilled throats are formed. soluble mineral particles eroded by acidic fluids form dissolution throats, which are connected with dissolution pores. the intercrystalline pores in chlorite or illite form the intercrystalline throats. compacted throats are necking and range in radius from 0.8 μm to 4 μm (fig. 2g), while dissolution 21combining sem and mercury intrusion capillary pressure in the characterization of pore-throat distribution in tight sandstone and its modification by diagenesis the mercury injection curves for the tight sandstone samples can be divided into two types. j12, j13, j17 and j20 belong to type i. these mercury intrusion curves contain one platform. when the mercury injection pressure is < 7 mpa, the corresponding mercury saturation increases at a low rate with the increase in pressure, forming an obvious platform in the curve. when the mercury injection pressure is > 7 mpa, the corresponding mercury saturation increases at a higher rate. the threshold pressure of this type is the higher of the two types, with a mean value of 2.35 mpa. this indicates that the maximum pore space that allows the fluid to enter the sample is small. the curves for j5, j6 and j10 belong to type ii. there are two platforms in mercury intrusion curves of this type. when the mercury intrusion pressure is < 3mpa, a platform appears due to a slow increase in the mercury intrusion saturation. when the mercury intrusion pressure is between 3 mpa and 10 mpa, the mercury intrusion saturation increases rapidly. when the mercury intrusion pressure is > 10 mpa, the mercury saturation increases more slowly, forming throats are flake-like and are in the range 0.3–0.8 μm (fig. 2h). intercrystalline throats are tube bundles, and their radii are mainly <0.2 μm (fig. 2i). pore-throat structures mercury injection is an effective way to obtain quantitative data on micro–nano-scale pores. the intrusion pressure pc was converted to pore or throat radius by applying the washburn equation (washburn, 1921), adopting an interfacial tension value of 485 dyn/cm and a wetting angle of 140°. pcp curves variation in the slopes of the mercury intrusion curves was attributed to variability in the pore size distribution (fig. 3); the corresponding parameters are given in table 2. figure 1. geological and lithological features. (a) location and tectonic units of the research area (modified from yang et al. (2005)); (b) comprehensive stratigraphic column of tight sandstone in the ordos basin (data on lithological features from wu et al., 2018) table 1. characteristics of seven tight sandstone samples. samples depth lithology porosity perm content of xrd(%) relative content of clay(%) (m) (%) (md) quartz feldspar carbonate clay kaolinite chlorite illite i/s mixed layer j5 2620 sandstone 9.54 0.19 43 47 4 6 27 22 11 40 j6 2404 sandstone 11.99 0.28 38 49 8 5 10 33 4 53 j10 2394 sandstone 11.92 0.14 37 48 2 13 10 33 4 53 j12 2270 sandstone 12.26 0.32 39 51 5 5 25 24 11 40 j13 2554 sandstone 10.58 0.11 31 46 8 15 30 26 8 36 j17 2367 sandstone 10.11 0.06 35 43 14 8 25 10 18 47 j20 2306 sandstone 9.16 0.22 34 40 20 6 27 8 16 49 22 wei wang, caili yu, le zhao, shuang xu, lei gao the second platform. the threshold pressure of this type of curve is low, with a mean value of 1.01 mpa. there is some discrepancy in the medium saturation pressures of the seven samples: the largest, 17.56 mpa, is for sample j17, and the smallest, 4.19 mpa, is for sample j6. this variation indicates that the pore-throat distribution of the samples is inhomogeneous. the maximum mercury intrusion saturations and residual mercury saturations of the samples differ greatly, with the average values being 88.78% and 57.13%, respectively. the difference between the mercury intrusion saturation and the mercury extrusion saturation indicates that a large amount of mercury residue remained in the pore-throat space of the samples. rcp curves rcp can provide information on the pore–throat structure because the data derived by this method reflects the pore space and throat space separately. figure 2. primary pore and throat types in sandstone samples in thin section and sem images. (a) residual intergranular pores preserved by chlorite (j6), (b) residual intergranular pores that are triangular and regularly polygonal (j20), (c) dissolution pores in feldspar (j6) (d) dissolution pores in debris (j10), (e) intercrystalline pores in illite (j12), (f) intercrystalline pores in chlorite (j13), (g) compacted throats (j5), (h) dissolution throats (j10) and (i) intercrystalline throats formed by clay-space necking (j17). figure 3. capillary pressure curves from pressure-controlled porosimetry (pcp) the samples were differentiated into two types based on the characteristics of the intrusion curve (fig. 4); the full rcp results are given in table 2. for sample j6, the total mercury saturation curve shows a similar trend with increasing pressure as the throat injection and pore injection curves, and the total mercury intrusion can be seen to be controlled by both throats and pores. for sample j20, the mercury intrusion saturation was controlled by pores at the beginning of mercury intrusion, but, with increasing pressure, throats became the controlling factor. the threshold pressures of the two types of samples are similar. the average value for type i is 1.04 mpa, and that for type ii is 0.97 mpa, indicating that the maximum throat radius controlling the intrusion of fluids into pores is similar in type i than in type ii. for type i, the final intrusion saturation ranges from 39.55% to 85.41%, with an average value of 62.26%, and for type ii, the final intrusion saturation varies from 45.13% to 73.07%, with an average value of 60.07%. the final mercury saturation of type i is similar to that of type ii, indicating that the macro-scale throat content of the samples is similar. the similar intrusion satruration and threshold indicates that micro-scale pore spaces is similar in two types i and ii. discussion features of the pcp results the pore-throat radius distribution can be obtained from the saturation values corresponding to different pressures along the mercury intrusion curve (fig. 5). the psd obtained by pcp shows that the pore radius ranges from 36 nm to 1.01 μm, and is mainly within the range of 0.01–0.4 μm. the pore-throat distribution curve for type ⅰ forms a single peak at a pore radius of 0.1 μm. the pore-throat distribution of type ⅱ is bimodal, indicating that there are two pore-throat development regions. the smaller pore-throats are 0.04 μm and the larger pore-throats are 0.8 μm. the pcp mercury intrusion curve shows a smaller pore radius than the thin section and sem observations and cannot identify residual intergranular pores with a larger radius. features of the rcp results when calculating the reservoir space by rcp, the pore space is considered to be spherical. however, because of compaction and cementation in tight 23combining sem and mercury intrusion capillary pressure in the characterization of pore-throat distribution in tight sandstone and its modification by diagenesis table 2. pore-throat structure parameters of seven samples on the basis of pcp and rcp samples pcp rcp threshold pressure (mpa) medium saturation pressure (mpa) maximum mercury intrusion saturation (%) residual mercury saturation (%) threshold pressure (mpa) final mercury intrusion saturation (%) pore mercury saturation (%) throat mercury saturation (%) j5 1.16 10.30 91.91 55.89 1.48 45.13 26.33 18.80 j6 0.71 4.19 90.53 60.46 0.88 62.00 41.43 20.57 j10 1.16 12.63 90.91 52.76 0.56 73.07 39.88 33.19 j12 1.81 10.55 86.17 59.12 0.79 85.41 49.66 35.75 j13 2.91 10.90 90.14 56.04 0.56 42.52 16.09 26.43 j17 2.91 17.54 87.18 53.71 1.98 39.55 24.09 15.47 j20 1.81 11.14 84.51 61.95 0.84 81.57 47.69 33.88 figure. 4 mercury intrusion curves for samples j6 and j20 (rcp) figure 5. pore radius distribution derived by pcp (a) type i and (b) type ii 24 wei wang, caili yu, le zhao, shuang xu, lei gao sandstones, the shape of pore space is dominantly tubular (xiao et al., 2017b). thus, the final pore distribution is obtained by first multiplying the initial pore radius given by rcp by a shape factor (2/3) to account for this tubular shape. the radii of the pores range from 50 μm to 200 μm, and the psd is characterized by two peaks (fig. 6a). the right peak ranges from 100 μm to 200 μm, which mainly corresponds with residual intergranular pores filled with cements, which have a large radius. the left peak ranges from 50 μm to 100μm, which is mainly attributable to feldspar dissolution pores, which have small radii. the throat sizes are widely distributed, ranging from 0.2 μm to 1.4 μm (fig. 6b), implying that the nano-scale reservoir spaces have been strongly influenced by diagenesis. the psd can be calculated by consideration of the pore distribution and throat distribution data from rcp (fig. 7). the pore-throat distributions of the seven samples are characterized by having two peaks. the distribution of the right peak is consistent with the information obtained via thin section and sem examination, ranging from 60 μm to 150 μm, while the distribution of the left peak is limited by injection pressure and ranges from 0.2 μm to 1.2 μm. psd calculated by pcp and rcp differences between the pcp and rcp results as discussed above, both pcp and rcp have limitations. it is impossible to obtain the full-range pore-throat distribution using just one of them. rcp can distinguish pores and throats, but the maximum mercury intrusion pressure is low, and throats with a radius of under 0.12 μm cannot be characterized. pcp can provide information on smaller pore-throats because of its high mercury intrusion pressure, but it cannot distinguish pores and throats, so it has poor ability to recognize pores with a large radius. therefore, it is best to combine the two test datasets to obtain the overall psd, taking advantage of the strengths of both pcp and rcp. since pcp and rcp are calculated under the same test principle, samples should have the same mercury intrusion curves under the same mercury intrusion pressure (fig. 8). in reality, the mercury intrusion curves obtained by pcp and rcp are not exactly the same due to the differences in contact angle and interfacial tension in pcp and rcp. figure 6. rcp-derived distributions of (a) pores and (b) throat bodies in seven tight sandstone samples figure 7. corrected-rcp-based pore size distribution in seven tight sandstone samples. the mercury saturation reached in the pcp tests varied from 84.51% to 91.91%, with an average value of 88.76%, which is larger than that arrived at through rcp (averaging 63.98%), indicating that there is a lot of nano-scale pore space in tight sandstone that cannot be probed by rcp. this is caused by the limitation on the mercury intrusion pressure that can be applied. full-range psd calculated by pcp and rcp the measurement range of rcp means that it can characterize pores but cannot identify nano-scale throat bodies, while pcp can characterize the overall distribution of nano-scale throats. thus, the combination of the throat distribution derived by rcp and the pore distribution derived by rcp is an efficacious method for revealing the pore-throat distribution. the mercury intrusion curve of pcp overlaps with that of rcp when the intrusion saturation is low. this portion of the data reflects the pore distribution of the sample, which can be obtained from the psd given by rcp. when the mercury intrusion saturation in rcp reaches its maximum, the distribution of the remaining throats in the sample is derived from the pcp mercury intrusion curve. thus, the distribution of micro-scale spaces is derived from rcp, and the distribution of nano-scale spaces is derived from pcp. the pore-throat sizes detected in the tight sandstone range from 36 nm to 200 μm in radius (fig. 9). the full-range psd curves are dominated by three peaks. the right semi-peak ranges from 50 μm to 200 μm, with a peak value of 90 μm; the corresponding pore types are residual intergranular pores and 25combining sem and mercury intrusion capillary pressure in the characterization of pore-throat distribution in tight sandstone and its modification by diagenesis dissolution pores. the middle semi-peak ranges from 0.2 μm to 1.5 μm; throats in this size range are compacted throats. the left peak is between 36 nm and 0.2 μm and reaches a maximum at different values for the two sandstone types: 0.1 μm for type ⅰ and 0.04 μm for type ⅱ. throats in this size range are dissolution throats and intercrystalline throats. the peak values of the middle peak and right peak to not differ significantly for the two types, indicating there is no obvious discrepancy between the content of pores and compacted throats in the two types. the higher peak value for the left peak for type ⅰ than for type ⅱ, meanwhile, implies that dissolution throats control more mercury intrusion space in type ⅰ samples than in type ⅱ samples. when pore radii are <1.5 μm, their pore-throats are dominantly nanothroats, including compacted throats, dissolution throats and intercrystalline throats. the sizes of compacted throats are between 0.2 μm and 1.5 μm, the sizes of intercrystalline throats are <0.05 μm, and dissolution throats usually range from 0.05 μm to 0.2 μm. the results show that the content of the different throat types differs significantly between the tight sandstone samples. the saturation of micro-scale pore bodies ranges from 15.87% to 49.66%, with an average value of 33.49%, and the average saturation of nanoscale throats is 66.54%, indicating that nano-scale throat bodies, rather than micro-scale pore bodies, dominate the pore spaces of the tight sandstones. with a decrease in the permeability of the samples, the nano-scale throat content increases, while the proportion of micro-scale pores decrease. the influences of diagenesis on the full-range pore-throat distribution diagenesis after sedimentation has an important influence on pore throat formation, giving rise to the complexity and heterogeneity of pore-throat space in tight sandstones. the effects of diagenesis on pores and throats can be inferred from the full-range pore-throat distribution. (1) compaction: samples j5, j6, j10, and j12 have similar contents of mineral particles, as well as a relatively low cement content (<15%) (table 1). with increasing sample depth from 2270 m to 2620 m, the pore distribution range remains stable and the pore-space saturation decreases gradually from 49.66% (sample j12, 2270 m) to 26.33% (sample j5, 2620 m) (fig. 10a, b and c). there is no obvious variation in the peak value of throats with a radius of under 1 μm in the four samples. with increasing depth, the porosity decreases from 12.26% to 9.54%, though the magnitude of the decline is relatively small for samples j12, j10 figure 8. comparison of pcp and rcp intrusion curves for samples j6 and j20 figure 9. pore size distribution derived by combining pcp and rcp results. (a) type i and (b) type ii 26 wei wang, caili yu, le zhao, shuang xu, lei gao and j6. this decrease may have been caused by feldspar dissolution and the formation of clay minerals during early cementation (figs. 2c and 10e, f), which can expand or preserve the pore space from compaction (berger et al., 2009). this implies that compaction has a considerable pore-space and porosity-reducing effect. (2) cementation the cements in the tight sandstones are mostly carbonate, while clay cements are less dominant. carbonates generally take a double-crystal shape, and clay appears both in the form of a coating on mineral particles and as crystals bridging spaces between grains. these cements have an important effect on reservoir quality. (i) carbonate. samples j10, j17, and j20 have similar burial depths and mineral particle content, and the cements in the three samples are dominantly carbonate and clay. with increasing carbonate content from 2% (j10) to 20% (j20) (table 2), the content of pore bodies becomes significantly lower, dropping from 39.88% to 15.88% (figs. 11a, b and c). porosity also shows corresponding changes. that may be caused by carbonate infilling of pore spaces during the later stages of cementation (fig. 11d), implying that carbonate plays an important role in reducing pore size. (ii) clay. with increasing clay content from 6% (j20) to 13% (j10), the distribution of the throat space becomes wider and the content of throat bodies becomes lower. that may be caused for throat space being filled and segregated by pore-bridging clay (figs. 2e and 11e). correspondingly, the permeability decreases. the clay minerals in samples j17 and j20 are mainly kaolinite and illite, while those in sample j10 are dominated by chlorite (table 2). with a decreasing relative chlorite content, the content of pore space becomes lower. the higher pore space content in the chlorite-rich sample may be attributable to porelining-type authigenic chlorite forming during the early stages of cementation (fig. 11a) preventing compaction and pore-bridging clay (e.g. kaolinite) from decreasing pore spaces (huang et al., 2004). this indicates that pore-bridging clay cements can reduce the throat size by filling the throat, while pore-lining clay cements can preserve pore space by coating the grains. figure 10. graphs shown the effect of compaction on the pore size distribution of (a) j12, (b) j10, (c) j6 and (d) j5, which have different burial depths but similar mineral compositions. (e-f) thin-section images showing (e) the residual intergranular pores in sample j12 and (f) isolated feldspar-dissolution pores in sample j5. note that there is evidence of a decrease in the pore space and porosity as compaction increases. figure 11. graphs showing the effect of clay and carbonate cementation on the pore size distributions of (a) j20, (b) j17 and (c) j10, three tight sandstone samples that have similar burial depths but various contents and types of cements. (d) thin-section image showing that sample j20 contains carbonate with a double-crystal shape. (e) sem image demonstrating that sample j10 contains abundant illite. with an increasing content of pore-bridging cement, the content of throat bodies and permeability decrease appreciably, while the content of pore bodies does not. with an increase in carbonate content, the pore body content and porosity decrease substantially. 27combining sem and mercury intrusion capillary pressure in the characterization of pore-throat distribution in tight sandstone and its modification by diagenesis conclusions sem, thin section, pcp and rcp analyses were performed on seven samples from the upper triassic yanchang tight sandstone from the ordos basin so as to bring the combined strengths of a range of methods to bear on determining the full-range distribution of pore bodies and throats and the effects of diagenesis upon them. the conclusions are as follows: (1) the sem and thin-section experimental results indicate that pores and throats formed in the tight sandstones by various diagenetic processes. these spaces can be distinguished between those with a compaction origin, a dissolution origin or a clay-dominated origin; these mainly formed by compaction, dissolution and cementation, respectively. pore-throats with different origins have different sizes and shapes. (2) the full-range pore size distribution of tight sandstones can be determined by combining the experimental data from pcp and rcp. the detected pore sizes of these tight sandstones range from 36 nm to 200 μm, and the psd curves of the samples are polymodal. the right peak remains stable across the sample set and corresponds to residual intergranular pores and dissolution pores. the middle and left peaks show variation between samples due to the complexity and heterogeneity of nano-scale throat bodies. the average content of micro-scale pores is 33.46%, and that of nano-scale throats is 66.54%. nano-scale throat spaces thus dominate the pore spaces in the tight sandstones. (3) the effects of diagenesis on pores and throats were analyzed by comparing the pore-throat distribution of samples with various burial depths and cement contents. pore-bridging clay cements can reduce throat sizes while pore-lining clay cementation coating mineral grains can preserve pore space. carbonate deposited during later stages of cementation, and compaction decreases the content of pore bodies. acknowledgments this work was financially supported by the national science and technology major project of china (nos 2011zx05013-006 and 2017zx05013-003), the doctoral scientific research foundation of yulin university (18gk22), and the shaanxi province science and technology resource open sharing platform project (2019t-18). references berger, a., gier, s. & krois, p. (2009). porosity-preserving chlorite cements in shallow-marine volcaniclastic sandstones: evidence from cretaceous sandstones of the sawan gas field, pakistan. aapg bulletin, 93(5), 595-615. daigle, h. & johnson, a. (2016). combining mercury intrusion and nuclear magnetic resonance measurements using percolation theory. transport in porous media, 111(3), 669-679. gane, p.a. et al., 2004. comparison of nmr cryoporometry, mercury intrusion porosimetry, and dsc thermoporosimetry in characterizing pore size distributions of compressed finely ground calcium carbonate structures. industrial & engineering chemistry research, 43(24), 7920-7927. gao, h., wang, c., cao, j., he, m. & dou, l. (2019). quantitative study on the stress sensitivity of pores in tight sandstone reservoirs of ordos basin using nmr technique. journal of petroleum science and engineering, 172, 401-410. ghanizadeh, a., clarkson, c., aquino, s., ardakani, o. & sanei, h. (2015). petrophysical and geomechanical characteristics of canadian tight oil and liquid-rich gas reservoirs: i. pore network and permeability characterization. fuel, 153, 664-681. hillier, s., 2000. accurate quantitative analysis of clay and other minerals in sandstones by xrd: comparison of a rietveld and a reference intensity ratio (rir) method and the importance of sample preparation. clay minerals, 35(1), 291-302. huang, s. j., xie, l. w. & zhang, m. (2004). formation mechanism of authigenic chlorite and relation to preservation of porosity in nonmarine triassic reservoir sandstones, ordos basin and sichuan basin, china. journalchengdu university of technology, 31, 273-281. jarvie, d. m., hill, r. j., ruble, t. e. & pollastro, r. m. (2007). unconventional shale-gas systems: the mississippian barnett shale of north-central texas as one model for thermogenic shale-gas assessment. aapg bulletin, 91(4), 475-499. jia, j., yin, w., qiu, n., wang, g., ma, l., liu, y., & liu, n. (2017). migration and accumulation of crude oil in upper triassic tight sand reservoirs on the southwest margin of ordos basin, central china: a case study of the honghe oilfield. geological journal, 53(5), 2280-2300. doi: https://doi.org/10.1002/gj.3065. klaver, j., desbois, g., urai, j. l. & littke, r. (2012). bib-sem study of the pore space morphology in early mature posidonia shale from the hils area, germany. international journal of coal geology, 103, 12-25. loucks, r. g. & ruppel, s. c. (2007). mississippian barnett shale: lithofacies and depositional setting of a deep-water shale-gas succession in the fort worth basin, texas. aapg bulletin, 91(4), 579-601. mayo, s., josh, m., nesterets, y., esteban, l., pervukhina, m., clennell, m., maksimenko, a., & hall, c. j. (2015). quantitative micro-porosity characterization using synchrotron micro-ct and xenon k-edge subtraction in sandstones, carbonates, shales and coal. fuel, 154, 167-173. nelson, p. h. (2009). pore-throat sizes in sandstones, tight sandstones, and shales. aapg bulletin, 93(3), 329-340. rezaee, r., saeedi, a. & clennell, b. (2012). tight gas sands permeability estimation from mercury injection capillary pressure and nuclear magnetic resonance data. journal of petroleum science and engineering, 88, 92-99. wang, x., peng, x., zhang, s., du, z. & zeng, f. (2018). characteristics of oil distributions in forced and spontaneous imbibition of tight oil reservoir. fuel, 224, 280-288. waples, d. w. (2002). evolution of sandstone porosity through time. the modified scherer model: a calculation method applicable to 1-d maturity modeling and perhaps to reservoir prediction. natural resources research, 11(4), 257-272. washburn, e. w. (1921). the dynamics of capillary flow. physical review, 17(3), 273. wu, h., zhang, c., ji, y., liu, r., cao, s., chen, s., zhang, y., ... liu, g. (2018). pore throat characteristics of tight sandstone of yanchang formation in eastern gansu, ordos basin. petroleum research, 3(1), 33-43. doi: https://doi.org/10.1016/j.ptlrs.2017.11.001 xi, k., cao, y., haile, b. g., zhu, r., jahren, j., bjorlykke, k., zhang, x., & hellevang, h. (2016). how does the pore-throat size control the reservoir quality and oiliness of tight sandstones? the case of the lower cretaceous quantou formation in the southern songliao basin, china. marine and petroleum geology, 76, 1-15. xiao, d., guo, s., xu, q., lu, z. & lu, s. (2017a). type and size distribution of nanoscale pores in tight gas sandstones: a case study on lower cretaceous shahezi formation in songliao basin of ne china. journal of nanoscience and nanotechnology, 17(9), 6337-6346. xiao, d., jiang, s., thul, d., huang, w., lu, z., & lu, s. (2017b). combining rate-controlled porosimetry and nmr to probe full-range pore throat structures and their evolution features in tight sands: a case study in the songliao basin, china. marine and petroleum geology, 83, 111-123. yan, w., sun, j., cheng, z., li, j., sun, y., shao, w., & shao, y. (2017). petrophysical characterization of tight oil formations using 1d and 2d nmr. fuel, 206, 89-98. yang, y., li, w. & ma, l. (2005). tectonic and stratigraphic controls of hydrocarbon systems in the ordos basin: a multicycle cratonic basin in central china. aapg bulletin, 89(2), 255-269. yao, y. & liu, d. (2012). comparison of low-field nmr and mercury intrusion porosimetry in characterizing pore size distributions of coals. fuel, 95, 152-158. yao, y., liu, d., che, y., tang, d., tang, s., & huang, w. (2010). petrophysical characterization of coals by low-field nuclear magnetic resonance (nmr). fuel, 89(7), 1371-1380. 28 wei wang, caili yu, le zhao, shuang xu, lei gao yuan, h. & swanson, b. (1989). resolving pore-space characteristics by ratecontrolled porosimetry. spe formation evaluation, 4(01), 17-24. zhang, l., lu, s., xiao, d. & li, b. (2017). pore structure characteristics of tight sandstones in the northern songliao basin, china. marine and petroleum geology, 88, 170-180. zhang, w., yang, h., hou, l. & liu, f. (2009). distribution and geological significance of 17α (h)-diahopanes from different hydrocarbon source rocks of yanchang formation in ordos basin. science in china series d: earth sciences, 52(7), 965-974. zhao, h., ning, z., wang, q., zhang, r., zhao, t., niu, t., zeng, y. (2015). petrophysical characterization of tight oil reservoirs using pressurecontrolled porosimetry combined with rate-controlled porosimetry. fuel, 154, 233-242. zhu, h., zhong, d., yao, j., niu, x., liang, x., & zhao, y. (2014). microscopic characteristics and formation mechanism of upper triassic chang 7 tight oil reservoir in the southwest ordos basin. journal of china university of mining and technology, 43, 853-863. zou, c., yang, z., tao, s. z., yuan, x. j., zhu, r. k., hou, l. h., wu, s. t. ... pang, z. l. (2013). continuous hydrocarbon accumulation over a large area as a distinguishing characteristic of unconventional petroleum: the ordos basin, north-central china. earth-science reviews, 126, 358-369. keywords: agricultural system; wheat; photosynthesis; cropping seasons palabras clave: sistema agrícola; trigo; fotosíntesis; cosechas how to cite item fateh, h., toube, a., & gholipuri, a. q. (2020). wheat physiological traits response in standard and ecological agricultural systems. earth sciences research journal, 25(1), 131-136. doi: https://doi.org/10.15446/esrj.v25n1.94403 this research was conducted on a farm of payam-e-noor university in piranshahr, in west azerbaijan province of iran, during 2014-2015 and 2015-2016 crop sessions to investigate the wheat physiological traits response to standard and ecological agricultural systems. the experiment was conducted as a split-plot design with two factors and three replications. the first factor, which was considered the main plot, included high, low-input, and ecological agricultural systems. subplots were seed pre-treatment, including control (without priming), hydro priming, and food priming with ascorbate. the results showed that the least photosynthesis rate and mesophilic conductance were related to lowinput agricultural system; however, in the ecological system, the transpiration, substomatal carbon dioxide, and water use efficiency of photosynthesis were the lowest. also, seed priming increased the photosynthesis rate, mesophilic conductivity, and water use efficiency of photosynthesis. abstract wheat physiological traits response in standard and ecological agricultural systems respuesta de los rasgos fisiológicos del trigo en sistemas agrícolas estándar y ecológico issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.94403 este estudio se realizó en la granja de pruebas de la universidad de payam-e-noor, en la ciudad de piranshahr, provincia de west azerbaijan, irán en la cosecha de los años 2014-2015 y 2015-2016, para analizar la respuesta de las características morfofisiológicas del trigo en sistemas de agricultura común y de agricultura ecológica. la fase experimental se diseñó en una área dividida, con dos factores y tres réplicas. el primer factor, que se consideró como el del lote principal, se dividió en los niveles de alta, baja contribución y sistemas de agricultura ecológica. los sublotes fueron sembrados antes de los tratamientos que incluyen control (sin preparación), preparación hidráulica y preparación de nutrientes con ascorbato. los resultados muestran que el menor índice de fotosíntesis y de conductividad mesófila estaba relacionado con los sistemas agrícolas de baja aportación. sin embargo, la transpiración, el dióxido de carbono subestomático y el uso eficiente del agua durante la fotosíntesis en el sistema ecológico fueron los más bajos. además, la preparación de la siembra incrementó el índice de fotosíntesis, la conductividad mesófila y el uso eficiente del agua durante la fotosíntesis. resumen record manuscript received: 05/02/2020 accepted for publication: 13/06/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 131-136 hamid fateh, ahmad toube*, abdul qayum gholipuri the university of mohaghegh ardebili, ardabil, iran. *corresponding author: ahmadtobeh1340@gmail.com e n v ir o n m e n ta l g e o l o g y https://doi.org/10.15446/esrj.v25n1.94403 https://doi.org/10.15446/esrj.v25n1.94403 mailto:ahmadtobeh1340@gmail.com 132 hamid fateh, ahmad toube, abdul qayum gholipuri introduction agricultural operations are based on the traditional knowledge and experience of human beings. an appropriate solution to increase agricultural products is to increase yields per unit area. now, in addition to increasing production, stability is also essential. therefore, various farming systems and their production potential should be investigated and revised (fateh, 2001). fateh et al. (2013) explained that the biggest challenge in the next 50 years is to double food in ways that do not harm the environment and consumer health. also, studies on agricultural systems are essential for understanding the developing factors and limiting factors. today, the global approach is towards ecological agriculture, which is a sustainable agricultural branch. this farming system tries to avoid chemical origin supplies (choghmaqi & moradi, 2007). the conventional agricultural system in the use and management of resources has not achieved acceptable success because pesticides and fertilizers have led to the emergence of unsustainable farming ecosystems, environmental pollution, and quantitative and qualitative reduction of the product and energy efficiency. lotter et al. (2003), while studying the standard agricultural and biological systems, reported that absorption and storage of water of rainfall in plots with the biological system are 100% higher than in the standard method, which could improve gas exchanges and consequently performance. they also reported that the performance of the ecological system is 38 percent higher than the standard system. jahan et al. (2009) reported that more nutritional elements, especially nitrogen, increase the water requirement proportionally to the plant’s growth in the high-input system. they also stated that moisture is more available to the plant for transpiration in a low-input ecological agricultural system than in a high-input system. and the plant can maintain a higher temperature difference with ambient temperature through more transpiration. higher soil capacity to retain moisture in ecological agricultural systems, especially biological systems, has been discussed in several sources (lotter et al., 2003; hole et al., 2005). meanwhile, the existence of animal manure in the ecological agricultural system has exacerbated this issue. jahan et al. (2009) stated that a high-input system is superior in terms of traits such as seed yield, number of maize, and soil respiration. in terms of plant height, the difference between systems is not significant. the canopy temperature in the ecological system was approximately 10% lower than the high-input system. the differences between ecological and low-input systems are shallow, and at the same time, they are in the middle between input and moderate-input systems. in other words, a nearly 10% reduction in the performance of ecological and low-input systems is compensated not only by lower supplies in terms of costs but also for the stability of these systems in the long term. oehl et al. (2004) reported that the consumption of fertilizers and energy in the biological farms decreases by 34-53 percent compared to the common ones. in contrast, the performance decrease of the biological method is only 20%. what is essential in ecological systems is forming mutual effects for increasing the efficiency of material and energy consumption, self-sufficiency of reversibility, and sustainability over time. economic performance in these systems among the high ecological values is not much attended (gosling et al., 2006; pimentel et al., 2005). materials and methods this research was conducted on a farm of payam-e-noor university, in piranshahr, west azerbaijan province, during 2014-2015 and 2015-2016 crop sessions. the experiment was conducted as a split-plot design with two factors and three replications. the first factor, which was considered the main plot, consisted of high-input, low-input, and ecological agricultural systems. subplots were seed pre-treatment, including control (without priming), hydro priming (seeding in distilled water for 8 hours), and food priming with ascorbate (100 ppm ascorbic acid solution for 8 hours) (harris et al., 2004). the wheat used is azar 2 certified, which a semi-water cultivar was prepared from oshnavieh gold cluster company. all crop operations, including plowing, pest, and weed control, were applied as registered in table 1. the traits related to gas exchange, including photosynthesis, transpiration, sub-stomatal carbon dioxide, and mesophilic conduction, the activity of antioxidant enzymes, leaf chlorophyll content, ion leakage percentage, the relative water content of leaves, and protein content of the seed, were measured. the infrared gas analyzer (irga) model lca4, manufactured by adc company of england, was used to measure photosynthesis per unit area of a leaf (µmol co2 m -2s-1), transpiration rate (mmol m-2s-1), and co2 sub-stomata concentration (µmol mol -1). all measurements were carried out at 10 to 12 hours in the light intensity of 1200-1400 micromoles photon per square meter per second. before the measurement began, the device was turned on for 20 minutes. in each treatment, the middle part of the leaf under the flag of the main stem was placed in the glass chamber in eight stages, from flowering to the hard seed; and after one minute, the device data was recorded. water use efficiency of photosynthesis (µmol co2 mol -1) was also determined by dividing the photosynthesis rate per unit area of the leaf to transpiration (siosemardeh et al., 2014). results and discussion the variance analysis results showed that the effect of year on photosynthesis was significant at 5% level, and the impact of agricultural system and seed priming system was significant at 1% level (table 2). as shown in figure 1, the photosynthesis rate in the first year was approximately 12 μmol co2 s -1, while in the second year, it decreased to 6.10. it should be noted that the precipitation in the second year was lower than in the first year. also, comparing the agricultural system’s average interactions showed no significant difference between the high-input and ecological agricultural systems regarding the photosynthesis of the wheat plant’s flag leaf. in contrast, the lowest photosynthesis rate was associated with the low-input agricultural system (fig. 2). the highest photosynthesis rate in the treatments was related to seed treatment with ascorbic acid. however, there was no statistically significant difference with hydro priming treatment, but there was a substantial difference with the control (fig. 3). this study’s results are consistent with fateh et al. (2013) on the positive effects of ascorbic acid priming on photosynthesis. plants from primed seeds have expanded their root system compared to plants from untreated seeds during a shorter time. they reach the autotrophic stage with the better absorption of water and food and the production of photosynthetic green sectors. the realization of biological and ecological conditions gives a particular position to plants primed with seeds. these conditions allow the plant to use environmental elements, such as water and light, more appropriately. these cases’ results can eventually lead to an increase in the photosynthetic rate and area in these plants. consequently, carbon dioxide stabilization and, naturally, the production of assimilates and the storage of non-structural hydrocarbons in different parts of the plant have increased. and as a result, higher biomass will be produced (duman, 2006). the main effects of seed priming include seed germination, initial seedling establishment, increasing grain yield and biomass through increased photosynthesis due to faster plant establishment (farooq et al., 2006; harris et al., 2001; duman, 2006). table 1. consumption of supplies and crop operations in different agricultural systems agricultural system high-inputlow-inputecologicalinput consumed 1--plowing times 11-cultivator times 400-200-150200-150-50biofertilizer fertilizer(n-p-k) kg/ ha --25manure (ton/ha) 21chemical struggle with pests and diseases(times) 21by handweed combat (times) 133tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm figure 1. comparison of the average effect of year: y1 (first year), y2 (second year) on the amount of photosynthesis of wheat. figure 2. comparison of the average effect of agronomic system: s1 (high-input), s2 (low-input) and s3 (ecological) on the amount of photosynthesis of wheat. figure 3. comparison of the average effect of seed pre-treatment: p1 (control), p2 (hydro priming) and p3 (ascorbic acid) on the photosynthesis of wheat. table 2. analysis of variance of studied traits (as=agricultural system. p=priming, y=year) mean of square degree of freedom (df) source of change grain yield water use efficiency mesophyll conductance sub stomata co2 transpiration photosynthesis 115/0 023/0 173/13 407/1625 018/0 336/0 replication 003/2 037/0 71/204** 074/1968 592/2* 582/27* y 543/0 014/0 796/0 630/135 119/0 286/0 main error 645/5** 929/2** 778/99** 352/5756** 343/4** 838/15** as 812/0** 021/0 992/3 574/668 069/0 247/0 p 562/0** 465/0** 386/14* 963/48 197/0 423/2** y× as 015/0 041/0 930/2 963/44 037/0 286/0 p ×y 062/0 085/0 506/5 713/599 050/0 261/0 as×p 020/0 036/0 429/1 602/99 059/0 132/0 y×p× as 052/0 053/0 327/4 060/875 131/0 227/0 total error 03/4 74/8 07/8 70/6 28/8 21/4 coefficient of variation (c.v %) * and ** there is a significant difference between them at the probability level of 5% and 1%, respectively. effect of year and agricultural system on transpiration rate was significant at 5% and 1% respectively (table 2). as shown in figure 4, the photosynthesis and transpiration rates were also higher in the first year than in the second year and, in the second year, the transpiration rate decreased by about 10% compared to the first year. however, unlike photosynthesis, transpiration rate was the lowest in ecological agricultural system (fig. 5). the main factor in reducing the photosynthesis and transpiration rates under stress conditions is due to this strategy that, the plant uses the limited amount of water that it has to avoid drought and, it tries to close its stomata to prevent the water loss. with the onset of the dry period, the plant maintains its transpiration at maximum level for a while but, with the continuation of the dry period, it tries to narrow its stomata and eventually close those (moradi et al., 2005). figure 4. comparison of the average effect of year, y1 (first year), y2 (second year) on the amount of transpiration rate of wheat. figure 5. comparison of the average effect of agronomic system: s1 (high-input), s2 (low-input) and s3 (ecological) on the amount of transpiration rate of wheat. 134 hamid fateh, ahmad toube, abdul qayum gholipuri the effect of high-input, low-input and ecological systems at the level of 1% on the sub-stomatal carbon dioxide was significant (table 2). the results showed that, under ecological conditions, the carbon dioxide content in the sub-stomatal cavity is less than that of the high-input system. application of chemical fertilizers and chemical pesticides seems to cause stress in the plant and has increased carbon dioxide levels in the sub-stomatal cavity in the highinput system (fig. 6). fateh et al. (2011) reported that the sub-stomatal co2 concentration under stress conditions was about 13% higher than the control condition. under stress conditions, co2 injected into leaves is not well used in photosynthesis. this suggests that any kind of stress action on the plant, in addition to closing the stomata and preventing the entry of co2, also prevents the leaves from being processed by co2 through the internal mechanism (siosemarde et al., 2003). the reason for this is the effect of metabolic factors during stress on the reduction of carboxylation and the efficiency of carbon dioxide emissions into chloroplast and mesophilic cells. under these conditions, it seems that non-stomatal limiting factors play an important role in reducing photosynthesis in drought stress. (nazemosadat & kazemini, 2008; siosemarde et al., 2003). figure 6. comparison of the average effect of agronomic system: s1 (high-input), s2 (low-input) and s3 (ecological) on sub-stomatal conductance rate of wheat. effects of crop year and system mesophilic conductivity was significant at 1% level and seed priming effect at 5% level on (table 2). as with the photosynthesis and transpiration rates, mesophilic conductivity was reduced in the second year. the decrease in mesophilic conductivity in the second year was about 14% compared to the first year, which was slightly higher than the decrease in transpiration (fig. 7). figure 7. comparison of the average effect of year: y1 (first year), y2 (second year) on the mesophilic conductance of wheat. figure 8. comparison of the average effect of agronomic system: s1 (high-input), s2 (low-input) and s3 (ecological) on mesophilic conductance rate of wheat also, the mesophilic conductivity in the ecological agricultural system was higher than the low-input system however, there was no statistically significant difference with the high-input system (fig. 8). in a study on durum wheat, in severe drought stress treatments, an increase in the concentration of sub-stomatal co2 was associated with a significant decrease in mesophilic conduction and thus reduction of photosynthetic capacity of chloroplast. in this case carbon dioxide injected into the leaves has not been properly used in the photosynthesis process (nazemosadat, & kazemini, 2008). in addition, with the mean comparison between priming treatments in this study it was found that, as with other traits studied so far, applying seed food priming with ascorbic acid or vitamin c has led to increased mesophilic conductivity in wheat plants due to faster growth and better growth and more efficient use of resources than control treatments (fig. 9). figure 9. comparison of the average effect of seed pre-treatment: p1 (control), p2 (hydro priming) and p3 (ascorbic acid) on mesophilic conductance of wheat the effect of agricultural system and also seed priming on the photosynthetic water use efficiency was significant at the level of 1%. comparison of mean of treatments showed that in ecological agricultural system, photosynthetic water use efficiency was higher in comparison with low-input and high-input systems and, this difference is significant. the photosynthetic water use efficiency in the ecological system was 2.79 mmol/ mol/sec which was reduced by 12 and 7 percent, respectively, in high-input and low-input systems (fig. 10). it was also determined that photosynthetic water use efficiency was higher in priming treatments (fig. 11). jahan et al. (2009) stated in their research that in general, in low-input and ecological agricultural systems, moisture available to the plant was higher than the highinput system for transpiration and the plant has been able to maintain a higher temperature difference with more transpiration. several sources have referred to the greater soil capacity to maintain moisture in ecological agricultural systems, especially biological systems (hole et al., 2005; lotter et al., 2003), while the existence of animal manure in the ecological agricultural system has exacerbated this issue. figure 10. comparison of the average effect of agronomic system: s1 (highinput), s2 (low-input) and s3 (ecological) on water use efficiency of photosynthesis of wheat. 135tracing method of rare earth elements in surrounding rock of geological formation based on three-dimensional positioning algorithm figure 11. comparison of the average effect of seed pre-treatment: p1 (control), p2 (hydro priming) and p3 (ascorbic acid) on water use efficiency of photosynthesis of wheat. the interaction between the crop year and the agricultural system and also the effect of seed pre-treatment on wheat grain yield was significant at the level of 1% (table 2). comparison of the average of treatments using duncan test at 5% probability level showed that in both the first and second year, the grain yield in the high-input system was higher than the other treatments. however, there was no significant difference between the low-input and ecological agricultural systems in both years (fig. 12). the results also showed that seed priming with ascorbic acid in comparison with control increased the grain yield, and this increase was statistically significant (fig. 13). figure 12. comparison of the interaction effect of year: y1 (first year), y2 (second year) and agronomic system, s1 (high-input), s2 (low-input) and s3 (ecological) on grain yield of wheat figure 13. comparison of the average effect of seed pre-treatment: p1 (control), p2 (hydro priming) and p3 (ascorbic acid) on grain yield of wheat the photosynthesis rate did not differ in ecological and high-input farming systems. the highest photosynthesis rate and mesophilic conduction among treatments were seed priming treatment with ascorbate and hydro priming. seed priming leads to accelerated germination and establishment of the green surface, and can eventually lead to an increase in the photosynthesis time and area (duman, 2006). under ecological system conditions, the intercellular co2 concentration (ci) was less than that of the high-input system conditions. it seems that the application of chemical fertilizers and chemical pesticides caused stress in the plant and increased the intercellular co2 concentration (ci) in the high-input system. under stress conditions, co2 injected into leaves is not well used in photosynthesis. in ecological farming system, the efficiency of photosynthetic water use was higher than the high and low input system. it was also determined that the relative water content of the leaf in the high-input system relative to the ecological system decreased by about 12%. it seems that in a low-input ecological farming system compared to high-input system, higher moisture has been available to the plant for transpiration and the plant could maintain higher temperature difference through more transpiration. higher photosynthetic water use efficiency is due to more soil capacity to maintain moisture and also the presence of manure in ecological farming system. however, there was no significant difference between the two years in the low-input and ecological farming system. common agriculture is based on maximum production. agricultural sources such as water, soil and genetic diversity have been exploited excessively and are being destroyed. the only remaining choice is to preserve productivity of agricultural land for a long time along with the change in consumption pattern of inputs. maintaining the long-term yield of agricultural land is a requirement for sustainable food production. this sustainability can be achieved through ecological processes on farms (kohnaward et al., 2011). the ecological farming system, while protecting soil fertility, increases the production of the product without relying on the use of chemicals and achieving a stable production. lower average yield in the long-term ecological system is possible through rising prices for these products. but the fact is that removal of chemicals is not enough to move from the common agricultural toward ecological farming (kohnaward et al., 2011). during the transition period of the common systems to low-input systems, problems such as nutrient deficiencies and weed and pest problems are among the major factors affecting yield loss (behdani et al., 2017). increasing yield in the common agricultural system is due to increased consumption of foreign inputs such as fertilizer, pesticide and herbicide. generally, the yield of ecological system is less than the common systems for reasons such as lack of soil nutrients, stress of weeds, pests and diseases. also, decreasing yields often occur during the transition period from common agriculture to ecological agriculture by removing chemical fertilizers, insecticides, fungicides and herbicides. although the use of chemical fertilizers in high-input system may lead to a significant increase in yield in the short term, instead, the sustainability and stability of the yield of ecological systems over time will be higher due to the lack of chemical inputs. the ecological system is, in fact, a step towards the long-term and principled productivity of inputs and the prevention of the degradation and erosion of funds such as water and soil (behdani et al., 2017). references behdani, m. a., zamani, g., fallahi, h. r., sayyari-zohan, m. h. & samadzadeh, a. (2017). evaluation of replacement corms growth criteria of saffron in response to different organic and conventional production systems. journal of saffron agronomy and technology, 5, 133-147 choghmaqi, y. & moradi, a. (2007). organic farming, human rescue. monthly journal of animal, crop and industry. duman, i. (2006). effect of seed priming with peg and k3po4 on germination and seedling growth in lettuce. pakistan journal of biological sciences, 9(5), 923-928 farooq, m., basra, s. m. a., tabassum, k., & ahmad, n. (2006). evaluation of seed vigor enhancement techniques on physiological and biochemical techniques on physiological bases in coars rice (oriza sativa). seed science and technology, 34, 741-750 fateh, h. (2011). effect of nitrogen and zinc fertilizers on physiological characteristics in two winter barley (hordeum vulgare) varieties under drought stress. university of kurdistan faculty of agriculture department of agronomy. fateh, h., siosemardeh, a., karimpour, m., & veysani, w. (2013). effect of seed priming on gas exchange and antioxidant enzyme activity of two wheat varieties. plant and ecosystem, 10(39), 101-121. gosling, p., hodge, a., goodlass, g., & bending, g. d. (2006). arbuscular mycorrhiza fungi and organic farming. agriculture, ecosystems and environment, 113, 17-35 harris, d., kumar, j. k. r., & kumar, j. (2004). onfarm seed priming. agriculture research, 44, 3-14 136 hamid fateh, ahmad toube, abdul qayum gholipuri hole, d. g., perkins, a. j., wilson, j. d., alexander, i. h., grice, p. v., & evans, a. d. (2005). does organic farming benefit biodiversity? biological conservation, 122, 113-130 jahan, m., koocheki, a., ghorbani, r., rejali, f., aryayi, m. & ebrahimi, e. (2009). the effect of biological fertilizers application on some agroecological characteristics of corn under conventional and ecological cropping systems. iranian journal of field crops research, 7(2), 375-390 kohnaward, p., jalilian, j., & pirzad, a. (2011). effect of foliar application of micro-nutrients on some agronomic characteristics of the safflower under conventional and ecological cropping systems. journal of plant science. 4(6), 14-25. lotter, d. w., seidel, r., & liebhardt, w. (2003). the performance of organic and conventional cropping systems in an extreme climate year. american journal of alternative agriculture, 18, 146-154 moradi, a., ahmadi, a., & jodi, m. (2005). photosynthetic reactions and stomatal conductance to severe and mild stress in various growth stages. the first national congress of legumes. ferdowsi university of mashhad institute of plant sciences, 272-268. nazemosadat, s. m. j., & kazemini, s. a. (2008). the effects of water stress and photosynthesis active radiation on photosynthesis rate of wheat. journal of agricultural sciences and natural resources, 14(6), 31-39. oehl, f., sieverding, e., mader, p., dubois, d., ineichen, k., boller, t., & wiemken, a. (2004). impact of long-term conventional and organic farming on the diversity of arbuscular mycorrhizal fungi. oecologia, 138, 574-583. pimentel, d., & burgess, m. (1980). energy inputs in corn production. in: pimentel, d. (ed.). handbook of energy. utilization in agriculture, 6784. crc press, inc. siosemardeh, a., ahmadi, a., poustini, k., & ebrahimzadeh, h. (2003). stomatal and nonstomatal limitations to photosynthesis and their relationship with drought resistance in wheat cultivars. iranian journal of agricultural science, 34(4), 93-106 siosemardeh, a., fateh, h., & badakhshan, h. (2014). responses of photosynthesis, cell membrane stability and antioxidative enzymes to drought stress and nitrogen fertilizer in two barley (hordeum vulgare) cultivars under controlled condition. iranian journal of field crops research, 12(2), 215-228. keywords: ridge-top landslide; debris avalanche; dynamic analysis; dan-w model. palabras clave: deslizamiento de tierra; avalancha de detritos; análisis de dinámicas; modelo dan-w. how to cite item yang, l., wang, w., zhang, n., & wei, y. (2020). characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china. earth sciences research journal, 24(2), 169-181. doi: https://doi. org/10.15446/esrj.v24n2.78990 a catastrophic landslide hit mount fugui, diexi township, mao county, sichuan province, at 05:38:58 on june 24, 2017. this landslide buried xinmo village, caused 83 deaths, and resulted in enormous losses of life and property. the landslide mass cut out and slid from a high position, loaded continuously, and accumulated at the top of the slope body. subsequently, the landslide mass was transformed into avalanche debris. this process was a typical chain disaster of avalanche debris triggered by a ridge-top landslide. the total volume, elevation difference, and horizontal distance of the landslide were 1637.6 × 104 m3, 1200 m, and 2800 m. in this research, the disaster-formation mechanism of the xinmo landslide was identified based on a geological field survey and remote sensing satellites. the disaster characteristics of the landslide source zone, debris avalanche zone, and accumulation zone were then numerically simulated. the entire process of the xinmo landslide movement was comparatively studied using dan-w, a dynamic landslide software program, and multiple rheological models. the research findings indicated that the frictional model favorably simulated the movement characteristics of various phases of the xinmo landslide. this landslide lasted approximately 120 s, and it had a maximum velocity of movement of 74 m/s. therefore, the frictional model and its parameters can be used in similar studies to investigate the dynamic disaster effects of ridge-top rock landslides. abstract characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china características y análisis del modelado numérico del deslizamiento de tierra en xinmo, provincia de sichuan, china issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.78990 un deslizamiento de tierra catastrófico afectó el monte fugui, de la localidad diexi, en el condado mao de la provincia china de sichuan, a las 05:38:58 hora local del 24 de junio de 2017. este deslizamiento enterró la población de xinmo, causó 83 muertos y significó enormes pérdidas de propiedades. la masa del deslizamiento se desprendió en lo alto de la montaña, se continuó cargando y se acumuló en la parte alta de la pendiente. seguidamente, la masa del deslizamiento se transformó en una avalancha de detritos. esta fue una cadena típica de un desastre ocasionado por una avalancha de detritos que comenzó por un deslizamiento en lo alto de la montañana. el volumen total, la diferencia de elevación y la distancia horizontal del deslizamiento fueron de 1637.6 × 104 m3, 1200 m, and 2800 m. en este trabajo se identificó el mecanismo de formación del deslizamiento de xinmo con base en un estudio de campo geológico y con el apoyo de satélites de teledetección. los autores simularon numéricamente las características del desastre en la zona fuente, en la zona de la avalancha y en la zona de acumulación. todo el proceso del deslizamiento de xinmo fue estudiado comparativamente con el programa dan-w, que analiza dinámicas de deslizamiento, y múltiples modelos reológicos. los resultados del estudio indican que el modelo friccional simuló favorablemente las características del movimiento en varias fases del deslizamiento de xinmo. este movimiento duró por lo menos 120 segundos, y tuvo una velocidad máxima de movimiento de 74 m/s. además, el modelo friccional y sus parámetros se pueden usar en estudios similares para investigar los efectos dinámicos de un movimiento de tierra en la parte alta de una montaña. resumen record manuscript received: 09/04/2019 accepted for publication: 07/04/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 169-181 longwei yang1,2,*, wenpei wang2, nan zhang2, yunjie wei2 1school of geological engineering and geomatics, chang’an university, china; 2china institute of geo-environment monitoring, cgs, beijing, china * corresponding author: yang0504@chd.edu.cn g eo te c h n ic a l en g in ee r in g https://doi.org/10.15446/esrj.v24n2.78990 https://doi.org/10.15446/esrj.v24n2.78990 https://doi.org/10.15446/esrj.v24n2.78990 mailto:yang0504@chd.edu.cn 170 longwei yang, wenpei wang, nan zhang, yunjie wei introduction landslides are a major geological hazard due to their long runout distance, high mobility, and entrainment effect (evans et al.,2007; zhang m& mcsaveney,2017; yin et al.,2009). in recent studies, due to the tremendous economy losses and casualties every year, catastrophic landslides have been addressed as environmental disasters (gao et al., 2020; warrick et al., 2019; igwe o, 2018; zhang et al., 2016). this research work has primarily concentrated on the following aspects (fan et al., 2018; hu et al., 2018; wang et al., 2018; zhao et al., 2018): first, potential landslide identification plays an important role in disaster prevention. methods for landslide identification include field surveys, aerial investigations (such as interferometric synthetic aperture radar, unmanned aerial vehicle, and light detection and ranging), acoustic emissions, and micro-seismicity. these methods have been used to analyze the deformation process of potential landslides. furthermore, landslide susceptibility mapping (lsm) has been used to predict potential landslides, such as in the alternating decision tree-adaptive boosting model and the recurrent neural network model. these new technologies could offer the necessary information (coordination and volume) (wang et al., 2020; wu et al., 2020; wang et al., 2019; althuwaynee et al., 2014). in addition, a considerable amount of indoor testing of landslide mechanisms has been performed (wang et al., 2018; huang et al., 2019). currently, the analyses and prediction of landslide hazards are based on indoor tests, field geological survey methods, and lsm (huang et al., 2009; wang et al., 2019). these tests and methods are essential for identifying and preventing landslides, and for providing security for emergency rescue and disaster relief (fan et al., 2018; wang et al., 2018). however, these methods cannot precisely calculate the hazardous effects of catastrophic landslide dynamics or conduct risk assessments such as the run-out distance, impact force, and entrainment that are basis for the design of remedial engineering measures. additionally, numerical simulation methods provide more disaster information because they are able to compute landslide motion parameters, such as flow speed, deposit thickness, and the danger zone. the numerical methods primarily include discrete element methods and the continuum method (mcdougall, 2017). the discrete element methods simulate a sliding mass as a muster of particles. according to newton’s second law, the interaction force among granules can be computed. this is appropriate for debris flow, such as the particle flow code (itasca, 2008) and matdem (liu & pollard, 2013). the continuum method is mainly based on the mass and momentum equations that incorporate the earth pressure theory and simulate the dynamic features of a sliding body (pudasaini & hutter, 2003; hungr, 2004). the continuum methods have been successfully applied in the simulation of geological hazards, such as landslides, avalanches, and debris flows (jeong et al., 2018; liu et al., 2014). the continuum methods have evolved into different models, such as d-claw (iverson & george, 2014), sassa-wang (wang & sassa, 2002), and dan (hungr & evans, 1996). based on the motion equation and fluid continuity equations, hungr established the dynamic analysis software of landslides (dan-w) along with different rheological models, such as the frictional model, the plastic model, the bingham model, and the voellmy model. in addition, a sliding body is regarded as an equivalent fluid for analysis in dan-w. in this research, a numerical dynamic simulation analysis of the 2017 xinmo landslide in maoxian, sichuan, china was conducted. it was characterized by its rapid velocity and long runout distance. according to field survey data, dan-w was used to simulate the entire landslide process. in previous work, a comparison of the motion features of several similar long runout and highspeed landslides in the western areas of china was conducted. these studies examined the dynamics of the sanxicun landslide (yin et al., 2016), the panjin landslide (yang et al.,2019) and the wenjiagou landslide (xing et al., 2017). in this study, it was shown that the frictional model was capable of a more accurate inverse analysis of the entire landslide process. the parameters and the numerical model scheme of this research can help precisely compute the motion features of similar potential landslides. methods we conducted field geological surveying in the songpinggou area before and after the landslide hazard, and analyzed the geological conditions of the xinmo landslide, including the geologic structure, formation lithology, and precipitation. we also used the multi-temporal remote sensing images before the hazard from the iknos sensor (resolution = 1.0 m), as well as the gf-2 sensor (resolution = 0.8 m). we use these images to find the key structures and the pre-sliding geomorphology before the landslide. according to the field geological survey data, unmanned aerial vehicle(uav) photograph, and the multi-temporal remote sensing images, the landslide area was divided into a landslide source zone, a debris avalanche zone, and an accumulation zone. in addition, dan-w, a landslide dynamic model, and multiple groups of rheological models were used to simulate the entire movement process of the xinmo landslide and to analyze its dynamic characteristics (i.e., movement velocity, movement distance, and thickness of the accumulation body). the reactivation mechanism and the disaster effects of the old landslide in the 1933 diexi earthquake were also analyzed. finally, it was determined that dan-w and the frictional-frictional-frictional (fff) model could effectively provide a reference for similar studies regarding the dynamic disaster effects of catastrophic rock landslides. the flowchart of this study can be described as shown in figure 1. figure 1. flow chart of the methodology. the xinmo landslide a catastrophic landslide hit diexi township, mao county, sichuan province, at 05:38:58 on june 24, 2017. located at the left bank of the firstgrade tributary of the minjiang river, songpinggou, the xinmo landslide lay at e103°39’ 03.4”, n32°04’ 09.4”, approximately 44.75 km from mao county (figure 2). the elevations of the landslide at the back and front edges were 3450 m and 2250 m, respectively. the elevation difference was 1200 m, the horizontal distance was 2800 m, and the volume of the accumulation body was 1640 × 104 m3. xinmo village was directly destroyed by the landslide, leading to 83 deaths. this landslide, which featured good concealment, a high altitude of initiation, and a huge impact, led to serious casualties and property losses. geologic conditions the xinmo landslide was located in longmenshan tectonic zone, which belongs to active area of tectonic movement (chen et al., 2015; jiang et al., 2002; wang et al., 2000; tang et al., 1991). the fugui mountain where the xinmo landslide was located was an earthquake-induced shattered mountain that formed in the epicenter zone of the 1933 ms7.5 diexi earthquake (with file:///e:/trabajos_2020/un/fc/revistas/esrj_24(2)/txt/javascript:; 171characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china an intensity level of x) and the strong motion zone of the ms 8.0 wenchuan earthquake (with an intensity level of ix) (figure 3). in terms of regional geology, this area has undergone s-n compressional tectonic stress several times (yin et al.,2017). the mountain slope behind xinmo village is in a dip slope, where the emergence strata consist of blocky, moderately thick and thin to moderately thick metamorphic quartz sandstone with phyllite and slate of the middle triassic zagunao formation (t2z), the occurrence of which is 184°∠ 53°. the strata show an undulating deformation because of compression. the steepdip structural planes, such as phyllite and slate, constitute a weak sliding zone. according to a field geological survey and the remote sensing image analysis (figure 4), there were two major preferred structural planes in the mountains of the sliding source area: (1) 100° ∠ 70° and (2) 350° ∠ 40°. the two figure 2. (a) location of the study area in china and (b) location of the xinmo landslide in the study area (base figure reproduced with kind permission from google earth). figure 3. regional seismic intensity map of the xinmo landslide.1 the location of the xinmo landslide, 2 city, 3 province, 4 provincial boundaries, 5 intensity contour of the diexi ms 7.5 earthquake in 1933, 6 intensity contour of the wenchuan ms 8.0 earthquake in 2008. sets of structural planes cut the rock into a grid on the plane and combined it with the weak structural plane of the bedding layer to separate the rock into a medium-thick layer and a thick layer of massive cataclastic rock. this caused the formation of readily sliding geological blocks. according to the rainfall data provided by the meteorological bureau of mao county (2011-2016), the multi-year average precipitation is 490.7 mm (figure 5) and the maximum daily precipitation is 75.2 mm. rainfall is concentrated from may to september when the cumulative precipitation can reach 353.3 mm or 72% of the annual precipitation. in 2017, data collected from meteorological stations at songpinggou and diexi township near the landslide showed that the cumulative precipitations from may 1 to june 23 exceeded 210 mm. the rain infiltrated along the mountain rocks and thus led to an increase in the hydrostatic pressure, which intensified the landslide failure. 172 longwei yang, wenpei wang, nan zhang, yunjie wei figure 4. characteristic zoning and comparison of the multi-temporal remote sensing images of the xinmo landslide(yin et al.,2017,modified). a overview zoning before sliding; b overview zoning after sliding; c-d remote sensing images before sliding, c---july, 2005 (iknos, r=1.0 m), d--april, 2017 (gf-2, r=0.8 m). e the photography before sliding(2017-06-22,to the northeast). figure 5. curve of the rainfall process around xinmo village zoning characteristics of typical chain disasters according to the movement and accumulation characteristics of the slide body, the landslide area was divided into a landslide source zone, a debris avalanche zone, and an accumulation zone (figures 4 and 6, table 1). an analysis of the remote sensing images before and after the landslide and a field survey indicated that the total area of these zones was 143.1 × 104 m2. the area’s longitudinal length, horizontal distance, elevation difference, maximum width in the front edge, and minimum width in the rear edge were approximately 3000 m, 2800 m, 1170 m, 1500 m, and 300 m, respectively. landslide source zone figure 7 shows the landslide source zone for the xinmo landslide. based on the field geological survey, the plane shape of the slip source region 173characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china was approximately “rectangular”. the landslide source zone had a dip length of approximately 400 m. the zone had a maximum thickness of 40 m (20 m thickness, on average), a width of 320 m, and a size of 320*104 m3. however, the rock mass in the landslide source zone could have increased up to 1.4 times its original volume when it cracked and broke up (yin et al., 2010). the mother rock strata were composed of blocky thick quartz sandstone, slate, and phyllite of the middle triassic zagunao formation. the elevation of the landslide source zone at the rear edge was 3450 m, the toe of the landslide source zone was about 3100 m, and horizontal distance of the landslide source zone was 350 m. the landslide source zone contained a group of nearly nne-ssw large joints or fracture zones that shared the same direction as the sliding direction, which provided the bulging deformation and the sliding of the rock mass with dominant surfaces of weakness. in addition, a deformation body lay on the west of the source area, with a volume of approximately 450 × 104 m3 (figure 7a). it was composed of bedding rock mass and an old residual slope accumulation, and it slid downward for 40 m. the inner part of the deformed body was broken, and its front and east sides were free faces. there was a small collapse that occurred in the front boundary of the deformed body. figure 6. engineering geological section of the xinmo landslide before and after sliding (yin et al., 2017, modified). (a) longitudinal profile before sliding, (b) longitudinal profile after sliding. 1 stratum of triassic zagunao, 2 accumulation area of the old landslide, 3 residual sliding mass of the old landslide, 4 accumulation area of the new landslide, 5 debris flow accumulation area, 6 sandwiched phyllite or schist, and 7 typical point. figure 7 characteristic of the sliding source zone of the xinmo landslide at the ridge-top. 1 elevation, 2 slip direction, 3 boundary of the sliding source zone, 4 the unstable slop zone, and 5 reverse fractured belt. 174 longwei yang, wenpei wang, nan zhang, yunjie wei debris avalanche zone figure 8 shows the debris avalanche zone of the xinmo landslide. according to a field survey, the original deposits on the slope were the residual accumulation body of an old landslide that was triggered by the ms7.5 diexi earthquake that took place in 1933, and the clastic debris cone was formed by the action of gravity from the top (cheng et al., 2008). this zone formed along the gully, and it was a pipeline-type debris avalanche for which the slope angle was 27°. the highest elevation of the debris avalanche zone was 3450 m, the toe of the debris avalanche zone was about 2400 m, and horizontal distance of the landslide source zone was 750 m. the zone had a maximum thickness of 30 m (15 m thickness, on average). the exposure of bedrock at a height between 2700 m and 2750 m divided the accumulation body into upper and lower sections, for which the upper section was about 540 × 104 m3 in size. the lower section was about 330 × 104 m3 in size. in total, the volume of this accumulation body was 870 × 104 m3. accumulation zone according to the remote sensing image analysis and a field survey (figure 8), the accumulation zone had a dip length of approximately 900 m, a width of 350 m, a maximum thickness of 50 m (20 m thickness on average), and a size of 580*104 m2 in size. the accumulation area was located at an elevation between 2440-2290 m. the accumulation zone fanned out and primarily accumulated on the left bank of songpinggou. it washed up on the right bank, where a landslide dam formed that had a gentle slope. the rock strata in the accumulation zone were composed of blocky thick quartz sandstone and slate. rock blocks in the accumulation zone were even in lumpiness and were generally cuboids whose side lengths were 0.5-1.5 m. the newly formed landslide accumulation zone overlapped the old landslide mass in the form of a diffusion flow. the old landslide mass was estimated to have an 80 m average thickness and a size of 200*104 m3. as a result, the total volume of the accumulated body in this zone was the volume of the avalanche debris body and that of the old landslide mass, for a total volume of 780 *104 m3. the initiation mechanism and post-failure behavior of the landslide the catastrophic landslide was analyzed by combining the geological conditions and the different zone characteristics as follows: there were multiple groups of discontinuous structural surfaces at the rear edge of the xinmo landslide, which cut the thick blocky rock mass into cataclastic rock blocks and formed a visible fracturing bulging zone at an elevation of 3150-3450 m. rainwater infiltrated the rock mass through the visible fracture cracks. the medium weathered quartz sandstone belonged to the middle triassic zagunao formation (t2z). this kind of rock is susceptible to strength reduction induced by rainwater infiltration. in particular, there were two groups of anti-dip joint zones that led to a ridge-top cutout. a sliding surface could then be easily formed. the rock was more broken, and the mountain surface of the sliding source area was much looser under the action of the wenchun earthquake. under the action of the gravity on the cataclastic rock mass, some collapses appeared in the sliding source prior to the landslide. the impact loading of the landslide mass resulted in the activation of an accumulation body. in this case, the accumulation body that formed after 1933 became unstable and changed into a pipeline-type debris avalanche after it broke up. when the avalanche debris slid downward toward the old landslide accumulation body at high speed, diffuse avalanche debris accumulation was formed due to the open terrain and the gentle slope in the front. a direct hit by the avalanche debris in the middle pushed the old landslide accumulation body to the top, buried xinmo village, caused 83 deaths, pushed dilapidated houses forward 200 m to the southern bank of songpinggou, and formed a landslide dam. a dispersed accumulation, together with the air blast impact, was formed by the avalanche debris on the eastern and western sides. shrubs at the front edge were flattened outward. the xinmo landslide was activated at the top of mount fugui, 3450 m above sea level. the landslide slid for 2.8 km, reached 1637 * 104 m3, and formed a landslide dam after it blocked songpinggou. it was a typical chain disaster of avalanche debris triggered by a ridge-top landslide, which was similar to the sanxi landslide in dujiangyan, sichuan province (yin et al., 2016) and the catastrophic landslide in southern leyte, the philippines (catane et al., 2007). figure 8. accumulation characteristics of the rockslide-avalanche debris in xinmo village, maoxian, sichuan. (a) debris flow overlapped with the old landslide mass and (b) huge block overlapped with the debris flow. table 1 zoning of the material sources and accumulation of the xinmo landslide, maoxian name of zone provenance/ scraping volume (×106 m3) accumulation volume (×106 m3) notes landslide source zone 5.5 1.5 the accumulation volume was 1.4 times the volume of the rock sliding mass (3.9 × 106m3) avalanche debris zone 8.7 3.3 the zone had a maximum thickness of 30 m (15 m thickness on average). accumulation zone 2.2 11.6 the newly formed landslide accumulation zone overlapped the old landslide mass in the form of a diffusion flow. total 16.4 16.4 unstable slope zone excluded (the volume was 4.50 × 106 m3). 175characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china dan-w numerical simulation analysis dan-w, a model used to study landslide dynamics developed by hungr (1995), can be used to simulate an entire landslide movement process. in this study, the 3d numerical dan-w model was developed based on the twodimensional conditions shown in the a-a’ profile in figure 6. the sliding path widths were confirmed according to field survey data (figure 9). following the theory of the lagrangian analytical solution of the saint-venant equation, a sliding mass is considered to be an “equivalent fluid,” and it is divided into several blocks made up of certain materials. the equilibrium equation and physical equation of each block are built in the curvilinear coordinate system with their solutions to obtain the movement characteristics of a landslide, such as the movement velocity and the sliding distance (hungr et al., 2004). this is shown in equations (1) (6): f h bds p ti i= +γ α -sin (1) where f is the sliding force,  is the unit weight of the sliding mass, h represents each block height, b represents each block width, ds is the nominal length of each block, α is the slope angle, p is the internal tangential stress, t is the base resisting force, and i represents each block index: v g f t m h bd vi i i i= + ( )' ∆ γ (2) where v is the new velocity of the sliding mass movement. the new speed at the end of a time step is gained from vi ' , that is, the old velocity, g is the gravitational acceleration, ∆t represents each time step, m is the momentum flux, and the other parameters are consistent with equation (1): h v s s b bi j i i i i = ( ) +( )+ 2 1 1+ (3) where h is the average thickness of the sliding mass, j represents each block boundary index, i represents each block index, s is the value of the curve displacement, and the other parameters are consistent with equation (1): v v v y lr point i i i n = + +∑ ∑ =1 (4) where v is the entire volume of the rock landslide deposit, vr is the volume of the initial rock landslide, vpoint is the volume of the unstable mass, y is the yield rate, l represents the length of each block, and i represents each block index: y b di i i= × (5) in the expression, d represents the erosion depth. the other parameters are consistent with equations 1-4. d depends primarily on the thickness of the loose layer in a sliding path. in this work, previous digital elevation and post digital elevation models were established by superimposing a 1:2000-scale pre-event topographic map onto a post-event unmanned aerial vehicle mage map (figure 10). the post-event source depths were subtracted from the previous dem to assess the thickness of the sliding rock mass that was scraped during runout. the initial volume of the xinmo landslide was estimated to be approximately 500 * 104 m3, the thickness of soils was approximately 25120 m, and the maximum erosion depth was approximately 6 m. dynamic model description the velocity of the sliding mass and the height of the landslide accumulation mass were calculated using equations (1)–(5). in addition, the amount of resisting force that was encountered was determined using different types of rheological models. dan-w provides different rheology relationship models, which, in most cases of landslide movement analysis, are the frictional model and the voellmy model (hungr & mcdougall, 2009). the former is primarily used for landslides when there is no turbulence, and the particle sizes of the residual body are large. the latter is suitable for the simulation of landslides with fractured particles where there is a visible liquidized layer in the sliding mass (aaron et al., 2017). in the frictional model, it is assumed that the flow of a sliding mass is subject to effective normal stress acting on each block. its resistance, τ, is denoted as follows: τ σ γ ϕ= ( )1 tan (6) where γ µ is the pore pressure coefficient, that is, the ratio of the pore pressure to the total pressure, φ is the internal friction angle, σ is the total stress perpendicular to the direction of the sliding path, and τ is the resistance at the bottom of the sliding mass. the shear stress of the voellmy model can be expressed as follows: τ σ γ ξ = +f v2 (7) where f is the friction coefficient of the sliding mass, γ is the unit weight of the materials, ν is the migration velocity of the sliding mass, ξ is the turbulence coefficient, and τ is the resistance at the bottom of the sliding mass. model selection and input parameters the movement track of a sliding mass, the appropriate rheology relationships, and the material parameters are three key factors that influence a numerical simulation of a landslide. first, the pre-landslide form line, the post-landslide form line, and the movement track of the sliding mass (figure 4b) were determined based on a field geological survey and data acquired by uav (unmanned aerial vehicle) photography. the rheological model that was best suited to the field geological conditions was selected from different models based on the characteristics of the xinmo landslide during various movement stages. it was appropriate to use the frictional model because the particle size of the residual body in the landslide source zone was large, there were noticeable marks of friction on the sliding surface, and the sliding velocity was low in the landslide source zone. the frictional (f) and voellmy (v) models were applied to the avalanche debris zone and the accumulation zone. according to the range of the avalanche debris, the simulation through three regions (the landslide source zone, the debris avalanche zone, and the accumulation zone) followed the frictional-frictional-frictional(fff), figure 9. the 3d numerical dan-w model employed for the landslide along the twodimensional sliding pathway illustrated by section a-a’ in figure.6 figure 10. a the previous digital elevation model and b the post-digital elevation model. 176 longwei yang, wenpei wang, nan zhang, yunjie wei frictional-voellmy-voellmy(fvv), frictional-frictional-voellmy(ffv), and frictional-voellmy-frictional(fvf) models, respectively, as shown in table 2. the numerical results that were obtained for different model combination were compared to the runout distance of the actual event. from these results, we could conclude that the fff model was the more practical model (see the “rheology model comparison” section). the result of the landslide dynamic analysis greatly depended on specialized knowledge of the authors and the data used (hungr & mcdougall, 2009). as a result, a trial-and-error method was used to perform considerable numerical calculations while following the findings of hungr and other researchers (hungr & mcdougall, 2009; sosio et al., 2008; xing et al., 2015) to probe the dynamic characteristics of the xinmo landslide. according to the results of previous research (hungr & evans, 1996; mcdougall & hungr, 2004), suitable values for the rheological model parameters were primarily found through trial and error. in the frictional model, the excess pore water pressure rose due to rainfall infiltration (musso, 2004; federico & cesali, 2017), and the γµ value was set to 0.6 through trial and error (with values ranging from 0.5 to 0.8). the dynamic friction angle b =17°(with values ranging from 10° to 30°) according to the literature (xing et al., 2015). subsequently, for the voellmy model, the friction coefficient f=0.3 (tan 24°) and the turbulence coefficient  = 450 m/s2 (equivalent in the range of 200-500 m/s2) were investigated. the erosion depth was set to 6 m, which was approximately the maximum erosion depth in the debris avalanche zone based on the field geological survey data. the longitudinal stress in the sliding mass was influenced by the internal friction angle, i. finally, using the engineering analogy method, the typical strength testing index was i =21°and the density was  =20 kn/m 3. the primary choices for values were due to previous experience and due to trial and error. in addition, individual values were chosen, which appropriately utilized the actual motion characteristics of the xinmo landslide based on the field geological survey data. as shown in table 3, the parameters and models were chosen based on the trial and error method and chosen in order to achieve simulations that mimicked the movement characteristics of the landslide. table 2. hydrodynamics model of the xinmo landslide model landslide source zone debris zone accumulation zone fff frictional frictional frictional ffv frictional frictional voellmy fvv frictional voellmy voellmy fvf frictional voellmy voellmy table 3. parameters of the frictional–voellmy model used for the xinmo landslide. model unit weight, (kn/m3) internal frictional angle, i(°) friction angle, b(°) pore pressure coefficient, γµ friction coefficient, f turbulivity, (m/s2) frictional 20 21 17 0.6 voellmy 20 21 0.3 450 results analysis analysis of movement velocity the movement velocity of the sliding mass was analyzed according to the results calculated by the fff model in dan-w. if the activation speed of the landslide in the source zone was zero, the movement velocity of the sliding mass showed an undulating increase at a horizontal distance of 500–2000 m (figure 11). the movement velocity of the sliding mass increased sharply upon the activation of the landslide, but the rate decelerated and even declined under the effect of friction against the sliding surface. however, the movement velocity of the sliding mass soared at horizontal distances of 1200 m and 1800 m, for which the friction decreased sharply and the slope steepened due to bedrock exposure. when the sliding mass reached a horizontal distance of 2000 m, the movement velocity peaked at 74 m/s, while the sliding mass provided continuous loading on the old landslide accumulation and pushed its upper portion to move. the sliding mass decelerated on the open terrain, but it still moved at 49 m/s when it destroyed and buried xinmo village at a horizontal distance of 2400 m. the sliding mass did not stop until it rushed to a horizontal distance of 2780 m. figure 11. variation of the velocity versus the sliding range. movement distance and thickness of the accumulation body as shown in figure 12, the numerical simulation of the xinmo landslide was performed using the fff model. subject to the fracturing bulging effect, the shear displacement of the landslide occurred at a horizontal distance of 400 m. in 34.5 seconds, the front edge of the landslide reached a horizontal distance of 1900 m, which was on top of the old landslide accumulation. the front edge of the landslide continued to move forward, scattered, and accumulated along the way when the terrain was open with a gentler slope, which lasted for 85.5 seconds. it finally stopped moving at a horizontal distance of 2780 m. the shear displacement of a rock mass with a size of 550 × 104 m3 was triggered with the action of the fracturing bulging effect in the landslide source zone (figure 12a). it reached a horizontal distance of 600 m and rushed to the upper part of the accumulation body on the slope in 9.7 seconds (figure 12b), scraping the slope surface at a depth of 2 m (the red line in figure12). when the sliding mass moved to the tail end of the upper part of the accumulation body on the slope at 17.2 seconds (figure 12c), the total volume of the landslide reached 1010.4 × 104 m3. the sliding mass entered the lower part of the accumulation body at 20.22 seconds after it moved horizontally for 1000 m (figure 16d) and continued to scrape the slope surface at a depth of 3–5 m. the sliding mass increased to 1390 × 104 m3 when it reached a horizontal distance of 1800 m (figure 12e). when it moved 100 m forward horizontally to 1900 m at 30.54 seconds, the sliding mass started loading on the upper part of the old landslide slope and pushed it, moving it along. after the sliding mass moved to a horizontal distance of 2400 m at 42.61 seconds (figure12f), xinmo village was buried by the sliding mass, with some of the houses pushed forward 100– 200 m. then the sliding mass moved forward and finally stopped at 2780 m at 60 seconds after the landslide took place. the sliding mass in the avalanche debris zone and the accumulation zone continued to scatter and accumulate at 60–120 seconds. the total volume of the landslide reached 1579.6 × 104 m3 at 120 seconds. in the end, the average thickness of the accumulation body was 20 m and the volume of the residual body reached 309.0 × 104 m3 in the landslide source zone, as shown by figure 12h. the average thickness and the volume of the landslide-avalanche debris accumulation zone were 18 m and 1270.2 × 104 m3, respectively. the total volume of the landslide was 1579.2 × 104 m3, whereas the total volume of the landslide shown by the field survey and the uav aerial photography was 1637.6 × 104 m3. the failure to fully reflect the complex topography in dan-w led to a deviation between the calculated result and the field geological survey, but the error was still acceptable. speed and characteristic analysis of typical points based on the actual situation as assessed during the field investigation, four points with horizontal distances x of 400 m, 1000 m, 1900 m, and 2400 m were selected as typical points for the analysis, as shown in figure 6. 177characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china (1) the point with a horizontal distance x = 400 m was located at the landslide shear outlet (figure 6). during the period from 0 to 10 seconds, the speed of this point increased abruptly, reaching 22.5 m/s at 10 seconds. this indicated that the landslide was initiated with an accelerated shear in a very short period of time (figure 13a). from 10 seconds to 120 seconds, the speed of this point decreased, and the rate of decrease became larger. this was because the residual sliding body was hindered by the negative topography. in the final steady state, the deposit could become 15 m thick (figure 13b). (2) the point with a horizontal distance x = 1000 m was located at the end of the upper portion of the old landslide (figure 6). the point began at 17.2 seconds (figure 13c). because the bedrock at this point was exposed, the sliding body was less hindered, leading to a speed as great as 52.5 m/s. from 20 seconds to 120 seconds, the speed gradually decreased. at 25 seconds, the deposit thickness at this point could be as great as 27.5 m. in the final steady state, the thickness of the deposit at this point could reach 6.5 m (figure 13d). (3) the point with a horizontal distance x = 1900 m was located at the end of the lower portion of the old landslide (figure 6). from this point, the terrain became wider and the slope became gentler, resulting in a maximum speed of 69 m/s at this point, and the speed remained high with very slow attenuation (figure 13e). the deposit thickness at this point reached 16 m from 34 seconds to 42 seconds and reached 20 m in the final steady state (figure 13f). (4) the point with a horizontal distance x = 2400 m was located in the xinmo village (figure 6). at approximately 40 seconds, the front edge of the sliding body reached the xinmo village with a speed close to 50 m/s (figure 13g). at the same time, due to the blockage by houses, the deposit thickness reached 27.5 m in 10 seconds (figure 13h), which was sufficient to cover seven-story houses, causing devastating damage to xinmo village. figure 13. changes in the speed and the deposit of typical points over time discussion rheology model comparison the diagram of the xinmo landslide speed variation with slippage obtained using different rheological models is shown in figure 14. the diagram of the xinmo landslide speed variation with slippage that was obtained using different rheological models is shown in figure 14. it can be seen that the runout distances of the ffv model and of the fvv model were relatively large. the sliding distances of the landslide that were calculated using both models exceeded 3 km, which disagreed with the measurement data. the fff model and fvf model results were more accurate. when estimating the sliding distance and the movement velocity of a landslide that was in the range of millions or even tens of millions of cubic meters in size, the average speed calculated using the sled model of the xinmo landslide reached 68 m/s (yin et al., 2017). the calculation of the sled model was similar to that of the fff model, whereas the movement liquidation effect of the sliding mass in the avalanche debris zone was considered in the fvf model, which resulted in a smaller result. the collision and friction marks between the blocks in the debris flow area that were found in the field geological survey were obvious, and the blocks displayed the state of the fragments. in addition, according to the reports of local residents, there was a gully in the debris flow area with a depth of approximately 1 m. the water in the gully dried up two years before the landslide occurred. there figure 12. scenario of the 2d depth profile of the rockslide-debris run-out of the xinmo landslide. 178 longwei yang, wenpei wang, nan zhang, yunjie wei was no loose liquefaction layer on the surface of the debris flow region based on the field survey. therefore, the frictional model was more of a good fit than the voellmy model for the avalanche debris zone. what is more, the voellmy model included a turbulence term to account for velocity-dependent resistance (eq. 7). the voellmy model limited flow velocity. in contrast, many studies have shown that the xinmo landslide was a high-speed and long-runout landslide. as indicated by the result (figure 15), the fff model was used to simulate the entire movement process of the xinmo landslide. figure 14. speed contrast diagram of five models of the xinmo landslide figure 15. maximum extent of the xinmo slope runout for different rheological models in dan-w . 1 calculation profile in dan-w. 2 the runout distance for different models, in m. 4 the actual maximum runout distance, in m. 4 the boundary of the xinmo landslide. numerous studies have been conducted on the xinmo landslide. xu and ouyang analyzed the dynamic disaster effects of the xinmo landslide using a number of numerical simulation software programs and the corresponding models. these included pfc software and the linear elastic rheology model, matdem software and the linear elastic rheology model, massmov2d software and the voellmy rheological model, and massflow software and the coulomb frictional model. the movement speed and deposit thickness of the xinmo landslide were calculated. the results showed that the landslide movement speeds that were calculated using massflow and massmov2d were relatively slow, under 50 m/s, and the landslide movement speeds that were calculated using pfc and matdem were in the range of 35-70 m/s (scaringi et al., 2017; ouyang et al., 2017; xu et al.,2017). chen et al. used dda software to simulate the dynamic disaster effect of the xinmo landslide and they analyzed the speed of the landslide movement to xinmo village (chen et al., 2018). unlike these earlier studies, this study was based on the features of the frictional model and the voellmy model, and the different disaster characteristics of the sliding source area, the debris flow area, and the deposit area were considered. appropriate models were selected for different locations along the landslide path. the fff model was found to be accurate at representing the dynamic disaster characteristics of the xinmo landslide. the results also showed that the ffv and fvv models led to calculation results, which indicated that the liquefaction phenomenon of the xinmo landslide was weak. hence, these models were not suitable for simulating the movement characteristics of the xinmo landslide. the fff model recommended in this study for the calculation of the xinmo landslide had a total movement time of 120 seconds and an average speed of 57 m/s. in addition, the results showed a maximum movement speed of 74 m/s at a horizontal distance of 1950 m for the sliding body. furthermore, the speeds at four characteristic points, such as the landslide shear outlet and xinmo village, were analyzed quantitatively, and the deposit thickness at each point was simulated. additionally, the scraping effect of the landslide was quantitatively analyzed and the scraping depth was calculated across a range of 3 m to 10 m during landslide movement. the fff model that was used in this study to simulate the xinmo landslide and its parameters can also be used to simulate the dynamic disasters of similar high-position rock landslides in the songpinggou region where the xinmo landslide was located and in other regions. investigation of the reactivation mechanism and the disaster effects of the old landslide in the 1933 diexi earthquake the diexi landslide in xinmo had a volume of 330 × 104 m3 due to its high position shear sliding. the continuous loading and deposition on the upper portion of the old landslide caused the upper portion of the old landslide to be reactivated. thus, the volume of loading that caused the old landslide body to figure 16. old slope geometry for the geo-slope simulation 179characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china be reactivated required investigation. in this study, the stability of the upper portion of the old landslide body was calculated using geostudio-slope/w. the upper portion of the old landslide body was modeled separately (figure 16) using the moor-coulomb model. the bedrock and cover layer were treated as two different materials, and the parameters were those shown in table 4. for the sake of computational convenience, the loading method was simplified by applying the force to the middle and rear portions of the old landslide body to simulate the loading in the calculation. table 4. parameters of the slope-w model used for the old xinmo landslide material unit weight (kn/m3) cohesion (kpa) internal friction angle (°) the old landslide(1933) 20 19 21 bedrock 23 21 32 the results showed the safety factor of the old landslide body prior to the disaster occurrence to be 1.15. after continuous loading, the safety factor of the upper portion of the old landslide body decreased gradually. when the load reached 125 × 104 m3, the safety factor(fos) was less than 1.0, and the upper portion of the old landslide body became unstable and slid. the relationship between the specific loaded volume and the safety factor is shown in figure 17. figure 17. relationship between the loading volume and the safety factor of the old landslide next, dan-w was used to conduct the dynamic simulation of the old landslide body only. there are many similar old landslide deposits in strong earthquake zones, and their reactivation is not caused by continuous loading only, but rather may also be induced by other factors, such as rainfall and earthquakes. for this reason, the sliding volume of the sliding source area was not considered in the calculation for this study, and the friction model was used. the parameters are shown in table 3. the results showed that the entire process of the old landslide movement lasted 80 seconds. the maximum movement speed was obtained as 53 m/s at a horizontal distance of x = 1900 m with a scraping depth of 3 m (figure 18a and b). at 42 seconds, the front edge of the sliding body reached xinmo village at a speed of 45 m/s with a deposit thickness of 45 m (figure 18c and d). the speed and the deposit thickness were sufficient to cause enormous damage to xinmo village. after the wenchuan earthquake, shattered mountains provided sources of material for landslides. therefore, it is very necessary to study the stability of old landslides in populated areas, calculate the scope of disaster mitigation for old landslides, and take the corresponding measures to prevent similar disasters. conclusion on june 24, 2017, a landslide struck xinmo village, mao county, sichuan province, buried the village, and caused 83 deaths. by focusing on the disaster-formation mechanism of the xinmo landslide based on a field geological survey, remote sensing satellite images, and probing the dynamic characteristics of this landslide using dan-w and geological rheological models, the following conclusions were reached: 1. on the basis of the field geological survey, the zoning characteristics of the xinmo landslide were analyzed by investigating the regional geological environmental conditions, such as the strata structure, easy-sliding structure, precipitation, and hydrogeology. accordingly, the landslide was divided into a landslide source zone, a debris avalanche zone, and a landslide-avalanche debris accumulation zone. 2. the entire movement process of the xinmo landslide was primarily divided into three stages: (1) the ridge-top landslide being sheared out, (2) the dynamic impact and the accumulation loading induced high-speed long-distance sliding of the residual soil body on the slope, (3) scattered accumulation along the western and eastern sides of the avalanche debris falling zone and the pushing of the upper portion of the old landslide accumulation body in the middle. 3. the movement characteristics of the xinmo landslide were successfully simulated using the landslide dynamic software dan-w and the fff rheological model. according to the calculation, the movement duration, maximum velocity, and average thickness of the accumulation body of the xinmo landslide were 120 seconds, 74 m/s, and 10–25 m, respectively. the reactivation initiation mechanism and the dynamic disaster effect of an old landslide under the continuous loading conditions were analyzed quantitatively. hence, dan-w and the fff model can effectively provide a reference for similar studies regarding the dynamic disaster effects of ridge-top rock landslides. acknowledgements this study was supported by the national key research and development program of china [no.2018yfc1505404], china geological survey (dd20190647, dd20179609, dd20190637). special thanks are given to prof. yueping yin, china institute of geo-environment monitoring. we are also grateful to prof.o.hungr for supplying a copy of the dan-w software. finally, we thank editors and reviewers for their thoughtful review and valuable comments to the manuscript. figure 18. dynamic disaster effect of the old landslide. (a) scenario of the 2d depth profile of the rockslide-debris run-out of the old landslide, (b) graph showing the velocity variation of the old landslide with slippage change, (c) velocity variation of the spot at xinmo village, and (d) thickness variation of the spot at xinmo village. prof.o.hungr 180 longwei yang, wenpei wang, nan zhang, yunjie wei references aaron, j., hungr, o., stark, t. d., & baghdady, a. k. (2017). oso, washington, landslide of march 22, 2014: dynamic analysis. journal of geotechnical and geoenvironmental engineering, 143(9), 05017005. althuwaynee, o. f., pradhan, b., park, h.-j., & lee, j. h. (2014). a novel ensemble decision tree-based chi-squared automatic interaction detection (chaid) and multivariate logistic regression models in landslide susceptibility mapping. landslides, 11(6), 1063–1078. catane, s. g., cabria, h. b., tomarong, c. p., saturay, r. m., zarco, m. a. h., & pioquinto, w. c. (2006). catastrophic rockslide-debris avalanche at st. bernard, southern leyte, philippines. landslides, 4(1), 85–90. chen, k. t., & wu, j. h. (2018). simulating the failure process of the xinmo landslide using discontinuous deformation analysis. engineering geology, 239, 269-281. chen, x., du, f. x., du, y. b., & jiang, l. w. (2015). research on the major gravity geological disasters distribution at upper minjiang river valley. journal of railway engineering society, 8, 20-24 (in chinese). cheng, g., wang, x., he, x., fan, j., & fan, j. (2008). outburst risk of barrier lakes in sichuan, china. journal of mountain science, 5(3), 189–193. evans, s. g., guthrie, r. h., roberts, n. j., & bishop, n. f. (2007). the disastrous 17 february 2006 rockslide-debris avalanche on leyte island, philippines: a catastrophic landslide in tropical mountain terrain. natural hazards and earth system sciences, 7(1), 89–101. fan, x., juang, c. h., wasowski, j., huang, r., xu, q., scaringi, g., van westen, c. j., & havenith, h.-b. (2018). what we have learned from the 2008 wenchuan earthquake and its aftermath: a decade of research and challenges. engineering geology, 241, 25–32. federico, f., & cesali, c. (2017). coupled effects of pore-water pressure evolution, slope curvature, and mass variation on the kinematics of rapidly sliding fine-grained materials. international journal of geomechanics, 17(10), 04017088. doi:10.1061/(asce)gm.19435622.0000976. gao, y., li, b., gao, h., chen, l., & wang, y. (2020). dynamic characteristics of high-elevation and long-runout landslides in the emeishan basalt area: a case study of the shuicheng “7.23”landslide in guizhou, china. landslides. doi:10.1007/s10346-020-01377-8 . huang, d., zhong, z., & gu, d. (2019). experimental investigation on the failure mechanism of a rock landslide controlled by a steep-gentle discontinuity pair. journal of mountain science, 16(6), 1258–1274. huang, r. q. (2009). some catastrophic landslides since the twentieth century in the southwest of china. landslides, 6(1), 69–81. hu, w., scaringi, g., xu, q., & huang, r. q. (2018). acoustic emissions and micro-seismicity in granular slopes prior to failure and flowlike motion: the potential for early warning. geophysical research letters, 45(19), 10406-10415. hungr, o. (1995). a model for the runout analysis of rapid flow slides, debris flows, and avalanches. canadian geotechnical journal, 32(4), 610– 623. hungr, o., & evans, s. g. (1996). rock avalanche runout prediction using a dynamic model. proceedings of 7th international symposium on landslides. a.a. balkema. trondheim, june, 233–238. hungr, o., & evans, s. g. (2004). entrainment of debris in rock avalanches: an analysis of a long run-out mechanism. geological society of america bulletin, 116(9-10), 1240-1252. hungr, o., & mcdougall, s. (2009). two numerical models for landslide dynamic analysis. computers & geosciences, 35(5), 978-992. igwe, o. (2018). the characteristics and mechanisms of the recent catastrophic landslides in africa under ipl and wcoe projects. landslides, 15(12), 2509-2519. iverson, r. m., & george, d. l. (2014). a depth-averaged debris-flow model that includes the effects of evolving dilatancy. i. physical basis. proceedings of the royal society a: mathematical, physical and engineering sciences, 470(2170), 20130819–20130819. itasca, consulting group inc. (2008). pfc3d particle flow code in 3 dimensions. user’s guide; itasca, consulting group inc.: minneapolis, mn, usa. jeong, s. w., wu, y.-h., cho, y. c., & ji, s. w. (2018). flow behavior and mobility of contaminated waste rock materials in the abandoned imgi mine in korea. geomorphology, 301, 79–91. jiang, l. w., wang, s. t., & liu, h. c. (2002). research on hypergene geological hazards in the upstream of min river. journal of geological hazards and environment preservation, 1, 13-17（in chinese）. liu, c., pollard, d. d., & shi, b. (2013). analytical solutions and numerical tests of elastic and failure behaviors of close-packed lattice for brittle rocks and crystals. journal of geophysical research: solid earth, 118(1), 71–82. liu, k. f., wei, s. c., & wu, y. h. (2014). the influence of accumulated precipitation on debris flow hazard area. in landslide science for a safer geoenvironment; springer: cham, switzerland; pp. 45–50. mcdougall, s. (2017). 2014 canadian geotechnical colloquium: landslide runout analysis current practice and challenges. canadian geotechnical journal, 54(5), 605–620. mcdougall, s., & hungr, o. (2004). a model for the analysis of rapid landslide motion across three-dimensional terrain. canadian geotechnical journal, 41(6), 1084-1097. musso, a., federico, f., & troiano, g. (2004). a mechanism of pore pressure accumulation in rapidly sliding submerged porous blocks. computers and geotechnics, 31(3), 209-226. ouyang, c. j., zhao, w., he, s. m., wang, d. p., zhou, s., an, h. c., wang, z. w., &cheng, d. x. (2017). numerical modeling and dynamic analysis of the 2017 xinmo landslide in maoxian county, china. journal of mountain science, 14(9), 1701–1711. pudasaini, s. p., & hutter, k. (2003). rapid shear flows of dry granular masses down curved and twisted channels. journal of fluid mechanics, 495, 193–208. scaringi, g., fan, x., xu, q., liu, c., ouyang, c., domènech, g., yang, f., & dai, l. x. (2018). some considerations on the use of numerical methods to simulate past landslides and possible new failures: the case of the recent xinmo landslide (sichuan, china). landslides, 15(7), 1359–1375. sosio, r., crosta, g. b., & hungr, o. (2008). complete dynamic modeling calibration for the thurwieser rock avalanche (italian central alps). engineering geology, 100(1-2), 11–26. tang, r. c., wen, d. h., huang, z. z., wu, x. g., & lin, w. f. (1991). the quaternary activity characteristics of several major active faults in the songpan-longmenshan region. earthquake research in china, 7(3), 64-71(in chinese). wang, y., fang, z., wang, m., peng, l., & hong, h. (2020). comparative study of landslide susceptibility mapping with different recurrent neural networks. computers & geosciences, 104445. doi:10.1016/j. cageo.2020.104445. wang, y., duan, h., & hong, h. (2019). a comparative study of composite kernels for landslide susceptibility mapping: a case study in yongxin county, china. catena, 183, 104217. doi:10.1016/j. catena.2019.104217. wang, l. s., yang, l. z., li, t. b., & wei, h. (2000). evolution mechanism of jiaochang earthquake landslide on min river and its controlling. journal of geological hazards and environment preservation, 3,195199 (in chinese). wang, x., yin, y., wang, j., lian, b., qiu, h., & gu, t. (2018). a nonstationary parameter model for the sandstone creep tests. landslides, 15(7), 1377–1389. wang, x., zhan, h., wang, j., & li, p. (2018). on the mechanical damage to tailings sands subjected to dry–wet cycles. bulletin of engineering geology and the environment, 78(6), 4647-4657. 10.1016/j.cageo 10.1016/j.cageo 10.1016/j.catena 10.1016/j.catena 181characteristics and numerical runout modeling analysis of the xinmo landslide in sichuan, china wang, f. w., & sassa, k. (2002). a modified geotechnical simulation model for the areal prediction of landslide motion. in: proceedings of the first european conference on landslides, prague, czech republic; rotterdam, balkema, pp. 735-740. warrick, j. a., ritchie, a. c., schmidt, k. m., reid, m. e., & logan, j. (2019). characterizing the catastrophic 2017 mud creek landslide, california, using repeat structure-from-motion (sfm) photogrammetry. landslides, 16(6), 1201-1219. wang, y., fang, z., & hong, h. (2019). comparison of convolutional neural networks for landslide susceptibility mapping in yanshan county, china. science of the total environment, 666, 975-993. wu, y. l., ke, y. t., chen, z., liang, s. y., zhao, h. l., & hong, h. y. (2020). application of alternating decision tree with adaboost and bagging ensembles for landslide susceptibility mapping. catena, 187, 104396, https://doi.org/10.1016/j.catena.2019.104396. xing, a., wang, g., li, b., jiang, y., feng, z., & kamai, t. (2015). long-runout mechanism and landsliding behaviour of large catastrophic landslide triggered by heavy rainfall in guanling, guizhou, china. canadian geotechnical journal, 52(7), 971-981. xu, q., fan, x., & scaringi, g. (2017). brief communication: post-seismic landslides, the tough lesson of a catastrophe. natural hazards and earth system sciences discussions, 1–11. yang, l., wei, y., wang, w., & zhu, s. (2018). numerical runout modeling analysis of the loess landslide at yining, xinjiang, china. water, 11, 1324.doi:10.5194/nhess-2017-363. yin, y. p., cheng, y. l., liang, j. t., & wang, w. p. (2016). heavy-rainfallinduced catastrophic rockslide-debris flow at sanxicun, dujiangyan, after the wenchuan ms 8.0 earthquake. landslides, 13(1), 9–23. yin, y., wang, f., & sun, p. (2009). landslide hazards triggered by the 2008 wenchuan earthquake, sichuan, china. landslides, 6(2), 139–152. yin, y. p., wang, w. p., zhang, n., yan, j.k., wei, y. j., & yang, l.w. (2017). long runout geological disaster initiated by the ridge-top rockslide in a strong earthquake area: a case study of the xinmo landslide in maoxian county,sichuan province. geology china, 44(5), 827–841 (in chinese). yin, y., wang, w., zhang, n., yan, j., & wei, y. (2017). the june 2017 maoxian landslide: geological disaster in an earthquake area after the wenchuan ms 8.0 earthquake. science china technological sciences, 60(11), 1762–1766. zhao, c. & lu, z. (2018). remote sensing of landslides a review. remote sensing, 10(2), 279. doi:10.3390/rs10020279. zhang, m., & mcsaveney, m. (2017). rock avalanche deposits store quantitative evidence on internal shear during runout. geophysical research letters, 44(17), 8814-8821. zhang, m., yin, y., & mcsaveney, m. (2016). dynamics of the 2008 earthquaketriggered wenjiagou creek rock avalanche, qingping, sichuan, china. engineering geology, 200(1), 75-87. https://doi.org/10.1016/j.catena.2019.104396 1324.doi 10.5194/nhess 10.3390/rs erikajaramillo.indd 73 earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 73-84 application of gravimetry and electric tomography methods to obtain stratigrafic profiles: case study at university of quindío and puerto espejo area, armenia-colombia erika cristina jaramillo giraldo1, ximena vega manchola1 and hugo monsalve jaramillo1 1centro de estudios e investigaciones de la facultad de ingeniería (research center at university of quindío)ceifi, quimbaya group e-mail: hugom@uniquindio.edu.co e-mail: manchol@hotmail.com. resumen se describe la metodología usada para desarrollar campañas geofísicas de resistividad y gravimetría aplicadas al subsuelo. estos son métodos totalmente indirectos cuyo propósito es determinar los contrastes de densidad y resistividad en el subsuelo, para relacionarlas con los diferentes tipos de suelos en las áreas de estudio, localizadas al sur y norte de la ciudad de armenia-quindío (puerto espejo y universidad del quindío respectivamente), para determinar la geometría de perfiles longitudinales basados en pozos geotécnicos existentes los cuales solo proveen información unidimensional. la importancia del subsuelo es analizada usando métodos eléctricos de resistividad y gravimetría en proyectos de ingeniería civil. los factores más importantes que influencian los cambios en la resistividad eléctrica y valores gravimétricos para diferentes materiales geológicos también son analizados, así como los rangos de variación de estos valores, dependiendo de las condiciones climáticas y la formación de capas en cada área de estudio. para la tomografía eléctrica, se usa el arreglo electródico wenner-schlumberger debido a su alta resolución vertical y horizontal. para el método gravimétrico, se usa la técnica microgravimétrica porque permite mayor detalle en capas superficiales del subsuelo. los resultados obtenidos son presentados en perfiles eléctricos y gravimétricos en puerto espejo y la universidad del quindío, los cuales fueron útiles como parámetros de calibración de los rangos de resistividad y gravimetría para los tipos de suelos encontrados en estas áreas de colombia, dependiendo también de los factores de humedad y temperatura. el análisis y discusión de los resultados es lo que finalmente permite obtener conclusiones, importante para la escogencia de métodos de exploración geofísica como una alternativa para familiarizarse con las estructuras localizadas en el subsuelo. palabras clave: gravimetría, tomografía eléctrica, perfiles estratigráficos, abanico volcaniclástico, escarpe, armenia (colombia). abstract the methodology used to develop resistivity and gravimetry geophysical surveying applied to the sub-surface is described. these are totally indirect methods whose purpose is to determine the density and resistivity (manuscript received may 2005, paper accepted october 2005) 74 erika cristina jaramillo giraldo, ximena vega manchola and hugo monsalve jaramillo contrasts in the sub-surface, in order to relate them with the different types of soils in the areas of study, located south and north in armenia-quindío (puerto espejo and university of quindío respectively), to determine the geometry of longitudinal profiles based on existing geotechnical boreholes which only provide unidimensional information.the importance of sub-surface is analyzed, using electric resistivity and gravimetry methods in civil engineering projects. the most important factors influencing the changes in electric resistivity and gravimetry values for differerent geological materials are also analyzed, as well as the ranges of variation of these values, depending on climate conditions and formation of layers in each area of study. for the electric tomography, the electrodic array wenner-schlumberger is used due to its high vertical and horizontal resolution. for the gravimetric method, the microgravimetry technic is used because this allows better detail in shallow layers of the sub-surface. results obtained are presented in electric and gravimetric profiles in puerto espejo and university of quindío, which are useful as calibration parameters of resistivity and gravity ranges for the types of soils found in these areas in colombia, depending also on the humidity and temperature factors. the analysis and discussion of results is what finally permits to obtain conclusions, important for choosing the methods of geophysical exploration as an alternative to get familiar with the structures located in the sub-surface. key words: gravimetry, electric tomography, stratigrafic profiles, volcaniclastic fan, scarp, armenia (colombia). © 2005 esrj unibiblos introduction the city of armenia, colombia has been the location of several studies (ingeominas, 1999, espinosa, 2000, monsalve y vargas, 2002) as a result of the most recent earhquake in quindío departament (25-01-1999, mw=6.2). for this reason, scientists and investigators have tried to improve general knowledge in geophysical and geological aspects in quindío departament. the most recent interest for the understanding of the sub-surface in armenia has been the geophysical and structural modeling of urban area and surroundings in armenia project, which includes the present study, where the physical methods of gravimetry and electric tomograpy are used, to obtain stratigraphic profiles. the geophysical surveying methods, among which we can mention the electric resistivity and gravimetry, measure a series of physical properties of objects or structures in the sub-surface, that make them different from the surrounding site. the exploration of these contrasts between the properties of diverse materials and the encasing rocks is common in the exploitation of natural resources (petroleum, underground water, minerals etc) and in prospection of support for civil engineering, just like the surveying for the study of conditions for dams or buildings foundations, or the level of soil saturation. the objective of geolectric and gravimetric prospection is to know the dimensions and depth of structures in the sub-surface, taking as reference the resistivity and gravity data, obtained in the site using different techniques. when geoelectric and gravimetric data are correctly analyzed with modern methods and using the appropriate mathematical and computational tools, they provide a cartography of gravity of the sub-surface, which permits to have a clearer idea of geometric characteristics of every layer below. the methodoly used to obtain geophysical imaging of sub-surface using gravimetry and electric tomography applied to university of quindío and puerto espejo, is described in general. in the same way, we present the results obtained from the the analyses and discussions are presented regarding these results. the present study finally considers the electric tomography and gravimetry as tools, having a high potential in the survey of sub-surface, allowing to obtain stratigraphic profiles in two dimensions that represent in a more exact way the real geometry of the layers. this is done in a better form than with currently available geotechnical methods. 75 aplication of gravimetry and electric tomography methods to obtain stratigrafic profiles: case study at university of quindío and puerto espejo area, armenia-colombia location the area under study is specifically located at university of quindío and puerto espejo, located northwest and southwest in armenia-quindío –colombia respectively, areas that have relatively flat topographies. inside each of these areas there is a geotechnical borehole, drilled during the seismic microzonification of armenia-quindío project (ais, 1999). figure 1 shows the location of the two zones under study and the soils that support them. figure 2 shows the geotechnical boreholes in them. the geographycal coordinates of gravimetric and tomographic profiles and the geotechnical boreholes are shown in chart 1. chart 1. geographical coordinates of final and initial stations for every gravimetric profile of geotechnical boreholes (p4: puerto espejo and p6: university of quindío) and the midpoints of every tomographic profile location puerto espejo university of quindío gravimetry n (m) 4° 31’ 5.65’’ 4° 31’ 10.02’’ 4° 33’ 14.93’’ – 4° 33’ 23.12’’ e (m) 72° 42’ 44.36’’ 72° 42’ 29.40’’ 72° 39’ 43.75’’ – 72° 39’ 30.85’’ electric tomography pto. medio n 4º 31.296’ 4º 33.466’ w 75º 42.849’ 75º 39.903’ geotechnical boreholes n (m) 4° 31’ 7.22’’ 4° 33’ 22.25’’ e (m) 72° 42’ 43.19’’ 72° 39’ 36.04’’ figure 1. location of tomographic and gravimetric profiles at university of quindío and in puerto espejo (modified from ingeominas, 1999). 76 erika cristina jaramillo giraldo, ximena vega manchola and hugo monsalve jaramillo figure 2. boreholes p4 and p6 corresponding to puerto espejo and university of quindío. geological setting the outcrops in the study area are geological formations belonging to cretaceous period (arquía complex), terciary period (fluvio volcanic and volcanic deposits) and quaternary period (quindío fan). these are, residual and saprofit deposits, coluvial deposits, pyroclastic fall deposits, alluvial sediments, and recent fluviotorrential sediments of quindío river. the area of interest is located on a large volcanosedimentary deposit, formed by alternation of pyroclastic flow deposits, laharic and fluviotorrential, alluvial and glacial materials in smaller proportion, mostly covered by volcanic tuffs (ash, , lapilli and volcanic dust), that covered a paleo-topography, generating round and low grade hills, which are the geographical shapes that characterize the region. the materials forming the big volcanic-sedimentary deposit come from the central cordillera, located east and northeast of quindío. these deposits are mostly products of volcanic activity originated at nevado del quindío, santa isabel, and cerro santa rosa volcanoes (gonzález and nuñez, 1991). espinosa (in press), assigns this unit the name of quindío fan and makes a morphological difference with others, concluding that its thickness is not constant, reffering to an increase from the apex, associated with distance to volcanic centers. espinosa’s work defines the terracing (see figure 3) of quindío and pereira fans as a series of descending levels from the apex to the distal part, for this reason, the quindío fan is divided in three sub-groups named as proximal, intermediate and distal fans, so that the closest fan to the apex is encased in an older unit. in the same form, the degree of incision in the oldest units (closest to the apex) is much bigger than that of recent units due to its bigger exposition to alluvial erosion. figure 3. formation of fans, note the terracing between the most recent units (4 and 3) in respect to the oldest (1 and 2) (espinosa, in press). 77 aplication of gravimetry and electric tomography methods to obtain stratigrafic profiles: case study at university of quindío and puerto espejo area, armenia-colombia methodology figure 4 shows the methodology used in gravimetric surveying. the objective of the study was to determine the contrast of densities and geometry (orientation and aproximate thickness) of sub-surface layers for the zones of puerto espejo and university of quindío. its reach permitted to classify the study as local type, for this reason the technique used was micro-gravimetry relative gravity testing was done every 10m along profiles of about 500m in longitude, using a worden gravimeter. the cycles were done in periods of less than 2 hours and by the same surveyor, seeking to minimize the level of human error. gravimetric profiles figure 4. flow diagram for the construction of gravimetric profiles. tomographic profiles figure 5. flow diagrams for the construction of tomographic profiles. 78 erika cristina jaramillo giraldo, ximena vega manchola and hugo monsalve jaramillo the bouguer anomaly data obtained, along with known geologic information of the region, were processed using a computer program for inversion of gravimetric data (gm-sys v.3.6), from which a geological model of the zones was obtained. figure 5 shows the methodology used in tomographic exploration. the objective of the study was to determine the contrast in resistivity in the subsurface layers in two dimensions, confirming the presence of materials associated to those found in geotechnical and geological studies for the zone. the method used to determine the stratigraphic profiles was the electric resistivity, since the objective was to find the geometry of the layers trough the contrast in their resistivities. the method of electric tomography was applied , using the wennerschlumberguer array due to the fact that bi-dimensional profiles were needed and for the necessity of obtaining complementation parameters to regional geotechnical and geological studies which are unidimensional. this array provides a good resolution in both vertical and horizontal directions. for the area of university of quindio, a 180m profile was made (see figure 1), with electrode spacing of 5 meters, reaching a 38.7 meter deep research , while in puerto espejo the profile reached 144 meters, with a spacing between electrodes of 4 meters and a 31 meter depth. the data of apparent resistivity obtained in the processing of the information were used to find the values of real resistivity using a program specialized in inversion (res2dinv, v. 3.48). results gravimetry measured bouguer anomaly values in the range of -33,23 to -22,29 mgal in puerto espejo and 57,79 a 65,27 mgal at university of quindío, resulted in the following models:. puerto espejo zone the profile shown in figure 3 is extended along a gravimetric line with a longitude of 543,68 meters, heights between 1327,52 and 1349,16 meters above sea level and the values for bouguer anomalies increase to the north of the profile as shown in figure 6. figure 6. model of densities distributions in puerto espejo and its anomaly adjustment (above). 79 aplication of gravimetry and electric tomography methods to obtain stratigrafic profiles: case study at university of quindío and puerto espejo area, armenia-colombia the geometry of the layers was stablished, obtainig values in density between 1.5 and 1.75 g/cm3, increasing with depth. four layers relatively horizontal for the profile were obtained (see figure 6.), which show an increase in thickness of the fan to the south, presenting a lightly dipping structure. the deposits found in this regions were associated to la florida fan, which belongs to the group of distals fans. the first layer, with a density of 1.5 g/cm3 shows uniformity in thickness (17 m. aproximately) as shown in geotechnical borehole (see figure 2) associated to volcanic ash. for this profile, the presence of arquia complex is observed below 100m depth approximately, in the north direction of the profile and shows a tendency to deepen in the south direction, due to accumulation of quaternary deposits. university of quindío the profile shown in figure 5 is extended along a gravimetric line with a longitude of 581,06 meters, heights between 1511.67 and 1544,725 meters above sea level, presenting bouguer anomalies in the range of 57,79 to 65,27 mgal, from north to south respectively, as shown in figure 7. figure 7. model of densities distributions at university of quindío and its respective anomalie adjustment(above). the values for density obtained are in the range 1.5 to 1.75 g/cm3 increasing as depth increases. five relatively horizontal layers were also obtained (see figure 7), these show an increase in the thickness of the layers toward south, showing a slightly dipping surface. the deposits found in this sector were associated to circasia fan, which belongs to the intermediate fan group. the first layer presenting a density of 1.5 g/cm3 shows a variation in depth from 11 to 24 m. confirming the thickness of volcanic ashes (21 m.) indicated in geotechnical boreholes p6 (see figure 2), increasing thickness toward north. the presence of arquia complex is after the 200m deep. aproximately in the north side, with tendency to deepen to the south direction of the profile, due to accumulation of quaternary deposits. 80 erika cristina jaramillo giraldo, ximena vega manchola and hugo monsalve jaramillo electric tomography a model with real distribution of resistivity in the layers of the sub-surface was obtained for every zone under study, which is represented by the variation in colors. after analyzing tomographic profiles with the corresponding mechanical boreholes, the resistivity ranges for every type of soil were computed. the variables involved were: natural humidity, unitary weight, bearing capacity, swave speed and rqd (r=?, q=?, d=?). puerto espejo the correlation of data obtained in mechanical boreholes p4 and the resistivity values found, allowed to obtain a definitive model for puerto espejo, which reflects a great influence of natural humidity in the values of resistivity in the upper layers. this can be observed in the graph obtained during the analysis (figure 8), while in lower layers, the dominant factor is the degree of cementation and meteorization. figure 8. variation in resistivity with natural humidity. although the variation of resistivity with natural humidity is not linear, there is a tendency in the resistivity, to decrease while humidity increases. a model was obtained in which 7 areas were found (see figure 9) every one with special characteristics as described: figure 9. areas differenciation according to resistivity and natural humidity ranges for tomographic profile in puerto espejo. 81 aplication of gravimetry and electric tomography methods to obtain stratigrafic profiles: case study at university of quindío and puerto espejo area, armenia-colombia area 1 presents the lowest humidity condition, being reflected in the largest resistivity range, which is, between 4500 and 9500 ohm-m and is associated to lime sandstones. area 2 exhibits an intermediate humidity condition and resistivity values between 2200 and 4500 ω.m., which is associated to sandy limestone and lime clays. area 3 exhibits the highest humidity conditions, as reflected by the smallest resistivity range, this is, between 335 and 2200 ω.m. and is associated to sandy limestone and lime clays. area 4 shows a high humidity condition and certain degree of cementation. due to the position of this area inside the layer of the volcanic ashes, cementation is dominant over humidity, which is reflected in an increase in the resistivity range (4500 a 9500 ω.m.), associated to cemented sandy limestone. below the volcanic ashes layer, there is a lithological material stratum which shows a constrast with the alter(?) and that corresponds to residual soil with resistivity values ranging between 9500 and 26300 ω.m., the aproximate thickness is 8 m. (see figure 2). finally, a layer presenting the highest resistivity values in the whole tomographic profile (from 26300 ω.m. and aheah), which is associated to volcanic flow after 26 m in depth (see figure 2). university of quindío the correlation of data obtained in borehole p6 along with the resistivity values found, determined the defínitive model for university of quindío, where a high influence of natural humidity is observed in resistivity values for upper layers, from which we produced the graph represented in fig. 10. on the other hand, it was possible to note that the dominant factor in lower layers is the degree of cementation and meteorization, same as in the model found for puerto espejo. figure 10. variation of resistivity with natural humidity. for this sector there is a linear variation of resistivity with the humidity, presenting an inverse variation of resistivity values with natural humidity. we can see here that the atypical values belonging to residual soil and volcanic flow do not make part of the chart, since these deposits are deeper, and there is no significant influence of natural humidity, but there is a high influence of cementation and metheorization. 82 erika cristina jaramillo giraldo, ximena vega manchola and hugo monsalve jaramillo this way, a definitive model was found, with seven defined areas (see figure 11) as described: area 1 presents the lowest humidity conditions, being this reflected in the highest resistivity values, with a wide variation range (from 12000 to 47030 ω.m.), the superior and inferior limits of resistivity values for this area are high, due to the high percentage of compactation, since it is located below a parking lot. this material is associated to lime sandstone. zona 2 presents a midterm to high humidity condition and resistivity values between 2000 and 7315 ohm-m, wich is associated to sandy limestone and lime clays. area 3 presents the highest humidity condition, being this reflected in the lowest resistivity range (from 449 to 2000 ω.m.). it is associated to sandy limestone and lime clays. area 4 presents a midterm humidity condition and certain degree of cementation. due to the position of this area inside the volcanic ashes layer, there is a marked tendency of cementation upon the humidity condition, which is reflected in an increase in the figure 11. differenciation of areas according to resistivity ranges and natural humidity for a tomographic profile at university of quindio. resistivity range (7,315 to 18,548 ω.m.), associated to cemented lime clays. below the volcanic ashes layer there is a stratum of lithic material which contrasts with the latter, and corresponds to residual soil. here, values of resistivity ranging between 18,548 and 47,030 ω.m. are observed. the aproximate thickness is 8 m. (see figure 2). finally, a layer with the highest resistivity values in the tomographic profile is observed. (after the 47030 ω.m.), which is associated to volcanic flow and is presented aproximately after 26 m in depth. (see figure 2). analysis and results discussion gravimetry taking into account the terraced topography of the area, the gravimetric models of puerto espejo and the university of quindío show the tendency of terracing of quaternary deposits like the arquia complex, reflecting a lightly dipping structure. it is noticeable that arquia complex at university of quindío is presented at a bigger depth (? dificil entender el significado de la expresion) than in puerto espejo, possibly due to the distribution of quaternary deposits. in puerto espejo, these deposits are less antique in comparisson to those at university of quindío, allowing to find the arquia complex closer to the surface. for both sectors there is a 83 aplication of gravimetry and electric tomography methods to obtain stratigrafic profiles: case study at university of quindío and puerto espejo area, armenia-colombia tendency in the layers to increase their thickness toward the south direction of the profiles, reflecting a slightly dipping surface, indincating the form how the deposits have been accumulated. the micro gravimetry method allowed to analyze the layers of the sub-surface in both sectors in a more detailed way, although superficial. the comparison between the different profiles found however, with the existing ones, was only established to a maximum depth of 40 meters aproximately, since this was the biggest depth found in geotechnical boreholes so as in the tomographic survey. electric tomography it is observed that the range of values in resistivity of the sub-surface at university of quindío is bigger with respect to that of puerto espejo, possibly due to soil characteristics and age of the deposits, those at university of quindío are older, which is reflected in the cementation range, one of the most influencing factors in the increase of resistivity. the age of shallow deposits analyzed at university of quindio is relflected in a higher degreee of compactation, which has a direct influence in the increase of resistivity ranges. both models show a dominant tendency of natural humidity in the variation of resistivity values in the volcanic ashes layer, allowing to diffrenciate permeable soil from those non permeable, while in the lower layers (residual soil and volcanic flow) the dominant factor is the degree of cementation and meteorization. the electric tomography method allowed to view the transition between the different layers of the soil. (volcanic ashes, residual soil and volcanic flow), this is evidenced by the strong contrast in the resistivity values mapped. the anomalies found in every profile demonstrates the effectivity of the electric resistivity method in the survey of underground structures since it is possible to observe a good resolution of geometry and depth, providing civil engineers with an reliable alternative in the analysis of foundations, piping, ducts, or underground structures. a limitation in the electric tomography works applied to civil engineering is the necessity to be carried out on a natural surface (soil) (??) conclusions -the ranges of apparent resistivity for certain types of geologycal material and minerals found reported by various authors differ with the values found in the present study, we suggest this is because in those refferences the conditions under which the resistivity values were obtained are not considered. moreover, these studies have been made in other countries were the composition of the layers of the sub-surface may be different to those of this geographycal region. -analyzing the tomographic profile, according to the geotechnical data from the area under study, its is observed that in the layers of volcanic ashes, (upper layers) the dominant factor in the variation of the resistivity is the natural humidity, while in the lower layers like residual soil and volcanic flow the metheorization degree is dominant. the electric tomography method is useful as a tool of sub-surface surveying at shallow levels and the construction of stratigraphic profiles, an appropriate calibration of the ranges of resistivity is necessary however for the soils of the municipality and for different seasons of the year in order to consider climate factors. it is useful to know the tendencies of the subsurface structures using the electric tomography in civil enginnering, and more specifically in geotechnics, because the level of uncertainty in excavations is reduced for the foundation. with this, the number of tests is also reduced as well as the cost of construction projects. the gravimetric method is of great help for geological characterization studies as well as for the seismic microzoning, because it permits to identify the more dense layers of the sub-surface associated to rock formation such as arquía complex and diffrenciate them from the less dense like quindío fan. due to the difference in characteristics between soils at university of quindío and puerto espejo, there is a variation in the ranges of resistivity, being reduced in puerto espejo. at university of quindío, the resistivity ranges obtained for the same type of soil in puerto espejo were wider. it can be observed that the thickness of the volcanicsedimentary deposits, corresponding to densities between 1.5 and 1.75 g/cm3, is strongly related to the location of each area under study in respect to volcanic centers that originated them. for this reason, 84 erika cristina jaramillo giraldo, ximena vega manchola and hugo monsalve jaramillo the gravimetric profile of puerto espejo presents smaller thickness for those deposits, since this area is located at a distal fan (la florida), while the thickness determined for deposits below university of quindío is larger because they are located in a midterm fan (circasia), being consequently older. references • ais, asociación colombiana de ingeniería sísmica. 1999. estudio de microzonificación sísmica para orientar la reconstrucción de armenia – quindío. informe técnico-científico. bogota. • burger h. r., 1992. exploration geophysics of shallow subsurface, prentice-hall. 489p. englewood cliffs-new jersey. • chica a., 1998. geofisica aplicada –magnetometría y gravimetría-. cuadernos de ingeniería geológica. no. 5. 125 p. medellín, colombia. • duyoner, m. y aguirre, y. (19997) identificación y caracterización macroscópica de la secuencia de tefras de la ciudad de armenia (crq). • espinosa, a., 2000. mapa preliminar de fallas activas del quindío, zona central. informe final proyecto 152, universidad del quindío, armenia, 70 p., inédito. • espinosa, a. (en prensa). un modelo del cuaternario del quindío y risaralda. revista de la academia colombiana de ciencias exactas, físicas y naturales. bogotá. • espitia, m. y caro, p. capítulo 1 geología local y tectónica en ingeominas, (1999) vol ii. • gm-sys, gravity and magnetics modeling system. versión no. 3.60. mike gemperle, gerry connard, milt sagen, and myungmun bae northwest geophysical associates, inc. • loke, m. h. 1999. electrical imaging survey for enviromental and engineerimg studies: a practical guide to 2-d and 3-d surveys. malasia. p 58. • monsalve-jaramillo, h. y vargas-jiménez, c.a. 2002. el sismo de armenia (25-01-99, mw=6.2): análisis telesísmico y características de las ondas de cuerpo. revista ipgh, no 57, 21-56. • núñez, a. y gonzález, h. 1991. memorias mapa geológico generalizado del departamento del quindío. ingeominas, bogotá. keywords: jadeite; color-causing; standard light source; color; quality grading. palabras clave: jadeíta; causante de color; fuente de luz estándar; color; clasificación de calidad. how to cite item pan, x., guo, y., liu, z., zhang, z., shi, y. (2019). impact of different standard lighting sources on red jadeite and color quality grading. earth sciences research journal, 23(4), 371-378. doi: https://doi.org/10.15446/esrj.v23n4.84113 the purpose of this paper is to investigate the standard light source for grading and displaying the color of red jadeite and to classify the color. with raman spectrometer, ultraviolet-visible spectrophotometer and x-ray fluorescence spectrometer, the results show that, the fe 3+ is the main chromogenic mineral of red jadeite, which negatively correlates with the tonal angle, while the color of red jadeite has a positive correlation with the hematite content. the color of 120 red jadeite samples was examined by collecting the reflective signaled from the sample surface using an integrating sphere with the portable x-rite sp62 spectrophotometer based on cie 1976 l*a*b* uniform color space. the color parameters of jadeite samples under d65, a and cwf standard light sources were analyzed. the light spectrum of d65 light source is continuous, relatively smoothed with high color temperature, which makes the sample color close to that under the natural light and can be used as the best evaluation light source. a light source contributes to improve the red tone of jadeite, which is the best light source for commercial display of red jadeite. cwf light source can be used as the auxiliary lighting for color evaluation. the color of red jadeite is divided into five levels from best to worst using k-means cluster analysis and fisher discriminant analysis under d65 light source: fancy vivid, fancy deep, fancy intense, fancy dark and fancy. abstract impact of different standard lighting sources on red jadeite and color quality grading impacto por diferentes fuentes de iluminación estándar en la clasificación de jadeíta roja y la calidad del color issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.84113 el propósito de este trabajo es investigar la fuente de luz estándar para calificar y mostrar el color de jadeíta roja y clasificarlo. con el espectrómetro raman, el espectrofotómetro ultravioleta-visible y el espectrómetro de fluorescencia de rayos x, los resultados muestran que el fe 3+ es el principal mineral cromogénico de la jadeíta roja, que se correlaciona negativamente con el ángulo tonal, mientras que el color de la jadeíta roja tiene una correlación positiva con el contenido de hematita. el color de 120 muestras de jadeíta roja se examinó recogiendo el reflectante señalado desde la superficie de la muestra utilizando una esfera de integración con el espectrofotómetro portátil x-rite sp62 basado en espacio de color uniforme cie 1976 l*a*b*. se analizaron los parámetros de color de las muestras de jadeíta bajo las fuentes de luz estándar d65, a y cwf. el espectro de luz de la fuente de luz d65 es continuo, relativamente suavizado con alta temperatura de color, lo que hace que el color de la muestra sea similar al de la luz natural y puede usarse como la mejor fuente de luz de evaluación. una fuente de luz contribuye a mejorar el tono rojo de jadeíta, que es la mejor fuente de luz para la exhibición comercial de jadeíta roja. la fuente de luz cwf se puede utilizar como iluminación auxiliar para la evaluación del color. el color de la jadeíta roja se divide en cinco niveles, de mejor a peor, utilizando el análisis de conglomerados k-means y el análisis discriminante de fisher bajo la fuente de luz d65: fancy vivid, fancy deep, fancy intense, fancy dark y fancy. resumen record manuscript received: 12/07/2019 accepted for publication: 07/10/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 371-378 xin pan, ying guo*, ziyuan liu, zikai zhang, yuxiang shi school of gemology, china university of geosciences (beijing), beijing, 100083 china * corresponding author: guoying@cugb.edu.cn m in er a lo g y 372 xin pan, ying guo*, ziyuan liu, zikai zhang, yuxiang shi introduction jadeite is generally considered to be a high-pressure mineral (yang, yang & long, 2015), which is called the “king of jade” in china. it generally occurs as block, and is a rare rock type formed in subduction zones with two origins: the wholesale metasomatic replacement and the vein precipitation (yui & fukuyama, 2015; mevel & kiénast, 1986). based on possible formation models, here we propose to classify jadeite into two types: fluid precipitates (p-type) and metasomatic replacement (r-type) (tsujimori & harlow, 2012). geologically, jadeite is a fibrous mineral aggregate composed of pyroxene minerals with jadeite and omphacite as the main minerals. it typically has different percentages of diopside (camgsi2o6)/ kosmochlor (nacrsi2o6 ), iron-rich pyroxene (cafe2 + si2o6 ) and aegirine (nafe3 + si2o6 ), which coexist with jadeite as solid solutions (harlow & olds, 1987 ). to petrologically define the jadeite, the jadeite pyroxene content in rock must reach 90-95%, or it may be too rigidity to be defined as jadeite-rich rock. beside basic mineral composition, the composition of jadeite varies with place of origin, e.g., jadeite from the burma contains soda amphibole (including ekman rutile, blue amphibole, sodium tremolite and amphibole), albite, zeolite, tremolite, rutile, clinochlore, barium albite and chromite, etc (htein & naing, 1994, 1995). jadeite, a polycrystalline heterochromatic jadeite, is transparent and colorless when purified, which usually white due to the scattering of cracks, tiny grain boundary edges and fluid inclusions. the color of jadeite can be divided into primary one and secondary one. the crystal field theory shows that the color mechanism of jadeite is mainly composed of the following two forms: (1) metamorphic differentiation: the color-causing trace elements such as cr3+ and fe3+ migrated and aggregated, which then replaced al3+ in the crystal of jadeite naalsi2o6, (2) ductile deformation: jadeite could be ductile deformed due to the tectonic compression under high pressure and with cr3+and fe3+rich fluid that flows into jadeite crystal lattice (ying, zong, qi, zhang & wang, 2018). the jadeite is mostly blue-green or lividity and blue-black when ferrous (fe2+) and iron (fe3+) are mixed into it. the darkest black-green is caused by a small amount of iron oxides and omphacite (rossman, 1974). furthermore, charge transferring (fe2+-o-fe3+) between ferrous and trivalent iron can also form purple in nearly pure jadeite (ponahlo, 1999; hughes et al., 2000). the secondary color of jadeite refers to the color formed by the exogenous geological process after being exposed to the surface, which is generally produced by the micro-mineral impregnation on the fine boundary, e.g., jadeite contaminated by water-bearing iron oxides is commonly red to orange-brown, iron-rich clay minerals often bring dark green stripes to jadeite, while graphite makes jadeite gray to black (guo, wang & du, 2016a). color plays a significant role in jadeite quality evaluation, typically determining the quality and value of jadeite. the color grading of jadeite was first summarized and described by professionals using subjective words. china launched the national standard of gb/ t 238852009 jade grading in 2009 to classify the jadeite color using the colorimetric method under standard environment (guo et al., 2016a). the main parameters include color hue, lightness and chroma (guo, 2017a). in this standard, the color is measured based on spectrophotometry with standard a or d light sources being recommended. however, standard a light source (color temperature is about 2856k) and standard d light source (color temperature is about 5000-7500k) are two completely different ones (china national bureau of technical supervision, 2009). also, purple and red jadeite are still graded according to the standard of green jadeite with no separate color grading standard for red jadeite. currently, all kinds of gemstones in the world have their own color grading methods, e.g., ruby, sapphire, emerald, etc., while most of the color grading systems refer to the color grading system of gia color diamonds (china national bureau of technical supervision, 2010; 2016a, 2016b). diamond color is classified and described into nine levels via gia according to their chroma and color hue, including fancy vivid, fancy deep, fancy intense, fancy dark, fancy, fancy light, light, very light, faint. materials and methods one hundred and twenty natural red jadeite with good texture, well purity and polished surface were selected and prepared in size of 5mm *6mm to 7mm *8mm, which varies in color from orange red to brownish red with continuous change of color hue and lightness. some typical samples are shown in figure 1. raman spectroscopy was carried out using hre volution laser raman spectrometer from japan horiba company. excitation light source wavelength: 532 nm; resolution: 1 cm-1; laser power: 30-40 mw; scanning times: 3; single integration time: 10 seconds; testing range: 100-2000 cm-1. ultraviolet-visible spectrums were analyzed using uv-3600 uv-vis spectrophotometer. measurement conditions were listed as followings: detector conversion wavelength: 200-900 nm; grating conversion wavelength: 830.00 nm; s/r shift: normal mode; light source: automatic; detector lock: automatic; slit program: normal; beam mode: double beam; scanning speed: high; sampling interval: 0.5s; scan mode: single. all-element analysis was performed on ourstex100fa portable energy dispersive x-ray fluorescence spectrometer. measuring mode: monochrome or continuous x-ray with pd target; tube voltage: 40kv; tube current: -0.051ma; detector: sdd (siliocon drift detector); detector refrigeration mode: peltier refrigeration mode (-28oc); sample environment: low vacuum (103 pa); diameter and height (expandable) of sample chamber: 20 cm and 10-30 m; measuring time: 100-300 s. x-rite sp62 was employed to collect reflective signals from forsterite surface through integrating sphere based on cie1976 l*a*b* uniform color space system (figure 2). test conditions were described as following: reflection, not include the specular reflection; light source: d65 standard one; observer view: 2°; measuring range: 400-700nm; measuring time:<2.5s, wavelength interval:10nm; voltage: 220v; current: 50-60hz. figure 2. schematic diagram showing experimental device quantitative color analysis is based on l*a*b* uniform color space (cie, 1976), which consists of plane color a*, b* axis and l* vertical axis. the positive and negative ends of a* axis represent red and green, respectively, the positive and negative ends of b* axis represent yellow and blue, respectively, and the l* axis represents brightness. the chroma cab* and color hue hab * can be calculated by color a* and b*, respectively. c a b* * *= ( ) ( )      2 2 1 2  (1) figure 1. images showing typical red jadeite samples 373impact of different standard lighting sources on red jadeite and color quality grading fe3+ is typical 3d5 configuration, whose compounds are usually brown, while the charge transfer band of fe3+ usually occurs at 200 nm in the far ultraviolet region, indicating that the peak at the former region (200-320 nm) is resulted from the charge transfer between fe2+ and fe3+. therefore, it can be confirmed that fe3+ is the main color-causing mineral of red jadeite, while the gradation varies as a function of iron content, which can be expressed by the absorption intensity in the spectra. figure 5. impact of fe content on hue angle of typical samples x-ray fluorescence spectrometer of four samples with gradient color shows that the main elements of red jadeite are na, al and si, which is h b a ° = arctan * * (2) cie de2000 (e00) color difference formula was employed to calculate the color difference of jadeite under different background (perez et al., 2011), which is the latest color difference formula recommended by cie. the formula includes not only weighting function of lightness difference, color hue difference and chroma difference, but also the cross term of color hue difference and chroma difference as well as the term of improving the prediction ability of neutral grey color difference. compared with cie lab (e*ab) color difference formula lacking of visual uniformity (guo, 2017b), cie de2000 is an improved version that can better express fine chromatic aberration in green area. cie de2000 (δe00) color difference formula is as following: ∆ ∆ ∆ ∆ ∆ e l k s c k s h k s r c k st t c c k k t c 00 2 2 2 = ′    ′    ′    ′    cc k k h k s     ′    ∆ where δl’, δc’ and δh’ are lightness difference, color hue difference and chroma difference for certain group of color data; rt is the conversion function for resolving the interaction between color hue and chroma in blue region; sl, sc and sh are position functions to correct the visual uniformity in cie lab; kl, kc and kh are calibration parameters under experimental environment, while two groups of parameter combinations are widely used:  cie de2000 (1:1:1): kl=1ˎ kc=1ˎ kh=1(cie technical report: improvement to industrial colordifference evaluation, 2001; luo et al, 2001; ghinea et al, 2010);  cie de2000 (2:1:1): kl=2ˎ kc=1ˎ kh=1(mangine et al, 2005; perez et al, 2011) (guo, wang, li & dong, 2016b). this paper employs sensitive parameters to evaluate color difference: cie de2000 (1:1:1). the standard light source box is selected as experimental environment. the inner wall of the box is composed of standard neutral gray matte material. the standard light source tube is put on top to simulate the natural light source, which can be switched freely. three lighting sources were used based on the national standard gb/t 20146-2006 requirements for standard lighting sources, as well as the requirements for the detection and quality evaluation of jewelry and jade, and the uniformity of lighting sources (guo, zhang, li & zhang, 2018a). the d65 light source (related color temperature: 6504 k) was used for simulating sunlight in cloudy days. the a light source (related color temperature: 2856 k), a typical incandescent lamp, was selected for lighting in households or stores, and finally, the cwf light source of the cold white fluorescent (related color temperature: 4150 k) for the business and office in the united states was used for reference. results and discussion color-causing analysis of red jadeite raman spectroscopy was carried out on the sections with uniform color that can represent the average color of each sample. the raman spectroscopy measurement shows that all samples have raman peaks of jadeite, a collective mineral phase, indicating that all samples are natural jadeite. raman spectra of three samples with obvious color gradient were analyzed to identify their peaks (figure 3). these peaks represent lepidocrocite (γ-feo(oh)) and hematite (α-fe2o3), which work together to express natural red-yellow. the lepidocrocite is the dominated mineral with relatively complete crystals, while hematite is of secondary importance. the sample color indicated by the peak intensity shows a positive correlation with the hematite content. ultraviolet-visible absorption spectra (figure 4) of three red jadeites with large color hue differences (hue angle: 35.11°, 43.52° and 50.46°, respectively) shows that the absorption spectra basically follows the same trend with slight shift of certain absorption peaks and minor difference in the intensity. it can be inferred that the lower the hue angle is, the lower the absorption of ultraviolet and visible light spectrum is, the darker the color is. generally, the weak absorption at 430-470 nm and absorption at 730-800 nm are associated with occurrence of fe3+. figure 3. raman spectra of red jadeite figure 4. uv-vis absorption spectra of red jadeite 374 xin pan, ying guo*, ziyuan liu, zikai zhang, yuxiang shi consistent with other jadeite colors. however, the fe, ca and mg contents are higher than other jadeite colors, while the fe content, one of main elements affecting the jadeite color, varies greatly among them. table 1 shows the fe content in the tested samples increases successively. it can be inferred that the fe content in red jadeite will increase with color concentration in the same color system under the same conditions. color variation of red jadeite under three light sources a portable x-rite sp62 spectrophotometer was used to measure the color parameters of red jadeite samples under standard light source d65, cwf light source and a light source, respectively. the color parameters are l* (lightness), a* (red-green coordinates), b* (yellow-blue coordinates), cab* (chroma) and h0 (hue angle). two sets of 2880 data from 120 jadeite samples were obtained, including tristimulus values (x, y and z) and 24 variables. each sample was measured 3 times to ensure the accuracy (guo, zong & qi, 2018b). range of color parameters firstly, the color data of jadeite samples measured under three different light sources were sorted out (table 1). the results show that, under the standard d65 light source, l* ranges from 37.06 to 52.82, a* ranges from 13.05 to 28.18), b* ranges from 8.19 to 30.02, ho ranges from 22.64° to 57.66°, c* ranges from 15.95 to 38.33. under the cwf light source, l* ranges from 37.94 to 54.89, a* ranges from 9.62 to 21.33, b* ranges from 9.56 to 34.04), ho ranges from 27.59° to 68.19°, c* ranges from 13.97 to 47.25. under the a light source, l* ranges from 39.27 to 58.68, a* ranges from 15.56 to 30.23, b* ranges from 10.75 to 34.90), ho ranges from 36.09° to 57.21°, c * ranges from 19.83 to 44.67. obviously, the color is concentrated in the reddish area with a yellowish hue with different brilliance and medium and low brightness. the maximum and minimum values of measured data under each light source were selected, and the average and variance were calculated. the comparison shows that, c* value under a light source is obviously higher than that under other two light sources, which means, chroma of jadeite samples under a light source is higher. the b* value under cwf light source is higher than a*, and their difference is the largest, indicating more yellow tone. however, the result under d65 light source is different from that, showing more red tone. the measurement matches well with the observation. to visualize the color distribution under three light sources, the measured color data is projected onto l*a*b* uniform color space system and a*b* chromaticity diagram, as shown in figure 6-2. the horizontal axis and the vertical axis represent a*axis (red-green direction) and b*axis (yellow-blue direction), respectively (shams-nateri, 2009). compared with d65 and cwf light sources, the chroma of sample is higher under a light source (figure 6). meanwhile, sample under d65 light source is generally red, and sample under cwf light source is commonly orange, which agrees with the color perception under naked eyes. the influence of spectrum on the jadeite color comparison of the theoretical values of spectrum of d65, cwf and a light sources (figure 7) suggests that the curve under a light source is a generally inclined straight line. the spectral radiation of red zone is the strongest in the visible light region with color-rendering index (ra) up to 92.39. importantly, the input of orange-red color makes it suitable for illuminating red jadeite (guo, zong, qi, zhang & wang, 2018c). the spectrum curve under d65 light source is uniform in green, yellow and red zone with the highest peak in the blue-green zone at 480 nm. the ra (80.02) is lower than that of a light source, and the spectrum curve is uneven, showing that its illuminance on red jadeite is worse than a light source. generally, both standard a light source and standard d65 light source are high quality color-rendering light source. the spectrum curve table 1. color parameters of jadeite samples under different light sources light source l* a* b* c* h0 d65 range average value 37.06~52.82 44.68 13.05~28.18 20.63 8.79~30.02 19.83 15.95~38.33 28.69 22.64~57.66 43.13 cwf range 37.94~54.89 9.62~21.33 9.56~34.04 13.97~47.25 27.59~68.19 average value 46.07 15.40 22.36 27.43 54.56 a range 39.27~58.68 15.56~30.23 10.75~34.90 19.83~44.67 36.09~57.21 average value 47.84 23.40 24.77 34.24 46.41 figure 6. l*a*b* uniform color space system and a*b* chromaticity diagram showing red jadeite under d65, cwf and a light sources. 375impact of different standard lighting sources on red jadeite and color quality grading of cwf light source is generally sharp with peaks at 440 nm and 540-580 nm, which is unsuitable to illuminate jadeite red. figure 7. spectrum of d65 light source, cwf light source and a light source the color temperature of a light source (2856k) is lower than that of d65 light source (6504k), and its energy in the long-wave red zone is obviously higher than that in the short-wave blue-purple zone with overall light color of yellowish as a whole. the energy distribution of d65 light source in the range of 360 nm to 780 nm is more uniform than that of a light source, and the relative spectrum of d65 light source is close to that of sunlight (guo, 2016a, 2016b). therefore, the light color of d65 light source is closest to that of natural sunlight with no other obvious hues, which can better express the object color. ten red jadeite samples were randomly selected to be observed by naked eye and be simulated by computer, and the color difference was calculated using formula (3). the result confirms that l* and c* of the samples are the highest under a light source, a good indicator of brighter color. the samples under cwf light source are generally dark yellow-green tone, and color under d65 light source is closer to the red of the jadeite observed under sunlight (table 2). comprehensive analysis two-way analysis of variance since the light source and sample are two independent factors determining the jadeite appearance, the influence of light source and sample on their color index is determined through two-way analysis of variance via spss software (table 3). table 3. two-way analysis of variance of different light sources and samples on red jadeite difference source samples light source l* c* ho l* c* ho p 0.000 0.000 0.000 0.000 0.000 0.000 f 100.174 39.036 15.578 768.864 920.394 714.955 the results show that the l*, c* and h0 are significantly governed by both light source and sample itself, because p << 0.01 suggests a significant level. f value can be used to represent the significance of difference in the variance analysis, e.g., f≥p is a good indicator of significant difference. the f values in the table 3 is higher than p values, verifying the conclusion that color index varies greatly with sample itself and light source (wang, li, you & zhao, 2011). f-statistic from the variance analysis can also be used to determine the difference of the impact of the sample/ light source on l*, c* and ho values, specifically, the impact of sample can be summarized as: l*> c* >ho, while the impact of light source can be described as following: c* > l* >ho. one-way anova it is essential to conduct the one-way anova to specify the light source transformation that has a great influence on color index (china national bureau of technical supervision, 2001, 2004; cha & lee, 2009). the p > 0.05 was used to confirm homogeneous variance in the homogeneity test of variance, where the lsd post-comparison was employed table 2. red jadeite samples under three standard light sources and color block simulated by computer no. 1 6 12 18 24 30 36 42 48 54 color parameters l* 37.41 40.55 42.65 44.25 51.37 49.18 40.26 41.90 43.09 48.20 a* 13.05 17.89 19.68 20.69 22.21 16.65 18.39 20.52 22.18 24.45 b* 9.17 14.31 14.33 19.64 26.91 21.54 15.72 17.71 19.39 24.67 samples d65 cwf a color difference δe00 d65 4.89 6.65 9.20 16.58 13.27 5.35 7.27 8.73 13.69 cwf 3.41 5.91 7.07 11.17 19.49 16.37 6.68 8.62 10.06 15.91 a 2.88 829. 10.10 13.01 21.02 16.76 8.75 10.82 12.52 18.14 simulated color block d65 cwf a 376 xin pan, ying guo*, ziyuan liu, zikai zhang, yuxiang shi because of its highest sensitivity. the p < 0.05 is an indicator of inhomogeneous variance, where tamhane method was used to compare the data (nayatani & sakai, 2007; shams-nateri, 2009). table 4. test of variance homogeneity of red jadeite with different light sources l* c* h0 p 0.630 0.568 0.007 the pl*>0.05 and pc*>0.05 in the homogeneity test indicate that lsd post-comparison method should be used to analyze the influence of different light sources on brightness and chroma, while ph0<0.05 suggests that tamhane method is supposed to be used to analyze the influence of different light sources on the red tone. table 5. one-way anova of light source on red jadeite calculation method (i) light source (j) light source averaged difference (i-j) p value l lsd d652 f22 -1.36100 0.005 f22 a2 -1.84292 0.000 a2 d652 3.20392 0.000 c lsd d652 f22 1.26300 0.044 f22 a2 -6.81683 0.000 a2 d652 5.55383 0.000 h0 tamhane d652 f22 -11.43317 0.000 f22 a2 8.15092 0.000 a2 d652 3.28225 0.000 the variance analysis (table 5) shows that the brightness, color hue and chroma of red jadeite are significantly affected by changing light source. the p values of these parameters after changing light sources are <0.05, which illustrates that the light source can exert significant impact on brightness, color hue and chroma. impact of standard light source on brightness, color hue and chroma of red jadeite figure 8. impact of three light sources on the l* of red jadeite three light sources have weak influence on the brightness (figure 8), while the brightness value under a light source is higher than that under d65 light source and cwf light source. also, a light source can increase the brightness value, making samples brighter, since its light color tends to be orange-red because of higher energy at long wave (figure 3). three light sources have great influence on the color hue that increases in an order: cwf light source, d65 light source and a light source. this can be explained by that, cwf light source, a cold and white light source in the united states, has a higher energy at wavelength of 550-620 with yellow and green tone. the red color is disturbed and the chroma value is reduced when the yellow-green light is superimposed on the red jadeite. d65 light source represents the average daylight with light distribution more uniform in all color light, so the red color of jadeite under d65 light source is more similar to that under sunlight, and the color displayed is more objective. as a color light, a light source has higher energy in a wide wavelength range with orange-red tone. the orange-red light can increase color value of the jadeite, making the red color more saturated. jadeite red has the largest tone angle under cwf light source, while d65 and a light sources can bring larger biases to samples with lower tone angle and smaller biases to samples with lower tone angle. thus, color hue of red jadeite varies greatly with light sources. this is because the cwf light source has higher energy in the purple band, enabling the purple light to neutralize the yellow tone of the sample and make the tone angle larger. orange-red light derived from a light source can make the color of the red jadeite more orange-red. summarily, d65 light source is suitable for the jadeite evaluation, a light source is suitable for exhibiting the red color of jadeite, which can be used to sale and display jadeite. the cwf light source can add more red tone to the orange-yellow jadeite, which can be used in special cases. color grading of red jadeite the light source has an important influence on the color grading of gemstones. as discussed above, the energy of d65 light source in the range of 360-780 nm is much more uniform than that of a light source and cwf light source, which is close to the relative light spectrum of the sunlight. thus, light color of d65 light source is the closest to the natural sunlight with no obvious other hues, which can better reflect the object color. therefore, we recommend d65 light source to be used for the color grading of red jadeite. three independent color parameters, e.g., l*, a*, b*, are used for cluster analysis based on the international grading scheme of gem color (guo, 2017). acceptably, the clustering performs best for the grading number 5 with uniform distribution of each category. also, the sig value of the clustering result is < 0.001 (table 1), suggesting that a valid grading. therefore, 120 red jadeite samples were classified into 5 categories. fisher discriminant function was used to exam the clustering, corresponding discriminant functions: f 1 = 22.417 l* + 10.690 a* 9.536 b* (12) f 2 = 23.384 l* + 12.586 a* -7.723 b* (13) f 3 = 23.033 l* + 11.648 a* -8.605 b* (14) f 4 = 24.546 l* + 11.623 a* -6.833 b* (15) f 5 = 23.864 l* +10.854 a* -7.966 b* (16) the significance (sig) of the five grading is < 0.001 with the accuracy of color data backdating of 100% (table 6), confirming that the scheme is valid and feasible (tang, guo & xu, 2018). table 6. fisher’s discriminant analysis of jadeite-jade red color. (%) 1 2 3 4 5 total 1 100.0 0.0 0.0 0.0 0.0 100.0 2 0.0 100.0 0.0 0.0 0.0 100.0 3 0.0 0.0 100.0 0.0 0.0 100.0 4 0.0 0.0 0.0 100.0 0.0 100.0 5 0.0 0.0 0.0 0.0 100.0 100.0 377impact of different standard lighting sources on red jadeite and color quality grading currently, high-quality red jadeite is characterized by pure color, bright and uniform distribution, thus, c* and l* were used to evaluate color. as we know, high lightness makes the jadeite bright. thus, with the same chroma, jadeite with middle lightness is regarded as high quality. consequently, jadeite with hue range of 22.64-57.66°is divided into five levels, including fancy vivid, fancy deep, fancy intense, fancy dark and fancy, respectively. a 3d model of red jadeite is established accordingly (figure 4), specifically, the l* of fancy vivid ranges from 35 to 47, c* ranges from 30 to 40; the l* of fancy deep ranges from 47 to 53, the c* ranges from 30 to 40; the l* of fancy intense ranges from 35 to 44, the c* ranges from 24 to 30; the l* of fancy intense ranges from 44 to 53, the c* ranges from 24 to 30; and the l* of fancy intense ranges from 35 to 53, the c*chroma ranges from 15 to 24 (pan et al., 2019). figure 9. color solid of red jadeite conclusion conclusions were given based on comprehensive analysis: (1) fe 3+ is the main chromogenic mineral of red jadeite. the fe content has negative correlation with the tonal angle, and the color of red jadeite positively correlates with the hematite content. (2) d65 light source, a light source and cwf light source can exert great influence on the brightness, color hue and chroma of red jadeite. the light color of d65 light source distributes more uniformly, allowing red color of jadeite more objective and more similar to that under sunlight, which is suitable for the jadeite evaluation. a light source is suitable for exhibiting the red color of jadeite, which can be used to sale and display jadeite. the cwf light source can add more red tone to the orange-yellow jadeite samples, which can be used in special cases. (3) k-means clustering analysis and fisher discriminant analysis were used to determine the color from best to worst under the d65 standard light source (the most suitable light source for jadeite color grading): fancy vivid, fancy deep, fancy intense, fancy dark, fancy. the l* of fancy vivid ranges from 35 to 47, c* ranges from 30 to 40; l* of fancy deep ranges from 47 to 53, the c* ranges from 30 to 40; l* of fancy intense ranges from 35 to 44, c* ranges from 24 to 30; l* of fancy intense ranges from 44 to 53, c* ranges from 24 to 30; and l* of fancy intense ranges from 35 to 53, c*chroma ranges from 15 to 24. acknowledgments the authors gratefully acknowledge the financial supports by the innovation and entrepreneurship training program for undergraduates of china university of geosciences (beijing) under project numbers x201911415404. it also thanks dr.guo and xue-ding wang for their suggestions and help in colorimetric research. references cha, h. s. & lee, y. k. (2009). difference in illuminant-dependent color changes of shade guide tabs by the shade designation relative to three illuminants. american journal of dentistry, 22(6), 350-356. china national bureau of technical supervision. (2010). diamond grading. chinese national standards. standards press of china, beijing. gb/t 16554-2010. china national bureau of technical supervision. (2009). jadeite grading. chinese national standards. standards press of china, beijing. gb/t 23885-2009. china national bureau of technical supervision. (2016a). sapphire grading. chinese national standards. standards press of china, beijing. gb/t 32862-2016. china national bureau of technical supervision. (2016b). ruby grading. chinese national standards. standards press of china, beijing. gb t 32863-2016. guo, y., wang, h., li, x., & dong, s. r. (2016b). metamerism appreciation of jadeite-jade green under the standard light sources d65 a and cwf. acta geologica sinica, 90(6), 2097-2103. doi: 10.1111/1755 6724.13024. guo, y., wang, h., & du, h. m. (2016a). the foundation of a color-chip evaluation system of jadeite-jade green with color difference control of medical device. multimedia tools and application 75(22), 1449114502. doi: 10.1007/s11042-016-3291-8. guo, y., zhang, x. y., li, x. & zhang, y. (2018a). quantitative characterization appreciation of golden citrine golden by the irradiation of [feo4]4. arabian journal of chemistry, 11(6), 918-923. doi: 10.1016/j. arabjc.2018.02.003. guo, y., zong, x., qi, m., zhang, y., & wang, h. (2018c). feasibility study on color evaluation of jadeite based on gemdialogue color chip images. eurasip journal on image and video processing, 95, 1-9. doi: 10.1186/s13640-018-0342-2. guo, y., zong, x., & qi, m. (2018b). feasibility study on quality evaluation of jadeite-jade color green based on gemdialogue color chip. multimedia tools and applications, vol. 67, no. 3, pp. 1-16, doi: 10.1007/s11042018-5753-7. guo, y. (2016a). metamerism appreciation of jadeite-jade green under the standard light sources d65, a and cwf. acta geologica sinica, 90(6), 2097-2103. guo, y. (2016b). the foundation of a color-chip evaluation system of jadeitejade green with color difference control of medical device. kluwer academic publishers, 75(22), 1-12. guo, y. (2017a). quality evaluation of tourmaline red based on uniform color space. cluster computing, 20(4), 3393-3408. doi: 10.1007/s10586017-1091-1. guo, y. (2017b). quality grading system of jadeite-jade green based on three colorimetric parameters under cie standard light sources d65 cwf and a. bulgarian chemical communications, 49(4), 961-968. harlow, g. e. & olds, e. p. (1987). observations on terrestrial ureyite and ureyitic pyroxene. american mineralogist, 72, 126–136. htein, w. & naing, a. m. (1994). mineral and chemical compositions of jadeite jade of myanmar. journal of gemmology, 24 (4), 269–276. htein, w. & naing, a. m. (1995). studies on kosmochlor, jadeite and associated minerals in jade of myanmar. journal of gemmology, 24(5), 315–320. 378 xin pan, ying guo*, ziyuan liu, zikai zhang, yuxiang shi hughes, r., galibert, o., bosshart, g., ward, f., oo, t., smith, m., … harlow, g. (2000). burmese jade: the inscrutable gem. gems & gemology, 36, 2-25. doi: 10.5741/gems.36.1.2. international commission on illumination. (2001). technical report: improvement to industrial colour-difference evaluation. cie publication, austria. international commission on illumination. (2004). technical report: colorimetry. cie publication, austria. mével, c. & kiénast, j.r. (1986). jadeite-kosmochlor solid solution and chromian sodic amphiboles in jadeitites and associated rocks from tawmaw (burma). bulletin de minéralogie, 109, 617–633. nayatani, y. & sakai, h. (2007). complexity of the color appearance of objects with nonuniform illumination. color research & application, 32(5), 372-377. perez, m., ghinea, r., herrera, l. j., ionescu, a. m., pomares, h., pulgar, r., & paravina, r. d. (2011). dental ceramics: a ciede2000 acceptability thresholds for lightness, chroma and hue differences. journal of dentistry, 39 (supp-s3). ponahlo, j. (1999). cathodoluminescence du jade. revue de gemmologie a.f.g., 137, 10–16. rossman, g. r. (1974). lavender jade: the optical spectrum of fe3+ and fe2+–fe3+ inter-valence charge transfer in jadeite from burma. american mineralogist, 59, 868–87. sanchez-marañon, m. & garcía p. a., huertas, r., hernandez-andres, j., & melgosa, m. (2011). influence of natural daylight on soil color description: assessment using a color-appearance model. soil science society of america journal, 75(3), 984. shams-nateri, a. m. (2009). estimation of cie tristimulus values under various illuminants color research & application, 34(2), 100-107. shen, s. & berns, r. s. (2009). evaluating color difference equation performance incorporating visual uncertainty. color research & application, 34(5), 375-390. tang, j., guo, y., & xu, c. (2018). light pollution effects of illuminance on yellowish green forsterite color under cie standard light source d65. ekoloji, 27(106), 1181-1190. tsujimori, t. & harlow, g. (2012). petrogenetic relationships between jadeitite and associated high-pressure and low-temperature metamorphic rocks in worldwide jadeitite localities: a review. european journal of mineralogy, 24, 371-390. doi: 10.1127/0935-1221/2012/0024-2193. wang, j., li, q., you, j., & zhao, q. (2011). fast kernel fisher discriminant analysis via approximating the kernel principal component analysis. neurocomputing, 74(17), 3313-3322. yang, j., yang, l., long, j. (2015). theoretical investigation of the electronic structure, optical, elastic, hardness and thermodynamics properties of jadeite. materials science in semiconductor processing, 31, 509-516. issn 1369-8001. ying, g.. zong, x., qi, m., zhang, y., & wang, h. (2018). feasibility study on color evaluation of jadeite based on gemdialogue color chip images. eurasip journal on image and video processing. doi: 10.1186/ s13640-018-0342-2. yui, t. f. & fukuyama, m. (2015). a revisit to the yorii jadeite–quartz rock, the kanto mountains, central japan: implications for petrogenesis. journal of asian earth sciences, 108, 58-67. doi: 10.1016/j. jseaes.2015.04.022. keywords: mine transient electromagnetic method; finite difference time domain; metal support; numerical analysis palabras clave: método de electromagnetismo transitorio en minas; método electromagnético en el dominio del tiempo; soporte de metal support; análisis numérico how to cite item yaoning l., shucai l., maofei l., xinming l., weihong g. (2021). influence of metal roadway supports on transient electromagnetic detection in mines. earth sciences research journal, 25(1), 109-114. doi: https://doi.org/10.15446/ esrj.v25n1.80156 to study the influence of metal supports in roadways on the detection of mines using the transient electromagnetic method, authors treated metal supports including anchor nets as a thin metal layer. according to the finite differences principle, the characteristics of the full-space transient electromagnetic response under the thin metal layer’s influence were calculated using a non-uniform grid. the thin metal layer's presence slowed the electromagnetic field’s diffusion rate and hindered the overall diffusion. the transient electromagnetic response curve observed under the thin metal layer’s influence was higher than that without the supports. thicker metal layers resulted in higher early response values and slower decay rates. the decay rate increased as a function of time, gradually approaching that of the curve without metal supports. the simulation of the transient electromagnetic response to the model of water-containing lowresistance structures showed that the metal roadway support reduced the sensitivity of the transient electromagnetic method and weakened its response to low-resistance anomalies. abstract influence of metal roadway supports on transient electromagnetic detection in mines influencia de los soportes metálicos en la calzada para la detección de electromagnetismo transitorio en minas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.80156 con el fin de estudiar la influencia de los soportes metálicos en las calzadas para detectar minas a través del método de electromagnetismo transitorio, los autores trataron estos soportes como capas finas de metal, incluidas las mallas de anclaje. de acuerdo con los principios de las diferencias finitas, las características de reacción del electromagnetismo transitorio bajo la incidencia de una capa fina de metal se calcularon con una cuadrícula no uniforme. la presencia de esta capa ralentizó el índice de difusión del campo electromagnético y dificultó la difusión en general. el resultado de la curva de reacción del electromagnetismo transitorio bajo la incidencia de la capa fina de metal fue mayor que en calzadas sin estos soportes. las capas más finas tienen valores de respuestas mayores e índices de deterioro menores. el índice de deterioro se incrementa como una función del tiempo, y se aproxima gradualmente a la curva de reacción de una calzada sin soportes. la simulación de la reacción del electromagnetismo transitorio en el modelo de estructuras de baja resistencia a contenidos de agua muestra que los soportes metálicos reducen la sensibilidad del método electromagnético transitorio y debilita su reacción a anomalías de baja resistencia. resumen record manuscript received: 06/06/2020 accepted for publication: 02/10/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 109-114 liu yaoning1, liu shucai2*, li maofei2, liu xinming3, guo weihong2 1institute of building intelligence, jiangsu vocational institute of architectural technology, xuzhou, jiangsu province 221116 china 2the school of resource and geoscience, china university of mine and technology, xuzhou, jiangsu province 221116 china. 3college of mining engineering, north china university of science and technology, tangshan, hebei province 063000 china. m in in g e n g in e e r in g https://doi.org/10.15446/esrj.v25n1.80156 https://doi.org/10.15446/esrj.v25n1.80156 https://doi.org/10.15446/esrj.v25n1.80156 110 yaoning liu, shucai liu, maofei li, xinming liu, weihong guo introduction with increasing coal mining depth, the hydrogeological conditions of mines are becoming increasingly complex, and coalmine water inrush disasters are a serious threat to the safe operation of coal mines (wu, 2014). the mine transient electromagnetic method considers proximity to the target for exhibiting a strong signal response and high resolution, this is widely used in the prediction of goaf water, column collapse, and mine disasters (cheng et al, 2016; danielsen et al, 2003; huang et al, 2017; wang et al, 2017a; wang et al, 2017b). however, transmission and receiving devices used by the mine transient electromagnetic method are located in underground roadways and hence, the received signal is a superposition of various low-resistance electromagnetic induction signals within the detection range. thus, distinguishing water body information from these signals for practical applications is difficult (chang et al, 2016; su et al, 2017; xue & yue, 2017; zhou et al. 2014). therefore, studying the influence of metal supports in the roadway on mine transient electromagnetic detection is important for eliminating metal interference and determining the distribution range of water-containing anomalies. many researchers have conducted experimental studies on the mine transient electromagnetic method considering metal interferences (liu et al, 2016; jin & wang. 2018, qin & zhao.2019, zhou et al. 2014); however numerical calculations of these phenomena have rarely been conducted. the direct time-domain finite difference method can directly simulate the propagation and diffusion of electromagnetic waves and can be used to analyze transient electromagnetic field signals (yan et al, 2002; yue et al, 2008). according to the maxwell equations and finite-difference principle, the metal support in the roadway can be considered as a thin layer of metal, and the finite difference iterative equation of this thin metal layer can be derived. the transient electromagnetic field under the influence of the metal roadway support can also be derived based on the finite-difference equation. this simulation was performed to study the electromagnetic field diffusion characteristics under the influence of a thin metal layer and generate a full-space transient electromagnetic response curve under the influence of these thin metal layers with different thicknesses. in addition, the transient electromagnetic response of the water-containing low-resistance structural model under metal the influence of the metal layer was analyzed. basic principles principles of the finite-difference numerical simulation quasi-static maxwell’s equations ∇× ( )= ( ) ( ) ∇× ( )=− ( ) ∇⋅ ( )= ∇⋅ ( )= h r r e r e r h r h r e r , , , , , , t t t t t t t σ µ ∂ ∂ 0 00 (1) where, h r, t( ) is the magnetic field strength, e r, t( ) is the electric field strength, σ r( ) is the dielectric conductivity, and  is the magnetic permeability. a passive diffusion equation for the electric field can be obtained. ∇ ( )= ( ) ∂ ( ) ∂ 2e r r e r , , t t t µσ (2) under 2d conditions, two infinitely long wires with positive and negative currents were used as the emission source and placed along the y-axis. thus, the induced electric field has only a y component and the magnetic field strength has only x and z components. figure 1 shows the two-dimensional finite-difference meshing mode where point ei j, and its surrounding ei j1, , ei j1, , ei j, 1 , ei j, 1 constitute a five-point difference format (liu. 2014). figure 1. two-dimensional finite difference model. connecting the center points of the four small rectangles around ei j, forms a rectangle, abcd. at any time, equation 2 can be integrated into the rectangle abcd to obtain equation 3 as follows: [ ] ∂ ∂ + ∂ ∂ = ∂ ∂∫∫ ∫∫ 2 2 2 2 e x e z dxdz e t dxdz abcd abcd µσ (3) the difference equation of the diffusion equation can be obtained according to green’s formula and the central difference method. e r r e r r z z ei j n i j i j i j n i j z i j j j i j n , , , , , , ,[ + − += − + + + ∆ ∆ + 1 1 1 1 4 1 4 2 1 4 ∆∆ ∆ + + ∆ ∆ + ∆ ∆ + − + + − z z e r r x x e x x e j j i j n i j x i j i i i j n i i i 1 1 1 1 1 2 1 4 , , , ,] [ ,, ]j n (4) where, ∆ = ∆ +∆+ z z z j j j1 2 , ∆ = ∆ +∆+x x x i i i1 2 , ri j z i j j j t z z, , = ∆ ∆ ∆ +µσ 1 , ri j x i j i i t x x, , = ∆ ∆ ∆ +µσ 1 , r r r i j i j z i j x , , , = + 2 . finite-difference simulation of the metal sheet metal supports can affect electromagnetic wave propagation, in turn affecting the detection results (yu et al, 2008). for numerical simulations, local fine mesh or subdomain algorithms are generally used for thin lamellar bodies (elsherbeni & demir, 2006; sun et al. 2013). however, the net thickness of the metal anchor is much smaller than the grid size and its resistivity is much smaller than the surrounding resistivity of the earth. thus, it is difficult to perform numerical simulations using conventional methods. therefore, a finitedifference iterative formula applicable to the metal structure must be derived. yan and shi (2004) derived the finite-difference iterative formula for thin wires by considering railroad tracks as thin wires. in this study, we consider the metal support to be a thin layer of metal with a finite thickness. the relationship between the electromagnetic field in the thin metal layer and the surrounding electromagnetic field can be derived using the maxwell equation boundary conditions. figure 2 shows the finite difference model of a thin layer of metal along the x-z plane (liu. 2014). 111influence of metal roadway supports on transient electromagnetic detection in mines figure 2. finite difference model of a thin layer of metal along the x-z plane the integral form of the two rotation equations in equation 1 can be written in the form of a difference equation. − −( ) = − ( )+ −( ) ∆ µ dh i j dt e i j e i j z x j , , ,1 1 (5) − ( ) = − +( )+ ( ) ∆ + µ dh i j dt e i j e i j z x j , , ,1 1 (6) ∆ +∆ ∆ +∆ ( )= ( )− −( )( ) × ∆ +∆ + + + x x z z e i j h i j h i j x x i i j j x x i i 1 1 1 2 2 1 2 σ , , , ++ ( )− −( )( ) ∆ +∆ + h i j h i j z z z z j j , ,1 2 1 (7) where  is the average conductivity within the rectangle, abcd. because the conductivity of the thin metal layer is much higher than that of the surrounding medium, the conductivity of the earth medium in abcd can be neglected as follows: ∆ +∆ ∆ +∆ = ∆ +∆+ + +x x z z x x di i j j i i1 1 1 2 2 2 σ σ (8) where  is the conductivity of the metal and d is the thickness of the metal layer. for the metal support, d should have an equivalent thickness. because of the influence of the metal layer, the magnetic field strength-x component hx -increases on both sides of the metal layer, and the magnetic field strength-z component hz-is continuous. thus, the second term on the right side of equation 7 can be neglected and the equation can be written as follows: σ d e i j h i j h i j x x x x i i⋅ ( )= ( )− −( )( ) ∆ +∆ +, , , 1 2 1 (9) substituting equations 5 and 6 into equation 9 yields: ∆ ∆ ⋅ ( ) =∆ −( ) +∆ +( )− ∆ +∆( + + + z z d de i j dt z e i j z e i j z z j j j j j j 1 1 1 1 1 µσ , , , )) ( )e i j, (10) the time derivative can be written in the form of a central difference: ∂ ∂ = − ∆ + −e t e e t i j n i j n i j n , , , 1 1 2 (11) linear interpolation of ei j n , e e e i j n i j n i j n , , , = ++ −1 1 2 (12) by substituting equations 11 and 12 into equation 10, the difference equation of the thin metal layer in figure 2 can be obtained. e ae be cei j n i j n i j n i j n , , , , + − + −= + +1 1 1 1 (13) where ∆ = ∆ +∆ + z z zj j 1 2 , a t z z t z z d j j j = ∆ ∆ ∆ ∆ +∆ ∆ + + 2 2 1 1µσ , b t z z t z z d j j j = ∆ ∆ ∆ ∆ +∆ ∆ + 2 2 1µσ , c z z d z t z z d z t j j j j = ∆ ∆ − ∆ ∆ ∆ ∆ + ∆ ∆ + + 1 1 2 2 µσ µσ . for the metal layer along the x-y plane, the difference equation can be obtained as follows: e ae be cei j n i j n i j n i j n , , , , + − + −= + +1 1 1 1 (14) where ∆ = ∆ +∆ +x x xi i 1 2 , a t x x t x x d i i i = ∆ ∆ ∆ ∆ +∆ ∆ + + 2 2 1 1µσ , b t x x t x x d i i i = ∆ ∆ ∆ ∆ +∆ ∆ + 2 2 1µσ , c x x d x t x x d x t i i i i = ∆ ∆ − ∆ ∆ ∆ ∆ + ∆ ∆ + + 1 1 2 2 µσ µσ equations 13 and 14 are the finite difference iterative equations for the thin metal layers. numerical calculations transient electromagnetic field distribution under the influence of the metal layer according to the principle of finite difference, using equations 4, 13, and 14, the full-space transient electromagnetic field under the influence of the metal support was simulated using the absorbing boundary condition (yang & yue, 2009). figure 3 shows the 2d space model used herein. the resistivity of the model was 50ω⋅m with a roadway width of 20 m, the launching point was at the center of the roadway. the two long wire sources were placed along the y-axis, while the normal direction was specified along the x-axis, space was 2 m, the emission current was 10a, and the metal mesh was set to a diameter of 5 mm— commonly used for the equivalent metal layer—which has an equivalent thickness of 0.2 mm. figure 3. two-dimensional space model. 112 yaoning liu, shucai liu, maofei li, xinming liu, weihong guo figure 4 shows the transient electric field contour map as a function of time with and without the metal support, where the electric unit is v m , and the horizontal line in figure 4 represents the roadway boundary. figures 4a and 4b show the electric field contour maps with and without the metal support, respectively, at t = 0.64 µs. here, the transient field which appears as a ‘smoke ring’ did not spread to the roadway boundary, the metal support at the boundary of the roadway compressed the electric field periphery. thus, the metal support of the roadway affected the electric field distribution. figure 4c and d show the electric field contour maps with and without metal support, respectively, at t = 2.21 µs. here, the ‘smoke ring’ without support reached the boundary of the roadway, whereas when support was present, the electric field became distorted at the boundary of the roadway. the presence of a thin layer of metal reduced the rate of electric field diffusion, squeezing the “smoke ring” at the metal boundary of the roadway. figure 4e and f show the electric field contour maps with and without the metal support, respectively, at t = 7.87 µs. here, the “smoke ring” without the support crossed the boundary of the roadway, whereas, with the support present, the ‘smoke ring’ remained at the boundary of the roadway and did not shift forward to a significant degree. this is due to the low resistivity of the thin metal layer (˂10-6). thus, the transient field diffused very slowly (yu et al, 2008). therefore, it can be concluded that the roadway metal support will greatly reduce the diffusion speed of the transient field, inevitably affecting the actual detection results. figure 4. transient electric field contour map at different times with and without the metal support. according to the relationship between the vertical induced electromotive force and electric field in the receiving coil per unit area, the electric field can be converted into a magnetic field as a function of time as follows: ∂ ∂ =− ∂ ∂ b t e x z (15) figure 5 shows the transient electromagnetic response curve influenced by the different thicknesses of the metal support. the observation point was located at the launch point and the wire diameter of the metal support net was varied at 2, 5, and 7 mm with equivalent thicknesses of 0.03, 0.2, and 0.4 mm, respectively. from fig. 5 it is clear that the response value with the metal support is significantly higher than that without the support, with a maximum value that differed by three orders of magnitude. moreover, after 2 ms, increasing the equivalent thickness of the metal support layer resulted in higher response values. initially, the response in the presence of the metal support is slowed. however, over time the decay rate gradually increases and the curve moves gradually closer to that of the unsupported curve, showing consistency with the physical simulation. figure 5. transient electromagnetic response curve under different metal shed supports. numerical simulation of a model with low resistance anomalies the mine transient electromagnetic method is mainly used to detect water-containing, low-resistance anomalous structures. to further study the effect of metal supports on the transient electromagnetic response, the response of a water-containing low-resistance structural model was simulated. here, the abnormal body resistivity was set to 1ω⋅m for a size of 10 m × 10 m. the model was located below the field source, 50 m away from the floor of the roadway, and the resistivity was set to 50ω⋅m . a metal mesh with a diameter of 5 mm was selected with an equivalent thickness of 0.2 mm. the simulation results are shown in figure 6 where the ordinate indicates the relative apparent resistivity r (chang et al, 2020). the relative apparent resistivity is defined as the ratio of the apparent resistivity in the presence of an abnormal body to that without an abnormal body, ρ ρ ρ τ τ r = ′ (16) where ρτ is the apparent resistivity without any low resistance anomalies and ρτ is the apparent resistivity with a low resistance abnormality. figure 6 shows that without the metal support, the relative apparent resistivity curve begins to decrease at t= 0.08 ms, and a low-resistance anomaly occurs. however, in the presence of metal support, an abnormality occurs at t= 3 ms, indicating that the metal support delayed the transient electromagnetic response. thus, when the abnormal body is at the same position, observed 113influence of metal roadway supports on transient electromagnetic detection in mines longer observation time is required in the presence of metal support. in figure 6, without the metal support, the low resistance abnormality caused the curve to drop to 0.32, whereas in the presence of the metal support, the lowest point of the curve was 0.90. this indicated that the metal layer reduced the sensitivity of the transient electromagnetic method and weakened its reaction to the low resistance abnormal body. figure 6. relative apparent resistivity contrast diagram with and without the metal support. therefore, in the actual exploration process, there should be a special focus on the interference of the surrounding metal environment and this work should be fully recorded. in the data interpretation process, it is necessary to perform time-sharing corrections on the exploration data according to the collected time series. conclusions by using a numerical simulation method to simulate the effect of metal support on tem detection, the following conclusions are drawn. 1) the metal anchor net support was treated as a thin metal layer and the finite difference equation of the metal layer was derived. a non-uniform grid was used to simulate the transient electromagnetic response characteristics of the entire space under the influence of the metal layers. the presence of a thin metal layer slowed the transient field diffusion rate and hindered the overall diffusion of transient electromagnetic fields. 2) the magnetic field as a function of the time derivative parameter curve observed under the influence of the thin metal layer was higher than that of the response curve without the metal support. increased thickness of the thin metal layer resulted in higher curves. 3) simulation of the transient electromagnetic response of the watercontaining, low-resistance anomaly model showed that the metal support layer delayed the transient electromagnetic response, requiring longer times for observation. moreover, the metal support reduced the sensitivity of the transient electromagnetic method and weakened the reaction to the lowresistance anomaly. therefore, performing relevant corrections during the actual exploration process is necessary. acknowledgment this research was supported by the national key r&d program of china under grant(2018yfc0807802-3) and doctoral specialty foundation (jyjbzx20-06). we are particularly grateful for the advice and help of editors and reviewers. in addition, we would like to thank editage (www.editage.cn) for english language editing. references chang, j., yu, j. & liu, z. (2016). three-dimensional numerical modelling of full-space transient electromagnetic responses of water in goaf. applied geophysics, 3, 539-552. chang, j., su, b., & malekian, r. (2020). detection of water-filled mining goaf using mining transient electromagnetic method. iee transactions on industrial informatics, 16(5), 2977-2984. cheng, j., chen, d. & xue, g. (2016). synthetic aperture imaging in advanced detection of roadway using the mine transient electromagnetic method. chinese journal of geophysics, 59(2), 731-738. danielsen, j. e., auken, j. e., sondergaard, v. & sorensen, k. i. (2003). the application of the transient, electromagnetic method in hydrogeophysical surveys. journal of applied geophysics, 53(4), 181198. elsherbeni, a. & demir v. (2006). the finite-difference time-domain method in electromagnetics: with matlab. raleigh: scitech publishing. huang, l., liu, s., wang, b. & zhou, f. (2017). quantitative calculation of aquifer water quantity using tem data. earth sciences research journal, 21(1), 51-56. jin, j., & wang, s. (2018). study on the low resistance interference test of coal mine transient electromagnetic. mining safety and environmental protection, 45(1), 93-97. liu, y. (2014). study on the application effect of mine transient electromagnetic technology under the metal interference. china university of mining and technology. liu, z., huang, w., & huang, j. (2016). interference of metal materials in tunnels on the detection results by transient electromagnetic method. modern tunnelling technology, 53(4), 116-122. su, b., yu, j., sheng, c., & zhang, y. (2017). maxwell-equations based on mining transient electromagnetic method for coal mine-disaster water. elektronika ir elektrotechnika, 23(3), 20-23. sun, h., li, x., & li, s. c. (2013). three-dimensional fdtd modeling of tem excited by a loop source considering ramp time. chinese journal of geophysics, 56(3), 1049-1064. wang, b., liu, s., li, s. & zhou, f. (2017a). double-transmitting and sextuplereceiving borehole transient electromagnetic method and experimental study. earth sciences research journal, 21(2), 77-83. wang, b., liu, s., zhou, f., zhang, j., & zheng, f. (2017b). diffraction characteristics of small fault ahead of tunnel face in coal roadway. earth sciences research journal, 21(2), 95-99. wu, q. (2014). progress, problems and prospects of prevention and control technology of mine water and reutilization in china. journal of china coal society, 39(5), 795-805. xue, g. & yu, j. (2017). new development of tem research and application in coal mine exploration (in chinese). progress in geophysics, 32(1), 0319-0326. yan, s., chen, m. & fu, j. m. (2002). direct time-domain numerical analysis of transient electromagnetic fields. chinese journal of geophysics, 45(2), 275-284. yan, s. & shi, x. (2004). thin bed and wires modeling in transient electromagnetic fields fdtd computation of tunnel whole-space. coal geology and exploration, 32(s1), 87-89. yang, h. & yue, j. h. (2009). application of absorbing boundary condition in whole-space computation of transient electromagnetic response. journal of china university of mining and technology, 38(2), 263-268. 114 yaoning liu, shucai liu, maofei li, xinming liu, weihong guo yu, j., liu, s. & wang, y. (2008). response characteristic of transient electromagnetic to metallic facilities in coal mines and the disposal technology. journal of china coal society, 33(12), 1403-1407. yue, j., yang, h. & hu b. (2008). 3d finite difference time domain numerical simulation for tem in-mine (in chinese). progress in geophysics 22(6), 1904-1909. zhou, s., yu, j. & jiang, z. (2014). study of mine transient electromagnetic response characteristics and correction method under support using bolt and wire mesh. china coal, 4, 45-48. zhou, j., cheng, j. & wen, l. (2017). response characteristics of metallic facilities and correction method on mine transient electromagnetic surveying. journal of the china coal society, 26(8), 146-150. keywords: polynomial regression; gps elevation fitting; quadric surface fitting; plane fitting; geodetic height. palabras clave: regresión polinómica; ajuste de elevación gps; ajuste de superficie cuadrática; montaje en plano; altura geodésica. how to cite item wang, j., & xie, d. (2020). gps elevation fitting study based on ternary polynomial regression. earth sciences research journal, 24(2), 201-205. doi: https://doi.org/10.15446/ esrj.v24n2.87228 for the traditional gps elevation fitting method, the accuracy has not been significantly improved in recent years, and the method has become increasingly complicated. this paper proposes to insert the geodetic height ‘h’ into the calculation system and use a ternary polynomial regression function to fit the gps elevation anomaly. the feasibility of the ternary polynomial regression method in gps elevation fitting was verified by an example, and compared with the results of the traditional plane fitting and quadric surface fitting method, it was proved that the proposed method is suitable for terrain with large terrain fluctuations. the fitting residual error and the standard deviation are smaller, and through example calculations, it is concluded that the ternary polynomial regression method under the seven parameters has the highest fitting accuracy. abstract gps elevation fitting study based on ternary polynomial regression estudio de ajuste de elevación gps basado en regresión polinómica ternaria issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87228 para el método tradicional de ajuste de elevación gps, la precisión no se ha mejorado significativamente en los últimos años, y el método se ha vuelto cada vez más complicado. este documento propone insertar la altura geodésica "h" en el sistema de cálculo y utilizar una función de regresión polinómica ternaria para ajustar la anomalía de elevación del gps. la viabilidad del método de regresión polinómica ternaria en el ajuste de elevación gps se verificó mediante un ejemplo y, en comparación con los resultados del método tradicional de ajuste de plano y ajuste de superficie cuadrática, se demostró que el método propuesto es adecuado para terrenos con grandes fluctuaciones. el error residual de ajuste y la desviación estándar son menores, y a través de cálculos de ejemplo, se concluye que el método de regresión polinómica ternaria bajo los siete parámetros tiene la mayor precisión de ajuste. resumen record manuscript received: 13/11/2019 accepted for publication: 07/04/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 201-205 jianmin wang*, dongping xie college of surveying and mapping and geographic sciences, liaoning technical university, fuxin, 123000, china * corresponding author: furuiteng123456@126.com gp s m ea su re m en ts a pp li ed t o ge op hy si ca l p ro bl em s https://doi.org/10.15446/esrj.v24n2.87228 https://doi.org/10.15446/esrj.v24n2.87228 https://doi.org/10.15446/esrj.v24n2.87228 mailto:furuiteng123456@126.com 202 jianmin wang, dongping xie introduction in recent years, china’s beidou navigation satellite system (bds) has made great progress and is expected to achieve the goal of global coverage by 2020. the coordinate system used is the beidou coordinate system (bdcs) is a coordinate system which is defined in the framework of the international terrestrial reference frame (itrs) as the current world geodetic coordinate system (wgs84) (li & huang, 2016). therefore, the research on elevation fitting of global positioning system (gps) is helpful for the effective conversion between beidou height and normal height in the future. at present, through the research of some scholars, the methods of fitting gps elevation data can be roughly divided into two categories. one is the traditional function fitting method, and the other is the new model fitting method. the traditional methods include weighted average method, plane fitting method, quadratic polynomial surface fitting method, moving quadratic polynomial surface fitting method and cubic polynomial surface fitting method, and the new models include neural network fitting method, support vector machine fitting method and curve cloud model fitting method (zhang et al., 2015; ren et al., 2015; guo, 2018; tang et al., 2016). in the actual operation, the new fitting method is not suitable for large-scale popularization because of its complex model, long data processing cycle and high requirements for hardware and software. however, the traditional fitting method only considers the relationship between the elevation anomaly and the plane coordinates x and y of the point, and the fitting accuracy is low, even in the practical engineering application, it can not achieve the experimental accuracy. in this paper, based on the traditional gps elevation fitting method, the geodetic height is included in the solution system, and a gps elevation fitting model based on ternary polynomial regression is proposed (zhou et al., 2016; ding & sun, 2016; liu et al., 2016; yang & zie, 2017; sun et al., 2014). ternary polynomial regression method the traditional elevation fitting method only considers that the elevation anomaly varies with the position of the plane. however, by understanding the elevation system (positive height, normal height, and geodetic height) and deducing the formula, it can be known that the elevation anomaly has a certain relationship with geodetic height. the ternary polynomial regression method can fully consider the relationship between elevation anomalies and x, y, h, so as to better fit the real elevation anomaly of ground points, and realize the conversion between the geodetic height and the normal height (zhao et al., 2011; wang et al., 2016; chen et al., 2014; song et al., 2014). interpretation of elevation anomaly and geodetic height in figure 1, the point p and the point ′p are on the same vertical line and have the same plane position. then there are h n hg + = (1) h n hg  + ′ = ′ (2) in the formulas, hg is the positive height lpqof point p, n is the geoid undulation lqv, h is the geodetic height lpu; hg is the positive height lp q′ of point ′p , ′n is the geoid undulation lqv, ′h is the geodetic height lp w′ . it can be seen from figure 1 that n n= ′, so there is ∇ − − ∇h h h h h hg g g= ′ = ′ = (3) that is, the positive elevation difference between the p point and the ′p point is equal to the geodetic height difference. in the formula: hg is the positive elevation difference between point p and point ′p , and h is the geodetic height difference between point p and point ′p . according to the literature (figure 1), the difference between the elevation anomaly  and the geoid n meets n g hm m m g= + ⋅η −γ γ (4) in the formula, gm is the real average gravity value on the vertical line between the geoid and the earth’s surface, and m is the average gravity value from the reference ellipsoid along the normal line to the approximate earth’s surface (li & huang, 2016). for point ′p , there is ′ = ′ + ′ ′ ′ n gm m m η −γ γ (5) for formula (4) and (5), there are h hg g≠ ′ ,g gm m≠ ′ and γ ≠ γm m′ . according to the complexity of the leveling surface and the earth’s gravity field, g h g hm m m g m m m g −γ γ −γ γ ⋅ ≠ ′ ′ ′ ⋅ ′ and combined with the formula n n= ′, figure 1. elevation system diagram 203gps elevation fitting study based on ternary polynomial regression parameters is carried out respectively. the fitting accuracy is shown in table 5, and the trend of fitting accuracy is shown in figure 2. from the results in table 3, it can be seen that the smallest value of the three-parameter ternary polynomial regression fitting residual is 0.17 cm, the average residual is 0.89 cm, and the smallest value of the plane fitting residual is 0.23 cm, the average residual is 2.61 cm. it is concluded that the ternary polynomial regression fitting method has strong feasibility in the actual gps elevation fitting.  ≠ ′ can be obtained (shijun & li, 2015; gong et al., 2014; škrekovski et al., 2019; zhao et al., 2017). looking back at the point p and point ′p , they have the same plane position, but the difference is their positive height. therefore, it can be seen that the elevation anomaly  is related to positive elevation hg. formula (3) shows that the change of positive elevation is equal to that of geodetic height. therefore, the elevation anomaly  is related to geodetic height h. ternary polynomial regression function if x, y and h are taken as three independent variables and elevation anomaly  is taken as dependent variable, then there are functions 1) three parameters f a x a y a h3 0 1 2= = + + (6) 2) four parameters f a a x a y a h4 0 1 2 3= = + + + (7) 3) seven parameters f a a x a y a h a xy a yh a xh7 0 1 2 3 4 5 6= = + + + + + + (8) 4) eight parameters f a a x a y a h a xy a yh a xh a xyh8 0 1 2 3 4 5 6 7= = + + + + + + + (9) 5) eleven parameters f a x a y a h a xy a yh a xh a x a y 11 0 1 2 3 4 5 6 7 2 8 = + + + + + + + + η= α 22 9 2 10+ +a h a (10) by calculating the known data set (x, y, h, ), the corresponding polynomial parameters can be obtained and the functional relationship can be determined. among them, three-parameter function requires at least three known point data, four-parameter function requires at least four known point data, seven-parameter function requires at least seven known point data, eight-parameter function requires at least eight known point data, and eleven-parameter function requires at least eleven known point data. if there is more redundant known data, it can be calculated according to the least square criterion (liu et al., 2016; wang et al., 2016; wu & miao, 2010). x a pa a plt t= ( ) ⋅1 (11) in the formula, a is a coefficient matrix composed of known data x, y, h and xy, p is a weight matrix, generally a unit matrix, l is an observation matrix composed of known data , and x is the parameter matrix obtained. case analysis in order to analyze the feasibility of the ternary polynomial regression fitting method proposed in this paper, the first 12 points are taken as the reference points and the last 7 points are taken as the points to be fitted by using the data provided in literature (li, 2013). planar fitting and three-parameter ternary polynomial regression fitting are performed separately. the known data is shown in table 1 and table 2. after calculation, the parameters of plane fitting function are respectively a0 34 1044= . , a e1 8 0703 6=  . , a e2 3 1533 5=  . and, and the parameters of three-parameter ternary polynomial fitting are respectively a e0 3 6290 7=  . , a e1 4 4750 5=  . , a e2 2 3390 4=  . and. the fitting results are shown in table 3. by using the above data, quadric surface fitting and seven-parameter ternary polynomial regression fitting are used respectively. the fitting results are shown in table 4. using the above data, the ternary polynomial regression fitting of three parameters, four parameters, seven parameters, eight parameters and eleven table 1. data of reference points (unit: m) point number ordinate x abscissa y geodetic height h 1 3564231.786 499937.723 30.442 2 3565858.080 499248.000 31.225 3 3564029.592 499613.378 30.134 4 3565549.066 498813.558 31.110 5 3563826.318 499348.917 30.059 6 3566091.401 499632.434 30.709 7 3564312.203 500321.498 31.206 8 3566375.346 499179.740 26.424 9 3564827.161 500392.773 29.487 10 3566324.251 498659.474 25.790 11 3564001.762 500035.270 31.338 12 3567660.247 499189.334 31.128 table 2. data of points to be fitted (unit: m) point number ordinate x abscissa y geodetic height h 13 3566814.699 499080.199 30.989 14 3567961.396 498691.018 30.160 15 3566854.849 498567.506 30.741 16 3567538.805 499757.287 31.074 17 3566774.076 499599.745 30.621 18 3567524.909 500219.017 31.183 19 3568016.260 499235.369 30.332 table 3. plane fitting and three-parameter ternary polynomial regression fitting results. point number elevation anomaly/m plane fitting three-parameter ternary polynomial regression fitting fitting value/m residual/ cm fitting value/m residual/cm 13 21.0273 21.0566 2.93 21.0322 0.49 14 20.9978 21.0351 3.73 21.0146 1.68 15 21.0076 21.0401 3.25 21.0093 0.17 16 21.0639 21.0721 0.8 21.0622 -0.17 17 21.0710 21.0733 0.23 21.0555 -1.55 18 21.0516 21.0868 3.52 21.0829 3.13 19 21.0138 21.0518 3.8 21.0389 2.51 204 jianmin wang, dongping xie from the results in table 4, it can be seen that the fitting effect of sevenparameter ternary polynomial regression is better than that of three-parameter ternary polynomial regression. the average residual is reduced to 0.27 cm, and the overall residual is mostly within 1 cm. similarly, the fitting effect of quadric surface is better than that of plane fitting. the minimum residual is reduced to 0.05 cm, and the average residual is reduced to 1 cm. it can be seen that the overall effect of seven-parameter ternary polynomial regression fitting is better than that of quadric surface fitting. conclusion in this paper, in the process of gps elevation fitting, the geodetic height is included in the calculation system, and the ternary polynomial regression method is used to solve the elevation anomaly. the method takes into account the influence of the geodetic height on the elevation anomaly and avoids the phenomenon of distortion caused by the omission of the geodetic height. the feasibility of the proposed method and its superiority in fitting the global elevation anomaly in a large range are verified by the experimental results. in addition, it can be seen from table 5 that in the actual engineering application, the fitting effects of the seven-parameter and eight-parameter are the best. the internal fitting accuracy is less than 0.2 cm, and the external fitting accuracy is about 1.5 cm. this paper puts forward new ideas and directions for gps elevation fitting. in order to solve practical problems better, it can be combined with other methods, and a series of problems related to accuracy need to be tested and discussed further. acknowledgement national natural science foundation project (no. 41474020). references chen, y., ma, q., liu, r., & mao, x. (2014). discussion on the relative height anomaly of gps elevation fitting method. bulletin of surveying and mapping, (12), 67-69. ding, h., & sun, j. (2013). research on total least-squares methods for transformation of gps elevation. journal of geodesy and geodynamics, 33(03), 52-55+63. gong, x., chen, q., & zhou, x. (2014). the application research of gps height fitting based on tls adjustment method. bulletin of surveying and mapping, (03), 6-8. guo, h. (2018). transformation of gps height and accuracy analysis based on combination model. science of surveying and mapping, 43(02), 34-38. li, j. (2013). research on regional gps height anomaly fitting and modeling method. kunming university of science and technology. liu, b., guo, j., shi, j., & wu, d. (2016). a gps height fitting method based on the egm2008 model and terrain correction. geomatics and information science of wuhan university, 41(04), 554-558. liu, y., zheng, n., zhang, x., & ge, l. (2016). robust weighted total least squares estimation for gps leveling fitting. journal of geodesy and geodynamics, 36(01), 30-34. table 4. quadratic surface fitting and seven-parameter ternary polynomial regression fitting results. point number elevation anomaly/m quadratic surface fitting seven-parameter ternary polynomial regression fitting fitting value/m residual/ cm fitting value/m residual/ cm 13 21.0273 21.0268 -0.05 21.0233 -0.4 14 20.9978 20.9943 -0.35 20.991 -0.68 15 21.0076 21.0062 -0.14 20.998 -0.96 16 21.0639 21.0351 -2.88 21.0525 -1.14 17 21.0710 21.0437 -2.73 21.0494 -2.16 18 21.0516 21.0447 -0.69 21.078 2.64 19 21.0138 21.0119 -0.19 21.0217 0.79 table 5. ternary polynomial regression fitting accuracy parameter form fitting accuracy /cm internal fitting external fitting three parameters 0.80 1.76 four parameters 4.28 4.09 seven parameters 0.16 1.53 eight parameters 0.09 1.58 eleven parameters 1.02 1.23 5 4 3 2 1 0 3 4 5 6 7 8 9 10 11 figure 2. ternary polynomial fitting accuracy trend 205gps elevation fitting study based on ternary polynomial regression li, z., & huang, j. (2016). gps measurement and data processing. wuhan: wuhan university press, 385-386. ren, c., liang, y., lan, l., & pang, g. (2015). influence of different combination methods of gps elevation fitting. journal of geodesy and geodynamics, 35(06), 1036-1040+1045. shijun, l., & li, s. (2015). gps height fitting research considering topographic correction. bulletin of surveying and mapping, (07), 6667+93. škrekovski, r., dimitrov, d., zhong, j.m., wu, h. l., & gao, w. (2019). remarks on multiplicative atom-bond connectivity index. ieee access, 7(1), 76806-76811. song, c., bei, jinzhong, & dang, y. (2014). application of principal component analysis in gps leveling data processing. science of surveying and mapping, 39(06), 90-93. sun, j., cui, x., & guo, s. (2014). fitting method of height anomaly by spherical cap harmonic function based on k-means cluster analysis. bulletin of surveying and mapping, (02), 20-22. tang, j., xu, t., & long, y. (2016). analysis of wavelet function used in the fitting of gps elevation. bulletin of surveying and mapping, (03), 58-60. wang, l., wu, f., & wu, l. (2016). total least squares combination method of gps height transformation. surveying and mapping engineering, 25(04), 11-14. wang, l., wu, f., & wu, l. (2016). total least squares fitting estimation model for gps height transformation. geomatics and information science of wuhan university, 41(09), 1259-1264. wu, j., & miao, h. (2010). ill-posed matrix and partial biased estimation method in measurement data processing. bulletin of surveying and mapping, (09), 9-11. yang, f., & xie, y. (2017). discussion on satellite positioning elevation fitting based on lssvm. journal of navigation and positioning, 5(01), 100102+114. zhang, p., yu, d., zhang y., & xu, x. (2015). research on gps elevation fitting method and accuracy comparison experiment. bulletin of surveying and mapping, (09), 54-56. zhao, h., zhang, s., & zhang, q. (2011). gps height fitting of weighted total least-squares adjustments. journal of geodesy and geodynamics, 31(05), 88-90+96. zhao, q., xu, a., & xu, z. (2017). gps levelling based on robust total least squares. journal of navigation and positioning, 5(01), 95-99. zhou, c., sun, j., & guo, s. (2016). height anomaly fitting method based on mobile-polyhedral function. bulletin of surveying and mapping, (12), 25-27+38. keywords: tight reservoir; volume fractured horizontal well; productivity model; start-up pressure gradient; productivity characteristics. palabras clave: reservorio compacto; pozo horizontal fracturado; modelo de productividad; presión inicial de gradiente; características de productividad. how to cite item fan, y., lifeng, l., qiquan, r., jinping, k., mengya, x., & jiaxin, d. (2020). the characteristics of unconventional tight oil reservoir and its modification technology and productivity prediction. earth sciences research journal, 24(4), 507-512. doi: https:// doi.org/10.15446/esrj.v24n4.91473 the zonal productivity prediction model of volume fractured horizontal wells in tight reservoirs is established in this study based on the physical model of volume fractured horizontal wells to study the near-well seepage law of volume fractured horizontal wells in unconventional tight reservoirs. the conformal transformation and iteration method are used to solve the problem. compared with fan zifei's and ning zhengfu's method, the optimization of fractured horizontal wells' production parameters is studied. the effects of starting pressure gradient and fracture cluster number on horizontal wells' productivity with volume fracturing are analyzed. multivariate regression analysis is used to rank the importance of influencing factors. the results show that the starting pressure gradient has the most significant impact on capacity. the application example of volume fractured horizontal wells in lucaogou tight reservoir shows that the zonal productivity model established in this study can be used to predict the initial productivity of volume fractured horizontal wells. thus, through theoretical analysis and numerical simulation, the productivity prediction model of horizontal wells with volume fracturing has certain guiding significance for the effective development of tight reservoirs. abstract the characteristics of unconventional tight oil reservoir and its modification technology and productivity prediction registro de enterramiento de carbones del carbonífero-pérmico y su impacto en la generación secundaria de hidrocarburos en la subcuenca de jiyang, china issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.91473 en este estudio se establece el modelo de predicción de la productividad zonal en los pozos horizontales fracturados en reservorios compactos con base en un modelo fìsico de pozos horizontales fracturados para estudiar el regimen de filtración en reservorios compactos no convencionales. los autores utilizaron la transformación conforme y el método de reiteración para resolver el problema. se estudió la optimización de los parámetros de producción en pozos horizontales fracturados en comparación con los métodos de fan zifei y de ning zhengfu. además, se analizaron los efectos de la presión inicial de la inclinación y el complejo de fracturas en la productividad de los pozos horizontales con el índice de fracturamiento. el análisis de regresión multivariante se utilizó para clasificar la importancia de los factores influenciadores. los resultados muestran que la presión inicial de la inclinación tiene el mayor impacto en la capacidad. el ejemplo de aplicación de pozos horizontales fracturados en el reservorio compacto de lucaogou muestra que el modelo de productividad zonal que se estableció en este estudio puede usarse para predecir la productividad inicial en estos pozos. de este modo, a través del análisis teórico y la simulación numérica, el modelo de predicción de productividad de pozos horizontales fracturados tiene una significancia para el desarrollo efectivo de reservorios compactos. resumen record manuscript received: 12/03/2019 accepted for publication: 20/09/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 507-512 yang fan1, liu lifeng1, ran qiquan1*, kong jinping2, xu mengya1, dong jiaxin1 1petrochina research institute of petroleum exploration and development, beijing 100083, china 2langfang branch, petrochina research institute of petroleum exploration and development, langfang, china 065007 *corresponding author: yang.fan@gmail.com pe t r o l e u m e n g in e e r in g https://doi.org/10.15446/esrj.v24n4.91473 https://doi.org/10.15446/esrj.v24n4.91473 https://doi.org/10.15446/esrj.v24n4.91473 508 yang fan1, liu lifeng1, ran qiquan, kong jinping, xu mengya, dong jiaxin introduction as an important part of energy, oil is the foundation of national economic development. in recent years, conventional energy has been unable to meet human needs, so unconventional energy has come into people’s vision, and people pay more and more attention to unconventional energy (luo et al., 2018). as an important part of unconventional energy, the proportion of tight sandstone reservoirs in energy utilization is also increasing. the exploration shows that the geological reserves of tight reservoirs are relatively high, but at present the recovery rate is relatively low, and there is a huge potential for exploration. the permeability of tight oil reservoir is very low, the pore structure is very complex and the heterogeneity is very serious. without natural modification, its seepage resistance is large and its natural productivity is very low, which makes it impossible to achieve large-scale and effective development. although many researchers have conducted in-depth research on the development of tight reservoirs, and at the same time, many development strategies have been put forward, some of which have achieved good results, but there are also some limitations (guo et al., 2018). when a tight reservoir is fractured, there is mutual interference between fractures, and it is unavoidable. segmental multi-cluster fracturing technology applied to tight reservoirs can be defined as volume modification technology. manpower fracture formed by reconstruction measures can activate natural micro-fracture in reservoir and form complex fracture net system (naik et al., 2018). however, the specific form of complex sewing nets, how to influence productivity, how to give full play to the advantages of sewing nets in reservoir development, so as to reduce the adverse effects of sewing nets, need more in-depth study. at present, the development of tight reservoirs is still in the exploratory stage. there is no specific description of seepage characteristics near wells of volume fractured horizontal wells in tight reservoirs. it is necessary to study relevant seepage theory and numerical simulation, establish relevant productivity prediction models, analyze the technical methods and fracturing development design methods for improving single well productivity of volume fractured horizontal wells in tight reservoirs, and provides theoretical support for the development of tight reservoirs. steady-state productivity of fractured horizontal wells in tight reservoirs comparison of productivity models for fractured horizontal wells in tight reservoirs the steady-state solution of rectangular boundary reservoir with vertical fracture is studied. then, on this basis, the steady productivity formula of fractured horizontal wells is established. the model assumes that a horizontal well exists in the center of a rectangular reservoir, and the streamline flows radially into the horizontal wellbore from the direction perpendicular to the horizontal section. the steady-state productivity formula of fractured horizontal wells can be expressed by image reflection method and superposition principle. q n k h p p b l h w h h nw h h l in hh o e wf o o o o o 0 1 2 2 1 2 4 2 = −( ) +     + + π µ β πα β β hh k h k c in h rf h o f f w +                 − 2 1 (1) h arcch ch b na l a f 1 2 2 =             p p sin (2) h in h rw 2 3 1 = +( ) β π β (3) the advantage of fan zifei's model is that the equation is simple, which can adjust the thickness of the fractured reservoir and consider the heterogeneity of the reservoir. however, the disadvantage of this model is that it does not consider the interaction between fractures and the process of fluid flowing from matrix to wellbore during production. compared with the radius of horizontal wellbore, the half-length of fracture is larger. when the influence of gravity is neglected, the flow process of fluid in fracture near wellbore can be simplified to plane radial flow, which means that formation thickness is w, flow radius is h/2, boundary pressure is pfj and bottom hole pressure is pwfj. when the skin factor of crack is neglected, the expression is as follows: p p q b k w in h rfj wfj fj f w − = µ π2 2 (4) if the pressure at the bottom of all fractures equals the pressure at the bottom of the well, that is to say p pwfj wf= , it can be obtained that: p p b k h q in r x id x r x id x jd id e wf h fi e f f e f f − = −       + + −       − µ π2 1 2 xx jd id x q b k w in h r f f fi f i n n w o o + + −      + = − ∑ 1 2 22 µ π (5) p p b k h q in r x id x r x id x jd id e wf h fi e f f e f f − = −       + + −       − µ π2 1 2 xx jd id x q b k w in h r f f fi f i n n w o o + + −      + = − ∑ 1 2 22 µ π according to the parity of the number of cracks n, the order from -no to no is adopted for j. therefore, a set of equations can be generated which can be solved in closed form. the output qfi of each crack can be obtained by using the column principal component gauss-jordan numerical solution method. the numerical value of the production of all fractures and the total production of fractured horizontal wells are as follows: q qfi i n n o o = = ∑ (6) ning zhengfu’s model assumes that each fracture has different production values, and the seepage resistance of fluid in the fracture is taken into account. compared with lang zhaoxin's equation, the advantage of this equation is that the hypothetical conditions of the model are more in line with the actual production situation of fractured horizontal wells, so it has stronger applicability (zhuoheng et al., 2018). at the same time, the disadvantage of this model is that the pressure drop in horizontal wellbore is not taken into account. based on the theory of seepage equivalent resistance, li et al. established the productivity equation of fractured horizontal wells. for the formation with fractured horizontal wells, fluid flow from formation to horizontal wellbore can be divided into two stages. in the first stage, fluid flows from formation to fracture. in the second stage, the fluid flows from the fracture to the horizontal wellbore. when n fractures are fractured in horizontal wells, the whole flow area can be divided into n regions. the fracture is the center of each flow area. the pressure drop of fluid flowing from formation to fracture is as follows: ∆p q l x kh l l q khx l l i f f f i f f 1 0 1 2 0 1 2 2 2 1 1 = −( ) + + +        * ( (7) the pressure drop of fluid flowing from fracture to horizontal wellbore is as follows: ∆p q x k hwi o f f 2 2 =  (8) the pressure drop produced by fluid flowing into the wellbore in a fracture is: 509the characteristics of unconventional tight oil reservoir and its modification technology and productivity prediction if any fracture is selected, the oil drainage area formed by fluid flowing from matrix to fracture is elliptical. if the long half-axis of the ellipse is a, the short half-axis is b, and the focal length, that is, the half-length of the crack, be xf 2 , the following equation can be obtained. a b xf2 2 2 2 = +       (13) here, the elliptical flow in z plane is transformed into radial flow in  plane by conformal transformation. the transformation coefficient is: z xf= +     / 2 2 1   (14) equation 14 is a rokovsky equation, in which z x iy= + (15) ξ ξ η θ= + =i rei (16) if equations 15 and 16 are brought into the upper equation, there are x iy x re r e x r r i r r f i i f+ = +     = +     + −     −/ / cos 2 2 1 2 2 1 1         sin (17) by calculating the steady-state productivity equations under various conditions, the results are analyzed. the results show that the calculated results by ning zhengfu’s method are the closest to those by the productivity equation deduced in this study, but these results are higher than the actual productivity (yin, 2018). this is because both methods are based on the assumption that there is no interference from other fractures in the control area of a single fracture. at the same time, the assumed conditions in this equation do not accord with the actual situation, that is, assuming that the formation is homogeneous reservoir, the fracture can completely penetrate the reservoir (wei-yao, & dong-xu, 2018). therefore, although the productivity equation in this study can roughly predict reservoir productivity, it needs to be further adjusted to meet the actual situation. influencing factors of productivity of volume fractured horizontal wells in dense reservoirs reservoir characteristics the oil-bearing formation is located in lucaogou formation of permian, which is divided into four units: p2121, p2122, p2111 and p2122 from bottom to top. the thickness of each unit layer group is shown in table 1. table 1. the thickness of each unit layer group layer group p2111 p2112 p2121 p2122 formation thickness (m) 28~75 63~129 8~34 50~120 average thickness (m) 52.3 104 17.4 90.5 upper sweet heart (p222) and lower sweet heart (p212) are developed in lucaogou group. reservoir physical parameters are shown in table 2, and the formation thickness between upper and lower desserts is 83.3m~131.7m, with an average of 108.4m. according to pore and permeability distribution, it is a typical tight oil reservoir. productivity increasing potential evaluation of volume fracturing horizontal wells with the continuous development of horizontal wells and fracturing measures, economic development of tight reservoirs has been realized under ∆p q k w in h rf w 3 01 2 2 2 = −    π µ π (9) since, p p p p pr wf− = + +∆ ∆ ∆1 2 3 (10) it can be obtained that: q p b l x kh l l khx l l x k hw i f f f f f f f f = − +( ) + +       + + 170 54 1 1 11 2 1 2 . * ∆ µ 11 2 2k w in h rf wπ π −                     (11) q p b l x kh l l khx l l x k hw i f f f f f f f f = − +( ) + +       + + 170 54 1 1 11 2 1 2 . * ∆ µ 11 2 2k w in h rf wπ π −                     the total flow rate of n vertical fractures is as follows: q qi i i n = = ∑ 1 (12) the advantage of li model is that it can control the perforation state of horizontal well section and distinguish the end fracture from the middle fracture. at the same time, the model also has shortcomings. it does not take into account the loss of pressure drop in horizontal wellbore and the decline of fracture conductivity with time. establishment of steady-state productivity model in order to describe the special percolation process of tight reservoir, the model is simplified as a single fracture. at this time, the flow process of crude oil is divided into three stages: the flow in matrix (non-linear flow from matrix to fracture wall), the flow from fracture wall to wellbore (plane radial flow) and the flow in wellbore (pressure drop due to resistance along the way) (qun et al., 2018). in the solution process, the first two processes are coupled to obtain the equivalent radius of the vertical well and the equivalent diameter in the presence of a single fracture. then, the multiple fractures are converted into corresponding production wells, and the oil production of each fracture under the condition of multiple fractures is obtained by pressure drop superposition. finally, the pressure drop flow in the horizontal wellbore is considered and combined with the previous output to carry out the iterative solution (peng et al., 2018). finally, the oil production under the control of each fracture can be obtained. the total oil production of horizontal wells after fracturing can be obtained by summing up the production of each fracture, as shown in figure 1. figure 1. fracture elliptic seepage if there is no interference between the cracks and the fluid in the matrix flows to the cracks, any crack can be regarded as an elliptical seepage flow. 510 yang fan1, liu lifeng1, ran qiquan, kong jinping, xu mengya, dong jiaxin the multi-stage fracturing horizontal wells. at present, the tight reservoirs developed in china are mainly in ordos basin and songliao basin. the cluster well development initially adopted has not achieved satisfactory results. in recent years, the application of volume fracturing horizontal wells in tight reservoirs has achieved good results (long et al., 2018). based on the characteristics of tight reservoirs, the application research of horizontal wells, conventional fractured horizontal wells and volumetric fractured horizontal wells is carried out in the depletion development of tight reservoirs. according to the physical characteristics of tight reservoir, the permeability of formation matrix is 0.05md. considering formation heterogeneity, kx = ky, kz = 0.1kx, porosity 10%, effective thickness of reservoir 15m, original formation pressure 18.6mpa, productivity and cumulative production of horizontal well, conventional fractured horizontal well and volumetric fractured horizontal well in tight reservoir are analyzed. from figures 2 and 3, it can be seen that conventional fractured horizontal wells in tight reservoirs have higher productivity than conventional fractured horizontal wells, while volumetric fractured horizontal wells have much higher productivity than conventional fractured horizontal wells, and the corresponding cumulative production is also much larger. the production of horizontal wells is always at a low level during production. the production of conventional fractured horizontal wells is higher at the initial stage, while that of volumetric fractured horizontal wells is higher at the initial stage, and then decreases with production (ren et al., 2019). the reason is that in horizontal wells, the use range is small, only near the horizontal wells. conventional volume fracturing horizontal wells are used only in fracture wave and small area. volume fractured horizontal wells have higher production and cumulative production because of the role of fracturing net and a wide use range. 0 500 1000 1500 2000 2500 3000 0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 d ai ly o ut pu t/m 3 d -1 time/day horizontal well conventional fracturing horizontal well volume fracturing horizontal well figure 2. daily production curve of different well types effect of starting pressure gradient on capacity because of the low porosity and permeability of tight reservoirs and the serious heterogeneity, the fluid from static to flowing needs to overcome certain binding force. only when the pressure gradient exceeds a certain critical value can the fluid flow. the critical value is called the starting pressure gradient. the starting pressure gradient varies with the change of reservoir physical permeability. the starting pressure gradients are taken as g=0.01mpa/m, 0.03mpa/m, 0.05mpa/m, 0.06mpa/m and 0.08mpa/m, respectively. the effect of start-up pressure gradient on productivity of horizontal well in volume fracturing is analyzed. figure 4 and figure 5 are daily production curves and cumulative production curves with different starting pressure gradients. from the daily production curve, it can be seen that the daily production decreases with the increase of starting pressure gradient, and the recovery degree decreases with the increase of starting pressure gradient. when the starting pressure gradient is 0.08mpa/m, the productivity and recovery degree are very low. this is because the bigger the start-up pressure gradient of reservoir is, the higher the pressure gradient required for fluid flow in reservoir will be, thus the pressure sweep range will be reduced, the use range will be reduced, and the recovery factor will be reduced. 0 250 500 750 1000 1250 1500 1750 2000 0 5 10 15 20 25 d ai ly o ut pu t/m 3/ d time/day g=0.01mpa/m g=0.03mpa/m g=0.05mpa/m g=0.06mpa/m g=0.08mpa/m data fitting figure 4. daily production curves with different starting pressure gradients table 2. physical parameters of tight oil reservoirs middle depth (m) formation temperature (°c) formation pressure (mpa) porosity (%) permeability (md) reservoir thickness (m) upper sweet heart 2980 85.6 37.9 5.09~25.79 average 1099 0.001~0.284 average 0.012 9.40~42.4 average 28.34 lower sweet heart 3250 92.0 41.4 5.51~20.72 average 11.62 0.002~2.76 average 0.010 7.50~67.24 average 31.26 0 500 1000 1500 2000 2500 3000 0 2000 4000 6000 8000 10000 12000 14000 16000 c um ul at iv e ou tp ut /m 3 time/day horizontal well conventional fracturing horizontal well volume fracturing horizontal well figure 3. cumulative production curves for different well types 511the characteristics of unconventional tight oil reservoir and its modification technology and productivity prediction 0 250 500 750 1000 1250 1500 1750 2000 0 2000 4000 6000 8000 10000 12000 c um ul at iv e ou tp ut /m 3 time/day g=0.01mpa/m g=0.03mpa/m g=0.05mpa/m g=0.06mpa/m g=0.08mpa/m data fitting figure 5. cumulative production curves with different starting pressure gradients effect of fracture cluster number on productivity volume fracturing is used to fracturing horizontal wells in many stages, and many clusters of fractures are produced in each stage. the number of fracture clusters affects the degree of opening natural fractures in each section and the degree of reformation in the fracturing reconstruction area. the number of fracture clusters in each section is designed as four schemes: 1 piece, 2 pieces, 3 pieces and 4 pieces, respectively. the effect of the number of fracture clusters on the productivity is analyzed. 0 250 500 750 1000 1250 1500 1750 2000 0 10 20 30 40 50 d ai ly o ut pu t/m 3/ d time/day cluster per paragraph two clusters per segment three clusters per segment four clusters per segment figure 6. daily production curve of different clusters figures 6 and 7 are daily and cumulative production curves for each segment with different cluster numbers. it can be seen from the daily production curve that the production increases with the increase of cluster number. moreover, the number of clusters has a great influence on the initial and intermediate stages of production. the more the number of cracks clusters in each section is, the greater the density of artificial and natural cracks communicated is, the greater the degree of fracturing reconstruction area transformation is, and the greater the productivity is. conclusion according to the characteristics of tight reservoir and considering the characteristics of volume fracturing, according to the basic principles of percolation mechanics and reservoir engineering, experimental research, theoretical derivation, numerical simulation and other research methods have been adopted to comprehensively study the percolation characteristics, productivity evaluation and utilization characteristics of volume fracturing in tight reservoir, analyze the development effects under different well pattern and different development modes, and establish the seepage model and productivity prediction model of horizontal well reservoir with compact oil volume fracturing, and form the productivity prediction module. compared with the horizontal wells without volume fracturing, the horizontal wells with multi-stage cluster fracturing enlarge the reconstruction volume, reduce the seepage resistance and increase productivity. because the permeability of tight reservoir is very low, the daily production is higher in the initial stage and the production decreases rapidly. under the same cluster number, the smaller the cluster spacing between intervals and the cluster spacing within intervals is, that is, the more uniform the distribution of fractures in horizontal wells is, the higher the productivity of horizontal wells is. for the same series, the more uniform the section number distribution is, the higher the production of the oil well is. therefore, in fracturing operation of tight reservoirs, in horizontal wells, the more uniform the distribution of interval and cluster spacing is, the better. in view of the seepage characteristics of unconventional tight reservoirs, on the basis of the derivation method of lang zhaoxin’s equation, considering the starting fracturing gradient, the productivity equation of fractured horizontal wells is derived. the productivity equation takes into account the loss of flow pressure caused by seepage resistance in cracks, and fully reflects the characteristic parameters of actual cracks, such as unequal length, unequal conductivity and unequal spacing. it is closer to the actual situation than the new equation of lang zhaoxin, and more suitable for productivity evaluation of fractured horizontal wells in tight reservoirs. the new productivity equation only considers the steady-state flow of single-phase fluid, does not consider the effect of pressure sensitivity on the production of horizontal wells in tight reservoirs, and ignores the pressure loss of fluid flow in horizontal wellbore, so it needs to be more perfect and closer to the actual productivity equation of fractured horizontal wells. references guo, c., wei, m., & liu, h. (2018). study of gas production from shale reservoirs with multi-stage hydraulic fracturing horizontal well considering multiple transport mechanisms. plos one, 13(1), pp.e0188480. long, j., cheng, y., han, z., gao, q., yan, c., wang, h., & fu, l. (2018). effect of liquid nitrogen cooling on the permeability and mechanical characteristics of anisotropic shale. journal of petroleum exploration & production technology, 9(1), 111-124. luo, h., li, h., zhang, j., wang, j., wan, k., xia, t., & zhu, x. (2018). production performance analysis of fractured horizontal well in tight oil reservoir. journal of petroleum exploration & production technology, 8(1), 229-247. 0 250 500 750 1000 1250 1500 1750 2000 0 1000 2000 3000 4000 5000 6000 7000 8000 9000 10000 11000 12000 13000 14000 c um ul at iv e ou tp ut /m 3 time/day cluster per paragraph two clusters per segment three clusters per segment four clusters per segment figure 7. cumulative production curves of different clusters 512 yang fan1, liu lifeng1, ran qiquan, kong jinping, xu mengya, dong jiaxin naik, s., malgaresi, g., you, z., & bedrikovetsky, p. (2018). well productivity enhancement by applying nanofluids for wettability alteration. appea journal, 58(1), pp.121-129. doi: https://doi.org/10.1071/aj17149 peng, w. j., li, q., song, a., & jin, z. j. (2018). spatial-temporal variations of co2 and ch4 flux through a water-air interface under the effect of primary productivity in wulixia reservoir. environmental sciences, 39(6), 2673-2679. doi: 10.13227/j.hjkx.201709119 qun, l., lifeng, y., yaoyao, d., dingwei, w., xin, w., baoshan, g., zhen, w., & ying, w. (2018). the “fracture-controlled reserves” based stimulation technology for unconventional oil and gas reservoirs. petroleum exploration and development, 45(04), pp.210-218. doi: https://doi. org/10.1016/s1876-3804(18)30080-6 ren, l., su, y., zhan, s., & meng, f. (2019). progress of the research on productivity prediction methods for stimulated reservoir volume (srv)-fractured horizontal wells in unconventional hydrocarbon reservoirs. arabian journal of geosciences, 12(6), 184. doi: 10.1007/ s12517-019-4376-2. wei-yao, z., & dong-xu, m. (2018). effective stress characteristics in shale and its effect on productivity. natural gas geoscience, 3(6), 339-346. doi: https://doi.org/10.1016/j.jnggs.2018.12.001 yin, c. (2018). test and analysis on the permeability of fractured fractures in shale reservoirs. natural gas industry, 5(5), 513-522. doi: https://doi. org/10.1016/j.ngib.2018.03.006 zhuoheng, c., chao, y., chunqing, j., dan, k., kezhen, h., xiaojun, l., & melinda, y. (2018). production characteristics and sweet-spots mapping of the upper devonian-lower mississippian bakken formation tight oil in southeastern saskatchewan, canada. petroleum exploration and development, 45(4), 102-112. doi: https://doi. org/10.1016/s1876-3804(18)30070-3 https://doi.org/10.1071/aj17149 10.13227/j.hjkx https://doi.org/10.1016/s1876-3804(18)30080-6 https://doi.org/10.1016/s1876-3804(18)30080-6 https://doi.org/10.1016/j.jnggs.2018.12.001 https://doi.org/10.1016/j.ngib.2018.03.006 https://doi.org/10.1016/j.ngib.2018.03.006 https://doi.org/10.1016/s1876-3804(18)30070-3 https://doi.org/10.1016/s1876-3804(18)30070-3 keywords: local t-s model; seismic surface; nonlinear response; switching; control; damper. palabras clave: modelo local t-s; superficie sísmica; respuesta no lineal; traspuesta; control; apagador. how to cite item jiang, y., luo, j., azarkhosh, h., & wu, e. (2020). the switching control method for the nonlinear response of seismic surface based on the local t-s model. earth sciences research journal, 24(3), 327-333. doi: https://doi. org/10.15446/esrj.v24n3.90432 based on the local t-s model, the switching control method for the nonlinear response of the seismic surface is studied, and the stiffness matrix of the composite interlayer element of viscoelastic damper and herringbone support is derived. the high-order single step ß method is applied to the nonlinear seismic response analysis and the instantaneous optimal active switching control. the local t-s model and controller are designed, and the approach and control of global nonlinear responses are realized through the multi-model switching control. the results show that the proposed method can accurately reflect the reduction of seismic ground load and the switching control of the main structure when the viscoelastic damper is installed. the switching control effect of the maximum spatial displacement becomes more and more obvious with the change of time history, and the control effect of all layers applying the switching control force is the most significant. whether the controller is disturbed or not, it is always good and stable performance. abstract the switching control method for the nonlinear response of seismic surface based on the local t-s model el método de control de conmutación para la respuesta no lineal de la superficie sísmica con base en el modelo local t-s issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n3.90432 este artículo estudia el método de control de conmutación para la respuesta no lineal de la superficie sísmica con base en el modelo local t-s, y deriva la matriz de rigidez del elemento compuesto de capa intermedia del amortiguador viscoelástico y el soporte de espiga. el método ß de un solo paso de alto orden se aplica al análisis de respuesta sísmica no lineal y al control de conmutación activo óptimo instantáneo. el modelo local t-s y el controlador están diseñados, y el enfoque y el control de la respuesta global no lineal se realizan a través del control de conmutación multimodelo. los resultados muestran que el método propuesto puede reflejar con precisión la reducción de la carga de tierra sísmica y el control de conmutación de la estructura principal cuando se instala el amortiguador viscoelástico. el efecto de control de cambio del desplazamiento espacial máximo se vuelve cada vez más obvio con el cambio del historial de tiempo, y el efecto de control de todas las capas que aplican la fuerza de control de cambio es el más significativo. ya sea que el controlador esté alterado o no, siempre es un rendimiento bueno y estable. resumen record manuscript received: 08/02/2020 accepted for publication: 11/07/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 3 (september, 2020): 327-333 yudong jiang1*, jinming luo2, hojatallah azarkhosh3, erjun wu3 1college of college engineering, nanjing university of aeronautics and astronautics, nanjing, 210016, china 2college of automation, shenyang institute of engineering, shenyang, 110136, china 3college of civil and transportation engineering, hohai university, nanjing, china *corresponding author: luojinming0830@163.com; hojatazarkhosh@gmail.com se is m o l o g y 328 yudong jiang, jinming luo, hojatallah azarkhosh, erjun wu introduction a large number of earthquake damage and existing records show that the earthquake will cause indelible damage to people’s lives and property. with the development of china’s national economy and the progress of science and technology, high-strength and light-weight building materials are widely used in building structures (altukhov et al., 2018), the stiffness and damping of the structures are constantly reduced, and the building structures are more sensitive to seismic excitation. in order to reduce the seismic response of the building structure, in addition to studying the seismic design structure of the bearing system of the building structure itself (zhang et al., 2017), some energy dissipation devices are set on the structure to increase the structural damping and consume the seismic energy through the nonlinear deformation of the energy dissipation materials, so as to reduce the seismic response of the main structure and effectively suppress the structural vibration (zhang et al., 2016). among these energy dissipation devices, the viscoelastic damper is the most concerned one (ding et al., 2016), because it has the following advantages compared with other energy dissipation devices: as long as the structure starts to vibrate under minor interference, it can consume energy immediately (zhang et al., 2018). therefore, even in the case of small elastic vibration, it also plays the role of earthquake control. this makes it not only suppresses the seismic response of the structure but also avoid the problem of how the initial stiffness of the damper matches the lateral stiffness of the structure existing in other energy dissipation dampers (wei et al., 2016); its “force-displacement” hysteretic curve is similar to the ellipse, so its energy dissipation is very strong (bande & choudhury, 2017). multi-model switching control is to approach the uncertainty and nonlinearity of response with multiple models. based on multiple models, a controller (viens et al., 2017) is established. multiple controllers can be selected to act on the same control object, so as to effectively control complex response (huang et al., 2018). it has nearly 40 years of development history. in recent ten years, some scholars have proposed multi-model control based on model switching. many achievements have been made in theory and practice. the t-s fuzzy system is a kind of dynamic response fuzzy identification method proposed by takagi and sugeno. its basic idea is to divide the input space into multiple fuzzy intervals, approximate each fuzzy interval with a linear model, and obtain the global model by local model interpolation. the larger the number of fuzzy rules is, the higher the precision is. however, when the number of rules of fuzzy reasoning is large, the calculation is complex and in some cases, no common positive definite matrix can be found to meet the stability conditions (acerbi et al., 2016; guo et al., 2017). in this paper, multi-model switching control of nonlinear response based on the local t-s model is proposed. the model controller is designed, and the viscoelastic damper damping device is added to the building. through the switching control, the switching control of the seismic surface response is realized. materials and methods combined stiffness matrix in the interlayer of viscoelastic damper and herringbone support stress-strain relationship of viscoelastic damping materials let and be the stress and strain of viscoelastic damping materials respectively. according to the research of tsai and lee, the stress-strain relationship satisfies the fractional derivative relationship f i t g g v i t g f i t∆ ∆ ∆( ) = +( ) ( ) + ( )0 1 1 (1) where f i t∆( ) is the original time effect of strain, and its expression is as follows: f i t i i t i v o i j v j tt t t∆ ∆( ) = ( ) + ( )    ( ) + ( )    (      1 1 11 1 )) = = ∑ j i 1 1 (2) g a t t e t u fdv t t 1 0 2 0 = ( ) ( ) + ( )  ∫ ∆ γ − − θ − (3) in the equation, g0, a0 and  are all material parameters, which can be measured by experiments. according to equation (1), the stress-strain curve of viscoelastic damping material is an ellipse. stiffness matrix and switching control force vector of interlayer element if the time interval is t, t = it, according to the layer model, only one direction (x-direction) of vibration is analyzed at a time, it means the relative displacement of viscoelastic damper nodes along the x direction, then: p b p g g a h b b u g af i t be t t t{ } =   = +( )     { } + ( )  ∆ ∆0 1 1 (4) in the equation, a is the acting area of the viscoelastic damper material; h is the thickness of the viscoelastic damper material; pe{ } is the load vector of the viscoelastic damper node, and only p p pe t{ } =  × ×1 2, along the x direction is considered; u u u t { } = { }1 2, is the displacement vector of the viscoelastic damper element node. from equation 4, it can be seen that the load vector of the viscoelastic damper element is composed of two parts. the former one is similar to the load vector generated by ordinary elastic elements, and the latter one is the restoring force vector unique to viscoelastic damper elements, namely the vector of energy dissipation switching control force of the element. let pev{ } be the vector of energy dissipation switching control force of the element. according to equation (4): p g af i t b g af i tev t{ } = ( )   = ( )       1 1 1 1 ∆ ∆  (5) let the element stiffness matrix of viscoelastic damper vibrate in one direction (x direction) be ke  , and it can get from equation (4): k g g a h b b a h g g ke t   = +( )     = +( )       =0 1 0 0 1 1 1 1 1 1 1 1 / /    1 1       (6) where k a h g g1 0 0 1= +( )/ is the element stiffness coefficient of the viscoelastic damper. if the coefficient of element lateral stiffness matrix is k2 and the lateral stiffness matrix of the element is ks , then: ks k  =      2 1 1 1 1   (7) where, k ea hs2 2 2 2 2 2 2 3 2 2 1 1 2= ( ) +  / / / ; as, e, 12 and h2 are the elastic modulus, cross-sectional area, span, and storey height of herringbone brace respectively. after the herringbone brace and viscoelastic damper are connected in series, the stiffness matrix of the interlayer element of the whole brace is as follows: k k k k kes   = ⋅ +       1 2 1 2 1 1 1 1   (8) high-order single step ß method for nonlinear response analysis of seismic surface considering nonlinearity, the equation of motion of shear-type structure expressed by the relative displacement between layers under the action of the seismic surface load is as follows: m x c x f m a xe s1 1 1  ′{ } +   ′{ } + { } =   { } ⋅ ⋅⋅ − (9) 329the switching control method for the nonlinear response of seismic surface based on the local t-s model controller design the parallel distributed compensation algorithm (pdc) is used to design the local state feedback controller in the same region. the fuzzy control rule i in the same region is as follows: if z t is m z t m u t ki p ip i and and is then 1 1( ) … ( ) ( ) =,  iix t( ) (18) therefore, the region controller (e.g., the q-th region rq, q  s) is expressed as: u h z t k x tq i i i r = ( )( ) ( ) = ∑ 1 (19) the global controller u t( ) is actually the output of the current regional controller selected by the multi-model switching strategy, which can be expressed as u t u t( ) = ( ) , where σ ∞t s s( )  ) → = …{ }: , , , ,0 1 2 is a piecewise constant function depending on time or state, u t( ) is generated by switching in the controller set u u us1 2, , ,…{ }, and has the form u u q st qσ ∈( ) = , , which can be further written as, u u h z t k x tt q j j r jσ −( ) = = = ( )( ) ( )∑ 1 (20) according to equation (20), the closed-loop fuzzy control system can be deduced as follows: (21) stability of nonlinear response of multi-model switching control based on local t-s model the lyapunov stability results of the multi-model switching response are all based on the following assumptions: limited switching times between local models in a limited time (wang, 2019). for the convenience of the following proof, the switching function of the multi-model switching control is described by the characteristic function q z t( )( ), namely: υ ∈ ∉ ∈q q q z t z t r z t r q s s( )( ) = ( )( )     = …{ } 1 0 1 , , , , , (22) it means that when the variable z(t) is in the region q, the controller q corresponding to the current region uq is selected by multi-model switching control. as a global controller u(t), controllers in other regions do not work. therefore, the global closed-loop control system and global controller of nonlinear reactive multi-model switching control based on local t-s model can be described as follows: (23) u h k xq qi qj i r q s = == ∑∑  11 (24) note: for the convenience of writing, x, q, hqi and u are used to represent the corresponding x t( ), q z t( )( ), h z tqi ( )( ) and u t( ), the same below. substituting equation (24) into equation (23), the global closed-loop fuzzy control system can be expressed as follows: (25) where, the restoring force of [m1], [c1] and {a} is calculated according to the following equation: f k x f f f k x f k xe ej ej ej i r i ef b i r b{ } =   ′{ } = = ′ + ′+1 1;   (10) p f k xpj r ej b i r b= ′ (11) where, f ej b are the current restoring force and the previous switching control restoring force generated by the i-th layer; ′xi and ′xb are the current displacement and the previous switching control displacement of the i-th layer; ppj r and ki r are the plastic load and the current stiffness of the i-th layer respectively; according to equation (9), (10) and (11), the motion equation after entering the nonlinearity can be obtained as follows: (12) where kr1   and pp r{ } represent the current stiffness matrix and plastic load matrix respectively. according to equation 12 and the assumption of the high-order singlestep method, the equation for step-by-step integration can be obtained as follows: (13) a new instantaneous optimal algorithm for active switching control the equation of motion for active switching control of structural seismic surface response is as follows: (14) where {u} represents the r1 order switching control force vector and [b] represents the nr order position matrix of switching control force action. the index function has energy meaning, and any instantaneous index function can be taken as: j x w k x x w m w t c x b t t= { }   { } + { }   +  ( ){ } +  1 2 1 2 1 2 1 2 3∆   { }( )     { }u k b ut 1 (15) where, w1, w2 and w3 are adjustable parameters of weight function. under the constraint of equation (14), let jn+ =1 0, and then it can get: b u d x d x d x d xn n n n  { } =   { } +   { }   ⋅{ } +   { } ⋅ ⋅+1 1 2 3 4 1 ggn+1 (16) in order to achieve closed-loop switching control, b u d x d x d x d xn n n n  { } =   { } +   { }   ⋅{ } +   { } ⋅ ⋅+1 1 2 3 4 1 ggn+1 is removed, so jn+1 of each step is no longer zero, that is, each step is only approximate optimization (17) from the above equation, the current applied switching control force u n{ } +1 can be obtained. 330 yudong jiang, jinming luo, hojatallah azarkhosh, erjun wu lemma 1 if at any time t, the number of activated fuzzy rules is less than or equal to r, then there are: υ − − υq qi q qi qj i ji r q s i r q s h r h h2 1111 1 1 2 0≥ ===== ∑∑∑∑∑ (26) prove according to this lemma: υ − − υ − υ −q qi q qi qj q qi qj ii r q s i j h r h h r h h2 1 2 11 1 1 2 1 1 0=    ≥ ==== ∑∑∑∑∑∑∑∑∑ ==== i r q s i r q s 1111 υ − − υ − υ −q qi q qi qj q qi qj ii r q s i j h r h h r h h2 1 2 11 1 1 2 1 1 0=    ≥ ==== ∑∑∑∑∑∑∑∑∑ ==== i r q s i r q s 1111 (27) from lemma 1, we can easily get lemma 2 as follows. lemma 2 if at any time is less than or equal to , 1 1<η ≤ , then there are: υ − − υq qi q qi qj i ji r q s i r q s h r h h2 1111 1 1 2 0≥ ===== ∑∑∑∑∑ (28) theorem it is assumed that at any time t, the number of activated fuzzy rules is less than or equal to  and . if there is a common symmetric positive definite matrix p and semi-positive definite matrix qq, the following linear matrix inequalities are obtained: g p pg qqii t qii q+ + ( )η− <1 0 (29) g g p p g g q i jqij qji t qji qji q +      + +      ≤2 2 0− <, (30) then the global closed-loop control system (25) based on the local t-s model is asymptotically stable. where, g a b kqij qj qi qj=  . prove let lyapunov function v x x pxt( ) = , where p is a positive definite matrix. along any trajectory of the closed-loop system (25), the derivative of v(x) with respect to time is: (31) where: g g g p p g gqij qji t qji qji= +      + +     2 2 (32) according to equation (29), (30) and lemma 2, there are: (33) from equation (29), when x t( ) ≠ 0, there is , so the global closed-loop control system (23) based on the local t-s model is asymptotically stable under the action of the global controller (24). according to the stability conditions of the above theorem, based on lmi, the design of feedback controller and local feedback gain kqi is to find a common symmetric positive definite matrix w, semi-positive definite matrix yq and matrix mqi, which makes the following linear matrix inequalities to be enabled: wa a w m b b m yqi t qi qi t qi t qi qi q+ + ( )− − η−1 0 (34) wa a w wa a w m b b m m b b m yqi t qi qj t qj qj t qi t qi qj qi t qj t qj qi q+ + +     2 0,i j wa a w wa a w m b b m m b b m yqi t qi qj t qj qj t qi t qi qj qi t qj t qj qi q+ + +     2 0,i j (35) where w p= 1, m k wqi qi= , y wq wq q= . with the help of the lmi toolbox in matlab, the feasp function is used to find the feasible solutions of w, yq and mqi which satisfy the inequality. then the local feedback gain , matrix p and qq are obtained, that is, k m wqi qi= 1, p w= 1, q py pq q= . results project overview a comprehensive building is a ten-story reinforced concrete frame structure, the height of the first to the third floor is 4.8m, the height of the fourth to the tenth floor is 4.2m, and the total height is 43.8m. the design of seismic crack resistance is 8 °, class ii site. the viscoelastic damper is placed on the top of herringbone support, beam kl1 is 550mm × 250mm, kl2 is 300mm × 250mm, column section size is 500mm × 500mm, and steel support area is 4260 mm2. in the design of viscoelastic damper support, under the principle of reducing project cost, the stiffness of viscoelastic damper support should be increased as much as possible to ensure that viscoelastic dampers can produce enough restoring control force. in this design, zn22 damping material produced by a factory is selected as the viscoelastic damping material, with material constants of g0, a0, and θ of 588, 2287, and 100.0, respectively. the unit is knm, and the area of single viscoelastic damping material is 0.7x0.7=0.49m2, the thickness is 3mm, and the working temperature is 20 ºc. calculation results and analysis in order to verify the performance of the method in this paper, the time history analysis and calculation of the nonlinear seismic response of the structure supported by the viscoelastic damper (ved), the structure only supported by the same type of steel herringbone support and the original pure frame structure are carried out respectively. in the calculation, the el-centro seismic surface acceleration record with a duration of 10.0s (its predominant period is 0.55s) is selected for the seismic surface wave. considering the rare seismic surface action, the seismic surface acceleration xg is 400gal. the calculation results of the maximum displacement of each floor of the three structures are shown in figure 1. figure 1. the horizontal displacement of each layer of the structure 331the switching control method for the nonlinear response of seismic surface based on the local t-s model table 1 shows the cracking and entering plasticity of the three structural layers. table 1. structural layer cracking and entering the plastic stage structural layer with ved cracking 1 ~ 3 layers the inelastic there is no only diagonally supported structural layers cracking 1 ~ 6 layers the inelastic 1 layer pure frame structure layer cracking 1 ~ 9 layers the inelastic 1 ~ 4 layers it can be seen from figure 1 and table 1 that the displacement of the top layer and the shear force of the bottom layer of the structure with viscoelastic damper support is reduced by about 35% compared with that of the structure with herringbone support only, and about 70% compared with that of the pure frame structure; the number of cracking layers of the structure with viscoelastic damper herringbone support is reduced, the plastic stage of the structure layer is avoided, and the collapse of the structure is prevented. the viscoelastic damper can produce a strong energy dissipation control force, significantly increase the energy dissipation capacity of the whole structure, weaken the seismic ground load transmitted to the mainframe structure, and achieve the purpose of structural vibration switching control. the results show that this method can accurately reflect the seismic response of the structure with a viscoelastic damper, and it is a more practical and accurate calculation method. comparison of maximum interlayer displacement the lateral stiffness k1 of each floor of the ten-story complex building is 3.4105, 3.26105, 2.85105, 2.71105, 2.69105, 2.43105, 2.23105, 2.07105, 1.69105, 1.37105kn/m. the layer damping coefficient ci of each floor is 490, 467, 410, 386, 375, 348, 298, 274, 243, 196 kn·s/m respectively. the ground motion acceleration lasts for 8.5 s, and the maximum acceleration is 0.1g. fig. 2 shows the switching control effect of the proposed method on the seismic surface response by comparing the maximum displacement time history under the condition of free and no switching control with the displacement time history under the switching control imposed on the top layer of the proposed method (weight function is w1= w2= w3=168). figure 2. maximum interlayer displacement response it can be seen from figure 2 that the switching control effect of the proposed method is more and more obvious with the change of time history. here, the selection of weight function is selected through comparison. for different structures, the selection of weight function is different. even for the same structure, different weight functions will lead to different switching control effects. when the weight function is not selected, it will lead to out of control. figure 3 shows the comparison of the maximum displacement time history of three different weight functions, where a is the applying switching control force on the top layer, and the weight function is w1= w2=168, w3=0; b is the applying switching control force on the top layer, and the weight function is w1= w2= w3=168; c is the simultaneously applying switching control force on the fourth, sixth, eighth and 10-th layers, and the weight function is w1= w2=1000,w3=10000. figure 3. maximum interlayer displacement response it can be seen from figure 3 that the change of w3 in the weight function has no obvious effect on the control of the maximum displacement. the effect of multi-layer simultaneous application of switching control force is more obvious than that of only the top layer. figure 4 shows the calculation results of switching control force applied to some layers and all layers. among them, c is the switching control force applied on the 4th, 6th, 8th, and 10th floor, with weight function of w1=w2=1000 and w3=10000; d is the switching control force applied on the 4th, 6th, 8th and 10th floor, with weight function of w1= w2=200 and w3=2000; e is the switching control force applied on the 10th floor, with weight function of w1= w2=200 and w3=2000. figure 4. maximum interlayer displacement response it can be seen from figure 4 that the selection of weight function has an impact on the control effect, and the control effect of switching control force applied to all layers is significant. maximum interlayer displacement response in a closed-loop state in the nominal case, that is, when there is no perturbation in the controller, the multi-model switching control method based on the local t-s model is applied to design the controller, and the reaction process is convergent 332 yudong jiang, jinming luo, hojatallah azarkhosh, erjun wu and stable; when there is a perturbation in the controller, the controller in the nominal case is still applied to realize the switching control for the reaction, and the reaction process is divergent and unstable. in both cases, the maximum interlayer displacement response in the closed-loop state is shown in figure 5. figure 5. the closed-loop maximum inter-layer displacement response of conventional switching control assuming that the controller has the same additive perturbation as in the previous case, the controller is designed by the method in this paper. the maximum interlayer displacement response of the closed-loop state is shown in figure 6, and the reaction process is convergent and stable. figure 6. in this paper, the closed-loop state maximum inter-layer displacement response of the controller is presented it can be seen from figure 5 and figure 6 that when the controller itself does not have perturbation or is disturbed by perturbation, the controller adopting the method in this paper can obtain good performance and keep stable. the simulation results show the effectiveness of the proposed method. discussions in the actual control process, the output variation range of the controller is limited (yoon et al., 2016; friederich et al., 2018), so its pole assignment is constrained when designing the controller. the multi-model switching control method proposed in this paper is used to divide the input space into several regions. considering the actual constraints, poles can be flexibly allocated in different regions (gao et al., 2016; liu et al., 2016). local controllers can be designed in each region, and then through the switching control, the approach and control of the overall nonlinear response can be realized, so that the controller can be avoided. the output saturation can also guarantee the convergence speed (cheng et al., 2016). for the reaction with a wide range of parameters, the effectiveness will be more obvious. high-order single step  method can obtain better switching control effect than other instantaneous optimization algorithms (in the past, the switching control effect of other algorithms is generally about 50% 70%) (li et al., 2016); adding damping term in the switching control index function has an impact on the switching control effect; for the value of weight function, there is the problem of optimal selection (zhan et al., 2016); there is a problem of optimal switching control layer combination in multi-level switching control. the proposed method can be applied to the design of controllers with a large range of response parameters and dynamic changes of subsystems. in addition to the application of this paper in the study of switching control of seismic surface response, it can also be applied to the air-to-air vehicle, for example. because the air-to-air vehicle has the characteristics of time-varying, uncertainty and nonlinear coupling, the parameters of system response vary greatly, and a single dynamic model has no way to accurately describe the actual changes of the system (yang & long, 2016). therefore, the method proposed in this paper can be used to approach the dynamic performance of the system, a multi-model controller is designed based on the multi-model, and multiple controllers can be selected to act on the same control object, so as to effectively control the complex system of an air space vehicle. the switching control of nonlinear response is still a problem that needs to be further studied. because the nonlinear response model is difficult to establish and control, it can use multiple linear models to approach the nonlinear model, or use multiple simple nonlinear models to approach the complex nonlinear model, and then establish the multi model controller, which will have a good control effect on some nonlinear systems. conclusions in this paper, a nonlinear response switching control method based on the local t-s model is proposed. by calculating the stress-strain relationship of viscoelastic damping materials, the stiffness matrix and switching control force vector of interlayer element, the herringbone support structure of viscoelastic damping material is applied to interlayer element, and the relative displacement of interlayer element is calculated by high-order single step  method combined with nonlinear seismic surface response analysis, which is used to represent the movement of shear structure under the seismic surface load. the controller is designed based on the local t-s model. by switching control, the nonlinear response of the whole seismic surface is controlled. in this paper, the switching control effect of the method for seismic surface response becomes more and more obvious with time. when the controller has additive perturbation, the convergence speed of the state response of the method in this paper is fast and stable, which shows the accuracy of the method in this paper for the switching control of seismic surface nonlinear response. acknowledgments this work is supported by the national nature science foundation under grant 61803271. references acerbi, e., fusco, n., julin, v., & morini, m. (2016). nonlinear stability results for the modified mullins-sekerka and the surface diffusion flow. journal of differential geometry, 113, 178-187. altukhov, v. i., sankin, a. v., sigov, a. s., sysoev, d. k., yanukyan, e. g., & filippova, s. v. (2018). schottky-barrier model nonlinear in surfacestate concentration and calculation of the i–v, characteristics of diodes based on sic and its solid solutions in the composite chargetransport model. semiconductors, 52, 348-351. bande, g. r., & choudhury, d. (2017). seismic passive earth resistance in submerged soils using modified pseudo-dynamic method with curved rupture surface. marine georesources & geotechnology, 35, 930-938. cheng, x., zhang, q., yu, x. j., du, w., liu, r., bian, q., wang, z., zhang, t., & guo, z. (2016). strike-slip fault network of the huangshi structure, sw qaidam basin: insights from surface fractures and seismic data. 333the switching control method for the nonlinear response of seismic surface based on the local t-s model journal of structural geology, 94, 1-12. ding, x. j., cheng, w., li, y., wu, j. l., li, x., cheng, q., & ding, s. (2016). an enzyme-free surface plasmon resonance biosensing strategy for detection of dna and small molecule based on nonlinear hybridization chain reaction. biosensors & bioelectronics, 87, 345-351. friederich, w., hunzinger, s., & wielandt, e. (2018). a note on the interpretation of seismic surface waves over three dimensional structures. geophysical journal international, 143, 335-339. gao, r. m., lv, t. h., & hu, j. (2016). explicit expression of limiting spectral density of high dimensional sample covariance matrices. journal of jilin university (science edition), 54, 499-505. guo, h. y., wang, l. f., ye, w. h., wu, j. f., & zhang, w. y. (2017). weakly nonlinear rayleigh-taylor instability in incompressible fluids with surface tension. chinese physics letters, 34, 60-63. huang, m., kong, l. l., cao, g. q., xiong, k., & lishen, t. (2018). analysis of short circuit failure of pouch-type lithium ion batteries. chinese journal of power sources, 42, 476-526. li, h. y., shi, p., & yao, d. y. (2016). adaptive sliding-mode control of markov jump nonlinear systems with actuator faults. ieee transactions on automatic control, 62, 1-1. liu, y., van der neut, j., arntsen, b., & wapenaar, k. (2016). combination of surface and borehole seismic data for robust target-oriented imaging. geophysical journal international, 205, 758-775. viens, l., denolle, m., miyake, h., sakai, s., & nakagawa, s. (2017). retrieving impulse response function amplitudes from the ambient seismic field. geophysical journal international, 210, 210-222. wang, c. w. (2019). the application of vr laser technology in architectural layout. automation & instrumentation, 5, 234-237. wei, y. f., wang, x. w., zheng, z., yang, m., & xiao, j. (2016). steady-state modeling and simulation for ac/dc hybrid power system based on modular multilevel converter . journal of power supply, 14, 112-120. yang, x. l., & long, z. x. (2016). seismic and static 3d stability of twostage rock slope based on hoek–brown failure criterion. canadian geotechnical journal, 53, 1-8. yoon, s. j., kim, j. w., & choi, w. y. (2016). an explicit data assimilation scheme for a nonlinear wave prediction model based on a pseudospectral method. ieee journal of oceanic engineering, 41, 112-122. zhan, c. b., luo, c., ding, z. k., & jin, x. l. (2016). research on seismic performance prediction of high-rise buildings. computer simulation, 33, 397-402. zhang, h. b., guo, q., liang, l., cao, c., & shang, z. (2017). a nonlinear method for multiparameter inversion of pre-stack seismic data based on anisotropic markov random field. geophysical prospecting, 66, 461-477. zhang, j., tang, z. h., ai, m. x., & gui, w. (2018). nonlinear modeling of the relationship between reagent dosage and flotation froth surface image by hammerstein-wiener model. minerals engineering, 120, 19-28. zhang, p., qin, z. c., & lu, z. (2016). an analytical model of space backbone network capacity for space information transmission. journal of china academy of electronics and information technology, 11, 66-72. keywords: gps; lava area; geology; exploration; remote; monitoring. palabras clave: gps; área de lava; geología; exploración; teledetección; sondeo. how to cite item xu, a. (2020). remote monitoring system of geological exploration in lava area based on gps. earth sciences research journal, 24(1), 207-214. doi: https://doi.org/10.15446/esrj. v24n2.87956 a remote monitoring system for geological exploration in a lava area based on gps is designed. the system mainly collects geological exploration data in a lava area using multi-s integration through the data acquisition module of geological exploration. the data collected is then transmitted for analysis to the data analysis and management module of geological exploration data in a lava area using a gps network. the monitoring data processing method based on the unascertained filtering method is used to remove gross errors in geological exploration, identify abnormal values, and transmit the abnormal values to the gps-based geological anomaly location module. the gps-based geological anomaly location module locates the geological anomaly according to the abnormal values, thus completing the remote monitoring of the lava area's exploration. the test results show that when the system is applied to remote monitoring of geological exploration in a lava area, the positioning error of small soil caves is low, and the reliability of monitoring data is high, which meets the needs of remote monitoring of geological exploration in a lava area. abstract remote monitoring system of geological exploration in lava area based on gps sistema de monitoreo remoto de exploración geológica en el área de lava basado en gps issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87956 en este artículo se diseña un sistema de monitoreo remoto para exploración geológica basado en gps en una área de lava. el sistema recopila principalmente datos mediante la integración multi-s a través del módulo de adquisición de datos de exploración geológica. luego, los datos recopilados se transmiten al módulo de análisis y gestión de datos de exploración geológica en el área de lava a través de una red gps. el método de procesamiento de datos de monitoreo basado en un método de filtrado no determinado se usa para eliminar errores graves en los datos de exploración geológica del área de lava, identificar valores anormales y transmitir los valores anormales al módulo de ubicación de anomalías geológicas basado en gps. el módulo de localización de anomalías geológicas basado en gps ubica la anomalía geológica de acuerdo con los valores anormales, completando así el monitoreo remoto de la exploración del área de lava. los resultados de la prueba muestran que cuando el sistema se aplica al monitoreo remoto de la exploración geológica en un área de lava, el error de posicionamiento de las pequeñas cuevas del suelo es pequeño y la confiabilidad de los datos de monitoreo es alta, lo que satisface las necesidades del monitoreo remoto de la exploración geológica en una zona de lava. resumen record manuscript received: 12/03/2019 accepted for publication: 17/03/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 207-214 aimei xu school of civil engineering and architecture, xinyu university, xinyu, china gp s m ea su re m en ts a pp li ed t o ge ol og y https://doi.org/10.15446/esrj.v24n2.87956 https://doi.org/10.15446/esrj.v24n2.87956 https://doi.org/10.15446/esrj.v24n2.87956 208 aimei xu introduction lava landform is also called karst landform. karst geomorphology refers to all kinds of landforms formed by dissolution, transportation, and sedimentation of soluble rocks (limestone) by water, such as stone buds, stone ditches, stone forests, peak forests, falling water caves, funnels, karst depressions, karst caves, underground rivers, etc. (masiokas et al., 2017). in the karst geomorphological development area, the ground is often crowded with odd peaks, the surface water system is relatively scarce, but the groundwater system is relatively developed (jian et al., 2017). guangxi, guizhou, yunnan, and other places in china widely distribute karst landforms, which is one of the most typical karst landforms in the world (zhu, 2018). engineering geological survey, geophysical prospecting, and drilling are mainly used in the geological exploration of lava area, in order to obtain the relevant information of lava landform, development degree, development law, hazard nature and geological activity law of highway section (gu et al., 2017). the first is the lava survey of the highway subgrade section. when the lava geomorphology is located on both sides of the highway subgrade, through the investigation and information collection of the direction of surface water runoff, the estimation of water volume, the development degree of surface karst groove and the scale of karst depression in a certain area on both sides of the subgrade, the development degree of lava in this area is determined, and then targeted drilling is arranged to reveal the buried depth of karst cave through drilling. the safety thickness of the lava roof can be obtained by the roof and floor thickness and filling species, which can provide corresponding excavation and filling parameters for highway subgrade construction. next is the lava survey in the bridge area (zhao et al., 2017). when the bridge crosses the lava area, through the geological survey in a certain range of the bridge site area, the data of surface water runoff direction, water flow, the development degree of surface dissolution ditches, the scale of dissolution depression, the development degree of bank slope dissolution on both sides and underground rivers are obtained. according to the characteristics, the related data such as drilling hole encounter rate, the vertical height of karst cave and depth of roof and floor are further obtained to determine the development degree of lava in the bridge site area and to provide the basis for the elevation design of bearing stratum of pile foundation (wu et al., 2018). finally, it is the tunnel lava survey. generally speaking, it is more difficult to survey tunnels in the lava area, and the safety problems in later construction and operation are more prominent. therefore, the investigation of tunnels in the lava area, especially long tunnels, is the focus of tunnel lava exploration (liu, 2017). firstly, engineering geology survey and hydrogeology survey are used to make a detailed investigation of surface water runoff direction, spring outcrop, surface dissolution depression and falling tunnel in a certain range of tunnel site area to draw engineering geology map and hydrogeology map and to preliminarily determine the relationship between surface water and groundwater, runoff model and fault structure (kugler, 2017). then geophysical exploration and drilling methods are used to further determine the relationship among the underground lava pipeline, underground river, karst cave, and the design cavern, and the connectivity and water flow of the underground river of lava pipeline are determined by means of connection test analysis. the surrounding rock of the cavern and the water inflow after excavation are accurately classified to provide reliable geological parameters for the design (wang et al., 2018). monitoring is widely used in a safe city, railway monitoring, bank monitoring, warehouse monitoring, construction site, and other occasions because of its intuitive, accurate, timely, and rich information content (khan et al., 2017). remote monitoring and control of geological exploration refer to the exploration of the geology of lava area by various means and methods, to determine appropriate bearing strata, identify geological details and the technical conditions for mining and utilization, provide the provision of geological information needed for highway construction and design, and investigate and study geological conditions such as the rocks, structures, minerals and strata in a certain area, so as to provide favorable data support for geological research in lava areas (zhao & miao, 2018). this paper designs a remote monitoring system for geological exploration in the lava area based on gps in order to contribute to construction data and exploration data in the lava area. materials and methods remote monitoring system of geological exploration in lava area based on gps the architecture of system hardware geological exploration in lava areas requires high equipment requirements. all remote monitoring equipment selection has been certified by the security department (riordan et al., 2017). the three core function modules of a remote monitoring system for geological exploration in lava area based on gps are the data acquisition module of geological exploration in lava area, the data analysis and management module of geological exploration in lava area, and the location module of geological anomaly based on gps (he & yan, 2017). the data acquisition module of geological exploration in the lava area is the monitoring end of the system. it transmits the collected geological data of the lava area to the data analysis and management module of geological exploration in the lava area using a gps network. after filtering and analyzing the acquired data, if there are abnormal values in exploration data, it transmits positioning instructions to the geology based on gps. the anomaly location module carries out positioning feedback to users (liu et al., 2017). the overall structure of the remote monitoring system for geological exploration in the lava area based on gps is shown in figure 1. (1) data acquisition module for geological exploration in the lava area the main task of the data acquisition module of geological exploration in lava area is to combine the old three pieces (geological hammer, compass, magnifier) and the new three pieces (notebook, digital camera, gps) to achieve multi-s integration and to collect formation information, rock information and structural information of lava area (geng et al., 2017). the structure of the data acquisition module for geological exploration in the lava area is shown in figure 2. (2) data analysis and management module of geological exploration in the lava area the data analysis and management module of geological exploration in the lava area is responsible for receiving, analyzing, and managing the geological data of the lava area collected by the data acquisition module of geological exploration in the lava area. after receiving the geological data collected by the data acquisition module of geological exploration in the lava area, the module filters the geological data in the lava area using the method of monitoring data processing based on unascertained filtering. according to the data, it judges whether there are abnormal values in geological exploration in the lava area. if there are abnormal values, it transmits positioning instructions figure 1. structural diagram of the remote monitoring system for geological exploration in the lava area based on gps 209remote monitoring system of geological exploration in lava area based on gps a. spatial part the space part of gps is composed of 24 gps working satellites, which together constitute the satellite constellation. among them, 21 are localizable satellites and 3 are standby microsatellites. the 24 satellites are evenly distributed in six orbital planes about 20183 km from the ground. the angle of intersection between each orbital plane and the equatorial plane is 55°, and the distance between the orbital planes is 60° (i.e. the ascending point of the orbits is 60° different from each other). there are four satellites in each orbit. each gps working satellite sends out signals for locating abnormal materials in the geology of lava areas. b. ground control part the ground control part is composed of a monitoring system consisting of several tracking stations distributed around the world. according to their different functions, these tracking stations are divided into main control stations, monitoring stations, and injection stations. c. user equipment section the user equipment is composed of gps receiver, data processing software, and corresponding user equipment. its function is to receive gps satellite signals to obtain necessary positioning information. monitoring data processing method based on unascertained filtering during the operation of the remote monitoring system for geological exploration in the lava area based on gps, there are more or fewer errors in monitoring data due to the influence of monitoring instrument accuracy and observation conditions. when analyzing the monitoring data, it will be found that there will be jumps in the monitoring data and local jumps in the timedisplacement curve drawn. on this basis, the analysis and survey of geological stability in the lava area will encounter difficulties in data screening. therefore, before analyzing the monitoring information, it is better to do filtering to minimize the systematic error, random error, and gross error in the monitoring data, so that the monitoring data can better reflect the geological conditions of lava areas. in the monitoring of geological exploration in the lava area, abnormal data processing is very important. there are generally two reasons for abnormal monitoring data: errors caused by monitoring, called gross errors; abnormal monitoring data caused by geological abnormalities, called abnormal values. in data processing, gross errors should be eliminated, abnormal values should be processed and analyzed, and effective measures should be taken to deal with the geological abnormal part of the lava area (hu et al., 2017). to the geological abnormal situation positioning module based on gps (kostopoulou et al., 2017). the structure of the data analysis and management module for geological exploration in the lava area is shown in figure 3. figure 3. structure diagram of data analysis and management module for geological exploration in the lava area (3) geological anomaly location module based on gps a geological anomaly location module based on gps consists of a space part, a ground control part, and a user part (alison et al., 2017). details are shown in figure 4. figure 2. structure diagram of geological data acquisition module in the lava area figure 4. gps based geological anomaly location module structure 210 aimei xu based on the modern error theory, this paper uses the unascertained filtering method to judge the gross errors and outliers of monitoring data, to effectively eliminate gross errors and identify abnormal values (song et al., 2017). there are essential differences between gross errors and abnormal values. the greatest difference is that gross errors are contingent, and the frequency of occurrence is not high. compared with the monitoring data of this group, gross errors are usually single and have a certain contingency. the abnormal values are continuous and accumulative after the catastrophic data are generated, which is essentially due to the geological anomalies in lava areas. in this paper, we use unknown rational numbers to distinguish gross errors from abnormal values (amiri-simkooei et al., 2017). taking the monitoring data sequence v v v vn1 2 3, , , ,… as an example, an unascertained rational number is defined. a v v f x=   ( ){ }min ,max ,1 1 (1) where, 1 1≤ ≤ ( )v n f x, , , f (x) is the distribution density function of the reliability of remote monitoring data for geological exploration in the lava area. the main purpose of processing monitoring data is to identify and screen anomalous data, eliminate the gross errors caused by various factors, and reduce their impact on data analysis. through the analysis of gross error and abnormal value of monitoring data, it is found that if the value of gross error appears in the monitoring sequence, the data must exist in isolation, and there is only one [a, b] in the remote monitoring data interval of geological exploration in lava area; if the value is abnormal value, the adjacent monitoring sequence has continuity. over time, this data appears more and more frequently. therefore, the combination of monitoring data and its reliability density function f (x) can effectively eliminate gross errors, identify outliers, and reduce its impact on monitoring data analysis. if the trend of monitoring data sequence changes too much, it will affect li number of neighborhoods of monitoring sequence, thus affecting the judgment of gross errors. in order to reduce the impact of sequence trend change on vi , n detection values of monitoring data can be divided into several data segments, each data segment has k + 1 detection value, k< n, then: when x = vi , f x i i i n( ) = ∑   (2) when x vi≠ , f x( ) = 0 (3) where, x represents the monitoring value in remote monitoring data interval [a, b] of geological exploration in the lava area, and i represents the number of (i  j) in vi field. the size of the credibility density function determines the probability of vi as the likelihood of gross error, the greater the reliability of vi is, the smaller the probability of gross error is; and the smaller the reliability of vi is, the greater the probability of gross error is. then the gross error is determined according to the value of f (x). after determining the gross error in the monitoring sequence, the gross error needs to be eliminated. in order to eliminate the gross error, the unascertained expectation of vi and its monitored values in the neighborhood is used as the final result of the monitoring sequence. it is not known that the expected e (a) calculation method is: e a l f v v i i l l k i l l k( ) = ( ) + + ∑ ∑ (4) where, l means that in the original monitoring data, the value of monitoring data starts from l, and the length of the data segment is k+1. l value varies with the location of vi in the selected monitoring data. let m be a definite even number. the selection of monitoring data is as follows: when i = 1, the first and second monitoring data are taken as the search area of i in the data sequence segment where vi is located. when formula (2) and formula (3) are used to calculate, k of confidence density function f (x) is taken as 1, and l = 1 of unascertained expectation is calculated by the formula (4). when 1 2 i m ≤ , the values of i m  2 to i m + 2 in the monitoring data are taken as the sequence search area, and then k = m, l i m =  2 . when n m i n− < 2 ≤ , the values from 2i ‒ n to n-th in the monitoring data are taken as the sequence search area, where k = 2 (n ‒ i), l l m =  2 . when i = n, the n-1 and n values in the monitoring data are taken as the search area of the sequence, and then k = 1 and l = n ‒ 1 are taken. gross error recognition depends on the reliability of monitoring data. the reliability depends on the number of monitoring data in the vi neighborhood, and the number of monitoring data depends on the size of  in the neighborhood radius. the value of  is determined according to the actual situation in the field. when the average difference between two adjacent monitoring values is large, the value of  is larger. when the average difference between two adjacent monitoring values is small, the value of  is smaller. in practical engineering, if the value of  is too large, the ability to identify gross errors will decrease; if the value of  is too small, the normal data may be misidentified as gross errors, resulting in increased workload and inaccurate calculation. therefore, the reasonableness of  value will directly affect the effectiveness of data processing by filtering method (yu & lu, 2017). for the value of , it is usually determined by the method of calculating the difference variance of the monitoring data sequence. with the help of the criterion of double or triple median error in gross error identification, the difference variance of double or triple is obtained. considering that  is bigger in the monitoring data sequence with gross errors,  is taken as doubledifference variance in the processing of remote monitoring data of geological exploration in the lava area. λ − − − = ( )+∑ 2 1 1 2 1 1 v v n i i n (5) the upper and lower limits of the neighborhood of the selected monitoring sequence are determined according to the neighborhood radius calculated by the formula. neighborhood upper limit = monitoring value + r (6) neighborhood lower limit = monitoringvalue – r (7) where  is the threshold parameter of gross error recognition. in the process of gross error identification, gross errors need to be eliminated and the remaining monitoring data need to re-calculate  until there is no gross error (li et al., 2017). because the value of k depends on the selection of m, m generally chooses all even numbers from 2 to n/2 and calculates the value of r iteratively. the value that makes  converge sufficiently and can identify gross errors is taken as the selection value of m. in concrete calculation, if m is too small and there are continuous gross errors, there will be zero gross errors, and the total reliability of the data segment is zero. at this time, e (a) cannot replace gross errors. we should select a larger data segment to obtain the unascertained expected value, or select more data to expand the neighborhood radius. for the monitoring data sequence with a small change range, m usually chooses a larger value to get a better result. if the data segment is wider, the ability of gross error recognition is stronger; for the monitoring data sequence with the large change, if m is selected, it is difficult to obtain satisfactory results. the reason is that the detection data sequence has volatility, which will result in the occlusion of gross error recognition, thus affecting the accuracy of data processing results (wang et al., 2017). 211remote monitoring system of geological exploration in lava area based on gps test settings the strata in the test area are mainly limestone of the tianziling formation in the devonian. in the detailed exploration stage of engineering geology in this area, besides ascertaining the stratum structure and geophysical and mechanical properties (gao et al., 2013; liu et al., 2014), the emphasis is on ascertaining the distribution, shape, size, scale of lava and the associated soil caves and ground subsidence in the overlying strata. this system is used to monitor the process of engineering geological exploration in this area. (1) location function test in the lava monitoring area used in the experiment, 12 small caves are set up. the system is used for remote monitoring of exploration, and the positioning function of the system is tested (hendrikx & gestel 2017). (2) reliability testing of monitoring data combined with the distribution of small soil caves in the experimental area, the monitoring data of 23 m and 3.2 m geological survey points in the lava area are extracted as the analysis objects. 23 m: a total of 22 monitoring data from may 2019 to july 2019 are selected for analysis. when the serial number of the monitoring data is greater than 3, the monitoring data is taken as the center and three values are taken forward and backward as their neighborhoods to calculate the reliability of the monitoring data in their neighborhoods (liu et al., 2018). 3.2 m: through collating and analyzing the monitoring data of this system, 22 monitoring data from august 29, 2019, to august 12, 2019, are selected for analysis. results test results of location function in this paper, the location function test results of 12 small soil caves in the geology of lava area are shown in table 1. the statistical results of the positioning errors of the system in this paper are shown in table 2. according to table 1 and table 2, we can see that the maximum error is 0.3 when the system locates 12 small soil caves, and the error is small. the reliability test results of monitoring data of the systemin this paper (1) reliability test results of monitoring data at 23m survey point the calculation process is shown in table 3, where i represents the number of monitoring data in the neighborhood. in this paper, the difference variance of r is  times. the reliability of the monitoring data of the system is tested several times by setting different monitoring dates. according to table 3, the reliability of monitoring data before may 28, 2019 is 1, and the reliability of monitoring data is high. during the period from may 28, 2019 to june 2, 2019, the reliability of monitoring data fluctuates to a certain extent, but it remains above 0.8, which can be considered that the monitoring data is basically in the normal range. from june 8, 2019 to june 23, the reliability of monitoring data is maintained above 0.8, and the reliability is basically stable. the reliability of monitoring data from june 28, 2019 is 0.81, and that of monitoring data from june 30, 2019 to july 2, 2019 is 0.83, which basically meet the requirements of the project. after july 2, 2019, the reliability of monitoring data has remained basically stable. according to the reliability analysis of the monitoring data of the system in this paper, we can see that most of the monitoring data have high reliability, which is basically maintained above 0.8. the reliability of the monitoring data in june-4, 2019 and june-28, 2019 are both 0.81, which indicates that the monitoring data in these two periods are encouraging, and the credibility is the lowest compared with other monitoring data. on the whole, the reliability of the monitoring data in this system is better, which meets the remote monitoring standard of geological exploration in lava area. (2) reliability test results of monitoring data at 3.2m survey point combined with the distribution of small soil caves in the experimental area, the monitoring data of 3.2m geological survey points in the lava area are table 1. the test results of the system positioning function are presented in this paper order number actual point coordinates the results of system monitoring in this paper x / cm y / cm h / cm x y h x y h 1 3824861.671 636956.611 286.178 3824861.681 636956.621 286.183 0.1 -0.2 -0.1 2 3824879.657 637304.244 331.949 3824879.649 637304.238 331.956 0.1 0.1 -0.1 3 3824940.057 637351.918 330.205 3824940.051 637351.912 330.214 0.1 0.1 -0.2 4 3824936.230 637412.885 335.384 3824936.225 637412.879 335.378 0.1 0.1 0.2 5 3824844.611 637524.535 324.801 3824844.618 637524.541 324.811 -0.1 -0.2 -0.2 6 3824926.865 637556.532 330.699 3824926.885 637556.542 330.685 -0.2 -0.1 0.4 7 3825026.065 637637.839 333.185 3825026.043 637637.821 333.171 0.2 0.1 0.4 8 3825145.265 636400.321 239.529 3825145.274 636400.328 239.533 -0.1 -0.1 -0.1 9 3825221.939 636475.104 248.448 3825221.945 636475.111 248.454 -0.6 -0.2 -0.1 10 3824771.613 638169.588 333.643 3824771.633 638169.598 333.625 -0.3 -0.2 0.2 11 3824874.477 637735.863 335.710 3824874.492 637735.882 335.723 -0.5 -0.1 -0.3 12 3824770.723 637906.581 336.103 3824770.729 637906.591 336.113 -0.2 -0.5 -0.2 table 2. systematic positioning errors in this paper roll call 1 2 3 4 5 6 7 8 9 10 11 12 positioning error 0.1 0.1 0.1 0.1 0.2 0.2 0.2 0.1 0.3 0.2 0.3 0.3 note: x and y are coordinate values of soil cave and h is height value. x, y and h are positioning differences. 212 aimei xu extracted as the analysis object. through collating and analyzing the monitoring data of this system, 22 monitoring data are selected from august 29, 2019 to july 2, 2019. details are shown in table 4. according to table 4, the reliability curve of monitoring data of 3.2m survey point is drawn, as shown in figure 5. among them, the abscissa represents the monitoring day, the ordinate represents the credibility data, and the maximum value is 1. during the period of august 29, 2018 to dec. 29, 2018, the credibility of monitoring data is 1, and the credibility of monitoring data is high in this period; during the period of january 1, 2019 to february 25, 2019, the credibility of monitoring data continues to decrease, to a minimum of 0.81; during the period of march 14, 2019 to may 7, 2019, the credibility of monitoring data continues to rise, to a maximum of 1; after may 7, 2019, the credibility of monitoring data remains stable, at this time, the credibility of monitoring data is high. discussions in this paper, the application of gps in remote monitoring of geological exploration in lava area is discussed. gps has the following advantages: (1) high accuracy in the past engineering surveying and mapping work in china, the relevant workers usually use traditional surveying and mapping technology, but the traditional surveying and mapping technology is gradually replaced by gps engineering surveying and mapping technology because of its shortcomings table 3. reliability calculation table of monitoring data at 23m survey point monitoring time displacement value neighborhood lower limit neighborhood upper bound i reliability 2019-5-21 1222.4 1218.0 1226.8 3 1 2019-5-25 1220.6 1216.2 1225.0 4 1 2019-5-28 1218.9 1214.5 1223.3 4 0.86 2019-5-30 1216.2 1211.8 1220.6 3 0.86 2019-6-2 1217.1 1212.7 1221.5 4 0.87 2019-6-4 1199.8 1195.4 1204.2 0 0.81 2019-6-8 1216.2 1211.8 1220.6 4 0.87 2019-6-10 1217.1 1212.7 1221.5 5 0.83 2019-6-13 1218.0 1213.6 1222.4 5 0.83 2019-6-16 1221.1 1216.7 1225.5 5 0.83 2019-6-20 1218.2 1213.8 1222.6 5 0.83 2019-6-23 1217.1 1212.7 1221.5 5 0.83 2019-6-28 1213.1 1208.7 1217.5 2 0.81 2019-6-30 1218.3 1213.9 1222.7 5 0.83 2019-7-2 1218.0 1213.6 1222.4 5 0.83 2019-7-4 1217.4 1213.0 1221.8 6 1 2019-7-6 1218.8 1214.4 1223.2 6 1 2019-7-9 1217.1 1212.7 1221.5 6 1 2019-7-13 1216.2 1211.8 1220.6 6 1 2019-7-23 1215.3 1210.9 1219.7 5 1 2019-7-26 1217.1 1212.7 1221.5 4 1 2019-7-31 1216.2 1211.8 1220.6 3 1 table 4. reliability calculation table of monitoring data at 3.2 m survey point monitoring time displacement value neighborhood lower limit neighborhood upper bound i reliability 2018-8-29 37.0 31.4 42.6 3 1 2018-9-3 35.7 30.1 41.3 4 1 2018-9-18 36.5 30.9 42.1 5 1 2018-9-21 36.4 30.8 42.0 6 1 2018-9-29 36.5 30.9 42.1 6 1 2018-10-29 36.7 31.1 42.3 6 1 2018-11-29 38.5 32.9 44.1 6 1 2018-12-15 38.4 32.8 44.0 6 1 2018-12-29 40.9 35.3 46.5 6 1 2019-1-6 40.2 34.6 45.8 5 0.83 2019-1-23 40.9 35.3 46.5 4 0.87 2019-2-25 40.1 34.5 45.7 3 0.81 2019-3-14 51.9 46.3 57.5 3 0.83 2019-4-10 52.1 46.5 57.7 4 0.83 2019-5-1 53.0 47.4 58.6 5 0.83 2019-5-7 53.3 47.7 58.9 6 0.83 2019-5-14 53.6 48.0 59.2 6 1 2019-5-21 55.8 50.2 61.4 6 1 2019-6-12 56.0 50.4 61.6 6 1 2019-7-12 56.4 50.8 62.0 5 1 2019-8-12 57.7 52.1 63.3 4 1 2019-9-17 59.3 53.7 64.9 3 1 figure 5. reliability curve of 3.2m monitoring data at survey point 213remote monitoring system of geological exploration in lava area based on gps in surveying accuracy and material consumption. for gps engineering survey technology, it has the perfect advantage of high accuracy. gps engineering surveying and mapping technology itself has the function of static surveying, and this function of static surveying can make the accuracy of surveying and mapping results to millimeter units, which is incomparable with traditional surveying and mapping technology in china. therefore, in this respect, gps engineering survey technology shows its unique advantages. (2) the measurement time is short and the efficiency is high. for gps engineering surveying technology, it not only has the characteristics of high accuracy and stronger adaptability, but also has the characteristics of short surveying time and high efficiency. in the traditional engineering surveying and mapping technology of china, due to the lack of relevant application software, it usually consumes a lot of manpower to complete the surveying and mapping work. however, gps engineering survey technology can solve this problem very well. gps engineering surveying and mapping technology is mainly the application of gps system, which can work uninterruptedly without the influence of time, geographical environment and weather, thus greatly saving working time and improving the overall surveying efficiency. (3) more adaptable gps engineering surveying technology adopts the grid method in the process of surveying, and the grid structure is very simple. the density of each point and the length of edge distance can be freely selected according to the actual situation of surveying and mapping, even if the known point is far away from the measuring point, it can also be measured. in addition, for gps engineering survey technology, this technology effectively solves the problem that points cannot be seen each other, and can select known points without elevation. for gps engineering survey technology, when surveying a large grid, gps engineering survey technology can show the advantages of high accuracy and uniform error of gps grid. therefore, for gps engineering survey technology, it has certain adaptability. conclusions at present, although china has achieved certain success in the development of geological survey and research work, the investigation of lava area is still the key and difficult problem that puzzles project researchers. how to carry out a series of economic, reasonable, technically feasible, effective and scientific exploration work and realization of remote and high-precision monitoring on the basis of fully understanding the actual geological background of local lava area and the limitation of existing resources of the project is a subject worthy of study. this paper designs a remote monitoring system for geological exploration in lava area based on gps. the system consists of three functional modules: data acquisition module for geological exploration in lava area, data analysis and management module for geological exploration in lava area, and positioning module for geological anomaly based on gps, so that the three major functions of geological exploration in lava area: data acquisition, analysis and management and anomaly positioning can be realized. the system is applied to the remote monitoring task of geological exploration in a lava area. several conclusions are drawn: (1) when the system locates 12 small soil caves, the maximum error is 0.3, and the error is small. (2) when the geological survey point in the lava area is 23m, according to the reliability of the monitoring data of this system, it can be seen that most of the monitoring data have high reliability, which basically maintains above 0.8. (3) when the geological survey point in the lava area is 3.2 m, the minimum and maximum reliability of the system monitoring is 0.81 and 1, which is high. with its unique and powerful functions and advantages, gps technology fully demonstrates its superiority in the field of development, as well as greater and broader development space. however, some shortcomings have also been exposed in the actual construction process and in the construction and monitoring of follow-up projects in this field. (1) the key of precise positioning of gps system lies in the accurate calculation of the distance between satellite and receiver. according to the fixed mode: distance = speed× time, after the time is determined, the speed is determined by the propagation speed of the electromagnetic wave. it is well known that electromagnetic waves propagate rapidly in vacuum, but the atmosphere is not a vacuum state, and the signals are subject to heavy interference from the ionosphere and troposphere. the gps system can only calculate the signal averagely and there must be some errors in some specific areas. because of the influence of high-rise buildings and trees on the signal in big cities or mountain areas, the non-linear propagation of the signal will also be caused, and some errors will be introduced in the calculation. (2) the difference of control point position in gps measurement directly affects the accuracy of observation point position. because gps measurement obtains point coordinates (including elevation) by receiving signals from satellites through data processing, any factors that may affect signal reception may cause errors when interference occurs. in the future research work, we should pay attention to the above shortcomings and optimize them. references alison, d. o., costa, e., & abdu, m. a. (2017). the variability of low-latitude ionospheric amplitude and phase scintillation detected by a triplefrequency gps receiver. radio science, 52, 439-460. amiri-simkooei, a. r., mohammadloo, t. h., & argus, d. f. (2017). multivariate analysis of gps position time series of jpl second reprocessing campaign. journal of geodesy, 91, 1-20. gao, y., jin, b., & zhang, h. (2013). wireless remote monitoring system for the agc svibration fault of rolling mill. journal of applied sciences, 13(21), 4875-4880. geng, j., jiang, p., & liu, j. (2017). integrating gps with glonass for highrate seismogeodesy. geophysical research letters, 44, 3139-3146. gu, y., yuan, l., & fan, d. (2017). seasonal crustal vertical deformation induced by environmental mass loading in mainland china derived from gps, grace and surface loading models. advances in space research, 59(1), 88-102. he, z. y., & yan, j. (2017). nlos mobile node localization algorithm based on error prior knowledge. journal of jilin university (science edition), 55, 1221-1226. hendrikx, w., & gestel, n. v. (2017). the emergence of hybrid professional roles: gps and secondary school teachers in a context of public sector reform. public management review, 19, 1105-1123. hu, j., li, z. w., & ding, x. l. (2017). derivation of 3-d coseismic surface displacement fields for the 2011 mw9.0 tohoku-oki earthquake from insar and gps measurements. geophysical journal international, 192, 573-585. jian, y., sun, f., & wang, p. (2017). vesicle distribution in acidic lava flow units in the rift basins of ne china: implications for petroleum reservoir exploration. geological journal, 52, 609-623. khan, m., yuksel, m., & winkelmaier, g. (2017). gps-free maintenance of a free-space-optical link between two autonomous mobiles. ieee transactions on mobile computing, 16, 1644-1657. kostopoulou, o., porat, t., & corrigan, d. (2017). diagnostic accuracy of gps when using an early-intervention decision support system: a highfidelity simulation. british journal of general practice, 67, e201-e208. kugler, l. (2017). why gps spoofing is a threat to companies, countries. communications of the acm, 60, 18-19. li, y., ye, y., & xiao, l. (2017). classifying community space at a historic site through cognitive mapping and gps tracking: the case of gulangyu, china. urban design international, 22, 127-149. liu, q. f. (2017). initial rotor position estimation method for interior permanent magnet synchronous motor. journal of power supply, 15,132-137. liu, x., chow, j. y., & li, s. (2018). online monitoring of local taxi travel momentum and congestion effects using projections of taxi gps-based vector fields. journal of geographical systems, 20, 253-274. liu, y., ye, s., & song, w. (2017). rapid ppp ambiguity resolution using gps+glonass observations. journal of geodesy, 91, 441-455. 214 aimei xu liu, z., meng, x., tang, y., gao, f., & wang, y. (2014). research of remote monitoring system in hemophilia information management. biotechnology: an indian journal, 10(11), 5490-5494. masiokas, m. h., delgado, s., & pitte, p. (2017). inventory and recent changes of small glaciers on the northeast margin of the southern patagonia icefield, argentina. journal of glaciology, 61, 511-523. riordan, d. o., byrne, s., & fleming, a. (2017). gps’ perspectives on prescribing for older people in primary care: a qualitative study. british journal of clinical pharmacology, 83, 1521-1531. song, t., capurso, n., & cheng, x. (2017). enhancing gps with lane-level navigation to facilitate highway driving. ieee transactions on vehicular technology, 66, 4579-4591. wang, q., zhang, y. c., & ban, y. m. (2018). research of distribution power supply fault location system based on android. chinese journal of power sources, 42, 130-131. wang, y. y., lin, h. h., & wang, y. s. (2017). what drives users’ intentions to purchase a gps navigation app: the moderating role of perceived availability of a free substitute. internet research, 28, 00-00. wu, p., steffen, h., & wang, h. (2018). optimal locations for gps measurements in north america and northern europe for constraining glacial isostatic adjustment. geophysical journal international, 181, 653-664. yu, d., & lu, h. (2017). the interactions between online shopping and personal activity travel behavior: an analysis with a gps-based activity travel diary. transportation, 44, 1-14. zhao, h. y., he, j. w., & wa, w. h. (2017). the single station geometric positioning technology of airborne infrared system for sea surface targets. journal of china academy of electronics and information technology, 12, 410-413. zhao. y. r., & miao, w. l. (2018). a mine personnel location system based on rfid technology. automation & instrumentation, 1, 191-194. zhu, x. y. (2018). simulation of target location signal detection in satellite gps geological exploration. computer simulation, 35, 445-448. palabras clave: magma; actividad térmica; calefacción magmática; movimiento de agua en canales de fractura-grieta; sistema de red; configuración jerárquica. how to cite item sun, k., tang, s., zhang, s., xi, z., & li, j. (2019). the magma thermal field and the shallow-level gas accumulation of tight gas reservoirs in the middle-eastern parts of the qinshui basin. earth sciences research journal, 23(1), 27-34. doi: https://doi.org/10.15446/esrj.v23n1.78805 in this study, the metal-non-metal mineral gas-water hydrothermal concept is used to analyze the movement channels of magmatic heating water. further, the concept of fractures-faults-cracks microfissures hierarchical configuration of the movement channels of magmatic heating water is proposed. in addition, the magma thermal field formed by magmatic heating water movement is studied and analyzed. based on the basin simulation method, which is combined with the paleo-tectonic evolution analysis and restoration of the ancient burial depth in the middle-eastern parts of the qinshui basin, the tectonic evolution history, thermal evolution history, and hydrocarbon generation and exhaustion history of tight gas reservoirs in the yushe-wuxiang block in the middle-eastern parts of the qinshui basin have been investigated. on the basis of the theories and methods that are proposed in this study, the hierarchical configuration of fractures-faults-cracks microfissures movement channels of magmatic heating water in the yushe-wuxiang block in the middle-eastern parts of the qinshui basin was studied and analyzed. it is observed that the magmatic heating water rises to the source formation through the movement channels of hierarchical configuration, heats the source rocks, accelerates the evolution of the source rock in the shallow layer, and forms a tight gas reservoir abstract keywords: magma; thermal activity; magmatic heating; water movement channel fracture-crack; network system; hierarchical configuration. the magma thermal field and the shallow-level gas accumulation of tight gas reservoirs in the middle-eastern parts of the qinshui basin el campo térmico magma y la acumulación de gas a nivel superficial de reservorios de gas en las partes del medio oriente de la cuenca de qinshui issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n1.78805 en este estudio, el concepto de mineral metal-no-metal hidrotermal gas-agua se utiliza para analizar los canales de movimiento del agua de calentamiento magmático. también se propone el concepto de configuración jerárquica de las microfisuras fracturas-fallas-grietas de los canales de movimiento del agua de calentamiento magmático. además, se estudia y analiza el campo térmico de magma formado por el movimiento de agua de calentamiento magmático. basado en el método de simulación de la cuenca, que se combina con el análisis de la evolución paleotectónica y la restauración de la profundidad de enterramiento en las partes del medio oriente de la cuenca qinshui, se ha investigado la historia de evolución tectónica, la historia de evolución térmica y la historia de generación y agotamiento de hidrocarburos en los depósitos de gases en el bloque yushe-wuxiang en las partes del medio oriente de la cuenca qinshui. sobre la base de las teorías y métodos que se proponen en este estudio, se analizó la configuración jerárquica de los canales de movimiento de microfisuras fracturas-fallas-grietas del agua de calentamiento magmática en el bloque yushe-wuxiang en las partes medio-orientales de la cuenca qinshui. se observa que el agua de calentamiento magmático sube a la formación de la fuente a través de los canales de movimiento de configuración jerárquica, calienta las rocas de origen, acelera la evolución de la roca de origen en la capa superficial y forma un depósito de gas estrecho. resumen record manuscript received: 12/12/2018 accepted for publication: 05/02/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 1 (march, 2019): 27-34 ke sun1,2,3, shuheng tang1,2,3*, songhang zhang1,2,3, zhaodong xi1,2,3, jun li4 1school of energy resource, china university of geosciences(beijing), beijing, 100083, china 2beijing key laboratory of unconventional natural gas geological evaluation and development engineering, beijing, 100083, china 3key laboratory of strategy evaluation for shale gas, ministry of land and resources, beijing, 100083, china 4college of geoscience and surveying engineering, china university of mining and technology (beijing), beijing, 100083, china. * corresponding author: tangsh@cugb.edu.cn g eo th er m a l en er g y 28 ke sun, shuheng tang, songhang zhang, zhaodong xi, jun li introduction for a long time, scholars at home and abroad generally believe that the magmatic and volcanic activities of oil-bearing basins depict a considerable influence on the generation, migration, and accumulation of oil and gas. however, it is only during the previous decade or so that some scholars have paid attention to this phenomenon. magma can invade the sedimentary formations in sedimentary basins in the form of dikes or caprocks and can heat the rocks that are located within a certain distance from the dikes (zhang, guan, & jian, 2014; li et al., 2016). if the sedimentary (argillaceous) shales are abundant in organic matter, they will accelerate the maturation of the source rock formations (rodriguez, villar, & baudino, 2007; polteau, el at., 2016; peace el at., 2017; alalade & tyson, 2013; hakimi et al., 2017). if magma invades in the form of dikes, either underneath or on both the sides of the dikes, there may be different heat conduction conditions that are caused due to the different heating effects; consequently, the source rocks will exhibit different maturity and heating ranges (zhu, jin, hu, song, & gao, 2007). the distance from the source rocks to the dikes is also an important factor that affects their maturity. due to the different heating effects, halos with different degrees of maturity can be formed around the dikes (muirhead, bowden, parnell & schofield, 2017). under special conditions, even asphalt veins that are parallel to the dikes will be formed (zanella, cobbold, ruffet, & leanza, 2015). in the atlantic research drilling project, it was observed that the intrusion of dikes significantly affects the maturity of the organic matter. magmatic intrusions and volcanism affect the thermal history of the sedimentary basins and the maturation history of the source rocks (spacapan et al., 2018). the magmatic intrusion thickness, duration of intrusion, and distance from the rock mass affects the hydrocarbon generation (aarnes, planke, trulsvik, & svensen, 2015). in addition, polyansky, reverdatto, khomenko & kuznetsovab (2009) postulated that the flow of the formation water is an auxiliary heat transfer method. several scholars have used various modeling methods to simulate magmatic intrusion and hydrocarbon generation after heating the source rocks (sydnes, fjeldskaar, løtveit, grunnaleite, & cardozo, 2018; iyer, svensen, & schmid, 2018). guangwen et al. conducted numerical thermal simulation studies on the heat dissipation process of invasive magma and classified the evolution of the source rocks into three modes: near, middle, and far (tian, jiang, pang, meng, & fu, 2005). several scholars have observed methane gas emissions while studying magmatic and volcanic activities (iyer, schmid, planke, & millett, 2017; frieling et al., 2016). evidence not only proves that magmatic and volcanic activities are conducive to generate hydrocarbons, such as light hydrocarbons and methane, but also proves that the generation of h2s in certain oil and gas fields is closely related with the magmatic intrusions (alberdi-genolet, cavallaro, hernandez, crosta, & martinez, 2013). the “magmatic thermal field” that is formed by magmatic activities not only heats the source rocks and promotes hydrocarbon generation but also exhibits a certain transformation effect on reservoirs (zhang, jin, li, & jiao, 2014; zhang et al., 2016; senger, planke, polteau, ogata, & svensen, 2014); magmatic and volcanic activities may also play a positive role in the migrations of oil and gas and the formation of hydrocarbon traps (rateau, schofield, & smith, 2013; cukur et al., 2010; xu, fang, zhang, hu, & zhu, 2006). the magmatic ascending channels of magmatic and volcanic activities can later become hydrocarbon migration channels (schofield et al., 2017) the magmatic intrusion into salt rock formations can form hydrocarbon migration channels (blažić & moreau, 2016); and a magmatic intrusion can form hydrocarbon storage spaces (witte, bonora, carbone, & oncken, 2012). the intrusion of caprocks deforms the strata and forms a trap structure that is similar to that of salt rock diapirism. after weathering the volcanic structure, it can also form a storage space that can be used as both oil and gas reservoirs (lee, kwon, yoon, kim, & yoo, 2006). additionally, magmatic activities exhibit a destructive effect on the existing reservoirs and can damage pores as well as fracture systems (salmachi, rajabi, reynolds, yarmohammadtooski, wainman, 2016; gao & liu, 2003). previous researches about the effect of magmatic activities on hydrocarbon accumulation mainly focused on the mass of a rock or the effect of the volcano itself on hydrocarbon generation in source rocks as well as hydrocarbon migration and accumulation. this study intends to use the metalnon-metal mineral gas-water hydrothermal concept to deeply analyze and study the movement of magmatic heating water that was formed in magmatic activities in the eastern qinshui basin and the effect of the formed magmatic thermal field on hydrocarbon accumulation, which can give rise to new ideas and methods concerning hydrocarbon accumulation due to magmatic activities. movement of magmatic heating water in the formation and the formed magma thermal field in ore-deposit, hydrothermal gas-water rises along a certain channel after being heated by magmatic rocks. mineralization is observed at the appropriate formational sites with surrounding rocks, which forms useful metal and nonmetallic deposits. inspired by this concept, the authors of this study believe that the hydrothermal gas-water heated by magma depicts a positive effect on hydrocarbon accumulation. the movement channel of magmatic heating water after the emplacement of magma, because the temperature of magma is as high as 800°c to 1300°c, the water in the surrounding formation is heated by magma to form magmatic heating water. the magmatic heating water flows in the strata and it not only carries the ore-forming material that can form ore deposits but also carries a large amount of heat in the strata that is located far from the magmatic emplacement location. when the water is heated, the physical properties are observed to alter. the most significant change is the increase in volume and the decrease in density; the water with less density will move through the appropriate channels to the shallow parts of the stratum under the action of buoyancy. at the macro level, various fracture systems developed in the basin include large-scale fractures and faults; at the micro level, the various cracks and microfissures that are observed to develop in the formation rocks are good magmatic heating water movement channels. additionally, all kinds of rock and debris particles will expand and become brittle when the temperature increases. it is easier to form tensile or extrusion cracks under the influence of structural movement stress and to improve the thermal water passing capacity of micro cracks and micro cracks. micro-level cracks and microfissures are extensively distributed in the strata, and their effects on thermal water movement cannot be ignored. the imbibition is a phenomenon of surface action that occurs when two media are in contact with each other. under normal circumstances, all the rocks in the formation are mostly characterized by water wetting and strong hydrophilia. therefore, magmatic heating water will enter the fracture along the walls of cracks or microfissures through the action of imbibition, thereby heating the formation and increasing the temperature of the formation. this is beneficial to improve the organic matter maturity in source rocks. in addition to fractures, faults, cracks, and microfissures of various sizes, the unconformity between formations and permeable sandstone formations are also good passageways for magmatic water heating, and play a role in communicating different fractures and faults systems. hierarchical configuration of magmatic heating water movement channel level configuration those who have studied the geological structure of the oil field have learned that the development of fractures and faults is mainly formed by the action of squeezing or tensile stress. this kind of squeezing or tensile stress is a kind of regional tectonic stress; that is, the tectonic stress in a certain area acts on the whole area; therefore, in the same area, the development of fractures and faults does not develop an isolated fracture or a fault, but a number of fractures and faults form a three-dimensional network structure at different levels and scales. the three-dimensional network structure composed of fractures and faults is a good magmatic heating water movement channel, and the magmatic heating water can be diverted from the vicinity of the magmatic emplacement position to a farther and more extensive range, thus heating source rocks in farther and wider areas. 29the magma thermal field and the shallow-level gas accumulation of tight gas reservoirs in the middle-eastern parts of the qinshui basin the distribution of fractures-faults-cracksmicrofissures in the strata is not an unordered combination, but there are certain rules to follow. large fractures and large faults develop earlier, with longer extension distances and deeper cuts through the formation. these fractures may have developed earlier than the magmation period, and may also develop in the same period as the magmation. the magma will also intrude into the shallower parts of the strata along the fractures, which is more conducive to the magmatic heating of the formation water and achieve better heating effects on source rocks. although the small faults, cracks, and microfissures are developed on a small scale, the development quantity is large and the distribution range is wide; hence, the magmatic heating water can heat source rocks across a wide range. therefore, fractures-faults-cracks-microfissures of this hierarchical configuration formed a good magmatic heating water movement network system (figure 1). in addition, the unconformity surface and the permeable sandstone formation interact between fractures and faults, allowing the magmatic heating water to move to fracture systems and formations further away from the magma. 1 magmatic heating water moves up along the deep fractures; 2 magmatic heating water enters large faults from deep fractures and moves up along the faults; 3 magmatic heating water moves up along sub-lever faults; 4 magmatic heating water enters cracks-microfissures along small faults; 5 magmatic heating water enters another fracture (fault) system from a fracture system along the sandstone permeable layer or unconformable surface magma thermal field formed by magmatic heating water magmatic heating water moves in a hierarchical configuration fracturescracks network system and is superimposed on the background of normal geothermal gradients; formations have higher temperatures in the deeper parts and lower temperatures in the shallow parts of the strata; heating water has higher temperatures in the deeper parts of the strata and the temperatures in the shallow parts of the strata are low. and the temperatures will not suddenly drop during the process of the magmatic heating water continuously moving to the shallow parts of the strata; the process of moving the magmatic heating water from the deep parts of strata to the shallow part of strata is a process by which the temperature is successively lowered. the heating effects of magmatic heating water on the formation can gradually reach the shallow parts of the strata. although the tectonic stress acts on the entire area, the development of fractures-cracks is not uniform throughout the region; in the tectonic stress accumulation area, the fractures-cracks system is well developed and the movement of magmatic heating water from the inside layers of the region to the shallow parts of the strata is not balanced; moreover, the heating effects of source rocks of the formation are also not balanced. due to the existence of the fractures-cracks network system, heating water can reach areas farther away from the magma; magmatic activities are regional activities controlled by the tectonic setting of the earth and are usually active during a geological period, such as the active period of yanshan magma in the qinshui basin, which is very favorable for the maturity of source rocks. the main sources of magmatic heating water are magma separation water, stratigraphic water, rock metamorphic water, and sea water. tectonic evolution and thermal evolution in the middle-eastern parts of the qinshui basin the characteristics of tight gas reservoirs in the middle-eastern parts of the qinshui basin the yushe-wuxiang coalfield block in the qinshui basin is located in the middle-eastern parts of the qinshui basin and has good exploration prospects for coalbed methane and tight sandstone gas. at present, 48 exploratory wells have been drilled for carboniferous-permian coalbed gas exploration, while only four exploration parameter wells have been drilled for the exploration of tight sandstone gas reservoirs. the research and analysis of the data and cores obtained from these four parameter wells are the key to a breakthrough in the exploration for tight sandstone gas reservoirs in the middle-eastern parts of the qinshui basin. the sandstone rock mass in the yushe-wuxiang block is relatively developed. the total thickness of the sand bodies such as rivers, deltas, tidal flats, and barrier islands have been deposited is 50 to 140m. tight sandstones are mainly composed of quartz, feldspar, and rock debris; cements are mainly siliceous cements, clay minerals, and carbonate cements. the pores of tight sandstone reservoirs are mainly secondary pores, and the primary pores basically disappear due to compaction and transformation; there are three types of secondary pores, namely, dissolved pores, hetero-micro pores, and microfissures. the physical properties of the tight sandstone in the yushe-wuxiang block are very poor. the porosity of the reservoir sample is 1.4%~11.9%, the average porosity is 3.74%, the lowest permeability is less than 0.02 md, the highest is 0.66 md, and the average permeability is less than 0.1md. in general, the overall layer is an ultra-low porosity and ultra-low permeability reservoir. the structural features of the middle-eastern parts of the qinshui basin the north china area is a stable craton basin; in the paleozoic, the degree of basal consolidation was high; in the early paleozoic, it was a stable neritic continental shelf-plateau facies carbonate sedimentation; in the middle ordovician-early carboniferous, it was overall uplift; and in the late paleozoic middle carboniferous-permian, it was a paralic coal-bearing deposit. in the north china areas, the mesozoic influence of the surrounding orogenic belt was in an extrusion environment, and there were magmatic activities in the deep crust. in the late mesozoic, a large mesozoic basin was formed in the north china areas. with the increase of the compressive force, a series of complex anticline and complex synclinal structures were formed in the eastern uplift of the qinshui basin, accompanied by intense magmatic activities and fractures. after the cenozoic era, the western parts of the north china areas (ordos, qinshui basin) were uplifted, which were the opposite of the rise of eastern regions and the fall of the western regions in the late mesozoic era. in the eastern margin of the yanshan qinshui basin, the taihang fault was thrust towards the east, and the eastern qinshui basin was strongly pressed. a series of thrust faults formed in the eastern part of the qinshui basin (right of figure 2). the burial and thermal evolution of middle-eastern parts of qinshui basin in this paper, we use the geochemical data of wells zk03-2 and zk091 to carry out basin simulations; based on results from comprehensive basin figure 1. diagram of fractures-faults-cracks-microfractures hierarchical configuration. 30 ke sun, shuheng tang, songhang zhang, zhaodong xi, jun li simulation, tectonic movement analysis, formation thickness, and ancient burial depth restoration, the evolution of the basin after late carboniferous can be divided into five stages, which are: i. rapid settlement stage from late carboniferous to the end late triassic; ii. the overall crustal uplifting stage of early jurassic; iii. secondary settlement stage of middle jurassic; iv. the uplift and magmatic activity stage of late jurassic-early cretaceous; v. the stage of uplift subjected to denudation after the mid-cretaceous (figure 3). different stages of basin development, with different characteristics of subsidence and different thermal evolution characteristics, play different roles in hydrocarbon generation and hydrocarbon expulsion of source rocks. the characteristics of the basin development and the geothermal evolution in the five stages are as follows: i. rapid settlement stage from late carboniferous to the end of late triassic: the maximum sedimentation rate at this stage was nearly 90m/ma, and reached the maximum depth at the end of the triassic, with a depth of about 4000m. at this stage, the yushe-wuxiang block was located in a normal paleogeothermal field. the paleogeothermal gradient was about 3°c/100m, and the formation temperature increased with the depth of the buried depth. ii. the overall crustal uplifting stage of early jurassic; in this stage, the entire crust was uplifted, the depth of burial was shallow, and the temperature was lowered. iii. secondary settlement stage of middle jurassic: the settlement rate at this stage was relatively small, around 15m/ma, and the temperature increased slowly. iv. the uplift and magmatic activity stage of late jurassic-early cretaceous: at this stage, the crustal body was uplifted, magmatic activities in the yanshan period were strongly affected, and the geothermal gradient was abnormal. the qinshui basin was in an anomalous paleogeothermal stage, and the paleogeothermal gradient was close to 6°c/100m, although the earth’s crust uplifted as a whole, but the temperature of the formation had increased significantly. in the slow uplift stage of the mid-cretaceous, the strata were lifted at a relatively low rate, the strata suffered erosion, and the temperature decreased. the paleogeothermal gradient was 3~4°c/100m, close to the normal paleo geotemperature. figure 2. the characteristics of structural location and faults of yushe-wuxiang block (the left picture is based on the study by wei shuhong et al.) figure 3. the burial evolution of the middle-eastern parts of the qinshui basin 31the magma thermal field and the shallow-level gas accumulation of tight gas reservoirs in the middle-eastern parts of the qinshui basin analysis of hydrocarbon generation and expulsion in middle-eastern parts of the qinshui basin temperature and pressure are the key factors in controlling the maturation of organic matter. the maturation of organic matter in the middle-eastern regions of the qinshui basin mainly occurs in stages i and iv (figure 4). during the period of rapid deposition from late carboniferous to the end of late triassic, the basin rapidly settled and the depth of the formation increased rapidly. as the depth of burial depth increased and the temperature increased, the organic matter became deeply metamorphic and the maturity gradually increased. during the middle stage of the triassic, organic matter has entered the mature stage. at the end of the triassic, the temperature was about 140°c and the ro% reached 0.9%±. the source rock transited from the immature stage to the mature stage and reached the first hydrocarbon generation peak. in the ii. early jurassic and iii. middle jurassic stage, the whole basin was uplifted and then slowly settled, organic matter heating temperature gradually decreased, temperature increased slowly, and maturation evolution of the organic matter of source rock was weak. in the iv. late jurassic-early cretaceous, due to the influence of magmatic events of the yanshan period, the thermal evolution of organic matter greatly improved from maturity to over-maturation, and gas began to form in large quantities. v. after the middle cretaceous, the organic matter heating temperature reduced, and there was little progress in the evolution of source rocks. the middle-eastern parts of the qinshui basin experienced two stages of hydrocarbon generation and expulsion: the first hydrocarbon generation stage was the late paleozoic to the end of the triassic, and the early strata slowly settled and the organic matter was in an immature stage. during the triassic period, the sedimentation rate was relatively fast, and the organic matter entered a deep metamorphic stage. at the end of the triassic, the maximum depth of the formation was attained, the temperature increased, and the organic matter entered the mature stage (ro was 0.7%~1.0%) and hydrocarbon production began. the first hydrocarbon expulsion occurred in the early jurassic period. at this point, the first stage of hydrocarbon generation ended, the basin was uplifted and denuded, and the pressure on the formation reduced, resulting in the desorption of oil and gas and expulsion of hydrocarbons to adjacent reservoirs. the second hydrocarbon generation stage occurred in the earlymiddle cretaceous. because of the magmatic activity in the late yanshan movement, the temperature of the coal-bearing formation increased rapidly and was higher than the temperature at the end of the triassic period, and the organic matter reached the over-mature stage (ro was greater than 1.3%) and the secondary hydrocarbon generation occurred. the secondary hydrocarbon expulsion began in the secondary main hydrocarbon generation period. a large amount of gas was generated in the hydrocarbon source rock formation. after the formation was saturated with gas, hydrocarbons were discharged to the adjacent reservoirs. through comprehensive analysis of the basin’s tectonic evolution, simulation, and inclusions, the main controlling factors, preservation conditions, hydrocarbon generation and expulsion history, and trap formation period in the qishe-wuxiang block constituted an effective match; the tectonic development history and the preservation period had formed a trap before the jurassic formation, and the preservation conditions were good. combined with the dynamic progress of hydrocarbon accumulation, it was known that the depth of the buried deep stratum at the end of the triassic strata was at its maximum, and the organic matter reached the hydrocarbon generation threshold and thus hydrocarbon generation began. however, because of the low maturation of organic matter, the amount of hydrocarbons generated was limited; also, because of the destruction of the tectonic movement, such as the uplift of the earth’s crust in the later period, the accumulation of organic matter had not been found on a large scale. magmatic activity occurred in the early cretaceous study area, which further increased the maturity of organic matter and thus the secondary hydrocarbon generation began. as all the elements of reservoir forming were available and well matched, large-scale gas reservoirs began to form. magmatic activity in qinshui basin the lithosphere has been thinning in the whole area since the mesozoic of north china platform, and the lithosphere is also significantly thinner in the qinshui basin which was characterized by the active mantle thermal in the deep figure 4. hydrocarbon generation history of well zk03-2 in the north of the yushe-wujie block. 32 ke sun, shuheng tang, songhang zhang, zhaodong xi, jun li qinshui basin during the late yanshan period as well as the upwelling of the mantle, thereby resulting in high geothermal background values. the north china platform was strongly transformed in the late mesozoic and formed extensive fold structures and thrust faults; the magmatic activity was strong and the activities after the orogenic period of the orogenic belt were evident. intense magmatic intrusion occurred on the margin of the qinshui basin, and many mesozoic intrusive rocks have been found on the margin of the qinshui basin. inside the qinshui basin, there were also magmatic intrusions in the same period. in the west, there were granite bodies in the erfengshan and tarshan between xiangfen, fushan, and yicheng. the jinzhong mantle has the qixian diorite body, the huyan granite body to the west of the xishan coalfield in taiyuan, and the zishan granite body in linxian. there were magmatic intrusions of the yanshan period in the eastern jincheng, gaoping, lingchuan pingshun, and xigou. in addition, magmatic hydrothermal dykes were found in jincheng, yangcheng, and yangquan. on the aeromagnetic map, intermittent positive magnetic aerodynamic anomalies in the east-west direction of the south end of the qinshui basin were observed. these positive magnetic anomalies may be a reflection in the magmatic rock mass. magma thermal field and shallow-level formation in middleeastern region of qinshui basin in the late jurassic-early cretaceous, magmatic activities were active during the yanshanian period. although no wells encountered magmatic rocks in the yushe-wuxiang block, the basin simulation results from the geochemical data of wells zk03-2 and zk09-1 show that the middle-eastern regions of the qinshui basin were obviously affected by magmatic activities and were in an anomalous paleo geotherm stage, with the paleogeothermal gradient approaching 6°c/100m. although the earth’s crust rose as a whole, the temperature of the earth’s stratum rose dramatically; the maximum temperature reached about 200°c, and the thermal evolution of organic matter also increased greatly from the mature stage to the overmaturation stage (ro% reaches 2.0%), and the gas began to form in large quantities. according to the analysis, the yushe-wuxiang block is located far from the taihang fault and the drilling did not penetrate the magmatic rock mass. therefore, there is no direct baking effect of the magmatic body; instead, due to the deeper and more distant magmatic heating water which reaches the source rocks’ strata through the three-dimensional network channel formed by fractures-faults-cracks microfissures arranged by stages and the consequent heating of the source rocks, the thermal evolution of organic matter reaches the over mature stage from the mature stage. under the microscope, the fluid inclusions were found in the microfissures of the quartz particles and they were all secondary two-phase fluid inclusions (fig. 5). the analysis shows that these quartz particles are brittle after being heated by magmatic heating water, causing the quartz particles to break under the action of diagenesis or multi-phase tectonic stress. the magmatic heating water carries fluids such as ch4, co2, and carbonaceous asphalt into the cracks; then, fill up, precipitation, and cementation cause the fractures to shrink, forming microfractures in the quartz grains; the trapped gas-liquid fluid forms secondary fluid inclusions in the microfractures. microscopically, it was also found that there was a small amount of carbonate cements in the detrital grain cements. these carbonate cements may be dissolved by magmatic water from the lower carboniferous-permian marine formations and be carried through the hierarchical configuration of fracturesfaults, which is also indirect evidence of the magmatic heating water movement. conclusions and recommendations in this paper, the tectonic evolution history, thermal evolution history and hydrocarbon generation and expulsion history of tight gas reservoirs was studied in the yushe-wuxiang block in the middle-eastern regions of the qinshui basin. this paper also puts forward the idea and method of magmatic heating water moving from deep to shallow strata, thereby heating the source rocks in the strata and improving the maturity of hydrocarbon source rocks and promoting hydrocarbon generation. also, applying this idea and method to study and analyze the magma thermal fields and characteristics of shallow-level oil and gas accumulation formed by the magmatic heating water in the tight gas reservoirs of the yushe-wuxiang block in the eastern parts of the qinshui basin has been discussed. based on these studies, the following conclusions are drawn: the three-dimensional network of fractures-faults-cracksmicrofissures in a hierarchical configuration is a good channel for magmatic heating water to rise from the deeper to the shallower parts of the formation, and the formed magma thermal field promotes the maturity of the source rocks. because the three-dimensional network of fractures-faults-cracksmicrofissures is unevenly distributed throughout the strata, the heating effects of magmatic heating water on the strata are not balanced. the development and evolution of the middle-eastern part of the qinshui basin after the carboniferous can be divided into five stages, namely: i. rapid settlement stage of the late carboniferous to the end of late triassic; ii. overall uplift stage of the crust of early jurassic; iii. secondary settlement stage of middle jurassic; iv. magmatic activity stage of late jurassic-early cretaceous; v. slow uplift stage of middle cretaceous; the hydrocarbon generation and expulsion mainly occurs in stages i and iv. the middle-eastern part of qinshui basin of the early cretaceous further increased the maturity of organic matter and secondary hydrocarbon generation began. the elements for hydrocarbon accumulation were available and wellmatched, large-scale gas reservoirs began to form, and magmatic activities of the yanshan period constitute the most critical factor in the formation of tight sandstone gas reservoirs in the middle-eastern parts of the qinshui basin. figure 5. the fluid inclusions in microfractures of quartz particle in well zk03-2. 33the magma thermal field and the shallow-level gas accumulation of tight gas reservoirs in the middle-eastern parts of the qinshui basin the middle-eastern parts of the qinshui basin were strongly squeezed in the late mesozoic and developed an overthrusting fracture system that cooperated with the fractures and microfissures to form the magmatic heating water movement channel of hierarchical configuration fractures-faults-cracks microfissures, heating the source rock and thereby facilitating the thermal evolution of organic matter from its mature stage to a stage of higher maturity. at present, the research on the role of magmatic activities in hydrocarbon accumulation in oil and gas accumulation basins is not deep enough and has not attracted widespread attention of scholars. however, in some basins, magmatic activities play key roles in hydrocarbon accumulation. the effective study of magmatic activities of oil and gas accumulation can be a breakthrough in the associated theory, and in the future, the study of this phenomenon can lead to new developments in oil and gas exploration. references aarnes, i., planke, s., trulsvik, m., & svensen, h. (2015). contact metamorphism and thermogenic gas generation in the vøring and møre basins, offshore norway, during the paleocene–eocene thermal maximum. journal of the geological society, 172(5), 588-598. doi: 10.1144/jgs2014-098. alalade, b., & tyson, r. v. (2013). influence of igneous intrusions on thermal maturity of late cretaceous shales in the tuma well, chad basin, ne nigeria. journal of african earth sciences, 77, 59-66. doi: 10.1016/j. jafrearsci.2012.09.006. alberdi-genolet, m., cavallaro, a., hernandez, n., crosta, d. e., & martinez, l. (2013). magmatic events and sour crude oils in the malargüe area of the neuquén basin, argentina. marine & petroleum geology, 43, 48-62. doi: 10.1016/j.marpetgeo.2012.11.005. blažić, l., & moreau, j. (2016). discovery of lower cretaceous hydrothermal vent complexes in a late rifting setting, southern north sea: insights from 3d imaging. journal of the geological society, 174(2), 233-241. doi: 10.1144/jgs2015-155. cukur, d., horozal, s., kim, d. c., lee, g. h., han, h. c., & kang, m. h. (2010). the distribution and characteristics of the igneous complexes in the northern east china sea shelf basin and their implications for hydrocarbon potential. marine geophysical researches, 31(4), 299313. doi: 10.1007/s11001-010-9112-y. frieling, j., svensen, h. h., planke, s., cramwinckel, m. j., selnes, h., & sluijs, a. (2016). thermogenic methane release as a cause for the long duration of the petm. proceedings of the national academy of sciences of the united states of america, 113(43), 12059-12064. doi: 10.1073/pnas.1603348113. gao, y., & liu, l. (2003). brief introduction of effect of magma intrusion activity on sandstone. geological science and technology information, 22(2), 13-16. doi: 10.3969/j.issn.1000-7849.2003.02.003. (in chinese) hakimi, m. h., nasher, m. a., saeed, s. a., al-hakame, h., al-moliki, t., & al-sharabi, k. q. (2017). metamorphosis of marine organic matter in the jurassic deposits from sharab area, taiz governorate of southwestern yemen: thermal effect of tertiary volcanic rocks. journal of the geological society of india, 89(3), 325-330. doi: 10.1007/s12594-017-0607-x. iyer, k., schmid, d. w., planke, s., & millett, j. (2017). modelling hydrothermal venting in volcanic sedimentary basins: impact on hydrocarbon maturation and paleoclimate. earth & planetary science letters, 467, 30-42. doi: 10.1016/j.epsl.2017.03.023. iyer, k., svensen, h., & schmid, d. w. (2018). silli 1.0: a 1-d numerical tool quantifying the thermal effects of sill intrusions. geoscientific model development, 11(1), 43-60. doi: 10.5194/gmd-11-43-2018. lee, g. h., kwon, y. i., yoon, c. s., kim, h. j., & yoo, h. s. (2006). igneous complexes in the eastern northern south yellow sea basin and their implications for hydrocarbon systems. marine & petroleum geology, 23(6):631-645. doi:10.1016/j.marpetgeo.2006.06.001. li, j., li, z., qiu, n., zou, y., yu, j., & liu, j. (2016). carboniferous-permian abnormal thermal evolution of the tarim basin and its implication for deep structure and magmatic activity. chinese journal of geophysics, 59(9), 3318-3329. doi:10.6038/cjg2016096. (in chinese) muirhead, d. k., bowden, s. a., parnell, j., & schofield, n. (2017). source rock maturation owing to igneous intrusion in rifted margin petroleum systems. journal of the geological society, 174(6), 979-987. doi: 10.1144/jgs2017-011. peace, a., mccaffrey, k., imber, j., hobbs, r., van hunen, j., & gerdes, k. (2017). quantifying the influence of sill intrusion on the thermal evolution of organic-rich sedimentary rocks in nonvolcanic passive margins: an example from odp 210-1276, offshore newfoundland, canada. basin research, 29(3), 249-265. doi: 10.1111/bre.12131. polteau, s., hendriks, b. w. h., planke, s., ganerød, m., corfu, f., faleide, j. i., . . . myklebust, r. (2016). the early cretaceous barents sea sill complex: distribution, 40 ar/ 39 ar geochronology, and implications for carbon gas formation. palaeogeography palaeoclimatology palaeoecology. 441(1):83-95. doi: 10.1016/j.palaeo.2015.07.007. polyansky, o. p., reverdatto, v. v., khomenko, a. v., & kuznetsovab e. n. (2003). modeling of fluid flow and heat transfer induced by basaltic near-surface magmatism in the lena-tunguska petroleum basin (eastern siberia, russia). journal of geochemical exploration, 78-79, 687-692. doi: 10.1016/s0375-6742(03)00079-7. rateau, r., schofield, n., & smith, m. (2013). the potential role of igneous intrusions on hydrocarbon migration, west of shetland. petroleum geoscience, 19, 259-272. doi: 10.1144/petgeo2012-035. ren, z., xiao, h., liu, l., & zhang, s. (2005). determination of mesozoic tectonic heat event in qinshui basin. petroleum exploration and development, 32(1), 43-47. doi:10.3321/j.issn:1000-0747.2005. 01.011. (in chinese) ren, z. (2000). comparison of thermal evolution history in sedimentary basins, north china. oil & gas geology, 21(1), 33-37. doi:10.3321/j. issn:0253-9985.2000.01.009. (in chinese) rodriguez, f., villar, h., & baudino, r. (2007). hydrocarbon generation, migration, and accumulation related to igneous intrusions: an atypical petroleum system from the neuquen basin of argentina. spe-107926-ms. doi: 10.2118/107926-ms. salmachi, a., rajabi, m., reynolds, p., yarmohammadtooski, z., & wainman, c. (2016). the effect of magmatic intrusions on coalbed methane reservoir characteristics: a case study from the hoskissons coalbed, gunnedah basin, australia. international journal of coal geology, 165, 278-289. doi:10.1016/j.coal.2016.08.025. schofield, n., holford, s., millett, j., brown, d., jolley, d., passey, s. r., . . . hole, m. (2017). regional magma plumbing and emplacement mechanisms of the faroe-shetland sill complex: implications for magma transport and petroleum systems within sedimentary basins. basin research, 29(1), 41-63. doi:10.1111/bre.12164. senger, k., planke, s., polteau, s., ogata, k., & svensen, h. (2014). sill emplacement and contact metamorphism in a siliciclastic reservoir on svalbard, arctic norway. norsk geologisk tidsskrift, 94, 155-169. spacapan, j. b., palma, j. o., galland, o., manceda, r., rocha, e., & odorico, a. d. (2018). thermal impact of igneous sill-complexes on organicrich formations and implications for petroleum systems: a case study in the northern neuquén basin, argentina. marine & petroleum geology, 91, 519-531. doi: 10.1016/j.marpetgeo.2018.01.018. sydnes, m., fjeldskaar, w., løtveit, i. f., grunnaleite, i., & cardozo, n. (2018). the importance of sill thickness and timing of sill emplacement on hydrocarbon maturation. marine & petroleum geology, 89(2), 500514. doi: 10.1016/j.marpetgeo.2017.10.017. tian, w., jiang, z., pang, x., meng, q., & gu, l. (2005). study of the thermal modeling of magma intrusion and its effects on the thermal evolution patterns of source rocks. journal of southwest petroleum institute, 27(1), 12-16. doi:10.3863/j.issn.1674-5086.2005.01.004. (in chinese) wang, d., & manga, m.. (2015). organic matter maturation in the contact aureole of an igneous sill as a tracer of hydrothermal convection. journal of geophysical research solid earth, 120, 4102-4112. doi: 10.1002/2015jb011877. wei, s., shen, y., & yang, x. (2017). assessment of coal measures shale gas reservoir features in yushe-wuxiang block, qinshui basin. coal geology of china, 29(8), 25-38. doi:10.3969/j.issn.16741803.2017.08.05. (in chinese) 34 ke sun, shuheng tang, songhang zhang, zhaodong xi, jun li witte, j., bonora, m., carbone, c., & oncken, o. (2012). fracture evolution in oil-producing sills of the rio grande valley, northern neuquén basin, argentina. aapg bulletin, 96(7), 1253-1277. doi:10.1306/10181110152. xu, h., fang, l., zhang, x., hu, j., & zhu, h. (2006). characteristics of early permian magmatic rocks and their impact on hydrocarbon accumulation in tarim basin. acta geoscientica sinica, 27(3), 235240. doi:10.3321/j.issn:1006-3021.2006.03.007. (in chinese) yuan, j. (1985). gitology. beijing: geological press. 72-103. (in chinese) zanella, a., cobbold, p. r., ruffet, g., & leanza, h. a. (2015). geological evidence for fluid overpressure, hydraulic fracturing and strong heating during maturation and migration of hydrocarbons in mesozoic rocks of the northern neuquén basin, mendoza province, argentina. journal of south american earth sciences, 62, 229-242. doi: 10.1016/j.jsames.2015.06.006. zhang, q., jin, w., li, c., & jiao, s. (2014). “magma thermal field” and its metallogenic significance (2). gansu geology, 23(2), 1-20. (in chinese) zhang, q., jin, w., wang, j., li, w., li, c., jiao, s., shao, g. (2016). relationship between magma-thermal field and hydrocarbon accumulation. progress in geophysics, 31(4), 1525-1541. doi:10.6038/pg20160416. (in chinese) zhang, w., guan, p., & jian, x. (2014). influence of the permian volcanicmagmatic activity in the tarim basin on the generation and storage conditions of the paleozoic. acta sedimentologica sinica, 32(1), 148158. doi:10.14027/j.cnki.cjxb.2014.01.017. (in chinese). zhu, d., jin, z., hu, w., song, y., & gao, x. (2007). effect of igneous activity on hydrocarbon source rocks in jiyang sub-basin, eastern china. journal of petroleum science & engineering, 59(3-4), 309-320. doi: 10.1016/j.petrol.2007.05.002. ernestogomezl.indd 139 (manuscript received march 2005, paper accepted october 2005) using the karhunen-loève transform to suppress ground roll in seismic data ernesto gómez londoño1, luís castillo lópez2 and thaís de souza kazmierczak2 1 ingeominas bogotá, recursos del subsuelo, e-mail: egomez@ingeominas.gov.co 2 universidade federal do rio grande do sul brazil. e-mail: luis.castillo@ufrgs.br; e-mail: thais.kazmierczak@ufrgs.br resumen en este estudio se implementó y modificó el algoritmo de sacchi (2002), basado en la transformada karhunen-loève (k-l), como una técnica para la supresión de ground roll, mostrando que aunque la amplitud del ground roll es mayor que la amplitud de la señal al usar la transformada k-l se tiene éxito en suprimirlo sin causar distorsión en las señales de reflexión, obteniendo mejores resultados frente a técnicas convencionales de remoción: f-k, paso alto y pasa banda. la transformada k-l ha sido utilizada en el campo de la detección y procesamiento de imágenes (levy and linderbaurn, 2000), en el reconocimiento de caras, iris y huellas dactilares como factores biométricos de identidad. una sección sísmica es una imagen del subsuelo por tanto esta transformada puede ser usada en el procesamiento sísmico, porque remueve partes correlacionadas espacialmente y proveyendo una imagen clara y coherente. el algoritmo se aplicó a registros generados con martillo, thumper y fuentes explosivas. se hizo un procesamiento convencional hasta el apilado, remplazando los filtros por le uso de la transformada k-l. los resultados muestran que la transformada k-l hace una mejor recuperación de la amplitud de los reflectores, eliminando refracciones que causan eventos someros irreales e incrementando la coherencia lateral de los eventos sísmicos, facilitando la interpretación geológica. palabras clave: amplitud, coherencia, filtro, onda de tierra, ruido. abstract the sacchi’s algorithm (2002) based on the karhunen-loève (k-l) transform was modified and implemented to suppress ground roll without distortion of the reflection signals, it provided better results than conventional techniques for noise removal like f-k, high-pass and band pass filters. the k-l transform is well known in other fields as image processing (levy and linderbaurn, 2000), face, iris and fingerprint identification. a seismic section is an image of subsurface where the k-l can be useful in seismic processing because spatially uncorrelated signals can be removed providing a clear and coherent image. the algorithm was applied to seismic data generated with hammer, thumper and explosive sources. conventional processing flows were used, but one replaced filters with k-l transform, providing stacked sections. the earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 139-147 140 ernesto gómez londoño, luis castillo lópez and thaís de souza kazmierczak k-l transform recovers better the reflector amplitudes when compared with others filters, also it removes refractions that cause unreal shallow events and increases the lateral coherence of seismic events showing a more interpretable geology. key words: amplitude, coherent, filter, ground roll, noise. © 2005 esrj unibiblos introduction the ground roll is (scales and snider, 1998) generated in shallow layers and is a typical rayleigh wave. it is characterized by low velocity and frequency and high amplitude, being considered a coherent noise in seismic exploration (d'agosto et al., 2002). the kl algorithm transforms the seismic data in a reasonable number of independent eigen functions that represents the most important signal characteristics. it has been considered a very advisable tool in the processing and analysis of reflection data (al-yahya, 1991). the conventional filtering techniques: f-k, high-pass and bandpass are applied in the frequency domain but in the case of ground roll amplitude stronger than reflection signals, f-k filters cause serious distortion of the signal (karsh and bayrak, 2004). high pass filter is based on the low frequency of the ground roll, with the disadvantage that also eliminates the low-frequency content of reflection signals. band pass filter depends on the amount of overlaps between the noise and the signal, therefore when the frequency bands concentrated the energy in each band without overlap, the signal/noise ratio will not change (yilmaz, 1987). the k-l transform separates the ground roll from the reflector signal, permitting to subtract it from seismic data. its application to real data showed that the ground roll suppression increased lateral coherence of seismic events in a stacked section. theoretical background the surface waves in the raw register in figure 1 travel with velocities ranging from 100 to 1000 m/s and with frequencies around 10 hertz and lower than those of reflections and refractions. note how the near surface effects distort travel time curve at the right flank of reflection "a". figure1. different waves in a raw shot gather. the karhunen-loève transform it makes a principal component analysis that is a mathematical way of determining that linear transformation of a sample of points in l-dimensional space which exhibits the properties of the sample most clearly along the coordinate axes. along the new axes, the sample variance are extremes and uncorrelated, see figure 2. using a cutoff on the spread along each axis, a sample may be reduced in its dimensionality. this way it can be used to transform independent coordinates into significant and independent ones. figure 2. k-l transformation principal eigenvector 141 using the karhunen-loève transform to suppress ground roll in seismic data k-l transform implementation the seismic traces x i (t) correspond to the rows of the named data matrix x n*m , n is the number of traces in the gather and m the number of gathers. the zero-lag covariance matrix *n nc is (jones and levy, 1987): x x* * * t n n n m m nc (1) this expression can be decomposed as .v a v* * * *n n n p n n p n tc (2) in equation 2 the columns of the matrix v n*p are the eigenvectors *n nc and an*n is a diagonal matrix with the eigen values on the diagonal. these eigen values are ordered in descending sequence along the principal diagonal of the matrix a n*n , v *n p t is the corresponding transpose matrix. the principal component of the data can be written by: m v x* * *p m p n t n m (3) if we form the matrix m’ p*m by selecting the upper m rows of the matrix m p*m , m is related with the order of the filter or principal row number, placing zeros in the remaining n-m rows, n is related with the cut of the filter, the ground roll is assumed to be represented by the matrix product between the eigenvectors matrix v n*p and the matrix m’ p*m , that include the principal components in agreement with the following expression: x v m* ' * * ' n m n p p m (4) finally, the ground roll x’ n*m is subtracted from the data matrix x n*m , obtaining the gather without ground roll x x x* * * ' n m n m n m0 (5) xo is a n*m matrix that represents the filtered seismic gather. as mentioned before x n*m is the input data matrix and x’ n*m is a matrix that represents the ground roll extracted from the input data. results the capacity of kl algorithm to suppress ground roll at gather level is evaluated using shot gathers with dispersive waves of high amplitude, obtained from shallow and dip seismic surveys using 3 types of source were used: hammer, dynamite and thumper. hammer the 102.data seismic gather (figure 3) is a record corresponding to the un-01 seismic line. note in the red square like the ground roll diminishes before (left) and after (right) kl algorithm application; the gather is much less noisy in the first 100 ms, with order of the filter: 2 and cut of the filter: 1. figure 4 shows the interactive spectral analysis for seismic record no 12 of un-01, before on top and after on bottom the application of the k-l transform. the k-l filtering gives good results because it compresses more the amplitude spectrum. additionally, the maximum amplitude of the raw power spectrum with the k-l is lower into the red circle (10000) compared with one of the same gather without this filter in the black circle (4,5 e + 07), indicating high amplitude concentrations in the spectrum; corresponding to the ground roll suppressed with k-l. the record with k-l displays seismic events with coherence and continuity. 142 ernesto gómez londoño, luis castillo lópez and thaís de souza kazmierczak figure 4. interactive spectral analysis of seismic data acquired in the un-01 seismic line before (top) and after (bottom) the application of the new kl filter. figure 3. a) seismic gather ‘102.data’ with content of ground roll of high amplitude, acquired with a symmetric spread. b) the ground roll extracted using the kl algorithm, with order of the filter: 2 and cut of the filter: 1. c) the gather after the subtraction of the ground roll. 143 using the karhunen-loève transform to suppress ground roll in seismic data explosive source the shot gather in figure 5, was generated using dynamite, it displays hyperbolic moveout, coherent noise masking the reflection signals. the analyses with f-k, high-pass and the kl filters gave the better results for the last one. the band-pass filter is used where the seismic trace contains low-frequency noise, such as ground roll and high-frequency ambient noise. the seismic reflection energy usually is confined to a bandwidth between 10 to 70 hertz, with dominant frequency of 15 hz. comparison between band-pass [15-30-55-70] (figure 6), and kl filter gave a better result for the kl, the ground roll cone has been suppressed and the lateral coherence of deeper seismic events increased. figure 5. common shot gather ‘1.segy’. comparison between a) f-k filter with the e3 test, b) high-pass filter with 30 hz and c) kl filter. note distortion in the reflection signals and noise presence with the f-k filtering; noise content with high-pass filtering; in contrast, better resolution and increasing the lateral coherence until of deeper seismic events applying the k-l transform. figure 6. common shot gather ‘2.segy’. comparison with filtering capacity between a) band-pass [15 – 30 – 55 70] filter and b) kl filter. note presence of remain ground roll with high amplitude once has been applied the band-pass filter, in contrast to the lower content of noise and increasing the seismic events lateral coherence applying the k-l transform. "a" corresponds to the ground roll cone. 144 ernesto gómez londoño, luis castillo lópez and thaís de souza kazmierczak thumper figure 7 shows an example, of the shot gather 481 with high content of ground roll. the seismic line rm-01, was acquired using a thumper source. a substantial increasing in the lateral coherence of the seismic events below the 200 ms is remarked. figure 8 shows the amplitude spectrum of the shot gather 481 before and after the kl filter application, a significant reduction of the high amplitude corresponding to ground roll is seen. figure 7. a) shot gather ‘481.segy’ containing high amplitude ground roll acquired with split-spread geometry. b) the ground roll extracted using the k-l transform, with a filter order 24 and principal rows 12. c) gather after the subtraction of the ground roll. “a” is the ground roll cone. figure 8. shot gather ‘481.segy’ amplitude spectrum before (left) and after (right) the application of the new kl filter. note in the red circle the existence of high amplitude and lower dips in the spectrum due to the great content of ground roll waves, which diminish with the k-l transform. shallow survey the 2d seismic data was acquired in october-2004 with a split spread configuration in the campus of the national university of colombia, with a geode seismic recorder with 24 channels for high resolution acquisition, the source consist in a hammer (8 kg) and geophones with vertical component. 145 using the karhunen-loève transform to suppress ground roll in seismic data the processing of the un-01 seismic line provided stacked sections shown in figure 9, on top conventional and on bottom with k-l. the section on bottom presents better results, increases lateral coherence and gives a clearer image. on the other hand, the kl suppresses unreal shallow events (indicated by the arrow), due to refractions that are observed with the conventional processing. figure 9. un-01 seismic line with conventional processing until the stacking before (top) and after (bottom) the application of the new kl filter. although a substantial amount of coherent noise has been filtered out, the general waveform character has been maintained. note that the section with the k-l displays one more a clearer and coherent image. deep survey this 2d data was acquired using a thumper in split spread configuration at southern of the middle magdalena valley, involving tertiary sediments from the gualanday group. the stacked section in rm-01 in figure 10 was obtained with conventional processing on bottom and with application of the k-l transform on top. the top image presents a great rank of amplitude (-3 to 3) compared with the conventional stacked on bottom (-2 to 2) as observed into the red circles; indicating a better amplitude recovering in reflectors with k-l use, although the processing sequence did not include fx decon. besides the kl shows coherence and it does not present the whitening effect due to the conventional processing flow (figure 10. red square, bottom). by the other hand, high dips (black arrow), are observed in conventional process image which are not true because the direction of seismic line was n-s, parallels to the beds strikes of the tertiary sediments with very smooth dips; in consequence they must be horizontals in the seismic section, as it is observed in the k-l image in agreement with the subsurface geology. conclusions the algorithm karhunen-loève is computationally economic and efficient to filter seismic data, providing coherent information of subsurface image. the interactive spectral analysis showed more concentrations of high amplitude due to the ground roll, and it verifies that k-l algorithm suppresses the ground roll waves. also it is verified than the amplitude of the ground roll is stronger than the reflection signals. by using k-l algorithm to extract the ground roll it suggests a form to suppress ground roll, without distorting the reflection signals. additionally, using k-l filter increases the lateral coherence of seismic events, in comparison with 146 ernesto gómez londoño, luis castillo lópez and thaís de souza kazmierczak the three conventional methods of suppression: filters f k, high-pass and bandpass. the algorithm was tested with raw gather containing ground roll (high amplitude) obtained from shallow and depth seismic land surveys with different source (hammer, dynamite and thumper. a lateral coherence increase is observed in stacked sections. the k-l use recovers the amplitude of the deep reflectors furnishing a resultant section more approximated to real geology. figure 10. rm-01 seismic line with conventional processing until the stacking before (bottom) and after (top) the application of the new kl filter. although a substantial amount of coherent noise has been filtered out, the general waveform character has be� conventional stacked that are suppressed by the kl. into the red square on the top the image enhancement of the seismic events with the kl is appraised. references • al-husseini, m., glover j. and barley b., 1981. dispersion patterns of the ground roll in eastern saudi arabia, geophysics, vol 46, p. 121-137. • al-yahya, k.m., 1991. application of the partial karhunen-loève transform to suppress random noise in seismic sections: geophys. prosp., 39, p. 77-93. • d’agosto, c., marfurt, k. and steven, j., 2002. modeling and removal of ground roll from horizontal component of c-waves. seg. • jones, i. f., and levy, s., 1987. signal – to – noise enhancement in multichannel seismic data processing: geophys. prosp., 35, p. 12-32. • karsh, h. and bayrak, y., 2004. using the wiener – levinson algorithm to suppress ground roll. journal of applied geophysics 55, p 187-197. • levy, a. and linderbaurn, m., 2000. sequential karhunen-loève basis extraction and its application to images, ieee transactions on image processing, vol 9, no 8, p. 1371-1374. 147 using the karhunen-loève transform to suppress ground roll in seismic data • nesvijski, e. g., 2000. on a possibility of rayleigh transformed sub-surface waves propagation. center of technology, federal university of santa maria, brasil. • sacchi, m.d., 2002. karhunen loève (kl) filtering of seismic data algorithm. signal analysis and imaging group (saig), department of physics, alberta university. • scales, j. and snider, r., 1998. ¿what is noise?: geophysics, vol 63, p. 1122-1124. • yilmaz, o., 1987. seismic data processing: society of exploration geophysicists, tulsa, p. 9-79. • zywicki, d., 1999. traditional seismic surface wave tests. georgia tech research institute (gtri), p. 11.34. levkarlincambio.indd 110 earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 110-122 simulation of near-surface layer of the colombian pacific ocean lev karlin1 and nancy villegas2 1 russian state hydrometeorological university, san petersburg, russia, e-mail: rector@rshu.ru 2 departamento de geociencias, universidad nacional de colombia, sede bogotá, e-mail:nlvillegasb@unal.edu.co resumen este artículo muestra los resultados obtenidos en el estudio de la estructura fina termohalina de la capa sub-superficial realizado en la cuenca del pacífico colombiano. se formularon modelos matemáticos para diferentes regímenes de capas sub-superficiales. basado en tres estaciones se realizó un análisis horario de información meteorológica, variación de espesor, temperatura y salinidad de la capa fina. las leyes de formación de la estructura fina termohalina para la capa sub-superficial en la cuenca del pacífico colombiano están relacionadas con convección nocturna, mezcla viento-ondas, absorción de energía radiante y precipitación. palabras clave: capa sub-superficial, estructura termohalina fina, termoclina, haloclina, mezcla vientoondas, mezcla convectiva. abstract this paper presents the results provided by the fine-scale thermohaline structure of near-surface layer (nsl) study done in the colombian pacific ocean (cpo). mathematical models for different regimes of the nearsurface layer are formulated. based on three stations, an hourly analysis of meteorological data, thickness change, temperature and salinity of this layer was done. the principles of formation of the thermohaline structure for nsl over cpo are related to night convection, wind-wave mixing, volume absorption of radiant energy and precipitation. key words: near-surface layer, fine-scale thermohaline structure, thermocline, halocline, wind-wave mixing, convective mixing. © 2005 esrj unibiblos manuscript received august 2004, paper accepted june 2005 introduction fedorov and ginsburg (1988) found for thermohaline structure formation in the ocean-nsl, the following regimes: intensive wind-wave mixing (wind speed > 8 m/s), intensive convection (night, autumn-winter, during the cooling or rise of salinity in ocean), langmuir circulation (wind speed between 3 and 10 m/s) and desalt due to precipitation. 111 simulation of near-surface layer of the colombian pacific ocean first, karlin et al. (1988) and later rodhe (1991) found another regime related to horizontal advection. finally, the regime connected to special features of the offshore thermohaline structure in polar and circumpolar regions was determined (timokhov, 1989; timokhov et.al, 1993). we described four mathematical models used for nsl regimes over cpo provided by data analysis in 3 stations. hourly weather data and vertical thermohaline profiles at these stations were examined in detail, obtaining nsl thickness values, top and bottom boundaries and the corresponding temperature and salinity (villegas, 2003). in conclusion, the predominant nsl regimes of cpo are night convection, wind-wave mixing, the volume absorption of radiant energy and regime with precipitation. methodology the former data was obtained at hydrometeorological stations 14 (4° n , 78° w), 49 (2° n, 80° w) and 111(3° n, 84° w), in august and september, 2001 (otero and pineda, 2001), recording data during 24 hours, trying to verify the models that describe the different nsl regimes, see figure 1. these stations covered the coastal (station 14), the offshore (station 111) and the mixing water zone (station 49), being selected according to the quasi-homogeneous classification zones (villegas, 2002). wind-wave mixing regime the quasi-homogeneity in temperature and salinity vertical distribution of the mixed layer in diurnal time, simplify the differential simulation without loss of accuracy, using integral methods. it is assumed the presence of mixed layer during the calculation period. the simplified expression of nsl thickness is: h g q mu2 * p t 0 3 = a (1) where, m is an empirical constant of proportionality; h is the upper mixed layer thickness, m; u c u * a p v = t t is the dynamic speed, m/s; cp is the resistance coefficient; uv is the wind speed, m/s; pa is the thermal expansion coefficient, 1/°c; qt0 is the heat flux through water-air boundary, °cm/s. the equation to calculate temperature evolution with constant heat flux is: t h q t t 0 0 =d d (2) where, t0d is the change of upper mixed layer temperature for the calculated step on the time td , °c. equation (1) determines the upper mixed layer thickness in wind-wave mixing regime, in the case of its decrease. this happens until maximum heat flux is observed, when diurnal thermocline is formed. thickness remain constant until the heat flux in airwater boundary decreases, then thickness increases and expansion of diurnal mixing layer is observed. in this case, the temperature and thickness evolution formulas of mixed layer must consider the formed diurnal thermocline thickness and heat flux qht on mixed layer bottom boundary. in accordance with the similarity theory of profile t in the thermocline (kitaygorodskiy, 1970; miropolskiy, 1970 and garnich and kitaygorodskiy, 1978), we have: t h g h t t mu t t q2 * p h t h t t 0 3 0 0 2 2 = -a a a^ ^h h (3) ; h q gh mu t t h h2 2 1* t p h t t0 2 3 0 a a a d ^ ] ] n h g g t t h q gh mu2 2 * t p 0 0 2 3 2 2 = a (4) figure 1. cpo hydrometeorological station locations. 112 lev karlin and nancy villegas where, t0 is the water temperature in upper mixed layer, °c; th is the water temperature in bottom boundary of mixed layer, °c; ta is the dimensionless coefficient of self-similarity determined according to observed data and h is the thermocline depth, m. formulas (1), (2), (3) and (4) calculate the thickness and temperature evolution of mixed layer in windwave mixing regime. convective mixing regime this type of convection, in absence of dynamic factors on water-air boundary, is also called free convection. in this work, convective-wind mixing took place at night, when heat flux on water-air boundary is negative, the mixed layer descends and heat flux through mixed layer bottom boundary becomes defined by the involvement hypothesis (karlin, 1988). the equations of thickness h ad temperature t evlution of nsl in convective mixing regime are: ; t h t t k q g h t t mu2 * h q t p h0 0 0 3 2 2 =+ -a ^ h 5) . . t t h q m g h u 12 2 * t p 0 2 3 2 2 = a 6) where, k q is dimensionless coefficient for free convection (woods and barkmann, 1986, varfolomeyev and sutyrin, 1981). convective mixing regime is characterized by layer mixed deepening even for weak wind, due to negative heat flux in water-air boundary. these two factors together (intensive refreshing and strong wind) contribute to the deepening of layer, causing the diurnal thermocline vanish. sunny and weak wind weather regime if in the ocean surface exists a convective mixed layer and the mixture of this layer is so strong that the temperature vertical gradient is zeroed, then we use heat equation in vertical direction, integrating from surface to inferior boundary of convective layer h: t t h q q q qt so h t h 0 02 2 = + -r 7) here, q so is the radiation flux through ocean-atmosphere boundary, °cm/s; q h is the radiation flux on bottom boundary of convective layer, °cm/s; qt0r is the heat flux through ocean surface caused by evaporation, turbulent heat exchange and long-wave radiation, °cm/s. equation (7) requires the expression of turbulent energy balance integrated in limits of convective layer. the production of energy (generated due to forces of archimedes) is made until z depth with the bouguer’s law (radiation flux determined exponentially) and considering dissipation according to works of zilitinkevich and dearfor (1974); dearfor et al. (1980); dearfor and willis (1985); kofi (1985) and zilitinkevich (1987), it is possible to write the energy balance turbulence equation integrated in the limits of convective layer as follows: q q h q h e r e 2 2 1 1 1 t h t s rh rh0 + + + -r ] ]g g: d 8) . q h04 t0= r where, q s is the short wave emission flux to ocean surface, °cm/s; r is the effective coefficient of weakening solar energy in sea water, m-1. in expression (8), qht is undetermined doing necessary to analyze two stages of nsl evolution, in the first one, layer thickness increases when solar energy flux decreases or decreases when heat loss increases; causing heat exchange (involvement), the previous processes make necessary to estimate qht . in the second stage, when solar energy flux increases or heat loss decreases, the mixed layer thickness decreases, i.e., a new convective mixing layer takes place with a thickness smaller than previous one. when h decreases q 0ht = then: . rh e q q e 1 02 2 1 rh s t rh 0 = + + r c ] m g 9) because rh is dimensionless, it is necessary to use the expression /h r1* = in order to calculate it. defining /hr h h h*= = u as the thickness of the mixed convective layer, so /h h r= u . from (9), hu is depending on / ;q q qt s s0 has important role because its value is maximum at noon and practically zero at night. in consequence h* decreases in sunrise to noon interval, producing a minimum convective layer thickness. after noon, mixed convective layer thickness increases and q 0ht ! , which is defined as: 113 simulation of near-surface layer of the colombian pacific ocean q t h t th t z02 2 = -^ h 10) where, t tz0 -^ h– is the difference between water temperature of convective layer and level ,z h 0= + ˚c. after noon /q qt s0 icreases and, it leads to an increase of h. this occurs until values qt0r and qs become equal, that indicates the warming up ends and usual convection begins due to surface cooling. precipitation regime nsl thickness according to experimental simulation (fedorov and ginsburg, 1988), is determined as: lnh gi s u t t* c 0 3 0 = b m 11) where, m – s the involvement coefficient, which is function of falling drops diameter; t – s the time counted from the beginning of freshened layer formation, s; t 0 – s the time interval from the beginning of rain to the beginning of the freshened layer formation, s; i – is the precipitation intensity, m/s; cb – s the salt compression coefficient, 1/ ‰; s 0 – is the top boundary salinity of mixed freshened layer, ‰. freshening value in layer generated during precipitation can be calculated by: ln s u is g t t t * c 3 0 2 0 =d m b^ h 12) these values depend on precipitation intensity, rain duration and wind speed. in katsaros and buettner (1969) work, it is shown that u * during precipitation will be equal to the sum of traditional dynamic speed u * with some u *1 : u iu 2* k 1 3 2 = 13) where, u*13 – is the dynamic speed in water due to turbulent influence of falling drops, m/s; uk – is the vertical speed of falling drops, m/s. the presence of thin freshened layer separated from great bulk of water over the mixed layer clearly can be observed due to a discontinuity layer that plays an important role in the evolution of this freshened layer after rain. when rain stops, salt flux through surface is equal to zero and thickness of freshened layer increases. then, its characteristic can be defined as follows (karlin et. al, 1988): h h g s h u s t5 * t 0 0 0 3 0= + d 14) where, h t is the thickness of the freshened layer at time t, easured since the end of the precipitation, m; h 0 is the thickness of the freshened layer at moment of end precipitation, m; s0d is the jump in salinity in the bottom of the freshened layer at the end of precipitation, ‰. this jump at time t ater end of precipitation is determined by the equation: s s h h t t 0 0=d d (15) equations (14) calculate the freshened layer thickness and (15) the evolution of the mentioned jump on its bottom boundary. results determination of regimes according to weather data and calculated fluxes on ocean-atmosphere boundary depending on change of wind, air temperature, water temperature, cloudiness and precipitation, a mathematical model for calculations is determined. inaccuracy in meteorological data can influence the behavior of used model in certain hydrometeorological station. in this section, obtained hour weather data and results of heat and salt calculated fluxes in each station, are analyzed. station 14 was observed from 5 am (8 september, 2001) to 4 am (9 september, 2001), with precipitation registered from 5 to 10 am and at 3 pm. in those hours, values of air (ta) and water temperature (tw) should be smaller than other hours, an upper thin freshening layer should appear, rain be accompanied by hard wind and cloudiness must be of 6 – 8 okt. (ginsburg et al., 1980). the figure 2 shows hourly weather characteristics and figure 3 shows results of heat and salt flux. the cloudiness data match with atmospheric aspects described below (figure 2a). during 6-7 am period, 114 lev karlin and nancy villegas wind speed was very low, and an increase in tw was observed (figure 2b). at night, convective process and a decrease in tw must occur, instead at 10 p.m., a tw increase was registered. similar behavior occurred from 1 am to 2 am (tw increased 0,2 °c), figure 2. weather data at station 14 on cpo figure 3. calculated heat and salt fluxes on water-air boundary at station 14 on cpo also at 4 am. calculated salt flux shows high values during rain (figure 3a). minimum calculated heat flux was observed at night and maximum from 10 am to 1 pm (figure 3b). a) cloudiness, okt ___ and wind speed, m/s --b) tw, °c ___ and ta, °c --a) calculated heat flux, °c cm/s b) calculated salt flux, ‰cm/s time, hours time, hours 115 simulation of near-surface layer of the colombian pacific ocean according to weather characteristics in station 14, nsl thickness and thermohaline structure will be properly described with precipitation and night convection regime models. after rain, nsl water must have less salt than before rain. at the beginning of rain nsl thickness will be small, and at the end, it will increase. due to wind, the nsl formed by precipitation will disappear immediately after rain. wind will create a new greater thickness nsl, more salinity and warmer water than previous one. night convection plays an important role in nsl thickness increase. hourly observations from 1 am (28 august, 2001) to 00 am (29 august, 2001) do not register precipitation at station 49 (figures 4 – 5) exhibiting a constant wind with relatively high speed. the nsl thickness will change depending on wind intensity: minimum thickness and warm water with low wind speed and maximum thickness and cold water with high wind speed. figure 4. weather data at station 49 on cpo figure 5. calculated heat and salt fluxes on water-air boundary at station 49 on cpo a) cloudiness, okt ___ and wind speed, m/s --b) tw, °c ___ and ta, °c --a) calculated heat flux, °ccm/s b) calculated salt flux, ‰cm/s time, hours time, hours time, hours time, hours 116 lev karlin and nancy villegas figure 4a shows very low cloudiness only from 10 am to 11 am, and the cloudiness at 8 am was 7 okt, figure 4b shows high ta and tw. ta and tw had highest values between 1 pm and 2 pm, although wind was constant and strong enough (5 m/s). since 6 pm to midnight there was no positive heat flux into the ocean, but an increase in tw at 7 pm, 10 pm and 00 am was registered. the anterior erroneous observations make hard to calculate nsl thermohaline characteristics. figure 5a shows calculated heat flux increasing due to erroneous data but at night this flux must be constant. figure 5b shows minimum salt flux associated to no rain. nsl thermohaline characteristics will be described by night convection and wind wave mixing regime models. figures 6 and 7 represent weather data and calculated heat and salt fluxes for station 111. this station observed from 00 (30 august, 2001) to 11 pm (31 august, 2001) registered three precipitations: from 6 to 7 am, from 9 am to 1 pm and from 6 pm to 9 pm. cloudiness and wind speed show values associated to rain, whereas ta and tw increased from 1 am to 5 am and from 8 pm to 11 pm, but usually ta and tw usually must decrease at this time. figure 6. weather data at station 111 on cpo according to early hourly weather observation nsl thickness must expand at 5 am and tw decrease. from 6 am, due to rain, a new thin layer should be generated and increase. after one hour, rain again is observed contributing to create a new thin layer with tw change, increasing its thickness at 1 pm due to wind. the described process will be repeated during night, appearing night convection with nsl thickness expansion and tw decreasing. taking into account this process, nsl thermohaline characteristics of station 111 can be well described by precipitation and night convection regime model. nsl thermohaline structure of cpo according to simulation results calculations of tw evolution were performed with all regimes, because temperature factor reacts immediately and more adequately to meteorological influences. calculations of salinity evolution were performed only with precipitation regime. station 14 results are represented in figures 8, 9 and 10. used regimes do not provided observed nsl thickness behavior, but shows properly tw and salinity difference evolutions. wind-wave mixing regime describes better tw evolution than a) cloudiness, okt ___ and wind speed, m/s --b) tw, °c ___ and ta, °c --time, hours time, hours 117 simulation of near-surface layer of the colombian pacific ocean figure 8. nsl characteristics evolution of cpo station 14: night convection and wind-wave mixing regimes figure 7. calculated heat and salt fluxes on water-air boundary at station 111 on cpo other ones. salinity difference evolution is well described by precipitation regime model from 5 am to 7 pm. until now it is impossible to say with confidence, that used models are not reliable to describe nsl characteristics in station 14 and verify data validness. a) calculated heat flux, °ccm/s b) calculated salt flux, ‰cm/s a) depth, cm b) tw, °c time, hours time, hours time, hours time, hours 118 lev karlin and nancy villegas figure 9. nsl characteristics evolution of cpo station 14: night convection and sunny and weak wind weather regimes figure 10. nsl characteristics evolution of cpo station 14: night convection and precipitation regimes calculated nsl characteristics at hydrometeorological station 49 are shown in figures 11-12. here, night convection regime describes well nsl depth behavior from 1 am to 5 am and from 7 pm to 12 pm. also there are good results with wind-wave mixing regime for the other hours. theoretically, when there is a constant wind, nsl depth should increase (in this case so it was calculated from 9 am to 2 pm), but according to observed data it decreased, allowing to conclude that oceanographic data or wind data are erroneous (wind should decrease). for this period, sunny and weak wind weather regime, gave a very sharp decrease of nsl depth, although this change is similar to natural data. here we also can assume not correctly registered wind data. a) depth, cm b) tw, °c a) depth, cm b) salinity, ‰ time, hours time, hours time, hours time, hours 119 simulation of near-surface layer of the colombian pacific ocean precipitation regime at this station did not adapt, since rain was not observed. nsl tw at station 49 is described well with examined regimes. sunny and weak wind weather regime shows abrupt changes in tw in day time, but preserves similarity with natural data. in general, calculations with different regimes at this station show better results, than at station 14. figure 11. nsl characteristics evolution of cpo station 49: night convection and wind-wave mixing regimes figure 12. nsl characteristics evolution of cpo station 49: night convection and sunny and weak wind weather regime figures 13, 14 and 15 show results of calculations at hydrometeorological station 111. here, evolution of nsl depth, tw and salinity difference are well described with all regimes. a) depth, cm b) tw, °c a) depth, cm b) tw, °c time, hours time, hours time, hours time, hours 120 lev karlin and nancy villegas figure 13. nsl characteristics evolution of cpo station 111: night convection and wind-wave mixing regimes figure 14. nsl characteristics evolution of cpo station 111: night convection and sunny and weak wind weather regimes simulation results of sunny and weak wind weather regimes, have a high differences from others regimes and preserve natural data behavior. calculated nsl depth with night convection regime from 0 to 5 am is more similar to natural data than those obtained from 7 pm to 11pm. the precipitation regime provided the best performance from 6 pm. in general, the sunny and weak wind weather regime was more suitable due to presence of precipitation with wind. salinity difference evolution was well enough described by precipitation regime. these models can be used for researches of nsl thermohaline characteristics of cpo, but it is necessary quality improvement of natural data. a) depth, cm b) tw, °c a) depth, cm b) tw, °c time, hours time, hours time, hours time, hours 121 simulation of near-surface layer of the colombian pacific ocean figure 15. nsl characteristics evolution of cpo station 111: night convection and precipitation regimes conclusions nsl thermohaline structure of cpo according to simulation results with different regimes was described. on basis of data analysis and simulation results, over cpo occur four nsl regimes: night convection, wind-wave mixing, volume absorption of radiant energy and precipitation. models that describe these regimes can be used for nsl thermohaline characteristics researches of cpo, but it is necessary to improve the quality of natural data for further experimental calculations. mixed layer thickness with night convection regime is deepened (in average) in 15 m, with tw change from 27,3 to 26.9°c. with wind-wave mixing regime mixed layer thickness changes in 10 m, with a not significant tw change. layer thickness of fresh water changes between 1 m and 5 m with a tw change of 0,2 °c and 0,3‰ of salinity with precipitation regime. during and after rain, thickness changes from 5 to 9 m with gradual disappearance of diurnal halocline. references • dearfor, j.w. and g.e. willis. further results from a laboratory studies of convective planetary boundary layer//boundary layer meteorol.– 1985.– v.32.– p. 205–236. • dearfor, j.w., g.e. willis, and strockton, p. laboratory studies of the entrainment zone of a convectively mixed layer//j. fluid mech.– 1980. v.100.– p. 41–64. • fedorov k. and a. ginsburg. near-surface layer of ocean.-l.:gidrometeoizdat, 1988.-304 p. • garnich k.g., and s.a. kitaygorodskiy. to the theory of sinking vks in the ocean as a result of processes of wind-induced turbulence// by izv. as ussr. serie. fao. 1978. t.14, no.10. – p.1062-1073 • ginsburg a. et. al. effects of precipitation in the nearsurface layer of the ocean/a.ginsburg, a. zatsepin, v. sklyarov, k. fedorov//oceanology-1980.t.20, no.5. p. 828–836. • karlin l.n., kluykov y.y. and v.p. kutko. small-scale structure of hydrophysical fields on the upper layer of ocean. m.: gidrometeoizdat, 1988. 164 p. • karlin, l.n. the processes of formation and interaction of the different-scale vertical thermohaline structure of the upper layer of ocean and their numerical simulation. thesis to the competition of the scientific degree of dr. of phys. and math. l.: lgmi. 1988. 289 p. • katsaros, k. and k.j. buettner. influence of rainfall on temperature and salinity of the ocean surface//applied meteorology. 1969. v.8, no.1. p. 15–18. • kitaygorodskiy s.a. physics of interaction of the atmosphere and ocean.l.: gidrometeoizdat, 1970. 285 p. • kofi s.d. experimental data of the atmosphere boundary layer//atmospheric turbulence and the simulation of the propagation of admixtures.l.: gidrometeoizdat, 1985. p. 126–172. • miropolskiy y.z. the nonstationary model of the layer of convectivewind mixing in ocean //izv. as ussr. serie. fao. 1970. t.6, no.12. p. 1284–1294. a) depth, cm b) salinity, ‰ time, hours time, hours 122 lev karlin and nancy villegas • otero l., y pineda r. cruceros oceanográficos pacífico xxxiv erfen xxxii y pacífico xxxv erfen xxxiii 2001: reporte final.– tumaco.: cccp, 2001.– 230 p. • rodhe, j. wind mixing in a turbulent surface layer in the presence of a horizontal density gradient//j. phys. oceanogr. – 1991. – v.21, no.7. – p. 1080–1083. • timokhov, l., golovin p. and kochetov, s. the special feature of the thermohaline structure of openings in summer in artic ices // oceanology. 1993. t.33, no.6. – p. 833-838. • timokhov, l.a. vertical structure and the dynamics of the under-ice layer of ocean. gidrometeoizdat, 1989. 141 p. • varfolomeyev a.a. and g.g. sutyrin. laboratory simulation of the free nonstationary penetrating convection// reports to the as ussr. 1981. t.261, no.1. p. 55-59. • villegas, n.l. definition of uniform zones in the surface water of the columbian pacific ocean// materials of the total session of academic council on 30 31 january 2002. st. petersburg: publ. rshu, 2002. – p. 126128. • villegas n.l. laws governing the formation of the thermodynamic structure of the columbian pacific ocean. thesis to the competition of the scientific degree of dr. of phys. and math..: rshu. 2003. 189 p. • woods, j.d. and w. barkmann. the response of the upper ocean to solar heating. p.i: the mixed layer//quart. j. roy. met. soc.– 1986.– v. 112, no.471.– p. 1–27. • zilitinkevich s.s. theoretical model of the penetrating turbulent convection// by izv. as ussr. serie. fao. 1987. t.23, no.6. p. 593–610. • zilitinkevich, s.s. and j.w. dearfor. similarity theory for the planetary boundary layer of time-dependent height// j. atmos. sci.– 1974.– v.31, no.5.– p.1449–1452. seismic site classification of the costa rican strong-motion network based on vs30 measurements and site fundamental period luis a. pinzón1, diego a. hidalgo leiva2, aaron moya-fernández2, victor schmidt-díaz2, luis g. pujades3. 1 research division, universidad católica santa maría la antigua, panama city, panama 2 earthquake engineering laboratory, universidad de costa rica, san jose, costa rica 3 department of civil and environmental engineering, universitat politècnica de catalunya, barcelona, spain keywords: site classification; site fundamental period; shear-wave velocity; costa rica; hvsr; vs30 palabras clave: clasificación de sitio; período fundamental de sitio; velocidad promedio de la onda de corte; costa rica; relaciones horizontales/ verticales; vs30 issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n4.93927 earth sciences research journal earth sci. res. j. vol. 25, no. 4 (december, 2021): 383-389 se is m o l o g y record manuscript received: 26/02/2021 accepted for publication: 01/10/2021 abstract in this paper, the soil profile of the costa rican strong-motion network (crsmn) stations is classified based on actual measurements and seismic regulations. the soil classification of the costa rican seismic code based on the average shear-wave velocity of the top 30 m (vs30) is used as a reference. the site fundamental period (tf) is included as a parameter to complement the existing characterization. for this, vs30 measurements from 52 accelerometric stations are related to the site fundamental period obtained through horizontal-to-vertical spectral ratios (hvsr) using ground motion records from the costa rican strong-motion database. the h/v ratios are estimated with 5% damped acceleration response spectra and with traditional fourier amplitude spectra from the s-wave window. from the relation between vs30 and tf, different ranges of tf are assigned to the existing soil profile classification and a graph with three-lines and four-areas is proposed to classify the stations of the crsmn. these lines are plotted from the intersection between values assigned to each site class. the assigned classification at each station will be the one corresponding to the area where the pair of values (vs30 – tf) falls. with this proposal, both parameters take relevance and are compensated, reducing the differences due to possible errors in measurements or interpretations. clasificación de sitio de la red de acelerógrafos de costa rica basada en mediciones de vs30 y en el período fundamental resumen en este artículo, se propone una nueva clasificación de sitio para la red de acelerógrafos de costa rica (racr). se utiliza como referencia la clasificación de sitio del código sísmico de costa rica basada en la velocidad promedio de la onda de corte de los 30 m superiores (vs30). el período fundamental del sitio (tf) se incluye como parámetro para complementar la caracterización existente. para ello, las mediciones de vs30 de 52 estaciones acelerométricas se relacionan con el período fundamental del sitio, este obtenido a partir de relaciones espectrales h/v utilizando registros de la base de datos de movimiento fuerte de costa rica. las relaciones h/v se estiman con espectros de respuesta de aceleración con 5% de amortiguamiento y con espectros de amplitud de fourier tradicionales obtenidos de la ventana de ondas s. a partir de la relación entre vs30 y tf, fueron asignados diferentes rangos de tf a la clasificación de perfil de suelo existente y se propone un gráfico con tres líneas y cuatro áreas para clasificar las estaciones de la racr. estas líneas fueron graficadas a partir de la intersección de los valores asignados a cada clase de sitio. la clasificación de cada estación será la correspondiente a la zona donde se encuentre el conjunto de valores vs30 y tf. con esta propuesta, ambos parámetros toman relevancia y se compensan, reduciendo las diferencias por posibles errores en las mediciones o interpretaciones. how to cite item: pinzon, l. a., hidalgo-leiva, d. a, moya-fernandez, a., schmidt-diaz, v., & pujades, l. g. (2021). seismic site classification of the costa rican strongmotion network based on v s30 measurements and site fundamental period. earth sciences research journal, 25(4), 383-389. https://doi.org/10.15446/esrj. v25n4.93927 https://doi.org/10.15446/esrj.v25n1.74167 https://doi.org/10.15446/esrj.v25n4.93927 https://doi.org/10.15446/esrj.v25n4.93927 384 luis a. pinzón, diego a. hidalgo leiva, aaron moya-fernández, victor schmidt-díaz, luis g. pujades. introduction costa rica is considered one of the countries with the greater seismicity of the region. its complex tectonic framework, subduction zone and active volcanism in the continental zone, has motivated to develop more studies to reduce the seismic risk of the country (protti and mcnally 1994; quintero and güendel 2000; moya-fernández et al. 2020). in seismic risk studies, three features must be taken into account: the hazard, vulnerability and damage value. a correct definition of the seismic hazard is essential to obtain reliable results. seismic sources, site characterization and ground motion prediction models must be defined (douglas 2017; pinzón et al. 2019a). several parameters have been used as proxy to define the seismic site conditions based on the topography, geology, site fundamental period (or frequency) and shear-wave velocity of the soil profile. among these, the most used is the average shear-wave velocity in the top 30 m of the soil profile (vs30). in united states, the nehrp (bssc 2003) and the asce 7-16 (asce 2017) proposed six site classes based on vs30, from a (hard rock, vs30 > 1500 m/s) to f (soil with special condition). the costa rican seismic code (crsc) (cfia 2016) uses an equivalent classification, based on vs30, varying only nomenclature (s1 to s4) (dobry et al. 2000). the vs30 is estimated using the following expression: (1) where di is the thickness of each soil layer in the profile until it reaches 30 m deep, vsi is the shear-wave velocity of each layer in m/s and n is the number of layers until it reaches the 30 m. recent studies show some limitations and problems of using vs30 to classify sites (steidl 2000; di alessandro et al. 2012; pinzón et al. 2019a). some of these limitations are the limited disposal of near-surface shear-wave data at strong-motion sites, the complexity of data acquisition, and the fact that these classifications do not consider the effect of the thickness of soft sediments. as an alternative, classifications based on the site fundamental period (tf) have been used (zhao et al. 2006; ghasemi et al. 2009; di alessandro et al. 2012; pinzón et al. 2019a). the site fundamental period is a parameter that characterizes the dynamics conditions of the soil and has been used as a proxy for soil amplification in ground-motion prediction models (di alessandro et al. 2012). there are different methods to estimate the fundamental period, but the most used are those based on the horizontal-to-vertical spectral ratios (hvsr). one is the conventional hvsr with the fourier amplitude spectra from the s-wave window. this approach has been used to determine different dynamic characteristics of the ground through micrometer measurements (nakamura 1989; alfaro et al. 2001; caselles et al. 2010) and strong-motions from earthquakes (nagashima et al. 2014). the ratio is defined by the following expression: (2) where hf and vf are the horizontal and vertical fourier amplitude spectra of ground-motion determined with the s-wave window, respectively. another of the most recently used technique is the one based on the proposal of zhao et al. (2006). with this approach, a site classification for japan was determined from all the seismic records available on each station (zhao et al. 2006). the method is based on the estimation of the spectral ratio h/v from the acceleration response spectra with a 5% of damping for the horizontal components (sah) and the vertical (sav) (see equation 3). (3) zhao et al. (2006) found that the use of averages among all the hvsr obtained at each station eliminates extreme peaks or anomalies. besides, they observed that these averages are not strongly affected if data is segregated by hypocentral distance, magnitude or depth. in costa rica, the strong-motion network (crsmn) operates since 1983 and is administrated by the earthquake engineering laboratory at the university of costa rica (lis-ucr for its acronym in spanish). the lisucr is in charge of recording, processing and storing all acceleration records for academic and research purposes. the network has more than 150 active accelerometric stations around the country. further details about the network can be found in moya-fernández et al. (2020). vs30 measurements on each station is essential in order to classify the site conditions of this network based on existing regulations. the main issue is that only a third of the stations have this kind of measures. the lack of data due to the high cost of measurements and other issues, introduces uncertainty in the estimation of site conditions that directly affect the seismic hazard studies carried out in the country. in recent studies, several relations between the site fundamental period, the topography and the vs30 were found (wald and allen 2007; cadet et al. 2011; pinzón et al. 2019b). these relations cannot be used to classify costa rican stations since they depend on the geomorphology of the region. the main objective of this study is to classify the sites of the accelerometric stations of the crsmn, considering the current seismic regulations and data limitation. to do this, a relation between the site fundamental period and the vs30 is defined to be able to classify stations that do not have vs30 measurements. average shear-wave velocity of the top 30 m (vs30) in the last two years, vs30 have been measured in 52 accelerometric stations (33% of the network) with the aim of improving the existing classification. to obtain vs30 values, the stations of the crsmn were scanned using the multichannel analysis of surface waves (masw) geophysical method. the masw is one of the most used seismic recognition methods where the elastic condition (stiffness) of the soil is evaluated. in this method, the velocity of the surface-waves is measured at different frequencies, obtaining the depth variation of the shear-wave velocity (vs) from the measured soil profile. the variation of velocity at different frequencies is mainly attributed to the stratification of the s-waves velocities whose values are obtained with a numerical inversion system. in figure 1, the shear-wave velocity profile from the accelerometric stations puntarenas-cóbano (pcob) and puntarenas-paquera (ppqr) are presented. these stations present vs30 values of 407.5 and 214.1 m/s respectively. figure 1. shear-wave velocity profiles from stations (a) pcob and (b) ppqr. 385seismic site classification of the costa rican strong-motion network based on vs30 measurements and site fundamental period the fundamental period through horizontal-to-vertical spectral ratios (hvsr) because almost 70% of the stations do not have measured values of vs30, it was decided to determine the fundamental period of each site (station) through horizontal-to-vertical spectral ratios (hvsr) and fill the gap produced by the lack of information. to estimate site fundamental periods, the average of the hvsr obtained from all the records available at each station is calculated with hvsrsa(5%) and hvsrfourier. then, the average of both fundamental periods (from both methods) is estimated, obtaining a unique value per station. records with high acceleration values were discarded (pga > 0.30 g), preventing fundamental period values from high deformations that could lead to a nonlinear response of the soil (pinzón et al. 2019a). in figure 2, the results obtained from the hvsr of the accelerometric stations pcob and ppqr are exposed. the hvsr are presented normalized by the maximum value. site fundamental periods of 0.25 and 0.63 s were obtained in pcob and ppqr stations respectively. hvsrfourier presents several peaks while the hvsrsa(5%) has a smoother shape. in both cases, the fundamental period was properly captured. this method was applied to the entire network, obtaining the site fundamental period at the 157 stations. a comparison of the results obtained with the hvsrsa(5%) and hvsrfourier for all the stations is shown in figure 3. figure 2. horizontal-to-vertical spectral ratios normalized by the maximum value and the site fundamental period from stations (a) pcob and (b) ppqr. figure 3. comparison between the site fundamental periods obtained with hvsrsa(5%) and hvsrfourier. vs30 and site fundamental period relation the vs30 values measured were related to their site fundamental period. results from the relation are observed in figure 4. as seen in the figure, the relation follows a bilinear trend with a constant value of vs30 (260 m/s) in sites with a fundamental period higher than 0.5 s. similar results were obtained in the united states from central and eastern north america (cena) sites (hassani and atkinson 2016). figure 4b shows the relation in linear scale. it is observed that the models fit the measured data reasonably well. the dispersion obtained is similar to that obtained in other studies (see figure 9 from hassani and atkinson, 2016, and figure 8 from ghofrani and atkinson, 2014). the bilinear model is defined by the following equations: (4) (5) as seen in the results, some sites with a null period have vs30 values that are not indicative of a rock outcrop. it is important to note that, in those cases, a tf = 0 s was assigned since the fundamental period was not identifiable, presenting a flat h/v with amplitudes < 2. for fundamental periods of the soil deposit larger than 0.5 s, the graph shows that sites with similar vs30 can have significantly different tf values. this outcome, observed also in other studies (hassani and atkinson, 2016), is quite significant. this behavior points a significant drawback of the vs30 approach for seismic site classification. for instance, in those cases, the considered 30 m might not be sufficient to characterize the vibration period of the soil deposit and, therefore, vs30 might not provide a reasonable proxy of the seismic response of the site. on table 1, we have made a proposal relation between the vs30 used on the crsc and the soil fundamental period. sites with fundamental periods lower than 0.15 s are considered rock sites (s1), between 0.15 and 0.35 s stiff soils (s2), 0.35 and 0.75 s soft soils (s3) and sites with fundamental period higher than 0.75 s are very soft soils (s4). to assign this classification, only equation 4 was used (see the dashed line of fig. 4). equation 5 was neglected since sites with very soft soils (s4) could not be classified in terms of vs30. although, in terms of safety (design), this proposal can be beneficial since sites with periods larger than 0.50 s will result in a lower classification compared to the vs30 approach, it is important to note that the use of equation 4 for periods larger than 0.50 s is arbitrary and not supported with data. probably, as mentioned before, the considered 30 m might not be sufficient to identify the fundamental period of the soil deposit. nevertheless, as can be seen in zhao et al. (2006) and di alessandro et al. (2012), a period of 0.60 s has been used as a 386 luis a. pinzón, diego a. hidalgo leiva, aaron moya-fernández, victor schmidt-díaz, luis g. pujades. figure 4. measured vs30 values at the station of the costa rican strong-motion network plotted against site fundamental period values obtained from hvsr. (a) the bilinear model (continuous line – equation 4 for periods < 0.5 s and equation 5 for periods ≥ 0.5 s) and the proposed model based on the costa rican seismic code (dashed line – equation 4) are depicted. (b) plot in linear scale. limit to classify soils s4 (equivalent to nehrp soil e), so our proposal at 0.75 s, is not very different from those already proposed. a graphic with three-lines and four-areas is proposed to classify stations with both vs30 and site fundamental period values (see fig. 5). each area (triangle) represents a site class based on the crsc. the lines are plotted from the intersection between values assigned to each site class (see table 1). the assigned classification at each station will be the one corresponding to the area where the pair of values (vs30 – tf) falls. although this definition is arbitrary, we believe is a reasonable limit according to the available information we have presented in this paper. with this proposal, both parameters take relevance and are compensated, reducing the differences due to possible errors in measurements or interpretations. in this way, we can guarantee that sites will have similar geodynamic characteristics, and consequently possible error sources in the use of accelerograms are reduced (e.g., in the development of ground motion prediction equations or the dynamic analysis of structures). using this classification, stations pcob and ppqr are classified as s2 and s3 respectively. figure 5. site classification proposal based in vs30 and site fundamental period values. crsmn site characterization from the 157 stations of the network, 52 were classified with the proposal described above and the other 105 were categorized only with their site fundamental period using table 1 ranges. in figure 6, a map of costa rica with the geographic distribution of the accelerometric stations and their assigned site class is shown. a total of 12 stations were classified as rock sites, 52 as stiff soils, 67 as soft soils and 26 as very soft soil (see fig. 7). a complete list with the assigned site class of each station is available at the lis-ucr website (http:// www.crsmd.lis.ucr.ac.cr/?id=estaciones). in figure 7b, the mean value of the hvsr per site class, estimated from ratios of all the stations is exposed. the shapes of the mean hvsr values correspond to the expected. the rocky sites have a flat shape with amplification table 1. site classification of the costa rican seismic code, nehrp equivalence, vs30 ranges and site fundamental periods proposed. site class based on costa rican seismic code (crsc) (cfia 2016) nehrp (bssc 2003) equivalence shear-wave velocity of top 30 m site fundamental period s1 (rock) a+b vs30 ≥ 750 m/s tf < 0.15 s s2 (stiff soil) c 360 < vs30 ≤ 750 m/s 0.15 ≤ tf < 0.35 s s3 (soft soil) d 180 < vs30 ≤ 360 m/s 0.35 ≤ tf < 0.75 s s4 (very soft soil) e vs30 ≤ 180 m/s tf ≥ 0.75 s http://www.crsmd.lis.ucr.ac.cr/?id=estaciones http://www.crsmd.lis.ucr.ac.cr/?id=estaciones 387seismic site classification of the costa rican strong-motion network based on vs30 measurements and site fundamental period −86˚ −85˚ −84˚ −83˚ 8˚ 9˚ 10˚ 11˚ 75 km nicaragua p a n a m a soil type s1 (rock) s2 (hard soil) s3 (soft soil) s4 (very soft soil) 10 km figure 6. map with the geographical distribution of the accelerometric stations with the assigned soil class. 388 luis a. pinzón, diego a. hidalgo leiva, aaron moya-fernández, victor schmidt-díaz, luis g. pujades. values less than two, while sites s2, s3 and s4 have amplification values around three in the range of periods assigned on each site class (see table 1). the h/v spectral ratio for s4 soils has a peak zone from 0.9 to 1.5 seconds that is reasonably consistent with the definition of this site class being dominant periods larger than 0.75 seconds. figure 7. (a) number of stations per site class and (b) mean hvsr from all the stations per site class. conclusions in this study, a site classification has been assigned to the stations of the costa rican strong-motion network. a total of 157 stations were classified based on vs30 measurements and the site fundamental period values obtained from hvsr of recorded ground-motions. this classification arises from the relationship between the tf and the vs30 values measured in 52 stations. because the current classification is based on the vs30, the proposed one has the advantage that stations that do not have vs30 measurements can be classified using the tf calculated from the earthquake acceleration records. with this classification, since it considers the fundamental period, we guarantee that sites assigned in the same class have an equivalent dynamic response. this fact plays an important role in the assessment of structures when ground-motion records are used for deterministic and/or probabilistic time-history analyses, decreasing uncertainty and improving the results for a specific region. the influence of this classification on the definition of new ground motion prediction model, compared to the existing/traditional ones remains a significant and interesting issue. although we are aware of the limitations of the vs30 as a parameter to classify sites, based on the results obtained in this study, we were forced to use it in this study since it is the basis of the current regulations. in this sense, we would like to recommend to those responsible of the seismic regulations to reconsider using vs30 as a measure to classify sites. from our perspective, this parameter has clear deficiencies, which do not allow us to accurately classify soft soil sites (t ≥ 0.5 s), as is observed in figure 4. therefore, we believe that it is necessary to use measures that guarantee a reliable site classification, grouping sites with similar geodynamic characteristics, as is the case of the use of the site fundamental period as proxy. acknowledgements this research was partially funded by the national emergency and risk prevention law n° 8933 from costa rica and the ucrea funds from the university of costa rica through the project referenced as b9780. references alfaro, a., pujades, l. g., goula, x., susagna, t., navarro, m., sanchez, j., & canas, j. a. (2001). preliminary map of soil’s predominant periods in barcelona using microtremors. pure and applied geophysics, 158, 2499–2511. doi: 10.1007/pl00001182 asce (2017). minimum design loads and associated criteria for buildings and other structures (asce/sei 7-16). bssc (2003). building seismic safety council, nehrp recommended provisions for seismic regulations for new buildings and other structures. report fema-450 (provisions), federal emergency management agency (fema). washington d.c., united states cadet, h., macau, a., benjumea, b., bellmunt, f., & figueras, s. (2011) from ambient noise recordings to site effect assessment: the case study of barcelona microzonation. soil dynamics and earthquake engineering, 31, 271–281. doi: 10.1016/j.soildyn.2010.07.005 caselles, j. o., pérez-gracia, v., franklin, r., pujades, l. g., navarro, m., clapes, j., canas, j. a., garcía, f. (2010). applying the h/v method to dense cities. a case study of valencia city. journal of earthquake engineering, 14, 192–210. doi: 10.1080/13632460903086069 cfia (2016). código sísmico de costa rica 2010 (revisión 2014). editorial tecnológica de costa rica, cartago, costa rica. di alessandro, c., bonilla, l. f., boore, d. m., rovelli, a., & scotti, o. (2012). predominant-period site classification for response spectra prediction equations in italy. bulletin of the seismological society of america, 102, 680–695. doi: 10.1785/0120110084 dobry, r., borcherdt, r. d., crouse, c. b., & idriss, i. m. (2000) new site coefficients and site classification system used in recent building seismic code provisions. earthquake spectra, 16, 41–67. douglas, j. (2017). ground motion prediction equations 1964-2016. glasgow, uk ghasemi, h., zare, m., fukushima, y., & sinaeian, f. (2009). applying empirical methods in site classification, using response spectral ratio (h/v): a case study on iranian strong motion network (ismn). soil dynamics and earthquake engineering, 29, 121–132. doi: 10.1016/j.soildyn.2008.01.007 ghofrani, h., atkinson gm (2014) site condition evaluation using horizontal-to-vertical response spectral ratios of earthquakes in the nga-west 10.1007/pl 10.1016/j.soildyn 10.1016/j.soildyn 10.1016/j.soildyn 389seismic site classification of the costa rican strong-motion network based on vs30 measurements and site fundamental period 2 and japanese databases. soil dynamics and earthquake engineering, 67, 30–43. doi: 10.1016/j.soildyn.2014.08.015 hassani, b., & atkinson, g. m. (2016). applicability of the site fundamental frequency as a vs30 proxy for central and eastern north america. bulletin of the seismological society of america, 106, 653–664. doi: 10.1785/0120150259 moya-fernández, a., pinzón, l. a., schmidt-díaz, v., hidalgo-leiva, d. a., & pujades, l. g. (2020). a strong-motion database of costa rica: 20 yr of digital records. seismological research letters, 91, 3407–3416. doi: 10.1785/0220200036 nagashima, f., matsushima, s., kawase, h., sanchez-sesma, f. j., hayakawa, t., satoh, t., & oshima, m. (2014). application of horizontal-to-vertical spectral ratios of earthquake ground motions to identify subsurface structures at and around the k-net site in tohoku, japan. bulletin of the seismological society of america, 104, 2288–2302. doi: 10.1785/0120130219 nakamura, y. (1989). a method for dynamic characteristics estimation of subsurface using microtremor on the ground surface. railway technical research institute, quarterly reports, 30, 25–33. pinzón, l. a., pujades, l. g., macau, a., carreño, e., & alcalde, j. (2019a). seismic site classification from the horizontal-to-vertical response spectral ratios: use of the spanish strong-motion database. geosciences, 9, 294. doi: 10.3390/geosciences9070294 pinzón, l. a., pujades, l. g., macau, a., & figueras, s. (2019b). increased seismic hazard in barcelona (spain) due to soil-building resonance effects. soil dynamics and earthquake engineering, 117, 245–250. doi: 10.1016/j.soildyn.2018.11.022 protti, m., gundell, f., & mcnally, k. (1994). the geometry of the wadati-benioff zone under southern central america and its tectonic significance: results from a high-resolution local seismographic network. physics of the earth and planetary interiors, 84, 271–287. doi: 10.1016/00319201(94)90046-9 quintero, r., & güendel, f. (2000). stress field in costa rica, central america. journal of seismology, 4, 297–319. doi: 10.1023/a:1009867405248 steidl, j. h. (2000). site response in southern california for probabilistic seismic hazard analysis. bulletin of the seismological society of america, 90, 149–169. doi: 10.1785/0120000504 wald, d. j., & allen, t. i. (2007). topographic slope as a proxy for seismic site conditions and amplification. bulletin of the seismological society of america, 97, 1379–1395. doi: 10.1785/0120060267 zhao, j. x., irikura, k., zhang, j., fukushima, y., somerville, p. g., asano, a., ohno, y., oouchi, t., takahashi, t., & ogawa, h. (2006) an empirical site-classification method for strong-motion stations in japan using h/v response spectral ratio. bulletin of the seismological society of america, 96, 914–925. doi: 10.1785/0120050124 10.1016/j.soildyn 10.3390/geosciences 10.1016/j.soildyn keywords: ecrowd-sensing technology; geographical location, limitation; high precision; global positioning system; measurement. palabras clave: btecnología de detección de multitudes; ubicación geográfica; limitación; alta precisión; sistema de posicionamiento global; medición. how to cite item zhang, y., wang, b., & zhang, l. (2020). highprecision gps measurement method without geographical restrictions using crowdsensing technology. earth sciences research journal, 24(4), 491-497. doi: https://doi. org/10.15446/esrj.v24n4.92151 in order to improve the flexibility of gps measurement, a high-precision gps measurement method that is not restricted by the geographical location under crowd-sensing technology was proposed. the performance of the crowdsensing network was improved through a regular hexagon-based crowd-smart big data sensing network deployment mechanism. the gps /sins/dr fast and high-precision combined measurement methods were used to achieve high-precision measurement without geographical restrictions. it has been verified that the proposed method in this paper has much better stability in the deployment strategy of a regular hexagon than that of the square. the proposed method can achieve fast acquisition of satellite signals and high-precision positioning, and its measurement accuracy in the low-latitude city and high-latitude city is higher than the online measurement method based on google earth, indicating that it has significant application value. abstract high-precision gps measurement method without geographical restrictions using crowd-sensing technology método de medición gps de alta precisión sin restricciones geográficas con tecnología de detección de multitudes issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.92151 para mejorar la flexibilidad de la medición gps se propuso un método de medición gps de alta precisión que no está restringido por la ubicación geográfica y que se ejecuta bajo tecnología de detección de multitudes. el rendimiento de la red de detección de multitudes se mejoró a través de un mecanismo regular de implementación de redes de detección de macrodatos basado en hexágonos. se utilizaron los métodos de medición combinados gps/sins/dr rápidos para lograr mediciones de alta precisión sin restricciones geográficas. se ha comprobado que el método propuesto en este trabajo tiene una estabilidad mucho mejor en la estrategia de despliegue del hexágono regular que la del cuadrado. el método propuesto puede lograr una rápida adquisición de señales de satélite y un posicionamiento de alta precisión, y su precisión de medición en ciudades de baja latitud y ciudades de alta latitud es mayor que el método de medición en línea basado en google earth, lo que indica que tiene un valor de aplicación significativo. resumen record manuscript received: 03/01/2020 accepted for publication: 10/08/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 491-497 yunxiang zhang, bin wang*, lei zhang college of information science & electronic technology, jiamusi university, jiamusi 154007, china *corresponding author: jmsuwang2019@163.com g ps m e a su r e m e n t s https://doi.org/10.15446/esrj.v24n4.92151 https://doi.org/10.15446/esrj.v24n4.92151 https://doi.org/10.15446/esrj.v24n4.92151 mailto:jmsuwang2019@163.com 492 yunxiang zhang, bin wang, lei zhang introduction crowdsensing refers to the large-scale and complex society sensing tasks that use the mobile devices (mobile phones, tablets, etc.) of ordinary users as the basic sensing unit, through conscious or unconscious collaboration on mobile internet, to realize the distribution of sensory tasks and the collection of sensory data as well as complete those complex social sensory tasks. at present, the application scope of crowd-sensing can be divided into three categories: public facilities, environment, and society (dube et al., 2017). in terms of public facilities, crowd-sensing can detect traffic congestion, road noise or potholes, search for parking spaces, maintain public facilities such as fire hydrants and traffic lights, and have also been used in monitoring and navigation of real-time traffic (lowe et al., 2016). for example, through the sensors installed on the car and the gps, empty parking spaces can be detected and shared on the internet. through the cms system that collects passenger information on the bus, the comfort level of the bus can be evaluated and finally posted on the website. individuals collect information including pictures and names of bus stations via mobile phones, which helps to predict the arrival time of buses on mobile devices (chang et al., 2017). in terms of environment, crowd-sensing technology can be used to collect information such as co2 and pm2.5, noise, and water quality in the air. for example, trajectories corresponding to the time and place of mobile users are collected, and known models are used to calculate and collect the content of co2 and pm2.5 in the atmosphere. this information is collected through the air quality sensor of the mobile phone to realize the function of environmental monitoring. because noise is harmful to hearing, a “track-noise” relationship is collected from mobile users to form a noise map and it will be shared on the network. because people can collect water quality information such as flow, flow velocity, and amount of garbage when passing by rivers, this information collected through mobile apps can be aggregated to a background server and published online (hoang et al., 2016). in society, crowdsensing has been applied in social networking, travel, food, and tourism. for example, much social software recommends friends by mutual friends among users. mobile devices are used to collect user traveling habits and can promote environmentally friendly transportation. through building a “location-photos” library, people can use the pictures taken to query their geographic information. by personally sharing their food and beverage in the circle of friends, you can not only improve yourself but also give others a good opinion. based on gps trajectories and environment information provided by bicycle users, the best cycling route is derived (zhang et al., 2016). although crowdsensing plays a great role in clothing, food, shelter, and transportation, researchers at home and abroad have not stopped studying it in depth. in the past two years, in several well-known academic conferences, many papers have adopted more or less the basic idea of crowd-sensing. the advantage of crowd-sensing is that it has a huge mass base, which makes it outstanding in many fields. in the near future, crowd-sensing will serve society in a smarter and more convenient way (hsueh & chen, 2017). crowdsensing is a new model that utilizes ubiquitous mobile devices to efficiently collect data and is suitable for many large applications. the entire workflow of crowd-sensing is to use groups to collect data, then analyze and process these data, and finally obtain useful information. mi gensuo et al. adopted gps/rfid combined location technology and used the cp so algorithm and bp neural network combined algorithm to filter the output, which achieved high-precision measurement. foreign researchers proposed using gnss and map matching algorithms to achieve precise positioning. liu jiang et al. use a combination of gps and inertial measurement units and perform data fusion through the hcx3 robust filtering method. however, the above methods have certain constraints on geographic location. with crowd-sensing technology, the method in this paper realizes a high-precision gps measurement that is not restricted by geographical location (cai et al., 2016). method big data crowd-sensing network deployment mechanism based on a regular hexagon in this paper, vehicle positioning measurement is taken as an example to study the high-precision gps measurement method under crowdsensing technology. crowd-sensing uses the carrying smart devices or wearable devices of people as the basic sensing unit, and wireless communication within a short distance is used to collect and transmit data (cu et al., 2016). the collection of user participation data is mainly reflected in offline mobile sensing, transmitting data through the human-in-the-loop sensing mode, online social media, and contributing data through various mobile social media. however, due to the limitations of mobile node capabilities and network framework, it is difficult for a crowdsensing network to form a large scale (li et al., 2019). regular hexagonal grid deployment can seamlessly cover the target detection area. figure 1 shows the details of the regular hexagonal grid deployment mechanism of crowdsensing big data network: figure 1. regular hexagon grid deployment supposing that the side length of the regular hexagonal grid is , and  is a constant. when the mobile node is located at the positive center of the cluster, the energy consumption of transmitting data between the clusters is: e k e k dsix elec fs= × + × ×ε β (1) where k describes the length of the transceiver circuit,  fsis the parameter under the free space channel model amplifier, eelec the energy consumed to process kbit data. the energy consumption eelec is mainly related to data send distance d. d rsix sixsix = ×2 3 2 2 (2) r rsix six= 3 3 2 2 (3) where rsix is the length of the regular hexagon. from formula (2) and formula (3), it can be obtained that: d rsix = ⋅ 3 3 2 (4) the data transmission distance between clusters divided by a regular hexagonal grid is shown in figure 2. 493high-precision gps measurement method without geographical restrictions using crowd-sensing technology where  indicates the direction parameter. relative to the earth coordinate system, the angular velocity of the vehicle geographic coordinate system is: w w w w v r v r v r let t etx t ety t etz t yt yt xt xt yt xt =             =  tan         (7) where v is the vehicle moving acceleration of the coordinate system, r is the crowd-sensing technology communication radius of the coordinate system, l is the vehicle moving distance, the corner mark i is the inertial coordinate system, the corner mark e is the earth coordinate system, the corner mark t is the geographic coordinate system, the corner mark b is the vehicle coordinate system. the specific force information fib b obtained by the accelerometer is the specific force of each axis in the carrier coordinate system, but the specific force fit t required in this paper is the specific force of each axis in the geographic coordinate system. the two specific force can be converted by the following matrix (liu et al., 2017): f c fit t b t ib b= (8) according to the proportions, the acceleration in each direction can be calculated. the motion of the vehicle can be regarded as two-dimensional motion in the plane, so only the acceleration in the x and y directions need to be calculated (coco et al., 2016). assume that the measured attitude angles are  and , and the odometer trip is s. the equations composed of sins/dr are: v f w w v w w v v ext t itx t iez t etz t ety t iez t etz t etz t etx t = +( ) +( ) =⋅ − −2 2 ff w w v w w vitx t iez t etz t ety t iez t etz t etz t− −2 2+( ) +( ) l l r= + cos θ λ λ γ= +          rcos (9) the combined measurement results are: l t v r dt r l t v r ldt r t yt yt xt xt = + + = + =< ∫ ∫ 0 0 0 0 cos sec cos arcsin θ λ γ+λ θ 223 13 33 γ −=                 arctan t t (10) where t describes the time. the above results were used as the update matrix for the second measurement of sins. gps satellite signal acquisition basic principles in order to track and decode the gps signal, the gps signal must be captured first, and the necessary parameters of the captured gps signal are immediately passed to the tracking process. the navigation message of the satellite can be obtained through the tracking process. signal acquisition is the process of matching the received satellite signal with the local replica signal. it is achieved by judging whether the carrier frequency and c/a code phase have the maximum correlation with the satellite signal. before the signal acquisition, the approximate doppler frequency shift and the initial position of the c/a code phase must be calculated, so the signal acquisition is a twodimensional search process (gao et al., 2018). at present, gps signal acquisition methods mainly include linear search, parallel code phase search, and parallel frequency search. parallel code phase search and parallel frequency search are realized by fourier parallel of one dimensional frequency code or phase code in the two-dimensional search, which can reduce the calculation amount of signal capture and improve the speed of two-dimensional signal search. in this paper, a faster parallel code phase search is used to further improve the acquisition speed (peyret et al., 2018). figure 2. distance between regular hexagon clusters in figure 2, r is a side length of the regular hexagon. the regular hexagonal division needs to keep the network connected, that is, d rsix ≤ . r is the communication radius. that is: d r r rsix = + ( ) =2 22 3 13 equation 3 is brought into the above equation: d rsix six= ⋅ 26 3 9 (5) to keep the network connected, d rsix ≤ is essential. gps/sins/dr-based fast and high-precision combined measurement method based on crowd-sensing technology sins/dr combined measurement algorithm the strap-down inertial navigation system (sins) has been widely used due to its comprehensive navigation information and complete autonomy in providing navigation information, but inertial navigation error will accumulate over time (kugler, 2017). for vehicle navigation applications, the dead reckoning (dr) system is as autonomous as the inertial navigation system, and the dr system error will slowly increase with the accumulation of time. in the case of long-term measurement, the measurement error of sins is much larger than that of dr (zhu et al., 2018). the advantages of the two were used to complement each other, and a sins/dr combined measurement algorithm was proposed. the algorithm in this paper introduces the strap-down inertial navigation attitude matrix into the dead reckoning state calculation matrix, so the update reckoning matrix can no longer be solved (golsorkhi et al., 2016). through this algorithm, the dead reckoning can suppress the divergence of a strap-down inertial navigation error. in this case, the error angle changes of the two are basically the same. in this method, the quaternion is calculated based on the initial attitude angle, and then the directional cosine matrix ct b is obtained. the result is: ct b = + ( ) ( ) +( )λ λ −λ −λ λ λ −λ λ λ λ −λ λ λ λ λ λ λ 0 2 1 2 2 2 3 2 2 2 0 3 2 3 0 2 1 2 0 3 02 2 2 22 1 2 2 2 3 2 2 3 0 2 1 3 0 2 2 3 0 1 0 2 1 22 2 2+ + ( ) ( ) +( )λ λ −λ λ λ −λ λ λ λ −λ λ λ λ λ λ λ −λ −−λ λ2 2 3 2+             (6) 494 yunxiang zhang, bin wang, lei zhang sins/dr-assisted satellite signal acquisition method the doppler frequency shift of a satellite signal is mainly due to the relative motion between the carrier and the satellite, the satellite clock error and the receiver clock error. its doppler frequency shift expression is: f f f fsh dy rc sc= +  (11) where fdy is the doppler frequency shift between the satellite and the motion carrier, frc is the clock error frequency of the receiver, fsc is the clock error frequency of the satellite. among them, the doppler frequency shift between the satellite and the moving carrier has the most significant influence on fsc. ignoring the receiver clock error frequency and the satellite clock error frequency, the doppler frequency between the satellite and the moving carrier can be predicted by introducing the carrier speed into the satellite navigation messages, which can be expressed as: f f fsh wdy zdy= + (12) therefore, the parallel code phase search method can be used to introduce sins/dr combined measurement data into a gps receiver, and predict the doppler frequency in advance. it improves the acquisition speed of satellite signals (gao, 2018). the implementation process is as follows: step 1: the doppler shift of satellite motion is calculated from the ephemeris: f f v r crdr r s e s = (13) where fr is the carrier frequency of the gps satellite, vs is the operating speed of the satellite in orbit, re is the radius of the earth, c is the speed of light, rs and is the average radius of the satellite orbit. step 2: calculate the doppler frequency shift of the carrier motion. from the above sins/d r combined measurement analysis, the velocity of the motion carrier can be expressed as: v v vz ets t ety t= ( ) + ( )2 2 (14) the doppler frequency shift of the motion carrier is: f f v v cz r ets t ety t = ( ) ( )2 2 (15) step 3: the doppler frequency shift can be obtained through the above calculation: f f c v r r v vsh r s e s etx t etx t= + ( ) + ( )        2 2 (16) step 4: the satellite signal was mixed with the local replica signal (with doppler frequency shift assistance) to obtain i and q signals (hoang et al., 2016). step 5: fourier transform was performed on the i and q signals. then they were multiplied with the local c/a code that completed the complex conjugate transform and fourier transform. and modulus operation was performed on the result, obtaining the last captured result (li & xu, 2016). data fusion processing algorithm when the vehicle is running at high speed, its high dynamics and complicated external environment will seriously interfere with the normal measurement work of the gps receiver. therefore, when improving the satellite signal acquisition capability, ensuring the sins/dr data connection if the satellite measurement data is interrupted is the key to improving measurement accuracy. this paper combines the least square method with the kalman filter algorithm to propose a new data fusion processing algorithm, lskf. the core idea of lskf is to determine whether the satellite measurement data is lost through the acquisition of satellite signals. in the case of data loss, the least square method is used to estimate the satellite measurement data, and the results are used as quasi observations and will be further processed by kalman filter. when the data is complete, kalman filter is directly used. this method avoids the non-linear errors in the observation equation and achieves high-precision tracking measurement of the moving vehicle. analysis of lskf algorithm model the least squares estimate is obtained from the navigation measurement data. it is assumed that the measurement data x yu u,( ) before gps satellite information loss can be measured, and u represents a constant. the route distance is regarded as an infinitely small value. for infinitely small distances, the route can be regarded as a straight line. with the least-square method, the fitting curve can be expressed as: y a bxn u= + (17) where a and b describe vectors. supposing that: j a b y a bxu u u n ,( ) = +( )  = ∑  2 1 (18) where n = 1,2,3 ..., n. the result is: nb b x y a x b x x y u u u n u n u u u u u n u n u n + = + =          == === ∑∑ ∑∑∑ 11 2 111 (19) the least squares solution can be obtained through equations (17) to (19). in order to realize the simulation calculation, the matrix is expressed as: a b x x x x x u t u       =                         1 1 1 1 1 1 1 1 1 1        xx y yu t u                       1  (20) the results obtained previously were used as quasi observations and a kalman filter calculation model was introduced. kalman filter is a recursive linear minimum variance estimation technique that uses recursive estimation to calculate continuous measurement data. the gps measurement data was processed by the least square method, and the obtained coordinates were used as quasi observations. because the processed data was approximately linear, kalman filter was used to optimize the measurement results (li et al., 2016). the kalman filter steps are as follows: (1) the initial values are calculated. when u =1 and u = 2, the train positions calculated by the least squares method are (x1, y1) and (x2, y2). and the system initial value x2 can be expressed as: x x y x x y y t2 2 2 2 1 2 1=    (21) (2) using the initial value and state variable in step 1, the estimated value of the system when u ≥ 3 is: x k k y k k x k p k k y k k p k y q k , , , , *  1 1 1 1 1 1 1 ( ) = ( ) −( ) −( ) = −( ) −( ) + −( )     (22) where k represents the state variable, p represents the observed value, q represents the estimated value. (3) the error of the observed value to the estimated value is corrected as: 495high-precision gps measurement method without geographical restrictions using crowd-sensing technology p k p k k p r k p x k x k k p k z t( ) = −( )  + −( )       ( ) = −( ) + ( ) − − − , , 1 1 1 1 1 1 kk px k k p k p k p r kt ( ) − −( )  ( ) = ( ) −( )          − , 1 11 (23) result network connectivity in order to verify the advantages of the deployment strategy proposed by this method, matlab2016a was used to analyze the connectivity stability of the crowd-sensing network. the square grid division was compared with the regular hexagonal grid division in this paper, as shown in figure 3. figure 3. maximum euclidean distance it can be seen from figure 3 that the maximum euclidean distance gradually increases with the increase of the area. the maximum euclidean distance of the square grid division is significantly higher than that of the regular hexagonal grid division. the quality gradually becomes unstable with the increase of the maximum euclidean distance. because regular hexagonal grid division has smaller maximum euclidean distance, and its communication quality is slightly better than that of square grid division, indicating the deployment strategy of the method in this paper has advantages. gps satellite signal acquisition verification according to the “specifications for the design of high-speed railways”, railways designed for passenger trains with a travel speed of 250 to 350 km/h are called high-speed railways. therefore, in this paper, a high-speed motion carrier is used to simulate the rapid driving of the vehicle, and the debugging simulation is carried out on computer. the doppler frequency shift of the signal carrier was set as ± 5 khz and the maximum travel speed of the carrier was set as 300km/h. satellite signal acquisition simulation analysis was performed on the traditional gps receiver and the gps receiver with carrier motion doppler frequency shift. the peak values of the signals captured by the two receivers are shown in figure 4. obviously, after doppler frequency shift is introduced in this method, the correlation of acquisition has been greatly improved, and the peak value has been improved. fig. 5 is the time curve of capturing 20 signals with different intensities in both cases. without the introduction of carrier motion doppler frequency shift, the gps satellite signal acquisition time is shown in figure 5. figure 5. gps satellite signal acquisition time in figure 5, the maximum value of the acquisition time for a conventional gps receiver is 0.36s. in the case of the method in this paper, the maximum value of the acquisition time is about 0.25s. the capture speed was improved by 0.11s. suppose that 10 test positions are the coordinate positions of 10 points evenly distributed on a road. the total length of the test road is 1.5 km. positions 2, 6, and 10 are on the road section without signal blocking. signals in positions 1, 3, 4, 5, 7, 8, and 9 are blocked by trees and houses to varying degrees, and the other test environments are exactly the same. fig. 6 shows the data of satellite capture measurement data using the traditional gps receiver and the gps receiver proposed in this paper: as can be seen in fig. 6, after using a gps receiver that introduces doppler frequency shift of the motion carrier, and the method in this paper successfully captures measurements more often than an ordinary gps receiver. measurement accuracy in order to further verify the measurement accuracy of the method in this paper. the method in this paper was applied to vehicle measurement in high figure 4. verification results of gps satellite signal acquisition 496 yunxiang zhang, bin wang, lei zhang latitude city london and low latitude city lhasa. the geographic location online measurement method based on google earth was used to make a comparison. the results are shown in figure 7 and figure 8. figure 7. measurement accuracy test results of two methods in high latitude city it can be seen from fig. 7 and fig. 8 that the accuracy of the method in this paper is as high as 98.89% and 98.99% in the measurement of vehicles in low-latitude city and high-latitude city, respectively. compared with the online measurement method based on google earth, this method has higher precision. discussion as an emerging research field, a crowd-sensing network is facing many challenges, which can be summarized as the following seven aspects. (1) the common platform of a crowd-sensing network. at present, academia and industry have designed and developed a variety of crowd-sensing applications. they usually have similar or partially overlapping functions and require the same or interrelated sensing data. they are facing a series of common issues and challenges such as data collection, resource allocation, energy conservation, user motivation, security and privacy. at present, this independent development model is very inefficient, resulting in a great waste of resources. therefore, building a common platform for the crowd-sensing network is a basic problem that should be solved urgently in this field. (2) front-end processing of crowd-sensing data. the raw sensing data collected by sensors such as gps, accelerometer, microphone, and camera are usually very noisy, incomplete or redundant, and it is difficult to use them directly. therefore, effective front-end processing algorithms need to be designed, mainly including two types. one is data quality enhancement, including operations such as eliminating noise, filtering abnormal data, recovering lost data, repairing and enhancing low-quality images. the other is context inference, which includes inferring traffic patterns, sports patterns, and social occasions (such as meetings, calls, watching tv, etc.) of users and the surrounding environment (such as road bumps, noise levels, etc.). (3) efficient transmission of crowd-sensing data. many crowd-sensing applications need to continuously collect the sensing data and transmit it to the data center. the transmission method based on the mobile cellular network will consume equipment power and data traffic of the users, and put a lot of pressure on the mobile cellular network. therefore, energy-efficient data transmission methods need to be designed. for example, in the short-range wireless communication methods, the opportunity of users to contact each other or users to contact with wifi hotspots is used to forward data. (4) value mining of crowd-sensing data. crowd-sensing data comes from different users, different sensors, and has features such as multi-modality and multi-association. these massive data must be intelligently analyzed and mined to effectively exert value and form a leap from data to information and even to knowledge. the technologies involved include big data storage and processing, data quality management, and multi-modal data mining. (5) resource optimization of crowd-sensing network. overcoming the resource constraints of mobile nodes in the terms of energy, bandwidth, and computing is the key to the practical application of the crowd-sensing network. first, the number of users and the availability of sensors change dynamically over time, and it is difficult to build models and make predictions accurately based on energy and bandwidth requirements to complete specific sensing tasks. second, it is necessary to select an effective subset from a large number of users with different sensing capabilities, and reasonably schedule the sensing and communication resources under resource constraints. (6) incentive mechanism of crowd-sensing network. crowd-sensing applications rely on the participation of a large number of ordinary users, and the users will consume their device power, computing, storage, communication and other resources and bear the threat of privacy leakage when participating. therefore, a reasonable incentive mechanism must be designed as repay for the users, which ensures the quality of required data. figure 6. gps comparison of satellite acquisition and positioning figure 8. measurement accuracy test results of two methods in low latitude city 497high-precision gps measurement method without geographical restrictions using crowd-sensing technology (7) security and privacy protection of the crowd-sensing network. sensing data may leak the privacy and sensitive information of the users, so a reasonable privacy protection mechanism must be designed to ensure the privacy of the users while completing data collection as much as possible. conclusion crowd-sensing has risen in recent years and is still in its early stages. its development depends largely on the development and popularity of mobile devices. in the process of crowd-sensing, the collection and use of data is very important. based on crowd-sensing technology, this paper proposes a high-precision gps measurement method that is not restricted by geographical location. a crowd-sensing big data network deployment mechanism based on regular hexagon was designed. based on this mechanism, a rapid high-precision gps/ sins/dr combined method was used to achieve high-precision measurement without geographical restrictions. the crowd-sensing network in this paper has high network connectivity and good communication quality. and after doppler frequency shift is introduced into this method, the correlation of its acquisition has been greatly improved, and the acquisition peak has been improved. the traditional gps receiver has a maximum acquisition time of 0.36s without the carrier motion doppler frequency shift. in this method, the carrier motion doppler frequency shift is used, and the maximum acquisition time is about 0.25s, which is increased by 0.11s. and the number of acquisition measurements is higher than traditional gps receivers. in lowlatitude city and high-latitude city, the measurement accuracy of the method in this paper is as high as 98.89% and 98.99%, which illustrates that it can achieve high-precision gps measurement without geographical restrictions. in the location service application, the crowd-sensing network system should pay attention to the setting of location privacy protection and incentive mechanism. location privacy leakage is the biggest threat to such applications. in location-based services, sensitive data is primarily user location information. an attacker can infer the sensitive information of the users based on the location information sent by them, including home address, commuting route, etc. in a crowd-sensing system based on location service, each user sends data to the server. this data often contains information about user privacy, such as their location and time. in the crowd-sensing system, any individual or organization can initiate tasks and motivate users to participate in sending data, so the application server is considered untrusted. if location information of the users is leaked to an untrusted server, it will cause great harm to the user. even if the user does not send data using their real name, the server can identify the specific user or reduce the anonymity of the user according to the relevance of the data sent by the user. the goal of user location privacy protection is that the server cannot establish the association between user identity and data, so the server cannot identify a specific user based on the data sent by the user. how to protect the location privacy of the users is a major problem faced by the crowd-sensing system. reference cai, a. h., he, j. w., & wang, p. (2016). experiment and error analysis of path planning for sonobuoy array deployment. journal of china academy of electronics and information technology, 11, 21-24. chang, h. k., sun, y. k., & chan, g. p. (2017). an adaptive complex-ekfbased doa estimation for gps spoofing detection. iet signal processing, 12, 174-181. coco, d. s., gaussiran, t. l., & coker, c. (2016). passive detection of sporadic e using gps phase measurements. radio science, 30, 1869-1874. dube, p., bernard, j. e., & gertsvolf, m. (2017). absolute frequency measurement of the 88sr+ clock transition using a gps link to the si second. metrologia, 54, 290-298. gao, l. (2018). urban strip road landscape sharp feature extraction integrity optimization simulation. computer simulation, 7, 261-264. gao, r. m., yang, w. j., & dai, q. (2018). characteristic polynomials of graphical arrangements corresponding to wheel graphs and join graphs. journal of jilin university (science edition), 56, 859-863. golsorkhi, m. s., lu, d., & guerrero, j. m. (2016). a gps-based decentralized control method for islanded microgrids. ieee transactions on power electronics, 32(2):1-1. gu, y., yuan, l., fan, d., you, w., & su, y. (2016). seasonal crustal vertical deformation induced by environmental mass loading in mainland china derived from gps, grace and surface loading models. advances in space research, 59, 88-102. hoang, g. m., denis, b., harri, j., & slock, d. t. m. (2016). breaking the gridlock of spatial correlations in gps-aided ieee 802.11p-based cooperative positioning. ieee transactions on vehicular technology, 65, 9554-9569. hsueh y. l., & chen, h. c. (2017). map matching for low-sampling-rate gps trajectories by exploring real-time moving directions. information sciences, 434, 55-69. kugler, l. (2017). why gps spoofing is a threat to companies, countries. communications of the acm, 60, 18-19. li, x., & xu, q. (2016). a reliable fusion positioning strategy for land vehicles in gps-denied environments based on low-cost sensors. ieee transactions on industrial electronics, 99, 1-1. li, h., guo, z. y., liu, c., & zhao, y. (2019). stability improvement method for cascaded dc-dc converters with additional inductor current linear feedback control. journal of power supply, 17, 103-110. li, z., filev, d. p., kolmanovsky, i., atkins, e., & lu, j. (2016). a new clustering algorithm for processing gps-based road anomaly reports with a mahalanobis distance. ieee transactions on intelligent transportation systems, 99, 1-9. liu, x. l.，chang, j. c., & liu, r. n. (2107). synthesis method of initial noise samples of oil pump cavitation fault based on characteristic frequency analysis. automation & instrumentation, 12, 62-164. lowe, s. t., labrecque, j. l., zuffada, c., romans, l. j., young, l. e., & hajj, g. a. (2016). first spaceborne observation of an earth-reflected gps signal. radio science, 37, 7-1-7-28. peyret, m., djamour, y., hessami, k., regard, v., bellier, o., vernant, p., daignieres, m., nankali, h., van gorp, s., goudarzi, m., chery, j., bayer, r., & rigoulay, m. (2018). present-day strain distribution across the minab-zendan-palami fault system from dense gps transects. geophysical journal international, 179, 751-762. zhang, d. h., xiao, z., igarashi, k., & ma, g. y. (2016). gps-derived ionospheric total electron content response to a solar flare that occurred on 14 july 2000. radio science, 37, 19-1-19-11. zhu, z. g., he, b. b., & xue, r. (2018). image feature recognition method for dangerous situation in intelligent substation. chinese journal of power sources, 42, 597-600. keywords: artificial neural network; bp neural network; regional water resources; water resources carrying capacity; carrying capacity prediction; palabras clave: red neuronal artificial; red neuronal bp; recursos hídricos regionales; capacidad de carga de recursos hídricos; predicción de la capacidad de carga. how to cite item shi, c., & zhang, z. (2021). a prediction method of regional water resources carrying capacity based on artificial neural network. earth sciences research journal, 25(2), 169-177. doi: https://doi.org/10.15446/esrj. v25n2.81615 to better predict the water resources carrying capacity and guide the social and economic activities, a prediction method of regional water resources carrying capacity is proposed based on an artificial neural network. zhaozhou county is selected as the research area of water resources carrying capacity prediction, and its natural geographical characteristics, social economy, and water resources situation are explored. according to the regional water resources quantity and utilization characteristics and evaluation emphasis, the evaluation index system of water resources carrying capacity is constructed to evaluate the importance and correlation of water resource carrying capacity. the pressure degree of water resources carrying capacity is divided into five grades. according to the evaluation standard of bearing capacity, the artificial intelligence bp neural network model is constructed. based on the main impact factors of water resources carrying capacity in this area, the water resources carrying capacity grade is obtained by weight calculation and convergence iteration by using neural network model and influence factor data to realize the prediction of water resources carrying capacity. the research results show that the network model can meet the demand for precision. the prediction results have a high degree of fit with the actual data, indicating that human intelligence can obtain accurate prediction results in water resources carrying capacity prediction. abstract a prediction method of regional water resources carrying capacity based on artificial neural network método de predicción de la capacidad de carga de los recursos hídricos regionales basado en una red neuronal artificial issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n2.81615 para predecir la capacidad de carga de los recursos hídricos con mayor precisión y orientar mejor las actividades sociales y económicas, se propone un método de predicción de la capacidad de carga de los recursos hídricos regionales basado en una red neuronal artificial. el condado de zhaozhou se selecciona como el área de investigación de la predicción de la capacidad de carga de los recursos hídricos, y se exploran sus características geográficas naturales, la economía social y la situación de los recursos hídricos. de acuerdo con las características regionales de cantidad y utilización de los recursos hídricos y el énfasis de la evaluación, el sistema de índice de evaluación de la capacidad de carga de los recursos hídricos se construye para evaluar la importancia y el grado de correlación de la capacidad de carga de los recursos hídricos, y el grado de presión de la capacidad de carga de los recursos hídricos se divide en cinco grados. de acuerdo con el estándar de evaluación de la capacidad de carga, se construye el modelo de red neuronal bp de inteligencia artificial. con base en los principales factores de impacto de la capacidad de carga de los recursos hídricos en esta área, el grado de capacidad de carga de los recursos hídricos se obtiene mediante el cálculo del peso y la iteración de convergencia utilizando el modelo de red neuronal y los datos de factores de influencia, para realizar la predicción de la capacidad de carga de los recursos hídricos. los resultados de la investigación muestran que el modelo de red puede satisfacer la demanda de precisión y los resultados de la predicción tienen un alto grado de ajuste con los datos reales, lo que indica que la inteligencia humana puede obtener resultados de predicción precisos en el proceso de predicción de la capacidad de carga de los recursos hídricos. resumen record manuscript received: 11/08/2018 accepted for publication: 08/12/2019 earth sciences research journal earth sci. res. j. vol. 25, no. 2 (june, 2021): 169-177 chaoyang shi1, zhen zhang2* 1information network center, zhengzhou yellow river nursing vocational college, zhengzhou, 450066, china 2school of intelligent engineering, zhengzhou university of aeronautics, zhengzhou, china *corresponding author: pzhangza@136.com h y d r o l o g y mailto:pzhangza@136.com 170 chaoyang shi, zhen zhang introduction water resources are an important material basis to support the development of human society. it is the most widely impact factor in nature. at the same time, it is irreplaceable and limited (li et al., 2018). with the geometric expansion of the world population, the demand for water resources in people’s daily life and industrial and agricultural production has changed greatly in quality and quantity. the shortage of water resources, water pollution and low utilization rate of water resources have been paid more and more attention. people realize that the problem of water resources not only restricts the sustainable development of social economy, but also threatens the survival of human beings (menhas et al., 2019). therefore, the theory of water resources carrying capacity based on the principle of sustainable development should come into being. the research on water resources carrying capacity can be used as a tool to judge whether the regional water resources supply and social development are in a coordinated state, and also can provide certain reference for scientific management of water resources (chenini et al., 2019). zhu et al. (2019) proposed to increase the water storage capacity of the reservoir as a separate index, used a cloud model to calculate the water resources carrying capacity of hubei province, and used the dematel method to determine the importance of reservoir water storage. finally, the degree of obstacles was considered to explore the main factors affecting the water resources carrying capacity of hubei province. in liu et al. (2019) yichang city was taken as an example, and based on the comprehensive evaluation of the carrying capacity of the “water resourcessocial economyecological environment” composite system in yichang city from 2005 to 2015, the gm (1,1) grey prediction model was used to predict the water resources carrying capacity in 2020, 2025 and 2030. based on the analysis of the uncertainty of the forecast results in the future level year, the sustainability of the socioeconomic development model of yichang city under four scenarios was discussed, combining with the strictest water resources management system and river length system. jia and long (2018) used conventional trend method to evaluate water resources carrying capacity of zhongshan city, constructed sd system model of water resources carrying capacity, and compared and analyzed water resources carrying capacity under conventional development mode and different policy development scenarios. guo (2020) aimed at the problems of long time-consuming and low accuracy of prediction results of traditional prediction systems for regional water resources sustainable carrying capacity, a new prediction system of regional water resources sustainable carrying capacity was designed based on big data technology. the system hardware was composed of four parts: general control layer, subsystem function layer, subsystem function layer and system support function layer. at the same time, real-time monitoring module, information management module, decision support module and remote control module were introduced. the software display field was designed by using the display button and exit button. through system initialization, sample data reading, sample internal zero matrix definition, the proportion of each index was calculated, and the software process was composed of the output of sustainable water resources carrying capacity. the artificial neural network method has a wide range of adaptability, learning ability and mapping ability. it can approach any nonlinear function in theory, and can usually obtain satisfactory results in the modeling and prediction of multivariable nonlinear systems. because the above methods in references only consider some useful information, lack of overall understanding, and the conditions of prediction methods have certain limitations, in order to further accurately predict water resources carrying capacity, this paper proposes a new prediction method of regional water resources carrying capacity based on an artificial neural network. the main work of this paper is described as follows: (1) zhaozhou county is selected as the research area of water resources carrying capacity prediction; (2) through water resources system evaluation index, social system evaluation index, economic system evaluation index and ecological environment system evaluation index, the evaluation index system of water resources carrying capacity is constructed, which truly reflects the relationship between supply and demand of water resources, social and economic development level and regional ecological environment, and then makes a reasonable judgment on the strength of regional water resources carrying capacity; (3) through the analysis of the definition and calculation formula of each index, the classification accuracy of evaluation index grading standard is improved; (4) according to the evaluation grade standard of water resources carrying capacity, the bp neural network model is more applicable; (5) the neural network model is used to predict the water consumption. the results show that the prediction accuracy of this method is high, and has good effectiveness and practicality. overview of the study area zhaozhou county is located in the southwest of heilongjiang province, the center of songnen plain, the north bank of songhua river, close to daqing oilfield. the geographical coordinates of zhaozhou county are 124 ° 48 ′ 12 ″ 125 ° 48 ′ 03 ″ e and 45 ° 35 ′ 02 ″ 46 ° 16 ′ 08 ″ n. the county covers an area of 2445 km2, with a cultivated land area of about 2.22 million mu and a vast grassland area of 850000 mu. zhaozhou county now governs 6 towns (zhaozhou town, erjing town, fengle town, yongle town, chaoyanggou town, xingcheng town), 6 townships (yushu township, chaoyang township, xinfu township, shuangfa township, tuogu township, yongsheng township), 2 pastures (star pasture and paradise pasture), 104 administrative villages and 732 natural villages. physical and geographical characteristics topography zhaozhou county is located in the hinterland of songnen plain. there are no rivers passing through the county. there are a small number of wetlands and small marshes in some low-lying areas. the surface water resources are small, and there is no railway for the time being. however, the highway traffic is relatively developed. the higher terrain area of zhaozhou county is changwu highland in the northeast. the terrain gradually decreases from the south and west of changwu highland. the altitude of the territory is about 118-232m, the average altitude of the agricultural production area is about 167m, the elevation of the southern and western border area is about 130m, and the ground slope is generally between 1-10 degrees. the county covers an area of 2445 km2, with 2.22 million mu of cultivated land, accounting for 60% of the total area. the grassland area is 850000 mu, accounting for about 23% of the total area. the length of the north-south direction is about 77km, and the width of the east-west direction is about 72km. the terrain is flat. there has never been a flood disaster in history. climatic characteristics zhaozhou county is located in a high latitude area, belonging to the mid temperate continental monsoon climate, with short spring and less rainfall, high temperature in summer but abundant rainfall, cool and moderate temperature in autumn as well as cold and dry climate in winter. zhaozhou county belongs to the first accumulated temperate zone of heilongjiang province, with an annual sunshine duration of 2900 hours and abundant light and heat resources. the annual average active accumulated temperature is about 2800 °c and the frost free period is about 143 days. in zhaozhou county, the temperature changes greatly in spring, summer, autumn and winter. the extreme high temperature in summer reaches 38.1 °c, the extreme low temperature in winter reaches 35.3 °c, and the annual average temperature is 4.2 °c. socio-economic profile the total population of zhaozhou county is 431387, including 82970 urban population, 348417 agricultural population and 158173 households. according to the registered residence population, the urbanization rate is 19.23%. the main food crops in zhaozhou county include rice, corn, soybean, etc. the economic crops include sugar beet, garlic, oil, sunflower, melon and vegetable, hemp and flue-cured tobacco. the total sown area of crops is 150100 hm2, the effective irrigation area is 66900 hm2, and the construction of ecological high standard farmland is 10000 mu. 171a prediction method of regional water resources carrying capacity based on artificial neural network overview of water resources surface water resources there is no natural river passing through zhaozhou county, and most of the surface water resources are supplied by rainfall. the surface runoff only occurs in the rainy season of a wet year with large rainfall, and limited by many conditions, the surface water resources that can be developed and utilized are less. according to the evaluation report of surface water resources in heilongjiang province, zhaozhou county is located in the low runoff value area of heilongjiang province. the annual average annual runoff depth is 25.53mm, the average annual surface water resources is 60 million m3, and the available runoff is 40 million m3. the average annual rainfall is 456.15mm. the minimum precipitation was 236.4mm in 2001 and 696.4mm in 1981, with a frequency of 1.3% and a corresponding runoff depth of 95mm. about 82% of the annual precipitation in zhaozhou county is mainly concentrated in june, july, august and september, and the precipitation in other months is less. the total annual inflow of the north main drainage channel is 43.13 million m3, including 16.15 million m3 in zhaodong city and 22 million m3 in the west main drainage channel. groundwater resources the aquifers in zhaozhou county are mainly composed of quaternary harbin formation, lindian formation confined aquifer, qiqihar formation phreatic aquifer, tertiary daan formation and cretaceous mingshui formation confined aquifer. the characteristics of different aquifers also have certain differences. the lithology of aquifers mainly includes sand gravel and quaternary pore phreatic aquifer. due to the characteristics of thin sand layers, uneven distribution, weak water yield and great difficulty in mining, centralized mining can not be carried out. however, the tertiary and quaternary confined aquifers in the lower part of the aquifer are characterized by large sand thickness and wide distribution. it can be used as the main mining layer for people’s daily life, agricultural irrigation and industrial production water. according to the equilibrium method and pumping test method, the exploitable amount of confined water and phreatic water in zhaozhou county is 63 million m3 / year and 30.6 million m3 / year respectively. total water resources the total amount of water resources is equal to the sum of surface water resources and groundwater resources. however, under the action of natural water cycle, there is a mutual recharge relationship between surface water and groundwater, and there is a part of repeated calculation between them. therefore, the repeated calculation amount of surface water and groundwater should be subtracted when calculating the total amount of water resources. the calculation formula is as follows: w w wtotal surface water quantity underground water quantity= + −− wrepetition quantity (1) where: wtotal is the total amount of water resources; wsurface water quantity is the amount of surface water resources; wunderground water quantity is the amount of groundwater resources; wrepetition quantity represents the amount of repeated calculation between surface water and groundwater. according to daqing statistical yearbook, the statistics of water resources in zhaozhou county over the years are shown in table 1. it can be seen from table 1 that the total amount of water resources in zhaozhou county will change greatly from 134 million m3 in 2010 to 294 million m3 in 2019. among them, the surface water resources have been relatively stable, maintaining below 100 million m3, reaching the highest value of 209 million m3 between 2010 and 2019 in 2013, indicating that the surface water resources in zhaozhou county are quite scarce. construction of evaluation index system of water resources carrying capacity in zhaozhou county at present, the research on water resources carrying capacity has been continuing, and various theories, concepts and research methods have also been emerging, but a complete theoretical system has not yet been formed. evaluation index of water resources carrying capacity there are many factors that affect the carrying capacity of regional water resources. the importance and influence of indexes, regional development trends and data acquisition should be considered in the selection of specific indexes of each system. according to the establishment principle of the above index system and combined with the specific situation of zhaozhou county, the evaluation indexes are selected as follows: the available amount of water resources per unit area (10000 m3 / km2); the available amount of water resources per capita (m3); the development and utilization degree of groundwater resources (%); and the population density (person / km2); urbanization rate (%); natural growth rate (%); rural per capita water consumption (l / (day · person)); urban per capita water consumption (l / (day · person)); per capita grain occupation (kg / person); per capita gdp (10000 yuan / person); cultivated land irrigation rate (%); water consumption of 10000 yuan gdp (m3 / 10000 yuan); the proportion of added value of primary industry in gdp (%); the proportion of the added value of three industries in gdp (%); gdp growth rate (%); ecological environment water consumption rate (%); forest coverage rate (%). these indexes can reflect the relationship between supply and demand of water resources, socio-economic development level and regional ecological environment in zhaozhou county, and then make a reasonable judgment on the strength of regional water resources carrying capacity. table 1. statistics of water resources in zhaozhou county over the years particular year total water resources surface water resources ground water resources repeated calculation of surface water and groundwater annual precipitation (mm) water production coefficient calculation unit area (km2) water production scale (10000 m2 / km2) 2010 1.34 0.4 0.94 0 283.2 0.19 2445 5.5 2011 1.24 0.3 0.94 0 328.1 0.19 2445 5.07 2012 2.01 0.59 1.4 0 408.1 0.2 2445 8.21 2013 2.09 0.61 1.48 0 431 0.2 2445 8.53 2014 2.07 0.61 1.46 0 427.1 0.2 2445 8.48 2015 1.37 0.43 0.94 0.15 468.8 0.12 2445 5.6 2016 2.66 0.68 2.05 0 465.6 0.24 2445 11.18 2017 3.32 0.73 2.59 0 587.7 0.23 2445 13.38 2018 2.67 0.68 1.99 0.68 452.2 0.24 2445 10.94 2019 2.94 0.71 2.23 0 507 0.24 2445 12.03 172 chaoyang shi, zhen zhang definition and calculation formula of each index based on the comprehensive analysis of the relationship among different factors affecting water resources carrying capacity in zhaozhou county, referring to the evaluation results of water resources carrying capacity in other similar areas, according to the selection principle of evaluation index system, 17 single indexes are selected. the connotation and calculation method of each single index are described as follows. evaluation index of water resources system (1) water resources per unit area (cheng et al., 2018) (10000 m3 / km2): water resources per unit area refers to the ratio of the total water resources in the current year to the total land area of the region. the index reflects the abundance of natural water resources in this area. the larger the index value is, the richer the water resources in the region are. it has a certain reference value for measuring the water shortage in this area. water resources per unit area total water resources total a  rrea of regional land (2) (2) per capita water resources occupation (hu et al., 2019) (m3): per capita water resources occupation refers to the ratio of the total regional water resources to the total regional population. to a certain extent, the index can truly reflect the relative abundance of water resources in the region, and is an evaluation of the actual utilization effect of regional water resources. water resources per capita total water resources total popu  llation (3) (3) development and utilization degree of groundwater resources (%): the development and utilization degree of groundwater resources refers to the percentage of groundwater resources development in the total amount of groundwater resources in the region. the index is a reverse index. the more the index value is, the greater the potential of groundwater development and utilization is. development and utilization degree of groundwater resourcess development and utilization of groundwater total amount  oof groundwater resources (4) evaluation index of social system (1) population density (person / km2): population density refers to the number of people per unit land area. this index can measure the population distribution of a region and reflect the population pressure of unit land area truthfully. population density total population total area of regional = lland %×100 (5) (2) urbanization rate (xu et al., 2020) (%): urbanization rate refers to the proportion of urban population in the total population in the region. this index can reflect the level of social and economic development, the living standard of residents and the quality of population. urbanization rate urban population total population %= ×100 (6) (3) natural growth rate (‰): the natural growth rate is generally the ratio of the number of births minus the number of deaths in the current year and the average total population in the same period, expressed in thousands. the natural growth rate of population is an index reflecting the natural growth of population. the change between birth rate and mortality rate directly affects the increase or decrease of natural growth rate of the population. natural population growth rate annual number of births ann = − uual death toll annual average population %×100 (7) (4) rural per capita domestic water consumption (l / (day · person)): the rural per capita domestic water consumption refers to the ratio between the total domestic water consumption of rural residents and the total rural population. this index can reflect the living water level of rural people in the whole region. rural per capita domestic water consumption domestic water = consumption of rural residents rural population* % 365 100× (8) (5) urban per capita domestic water consumption (l / (day · person)): urban per capita domestic water consumption refers to the ratio of the total domestic water consumption of regional urban residents to the total urban population. this index can reflect the living water level of urban people in the whole region. urban per capita domestic water consumption domestic water = consumption of urban residents urban population* % 365 100× (9) (6) per capita grain occupation (kg / person): the per capita grain occupancy refers to the ratio of the total grain production and the total population in the region in the current year. this index can reflect the level of people’s life in the region as a whole. grain per capita domestic water consumption of urban resid = eents urban population* % 365 100× (10) evaluation index of economic system (1) per capita gdp (10000 yuan / person): per capita gdp refers to the ratio of the gdp of the region to the total population at the end of the year. this index can measure regional economic development. per capita gdp gdp total population at the end of the year = ×1100% (11) (2) farmland irrigation rate (%): farmland irrigation rate refers to the ratio of irrigation area to total cultivated land area, which can reflect the actual situation of regional agricultural development. irrigation rate of cultivated landirrigation rate of cultivated land irrigation area total ar = eea of cultivated land %×100 (12) (3) water consumption per 10000 yuan gdp (m3 / 10000 yuan): the water consumption per 10000 yuan gdp is the ratio of the total annual gdp to the total water consumption in 10000 yuan. the index can reflect the regional industrial structure, economic development level and water-saving measures to a certain extent. at the same time, it can also reflect the coordination of regional economic development and water resources. water consumption per 10000 yuan gdpwater consumption per yuan gdp gdp total water consumpt 10000  iion (13) (4) the proportion of the added value of the primary industry in gdp (%): the proportion of the added value of the primary industry in gdp is the ratio of the added value of the primary industry to the annual gdp in this year. the primary industry is the core and foundation of national economic development. this index is the foundation of the whole social economy, and it is related to the healthy operation of economic development. 173a prediction method of regional water resources carrying capacity based on artificial neural network proportion of added value of primary industry in gdp added = value of primary industry gdp %×100 (14) (5) the proportion of the added value of the tertiary industry in gdp (%): the proportion of the added value of the tertiary industry in gdp is the ratio of the added value of the tertiary industry to the gdp of the current year. the index can reflect the level and stage of regional economic development. the proportion of added value of tertiary industry in gdp = aadded value of primary industry gdp %×100 (15) (6) gdp growth rate (%): gdp growth is the ratio of gdp added value of this year to gdp of last year. �gdp growth rate gdp growth in this year gdp of last year = ×1000% (16) evaluation index of ecological environment system (1) ecological environment water consumption rate (%): ecological environment water consumption refers to the minimum water consumption required to ensure the benign cycle of the ecosystem. ecological environmental water use rate refers to the ratio of water consumption allocated to the ecological environment to total water consumption. water consumption rate of ecological environment water con = ssumption of ecological environment total water quantity ×1000% (17) (2) forest coverage (%): forest coverage refers to the ratio of forest area to regional land area. the index reflects the regional forest vegetation coverage and is an important index to reflect the operation of the ecosystem as a whole. forest coverage forest coverage total area of regional land = ××100% (18) determination of privacy classification criteria in order to reflect the state of regional water resources carrying capacity scientifically, reasonably, accurately and objectively, the pressure degree of water resources carrying capacity is divided into five grades according to the importance and correlation degree of the above evaluation indexes to the evaluation of regional water resources carrying capacity, and referring to the grading standards of water resources carrying capacity designed by other similar areas, the pressure degree of water resources carrying capacity is divided into five grades, respectively, v1, v2 , v3, v4 and v5. from v1 to v5 means that the carrying capacity of water resources is gradually becoming fragile. where: v1 means that the regional water resources carrying capacity is strong, the regional water resources are relatively rich and the degree of development and utilization is low, which can meet the demand of regional social and economic development for water resources, and the water resources supply situation is good. v2 indicates that the carrying capacity of regional water resources is strong. the regional water resources supply and development have reached a certain degree, but there is still a large space for development and utilization. v3 indicates that the carrying capacity of water resources is medium, and the carrying capacity of water resources is in a medium level, and the water resources and social economy are in a state of coordinated development. v4 indicates that the carrying capacity of water resources in the region is light and weak, which is close to the critical value, and the potential for development and utilization is small. in the long run, problems such as water shortage and water pollution will occur. v5 indicates that the carrying capacity of water resources in the region is weak, indicating that the carrying capacity of water resources has reached the limit or is overloaded, and the development potential is extremely low. therefore, water resources will restrict the development of the social economy, and corresponding countermeasures should be taken. the classification criteria of each index are shown in table 2. analysis and calculation process of neural network model the theory of neural network systems is a frontier research field of artificial intelligence in recent years. it is based on the connection mechanism of large-scale parallel processing and distributed information storage. depending on the connection of a large number of neurons and the different excited states of neurons caused by this connection and the overall behavior of the system, it is closer to the processing mode of the human brain (pelusi et al., 2018). by using the nonlinear mapping relationship of neural networks, the complex table 2. classification standard of evaluation index evaluating index v1 v2 v3 v4 v5 water resources per unit area (10000 m2 / km2) >60 30-60 20-30 10-20 <10 available amount of water resources per capita (m2) >3000 1700-3000 1000-1700 500-1000 <500 development and utilization degree of water resources <10 10-20 20-40 40-60 >60 population density (person / km2) <40 40-80 80-140 140-300 >300 urbanization rate (%) <25 25-40 40-60 60-70 >70 natural growth rate (‰) <2 2-10 10-15 15-20 >20 rural domestic water consumption per capita (l / (day · person)) <50 50-80 80-100 100-120 >120 urban domestic water consumption per capita (l / (day · person)) <150 150-210 210-300 300-400 >400 grain per capita (kg / person) >2000 1000-2000 500-1000 300-500 <300 per capita gdp (10000 yuan / person) <0.4 0.4-0.6 0.6-2.5 2.5-8 >8 lower irrigation rate (%) <10 10-20 20-50 50-60 >60 water consumption per 10000 yuan gdp (m3 / 10000 yuan) <30 30-60 60-100 100-300 >300 proportion of positive value of primary industry in gdp <5 5-12 12-15 15-30 >30 proportion of positive value of tertiary industry in gdp >65 50-65 40-50 30-40 <30 gdp growth rate <5 5-10 10-15 15-20 >20 water consumption rate of ecological environment >5 3-5 2-3 1-2 <1 forest coverage rate (%) >50 40-50 20-40 10-20 <10 174 chaoyang shi, zhen zhang coupling system of water resources carrying capacity is abandoned to find out the inevitable relationship between them, which can avoid the difficulties encountered in finding the relationship between them by other quantitative methods. therefore, neural network theory provides an opportunity for further development of water resources carrying capacity research. the introduction of neural network theory into the field of water resources carrying capacity will promote the development of water resources carrying capacity research. artificial neurons neuron is the basic processing unit of an artificial neural network and a mathematical model of biological neurons in function and structure. it was first proposed by mcculloch-pitts in 1943, so it is also called mp model (todo et al., 2019). the mp model is a multi-input single output nonlinear element, as shown in figure 1. figure 1. mp neuron model suppose that x1, x 2, … x n is the n inputs of neuron i, wji is the binding strength between the neuron and the j-th neuron from other layers, which is called the weight (par & kar, 2019); ui is the sum of inputs of neurons, i is the threshold of neurons, yi is the output of neurons, that is, the output equation of neurons is: u w xi ji i i= −=∑i n 1 θ (19) y f ui i= ( ) (20) where: f ui( ) is a nonlinear function between input and output, which is called excitation function (wang et al., 2019). the commonly used excitation functions can be classified into three types: threshold type, s-type and linear type. the mathematical expressions are as follows: f u u ui i i ( ) = > ≤    1 0 0 0 , , (21) f u ei cui ( ) = + − 1 1 (22) f u ui i( ) = (23) bp neural network model bp network model is the most widely used and successful artificial neural network. in 1989, robert hecht nielson proved that under certain conditions, for any given e < 0, there exists a three-layer neural network, which can approach any nonlinear function with the precision of e-mean square error. therefore, the mapping from arbitrary n-dimensional feature space to m-dimensional feature space can be completed by using a three-layer bp network. bp neural network is a multi-layer forward network with unidirectional propagation, composed of input layer ( ,..., )x xn1 , output layer ( ,..., )y ym1 and one or more hidden layers ( ,..., )h ht1 (figure 2). the input layer consists of n neurons, x i ni ( , , , )= …1 2 is the input; hidden layer consists of t neurons; the output layer consists of m neurons, y g mg ( , , , )= …1 2 is its output. there is no coupling between neurons in the same layer, and the neurons between layers are connected by connection weight w and threshold b (wang et al., 2018; kamimura, 2018). wip, b i n; p tp ( , , , , , , )= … = …1 2 1 2 denotes the connection weight and threshold between the input layer and the hidden layer respectively, and w b p t; g m, pg g ( , , , , , , )= … = …1 2 1 2 denotes the connection weight and threshold between the hidden layer and the output layer, respectively. figure 2. typical topology structure of three layer bp neural network in general, there is no threshold and activation function in the input layer of the bp network, that is, the output of neurons in the input layer is equal to the input of input neurons, while the output of neurons in the hidden layer and output layer must be processed by weight and threshold. the calculation model is as follows: the output of the hidden layer is as follows: z f w x bp h ip i p= +( )∑ (24) where i n= …1 2, , , ; p t= …1 2, , , ; xi is the input value of neurons in the input layer; zp h is the output value of neurons in the hidden layer; wip h and bp is the weight and threshold between the input layer and the hidden layer. the output of the output layer is as follows: z f w z bg y pg p h g= +( )∑ (25) where, p t= …1 2, , , ; g m= …1 2, , , ; w pg h and bg is the weight and threshold between the hidden layer and the output layer. if there is an error between the output layer actually outputs zg y and expected output yi , and the error e is greater than the specified error e0, the error signal is returned along the original path and propagates to the input layer successively by modifying the connection weights between neurons in each layer, and then goes through the forward propagation process. these two processes are repeated until e e≤ 0. at this time, the network learning process will automatically end, and the corresponding network structure and parameter setting will constitute the bp model of the problem. the commonly used error control criterion is the least square error, that is, the energy function is (li & wang, 2018): e y zi g y= −( ) = ∑ 1 2 1 2 i m (26) network learning is based on gradient descent, by adjusting the weight wip and w pg, the formula of weight adjustment is as follows 175a prediction method of regional water resources carrying capacity based on artificial neural network ∆ = −( ) −( ) + ∆ −( ) ∆ = −( ) = w · y z z z y ± w t w · z z pg i g y g y g y › i pg ip p h p h m 1 1 1 1i ∑∑ + ∆ −( )      ´ w ± w tp pg ip 1 (27) where, wip and w pg is the weight between the input layer and the hidden layer, and the weight between the hidden layer and the output layer.  and  are called learning rate and inertia factor, respectively (wen et al., 2019; jiang & li, 2018). when  is large, the convergence is fast but unstable; when  is small, the convergence is slow;  is just opposite to  . model form and step establishment the evaluation index sequence of regional water resources carrying capacity is xij, the evaluation grade sequence is zi , i ...,n1 2, , ; j ...,m1 2, , . n and m represent the capacity of index sequence and the number of evaluation indexes respectively. there are: z f xi ij= ( ) (28) the above formula shows that the input layer of ann model has m units ( ; , ..., )x j mij 1 2 , and the input layer has a unit ( )zi . (1) according to the evaluation grade standard of water resources carrying capacity, a sufficient number of evaluation index sequences are generated by using visual basic programming technology, and the modeling sequence is composed of these index generating sequences and their evaluation grade values. the upper and lower limits of the evaluation index value in the p -th evaluation grade are respectively a j p and b j p , then the random simulation formula of evaluation index value is as follows: x b a *rnd x aij p j p j p j p= −( ) ( )+ (29) (2) standardization of evaluation index in order to eliminate the influence of different dimensions on the evaluation results, the evaluation index sequence is normalized and the following formula is used for data processing: cost indexes: u x xij ij ij / max (30) pressure type index: u x xij ij ij / min (31) (3) the number of hidden layers and units of each hidden layer is selected, and the error limit value e0 between the output and the actual output is given. according to the sequence xij and zi, the neural network is trained. when e e≤ 0, the training is over. the structure parameters, weight and threshold value of the corresponding neural network constitute the artificial neural network model for the evaluation of water resources carrying capacity. prediction of water resources carrying capacity in zhaozhou county based on neural network model the purpose of water resources carrying capacity analysis is to provide guidance and countermeasures for regional development, and the prediction of water resources demand is of great significance for the prediction of regional water resources carrying capacity. in this paper, the neural network model is used to predict the water consumption of the planning year 2025 with 2019 as the base year. selection of prediction indexes as the utilization rate of water resources, water supply modulus, per capita daily domestic water consumption, cultivated land irrigation rate, forest coverage rate, water consumption of 10000 yuan industrial output value and water consumption of 10000 yuan agricultural output value are the main impact factors affecting the carrying capacity of water resources in this area. therefore, it is used to predict the carrying capacity of water resources in the study area in this section. prediction of water resources carrying capacity of zhaozhou county based on bp neural network model preparation of e-learning samples and prediction data according to the statistical yearbook of zhaozhou county and the water resources bulletin of zhaozhou county, the original data needed to calculate the above seven impact factors are calculated, as shown in table 3. through calculation, the data of seven impact factors of water resources carrying capacity in zhaozhou county are obtained, as shown in table 4. where, the utilization rate of water resources = total annual water consumption / annual water resources, water supply modulus = 104 × water supply / area, per capita daily domestic water consumption = 107 × domestic water consumption / (population × 365), and water consumption of 10000 yuan industrial (agricultural) output value = 104 × industrial (agricultural) water consumption / industrial (agricultural) output value. combined with zhaozhou county planning outline and social development science and technology planning, based on some data in zhaozhou county statistical yearbook, and selecting 2019 as the base year, the data of seven impact factors of water resources carrying capacity in zhaozhou county in the planning level year 2025 can be obtained, as shown in table 5. table 3. original data of impact factors of water resources carrying capacity in zhaozhou county (2013-2019). particular year 2019 2018 2017 2016 2015 2014 2013 annual total water consumption (100 million m3) 97.22 94.3182 93.6993 89.9 87.164 84.17 87.67 annual water resources (100 million m3) 834.63 990.999 916.074 1117.5 972.5 1217.46 1206.2 water supply (100 million m3) 97.22 94.3182 93.6993 89.9 87.164 87.17 87.67 area (km2) 176167 176167 176167 176167 176167 176167 176167 domestic water consumption (100 million m3) 12.82 10.5603 10.4997 15.35 15.043 15.11 15.28 population (10000) 3931.12 3878.47 3869.66 3837.27 3798.51 3755.72 3605.81 effective irrigation area (10000 mu) / 1040.2864 1021.065 1007.295 989.72 980.06 965.81 common cultivated land area (10000 mu) / 2723.0128 2625.81 2654.11 2748.42 2756.22 2772.46 forest coverage rate (%) 35.9 34.9 34.9 30.83 30.83 30.83 30.83 industrial water consumption (100 million m3) 28.14 26.9346 26.3888 23.22 20.14 18.88 24.05 industrial output value (100 million yuan) 849.4621 731.7347 600.817 474.68 387.85 348.59 326.03 agricultural water consumption (100 million m3) 54.51 55.87 55.8676 51.34 51.976 50.18 48.34 agricultural output value (100 million yuan) 363.149866 329.170207 298.37 280.83 274.17 270.99 267.57 176 chaoyang shi, zhen zhang table 5. data of seven impact factors of water resources carrying capacity in zhaozhou county in 2025 impact factors base year (2019) planning year (2025) utilization rate of water resources 11.6483 18.35 water supply modulus (10000 m3 / km2) 5.5186 60.269 daily domestic water consumption per capita (l / person / d) 89.3468 200.0000 probability of cultivated land management 38.7854 55.49 forest coverage rate (%) 35.9 40.0000 water consumption of 10000 yuan industrial output value (m3 / 10000 yuan) 331.2685 173.3227 water consumption of 10000 yuan agricultural output value (m3 / 10000 yuan) 1501.0332 1298.0196 where, the utilization rate of water resources, water supply modulus, water consumption of ten thousand yuan industrial output value and ten thousand yuan agricultural output value are predicted and calculated by natural growth rate method; r1, r2, r6 and r7 represents the average annual growth rate of water resources utilization rate, annual average growth rate of water supply modulus, annual average growth rate of water consumption per 10000 yuan of industrial output value and annual growth rate of water consumption of 10000 yuan of output value, respectively; r %1 9 5217 . , r %2 1 7778 . , r %6 12 151=− . , r %7 2 864=− . ; r4 is the average annual growth rate of cultivated land from 2013 to 2019, the value is − 0.147%, and the irrigation area in 2025 is predicted to be 15 million mu. network training process and result test in this paper, seven indexes of water resources utilization rate, water supply modulus, per capita daily domestic water consumption, cultivated land irrigation rate, forest coverage rate, water consumption of ten thousand yuan industrial output value and ten thousand yuan agricultural output value are selected as the input layer of the neural network prediction model of water resources carrying capacity in zhaozhou county, and the water demand is taken as the network output; the hidden layer neuron is 7; therefore, the network topology structure is: 7 7 1  . the seven impact factor values from 2013 to 2018 in table 4 and the water consumption in corresponding years in table 3 are normalized and used as network training samples to train the network. when learning rate η = 0 1. , inertia factor α = 0 2. , error tolerance e ≤ 0 1. , that is, the accuracy is 90% (dos santos et al., 2019); when the number of iterations is 34735, the network reaches the required accuracy, e 0 0999924. , that is, e e≤ 0 . table 4. data of seven impact factors of water resources carrying capacity in zhaozhou county (2013-2019) particular year current year 2019 2018 2017 2016 2015 2014 2013 utilization rate of water resources 11.6483 9.5175 10.2284 8.0447 8.9629 6.9136 7.2683 water supply modulus (10000 m3 / km2) 5.5186 5.3539 5.3188 5.1031 4.9478 4.7779 4.9765 daily domestic water consumption per capita (l / person / d) 89.3468 74.5973 74.3381 109.5956 108.4996 110.2246 116.0988 probability of cultivated land management 38.7854 38.2035 39.0000 37.9524 36.0105 35.5581 34.8358 forest coverage rate (%) 35.9 34.9 34.9 30.83 30.83 30.83 30.83 water consumption of 10000 yuan industrial output value (m3 / 10000 yuan) 331.2685 368.0924 439.2153 489.1717 519.2729 541.6105 737.6622 water consumption of 10000 yuan agricultural output value (m3 / 10000 yuan) 1501.0332 1697.2982 1872.4269 1828.1523 1895.7581 1851.7288 1806.6300 the seven impact factor values in 2018 and 2019 in table 4 and the water consumption in corresponding years in table 4 constitute test samples to test the trained network. the test results are shown in table 6. table 6. prediction test and prediction results of water resources carrying capacity neural network model in zhaozhou county particular year water consumption (100 million m3) learning outcomes actual value error 2012 84.295853 87.67 -0.03849 2013 84.240055 84.17 0.0008317 2014 87.161970 87.164 -0.0000233 2015 89.900511 89.9 0.0000057 2016 93.712300 93.6993 0.0001387 2017 94.218207 94.3182 -0.0010602 2018 94.214266 97.22 -0.0303 2019 94.277907 / / the actual value of annual water consumption in 2019 is 9.722 billion m3, and the prediction value of neural network model is 942742.66 billion m3, with the accuracy of 96.97%; then, the data of the planning level year 2025 is input into the network for prediction, and the prediction result is that the water demand in 2025 will be 94277907 million m3. the change trend chart of predicted and actual water consumption in 2013-2019 is shown in figure 3. figure 3. variation trend of water consumption predicted by neural network model and actual value in 2013-2019 conclusions as an important natural resource, water resource is the key factor to maintain the integrity and virtuous cycle of ecosystem function, and is also the basic resource and strategic economic resource to maintain human survival and 177a prediction method of regional water resources carrying capacity based on artificial neural network social and economic development. in recent years, with the increase of population and the rapid development of social economy, the water problem has become increasingly prominent. as an important part of sustainable development and water resources security strategy research, water resources carrying capacity plays an important role in realizing the harmonious development of economic, social and environmental system, and is also the focus and hot issue in the field of water resources science research. in this paper, the artificial neural network is used to predict the regional water resources carrying capacity (1) based on the analysis of the general information of the study area, the water resources of the region are determined; (2) an evaluation index system of water resources carrying capacity suitable for the study area is constructed to give full play to the role of evaluation indexes; (3) the experimental results show that: this method can effectively predict the carrying capacity of water resources, and the prediction results fit well with the actual data, which has important practical value. to sum up, the prediction method has certain feasibility conditions. using artificial neural network model to predict water resources carrying capacity can obtain more accurate prediction results and lay a data foundation for the followup measures. acknowledgments this work was supported by the humanities and social sciences research projects of henan province education department (no.2021zzjh-432) and the key r & d and the promotion projects of henan province (no. 202102210325). references cheng, k., fu, q., meng, j., li, t. x., & pei, w. (2018). analysis of the spatial variation and identification of factors affecting the water resources carrying capacity based on the cloud model. water resources management, european water resources association (ewra), 32(8), 1-15. https://doi.org/10.1007/s11269-018-1957-x chenini, i., msaddek, m. h., & dlala, m. (2019). hydrogeological characterization and aquifer recharge mapping for groundwater resources management using multicriteria analysis and numerical modeling: a case study from tunisia. journal of african earth sciences, 154(6), 59-69. https://doi.org/10.1016/j.jafrearsci.2019.02.031 dos santos, f. f., pimenta, p. f., lunardi, c., draghetti, l., carro, l., kaeli, & d., rech, p. (2019). analyzing and increasing the reliability of convolutional neural networks on gpus. ieee transactions on reliability, 68(2), 663-677. doi: 10.1109/tr.2018.2878387 guo, l. (2020). design of prediction system for sustainable carrying capacity of regional water resources based on big data. modern electronic technology, 043(009), 117-121. hu, m. j., li, z. j., ding, z. s., zhou, n., & shen, y. (2019). the spatio-temporal heterogeneity and driving mechanism of china’s water transfer from agriculture to non-agriculture. geographical research, 38(06), 15421554. doi: 10.11821/dlyj020171262 jia, j. h., & long, x. j. (2018). research on prediction model of water resources carrying capacity. water resources and hydropower engineering, 49(10), 24-30. jiang, y. t., & li, p. (2018). short-term load forecasting based on improved particle swarm neural network. electrical engineering, 19(02), 87-91. doi: 10.1109/ceect50755.2020.9298636 kamimura, r. (2018). neural self-compressor: collective interpretation by compressing multi-layered neural networks into non-layered networks. neurocomputing, 323(1), 12-36. https://doi.org/10.1016/j. neucom.2018.09.036 li, s. h., & wang, l. (2018). neural network renormalization group. physical review letters, 121(26), 260601.1-260601.7. doi: 10.1103/ physrevlett.121.260601 liu, z. m., zhou, z. z., & wang, y. q. (2019). forecast and analysis of regional water resources carrying capacity based on grey prediction model. journal of yangtze river scientific research institute, 36(9), 34-39. li, w. k., wang, w. l., & li, l. (2018). optimization of water resources utilization by multi-objective moth-flame algorithm. water resources management, 32(10), 3303-3316. https://doi.org/10.1007/s11269018-1992-7 menhas, r., mahmood, s., tanchangya, p., & safdar, m. n. (2019). sustainable development under belt and road initiative: a case study of chinapakistan economic corridor’s socio-economic impact on pakistan. sustainability, 11(21), 1-24. par, s. s., & kar, s. (2019). a hybridized forecasting method based on weight adjustment of neural network using generalized type-2 fuzzy set. international journal of fuzzy systems, 21(1), 308-320. https://doi. org/10.1007/s40815-018-0534-z pelusi, d., mascella, r., tallini l., nayak, j., naik, b., & abraham, a. (2018). neural network and fuzzy system for the tuning of gravitational search algorithm parameters. expert systems with applications, 102(7), 234244. https://doi.org/10.1016/j.eswa.2018.02.026 todo, y., tang, z., todo, h., ji, j., & yamashita, k. (2019). neurons with multiplicative interactions of nonlinear synapses. international journal of neural systems, 29(08), 115-133. doi:10.1142/ s0129065719500126 wang, l., xiong, y. n., li, s. w., & zeng, y. r. (2019). new fruit fly optimization algorithm with joint search strategies for function optimization problems. knowledge-based systems, 176(7), 77-96. https://doi.org/10.1016/j.knosys.2019.03.028 wang, l., zhao, h. y., & sha, c. l. (2018). dynamical stability in a delayed neural network with reaction–diffusion and coupling. nonlinear dynamics, 92(3), 1197-1215. https://doi.org/10.1007/s11071-0184118-0 wen, s., xiao, s., yang, y., yan, z., zeng, z., & huang, t. (2019). adjusting learning rate of memristor-based multilayer neural networks via fuzzy method. ieee transactions on computer-aided design of integrated circuits and systems, 38(6), 1084-1094. doi: 10.1109/ tcad.2018.2834436 xu, x., zhang, y., & chen, y. (2020). projecting china’s future water footprint under the shared socio-economic pathways. journal of environmental management, 260(4), 110102.1-110102.13. doi:10.1016/j. jenvman.2020.110102 zhu, l., li, x., bai, y., yi, t., & yao, l. (2019). evaluation of water resources carrying capacity and its obstruction factor analysis: a case study of hubei province, china. water, 11(12), 2573.1-2573.14. doi:10.3390/ w11122573 https://doi.org/10.1007/s11269-018-1957-x https://doi.org/10.1016/j.jafrearsci.2019.02.031 https://doi.org/10.1109/tr.2018.2878387 https://doi.org/10.1109/ceect50755.2020.9298636 https://doi.org/10.1016/j.neucom.2018.09.036 https://doi.org/10.1016/j.neucom.2018.09.036 https://doi.org/10.1007/s11269-018-1992-7 https://doi.org/10.1007/s11269-018-1992-7 https://doi.org/10.1007/s40815-018-0534-z https://doi.org/10.1007/s40815-018-0534-z https://doi.org/10.1016/j.eswa.2018.02.026 https://doi.org/10.1016/j.knosys.2019.03.028 https://doi.org/10.1007/s11071-018-4118-0 https://doi.org/10.1007/s11071-018-4118-0 articulo 3c.indd earth sciences research journal seismology earth sci. res. sj. vol. 16, no. 2 (december, 2012): 103 108 introduction since fractal description was developed for the geometry of natural objects (mandelbrot b.b., 1982), it has been recognised that many complex space-time phenomena, such as seismicity, may be described and interpreted in terms of fractal distributions with power-law scaling (e.g. hirata t., 1989; öncel a.o. et al., 1995; caneva a., smirnov v., 2004; öncel a.o., wilson t.h., 2007; roy s. et al., 2011). however, studies of possible correlation between seismicity and fault distribution are limited. applying seismic hazard assessment to fractal association between seismotectonic variables requires distinguishing normal and anomalous correlations between such data sets’ fractal attributes. since the statistical behaviour of seismicity may potentially represent sensitive short-term predictors of major earthquakes, this study’s main aim was to find an empirical relationship between seismic b-values and the fractal distribution of epicentres for turkish earthquakes. data regarding earthquakes and tectonic zoning several earthquake catalogues are available for turkey from both national and international sources. earthquakes occurring between 1970 and 2006 were taken from öztürk’s catalogue of instrumental data statistical correlation between b-value and fractal dimension regarding turkish epicentre distribution serkan öztürk gümüşhane university. turkey. e-mail: serkanozturk@gumushane.edu.tr abstract this study was aimed at analysing the relationship between seismic b-value and fractal dimension dcvalue for turkish epicentres. the earthquake catalogue consisting of 99,737 instrumentally-registered turkish events was analysed for the period between 1970 and 2011; turkey was divided into 55 tectonic zones for making a detailed comparison. the b-values were calculated by the maximum likelihood method and the dc-values were obtained with 95% confidence limits by linear regression. the results showed that higher dc-values were associated with lower b-values and this could have been an indication of relatively high stress intensity and stronger epicentre clustering in these regions. orthogonal regression was used to estimate a suitable statistical correlation between two seismotectonic parameters; the dc = 2.44 0.30*b relationship was obtained with a strong negative correlation (r = -0.82) between b-value and dc-value for turkish earthquake distribution. this seemed to agree with other regional results obtained for different parts of turkey and the rest of the world. resumen este estudio tuvo como objetivo analizar la relación entre el valor b y la dimensión del valor fractal dc para los epicentros de turquia. el catálogo sísmico que consiste en 99.737 eventos registrados instrumentalmente fue analizado para el período comprendido entre 1970 y 2011; turquía se dividió en 55 zonas tectónicas para hacer una comparación detallada. los valores de b se calcularon por el método de máxima verosimilitud y la dc-valores se obtuvieron con límites de confianza del 95% mediante regresión lineal. los resultados mostraron que el aumento del valor fractal dc se asocia con menores valores b y esto podría haber sido un indicador de intensidad de tensiones relativamente altas y una más fuerte agrupación de epicentro en estas regiones. se utilizó una regresión ortogonal para estimar la correlación estadística adecuada entre dos parámetros sismotectónicos la relación de la dimensión fractal dc = 2,44 a 0,30*b, se obtuvo con una fuerte correlación negativa (r = -0,82) entre valores b y valores fractales dc para la distribución de terremotos en turquía. esto parece estar de acuerdo con otros resultados regionales obtenidos para las diferentes partes de turquía y el resto del mundo. palabras claves: terremoto de turquía, dimensión fractal, valor-b, correlación. keywords: turkish earthquake, fractal dimension, b-value, correlation. record manuscript received: 13/02/2012 accepted for publications: 06/05/2012 serkan öztürk104 (2009), including duration magnitude (m d ) for 73,530 earthquakes occurring in turkey between 1970 and 2006; öztürk used empirical relationships to compile a homogenous and complete earthquake catalogue (1970 to 2006). the bogazici university’s kandilli observatory and research institute (koeri) catalogue was also used for 2006 to 2011; koeri usually gives local magnitudes (m l ) for local earthquakes having missing m d . if an m d was unknown in the koeri catalogue for 2006 to 2011, then m d were calculated using the relationships given in öztürk (2009). 26,207 earthquakes in and around turkey were thus selected for the aforementioned period. this earthquake catalogue was homogeneous for m d and contained 99,737 earthquakes occurring between 1970 and 2011. many authors have suggested tectonic zoning as a widely-used methodology for evaluating the hazard of an earthquake occurring (e.g. erdik m. et al., 1999; jiménez m. et al., 2001; bayrak y. et al., 2009). the present study’s new seismic source zones for turkey and its adjacent areas were based on tectonic zoning studies by erdik (1999) and bayrak (2009), the existing tectonic structure and earthquake epicentre distribution (figure 1). turkey was divided into 55 different zones, as shown in figure 1. new smaller zones were selected to compare such tectonic zoning in detail, in the same regions. figure 1 shows turkey’s active tectonics and many details regarding turkey’s tectonic structure can be found in şaroğlu (1992), mcclusky (2000) and bozkurt (2001). the numbers of earthquakes occurring in each zone were sufficient for analysis; table 1 shows the seismic zones in this study (numbered 1 to 55) with their tectonic environments. earthquakes’ frequency-magnitude distribution (seismic b-value) earthquakes’ magnitude distribution (md) is usually parameterised using gutenberg-richter’s (g-r) power law relationship (gutenberg b., richter c.f., 1944); such frequency-magnitude relationship has been found to be applicable (in simplified form) as follows: log 10 n(m) = a – bm (1) where n(m) is the cumulative number of events having a magnitude greater than m, b describes the slope of the size distribution of events and a is proportional to the productivity of a volume, or the seismicity rate. the b-value is one of the most important statistical parameters for describing the size scaling properties of seismicity; b-values change figure 1. tectonic areas and active fault systems in turkey. tectonic structures have been modified from şaroğlu f. et al., (1992) and bozkurt e. (2001). roughly in the range of 0.3 to 2.0, depending on the different regions. however, average regional scale estimates of b-value are usually equal to 1 (frohlich c., davis s., 1993). many factors can cause perturbation of average b-value (b~1.0); regions having lower b-values are probably regions subjected to higher applied shear stress after the main shock, whereas regions having higher b-values are areas which have experienced slip. high b-values are reported from areas having increased geological complexity, indicating the importance of multi-fracture areas; a low bvalue is thus related to a low degree of heterogeneity of cracked medium, large stress and strain, high deformation speed and large faults (bayrak y., öztürk s., 2004). many methods can be used for calculating any region’s b-value but the maximum likelihood method is the most robust and widely-accepted method for estimating b-values (aki k., 1965): b = 2.303 /(m — – m min + 0.005) (2) where m — is average magnitude value and m min is minimum magnitude value. 0.05 in equation (2) is a correction constant. if b=1 and m min=1 is used, maverage=3.25 will be obtained. however, this is an extreme mmin value. mmin=1 did not occur in previous turkish earthquake catalogues; mmin in turkish earthquake catalogues was around 3.0 until the beginning of the 2000s. many stations have been built in turkey, especially after two destructive earthquakes in 1999, minimum earthquake magnitude now being seen to be around 1.5 (average value may sometimes be higher than 3.25 which is not high value for turkish earthquakes). the standard deviation of seismic b-value (95% confidence limit) may be determined using the equation suggested by aki (1965) as  1.96b / √n, where n is the number of earthquakes used to make the estimate. this would yield ±0.1–0.2 confidence limits regarding b-value for a typical sample consisting of n=100 earthquakes. such sample consisting of n=100 events represents a specific calculation. the number of earthquakes in this study was 97,737; table 1 shows that sometimes 110 earthquakes were used (in region 13) or 10,328 earthquakes on other occasions (in region 34). this value for n=100 earthquakes is thus a typical sample for the aforementioned calculation. fractal dimension (correlation dimension, dc) of epicentres’ spatial distribution earthquake distribution spatial patterns and temporal patterns of occurrence were demonstrated to be fractal using a two-point correlation dimension (dc). analysing correlation dimension is a powerful tool for quantifying a geometrical object’s self-similarity. grassberger and procaccia (1983) defined dc and correlation sum c(r) as follows: dc = lim[logc(r)/logr] (3) r  0 c(r) = 2nr<r / n(n – 1) (4) where c(r) is the correlation function, r is the distance between two epicentres and n is the number of pairs of events separated by distance r<r. if the epicentre distribution has a fractal structure, the following relationship would be obtained: c(r)  rdc (5) where dc is the fractal dimension (more strictly, the correlation dimension). distance r between two earthquakes would be calculated (in degrees) from: r = cos–1(cosi cosj + sini sinj cos (fi – fj)) (6) statistical correlation between b-value and fractal dimension regarding turkish epicentre distribution 105 table 1. seismotectonic parameters b-value and dc-value, as well as the number of earthquakes occurring in entire seismic areas of turkey. region tectonics earthquake number b-value dc-value 1 north east anatolian fault zone (neafz) (horosan fault, dumlu and çobandede fault zones) 338 0.62±0.03 2.26±0.09 2 1234 1.05±0.03 2.08±0.07 3 ağrı, tutak, balıklıgöl and kağızman faults (atbkf) 219 0.86±0.06 2.30±0.04 4 iğdır, doğubeyazıt and çaldıran faults (idçf) 268 0.63±0.03 2.23±0.02 5 başkale, erciş, muradiye and süphan faults (bemsf) and hasan timur fault zone (htfz) 159 0.98±0.09 2.09±0.03 6 350 0.95±0.05 2.12±0.03 7 malazgirt and bulanık faults (mbf) 320 1.13±0.07 2.08±0.04 8 bitlis-zagros thrust zone (bztz) (kavakbaşı fault, muş thrust zone and yüksekova-şemdinli fault zone) 442 1.05±0.06 2.08±0.04 9 238 0.77±0.05 2.28±0.01 10 116 1.21±0.10 2.03±0.01 11 254 0.63±0.04 2.28±0.05 12 251 0.83±0.06 2.25±0.05 13 karacadağ extension zone (kez) 110 1.17±0.04 2.00±0.03 14 510 1.49±0.08 1.94±0.04 15 east anatolian fault zone (eafz) 2272 0.96±0.02 2.11±0.06 16 junction of a part of dead sea fault and eafz 298 0.84±0.05 2.31±0.06 17 1118 0.93±0.04 2.12±0.03 18 northern part of cyprus 179 1.61±0.05 2.00±0.03 19 southern part of cyprus, including eastern part of cyprus arc 369 1.20±0.07 2.07±0.02 20 340 1.28±0.05 1.99±0.04 21 western part of cyprus arc 350 1.05±0.07 2.10±0.03 22 1161 1.15±0.05 2.10±0.02 23 acıgöl, dinar and çivril faults (adçf) 2154 1.36±0.09 2.06±0.04 24 muğla and rhodes region 1534 1.64±0.09 1.94±0.04 25 5392 1.38±0.09 1.97±0.03 26 aegean arc 1914 1.38±0.09 2.21±0.03 27 burdur fault zone (bfz) 1787 1.01±0.02 2.15±0.05 28 büyük and küçük menderes grabens (bkmg) 3428 1.24±0.08 2.02±0.04 29 1408 1.34±0.07 2.08±0.04 30 aliağa and dumlupınar faults (adf) and zeytindağ-bergama faults (zbf) 926 1.30±0.07 2.08±0.04 31 3770 1.13±0.02 2.10±0.07 32 alaşehir and gediz grabens (agg) 1782 1.24±0.06 2.03±0.02 33 sultandağı, beyşehir and tatarlı faults (sbtf) 3252 1.35±0.05 2.05±0.04 34 kütahya and simav grabens (ksg) 10328 1.15±0.01 2.08±0.05 35 8036 1.62±0.04 2.03±0.05 36 soma and bakırçay grabens (sbg) 6422 1.93±0.05 1.95±0.07 37 1710 1.65±0.08 2.04±0.05 38 1168 1.18±0.04 2.06±0.05 39 eskişehir, i̇nönü-dodurga and kaymaz faults (ei̇dkf) 2527 1.41±0.03 1.98±0.05 40 manyas and ulubat faults (muf) 1432 1.44±0.03 2.04±0.04 41 yenice-gönen and sarıköy faults (ygsf) 4220 1.41±0.04 2.03±0.04 42 etili fault (ef) 1398 1.65±0.07 1.92±0.04 serkan öztürk106 where (i,fi) and (j,fj) are the latitudes and longitudes of the i th and jth events, respectively (hirata t., 1989). by plotting c(r) against r on a double logarithmic coordinate, fractal dimension dc would be obtained from the slope of the graph. fractal analysis is often used to quantify size scaling attributes and seismotectonic variables’ clustering properties. fractal dimension dc can be calculated to determine possible unbroken sites and such unbroken sites have been suggested as being potential seismic gaps to be broken in the future. variations in fractal correlation dimension mainly depend on the complexity or quantitative measurement of the degree of heterogeneity of seismic activity in a fault system. in areas of increased complexity in an active fault system (higher dc) associated with lower b-value, stress release occurs on fault planes having smaller surface area (öncel a.o., wilson t.h., 2002). results and discussion an investigation of seismotectonic parameters in and around turkey involved using the b-value of the g-r relationship (equation 1) and fractal correlation dimension dc-value of turkish earthquake distribution from 1970 to 2011; turkey was divided into 55 seismic source zones in an attempt to find an empirical relationship between b-value and dc-value. zmap software was used for calculating b and dc-values for each region (wiemer s., 2001). the b-values were calculated by using the maximum likelihood method (95% confidence limit) because it yields a more robust estimate than the least-square regression method (aki k., 1965). the dcvalues for all turkish areas were obtained with 95% confidence limits by linear regression. table 1 shows the results for two seismotectonic parameters with their standard deviations as well as the number of earthquakes for entire tectonic areas. the first seismic parameter b-value ranged from 0.62 to 1.93 for 55 turkish areas. b-values smaller than 0.8 were found in regions 1, 4, 9 and 11, including the north-eastern anatolian fault zone (neafz) including the dumlu and çobandede fault zones, iğdır, doğubeyazıt and çaldıran faults (idçf) and two parts of the bitlis-zagros thrust zone (bztz). such low values may have depended on the position of these regions neighbouring the easternmost part of the north anatolian fault and the bztz. the conjugate strike-slip fault system related to these areas dominates eastern anatolia’s active tectonics and thrust faults can produce destructive earthquakes in the bitlis thrust which forms a complex continent-continent and continent-ocean collision boundary (bozkurt e, 2001). these systems generate major earthquakes, such as patnos in 1903 (m s=6.3), pasinler in 1924 (ms=6.8), lice-diyarbakır in 1975 (ms=6.6), çaldıran in 1976 (ms=7.5) and horasan in 1983 (ms=6.8). global positioning system (gps) data gives 10±2 mm/yr for total shortening between the strike slip faults in eastern turkey and thrusting along the caucasus. the bztz’s north-western motion is 18±2 mm/yr and such motion is related to that of eurasia (mcclusky s. et al., 2000); small b-values in these regions are thus related to a low degree of heterogeneity and large faults resulting in strong earthquakes. the other small b-values changing between 0.8 and 1.0 were observed in regions 3, 5, 6, 12, 15, 16, 17 and 50 covering the ağrı, tutak, balıklıgöl and kağızman faults (atbkf), the başkale, erciş, muradiye and süphan faults (bemsf) and the hasan timur fault zone (htfz), the western part of bztz, the east anatolian fault zone (eafz), the junction of a part of the dead sea fault and eafz, the eastern part of the north anatolian fault zone (enafz) including the bingöl-karakoçan and sancakuzunpınar fault zones. table 1 shows that regions 15, 16 and 17 are connected with the east eafz which is not as seismically active as the nafz. unlike nafz, the eafz in region 15 has been relatively quiescent in the instrumental period compared to the historical period (mcclusky s. et al., 2000). the data used in this study only included the instrumental period for earthquakes occurring from 1970 to 2011. according to the fault slip rate and observed seismicity during recent years, relatively larger b-values were observed in these regions than in regions 1, 4, 9 and 11. according to scholz (1968), low b-values indicate that the state of stress is high. special interest should thus be given to whole regions where low b-values have been observed, especially after the occurrence of the m w=6.0 elazığ earthquake (in region 15) on 8th march 2010 and the mw=7.2 van lake earthquake (in region 6) on 23rd september 2011. regions such as 2, 8, 21, 27, 46, 48, 49 and 52 have almost the similar b-values (very close to 1.0); these regions are related to the neafz, the eastern part of bztz, the western part of the cyprus arc, the burdur fault zone (bfz), the düzce fault (df), the yağmurlu-ezinepazarı fault zone (yefz) the another part of the enafz and sürgü fault. the maximum b-value was estimated as being 1.93 in region 36, such region lying around the soma and bakırçay grabens. other largest b-values greater than 1.5 were calculated in regions 18, 24, 35, 37 and 42; these zones were covered by the northern part of cyprus, the muğla and rhodes region, the kütahya and simav grabens (ksg) and the etili fault (ef). the b-values obtained for the rest of the zones were between 1.1 and 1.5. the nafz is a very active structure and, according to geodesy, accommodates 24±2 mm/yr of dextral motion (mcclusky s. et al., 2000). two large earthquakes occurred in regions 44 and 46 during august and november 1999 and, of course, the stress level in these region tectonics earthquake number b-value dc-value 43 marmara part of north anatolian fault zone (mnafz) 1378 1.10±0.07 2.10±0.06 44 7859 1.26±0.02 2.05±0.03 45 1430 1.35±0.07 2.05±0.03 46 düzce fault (df) 2011 1.02±0.03 2.14±0.05 47 ismetpaşa segment (is) 1684 1.38±0.09 2.04±0.05 48 yağmurlu-ezinepazarı fault zone (yefz) 1369 1.04±0.03 2.10±0.04 49 eastern part of north anatolian fault zone (enafz) (bingöl-karakoçan, sancak-uzunpınar fault zones) 794 1.03±0.05 2.13±0.03 50 2236 0.87±0.07 2.28±0.06 51 ovacık and malatya fault zone (omfz) 1340 1.20±0.04 2.05±0.03 52 sürgü fault (sf) 577 1.06±0.05 2.06±0.03 53 central anatolian fault zone (cafz) (yakapınar-göksun and yıldızeli fault zones ) 659 1.27±0.05 1.99±0.03 54 738 1.25±0.04 2.02±0.03 55 tuzgölü fault zone (tfz) 2178 1.19±0.03 2.03±0.06 statistical correlation between b-value and fractal dimension regarding turkish epicentre distribution 107 regions would thus not have been so high in recent years. the regions characterised by large b-values had a greater proportion of low magnitude earthquakes whereas regions having low b-values represented areas in which large magnitude earthquakes occurred. b-values estimated using the maximum likelihood approach for the g-r method seemed to have a good relationship to the tectonics and level of seismicity. the second seismotectonic parameter dc-value ranged from 1.92-2.31 for the 55 tectonic source regions in turkey and the surrounding areas. dcvalues smaller than 2.0 were estimated in regions 14, 20, 24, 25, 36, 39, 42 and 53. these zones were related to the karacadağ extension zone (kez), the southern part of cyprus including the eastern part of the cyprus arc, the muğla and rhodes region, the sbg, the eskişehir, i̇nönü-dodurga and kaymaz faults (ei̇dkf), the ef, the central anatolian fault zone (cafz) including the yıldızeli fault zone. dc-values greater than 2.2 were observed in regions 1, 3, 4, 9, 11, 12, 16, 26 and 50 which were related to the neafz including the dumlu and çobandede fault zones, the atbkf, the idçf, the bztz, the junction of part of the dead sea fault and the eafz, the aegean arc and the enafz. dc-values ranged from 2.0 to 2.2 in the rest of the seismic source areas. the results obtained from analysis of data along the nafz (regions 43-49) suggested that epicentre distribution became less clustered (higher dc) as the probability of larger magnitude earthquakes became smaller (larger b-value). this implied greater fracture toughness in the central portions of the nafz. as stated in öncel and wilson (2002), stress becomes released on fault planes having smaller surface area in regions of increased complexity in an active fault system (higher dc) associated with lower b-values. this means that areas in a fault network characterised by higher dc-values would be associated with greater complexity in a fault pattern and the persistence of such complexity on smaller scales. higher order fractal dimension (especially greater than 2.0) is increasingly sensitive to heterogeneity in magnitude distribution, suggesting that seismicity is more clustered at larger scales (or in smaller areas) in these fault zones. it may also be related to the number of events to a certain extent. since the uniform distribution of earthquakes decreases with an increase in the clustering of events, it is reasonable to assume that higher dc values (≥2.2) and lower bvalues (0.8≤) are the dominant structural feature in the study area and may have arisen due to clusters. correlation between b and dc-value this study was aimed at determining a statistical relationship between seismic b-values and fractal dimension dc-value for turkish epicentres. the orthogonal regression (or) method (e.g, carroll r.j., ruppert d., 1996) was used to find the most suitable correlation between both seismotectonic parameters selected here. as the standard least squares method figure 2. dc-value compared to b-value for turkish earthquakes occurring between 1970 and 2011. the statistical relationship (with its coefficient of correlation) is also given is based on the assumption that horizontal axis values are estimated without error (carroll r.j., ruppert d., 1996), or was applied in fitting the relationship. figure 2 gives a graphical representations of or fit between b and dc-values for turkish earthquake epicentres. figure 2 shows that the correlation coefficient of regression fit was very strong (r=-0.82). linear fit was used for regression and the following equation was derived: dc = 2.44 – 0.30*b (7) the correlation between fractal dimension and b-value has been studied in several parts of the world (e.g. aki k., 1981; hirata t., 1989; henderson j. et al., 1992; caneva a., smirnov v., 2004; roy s. et al., 2011) and particularly the turkish region (e.g. öncel a.o. et al., 1995 and 1996; öncel a.o., wilson t.h., 2002 and 2007). since aki (1981) proposed a simple relationship between b-value and fractal dimension having d=2b positive co-relationship, both positive (e.g. öncel a.o., wilson t.h., 2004; roy s. et al., 2011) and negative co-relationships (e.g. hirata t., 1989; henderson j. et al., 1992; öncel a.o. et al., 1995 and 1996) between these two scaling exponents have been reported and debated in the pertinent literature. such co-relationships could even change from negative to positive (e.g. öncel a.o., wilson t.h., 2002 and 2007). hirata’s results (1989) did not support aki’s assumption that d=2b but showed, on the contrary, a negative co-relationship dc=2.3-0.73*b (r=-0.77) between b and fractal dimension of epicentres in the tohoku region of japan. henderson (1992) obtained a similar result for the riverside catalogue in southern california. similarly, a study of seismicity in the nafz, turkey, revealed a long-term negative co-relationship between b and dc (öncel a.o. et al., 1995). the b-value was found to be weakly negatively correlated with fractal dimension dc=2.74-1.52*b (r=-0.56) for the nafz (including the northern aegean sea) by öncel (1995). öncel (1996) has also observed a strong negative correlation (r=0.85) between dc and b-values as dc=2.32-1.09*b was associated with the nafz. by contrast, öncel and wilson (2002) found weak positive correlation (r=0.48) between variations in b and dc in the western nafz. however, variability between b and dc along the length of the fault zone in this study revealed divergence between b and dc in the central nafz and spatial variation yielded a strong negative correlation (r=-0.85) between b and dc. analysis presented in öncel and wilson (2004) revealed a strong positive correlation (r=0.81) between dc and b-values along the nafz preceding the 1999 izmit earthquake. öncel and wilson (2007) observed a strong positive co-relationship between dc and b-value during 1992-1994 (r=0.84) and 1996-1998 (r=0.94) and negative correlation (r=-0.71) extending from mid-1994 to mid-1996 in north-western turkey. table 2 gives previous studies on the b-value and dc-value relationship. reference relation aki k., (1981) d=2*b hirata t., (1989) dc=2.3-0.73*b öncel a.o. et al., (1995) dc=2.74-1.52*b öncel a.o. et al., (1996) dc=2.32-1.09*b öncel a.o., wilson t.h., (2002) positive correlation öncel a.o., wilson t.h., (2004 positive correlation öncel a.o., wilson t.h., (2007) positive and negative correlation table 2. some examples of previous studies of the relationship between b-value and dc-value. serkan öztürk108 conclusions this study tried to estimate a suitable and reliable correlation between two seismotectonic parameters: b and dc-values for turkish earthquakes. statistical analysis of the available data included 99,737 earthquakes from 1970 to 2011. the maximum likelihood method was used for calculating b-values and the linear regression technique for obtaining dc-values (95% confidence limit). it was observed that the largest events were associated with low b and high dc-values, respectively, implying relatively high stress intensity and stronger epicentre clustering. orthogonal regression was used for correlating the chosen seismotectonic parameters. the results showed strong negative correlation between b-value and fractal dimension dc-value for turkish earthquakes dc = 2.44 0.30*b (r=-0.82) given by or. the results had good agreement with previous studies for different parts of turkey and the rest of the world. acknowledgements the author would like to thank to prof. dr. stefan weimer for providing zmap software and anonymous reviewers for their useful and constructive suggestions in improving this paper. i am grateful to dr. doğan kalafat (koeri) for providing us with the earthquake catalogue. i would also like to thank koeri for providing an earthquake database via internet. references aki k. (1965): “maximum likelihood estimate of b in the formula log n = a – bm and its confidence limits”. bull earthq res inst tokyo univ. 43, 237-239. aki k. (1981): “earthquake prediction”. am geophys union. vol 4. washington, pp 566-574. bozkurt e. (2001): “neotectonics of turkey -a synthesis”. geodinamica acta. 14, 3-30. bayrak y., öztürk s. (2004): “spatial and temporal variations of the aftershock sequences of the 1999 i̇zmit and düzce earthquake”. earth planets space. 56, 933-944. bayrak y., öztürk s., çınar h., kalafat d., tsapanos t.m., koravos g.ch., leventakis g.a. (2009): “estimating earthquake hazard parameters from instrumental data for different regions in and around turkey”. eng geol. 105, 200-210, carrol r.j., ruppert d. (1996): “the use and misuse of orthogonal regression estimation in linear errors-in-variables models”. the american statistician. 50, 1-6. caneva a., smirnov v. (2004): “using the fractal dimension of earthquake distributions and slope of the reccurence curve to forecast earthquakes in colombia”. earth sci res j. 8(1), 3-9. erdik m., alpay b.y., onur t., sesetyan k., birgoren g. (1999): “assessment of earthquake hazard in turkey and neighboring regions”. annali di geofisica. 42, 1125-1138. frohlich c., davis s. (1993): “teleseismic b-values: or, much ado about 1.0”. journal of geophysical research 98 (b1), 631-644. grassberger p., procaccia i. (1983): “measuring the strangeness of strange attractors”. physica 9(d), 189-208. gutenberg r., richter c.f. (1944): “frequency of earthquakes in california”. bull seismol soc am. 34, 185-188. henderson j., main i.g., meredith p.g., sammonds p.r. (1992): “the evolution of seismicity-observation, experiment and a fracturemechanical interpretation”. j struct geol. 14, 905-913. hirata t. (1989): “correlation between the b-value and the fractal dimension of earthquakes”. j geophys res. 94, 7507-7514. jiménez m.j., giardini d., grünthal g., sesame working group (erdik m., garcíafernández m., lapajne j., makropoulos k., muson r., papaioannou ch., rebez a., riad s., sellami s., shapira a., slejko d., van eck t., el sayed a.) (2001): “unified seismic hazard modelling throughout the mediterranean region1. bolletino di geophysica teorica ed applicata. 42, (1-2), 3-18. mandelbrot b.b. (1982): “the fractal geometry of nature”. freeman press, san francisco. mcclusky s., balassanian s., barka a., demir c., ergintav s., georgiev i., gürkan o., hamburger m., kahle k.h.h., kastens k., kekelidze g., king r., kotzev v., lenk o., mahmoud s., mishin a., nadariya m., ouzounis a., paradissis d., peter y., prilepin m., reilinger r., şanlı i., seeger h., tealeb a., toksöz m.n., veis g. (2000): “global positioning system constraints on plate kinematics and dynamics in the eastern mediterranean and caucasus”. j geophys res. 105 (b3), 5695-5719. öncel a.o., alptekin ö., main i.g. (1995): “temporal variations of the fractal properties of seismicity in the western part of the north anatolian fault zone: possible artifacts due to improvements in station coverage”. nonlinear processes geophys. 2, 147-157. öncel a.o., main i.g., alptekin ö. cowie p.a. (1996): “temporal variations of the fractal properties of seismicity in the north anatolian fault zone between 31°e and 41°e”. pure appl geophys. 146, 148-159. öncel a.o., wilson t.h. (2002): “space-time correlations of seismotectonic parameters: examples from japan and from turkey preceding the izmit earthquake”. bull seism soc am. 92(1), 339-349. öncel a.o., wilson t.h. (2004): “correlation of seismotectonic variables and gps strainmeasurements inwestern turkey”. j geophys res. 109 (b11), b11306. öncel a.o., wilson t.h. (2007): “anomalous seismicity preceding the 1999 izmit event, nw turkey”. ggeophys j int. doi: 10.1111/j.1365246x.2006.03298.x. öztürk s. (2009): “an application of the earthquake hazard and aftershock probability evaluation methods to turkey earthquakes”. doctoral dissertation, karadeniz technical university, trabzon, turkey (in turkish with english abstract). roy s., ghosh u., hazra s., kayal j.r. (2011): “fractal dimension and bvalue mapping in the andaman-sumatra subduction zone”. nat hazards. 57, 27–37. şaroğlu f., emre o., kuşcu i. (1992): “active fault map of turkey”. general directorate of mineral research and exploration, ankara, turkey. scholz c.h. (1968): “the frequency-magnitude relation of microfracturing in rock and its relation to earthquakes”. bull seism soc am. 58, 399-415. wiemer s. (2001): “a software package to analyze seismicity: zmap”. seismol res lett. 72(3), 373-382. search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 reconstruction of an ancient volcanic edifice near to paipa (boyac, colombia) by means of computer-assisted models for pyroclastic flows earth sciences research journal earth sci. res. j. vol. 8, no. 1 (dec. 2004): 56 – 62. manuscript received may 2004. paper accepted october 2004. 56 pyroclastic flow modeling to reconstruct a volcanic edifice in paipa (boyacá-colombia). óscar rodríguez 1, milton obando1, luis castillo1 and héctor cepeda2. 1 departamento de geociencias, universidad nacional de colombia, bogotá. e mail: lacastillol@unal.edu.co 2 ingeominas, bogotá. e-mail: hcepeda@ingeomin.gov.co abstract pyroclastic deposits produced by the domes collapse (resurgence of a caldera collapse), at the west of the honda grande creek (paipa, boyacá-colombia) were related by ingeominas. these deposits fill the valleys of olitas, calderitas and a creek at the south of the alto de los volcanes reaching distances near to 3 km from the focus between the alto de los volcanes and el mirador hill. the flows were modeled using 3d software (sheridan and kover, 1996). a volcanic simulation was done obtaining the height and morphology of the volcanic edifice before the collapse during the last eruptive event. keywords: volcanic simulation, pyroclastic flows, computers model resumen depósitos producidos por el colapso de domos (resurgencia del colapso de la caldera), fueron relacionados por ingeominas al oeste de la ensenada honda grande (paipa, boyacácolombia -). estos depósitos llenaron los valles de "olitas" y calderitas, como también la ensenada ubicada al sur del alto de los volcanes, alcanzando distancias de 3 kilómetros desde el foco en un sector entre el alto de los volcanes y el mirador hill. las flujos se modelaron usando software 3d (sheridan y kover, 1996). la simulación del evento volcánico suministró la altura y la morfología que tenía el edificio antes del colapso en el último evento eruptivo palabras clave: flujos piroclásticos, modelos por computador, simulación volcánica. © 2004 esrj unibiblos. pyroclastic flow modeling to reconstruct a volcanic edifice in paipa (boyacá-colombia). introduction in paipa town (boyacá-colombia), a volcanic caldera was discovered and defined by velandia et al. (2004) and also pardo (2004). after the caldera collapsed a volcanism was produced over the cordillera oriental, and pyroclastic flows conformed by pumice and ash ignimbrite were generated. these deposits represents a first volcanic episode. then a new volcanic edifice was formed, representing a volcanic resurgence episode. their products are block and ash pyroclastic flows formed during the dome collapse. pyroclastic flow deposits were mapped and detailed to reconstruct a volcanic edifice generated during the resurgence. the formation conditions were simulated using the flow3d software. the models run many topographic reconstruction until the simulation results were coherent with dispersion of the volcanic products observed in the field, i.e. dome collapse pyroclastic flow. geological setting paipa area is over the cordillera oriental, 120 km north of bogotá d.c. southeastern side of paipa, at boyacá department, between the coordinates x = 1103000e and 1113000e and y = 1116000n and 1127000n, utm bogotá origin. (figure 1) in the area there are cretaceous, paleogene, neogene rocks and sediments corresponding to quaternary deposits. cretaceous older rocks in the interest zone are from the upper cretaceous, which include une formation, churuvita group, conejo, plaenners, arenisca de labor, arenisca tierna and guaduas formation. these units are constituted by a thick section of sedimentary bodies with sandstone layers and shale intercalations. the last formation is characterized by coal seam. cenozoic it consist of volcanic and sedimentary deposits. some of the volcanic deposits are: el volador unit (neogene), paipa volcanic unit and piroclastic rocks (guarrus section, tobas del garrus divided in a1 and a2 deposits, tephras el guarrus deposits: b, c; olitas creek section: volcanic domo, pyroclastic 1,2 flows). the sedimentary sequences are fluvial and alluvial deposits: conglomeratic rocks and sandstone planar and cross structures. two geologic sections: olitas and caldera creek were measured in formed by pyroclastic flows unconformities on the cretaceous formation. figure 1. survey area location 57 oscar rodríguez et al. 58 fundaments to simulate the pyroclastic flows, two models were used: the energy line and the flow3d model. figure 3. energy line model. figure 2. sliding block model. figure 4. energy line for pyroclastic flows. energy line model the energy line principle is the ratio between height of the starting point of the flow (h) and the length of the runout (l), using a type of friction parameter named heim coefficient (µ), after albert heim (figures 2, 3 and 4), who proposed the energy line concept (heim, 1932). energy line inclination is the angle defined by arctan (h/l). the intersection of the energy line defines the distal limits of the flow, being originated at the eruptive source with the ground surface. the vertical distance (h) between the ground surface and the energy line provides a way to estimate the flow velocity in this model. v 2 = 0.5 g h, (1) where v is flow velocity and g is gravitational acceleration. flow3d model. the flow3d is based on generation of a digital elevation model (dem) representing the topographic surface along which the gravity flows move on a triangulated irregular network (tin) of elevations. this kinematics model is constructed from data sets and a geometric configurations (point source, radial distribution, linear or random) used to set the former model. the triangles are contiguous at their boundaries, so that there are no discontinuities such as those that exist with raster data. the tin also serves as the basis for the computations where the gravitational acceleration is assigned to each triangular element of the network. the algorithm considers previous flow models for gravity slides and assumes a constant mass, thickness, and density of the flows. within each specified triangle a single vector represents the driving acceleration due to gravity. the flow3d code (kover, 1995; sheridan and kover, 1996) provides velocity histories of particle streams along flow paths in three dimensions. the calculus of frictional resistance is similar to the mcewen and malin “flow” model (1989), acceleration along the path of steepest descent for each terrain element and the assumption that the center of mass is focused at the ground surface. however, flow3d calculates viscous and turbulent resistances by multiplying the user-defined pyroclastic flow modeling to reconstruct a volcanic edifice in paipa (boyacá-colombia). coefficients of energy dissipation by the flow velocity, which is determined incrementally. multiple flow paths are incremented every 0.1 seconds across triangular elements using as many as three parameters to calculate shear resistance (τr ): basal friction (a0), viscosity (a1) and turbulence (a2) (mcewen and malin, 1989). τr = a0 + a1v+ a2v 2. (2) flow3d simulations have been verified by laboratory tests and field studies in the 1991 unzen eruption (kover and sheridan, 1993; kover, 1995) and the 1980 mount st. helens pyroclastic flows (kover, 1995). it has been applied to risk assessment at several other volcanoes including the creation of hazard maps at popocatépetl (macías et al., 1995) and volcán colima (martin del pozzo et al., 1995) and risk probability at volcán colima (sheridan and macías, 1995; saucedo et al., 1997). recently, flow3d was used to simulate the velocity history and runout of block-and-ash flows from “merapi type” dome collapses at soufriere hills montserrat (hooper and mattioli, 2001). in colombia the flow3d has been used in hazard evaluation for pyroclastic flows in the cerro machín volcano (obando and ramos, 2003 and obando et al, 2003). reconstruction procedures the first step in reconstruction procedures is to know the distribution of the pyroclastic flows deposits (figure 1). these deposits are characterized by the presence of fragments of blocks and ash, corresponding to the fourth volcanic event, (pardo, 2004). the pyroclastic flows were formed by dome collapse without the presence of an eruption column, dispersing calderitas and olitas in the valleys and creek, towards the northeast and southeast flanks from the volcanic source (between the alto el mirador and alto los volcanes, map 1). in this work, the term pyroclastic flow 2 is used indicating the pyroclastic flow1 was deposited before. flow3d models were simulated on a topographic model constructed from igac topographic mapsscale 1:25000: 117-iv-c, d, y 191-ii-a, b, which were digitalized using the program autocad. once they were digitalized to generate a tin, taking into account the actual topography expressed by the alto el mirador and alto los volcanes to reconstruct the ancient volcano edifice and saved in dxf format (figures 5 and 6), they were converted to xyz format and loaded to flow3d. 59 x: 1'120.000 y : 1 '1 05 .0 00 alto los volcanes el uvo olitas el molino vereda el tunal q ue br ad a a gu a fr ía q uebrada honda g rande alto el mirador q ue br ad a c al de rit as alto los volcanes el uvo olitas el molino vereda el tunal q ue br ad a a gu a fr ía q uebrada honda g rande alto el mirador q ue br ad a c al de rit as x: 1'120.000 y : 1 '1 05 .0 00 figure 5. reconstruction of the volcanic edifice. actual topographic lines (black lines ), and the prolongation of the topographic lines ( gray line ). oscar rodríguez et al. 60 x: 1'118.000 y : 1 '1 05 .0 00 y : 1 '1 10 80 00 x: 1'123.000 alto los volcanes el uvo olitas el molino vereda el tunal alto las peñas q u eb rada honda grande q uebrada el ceron q u eb ra d a q u in ta l 2800 2800 26 00 28 00 quebr ada el tunal alcantarilla el banco el durazno locita blanca 2600 la mesa q u eb ra da ca lde rita s alcantarilla alcantarilla 106 a 2871 hato viejo la esperanza 1316 2697 alto el mirador escuela el tunal universidad nacional de colombia facultad de ciencias departamento de geociencias map 1 scenary map made in order to identify parameters escale. 1 : 25.000 0 250 500 750 1000 explanation drainage ways contours pyroclastic flow 2 volcanic dome emplacement area figure 6. final aspect of the volcanic edifice, including a topographic barrier at sw. the distribution of the pyroclastic flows deposits found in the area is important to achieve the best topographic configuration. to found the distribution that getting the best match respect to paipa distribution (figures 7´s and 8), different values for a0, a1, a2, were used for each model so a0 = 0.18, a1 = 0 and a2 = 0, were the best values, for all the cases the value of initial velocity was zero assuming that pyroclastic flows generated by dome collapse were initially static. discussion and conclusions it was found that the altitude of the volcanic build was 3100 m.o.s.l., and the friction coefficient a0 is 0.18, value coherent with observations made in other sites with pyroclastic flow deposits generated by dome collapse (figures 7´s and 8). the lack of evidence of pyroclastic flow thickness occurs because of the few outcrops in the area. then the reconstruction of the paleotopography is difficult, so it would be easier if exist a topography to measure these values. the maximum velocities reached by these flows were about 75 km/h, but it is not enough to override topographic barriers 3150m.a.s.l. to a great altitude localized 12,5km far from the focus of the pyroclastic flow (figure 7a) figure 7a. final model by flow 3d oscar rodríguez et al. vi = 0 m/s figure 7b. final model by flow 3d figure 8. dem corresponding to the reconstruction of the volcanic edifice. 61 oscar rodríguez et al. 62 acknowledgments the authors wish to express their gratitude to the geologists and vulcanologist michael sheridan and the ph.d. student keith dalbey by supplying the flow 3d code as well as the comments and help along this work. special thanks to sloan moreno i.t engineer of laboratorio de geomatica y computacion gráfica of the universidad nacional de colombia. references. heim, a.: (1932). der bergstruz von elm, deutch. geol. gesell. zeitschr. 34, 74–115. hooper, d. m. and mattioli, g. s.: (2001). kinematic modeling of pyroclastic flows produced by gravitational dome collapse at soufriere hills volcano, montserrat, natural hazards 23, 65–86. kover, t. p. (1995). application of a digital terrain model for the modeling of volcanic flows: a tool for volcanic hazard determination, msc thesis, state university of new york at buffalo. kover, t. p. and sheridan, m. f. (1993). numerical models for pyroclastic flows of mt. unzen, japan. geol. soc. amer. abstracts with programs 25, 268. macías, j. l., carrasco-nuñez, g., delgado, h., martin del pozzo, a. l., siebe, c., hoblitt, r. p. martin del pozzo, a. l., sheridan, m. f., barrera, d., hubp, j. l., and selem, l. v. (1995), mapa de peligros volcán de colima, instituto de geofisica, unam, méxico. mcewen, a. s. and malin, m. c., (1989). dynamics of mount st. helens’ 1980 pyroclastic flows, rockslide-avalanche, lahars, and blast, j. volcanol. geotherm. res. 37, 205–231. obando q., m. g. and ramos p., r. a. (2003). evaluación de la amenaza por flujos piroclásticos en la localidad de cajamarca (tolima) ante una probable erupción del volcán cerro machín (trabajo de grado). universidad nacional de colombia, bogotá. obando q., m. g., ramos p., r. a., portilla g, m. e. and cepeda v, h. (2003). modelamiento asistido por computador de flujos piroclásticos en el volcán cerro machín (tolima, colombia). geología colombiana, 28,pp. 2137, bogotá. pardo, n., (2004). estratigrafía de las vulcanitas asociadas al volcán de paipa, municipios de paipa y tuta, departamento de boyacá, colombia. (trabajo de grado), facultad de ciencias, departamento de geociencias, universidad nacional de colombia. bogotá. saucedo, r., komorowski, j-c., macías, j. l., and sheridan, m. f. (1997), modeling of pyroclastic flows generated during the 1913 eruption of colima volcano, méxico. eos, trans. amer. geophys. union 78, f823. sheridan, m. f., and tilling, r. i. (1995), mapa de peligros del volcán popocatépetl, informe final, instituto de geofisica, unam, méxico. 12 pp. & map. sheridan, m. f. and kover, t.: (1996), flow3d: a computer code for simulating rapid, openchannel volcanic flows. in proc. ujst workshop on the technology of disaster prevention against local severe storms. norman ok, pp. 155–163. sheridan, m. f. and macías, j. l. (1995), estimation of risk probability for gravitydriven pyroclastic flows at volcán colima, méxico, j. volcanol. geotherm. res. 66, 251– 256. velandia, f., cepeda, h., alfaro, c., vázquez, l., bernal, n., aguirre, a. (2004). integración geocientífica para la obtención del modelo geotérmico de paipa. primer congreso colombiano de hidrogeología, memorias, medellín. issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n2.89819 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (june, 2022): 131 138 g e o ph y si c s effects of the continuously operating reference station (cors) network on the accuracy, precision, and time to fix ambiguity (ttfa) performance ömer gökdaş1, m. tevfik özlüdemir2 1. department of geomatics engineering, istanbul water and sewerage administration, 34060, eyupsultan, istanbul, turkey. email: ogokdas@iski.gov.tr orcid id: 0000-0003-0109-8073 (correspondence) 2. department of geomatics engineering, igs ista satellite observation and processing laboratory, istanbul technical university, 34469, maslak, istanbul, turkey. email: tozlu@itu.edu.tr orcid id: 0000-0002-1413-9244 subject: gps measurements applied to geophysical problems abstract the geometric design of the continuously operating reference station (cors) network is one of the most critical factors that impact accuracy, precision, and time to fix ambiguity (ttfa) performance. in this study, the authors investigate the subject of geometric design by using both local iski cors and national cors-turkey (cors-tr) networks, and they redesign the iski cors network by increasing interstation distances. for three systems, real-time virtual reference station (vrs) solutions have been obtained and tested with analysis of variance (anova). as a result, the 8-station iski cors provides the most accurate results in the vertical component. in contrast, for ttfa performance, the cors-tr network shows worse outcomes than the others. due to the increase in the interstation distances, the increase in the base length caused worse results in fix solution ratios for redesigned iski cors. in summary, the authors have stated that the 8-station iski cors network performs better, especially regarding vertical accuracy. furthermore, the authors state that the ttfa performance and fix solution ratios are high, and the iski cors geometric design is optimal. keywords: iski cors; cors-tr; network; accuracy; precision. efectos de una red de estaciones de referencia de operación continua en exactitud, precisión y desempeño de la resolución de ambigüedades resumen el diseño geométrico de una red de estaciones de referencia de operación continua (cors, del inglés continuously operating reference station) es uno de los factores más críticos que impactan la exactitud, precisión y desempeño de la resolución de ambigüedades (ttfa, del inglés time to fix ambiguity). en este estudio los autores investigan el diseño geométrico de la red local iski cors y la nacional cors-turkey (cors-tr) y rediseñan la red iski cors al incrementar las distancias entre estaciones. en los tres factores se obtuvieron las soluciones en tiempo real de las estaciones de referencia virtual (vrs, virtual reference station) y se evaluaron con el análisis de la varianza (anova). como resultado en el componente vertical, la red iski cors de ocho estaciones registra los resultados más exactos. en contraste, para el desempeño ttfa la red cors-tr muestra los peores resultados frente a los otros factores evaluados. debido al incremento de las distancias entre estaciones, el incremento en la longitud de base generó los peores resultados en los radios de resolución para la red rediseñada iski cors. en resumen, los autores han establecido que la red iski cors de ocho estaciones presenta el mejor desempeño, especialmente en términos de precisión vertical. además, los autores definieron que el desempeño ttfa y los radios de resolución son altos, y que el diseño geométrico iski cors es el óptimo. palabras clave: iski cors; cors-tr; redes; precision; exactitud; record manuscript received: 08/08/2020 accepted for publication:07/06/2022 how to cite item: gokdas, o., & ozludemir, t. (2022). effects of the continuously operating reference station (cors) network on the accuracy, precision, and time to fix ambiguity (ttfa) performance. earth sciences research journal, 26(2), 131-138. https://doi.org/10.15446/esrj.v26n2.89819 https://doi.org/10.15446/esrj.v26n2.89819 https://doi.org/10.15446/esrj.v26n2.89819 132 ömer gökdaş, m. tevfik özlüdemir introduction high-accuracy positioning can be provided through a differential global positioning system (dgps) approach. this approach is also known as relative positioning, and it is based on simultaneous satellite observation and assessment from a reference station with known coordinates, and stations which the positioning will be achieved. within this framework, post-process or real-time kinematic (rtk) methods can be employed. in single base systems with a single reference station, distance-based (atmospheric refraction, or orbit) errors decrease the accuracy. in these systems, a high-accuracy position can be determined with an interstation distance of up to 10-20 kilometres. with multiple reference stations using carrier phase networks instead of a single reference station, real-time positioning can be achieved across all cities and countries with the desired accuracy. using data from multiple stations significantly decreases distance-based errors and increases ttfa performance and positional reliability. for this purpose, the distance-based error model is created in multiple stations’ adjustment software and transferred to the user in real-time (lachapelle et al., 2000; raquet, 1998; wübbena et al., 1996). correction data is transmitted to the user in the radio technical commission for maritime services (rtcm) format. rtcm started its operations as a usa government advisory committee in 1947 before transforming into its current status as a globally supported independent organisation. rtcm creates a standard that facilitates all transportation modes, including satellite positioning and navigation systems. data transmission is achieved through network transport of rtcm via internet protocol (ntrip). this enables a safe flow control mechanism for rtcm data on a network (rtcm, 2019). a frequently used correction model in network rtk (nrtk) is the vrs; the operating principle of this is shown in figure 1. in short, reference stations transmit carrier phase observation and broadcast ephemeris data to the control centre in rtcm format via ntrip. subsequently, a rover transmits approximate coordinates to the control centre in the national marine electronics association (nmea) format via ntrip. there is two-way communication between the rover and the control centre. cors software in the control centre creates a vrs point using the approximate coordinates of the rover. the rover then achieves precise positioning by retrieving correction data based on the vrs location and tropospheric conditions from the internet via ntrip (wanninger, 1999). the advantages of this method are that they support current rtcm formats (rtcm2.1 and rtcm 3.1), calculate troposphere and ionosphere models at the control centre, and create separate corrections for each rover. there are, however, also disadvantages of this method. these include a limited number of concurrent users (depending on the control centre server capacity) and the necessity to re-initialise when the distance between rover position and vrs point exceeds certain distance limits (gumus and selbesoglu, 2019). figure 1. vrs operation principle. studies have already been carried out into the positioning accuracy and reliability of single rtk for different interstation distances using horizontal and vertical components. in these studies, it was found that precision in the vertical components changes as horizontal components change and that as interstation distance increases, the accuracy of single rtk method decreases along with it (featherstone and stewart, 2001; satalich and ricketson, 1998). in some studies, different geometric designs were adopted to test global navigation satellite system (gnss) receiver brands. it was found that when 90 km interstation distance is sufficient, ± 2.5cm accuracy was achieved in horizontal components and ± 5cm accuracy in vertical. different gnss receiver brands had no significant effect on the analysis (edwards et al., 2010; eren et al., 2009; garrido et al., 2011). in terms of temporal correlation for nrtk methods, it was shown that there is no significant difference for interstation distances of between 15 and 45 kilometres (ogutcu, 2019). other studies have observed that galileo and glonass usage with gps has positive impacts on integer ambiguity solution and ttfa (ogutcu and kalayci, 2016; tsakiri, 2011). in studies conducted to investigate the effect of nrtk measurement time on accuracy and precision, it was observed that while measurement durations increase precision, there is no significant difference for accuracy (bae et al., 2015; ogutcu and kalayci, 2018). similarly, studies in south carolina showed that gnss measurement accuracy increases with measurement duration; however, it does not significantly change after 180-300 seconds (allahyari et al., 2018). this paper aims to investigate the impacts of cors network geometry on precision and accuracy by using different cors networks. additionally, the effect of the geometric design on ttfa will be analysed. figure 2 shows the position of two different cors network stations for the study region. figure 2. iski cors (×) and cors-tr (*) network stations in istanbul and its immediate surroundings. during the design phase of the cors-tr network, there was a discussion in the technical and scientific environment regarding point density. following these discussions, the number of stations in the project rose from the 80 that had initially been planned – firstly to 145, and then again to a final total of 158 stations (cors, 2019). in addition to cors-tr, local cors networks such as iski cors network have been put into service in turkey. it has not yet been established how these local and national networks will be integrated, and what the network design principles and point density should be. the solutions developed in this regard are essential both for the efficient use of resources and to guarantee the performance expected from the network. this study was carried out in order to contribute to the formation of the aforementioned solution strategy. within the scope of the study, the effects of baseline lengths on the solution performance related to accuracy and precision will be discussed, as well as the ensuing effect on ttfa, which is a crucial component of gnss networks. in this context, the network geometry was changed for iski cors, with the interstation distances being increased. the analysis was conducted into the accuracy and precision rate for the new interstation distance, and the effects of the redesigned network geometry on integer ambiguity solution and ttfa were investigated. in addition, the sufficiency of iski cors network station frequency was tested using two different network geometries. data and methods this study was based on the correction data of the local iski cors in istanbul and the national cors-tr network. reference stations for these two 133effects of the continuously operating reference station (cors) network on the accuracy, precision, and time to fix ambiguity (ttfa) performance networks have been positioned in locations without signal obstruction to limit satellite visibility. similarly, there are no reflecting surfaces at station locations that might lead to multipath error. as rover stations, two stations that were added to the iski cors network in 2013 and 2014 were used (kest and pasa) (iski, 2019). the average altitude difference between kest-pasa stations and the iski cors network is 47m; for cors-tr, it is 80m. a low altitude difference between the rover and the reference stations decreases the tropospheric effect. in kest and pasa stations, there are topcon tps netg3a model receiver and topcon tpscr.g5 antenna. other stations in the iski cors network have a topcon tps net-g3 model receiver and a topcon tpscr.g3 model antenna. in the cors-tr network, the stations are in the trimble system. the study examined the effect of network geometry on accuracy, precision and ttfa, carrying out analysis on open source rtklib software (rtklib, 2019) with a real-time vrs solution. to introduce stations as a rover on topnet live software and to connect with ntrip, an rtk mount point was defined on these stations and 1004 (gps) and 1012 (glonass) formats in rtcm 3.0 were selected as observation message type. correction data was obtained simultaneously via iski cors and corstr 2101 port external ip over rtklib software, and ntrip connection was provided over a dsl internet line. broadcast ephemeris data were defined with free-broadcasted ntrip. use-snip.com:2101/rtcm3eph service. because introducing these changes in topnet live software (such as defining stations as a rover, and changing network geometry by interrupting correction broadcast of some stations) will impact all users, the measurements were conducted between 11 pm and 4.30 am in order to minimise disruption. in addition, recording measurements at night time resulted in a minimum level of ionospheric effect. figure 3 shows the data communication between cors software in the iski cors and corstr network control centre (topnet live and trimble cors) and rtklib software. in figure 3, it can be seen that solution 1 and solution 2 were obtained by using two different base corrections (iski cors and cors-tr) for kest and pasa stations. figure 3. data communication scheme in vrs solution. to carry out our study, firstly the two reference stations selected as rovers (kest and pasa) were inactivated on topnet live software in the iski cors network centre server. an 8-station iski cors network was used to obtain 17,500 epoch vrs solutions with a one-second period on rtklib software. these data were collected on 24, and 25 october 2019, between 11 pm and 4.30 am (figure 4b). subsequently, four stations inside the iski cors network (beyk, kcek, pala and terk) were inactivated on topnet live software, and baseline lengths were increased (figure 4a). using this new iski cors network geometry, vrs solutions for kest and pasa stations were obtained as 17,500 epochs; these were collected on 25 and 26 october 2019 using the same timeframe. similarly, 6-station national cors-tr network geometry was used, and vrs solutions were simultaneously obtained for kest and pasa stations (figure 5). figure 4. a: 8-station iski cors network geometry. b: 4-station iski cors network geometry. (+: rover). figure 5. national cors-tr network geometry with six stations in the study region (*: cors-tr, +: rover). figure 4 shows the 8-station iski cors network and the new 4-station geometric design. accordingly, the average interstation distance for a 4-station iski cors network was 88.90 km, and the average interstation distance for an 8-station iski cors network was 48.60 km. figure 5 shows the geometric design for cors-tr network (average of 89.60 km interstation distance). since iski cors and cors-tr networks are defined with different datums, itrf2005 itrf2000 and itrf2000 itrf1996 transformations were conducted with the itrf online tool (itrf, 2019) in order to achieve datum unity. kest and pasa station coordinates were determined with bernese (v5.2) software (dach et al., 2015) by using igs stations as a reference, and 134 ömer gökdaş, m. tevfik özlüdemir these values were used as true values (gökdaş and özlüdemir, 2020). vrs cartesian coordinates obtained with three different network geometries for kest and pasa stations were subtracted from the true values determined with bernese software. local geodetic system (lgs) transformation of the differences between the vrs solution and true values were calculated in equation 1. ℎ = −sin cos − sin sin cos − sin cos 0 cos cos cos sin sin × ( − ) (1) in equation 1, n (north), e (east) and h (up) are the coordinate axes in lgs, and and are the latitude and longitude values of receiver positions, respectively. is the true receiver value in the ecef coordinate system. recef are measurement values obtained with vrs. root mean square error (rmse) values for n, e, u (lgs) coordinates are calculated with equation 2. = √ ∑ ( )2=1 −1 = √ ∑ ( )2=1 −1 = √ ∑ ( )2=1 −1 (2) here, k is the total number of vrs epochs. rmse horizontal components (2d) are calculated based on the equation below. 2 = √ 2 + 2 (3) matlab software was adopted to evaluate the results. an anova test was selected, as the population had a normal distribution and independent samples. two different hypotheses were tested in order to establish whether different cors geometric design had caused significant difference for accuracy and precision (equation 4). ho : m1 = m2, ha : ho, is rejected (4) = ∑ ( ̅ − ̅ ) 2 ( −1)⁄ ∑∑( − ̅ ̅ ̅ )2 ( − )⁄ (5) in equation 4, m1 values are vrs solutions (n, e and u) of the kest and pasa stations obtained with the 8-station iski cors network. the m2 value is the vrs solution for cors-tr and the 4-station iski cors network. in equation 5, represents the number of groups. here, iski cors and cors-tr networks are represented as a group. the numerator of the equation represents the sum of squares, and the denominator represents the within-group sum of squares. n is the sample size (the number of epochs). ftest – the value obtained from equation 5 at (k 1) and (n k) degrees of freedom – was compared with ftable value at α: 0.05 significance level. the value from the f table was f1,inf = 3.84. results and discussion as explained in the previous section, the local iski cors network geometric design was analysed. for this purpose, 8and 4-station iski cors networks were used, and vrs solutions were obtained for two stations (kest and pasa) with two different geometric designs (figures 4a and 4b). again looking at two stations, the national cors-tr network was also used to obtain a vrs solution for a longer interstation distance than that of the iski cors network (figure 5). vrs solutions with 17,500 epochs for the 8-station iski cors network, the 4-station iski cors network and the national cors-tr network with six stations in the study region were obtained with rtklib software. as explained in the previous section, this was maintained by calculating the difference between obtained vrs solutions and true values determined with bernese. by using equation 1, lgs transformation was applied to the differences and north, n, e and h values were calculated (figure 6). table 1 shows mean, max, min, rmse and std. values for the north, east and up in lgs. accordingly, the 4-station iski cors showed lower accuracy than the other two cors geometries for n axis mean and rmse value (1.57cm and ±1.76cm). on the e axis, the 6-station cors network provided better results than the other two cors geometries (mean: 0.21cm, rmse: ±0.63cm). on the u axis, the 8-station iski cors network showed the best performance for mean and rmse values (-1.19cm, ±1.66cm), whereas the cors-tr network showed the worst performance (mean: -3.26cm, rmse: ±3.38cm). for std. values calculated for the three different cors geometries, similar results were determined for each axis. figure 6. n, e and u axis average accuracy and precision for three different cors geometries (a: n axis, b: e axis and c: u axis). 135effects of the continuously operating reference station (cors) network on the accuracy, precision, and time to fix ambiguity (ttfa) performance table 1. mean, max, min, rmse and std. values on the n, e and u axes for three different geometric designs. average 4-stations iski 8-stations iski cors-tr (cm) n e u n e u n e u mean 1.57 0.88 -2.71 0.94 0.85 -1.19 1.03 0.21 -3.26 max. 3.17 2.08 -0.97 2.51 2.23 0.64 2.87 1.51 -0.76 min. 0.27 -0.17 -5.62 -0.08 -0.04 -3.26 0.05 -0.68 -5.15 rmse ±1.76 ±1.04 ±2.88 ±1.14 ±1.02 ±1.66 ±1.17 ±0.63 ±3.38 std. 0.80 0.55 0.99 0.65 0.57 1.16 0.56 0.59 0.87 figure 7. normal distribution graphic of vrs solutions with three different cors geometries (a: north, b: east, c: up). figure 8. anova test graphic for three different cors network geometries for the north, east and up components. figure 7 shows normal distribution graphics for the north, east and up components calculated with three different cors geometries. according to figure 7 and table 1, the error margin on the u axis is more than for the other axes. the 8-station iski cors geometry showed the best performance on the u axis. figure 8 shows the box plot of the error distribution of three different cors network solutions. the line in the box plot shows the median value, and the red region outside the upper and lower boundary notes the outliers. vrs solutions on cors networks with three different geometries were compared and analysed using an anova test at a 95% confidence interval on matlab software. as can be seen from table 2, the anova test showed that the 8-station iski cors and 6-station cors-tr had lower f:256 values than the other results on the horizontal component. however, at a 95% confidence interval, there was a significant difference between the three different cors geometry solutions (p>f:0). in table 2, ss is the sum of squares, df is the degree of freedom, ms is mean square, f is ftest value, and ftable value is 3.84. 136 ömer gökdaş, m. tevfik özlüdemir figure 9 shows the average change on the n, e and u axes for the first 100 seconds for three different cors geometries. this shows that, regardless of the number of stations in the iski cors network solution, an average of 5 seconds was obtained; for the cors-tr network solution an average ttfa value of 65 seconds was obtained. figure 10 shows the fix solution rate of kest and pasa stations for three different cors geometries. here it can be seen that the integer ambiguity solution rate of kest station was higher than that of pasa station. while integer ambiguity solution rates of 8-station iski cors and cors-tr network geometries provided similar results for both stations, the ratio was low for 4-station iski cors network geometry. the lower integer ambiguity solution rate in pasa station (compared to kest station) was analysed. low-level integer ambiguity solution in each of the three cors geometries of pasa station was observed between 11 pm and 11.30 pm. between these hours, it was found that only a small number of gps and glonass satellites were observed at pasa station, and they did not show homogenous distribution. figure 11 shows the gps and glonass satellites at 11 pm, 1.30 am, and 4 am. table 2. anova test statistics. anova source ss df ms f prob>f is k i 8 r ef . / c o r str horizontal between groups 0.01157 1 0.01157 256 0 within groups 1.5955 35000 0.00045 total 1.6071 35001 > 3.84 vertical between groups 3.8138 1 3.8138 36369 0 within groups 3.70668 35000 0.00001 total 7.52044 35001 > 3.84 is k i 8 r ef . / 4 r ef . horizontal between groups 0.47293 1 0.47293 11287 0 within groups 1.48098 35000 0.00042 total 1.95391 35001 > 3.84 vertical between groups 2.12132 1 2.12132 17665 0 within groups 4.24460 35000 0.00012 total 6.36591 35001 > 3.84 figure 9. ttfa graph. figure 10. integer ambiguity solution rates. 137effects of the continuously operating reference station (cors) network on the accuracy, precision, and time to fix ambiguity (ttfa) performance summary and conclusion in this paper, the authors investigated the effect of cors network geometric design on the accuracy, precision, ttfa and integer ambiguity solution in nrtk. in addition to the local iski cors network operating in istanbul with ten stations, data were also used from the national cors-tr network with 158 national and six regional stations (figure 2). in addition to the two current network geometries, a new geometric model was developed for iski cors network geometry analysis, and a 4-station iski cors network with a lower number of stations and higher interstation distances was created. for this purpose, only four stations (sile, slvr, tuzl and yali) were activated in topnet live software where iski cors network correction parameters are calculated (figure 4a). vrs solutions were obtained for kest and pasa stations inactivated at the iski cors network central server with three different network geometries. observation files for the inactivated kest and pasa stations and vrs correction data created on topnet live software for kest and pasa stations were transferred to rtklib software. broadcast ephemeris data were obtained from the online service. all data were communicated in rtcm data format and with ntrip data transmission protocol (figure 3). since it is not possible to simultaneously broadcast correction data in 4-station iski cors and 8-station iski cors geometries, the vrs solution was conducted after waiting for one day between the analyses (24-25 october and 25-26 october 2019). in cors-tr geometry, vrs analysis for both days was conducted simultaneously. since increasing baseline lengths by changing the iski cors network geometric design and inactivating two stations as rover will impact the nrtk users, this study was conducted between 11 pm and 4.30 am. the true coordinates of kest and pasa stations were determined with bernese software, and these values were used in accuracy analysis. since iski cors and cors-tr networks are defined on different datums (itrf05 itrf96), datum transformation on kest and pasa stations coordinates were applied for datum unity. the true value of kest and pasa stations and the differences between vrs solutions of three different cors geometries were transformed into lgs with equation 1. north, east and up values calculated in lgs were statistically analysed (table 1). additionally, these values were tested with anova on matlab (table 2). after conducting accuracy and precision analysis on three different cors geometry solutions, integer ambiguity solution rate and ttfa performance were analysed (figure 9, 10). based on these analyses, it was seen that the horizontal and vertical components of three different cors geometry solutions were different from each other at a 95% confidence interval. when compared for accuracy, it was seen that for horizontal components interstation distances of three different cors geometries had no significant effect on results. for vertical components, the 8-station iski cors network with the shortest interstation distance produced the best results for accuracy. for std. values, it was found that interstation distances had no significant effect on the solutions. within the scope of this study, it was seen that although interstation distances had no effect on ttfa, 4and 8-station iski cors geometries ensure precise positioning in the same time interval, and cors-tr geometry provided a solution with a 60-second delay. in the integer ambiguity solution rate, the 4-station iski cors geometry performed less well than the other two cors geometries. additionally, gps and glonass satellite numbers, elevation angle and distribution were investigated for low integer ambiguity solution rates in pasa stations. it was found that these results could be accounted for by the low number of satellites observed between 11 pm and 11.30 pm at pasa station, and their non-homogenous distribution (figure 11). as a result, the authors express that by considering precise positioning at vertical components, integer ambiguity solution rate and ttfa performance factors; the 8-station iski cors has sufficient point density and optimal geometric design. cors-tr also provided similar results to the 8-station iski cors network, especially in the horizontal component and fix solution rate. as seen in this study, iski-cors network gave optimum results with an average interstation distance of 48.60 km. it is recommended not to exceed this base distance in newly established cors networks. acknowledgements the authors would like to thank the istanbul water and sewerage administration for providing full access to iski cors network data and topnet live software. also, the authors thank everyone who created the rtklib software and enabled open source development. references allahyari, m., olsen, m. j., gillins, d. & dennis, m. d. (2018). tale of two rtns: rigorous evaluation of real-time network gnss observations. journal of surveying engineering, 144, 2. http://dx.doi.org/10.1061/(asce) su.1943-5428.0000249 bae, t. s., brzezinska, d. g., mader, g., & dennis, m. (2015). robust analysis of network-based real-time kinematic for gnss-derived heights. sensors, 15, 27215-27229. http://dx.doi.org/10.3390/s151027215 cors-tr. (2019, december 21). http://www.tusaga-aktif.gov.tr/map/sensormap.aspx dach, r., s., lutz, p., walser, p. & fridez. (2015). bernese gnss software version 5.2. user manual. astronomical institute, university of bern, bern open publishing. isbn: 978-3-906813-05-9. http://dx.doi.org/10.7892/ boris.72297 edwards, s. j., clarke, p. j., penna, n. t., & goebell, s. (2010). an examination of network rtk gps services in great britain. survey review, 42(316), 107-121. http://dx.doi.org/ 10.1179/003962610x12572516251529 eren, k., uzel, t., gülal, e., yıldırım, ö., & cingöz, a. (2009). results from comprehensive global navigation satellite system test in the cors-tr network: case study. journal of surveying engineering, 135(2), http:// dx.doi.org/ 10.1061/(asce)0733-9453(2009)135:1(10) figure 11. gps (green) and glonass (orange) satellite graphics observed for three different times on pasa station. http://dx.doi.org/10.1061/(asce)su.1943-5428.0000249 http://dx.doi.org/10.1061/(asce)su.1943-5428.0000249 http://dx.doi.org/10.3390/s151027215 http://www.tusaga-aktif.gov.tr/map/sensormap.aspx http://www.tusaga-aktif.gov.tr/map/sensormap.aspx http://dx.doi.org/10.7892/boris.72297 http://dx.doi.org/10.7892/boris.72297 http://dx.doi.org/%2010.1179/003962610x12572516251529 http://dx.doi.org/%2010.1061/(asce)0733-9453(2009)135:1(10) http://dx.doi.org/%2010.1061/(asce)0733-9453(2009)135:1(10) 138 ömer gökdaş, m. tevfik özlüdemir featherstone, w. e., & stewart, m. p. (2001). combined analysis of real-time kinematic gps equipment and its users for height determination. journal of surveying engineering, 127(2). http://dx.doi.org/ 10.1061/ (asce)0733-9453(2001)127:2(31) garrido, m. s., gimenez, e., lacy, m. c., & gil, a. j. (2011). testing precise positioning using rtk and nrtk corrections provided by mac and vrs approaches in se spain. journal of spatial science, 56(2), 169-184. http://dx.doi.org/ 10.1080/14498596.2011.623341 gökdaş, ö., & özlüdemir, m. t. (2020). velocity estimation performance of gnss online services (apps and auspos). survey review, 53(378), 280-288. https://doi.org/10.1080/00396265.2020.1809233 gumus, k., & selbesoglu, m. o. (2019). evaluation of nrtk gnss positioning methods for displacement detection by a newly design displacement monitoring system. measurement, 142, 131-137. http://dx.doi.org/ 10.1016/j.measurement.2019.04.041 iski cors. accessed december 21, 2019. https://ukbs.iski.gov.tr/ukbs-nedir. aspx itrf. accessed december 21, 2019. http://www.epncb.oma.be/_productsservices/coord_trans/index.php lachapelle, g., alves, p., fortes, l. p., & cannon, m. e. (2000). dgps rtk positioning using a reference network. proceedings of the 13th international technical meeting of the satellite division of the institute of navigation (ion gps). session c3, salt lake city, pp. 1165-1171. ogutcu, s. (2019). temporal correlation length of network based rtk techniques. measurement, 134, 539-547. http://dx.doi.org/ 10.1016/j.measurement.2018.10.099 ogutcu, s., & kalaycı, i̇. (2016). investigation of network-based rtk techniques: a case study in urban area. arabian journal of geosciences, 9(199), 539-547. http://dx.doi.org/ 10.1007/s12517-015-2262-0 ogutcu, s., & kalaycı, i̇. (2018). accuracy and precision of network-based rtk techniques as a function of baseline distance and occupation time. arabian journal of geosciences, 11, 354. http://dx.doi.org/ 10.1007/s12517018-3712-2 raquet, j. (1998). development of a method for kinematic gps carrier-phase ambiguity resolution using multiple reference receivers. phd thesis, ucge report number 20116, university of calgary. rtcm. accessed december 21, 2019. www.rtcm.org rtklib. accessed december 21, 2019. http://www.rtklib.com satalich, j., & ricketson, r. (1998). field test of trimble 4000 real-time kinematic gps survey system. journal of surveying engineering, 124(1). http://dx.doi.org/10.1061/(asce)0733-9453(1998)124:1(40) tsakiri, m. (2011). evaluation of gps/galileo rtk network configuration: case study in greece. journal of surveying engineering, 137(4). http://dx.doi. org/ 10.1061/(asce)su.1943-5428.0000054 wanninger, l. (1999). the performance of virtual reference stations in active geodetic gps-networks under solar maximum conditions. proceedings of the 12th international technical meeting of the satellite division of the institute of navigation (ion gps), 1419-1427. wübbena, g., bagge, a., seeber, g., boder, v., & hankemeier, p. (1996). reducing distance dependent errors for real time precise dgps applications by establishing reference stations networks. proceedings of the 9th international technical meeting of the satellite division of the institute of navigation (ion gps), 1845-1854. http://dx.doi.org/%2010.1061/(asce)0733-9453(2001)127:2(31) http://dx.doi.org/%2010.1061/(asce)0733-9453(2001)127:2(31) http://dx.doi.org/%2010.1080/14498596.2011.623341 https://doi.org/10.1080/00396265.2020.1809233 http://dx.doi.org/%2010.1016/j.measurement.2019.04.041 http://dx.doi.org/%2010.1016/j.measurement.2019.04.041 https://ukbs.iski.gov.tr/ukbs-nedir.aspx https://ukbs.iski.gov.tr/ukbs-nedir.aspx http://www.epncb.oma.be/_productsservices/coord_trans/index.php http://www.epncb.oma.be/_productsservices/coord_trans/index.php http://dx.doi.org/%2010.1016/j.measurement.2018.10.099 http://dx.doi.org/%2010.1016/j.measurement.2018.10.099 http://dx.doi.org/%2010.1007/s12517-015-2262-0 http://dx.doi.org/%2010.1007/s12517-018-3712-2 http://dx.doi.org/%2010.1007/s12517-018-3712-2 http://www.rtcm.org/ http://www.rtklib.com http://dx.doi.org/10.1061/(asce)0733-9453(1998)124:1(40) http://dx.doi.org/%2010.1061/(asce)su.1943-5428.0000054 http://dx.doi.org/%2010.1061/(asce)su.1943-5428.0000054 _hlk28987430 _hlk29023662 earth sciences research journal earth sci. res. j. vol. 20, no. 3 (september, 2016) letter from the editor the earth sciences research journal volume 20, number 3, with eight high-quality scientific articles, is now available online for all our readers, all the related scientific community. the publishing committee has decided to increase to four the numbers issued per year starting 2016. that will be just possible with your support. this time we want to call the attention to the necessary balance between good judgment and environmentalism. just for a different way of thinking about environment and management resources, some self-proclaimed environmentalists take a right for pointing out, for blocking those aimed at development or exploiting resources. their origin is located in germany during the 60's with the vorsorgeprinzip, or the precautionary principle. green activists depict themselves as victims of economic empires of government's neglected policies to turn into persecutors over development, competitivity, and innovation plans. as it was recently stated by the social issues research centre, this principle poses of a harmless policy. is not our right to be protected when the unscrupulous science is applied, especially when the environment and our lives are at risk? but the use of the principle goes beyond the protection of known or foreseen risks. this principle has been recalled to block plans with just little risks which will be balanced out thanks to the reached advantages. the precautionary doctrine is claimed whenever the little risk perception appears, no matter if the actions to take are necessary to allow future benefits. thousands of cases around the world would be checked. not only in mining or petroleum exploitation but genetically modified food, technology gadgets, fracking. this principle is a serious obstacle for the rational analysis. if the risk can not be demonstrated, the vorsorgeprinzip is baseless. the science works assuming the hypothesis can not be probed, but refuted. the evidence and probes are provided to the scientific community, and they will proceed under logic and rational methods; usually, the precautionary principle is not an important matter at this stage of the process. but is there when that policy can turn into dangerous. it origins a fanatism, close to a religious belief. the senseless debate shows a society with a limited future, without options. all the progress is neglected for standing up at the "safe place." our present has been just possible thanks to the bold decisions by some people with the enough vision to calculate the risks by the advantages. the trip to the moon and the sun system study, development of communications and treatment to sicknesses have been no possible under the precautionary principle. our scientific community should pick the good judgment up to allow all the small steps for humanity. carlos alberto vargas jiménez. editor in chief, earth sciences research journal l e t t e r f r o m t h e e d it o r issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n3.96985 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (september, 2022): 221 230 pl a n e ta r y s c ie n c e s life on mars (lomars): history, advances, current research, and perspectives arshad aliad*, muhammad s. shahidb, iffat jabeenc, mohamed a.k. el-ghalia a. earth sciences research centre, sultan qaboos university, al khod, muscat 123, oman b. department of plant sciences, college of agricultural and marine sciences, sultan qaboos university (squ), al-khod, muscat 123, oman c. 1952 jubilee dr, london, ontario, n6g 0h7, canada d. department of earth sciences, college of science, sultan qaboos university, alkhod, muscat 123, oman * corresponding author: arshad ali e-mail: arshadali@squ.edu.om; abstract a quest for life on mars (lomars) started in the early 1960s when the most prestigious scientific journals published several research articles. according to elsevier’s scopus database, the rise in annual literature production started in the late 1990s, most likely associated with the launch of the national aeronautics and space administration’s (nasa) first rover, sojourner, in 1996. the number of articles on mars will likely continue to rise sharply, given that the launch and landing of the mars 2020 perseverance rover are critical to discovering and understanding the present or past life on the planet. thus far, the usa has dominated in the number of citations and collaborations related to mars research, given its facilities equipped with relevant infrastructure and researchers’ capacity to explore the solar system beyond earth. most of these frequently cited papers report observational and theoretical research results. however, a highly cited article is based on analytical studies of a unique martian meteorite (i.e., allan hills 84001) found in antarctica. it is expected that the future sample return mission associated with the perseverance rover caching system could increase the lomars research exponentially in the coming decades if martian samples are successfully brought to earth. based on the total number of publications on lomars, the most influential institute, the author, and the journal are caltech, c.p. mckay, and icarus, respectively. both the institute and the author are directly affiliated with nasa, indicating the leadership offered by the organization in lomars research. keywords: life on mars; water on mars; perseverance rover; mars 2020; zhurong rover; alh84001 meteorite vida en marte: historia, avances, investigación actual y perspectivas resumen las preguntas sobre la vida en marte comenzaron en los años sesenta cuando varios artículos de investigación se publicaron en las revistas científicas más prestigiosas. de acuerdo con la base de datos de elsevier, el aumento en la producción literaria sobre este tema comenzo a finales de los noventas, debido al lanzamiento del sojourner, el primer astromóvil de la administración nacional de aeronáutica y el espacio (nasa). es previsible que el número de artículos sobre marte continuará creciendo debido a que el lanzamiento y amartizaje del astromóvil perseverance en 2020 es determinante en el descubrimiento y exploración del pasado y presente de este planeta. los estados unidos han dominado en número de citaciones y colaboraciones relacionadas a la investigación de marte, dada sus instalaciones equipadas con infraestructura relevante y con investigadores capacitados para explorar el sistema solar más allá de la tierra. la mayoría de estos artículos, frecuentemente citados, presentan el resultado de investigaciones teóricas y de observación. sin embargo, uno de estos trabajos se basa en los estudios analíticos del único meteórito proveniente de marte y que fue hallado en la antártida. se prevee que el sistema de toma de muestras del perseverance rover incrementará exponencialmente la investigación sobre la vida en marte en las próximas décadas siempre y cuando estas muestras puedan traerse a la tierra. con base en el número total de estudios sobre la vida en marte, el instituto caltech, el autor c.p. mckay y la publicación icarus son las más influyentes de la materia. tanto el instituto como el autor están afiliados directamente con la nasa, lo que indica el liderazgo ofrecido por la organización en las investigaciones sobre la vida en marte. palabras clave: vida en marte; agua en marte: astromóvil perseverance; zhurong rover; meteorito alh84001 record manuscript received: 29/06/2021 accepted for publication: 18/10/2022 how to cite item: ali, a., shahid, m. s., jabeen, i., & el-ghali, m. a. k. (2022). life on mars (lomars): history, advances, current research, and perspectives. earth sciences research journal, 26(3), 221-230. https:// doi.org/10.15446/esrj.v26n3.96985 https://doi.org/10.15446/esrj.v26n3.96985 about:blank https://doi.org/10.15446/esrj.v26n3.96985 https://doi.org/10.15446/esrj.v26n3.96985 222 arshad ali, muhammad s. shahid, iffat jabeen, mohamed a.k. el-ghali 1. introduction according to the elsevier’s abstract and citation database, scopus, the first paper about ‘life on mars’ appeared in the scientific journal science (levin et al., 1962), opening the doors for a concerted multidisciplinary approach to explore another planet. in order to address the possibility of the presence of water and the existence of life on mars, several more studies came out during the 1960s (e.g., rea, 1963; oyama, 1963; shirk et al., 1965; hawrylewicz et al., 1967). contemporaneous to these studies, the national aeronautics and space administration (nasa) successfully launched its first flyby of mars (mariner iv) on 28 november 1964 after a few unsuccessful attempts by the soviet union and nasa. later, it captured and sent the first close-up images of mars to earth. the next decade (the 1970s) highlighted the successful launch of viking 1 and viking 2 landers in 1975 that provided important information about the atmosphere and surface of mars with no conclusive evidence regarding the existence of life on the planet (levin and straat, 1976). in addition to the use of remotely controlled unmanned aerial vehicles (uavs) for mars exploration (sharma et al., 2022 and refs. therein), detailed investigations of meteorites have provided a complimentary understanding of mars. up till the early 1980s, some meteorites were thought to be of martian origin due to their chemical composition matching that of the viking lander’s analyses (levin and straat, 1976) and differences in geochemical characteristics from other common groups of meteorites. bogard and johnson (1983) provided direct evidence of a link between these anomalous meteorites and mars after analyzing the trapped gasses in impact-related glass from the elephant moraine 79001 (eet 79001) meteorite collected by ansmet (i.e., us antarctic search for meteorites program) in antarctica. the obtained compositions from these gases (e.g., argon, krypton, xenon) closely resembled those of the martian atmosphere analyzed by the viking landers (bogard and johnson, 1983). martian meteorites are rare (i.e., as of 28 july, 2022, the meteoritical bulletin database contains 335 samples registered out of 69949 = 0.48%) compared to other common types and have so far provided important clues about water and life on mars (e.g., mckay et al., 1996). after launching flybys and landers, in the mid-1990s (4 december 1996), nasa successfully launched its first rover (i.e., a wheeled vehicle), sojourner, which was in a lander called pathfinder to take the martian exploration to the next level. since then, other rovers, including spirit (10 june 2003), opportunity (8 july 2003), and curiosity (26 november 2011), have also made successful landings on mars. all these missions have provided an unparalleled understanding of the martian geology and landscape. our motivation for the present study is the recent launch and landing of nasa’s perseverance rover, also known as mars 2020, designed specifically to search for microbial life on mars and to store sample cores in tubes on the martian surface for future sample return missions. with the exciting current and future developments in mars research and prospects of human habitability on mars, it is vital to evaluate the history and impact of the research on lomars using statistical (i.e., bibliometric) methods. in the present study, we provide a literature overview of the lomars research with the main emphasis on bibliographic variables including author(s), source journal, subjective keywords as well as citation scores (young and belanger, 1983) and recognize the most prominent articles, authors, journals, collaborations, and growth trends. 2. methodology literature search, data extraction and analysis a literature collection was initiated with a wide-ranging search of relevant articles on lomars. the initial step was to determine appropriate databases, which would provide a full range of published articles on lomars. elsevier’s scopus (esc), an extensive database containing peer-reviewed literature, was used for bibliographic information of each document along with other publishing information necessary to carry out citation analysis of the fields relevant to lomars articles. in addition, esc has a wider article coverage than other web sources, thus being most suitable for our objective to extract as much relevant information as possible. abstracts, titles and keywords were searched using a comprehensive list of important words used by the researchers involved in the literature on lomars. the following final boolean search has been carried out in esc during august 2020, after the launch of mars 2020 perseverance rover by nasa, and later updated on march 14, 2021: title-abs-key ((“life on mars*” or “martian life*” or “water on mars*” or “martian soil*” or “snc meteorites*” or “shergottite-nakhlite-chassignite*” or “shergottite nakhlite chassignite*” or “shergottites*” or “nakhlites*” or “chassignites*” or “martian meteorites*”). this method yielded more than seven thousand documents. in order to extract relevant articles, several filters were applied; we kept only english journal articles while removing books, book chapters and conference proceedings. following the application of all these filters, as well as the manual removal of irrelevant articles, the final sample contained 660 articles published between 1962 and 2020. the academic data file records, such as the publication over time, authors’ production over time, authorship, institutions, countries, journals, publication count, keywords, institution collaborations network and citation count were extracted from the scopus database. the downloaded data file (full records and cited references) were exported into the r-based bibliometrix interface, which is widely used for citation analysis (marty et al., 2020; aria and cuccurullo, 2017). these articles have been published in diverse peer-reviewed research journals, written by single or multiple authors from numerous countries. the vosviewer program was used to construct the network analyses between citation, country, and authors’ collaborations (perianes-rodrigues et al., 2016). 3. results and discussion overall results eleven search strings were used to find the 660 documents on lomars research, which have been cited 22,242 times based on the average number of citations per document (33.7). more than two-thousand authors (2006) have been identified in these documents, averaging three authors per document. the annual number of articles began to rise in the mid-1990s and continued to rise until 2002, at which point it mostly remained below thirty publications per year, with the exception of 2003 (35), 2011 (33), and 2019 (34) (figure 1). the introduction of the first rover, sojourner (4 december 1996), may have contributed to the increase in the number of publications in the 1990s. similarly, spikes in the number of articles published in 2003 and 2011 could be attributed to the launches of nasa rovers such as spirit (10 june 2003), opportunity (8 july 2003), and curiosity (26 november 2011). presumably, the exceptional features of the perseverance rover in comparison to previous rovers will result in a higher publication throughput in the future, and the spike in 2019 could be the start of yet another surge in lomars research as previously seen in the 1990s (figure 1). out of the twenty-three publications in 1998 (figure 1 inset), the only two articles cited more than a hundred times in 22 citable years (i.e., cy) were jakosky and shock (1998), and sleep and zahnle (1998), with the mean total citations per article (i.e., mtcpa) and mean total citations per year (i.e., mtcpy) equal to 32.9 and 1.5, respectively (figure 2). when compared to years with np > 20 per year, the years 2000 (cy = 20) and 2015 (cy = 5) have the highest mtcpa (74.9) and mtcpy (6.3) values (figure 2). four of the twenty articles published in 2000 (cy = 20, figure 2) were cited more than 150 times (yen et al., 2000; skidmore et al., 2000; benner et al., 2000; wynn-williams and edwards, 2000), which could explain the higher mtcpa (75). a paper on hydrated salts on mars (ojha et al., 2015) has been cited over 300 times, giving the year 2015 the highest mtcpy value (6.3). the same paper is also one of the top ten most cited publications in terms of average number of citations per year (i.e., apyc; see table 1). 223 life on mars (lomars): history, advances, current research, and perspectives table 1. a summary of the top ten highly cited articles on lomars research. abbreviations: r = article rank. tc = total citations. apyc = average per year citations. py = publication year. 5yif = 5-year impact factor. jgr = journal of geophysical research. epsl = earth and planetary science letters. r tc apyc py article title brief description journal/5yif reference 1 312 52.0 2015 spectral evidence for hydrated salts in recurring slope lineae on mars several hydrated salts were discovered, which is consistent with the formation of recurring slope lineae as a result of contemporary water activity on mars nature geoscience/ 15.781 ojha et al., 2015 2 1199 48.0 1996 search for past life on mars: possible relic biogenic activity in martian meteorite alh84001 the discovery of polycyclic aromatic hydrocarbons, carbonate globules, and secondary minerals from the fresh fracture surfaces of the alh84001 martian meteorite could be the fossil remains of a past martian biota science/ 44.372 mckay et al., 1996 3 124 41.3 2018 radar evidence of subglacial liquid water on mars the planum australe region was surveyed using mars advanced radar for subsurface and ionosphere sounding (marsis), which revealed bright features with high dielectric permittivity (>15), indicating the presence of a stable body of liquid water on mars science/ 44.372 orosei et al., 2018 4 642 40.1 2005 phyllosilicates on mars and implications for early martian climate the discovery of phyllosilicates by the mars express spacecraft’s omega imaging spectrometer suggests the presence of aqueously altered products during the noachian period (4.1-3.7 ga) nature/ 45.819 poulet et al., 2005 5 267 29.7 2012 the sample analysis at mars investigation and instrument suite the analytical capabilities of the sample analysis at mars (sam), one of the instruments of the mars science laboratory (msl) on board the curiosity rover, have been discussed for the investigation of mars’ habitability potential through determination of chemical and isotopic compositions of atmospheric gases and volatiles space science reviews/ 7.885 mahaffy et al., 2012 6 148 21.1 2014 recurring slope lineae in equatorial regions of mars the mars reconnaissance orbiter data for the deep canyons of valles marineris indicate that recurring slope lineae activity has been widespread on mars, with intermittent flow of briny water in some near-surface equatorial regions nature geoscience/ 15.781 mcewen et al., 2014 7 404 20.2 2001 evidence for recent climate change on mars from the identification of youthful nearsurface ground ice the observational evidence suggests that the nearsurface water ice, which occupies the pore space of soils, formed 100,000 years ago, corresponding to a global layer of water 10-40 cm thick nature/ 45.819 mustard et al., 2001 8 496 17.7 1993 a model for the hydrologic and climatic behavior of water on mars the hydrologic models are used to interpret the largescale vertical and horizontal transport of water between the atmosphere, polar caps, and near-surface crust on mars over the course of its geologic history jgr planets/ 3.858 clifford, 1993 9 272 17.0 2005 sedimentary rocks at meridiani planum: origin, diagenesis, and implications for life on mars the mars exploration rover (mer) opportunity investigated the ancient sedimentary rocks of meridiani planum, which included sandstones, silicates, sulfate minerals, cements, hematite, and crystal molds; formed by chemical weathering and diagenesis, mediated by ambient ground waters epsl/ 5.273 squyres and knoll, 2005 10 122 15.3 2013 impact-generated hydrothermal systems on earth and mars the study of impact craters on earth and mars in relation to the generation of hydrothermal systems, which may have provided nutrients and habitats for life on both planets icarus/ 3.345 osinski et al., 2013 224 arshad ali, muhammad s. shahid, iffat jabeen, mohamed a.k. el-ghali figure 1. plot showing the annual number of articles published on lomars research from 1962 to 2020. the inset bar graph depicts the gradual increase in annual output since the early 1990s. the numbers next to each kink represent the number of articles published during that specific year. figure 2. plot of the number of publications (np), mean total citations per article (mtcpa), and mean total citations per year (mtcpy) against the number of citable years (cy). it is important to keep in mind that interesting studies may have been published in non-elsevier journals, and particularly in non-english journals. this may certainly be true for studies, in particular, published in russian, japanese, and chinese journals, which might influence the statistics. figure 3. (a) comparison of lomars research articles with corresponding authors from different countries, affiliation of co-authors from a single country (scp), and affiliation of co-authors from multiple countries (mcp). scp = single country publications. mcp = multiple country publications. (b) comparison of top-cited countries with a minimum of five citations. most of the g7 countries (united states, japan, germany, the united kingdom, france, italy, and canada) ranked in the top ten. table 2. a list of the top 10 most influential authors who have published the most papers on lomars research. abbreviations: np = number of publications. tc = total citations. authors affiliation np tc h -index g -index m -index c.p. mckay nasa, usa 29 1131 17 29 0.53 f. westall nasa, usa 16 567 13 16 0.57 r.l. mancinelli nasa, usa 14 326 11 14 0.34 d. schulzemakuch technische universität, germany 14 275 9 14 0.60 m.a. sephton imperial college london, uk 14 149 6 12 0.46 a. steele carnegie institute of science, usa 13 873 12 13 0.52 a.f. davila nasa, usa 10 418 9 10 0.69 a.g. fairén centro de astrobiología, spain 10 353 8 10 0.57 j.d. farmer arizona state university, usa 10 598 8 10 0.35 s. fonti università del salento, italy 10 678 6 10 0.29 analysis of prominent authors and countries the productivity of lomars researchers has been assessed using h-, g-, and m-indices, which are based on the number of citations (table 2). six of the top ten most renowned authors are americans (the majority of whom work 225 life on mars (lomars): history, advances, current research, and perspectives for nasa; table 2), while the remaining four are from european countries such as the united kingdom, germany, spain, and italy (members of the g7 countries). in terms of the total number of publications, total citations, h-, and g-index, c.p. mckay is firmly ranked in first place among the top ten authors (table 2). however, because the author has been active for over 30 years, the m-index (i.e., h-index divided by the number of years the scientist has been active) is lower. in addition, the united states leads the world in two categories: multiple country publication (mcp) and single country publication (scp) (scp; fig. 3a). as evidenced by the highest number of documents for the mcp category (i.e., an indicator for collaborative research; figure 3a), nasa ranks sixth on the list of the world’s most collaborative institutions (nature index 2016), indicating the united states’ dominance in collaborative research (i.e., total link strength) in the field of lomars in particular and planetary research in general (table 3). furthermore, the united states ranks first among highly cited countries in lomars research (figure 3b). table 3. the top-15 countries ranked based on the number of publications with their citation scores and collaboration link strength. abbreviations: np = number of publications. tc = total citations. rank country np tc total link strength 1 united states 419 17550 206 2 united kingdom 109 2731 129 3 france 62 3062 100 4 germany 56 2309 110 5 spain 36 1158 79 6 italy 31 1078 48 7 canada 28 2025 59 8 netherlands 27 809 77 9 australia 20 542 15 10 russia 20 1353 23 11 switzerland 13 774 41 12 china 10 122 4 13 japan 10 90 11 14 austria 8 287 32 15 denmark 8 316 18 collaborative research analysis despite a significant difference in the number of documents published as mcp (i.e., co-authors’ affiliations are in different countries) and scp (i.e., all authors’ affiliations are in the same country), the united states remains the topranked country in terms of the corresponding author’s affiliation (figure 3a). if the mcp is used as a proxy for collaborative research, the united states would be the most collaborative country in lomars research (figure 4). the majority of the other top-ranking countries are g7 members, supporting the idea that scientific collaborations are inextricably linked to scientific infrastructure, basic research, and collaborative publishing output. figure 4. comparison of collaborative research in lomars, with a minimum of two articles published. the united states is the main international collaborator on lomars research, particularly in europe and asia. the majority of the g7 countries (japan, germany, the united kingdom, france, italy, and canada) ranked in the top ten. for instance, china has been leading the g7 countries in terms of publications since 1996 (he, 2009), despite the fact that it was not previously prominent in lomars research (figure 4). the author emphasized that political, economic, educational, and scientific policies have successfully transformed the country into a world giant in international scientific publications. collaborative work, which is one of the most important aspects of modern research, opens up international opportunities for scientists, as evidenced by the exponential growth of international collaboration publication output between china and g7 nations (he, 2009). evidently, the majority of african and asian nations’ contributions to lomars research are either absent or insignificant (figure 4). however, a few african and asian countries have recently emerged as leaders in the field of planetary sciences. for instance, the japan aerospace exploration agency (jaxa)and the united arab emirates (uae) jointly launched the hope probe on july 19, 2020. on february 9, 2021, the probe was successfully placed into mars orbit, only days before the mars 2020 perseverance rover landed on the martian surface (https://www.emiratesmarsmission.ae/). the hope probe will investigate various aspects of the martian atmosphere for the first time in order to provide a complete compositional picture using cuttingedge instruments such as the emirates mars infrared spectrometer (emirs), emirate exploration imager (exi), and emirates mars ultraviolet spectrometer (emus). furthermore, the china national space agency (cnsa) launched china’s first interplanetary mission, the tianwen-1 robotic spacecraft to mars, on july 23, 2020, which successfully entered the martian orbit on february 10, 2021 to examine the target site for a rover landing on may 14, 2021. this rover’s scientific goals include studying martian geology with a focus on the planet’s interior structure; identifying minerals and rocks; examining the atmosphere; and looking for evidence of water using 13 scientific equipment (https://en.wikipedia.org/wiki/tianwen-1). as a result of these missions, the number of international collaborative research publications (i.e., mcp) on lomars and related themes could increase exponentially in coming years. international collaborations have lately been proven to be ineffective unless an underdeveloped partner has some local research capabilities to benefit from the developed counterpart (scarazzati and wang, 2019). collaborative research is typically based on partners exchanging knowledge, skills, and techniques, resulting in economic progress and growth (luukkonen et al., 1992; katz and martin, 1997). in contrast, partners with more advanced scientific fields do not seek the same kinds of collaboration as partners with less advanced fields (scarazzati and wang, 2019). according to reports, the planetary and space sciences (pss) potential in africa has largely remained untapped due to the common misconception in african countries that funding fundamental science costs society more than it benefits (baratoux et al., 2017). despite the emergence of pss in africa through the launch of several pss projects for regional 226 arshad ali, muhammad s. shahid, iffat jabeen, mohamed a.k. el-ghali research capacity building, research groups remain dispersed and underfunded. the publication survey (2000-2015) reveals that africa has published less than 1% of the worldwide output of scientific articles published in the four pss representative journals (viz. icarus, journal of geophysical research: planets, journal of geophysical research: space physics, and meteoritics & planetary science), despite africa making up more than 15% of the world’s population (baratoux et al., 2017). however, there have been numerous pss advancements across africa that show promise for the future and the growth of pss-related basic science. for instance, both the south african national space agency and south african astronomical observatory are active in the pss sector. the largest radio telescope in the world, the square kilometre array (ska), will soon be established in the australian and south african deserts (baratoux et al., 2017). morocco has become well-known for various meteorite discoveries and impact studies (chennaoui aoudjehane et al., 2016), attracting international scientists through conferences and outreach activities (e.g., space bus morocco). furthermore, the ibn battuta center and the first astronomical observatory in marrakech have been established to test instruments related to martian exploration and meteorite impacts on the moon (ait moulay larbi et al. 2015), as well as the search for exoplanets (benkhaldoun et al., 2005; gillom et al., 2017). algeria is also focusing on meteorites and impact craters with the support of international collaboration (lamali et al., 2016; sahoui et al., 2016). since 2015, the arabian geosciences union has launched a special section on planetary science and astrobiology for pss scientists across northern africa and the middle east. the sultanate of oman, which has been the site of numerous meteorite discoveries (ali et al. 2017a, 2017b), including martian meteorites like dhofar 019, sayh al uhaymir (sau) 005, and sau 008 (ali et al. 2016, 2017c, 2018), can act as a catalyst for regional collaboration to promote pss. the most collaborative author on lomars research (c.p. mckay; figure 5) has recently discussed five prerequisites for detecting life on other bodies in space, including mars (mckay, 2020). this author identified the following compelling evidence to be targeted in the pursuit of the search for life beyond planet earth: i) signs of liquid water of suitable salinity, ii) carbon in the water, iii) biologically available nitrogen in the water, iv) biologically useful energy in the water, and v) organic material that can possibly be of biological origin and an applicable sampling strategy for this material. figure 5. co-authorship network of the top authors with a minimum of five collaborative publications on lomars research. analysis of productive journals according to bradford’s law (the estimation of exponentially decreasing returns from searching for references in scientific journals), 162 journals have been classified into three categories (representing 4, 16, and 142, respectively), with roughly one-third of all the publications (i.e., 229, 215, and 216) in each category (figure 6a). icarus, astrobiology, planetary and space science, and advances in space research are the four core journals in zone 1. these journals jointly published 229 papers (figure 6a), which were cited 3698 times (figure 6b). the total number of citations in zones 1, 2, and 3 is distributed in increasing order as 3698 < 5302 < 13239, respectively (figure 5b). however, the average number of citations per journal in each category is in the order of 925> 331> 93, respectively. figure 6. (a) distribution of scientific journals according to the bradford’s law distribution of journals in categories; number of publications in each category; and (b) total citation share of each category. note that core journals in zone 1 (icarus, astrobiology, planetary and space science, advances in space research) published 34.7% of the total articles. table 4. the ranking of top-15 most productive journals for lomars research based on the number of publications (np). other parameters, including percent publications (%p), total citations (tc), and h-, g-, and m-indices are given for reference. source np %p tc h -index g -index m -index icarus 75 11.4 3660 34 59 0.68 astrobiology 73 11.1 1923 27 41 1.35 planetary and space science 44 6.7 1087 19 32 0.42 advances in space research 37 5.6 651 15 24 0.43 journal of geophysical research e: planets 35 5.3 1811 24 35 0.96 international journal of astrobiology 29 4.4 465 14 20 0.74 science 22 3.3 2869 16 22 0.27 geophysical research letters 20 3.0 513 14 20 0.64 nature 16 2.4 1832 13 16 0.22 earth and planetary science letters 14 2.1 655 10 14 0.63 geochimica et cosmochimica acta 12 1.8 431 10 12 0.42 geology 10 1.5 398 10 10 0.53 acta astronautica 10 1.5 74 6 8 0.14 origins of life and evolution of the biosphere 9 1.4 497 8 9 0.22 meteoritics and planetary science 8 1.2 274 6 8 0.26 227 life on mars (lomars): history, advances, current research, and perspectives among the journals listed in table 4, icarus is one of the oldest, following science and nature. the top-15 journals have published 62.7% of all lomars research articles. icarus (11.4%) is the most productive journal, just ahead of astrobiology (11.1%). the journal icarus leads in all categories, including the number of publications (np), citations (ct), the h-, and g-indexes, but not the m-index (h-index divided by the number of years since it published its first paper on lomars), because it began publishing on lomars in 1971 (table 4). based on the total number of publications related to the lomars research, the top ten lists of journals, institutes, and authors are compared in a bar diagram (figure 7). the number of articles associated with the journals, institutes, and authors ranges from 14 to 75, 17-55, and 10-29, respectively. there is a close competition between the top two spots for icarus (75) and astrobiology (73), as is seen at the california institute of technology (55) and nasa ames research center (51). however, cp mckay (29) is leading in front of f westall (16) by a substantial margin. therefore, icarus (75), the california institute of technology, and c.p. mckay have been recognized as the most influential journal, institute, and author, respectively (figure 7). elsevier’s journal icarus, with a 2021 impact factor of 3.508 (clarivate analytics, 2022), is primarily focused on the publication of original contributions in the field of solar system studies. caltech established the jet propulsion laboratory (jpl) in the 1930s, and nasa has managed it since 1958. c.p. mckay, the highly influential author of lomars research, is a senior scientist at nasa. keywords analysis according to figure 8, the top five keywords (mars, astrobiology, exobiology, mars’ surface, life on mars) are directly related to mars and the study of martian life. the mars 2020 rover was specifically designed to study past and present life on mars, and future sample return missions are critical to understanding lomars. so far, meteorites have been investigated for this purpose, but because of their journey under severe conditions, some physical changes occur to their surface (i.e., fusion crust) due to frictional heating followed by possible terrestrial contamination. thus, pristine samples returning from mars’ surface and subsurface are predicted to introduce keywords such as surface, mineralogy, carbonates, habitability, and climate in the future (figure 8). figure 7. the top ten highly influential journals, institutes, and authors based on the total number of publications in lomars research. figure 8. co-occurrence network of the most commonly used author keywords in lomars research. 4. conclusions the search for life on other planets, particularly mars, has been going on since the mid-twentieth century, with articles published in prestigious scientific journals beginning in the early 1960s. based on bibliometric analysis using elsevier’s scopus database, 660 articles on life on mars (lomars) research have been published by over 2000 authors and have been cited over 22,000 times, averaging 33.7 citations and 3 authors per article. furthermore, lists of highly cited authors, most productive authors, institutes, and journals are 228 arshad ali, muhammad s. shahid, iffat jabeen, mohamed a.k. el-ghali identified based on citations and number of publications. according to these lists, c.p. mckay, caltech, and icarus emerged as the most productive author, institute, and journal, respectively. the highest ranked author and institute are both affiliated with nasa, demonstrating the organization’s dominance, which, however, is challenged by the recent launches of missions to mars by organizations from other nations, particularly, the china national space agency (cnsa). using multiple nation co-authorship as an indicator of collaborative research, the united states has dominated the world to date due to the availability of infrastructure and human scientific resources. recent collaborative initiatives among asian and african nations have broadened the field of planetary and space sciences. the rate of scientific publications has been found to be closely linked with nasa’s launch of various rovers. we used this tendency as a proxy to predict that future lomars research will grow following the recent landings of nasa’s mars 2020 perseverance rover and china’s zhurong rover. parallel missions launched by the uae space agency in partnership with jaxa would perpetuate this trend. the findings of observational and theoretical research are reported in majority of the top-cited papers. however, the highly cited paper by d.s. mckay et al. (1996) is based on analytical studies of a unique martian meteorite (i.e., allan hills 84001), implying that the future sample return mission associated with the perseverance rover could potentially enhance lomars research. the study of pristine samples from mars’ surface and subsurface would influence the keywords selection seen in previous papers. acknowledgments the authors would like to thank dr. nisar ahmed from sultan qaboos university for help in bibliometric analysis in this study. we are indebted to editor-in-chief dr. alexander caneva for his editorial handling and three anonymous reviewers for their critical and constructive comments to improve the quality of this contribution. references ait moulay larbi, m., daassou, a., baratoux, d., bouley, s., benkhaldoun, z., lazrek, m., garcia, r., & colas, f. (2015). first lunar flashes observed from morocco (iliad network): implications for lunar seismology. earth, moon, and planets, 115, 1-21. https://doi.org/10.1007/s11038015-9462-1 ali, a., jabeen, i., gregory, d., verish, r., & banerjee, n. r. (2016). new triple oxygen isotope data of bulk and separated fractions from snc meteorites: evidence for mantle homogeneity of mars. meteoritics & planetary sciences, 51, 981-995. https://doi.org/10.1111/maps.12640 ali, a., nasir, s., jabeen, i., & al rawas, a. (2017a). chemical and oxygen isotopic properties of ordinary chondrites (h5, l6) from oman: signs of isotopic equilibrium during thermal metamorphism. meteoritics & planetary sciences, 52(10), 2097-2112. https://doi.org/10.1111/maps.12910 ali, a., nasir, s., jabeen, i., & al rawas, a. (2017b). geochemical and o-isotope perspective of a new r chondrite dhofar 1671: affinity with ordinary chondrites. meteoritics & planetary sciences, 52(9), 1991-2003. https:// doi.org/10.1111/maps.12903 ali, a., nasir, s., jabeen, i., & al rawas, a. (2017c). review of the sayh al uhaymir (sau) 005, plus pairings, martian meteorite from al wusta, oman. sultan qaboos university journal for science, 22(1), 29-39. https://doi. org/10.24200/squjs.vol22iss1pp29-39 ali, a., jabeen, i., nasir, s., & banerjee, n. r. (2018). oxygen isotope thermometry of dag 476 and sau 008 martian meteorites: implications for their origin. geosciences (special issue: martian meteorites). https://doi. org/10.3390/geosciences8010015 aria, m., & cuccurullo, c. (2017). bibliometrix: an r-tool for comprehensive science mapping analysis. journal of informetrics, 11, 959-975. https:// doi.org/10.1016/j.joi.2017.08.007 baratoux, d., chennaoui aoudjehane, h., gibson, r., lamali, a., reimold, w. u., selorm sapah, m., charaf chabou, m., habarulema, j. b., jessell, m. w., mogessie, a., benkhaldoun, z., nkhonjera, e., mukosi, n. c., kaire, m., rochette, p., sickafoose, a., martínez-frías, j., hofmann, a., folco, l., rossi, a. p., … abdeen, m. m. (2017). the state of planetary and space sciences in africa. eos, 98. https://doi.org/10.1029/2017eo075833 benkhaldoun, z., abahamid, a., el azhari, y., & lazrek, m. (2005). optical seeing monitoring at the oukaïmeden in the moroccan high atlas mountains: first statistics. astronomy & astrophysics, 441(2), 839-843. https:// doi.org/10.1051/0004-6361:20042515 benner, s. a., devine, k. g., matveeva, l. n., & powell, d. h. (2000). the missing organic molecules on mars. proceedings of the national academy of sciences, 97(6), 2425-2430. https://doi.org/10.1073/pnas.040539497 boamah, d., & koeberl, c. (2007). the lake bosumtwi impact structure in ghana: a brief environmental assessment and discussion of ecotourism potential. meteoritics & planetary sciences, 42, 561-567. https://doi. org/10.1111/j.1945-5100.2007.tb01061.x bogard, d. d., & johnson, p. (1983). martian gases in an antarctic meteorite? science, 221, 651-654. doi: 10.1126/science.221.4611.651 chennaoui aoudjehane, h., el kerni, h., reimold, w.u., baratoux, d., koeberl, c., bouley, s., aoudjehane, m. (2016). the agoudal (high atlas mountains, morocco) shatter cone conundrum: a recent meteorite fall onto the remnant of an impact site. meteoritics & planetary sciences, 51(8), 1497–1518. https://doi.org/10.1111/maps.12661 clifford, s. m. (1993). a model for the hydrologic and climatic behaviour of water on mars. journal of geophysical research planets, 98(e6), 1097311016. https://doi.org/10.1029/93je00225 gillon, m., triaud, a.h.m.j., demory, b.–o., jehib, e., agol, e., deck, k.m., lederer, s.m., de wit, j., burdanov, a., ingalls, j.g., bolmont, e., leconte, j., raymond, s.n., selsis, f., turbet, m., barkaoui, k., burgasser, a., burleigh, m.r., carey, s.j., chaushev, a., … queloz, d. (2017). seven temperate terrestrial planets around the nearby ultracool dwarf star trappist-1. nature, 542, 456-460. https://doi.org/10.1038/nature21360 hawrylewicz, e. j., hagen, c., tolkacz, v., & ehrlich, r. (1967). effect of reduced barometric pressure on water availability related to microbial growth. life sciences in space research, 5, 174-186. he, t. (2009). international scientific collaboration of china with the g7 countries. scientometrics, 80(3), 571-582. https://doi.org/10.1007/s11192007-2043-y jakosky, b. m., & shock, e. l. (1998). the biological potential of mars, the early earth, and europa. journal of geophysical research, 103(e8), 1935919364. doi: 10.1029/98je01892 katz, j. s., & martin, b. r. (1997). what is research collaboration? research policy, 26(1), 1-18. https://doi.org/10.1016/s0048-7333(96)00917-1 lamali, a., rochette, p., merabet, n., abtout a., maouche, s., gattacceca, j., ferrière, l., hamoudi, m., aster team, meziane, e. h., & ayache, m. (2016). geophysical and magneto-structural study of the maâdna structure (talemzane, algeria): insights on its age and origin. meteoritics & planetary science, 51(12), 2249–2273. https://doi.org/10.1111/ maps.12715 levin, g. v., heim, a. h., clendenning, j. r., & thompson, m. f. (1962). “gulliver” ̶ a quest for life on mars. science, 138, 114-121. levin, g. v., & straat, p. a. (1976). viking labelled release biology experiment: interim results. science, 194, 1322-1329. doi: 10.1126/science.194.4271.1322 luukkonen, t., persson, o., & sivertsen, g. (1992). understanding patterns of international scientific collaboration. science, technology, & human values, 17(1), 101–126. https://doi.org/10.1177/016224399201700106 mahaffy, p. r., webster, c. r., cabane, m., conrad, p. g., coll, p., atreya, s. k., arvey, r., barciniak, m., benna, m., bleacher, l., brinckerhoff, w. b., eigenbrode, j. l., carignan, d., cascia, m., chalmers, r. a., dworkin, j. p., errigo, t., everson, p., franz, h., & mumm, e. (2012). the sample analysis at mars investigation and instrument suite. space science reviews, 170, 401-478. https://doi.org/10.1007/s11214-012-9879-z https://doi.org/10.1007/s11038-015-9462-1 https://doi.org/10.1007/s11038-015-9462-1 https://doi.org/10.1111/maps.12640 https://doi.org/10.1111/maps.12910 https://doi.org/10.1111/maps.12903 https://doi.org/10.1111/maps.12903 https://doi.org/10.24200/squjs.vol22iss1pp29-39 https://doi.org/10.24200/squjs.vol22iss1pp29-39 https://doi.org/10.3390/geosciences8010015 https://doi.org/10.3390/geosciences8010015 https://doi.org/10.1016/j.joi.2017.08.007 https://doi.org/10.1016/j.joi.2017.08.007 https://doi.org/10.1029/2017eo075833 https://doi.org/10.1051/0004-6361:20042515 https://doi.org/10.1051/0004-6361:20042515 https://doi.org/10.1073/pnas.040539497 https://doi.org/10.1111/j.1945-5100.2007.tb01061.x https://doi.org/10.1111/j.1945-5100.2007.tb01061.x https://doi.org/10.1111/maps.12661 https://doi.org/10.1029/93je00225 https://doi.org/10.1038/nature21360 https://doi.org/10.1038/nature21360 https://doi.org/10.1007/s11192-007-2043-y https://doi.org/10.1007/s11192-007-2043-y https://doi.org/10.1016/s0048-7333(96)00917-1 https://doi.org/10.1111/maps.12715 https://doi.org/10.1111/maps.12715 https://doi.org/10.1177/016224399201700106 https://doi.org/10.1007/s11214-012-9879-z 229 life on mars (lomars): history, advances, current research, and perspectives marty, t., vanstone, b., & hahn, t. (2020). news media analytics in finance: a survey. accounting & finance, 60, 1385-1434. https://doi.org/10.1111/ acfi.12466 mckay, d. s., gibson jr., e. k., thomas-keprta, k. l., vali, h., romanek, c. s., clemett, s. j., chillier, x. d. f., maechling, c. r., & zare, r. n. (1996). search for past life on mars: possible relic biogenic activity in martian meteorite alh84001. science, 273, 924-930. doi: 10.1126/science.273.5277.924 mckay, c.p., 2020. what is life ̶ and when do we search for it on other worlds. astrobiology, 20(2), 163-166. doi: 10.1089/ast.2019.2136 mcewen, a. s., dundas, c. m., mattson, s. s., toigo, a. d., ojha, l., wray, j. j., chojnacki, m., byrne, s., murchie, s. l., & thomas, n. (2014). recurring slope lineae in equatorial regions of mars. nature geoscience, 7, 53-58. https://doi.org/10.1038/ngeo2014 mustard, j. f., cooper, c. d., & rifkin, m. k. (2001). evidence for recent climate change on mars from the identification of youthful near-surface ground ice. nature, 412, 411-414. https://doi.org/10.1038/35086515 ojha, l., wilhelm, m. b., murchie, s. l., mcewen, a. s., wray, j. j., hanley, j., massé, m., & chojnacki, m. (2015). spectral evidence for hydrated salts recurring slope lineae on mars. nature geosciences, 8, 829-832. https:// doi.org/10.1038/ngeo2546 orosei, r. lauro, s. e., pettinelli, e., cicchetti, a., coradini, m., cosciotti, b., di paolo, f., flamini, e., mattei, e., pajola, m., soldovieri, f., cartacci, m., cassenti, f., frigeri, a., giuppi, s., martufi, r., masdea, a., mitri, g., nenna, c., … seu, r. (2018). radar evidence of subglacial liquid water on mars. nature, 361, 490-493. doi: 10.1126/science.aar7268 osinski, g. r., tornabene, l. l., banerjee, n. r., cockell, c. s., flemming, r., izawa, m. r. m., mccutcheon, j., parnell, j., preston, l. j., pickersgill, a. e., pontefract, a., sapers, h. m., & southam, g. (2013). impact-generated hydrothermal systems on earth and mars. icarus, 224(2), 347-363. https://doi.org/10.1016/j.icarus.2012.08.030 oyama, v. i. (1963). use of gas chromatography for the detection of life on mars. nature, 200, 1058-1059. https://doi.org/10.1038/2001058a0 perianes-rodrigues, a., waltman, l., & jan van eck, n. (2016). constructing bibliometric networks: a comparison between full and fractional counting. journal of informetrics, 10, 1178-1195. https://doi.org/10.1016/j. joi.2016.10.006 poulet, f., bibring, j. p., mustard, j. f., gendrin, a., mangold, n., langevin, y., arvidson, r. e., ondet, b., gomez, c., & the omega team. (2005). phyllosilicates on mars and implications for early martian climate. nature, 438, 623-627. https://doi.org/10.1038/nature04274 rea, d. g. (1963). evidence for life on mars. nature, 200, 114-116. https://doi. org/10.1038/200114a0 shirk, j. s., haseltine, w. a., & pimentel, g. c. (1965). sinton bands: evidence for deuterated water on mars. science, 147, 48-49. doi: 10.1126/science.147.3653.48 squyres, s. w., & knoll, a. h. (2005). sedimentary rocks at meridiani planum: origin, diagenesis, and implications for life on mars. earth and planetary science letters, 240, 1-10. https://doi.org/10.1016/j.epsl.2005.09.038 sahoui, r., belhai, d., & jambon, a. (2016). impact-generated carbonate melts in the talemzane impact structure (laghouat, algeria). arabian journal of geosciences, 9, 641. https://doi.org/10.1007/s12517-016-2665-6. scarazzati, s., & wang, l. (2019). the effect of collaboration on scientific research output: the case of nanoscience in chinese regions. scientometrics, 121, 839-868. https://doi.org/10.1007/s11192-019-03220-x sharma, m., gupta, a., gupta, s. k., alsamhi, s. h., & shvetsov, a. v. (2022). survey on unmanned aerial vehicle for mars exploration: deployment use case. drones, 6, 4. https://doi.org/10.3390/drones6010004 skidmore, m. l., foght, j. m., & sharp, m. j. (2000). microbial life beneath a high arctic glacier. applied and environmental microbiology, 66(8), 32143220. https://doi.org/10.1128/aem.66.8.3214-3220.2000 sleep, n. h., & zahnle, k. (1998). refugia from asteroid impacts on early mars and the early earth. journal of geophysical research planets, 103(e12), 28, 529-28, 544. https://doi.org/10.1029/98je01809 wynn-williams, d. d., & edwards, h. g. m. (2000). proximal analysis of regolith habitats and protective biomolecules in situ by laser raman spectroscopy: overview of terrestrial antarctic habitats and mars analogs. icarus, 144, 486-503. https://doi.org/10.1006/icar.1999.6307 yen, a. s., kim, s. s., hecht, m. h., frant, m. s., & murray, b. (2000). evidence that the reactivity of the martian soil is due to superoxide ions. science, 289, 1909-1912. doi: 10.1126/science.289.5486.1909 young, h., belanger, t., corbin, j. b., magrill, r. m., peterson, f. m., thomas, d. m., torok, a. g., & wools, b. (1983). the ala glossary of library and information science. chicago: american library https://doi.org/10.1111/acfi.12466 https://doi.org/10.1111/acfi.12466 https://doi.org/10.1038/ngeo2014 https://doi.org/10.1038/35086515 https://doi.org/10.1038/ngeo2546 https://doi.org/10.1038/ngeo2546 https://doi.org/10.1016/j.icarus.2012.08.030 https://doi.org/10.1038/2001058a0 https://doi.org/10.1016/j.joi.2016.10.006 https://doi.org/10.1016/j.joi.2016.10.006 https://doi.org/10.1038/nature04274 https://doi.org/10.1038/200114a0 https://doi.org/10.1038/200114a0 https://doi.org/10.1016/j.epsl.2005.09.038 https://doi.org/10.1007/s12517-016-2665-6 https://doi.org/10.1007/s11192-019-03220-x https://doi.org/10.3390/drones6010004 https://doi.org/10.1128/aem.66.8.3214-3220.2000 https://doi.org/10.1029/98je01809 https://doi.org/10.1006/icar.1999.6307 _hlk110424112 _hlk110424427 keywords: multifractal characteristics of regional geomorphology; development of geological disasters; correlation; accuracy; palabras clave: características multifractales de la geomorfología regional; desarrollo de desastres geológicos; correlación; exactitud how to cite item xiong, y. (2021). correlation analysis between multifractal characteristics of regional geomorphology and development of geological disasters. earth sciences research journal, 25(1), 49-55. doi: https://doi.org/10.15446/ esrj.v25n1.93700 in view of the inaccuracy of the traditional correlation analysis method, this paper proposes a correlation analysis method between the multifractal characteristics of regional landforms and the development of geological disasters. firstly, the multifractal characteristics of regional landforms are described by using the basic fractal characteristics of self-similarity or scale invariance. then the corresponding relation table is established according to the width of the fractal spectrum and the number of landslides and hidden dangers, and the spatial relationship of geological disaster development is analyzed. combined with the above-mentioned spatial relationship of geological disaster development and the multifractal characteristic data of regional landforms, the correlation coefficient between the two is calculated to complete the correlation analysis between the multifractal characteristics of regional geomorphology and the development of geological disasters. the experimental results show that compared with the traditional correlation analysis method, the correlation analysis results of the multifractal characteristics of regional geomorphology and the development of geological disasters are more accurate. abstract correlation analysis between multifractal characteristics of regional geomorphology and development of geological disasters análisis de correlación entre las características multifractales de la geomorfología regional y el desarrollo de desastres geológicos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.93700 debido a la inexactitud del método de análisis de correlación tradicional, este artículo propone un método de análisis de correlación entre las características multifractales de accidentes geográficos regionales y el desarrollo de desastres geológicos. en primer lugar, las características multifractales de los accidentes geográficos regionales se describen a través de las características fractales básicas de autosemejanza o invariancia de escala. luego se establece la tabla de relación correspondiente de acuerdo con el ancho del espectro fractal y el número de deslizamientos de tierra y peligros ocultos, y la relación espacial del desarrollo de desastres geológicos. combinado con la relación espacial mencionada anteriormente del desarrollo de desastres geológicos y los datos característicos multifractales de accidentes geográficos regionales, el coeficiente de correlación entre los dos se calcula para completar el análisis de correlación entre las características multifractales de la geomorfología regional y el desarrollo de desastres geológicos. los resultados experimentales muestran que en comparación con el método de análisis de correlación tradicional, los resultados del análisis de correlación de las características multifractales de la geomorfología regional y el desarrollo de desastres geológicos son más precisos. resumen record manuscript received: 16/03/2020 accepted for publication: 27/11/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 49-55 yiying xiong hnu-asu joint international tourism college, hainan university, haikou, 570228, china xiongyiying88@163.com g e o l o g ic a l h a z a r d s https://doi.org/10.15446/esrj.v25n1.93700 https://doi.org/10.15446/esrj.v25n1.93700 https://doi.org/10.15446/esrj.v25n1.93700 mailto:xiongyiying88@163.com 50 yiying xiong introduction china is a country with many geological disasters. there are many kinds, wide distribution, frequent activities, and serious geological disasters, which is one of the most serious countries in the world. geological disaster refers to the event or phenomenon where the geological environment is destroyed suddenly or gradually due to geological processes (natural, artificial, or comprehensive), and human life and property are lost (kleemann et al., 2019). for the classification of geological disasters, there are different angles and standards, which are more complex. in terms of its causes, the geological disasters mainly caused by natural variation are called natural geological disasters; the geological disasters mainly induced by human actions are called human geological disasters. according to the different ways of energy release, its main types include earthquake, volcanic eruption, mountain collapse, landslide, debris flow, land subsidence, ground fissure, etc. although geological disaster is an inevitable phenomenon in the process of earth evolution, its occurrence will bring great influence and destructive power to people (masson et al., 2018). due to the impact of geological disasters on human society, many scholars have been exploring the general laws of these disasters, trying to find a way to prevent or evaluate geological disasters. at present, the research on geological disasters mainly focuses on risk assessment, risk analysis, early warning, and prediction model, while the research on the relationship between different combination forms of geomorphic morphology and spatial distribution of geological disasters is relatively less (galbraith & de lavergne, 2019). at the same time, slope, length, and density of ravines are often used in the study of geomorphic form, which is characterized by only one aspect of a single geomorphic type and highly generalizing the linear characteristics of geomorphic form. it not only causes partial distortion of geomorphic form information but also does not fully reflect the essential characteristics of complex and broken geomorphic form, namely, the amount of its comprehensive properties is not enough. however, with the development of fractal theory, its purpose is to explore the common, unpredictable, unstable, and very random systems, phenomena, and processes in nature, and try to find a new order between order and disorder, macro-micro and global-local. using the simple fractal function, we can depict the mandelbrot landscape which can reach the truth. like the real natural landscape, they have no characteristic length, which is difficult to realize by any mathematical physical model, and have been widely used in many subjects. although the fractal theory is still in its infancy in the study of geological disasters, it is gradually found that the geomorphic morphology shows fractal characteristics, and fractal can be used to quantify the geomorphic morphology of different structures (hansen et al., 2018). the geomorphic form is the final form of modern tectonic movement (cui & wang, 2019), one of the important underlying factors affecting the rainfall, runoff, and sediment yield process, and one of the important factors for the prediction of water and soil loss in the basin. due to the diversity of basin geomorphology and the complexity of spatial combination, the scientific and accurate quantification of basin geomorphology has become one of the key problems to be solved in the establishment of a widely applicable basin water and soil loss prediction model. domestic and foreign researchers generally choose one or more of the average slope, average slope length, watershed height difference ratio, gully density, and gully cutting depth as the quantitative parameters of the geomorphic factors in the watershed water and soil loss prediction model, but these quantitative parameters only represent the characteristics of one aspect of the watershed topography, and the expression of the integrity and comprehensiveness of the watershed topography is not very reasonable. therefore, there are obvious limitations in the application of this method in the field of water and soil loss assessment and prediction (soloviev et al., 2018). the fractal theory is an important branch of mathematics in nonlinear science. it was founded by american mathematician b. b. mandelbrot in the mid-1970s. the fractal theory aims to explore the common, unpredictable, unstable, and very irregular systems, phenomena, and processes in nature, and try to find a new order between order-disorder, macro-micro, and whole part, which is an important part of the modern scientific theory of complexity. although the geomorphic morphology (such as mountains, water systems, coastlines, etc.) seems to be messy and complex, it can be found that they still have certain mathematical (geometric) laws by fractal theory analysis and can be quantitatively described by fractal dimension (benammi et al., 2018). the fractal theory has opened up a new way for the non-linear study of geomorphic morphology description and simulation, geomorphic development process, etc. however, continuous research and practice have found that the complexity of geomorphic phenomena can not be completely described only by a single fractal dimension which depends on the overall characteristics. the complexity of geomorphic morphology shows obvious multifractal characteristics, which can be regarded as a large number of fractal sets with different scale-free areas. therefore, in order to solve the complexity problem in geomorphology effectively, it must focus on the complex multifractal problem, which is of great significance to comprehensively grasp the characteristics and formation laws of various geomorphic phenomena. multifractal theory, as the leading edge of fractal theory, is the further development of fractal theory. it is proposed to study the phenomenon of non-uniformity and dissimilarity in nature. it uses spectrum function to describe the special structure behavior and characteristics caused by different local conditions or different levels in the evolution process of the fractal structure and discusses the distribution law of probability measure of characteristic parameters by the method of statistical physics. the introduction of multifractal can deeply describe the complex features of landforms, thus revealing more information on landforms. with the development of computer and gis technology, a digital elevation model (dem) can directly display and express the three-dimensional shape of the basin landform. taking the basin landform dem as the research object, according to the fractal theory and the multifractal correlation calculation model, it is possible to directly quantify the multifractal characteristics of the basin three-dimensional landform (ali & lee, 2018). the correlation analysis between the multifractal characteristics of regional geomorphology and the development of geological disasters description of multifractal characteristics of regional landform when describing the multifractal characteristics of regional geomorphology, we use the basic fractal characteristics of self-similarity or scale invariance to describe quantitatively through fractal dimension index. it is a common phenomenon in many fields such as geomorphology to synthesize self-similarity and invariance. therefore, for many fractal bodies with nonuniform and singularity, one dimension cannot describe all their characteristics, so it needs to use multiple dimensions fractal measures (chardon et al., 2018). let  be a probability measure distributed on region o. if the concentration of  changes greatly, that is, for a very small r, it has the following relations:  b x r r a( , )( ) (1) in the above equation, b ( x, r ) represents the local neighborhood ball with radius r, a is the singular index, and the singular index a corresponding to different subsets is different. therefore,  is a multifractal measure, multifractal measure is a singular measure, and it does not have a density function. therefore, multifractal is an infinite set of singular measures defined on fractals and composed of multiple scale indexes (salin et al., 2019). in the study of the distribution of measure yields on fractal volume f, fractal volume f can be divided into several units with scale r, and i is used to represent the average value of measure yields in the i-th unit. there is the following scaling relationship between the average measure  and scale r of each cell: µi ar i→ (2) in the above equation, ai is the singularity index. if some elements have the same singularity index a, their measures can be expressed by (a). if fa is used to represent the subsets of the elements with the same singularity index a, then the s-dimensional hausdorff measure of fa can be defined as: 51correlation analysis between multifractal characteristics of regional geomorphology and development of geological disasters h fa a h fa a h fa a u fa s r r s r s i s i ( , ( )) lim ( , ( )) ( , ( )) inf : µ µ µ = = ⊂ → = ∞ ∑ 0 1 i i ∞                   (3) moreover, the s-dimensional hausdorff measure h s(fa,  (a)) satisfies the following relations: h fa a s f a s fs( , ( )) , ( ); (µ = ∞ < = when finite positive value when aa s f a ); , ( )0 when >      ; (4) in the equation above, f a( ) is the hausdorff dimension of the subset. if in the interval [a,a+da], the number of elements whose measure is  (a) is n(a), then h fa a n a rs s( , ( )) ( )µ = (5) thus, it exists h fa a h fa a n a rs r r s s( , ( )) lim ( , ( )) ( )µ µ= = →0 (6) according to the above equations 4 and 6, only when the following equation is satisfied, n a r f a( ) ( )→ − (7) so that h s(fa, (a)) can be taken as a finite value, from which we can get: f a n a rr ( ) lim ln ( ) ln =− →0 (8) for uniform measure distribution, there is only a value; for non-uniform measure distribution, f a( ) is a convex function of a. in multifractal theory, f a( ) is the singular spectrum of multifractal. a is the singularity of different sub regions, and its size is determined by the measurement distribution of the generating unit of the corresponding sub regions. f a( ) is the fractal dimension of the sub regions with the same singularity a. from equation 6, we can see that its size is related to the number of subregions with the singularity a. f a( ) reveals the distribution structure and change of different singularity a in the fractal body, which is a measure of the complexity, irregularity and unevenness of the fractal body. it describes the physical characteristics of different localities or layers on the fractal body (alejo & montes-rojas, 2018). here f a( ) is a basic language to describe multifractal. the multi-fractal shape of regional geomorphology is divided into small units with scale of r, pi (r) is defined as the probability of unit i,  is the probability measure, and when q >> 1 and pi >> pj, the q step of probability measure  can be defined as: n q r p ri q i ( , ) ( )=∑ (9) in this case, the generalized s-dimensional measure can be defined as: m q r p r r n q r r m q m q r s i q i s s s r r s ( , ) ( ) ( , ) ( ) lim ( ) = = =        ∑ →0 (10) if there is a critical index t (q) that depends on the q-order distance of pi , when t (q) is defined as the mass index, then the generalized dimension can be defined according to the q-order distance of  , and the calculation results are as follows: d q n q r r q p p r q r r i i i = − ≠          → → ∑ 1 1 1 0 0 lim ln ( , ) ln , lim ln ln (11) in the above equation, dq is the sum of the q-power of each subset of space’s singular measures, reflecting the degree of singularity of each subset from the overall point of view, and calculating the generalized multifractal characteristic dimension curve of regional landform, as shown in the following figure: dq dq q q = +∞ q = 0 00 q = −8 figure 1. generalized multifractal characteristic dimension curve of regional geomorphology according to the characteristic curve shown in the figure above, dq and t (q) satisfy the following relationship: d q q q q q = − ≠ − =      1 1 1 1 1 τ τ ( ), '( ), (12) when dq and t (q) are differentiable, there is a legendre transformation, as follows: f a qa q a d q dq ( ) ( ) ( ) = − =−      τ τ (13) in the above equation,  is the legendre transformation coefficient. so for the general multifractal spectrum f a( ) is a single peak convex function about a, it increases monotonically in q > 0 part, decreases monotonically in q < 0 part, when q = 0 , f a( ) gets the maximum value, and the multifractal spectrum image obtained is shown in figure 2 below: q = −8 f a( ) f a( )max amin amax a q = 0 q < 0 a(0) q > 0 q = +∞ figure 2. multifractal spectrum image it can be seen from the above figure that the physical meaning of f a( ) is a measure of the complexity, irregularity and unevenness of fractal structure. 52 yiying xiong different singular indexes a reflect the properties of different probability measure subsets (sachica & martinez, 2018). for a simple fractal, the multifractal spectrum a ~ f a( ) is transformed into a point. at this time, this point is the multifractal feature of the regional landform described. retaining the multifractal feature points of the regional landform, combining the spatial relationship of the development of geological disasters, and calculating the correlation coefficient between the multifractal feature of regional landform and the development of geological disasters, the correlation analysis of the two can be completed. analysis of spatial relationship of geological disaster development before analyzing the spatial relationship of geological hazard development, 12 geological points are selected first, and then the geological hazard is divided into landslide, collapse and debris flow by using the fractal characteristic parameters of landform and the spatial distribution characteristic parameters of geological point landslide hazard.firstly, the corresponding relationship between the width of fractal spectrum a and the number of landslides and hidden dangers is established, as shown in the following table: table 1. width of fractal spectrum and quantity of landslides and hidden dangers geological point a n 1 0.045 0 2 0.066 19 3 0.131 51 4 0.120 49 5 0.130 96 6 0.116 134 7 0.084 53 8 0.107 78 9 0.094 48 10 0.134 221 11 0.157 191 12 0.124 134 according to the data shown in the above table, the relationship among a and the number of landslides and hidden geological disasters is obtained, as shown in the following figure: 0.160.120.080.04 0 50 100 150 200 250 a number of landslides and hidden dangers/piece figure 3. relationship among a and the number of landslides and potential geological disasters according to the above diagram, the following landslide coupling spatial relations can be concluded: in general, with the increase of a, the number of landslides also increases, which indicates that there is a significant correlation between the multifractal characteristics of geomorphic morphology and geological disasters of landslides. furthermore, with the increase of a, the complexity of geomorphic morphology and the probability of occurrence of landslides increase, the number of landslides will also increase (zhang et al., 2019). the number of landslides can be divided into three levels: when a<0.100, there is a positive correlation between a and the number of landslides. that is to say, with the increase of a, the number of landslides and potential disasters increases, but the number of landslides in this section is less, up to 53.when 0.100 < a < 0.134, the number of landslides fluctuates relatively large, indicating that the change range of geomorphic singular index is in this area, and the occurrence of landslides is very unstable, showing a certain degree of volatility.when 0.134 < a < 0.157, the number of landslides is about 200, indicating that when the change range of geomorphic singular index is in this area, landslides are more likely to occur. in the case of collapse, according to the fractal characteristic parameters of geomorphology and the spatial distribution characteristic parameters of particle collapse and hidden geological disasters in different places, the corresponding relationship table between the depth of geomorphic pleomorphic spectrum and the number of collapses n is established in the same way as landslide, as shown in the following table: table 2. width of fractal spectrum, collapse and number of hidden dangers geological point a n 1 0.045 0 2 0.094 5 3 0.084 5 4 0.066 0 5 0.107 34 6 0.116 11 7 0.120 7 8 0.130 67 9 0.131 13 10 0.124 65 11 0.138 20 12 0.075 5 according to the data shown in the above table, the relationship among a and the number of collapse and potential geological disasters is obtained, as follows: 0.160.120.080.04 0 20 40 60 80 a number of collapses and hidden dangers/piece figure 4. relationship among a and the number of geological disasters caused by collapse and hidden danger from the diagram above, the following spatial relation of collapse coupling can be concluded: generally speaking, with the increase of a, the number of collapses also increases, indicating that there is a significant correlation between the multifractal characteristics of geomorphic morphology and the geological disasters of collapses, and further indicating that with the increase of a, the complexity of geomorphic morphology increases, the probability of collapses increases, and the number of collapses will also increase.when the number of collapse disasters changes with the change range of geomorphic singular index a, the change characteristics can also be divided into three levels: when a < 0.100, the number of collapse and hidden danger disasters is very small, and there is almost no collapse geological disaster phenomenon in this section (leonarduzzi et al., 2018).when 0.100 < a < 0.134, although the number of collapses increases with a, it also shows some volatility in this range, especially between 0.130 and 0.134, the number of collapses first increases, then decreases and then increases, indicating that the change range of geomorphic singular index is in this region, and the disaster phenomenon of collapses is very unstable; when 0.134 < a < 0.157, the number of collapse disasters is large, but there is a downward trend, which may be related to the disaster prevention measures taken by people in engineering activities. 53correlation analysis between multifractal characteristics of regional geomorphology and development of geological disasters similarly, when analyzing the spatial relationship of the development of debris flow geological disasters, the corresponding relationship table between the depth of geomorphic pleomorphic spectrum and the number of debris flows is established, as shown in the following table: table 3. width of fractal spectrum and quantity of debris flow and its hidden dangers geological point a n 1 0.045 0 2 0.131 5 3 0.084 1 4 0.094 5 5 0.157 4 6 0.116 8 7 0.120 2 8 0.130 0 9 0.066 36 10 0.134 2 11 0.107 28 12 0.137 8 similarly, the relationship among a and the number of debris flow and potential geological disasters is obtained as follows: 0.160.120.080.04 0 6 12 18 24 30 36 a debris flow quantity / piece figure 5. relationship among a and quantity of debris flow and potential geological disasters from the above diagram, the following coupling spatial relations of debris flow disaster can be concluded: in general, with the increase of a, the number of debris flow increases, which indicates that there is a significant correlation between the multifractal characteristics of geomorphic morphology and the geological disasters of debris flow. furthermore, with the increase of a, the complexity of geomorphic morphology and the probability of debris flow disasters increase, the number of debris flow disasters will also increase. the number of debris flow disasters can also be divided into three levels according to the variation range of geomorphic singular index a. when a<0.120, there is a positive correlation between a and the number of debris flow. that is to say, with the increase of a, the number of debris flow disasters increases, but the number of debris flow in this section is less.when 0.120<a<0.134, although the number of debris flow increases with the increase of a, it also presents certain volatility in this range, especially between 0.130 and 0.134, the number of debris flow increases first, then decreases and then increases, indicating that the change range of geomorphic singular index in this region, the disaster phenomenon of debris flow is very unstable.when 0.134 < a < 0.157, the number of debris flow disasters is large, but there is a downward trend, which may be related to the disaster prevention measures taken by people in engineering activities (fan et al., 2018). to sum up, the change characteristics of the number of geological disasters at the twelve selected geological points in the change range of geomorphic singular index a can be roughly divided into three levels: when a<0.100, there is a positive correlation between a and the number of disasters, that is, with the increase of a, the number of disasters increases, but the number of disasters in this section is less; when 0.100 <a <0.134, the fluctuation of the number of disasters is relatively large, indicating that the change range of the geomorphic singular index is in this area, the landslide disaster phenomenon is very unstable, showing a certain degree of volatility; when 0.134<a<0.157, the number of disasters is large, indicating that the change range of the geomorphic singular index is in this area, the geological disaster phenomenon is more likely to occur. according to the spatial analysis of the spatial relationship of the development of geological disasters, combined with the multifractal characteristics of regional landforms, the correlation coefficient between the two is calculated, and the correlation analysis between the multifractal characteristics of regional landforms and the development of geological disasters is completed. calculation of correlation coefficient when calculating the correlation coefficient between the multifractal characteristics of regional landforms and the development of geological disasters, combining the above-mentioned spatial relations of geological disasters and the multifractal characteristics of regional landforms, the weights of the two are calculated firstly, and then the area is divided into t units of the same area, among them, d units are landslide units and d̄ units are non landslide units. in this case, the weight in the geological space can be defined as: w p b d p b d w p b d p b d + − = =        ln ( / ) ( / ) ln ( / ) ( / ) (14) where, w+ and w− are weight values of evidence factor existence and nonexistence respectively, and for the missing area of original data, their weight values are 0. b is the number of units in the area where the evidence factor exists, and b̄ is the number of units in the area where the evidence factor does not exist. the above equation (14) is processed by bayes rule, and the change of the above equation (14) is as follows: ln ( / ... )( ) ( ) ( )o d b b bk k n k n 1 1 2 2 (15) where, o is the probability, o = p / ( 1 − p ), d is the number of unit grids with landslides, bj is the j-th evidence layer, and k (j) is the j-th evidence layer. assuming that there are related variables in equation (15), the logical function of the correlation coefficient is established as follows: f x e e x x( )= +1 (16) according to the definition of expected value of discrete random variable, when p represents the probability of y = 1, we can get: f p e e p p( )= +1 (17) from the above equation, the change of function f p( ) to x is insensitive and slow near f p( ) = 0 or f p( ) = 1, and the degree of nonlinearity is high. in this case, the logistic transformation of f p( ) is introduced, namely: log ( ) log ( )it p it p p = −1 (18) therefore, at this time, the correlation coefficient between the multifractal characteristics of regional geomorphology and the development of geological disasters can be calculated p it p p e e e e x x p p = − + + + log ( ) 1 1 1 (19) 54 yiying xiong from the above equation, p is the final correlation coefficient. using the above calculation and derivation process, when analyzing the correlation between the multifractal characteristics of regional landforms and the development of geological disasters, the correlation is converted into numerical value, and the data between the two are constantly transformed to obtain the change of the other caused by the change of one party, so as to obtain the correlation between the multifractal characteristics of regional landforms and the development of geological disasters, and complete the analysis of the correlation between the two (valenza et al., 2018). experiment experiment preparation the computer experiment environment for experiment preparation is shown as follows: figure 6. experimental environment the of geomorphic multiplicity characteristic and the coefficient of geological disaster development of the experimental preparation area are shown in the following table: table 4. parameters of regional geomorphic multiplicity characteristic and geological disaster development data r n(r) ln(r) lnn(r) 5000 3 8.517193191 1.098612289 4000 5 7.919356191 1.609437912 4250 4 7.824046011 1.609437912 4000 4 7.718685495 1.609437912 3800 4 7.60090246 1.791759469 3500 5 7.522940918 1.791759469 3300 5 7.313220387 1.945910149 2750 5 7.170119543 1.945910149 2500 6 6.907755279 2.197224577 2250 7 6.684611728 2.302585093 2000 7 6.476972363 2.564949357 1850 9 6.214608098 2.772588722 1500 10 5.703782475 2.944438979 1300 13 5.298317367 3.218875825 1000 16 5.010635294 3.713572067 800 19 4.787491743 4.110873864 650 25 4.605170186 4.394449155 500 41 4.442651256 4.615120517 300 61 4.317488114 4.795790546 based on the above data preparation, two traditional correlation analysis methods and the correlation analysis methods for regional geomorphic multifractal characteristics and geological disaster development are used for experiments, and the accuracy of the three correlation analysis methods is compared (yang et al., 2020). analysis of experimental results using the correlation coefficient calculation equation, the coefficients in table 4 is calculated, and finally the experimental results obtained by the three correlation analysis methods are as follows: 0.6 0.8 1.00.40.20.0 0.0 0.2 0.8 0.6 0.4 1.0 traditional analysis method 2 traditional analysis method 2 analysis method in this paper numerical value a na ly si s re su lt va lu e figure 7. experimental results of three analysis methods from the above experimental results, it can be seen that with the increase of the parameters of regional geomorphic multiplicity characteristic and geological disaster development data, the analysis results obtained by traditional analysis method 1 show irregular changes. compared with the traditional analysis method 2 (wu et al., 2020), the correlation analysis method of regional geomorphic multifractal characteristics and geological disaster development is larger, and the error of correlation analysis is larger.however, the results of traditional correlation analysis method 2 show linear alternation between regular and irregular, and the error of correlation analysis is small. however, the correlation analysis method of the multifractal characteristics of regional geomorphology and the development of geological disasters shows a linear change, which shows that there is a strong correlation between the multifractal characteristics of regional geomorphology and the development of geological disasters, and there is almost no error in the actual analysis, so it is suitable for the actual correlation analysis (cheng et al., 2020). conclusions taking topography and geomorphology as the research object, using fractal, multifractal theory and gis technology, combined with geomorphology, disaster statistics and other interdisciplinary theories and methods, taking digital elevation model as the basic data source, the multifractal calculation model of temperature geomorphology and the realization method based on gis technology are proposed, and the spatial-temporal characteristics of main types of geological disasters are analyzed (gao et al., 2018). the related parameters such as the multifractal spectrum, linear interval and surface area distribution probability of the three-dimensional landforms in this area are calculated, the multifractal characteristics of the three-dimensional landforms are revealed, and a coupling model based on the relationship between the multifractal characteristics of the landforms and the number of geological disasters is established (gao et al., 2019). the coupling relationship between the multifractal characteristics of the landforms and the number of main types of geological disasters is analyzed, and the rationality of multifractal feature of geomorphology as a comprehensive quantitative index to study geological hazards is revealed. references alejo, j., montes-rojas, g., & sosa-escudero, w. (2018). testing for serial correlation in hierarchical linear models. journal of multivariate analysis, 165, 101-116. 55correlation analysis between multifractal characteristics of regional geomorphology and development of geological disasters ali, m. y., & lee, j. (2018). petroleum geology of the nogal basin and surrounding area, northern somalia: part 1, stratigraphy and tectonic evolution. journal of petroleum geology, 42(1), 91-120. benammi, m., adnet, s., marivaux, l., yans, j., noiret, c., tabuce, r., surault, j., el kati, i., enault, s., baidder, l., saddiqi, o. (2019). geology, biostratigraphy and carbon isotope chemostratigraphy of the paleogene fossil-bearing dakhla sections, southwestern moroccansahara. geological magazine, 156(1), 117-132. chardon, d., grimaud, j. l., beauvais, a., & bamba, o. (2018). west african lateritic pediments: landform-regolith evolution processes and mineral exploration pitfalls. earth-science reviews, 179(4), 124-146. cheng, h. y., & liu, y. n. (2020). an improved rsu-based authentication scheme for vanet. journal of internet technology, 21(4). cui k., & wang y. x. (2019). structural styles and origin of the dabashan foreland arcuate belt and basin-mountain system in central china. journal of asian earth sciences, 176, 244-252. fan, a. h., liao, l. m, & wu, m. (2018). multifractal analysis of some multiple ergodic averages in linear cookie-cutter dynamical systems. mathematische zeitschrift, 290(1-2), 63-81. galbraith, & de lavergne, d. (2019). response of a comprehensive climate model to a broad range of external forcings: relevance for deep ocean ventilation and the development of late cenozoic ice ages. climate dynamics, 52(1-2), 653-679. gao, n. s., cheng, b. z., hou, h., & zhang, r. h. (2018). mesophase pitch based carbon foams as sound absorbers. materials letters, 212, 243246. doi: 10.1016/j.matlet.2017.10.074 gao, n. s., guo, x. y., cheng, b. z., zhang, y. n., wei, z. y., & hou, h. (2019). elastic wave modulation in hollow metamaterial beam with acoustic black hole. ieee access, 7, 124141-124146. doi: 10.1109/ access.2019.2938250 hansen, n. e., noesen, b. t. nador, j. d., & harel, a. (2018). the influence of behavioral relevance on the processing of global scene properties: an erp study. neuropsychologia, 114, 168-168. kleemann, t., strauss, m., kouraki, k., werner, n., & zahn, r. (2019). prognostic relevance of new onset arrhythmia and icd shocks in primary prophylactic icd patients. clinical research in cardiology(337), 1-7. masson, m., angot, h., le bescond, c., launay, m., dabrin, a., miege, c., le coz, j., & coquery, m. (2018). sampling of suspended particulate matter using particle traps in the rhône river: relevance and representativeness for the monitoring of contaminants. science of the total environment, 637-638, 538-549. leonarduzzi, r. f., abry, p., wendt, h., jaffard, s., touchette, h. (2018). a generalized multifractal formalism for the estimation of nonconcave multifractal spectra. ieee transactions on signal processing, 99, 1-1. sachica, m., & martinez, a. (2018). analysis of correlation in the intercomparison of dc voltage reference standards. ieee transactions on instrumentation and measurement, 99, 1-4. salin, b. m., kemarskaya, o. n., & salin, m. b. (2019). estimating the amplitude of wind waves by correlation analysis of the reverberation signal. acoustical physics, 65(1), 41-51. soloviev, s. g., kryazhev, s., & dvurechenskaya, s. (2018). geology, mineralization, and fluid inclusion study of the kurutegerek au-cu-mo skarn deposit in the middle tien shan, kyrgyzstan. mineraliumdeposita, 80, 79-102. valenza, g., wendt, h., kiyono, k., hayano, j., watanabe, e., yamamoto, y., abry, p., & barbieri, r. (2018). mortality prediction in severe congestive heart failure patients with multifractal point-process modeling of heartbeat dynamics. ieee transactions on biomedical engineering, 65(10), 2345-2354. wu, z., liu, y. n., & jia, x. x. (2020). a novel hierarchical secret image sharing scheme with multi-group joint management. mathematics, 8(3), 12. doi: 10.3390/math8030448 yang, f., yang, f., wang, g. y., kong, t., wang, h., & zhang, c. s. (2020). effects of water temperature on tissue depletion of florfenicol and its metabolite florfenicol amine in crucian carp (carassius auratus gibelio) following multiple oral doses. aquaculture, 515, 9. doi: 10.1016/j. aquaculture.2019.734542 zhang, l., shi, z. l., cheng, m. m., liu, y., bian, j. w., zhou, j. t., zheng, g., & zeng, z. (2019). nonlinear regression via deep negative correlation learning. ieee transactions on pattern analysis and machine intelligence, 99, 1-1. earth sciences research journal earth sci. res. j. vol. 20, no. 4 (december, 2016) letter from the editor at the end of the 2016 and early 2017, earth sciences research journal wants to thank you for making this project a successful journal. four numbers in one year were our principal milestone for 2016. in total, we issued 44 papers expecting to sum our efforts in the earth sciences' state of the art. comparing to recent years, we have increased our figures in publishing and citation close to the 200 percent. for 2017, the commitment is the same: hard-working for the best of your papers and the journal. as part of our mission, we'd like to make a brief list of the highlighted events for 2017 around the world on geophysics and geology, as invitation letter to pay attention them. for march is scheduled in cyprus the fifth international conference on remote sensing and geo-information of the environment. the conference sessions and workshops include presentations regarding the state-of-the-art methods and innovative projects in various areas of remote sensing and geoinformation. one of the most important meetings in the field will be celebrated in april, the association of petroleum geologists annual convention and exhibition 2017, in houston. the aapg is positioned to attract a focused audience of geoscience professionals. the event has an average of 6900 global attendees from 78 countries over the las five years, according to organizers reports. same on april is scheduled the european geoscience union (egu) general assembly, in viena, austria. the egu aims to provide a forum where scientists, especially early career researchers, can present their work and discuss their ideas with experts in all fields of geoscience. in calgary, canada, between 15 and 19 may 2017, will take place the geoconvention. the event expects going beyond the traditional technical sessions, geoconvention 2017 will take a holistic approach to a vast spectrum of topics including financial market and long-term strategy, infrastructures, regulations, safety and environmental protection, among others. for those how are students with innovation ideas, paris will be the host of the european association of geoscientists and engineers 79 conference, student programme. organizers are expecting 6000 experts attending, more than 1000 presentations in facilities over the 11.000 square meters. l e t t e r f r o m t h e e d it o r for september 2017, in australia, will be held the 18th annual conference of international association for mathematical geosciences. the meeting will provide a venue for the presentation of research and developments in mathematical geosciences in the form of oral and poster presentations, in the historic port city of fremantle in western australia. in october is expected the joint meeting of the international sedimentological association (ias) and the association des sédimentologistes français (asf). this event will bring the opportunity to gather the sedimentological and associated geological research communities to share experiences, to discuss new challenges, and to stimulate future rewarding collaborations. in turkey, a country from many papers are coming from to earth sciences research journal, will celebrate the fourth international conference on earthquake engineering and seismology, on october. the principal aim of the conference is to bring all national and international researchers within the earthquake engineering and seismology disciplines together and to facilitate their participation in a scientific environment. in this same month will be celebrated the american association of petroleum geologists international conference and exhibition 2017, in london, and the geological society of america (gsa) 2017 annual meeting, in seattle, the united states. we hope all this information is usable for you and expect the papers we are presenting can attract your attention. our next volume will be ready soon! carlos alberto vargas jiménez. editor in chief, earth sciences research journal runoff curve number (cn model) evaluation under tropical conditions daniela calero mosquera, fanny hoyos villada, enrique torres prieto universidad nacional de colombia *corresponding author: dcalerom@unal.edu.co keywords: flow; hydrology; rainfall; palabras clave: caudal; hidrología; precipitación; issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n4.95321 earth sciences research journal earth sci. res. j. vol. 25, no. 4 (december, 2021): 397-404 h y d r o l o g y record manuscript received: 27/04/2021 accepted for publication: 02/11/2021 abstract in tropical countries the curve number method (cn) of the soil conservation service (scs) is widely used in civil engineering (to calculate drainage infrastructure) as well as in agricultural and environmental studies. however, little is known about the accuracy of cn method in tropical watersheds. to reveal this accuracy, this study compares the scscn model with the methodology proposed by hawkins (1993) where cn (cnq) values are determined with field data measured in an andean micro-basin. for this, cn values for the tropical micro watershed “la vega” were identified using the tables proposed by the scs and its respective antecedent moisture and slope corrections (cnamc2s); subsequently cnq values were derived from 55 events in which runoff and rainfall were measured. according to the results obtained, it was found that in the tropical micro watershed evaluated in this study, the scs-cn model overestimated runoff. the scs-cn model is strongly influenced by the antecedent humidity and its impact becomes stronger in high rainfalls records; however, the antecedent humidity influence was not observed in measured runoff data. despite the scs-cn model did not show a direct relationship between cn and rainfall, this relationship was observed in measured data. furthermore, a clear relationship between runoff and maximum rainfall intensity was found, so it is considered that including this factor in future research can improve the model’s predictions. this study shows that scs-cn model has some inaccuracies and it requires further studies to know its applicability in tropical conditions. evaluación de la escorrentía del número de curva (cn) en condiciones tropicales resumen en países tropicales el número de curva (cn) del servicio de conservación de suelos (scs) es ampliamente usado en ingeniería civil (para calcular infraestructura de drenaje), así como en estudios agrícolas y ambientales. sin embargo, se conoce poco sobre la precisión del modelo scs-cn en cuencas tropicales; para revelar esta precisión, este estudio compara la metodología del scs-cn con la metodología propuesta por hawkins (1993), donde se determinan valores de cn (cnq) a partir de datos de campo medidos en una microcuenca andina. para esto, se identificaron los valores de cn para la microcuenca tropical “la vega” utilizando las tablas propuestas por el scs con sus respectivas correcciones por humedad antecedente y pendiente (cnamc2s); posteriormente, los valores de cnq se derivaron de 55 eventos en los que se midieron lluvia y escorrentía. de acuerdo con los resultados obtenidos, se encontró que, en la microcuenca tropical evaluada en este estudio, el método del scs sobreestimó la escorrentía. el modelo del scscn está fuertemente influenciado por la humedad antecedente y su impacto es más fuerte en registros de altas precipitaciones, sin embargo, la influencia de la humedad antecedente no se observó en las escorrentías medidas. a pesar de que el modelo scs-cn no presentó una relación directa entre cn y precipitación, esta relación si se observó en los datos medidos, además, se encontró una clara relación entre escorrentía e intensidad máxima de precipitación, por lo que se considera que incluir este factor en futuras investigaciones puede mejorar las predicciones del modelo. este estudio muestra que el modelo scs-cn tiene algunas inexactitudes y requiere más estudios para conocer su aplicabilidad en condiciones tropicales. how to cite item: calero mosquera, d., hoyos villada, f., & torres prieto, e. (2021). runoff curve number (cn model) evaluation under tropical conditions. earth sciences research journal, 25(4), 397-404. https://doi. org/10.15446/esrj.v25n4.95321 mailto:dcalerom@unal.edu.co https://doi.org/10.15446/esrj.v25n1.74167 https://doi.org/10.15446/esrj.v25n4.95321 https://doi.org/10.15446/esrj.v25n4.95321 398 daniela calero mosquera, fanny hoyos villada, enrique torres prieto introduction the rainfall – runoff relationship in a watershed involves a set of physical and chemical processes which intervene at different spatial and temporal scales (plesca et al., 2012; ocampo et al., 2014;), and in turn, the watershed is influenced by weather conditions, soil properties, land use and topographic characteristics (marín et al., 2014; martínez et al., 2017). however, the scarce information in the andean watershed is a limitation in order to analyze these relationships, which is why hydrological models have been implemented, as a key tool to understand the runoff generation, provide hydrological forecast and therefore, contribute to the planning and management of hydrological resources. (chen et al., 2020; plesca et al., 2012). one of the widely used models is the cn curve number method was developed in 1972 by the soil conservation service (scs) of the u.s. department of agriculture (usda) as a simple application methodology for calculating runoff in a rainfall event and whose criterion does not require the study watersheds to be gauged or instrumented (aparicio mijares, 1992; chow et al., 1994; ibáñez-castillo et al., 2014; pérez nieto et al., 2015). in the scs-cn model, the generated runoff is obtained from the recorded rainfall, the initial abstractions or losses (considered as 20% of maximum soil retention) and curve number (cn), which is obtained from tables designed according to soil type, land use, conservation practices in agricultural areas, and hydrological condition (chow et al., 1994; ibáñez-castillo et al., 2014). according to ibáñez-castillo et al. (2014) and montiel gonzaga et al. (2019), the scs-cn model is one of the most used techniques in the hydrological field due to its simplicity and medium information requirement (pacheco moya, quiala ortiz, & martinez hernandez, 2018), even it is one of the methods used by soil and water assessment tool (swat) model to predict runoff (neitsch et al., 2011). nevertheless, the model has been subject to several critical revisions by authors as hawkins (1993), ponce & hawkins (1996), lopez (2001), mishra et al. (2007) y paz-pellat (2009). that is how hawkins (1993) stated that it is difficult to accurately select the appropriate cn from the tables and that the runoff calculation is much more sensitive to the chosen cn than to the rainfalls presented because rainfall records are more closely tracked, while the cn values are unique to the area studied or analyzed and are only found in the tables developed by the scs. in view of the foregoing, hawkins (1993) proposed to use the scs-cn model in an inverse way to calculate cn. in order to do this, hawkins solved the mathematical equations for the scs-cn for model (eq. 1-2) and let rainfall and runoff as model variables (eq. 7). in this way, it is possible to calculate cn with the rainfall and runoff data in a watershed. however, this methodology shows that there is not a unique cn value for a watershed, but a secondary relationship emerges between cn and the rainfall, giving rise to a graph which characterizes the type of response pattern of the watershed and is classified as “complacent”, “violent” or “standard”, the latter being the most common scenario (hawkins, 1993). the evaluation proposed by hawkins (1993) have been implemented in different parts of the world which had runoff observed data as usa (hawkins, 1993; durán-barroso et al., 2016; santikari & murdoch, 2019), argentina (ares, varni, et al., 2012), mexico (ibáñez-castillo et al., 2014; pérez nieto et al., 2015), brasil (da cunha et al., 2019; durán-barroso et al., 2016) india (gundalia & dholakia, 2014; lal et al., 2015, 2017, 2019) and south korea (ajmal et al., 2015); in all these studies a great difference between the cn of scs-cn model and the cn derived from observed data was found, showing that scs-cn method cannot calculate runoff properly. in addition, other deficiencies and limitations in the model have been considered, ponce & hawkins (1996) mentioned that the method does not take into account the spatial and temporal variability of infiltration and other losses by abstractions. in this regard, merz & bárdossy (1998) said that spatial variability might have a strong influence in the runoffrainfall relationship. likewise, lópez (2001) stated that calculation methodology does not take into account the effects of rainfall intensity, it means, that for the calculation carried out in two selected events, with the same rainfall but of different duration, the runoff will be the same. in the case of tropical conditions, no evaluation studies of the scs-cn model have been found, but the scientific evidence presented here highlights the importance of evaluating the scs-cn model, where the frequency of intense rainfall has increased, its records show high spatio-temporal variability on low scales, and in addition, temporal information is scarce (marín et al., 2014). in colombia, the scs-cn model is widely used to calculate runoff, mainly due to the easy implementation of the model. however, the absence of flow monitoring stations (cvc, 2015) has been an impediment to the validation of the scs-cn model, therefore, it is not really known whether the model works properly under tropical conditions. then, the current investigation aims at evaluating the curve number model (cn) under conditions of a tropical andean micro watershed using the methodology proposed by hawkins (1993). materials and methods micro watershed study: description of la vega la vega micro watershed is located in the municipality of palmira, valle del cauca, colombia; its drainage area is immersed in the aguaclara sub watershed, which is the main tributary of the bolo river. the study area comprises 3°30’33.12’’ to 3°29’53.52’’ north and 76° 10’ 54.48’’ to 76°11’44.52’’ west, with an extension of 114.08ha, an altitudinal range of 1400 to 1900 m.a.s.l. an average slope of 21% and average multiannual monthly rainfall of 1800 mm. the average multi-annual monthly rainfall in the area presents a bimodal regime, with two wet periods in the months of march-may and octoberdecember, as well as two periods of less rainfall in the months of januaryfebruary and june-september. groundcover in the micro watershed (fig. 1) are permanent shrub crops (coffee crops) (53%), cultivated grass (34%), shrubs and bushes (13%). soils (fig. 2) are typic dystrudepts (85%) and typic humudepts (7%), which are characterized by being deep, well drained, with fine textures, medium to low organic carbon content in depth and low moisture retention; meanwhile, typic udorthents soils (8%) are superficial, limited by the presence of rock fragments, moderately well drained and low moisture retention (cvc; igac, 2014). figure 1. coverage map of la vega microwatershed. rainfall and flow measurement the study micro watershed kept a detailed record of rainfall and flow with a time resolution of 15 minutes. the information was collected from june 2015 to july 2017 by the centro de investigación de la caña de azúcar de colombia – cenicaña. rainfalls in la vega were recorded using a davis vantage pro-2 weather station located into the micro watershed at an altitude of 1607 m. flow monitoring was performed continuously at the micro watershed hydrometric station, consisting of a water level sensor (through a keller brand acculevel pressure sensor), installed in a narrow crest weir to ensure a control section and accurate flow measurement. 399runoff curve number (cn model) evaluation under tropical conditions figure 2. soil map of la vega microwatershed. runoff calculation with scs-cn method the scs-cn method presents runoff as a function of rainfall (daily rainfall or 24-hour rainfall) and soil water storage capacity, calculated by the expression: where p is the rainfall in mm; is the initial abstraction of the watershed in mm; q is the runoff generated in mm and s is the maximum soil retention in mm, the latter related to cn by: the cn was obtained from tables according to soil type, land use, conservation practices in agricultural areas, and hydrological condition (chow et al., 1994; neitsch et al., 2011b; ibáñez-castillo et al., 2014). for this purpose, information was obtained from the semi-detailed soil survey at a scale of 1:25000 (fig. 1) of the watersheds prioritized by the environmental authority of the department: corporación autonoma regional del valle del cauca (cvc) (cvc; igac, 2014); in the same way, information on land use was obtained (fig. 2), which was validated and compared with information collected in field visits. to select the hydrological soil group, the four groups of soils described by chow et al. (1994) were taken into account, these groups are presented in table 1, also in this table can be observed the soil type associated with the hydrological soil group. table 1. hydrological soil group soil groups description la vega soils group a deep sand, deep soils deposited by the wind, aggregated silts group b shallow deep soil deposited by the wind, sandy loam typic udorthents group c clay loams, shallow sandy loam, soils low in organic content, and soils usually high in clay typic dystrudepts and typic humudepts soil groups description la vega soils group d soils that swell significantly when wet, heavy plastic clays, and certain saline soils. source: adapted from chow et al. (1994) once the hydrological soil group for each soil type was found, the cn was determined for the different land uses in the micro watershed according to the cn tables presented by neitsch et al. (2011b). in order to obtain an average cn for the hole watershed, cn was weighted by area using arcgis 10.4.1 following the methodology proposed by chow et al. (1994). the selected cn was corrected for antecedent humidity (eq. 3,4) and slope (eq. 5) (ibáñez-castillo et al., 2014; pérez nieto et al., 2015); in terms of humidity, there are three types of cn classified according to previous rainfall (during the 5 days prior to the runoff event), to determine the adjustment. thus, cn data recorded in tables by the scs are assumed in normal antecedent moisture condition (amc ii) (chow et al., 1994). criteria for defining the antecedent moisture condition are presented in table 2. table 2. antecedent moisture condition classification amc group total 5-days antecedent rainfall (mm) dormant season growing season i ≤ 13 ≤ 36 ii 13 28 36 – 53 iii ≥ 28 ≥ 53 source: adapted from (ferrer polo, 2000) for the correction for antecedent moisture (table 2), the growing station was selected, since in both periods of the bimodal regime there are rainfall records of significant magnitude. calculation for equivalent cn is presented in equations 3-4. the slope correction (eq. 5) was introduced because cn tables proposed by the scs were developed for use in watersheds with slopes of less than 5% (nash & sutcliffe, 1970): where is the curve number corrected for slope; is the curve number for the amc (iii) group; is the curve number as read in the scs tables; slp is the mean slope of the watershed (m/m). in this study, the cn derived from scs tables and adjusted by slope and moisture will be termed cnamc2s runoff calculation from observed flow (baseline method j + n) to determinate actual runoff in the micro watershed, 55 rainfall events and its flow responses were analyzed during june 2015 and july 2017. these 55 events were considered because presented a clear flow response as a result of 400 daniela calero mosquera, fanny hoyos villada, enrique torres prieto a single rainfall event. these clear flow response events were easily identified in the hydrograph by an ascent curve (qa), a single peak flow (qmax) and finally a descent curve (qd). in the present study, only events of a single peak were considered and were referred to as unique events. figure 3 represents a typical case of a unique event presented in the study area. different kinds of flow responses were not considered because runoff cannot be calculated by the baseline method j + n. the start of the event was determined when the level sensor showed a significant increase in river flow as a consequence of rainfall; this start was projected horizontally until the moment of peak on the hydrograph (j) and ended with a diagonal according to the fixed base flow method (chow et al., 1994), where it is assumed that runoff ends in a fixed time n (eq. 6) in the recession segment. where n is the number of days after the peak on the hydrograph, and a corresponds to the watershed area, given in km2. figure 3 graphically explains this method. the event initial flow was determined as an average for flows recorded two hours prior to the start of the event; runoff was calculated as the area under the hydrograph curve and delimited below by the j and n lines (inclined); runoff (mm) was calculated by dividing the total runoff volume by the micro watershed area. influence of rainfall intensity on runoff generation to evaluate the influence of precipitation intensity in runoff generation, the methodology proposed by wischmeier y smith (1978) to determine maximum 30 minutes precipitation intensity (i30) in an erosive event was used. this methodology is considered into the erosivity factor of the universal soil loss equation (usle). i30 values were obtained for the 55 analyzed events and were related with observed runoff (baseline method j+n). determination of the observed cn (cnq) to determine actual cn (cnq) an inverse scs-cn modeling was performed (hawkins, 1993). for that, each of the 55 actual rainfall events data and its runoff responses (calculated by j + n methodology) were introduced in equation 7. the equation 7 results from solving the maximum retention of the soil (s) in equation 1a and substituting this value in equation 2. 2.6 estimation of adjustment between cnamc2s and cnq to measure the level of adjustment between cnamc2s and cnq, the nash efficiency index (e) was used (nash & sutcliffe, 1970), which has been widely used in the evaluation of hydrological models (yusop, chan, & katimon, 2007; pérez nieto et al., 2015), calculated from the following equation: the nash efficiency index can have values from -∞ to 1; when the nash index is closer to 1, hydrological model prediction (in this case scs-cn) is better. asymptotic adjustment method to determine the type of response pattern of the studied micro watershed, the rainfall and runoff values measured in the watershed (55 unique events summarized in section 2.4) were ordered independently and in descending order, subsequently the cn values were determined for each pair of data (rainfall and runoff) using the equation 7 then, the relationship between rainfall and cn was plotted. according to hawkins (1993), three types of patterns can occur in this rainfall-cn relationship: “standard”, “violent” and “complacent” behavior. in cases of responses patterns “standard” or “violent”, the cns approach a constant value with increasing rainfall, and these are given by the equations 9-10 respectively. figure 3. graphical explanation of baseline method j+n for base flow separation 401runoff curve number (cn model) evaluation under tropical conditions where is a constant value approached as and is a fitting constant (different for each behavior).the variable describes the data set for larger rainfall events (hawkins, 1993). in the “complacent” behavior, evidences no appreciable tendency to achieve a stable cn value (hawkins, 1993). deeper mathematical analyses of equations 1-2 were developed, where the effects of the variables ia and s were shown to be minimal; thus, the cn was rethought as a rainfall-dependent variable (eq. 11) and was considered the lower limit of the curve number definition (hawkins, 1973). results and discussion in the 55 unique events analyzed in the micro watershed, the maximum recorded rainfall was 74 mm and the minimum was 0.3 mm, while the antecedent moisture (five previous days) varied from 0 to 105 mm. scs-cn method cn values obtained from scs tables the cn values were calculated by tables according to scs method. due to micro watershed heterogeneity (soil and land use), there were several cn values for the micro watershed. table 3 presents the cn weighted by area for “la vega” micro watershed, its slope correction (cn2s), and its antecedent humidity correction (cnamc2s). table 3. cn found by the scs methodology. weighted cnamc2s micro watershed cn cn2s cn i cn ii cn iii la vega 73 76.8 58.2 76.8 88.7 runoff calculated from cn tables (cnamc2s) the runoff scs-cn method calculated from cn tables was 2.3 mm on average, with a maximum value of 39.6 mm and a minimum of zero; in figure 4, lines present the runoff calculated from cn tables (cnamc2s) with respect to rainfall in the la vega micro watershed in its three humidity conditions (cn i, cn ii, cn iii) (feyereisen et al., 2008). the points represent 55 events studied in the micro watershed and the runoff generation according to the scs-cn model. figure 4. evaluation of runoff (cnamc2s) rainfall relationship in la vega, according to the scs-cn model. figure 4, shows that 42 events out of the 55 analyzed correspond to low antecedent moisture (cn i line), it means most of the cn data were below the “mean condition” (cn ii) defined by the scs, just 5 events corresponded to cn ii condition and 8 to cn iii condition it is important to note that scs-cn model is strongly influenced by the antecedent moisture because the correction of the cn from the antecedent humidity affects in an inverse way ia (as described in equations 1 and 2) and, consequently, in the generation of runoff. to show the humidity correction effect, similar rainfall events recorded in the micro watershed (low: 15mm and high: 70mm) were taken under the three antecedent moisture conditions, which are presented in table 4. in this table, it can be observed that rainfalls close to 70 mm (high rainfall) presented very different runoffs (from 5 to 39.6mm), this was due to the fact that from condition i to condition ii the cn changes 32%, but the impact in runoff calculated by the model was 414%. from condition i to condition iii the impact was even stronger, with a cn change of 52%, runoff increases by 692%. in low rainfall, records around 15mm were taken. for this case, the impact was not so strong. all this indicates that the impact of the antecedent moisture condition becomes stronger in high rainfall. runoff calculation from observed flow (baseline method j + n) runoff calculated from observed flow (j+n baseline methodology) had lower result than runoff of scs-cn methodology. runoff average value was 0.6 mm with a maximum of 6.5 mm and a minimum of zero. in figure 5 the three antecedent moisture curves were plotted (as shown in fig. 4) and over this data was plotted the relation between rainfall and runoff calculated from observed flow (j+n baseline methodology) as black dots. here, it can be observed that the runoff calculated from j+n methodology (runoff j+n) presented a trend toward the cni curve set forth in the scs-cn model; likewise, its values were lower than those obtained in the scscn methodology. for this reason, no points were observed in the cnii and cniii lines despite the fact that runoffs in both methodologies were calculated from the same rainfall events observed in the micro watershed. figure 5. evaluation of the runoff (j+n) – rainfall relationship in “la vega”, according to the scs-cn model. similar results were found by lópez (2001), who conducted a literature review where he stated that runoffs less than 12.7 mm (half an inch) make the scs-cn method less accurate. on the other hand, studies from ares et al. (2012) and da cunha et al. (2019) found that events in hydrological condition iii overestimated modelled runoff. based on this, it can be deduced that the cnamc2s selected from tables, corrected for slope and antecedent humidity, would be overestimating the micro watershed runoff. in addition, rainfall events where found the scs-cn model assume runoff as (zero), while the observed data reported runoff. in cases of low runoff, the limit imposed by the scs-cn model (p> = ia) based on the relationship between ia and s (λ = 0.20 in equation 1c) may not be adequate. this agrees with authors such as ajmal et al. (2015) and lal et al. (2015, 2017, 2019) who stated that the optimized initial abstraction ratio (λ) values showed that the original assumption of the λ of 0.20 is unusually high. 402 daniela calero mosquera, fanny hoyos villada, enrique torres prieto influence of rainfall intensity on runoff generation when evaluating the influence of the intensity of rainfall on the generation of runoff, the results showed that there is a clearer relationship between maximum rainfall intensity and runoff (r2 = 0.96) than between amount of rainfall and runoff (r2 = 0.42) (fig. 6). since the runoff process is the result of the interaction of the rainfall rates (intensity) and the infiltration rate, it makes more sense to use the rainfall intensity as a variable in the model instead of the amount of rainfall as expressed by santikari & murdoch (2019), who consider that taking temporary variations into account can improve the performance of the scs-cn model. however, it is more complex to measure intensity than amount of rainfall, especially at the time the scs-cn model was implemented but now with the current development of automatic stations, it may be feasible to incorporate rainfall intensity in runoff calculation. figure 6. relationship between runoff and rainfall (a) and maximum rainfall intensity (b). cn derived from runoff (cnq) the cnq was calculated using the scs-cn model in an inverse way (eq. 7), which already had the influence of slope and antecedent moisture, due to the fact that its calculation was made from real runoff data in the micro watershed for the 55 unique events of the micro watershed, the corresponding cnq were found; their values oscillated between 48 and 99 with a mean of 80, indicating that most of the cnq found were high (>70) unlike the cnamc2s which showed a tendency of 58.2 (cni). in addition, each event had a different cnq value and it was not possible to classify them in unique values of cni, cnii, cniii as in the case for cnamc2s. similarly, the nash efficiency index (nash & sutcliffe, 1970) was -3.04, indicating that there was no similarity between the values of cnamc2s and cnq, therefore, the two cn were not comparable, even though both start from the same model (eq. 1 2) proposed by the scs. in this case, by finding a nash efficiency of less than 0 it can be inferred that the cnq is a better predictor than cnamc2s (yusop et al., 2007; pérez nieto et al., 2015) ares et al. (2012) stated that the variety of cnq values could initially be related to the heterogeneity of soils and micro watershed coverages and secondly, to the antecedent moisture of each event. however, in the figure 7 it can be evidenced that, unlike cnamc2s which is stratified by the antecedent moisture (three horizontal lines), the cnq was not influenced by this factor (white triangles), for example, for an antecedent moisture of 18 mm, the cnq varies between values of 62 and 96 (fig. 7a). thus, it was deduced that the antecedent moisture correction proposed in the scs-cn methodology is a parameter that does not work properly in the actual micro watershed conditions. as mentioned in the previous section, it is possible that other aspects have a great impact on runoff, such as rainfall distribution and intensity which is not considered by scs-cn model. figure 7. cnamc2s (black circles, rhombus and squares) and cnq (white triangles) variation with respect to antecedent moisture condition (amc) in la vega. a. amc i. b. amc ii. c. amc iii. 3.4 standard asymptotic adjustment to identify the micro watershed response type, the cnq values found were related to the recorded rainfall, as proposed by hawkins (1993), giving table 4. example of some rainfall events differences in rainfall date (dd/mm/yyyy) pp (mm) antecedent. moisture (mm) antecedent. moisture condition (amc) cnamc2s initial abstractions ia (0.2*s) runoff (mm) 30/09/2016 69.3 13.2 i 58.2 36.54 5.0 high rainfall 13/09/2016 74.4 37.1 ii 76.8 15.35 25.7 13/03/2017 67.8 105.2 iii 88.4 6.67 39.6 low rainfall 07/05/2017 15.0 20.8 i 58.2 36.54 0.0 08/05/2017 15.5 35.8 ii 76.8 15.35 0.0 10/05/2017 13.7 56.9 iii 88.4 6.67 1.2 403runoff curve number (cn model) evaluation under tropical conditions rise to the graph presented in figure 8. further, cn0 (eq. 11) was calculated and was found that it is always lower than the cnq (eq. 7), which allowed to verify that variable ia was actually a runoff lower limit as suggested by the model. figure 8. cn0 and cnq variation with respect to rainfall. figure 8 showed an inverse proportional relationship between the cnq and the rainfall recorded, as the latter increases, the cnq tends to decrease. this allow to conclude that there is a clear relationship between the cnq and the recorded rainfall, as it was expected. although, this relationship did not reach a stable cnq value, thus coinciding with the “complacent behavior” type response pattern (hawkins, 1993), whose definition suggests that micro watershed of the present study did not respond in accordance with equation 1a, at least not within the range of the analyzed data. in these cases, hawkins (1993) stated that it was possible to fit the equation for the standard behaviour (eq. 9) with the least squares method, assuming that the data set is merely the lower rainfall end of the standard behavior pattern, and that the watershed will perform in the extrapolated standard pattern when these larger storms are eventually encountered. therefore, the “standard asymptotic adjustment method” was used, obtaining the following adjustment =0.95): in equation 12, a cn∞ of 44.62 was defined for high rainfall events and it describes the cnq trend in the case of rainfall events of significant magnitude, however it is a much lower cn value than those analyzed in the set of data presented in figure 8 (hawkins, 1993). according to figure 8, events with rainfalls between 20 to 50 mm were related to a cn between 60 to 80, and rainfall lower than 20 mm with a cn greater than 80, the latter being the most frequent events in the micro watershed studied here. according to hawkins (1993), values such as those observed at the beginning of figure 8, are due to the fact that low rainfall with low cn do not generate significant runoff. similar results were found by ares et al. (2012), in the videla stream watershed, they found a “standard” category response pattern (hawkins, 1993), and thus obtained a single cn of 57 for large events in their study area. ares et al. (2012) had similar rainfalls in the analyzed events compared to our data (0-120mm), although ares et al. (2012) average events were longer (69h) than the average events analyzed here (20h). data presented here showed that cnamc2s presents a direct relationship with the antecedent moisture condition (fig. 7), however, the relationship between amc-cnq was null; on the other hand, the variation of the cnq obtained from observed data (fig. 8) might be attributed to the effect of the intensity rainfall in the generation of runoff, evidenced in figure 6a; thus, the intensity runoff could be the object of study in future investigations of the scscn model. conclusions the estimation of the adjustment between the curve numbers obtained from the table and those calculated from the observed runoff (cnamc2s and cnq) showed a high difference between these values (e=-3.04), indicating that scs-cn method does not calculated runoff properly in la vega microwatershed. according to the calculations made for the runoff estimation, it was found that the cn-scs model overestimates the runoff generated in the micro watershed, nevertheless, in some rainfall events where the scs-cn model assume no runoff (zero), the observed data reported runoff, this was related to the limit imposed by the scs-cn model (p> = ia) based on the relationship between ia and s (λ = 0.20) which may not be adequate. the scs-cn model is strongly influenced by the antecedent moisture correction; however, according to observed data (cnq), the amc does not work properly in “la vega” micro watershed real conditions. a cn-rainfall relationship emerges when cns are calculated from rainfall-runoff data observed (cnq), although this relation is not considered into de scs-cn model. thus, the variation of the cnq in the micro watershed might be attributed to the fact that the intensity rainfall is not taken into account in the scs-cn model. in “la vega” micro watershed a clear relationship between maximum rainfall intensity and runoff (r2 = 0.96) was found. this relationship is not considered in scs-cn model. thus, taking rainfall intensity into account in future investigations, can improve the performance of the scs-cn model. this study highlights the importance of assessing the applicability and limitations of the models that were developed a long time ago (such as the scscn model) and which are still widely used without the knowledge of the real impact of being applied in different conditions of its development, in this case, tropical conditions. references ajmal, m., moon, g. w., ahn, j. h., & kim, t. w. (2015). investigation of scs-cn and its inspired modified models for runoff estimation in south korean watersheds. journal of hydro-environment research, 9(4), 592–603. https://doi.org/10.1016/j.jher.2014.11.003 aparicio mijares, f. j. (1992). fundamentos de hidrología de superficie. limusa (noriega editores), mexico d.f., 291 págs. ares, m. g., chagas, c., & varni, m. (2012). predicción de la escorrentía y estimación de la humedad antecedente en una cuenca aforada. asociación argentina ciencia del suelo, 30(2), 151–160. ares, m. g., varni, m., chagas, c., & entraigas, i. (2012). calibración del número n de la curva de escurrimiento en una cuenca agropecuaria de 116km2 de la provincia de buenos aires, argentina. agrociencia, 46(6), 535–541. chen, x., lee, r. m., dwivedi, d., son, k., fang, y., zhang, x., graham, e., stegen, j., fisher, j. b., moulton, d., & scheibe, t. d. (2021). integrating field observations and process-based modeling to predict watershed water quality under environmental perturbations. journal of hydrology, 602(125762). https://doi.org/10.1016/j.jhydrol.2020.125762 chow, v., maidment, d., & mays, l. (1994). hidrología aplicada. mcgraw-hill, 584 págs. cvc, & igac. (2014). levantamiento semidetallado de suelos escala 1:25.000 de las cuencas priorizadas por la corporación autónoma regional del valle del cauca cvc. da cunha, z. a., beskow, s., de moura, m. m., & steinmetz, a. a. (2017). aplicabilidade do método número da curva (cn/scs) com base em valores de cn tabelados: estudo de caso na bacia hidrográfica do arroio cadeia. xxii simpósio brasileiro de recursos hídricos, florianópolis – sc, novembro 1-8. durán-barroso, p., gonzález, j., & valdés, j. b. (2016). improvement of the integration of soil moisture accounting into the nrcs-cn model. journal of hydrology, 542, 809–819. https://doi.org/10.1016/j.jhydrol.2016.09.053 https://doi.org/10.1016/j.jher.2014.11.003 https://doi.org/10.1016/j.jhydrol.2020.125762 https://doi.org/10.1016/j.jhydrol.2016.09.053 https://doi.org/10.1016/j.jhydrol.2016.09.053 404 daniela calero mosquera, fanny hoyos villada, enrique torres prieto ferrer polo, f. j. (2000). recomendaciones para el cálculo hidrometeorológico de avenidas. centro de estudios y experimentación de obras públicas, madrid, españa, 76 págs. feyereisen, g. w., strickland, t. c., bosch, d. d., truman, c. c., sheridan, j. m., & potter, t. l. (2008). curve number estimates for conventional and conservation tillages in the southeastern coastal plain. journal of soil and water conservation, 63(3), 120–128. https://doi.org/10.2489/ jswc.63.3.120 galvão do valle junior, l. c., bicca rodrigues, d. b., & sanches de oliveira, t. p. (2019). initial abstraction ratio and curve number estimation using rainfall and runoff data from a tropical watershed. revista brasileira de recursos hídricos, 24(5). https://doi.org/10.1590/23180331.241920170199 gundalia, m., & dholakia, m. (2014). impact of monthly curve number on daily runoff estimation for ozat catchment in india. open journal of modern hydrology, 4(4), 144–155. http://dx.doi.org/10.4236/ ojmh.2014.44014 hawkins, r. h. (1973). improved prediction of storm runoff in mountain watersheds. journal of the irrigation and drainage division, 99(4), 519–523. https://doi.org/10.1061/jrcea4.0000957 hawkins, r. h. (1993). asymptotic determination of runoff curve numbers from data. journal of irrigation and drainage engineering, 119(2), 334–345. https://doi.org/10.1061/(asce)0733-9437(1993)119:2(334) ibáñez-castillo, l. a., alonso-sánchez, h., arteaga-ramírez, r., & vázquez-peña, m. a. (2014). identificación de curva de escurrimiento en tres microcuencas del río coatán. tecnología y ciencias del agua, 5(4), 153–161. lal, m., mishra, s. k., & kumar, m. (2019). reverification of antecedent moisture condition dependent runoff curve number formulae using experimental data of indian watersheds. catena, 173, 48–58. https://doi. org/10.1016/j.catena.2018.09.002 lal, m., mishra, s. k., & pandey, a. (2015). physical verification of the effect of land features and antecedent moisture on runoff curve number. catena, 133, 318–327. https://doi.org/10.1016/j.catena.2015.06.001 lal, m., mishra, s. k., pandey, a., pandey, r. p., meena, p. k., chaudhary, a., jha, r. k., shreevastava, a. k., & kumar, y. (2017). evaluation of the soil conservation service curve number methodology using data from agricultural plots. hydrogeology journal, 25(1), 151–167. https://doi. org/10.1007/s10040-016-1460-5 lopez, r. (2001). consideraciones acerca de los límites de aplicación del método del número de curva del soil conservation service. ciencia y técnica, 66, 92–97. http://hdl.handle.net/10459.1/46546 marín, j. p., ocampo, o. l., & velez, j. j. (2014). eventos hidroclimáticos extremos en cuencas andinas: caso de estudio cuenca del río chinchiná. xxi seminario nacional de hidráulica e hidrología, boyacá colombia, septiembre. martínez, e., coello, c., & feyen, j. (2017). análisis comparativo del comportamiento de la escorrentía de tres microcuencas andinas con diferente régimen de precipitación y cobertura vegetal. maskana, 8(1), 129–144. https://doi.org/10.18537/mskn.08.01.12 merz, b., & bárdossy, a. (1998). effects of spatial variability on the rainfall runoff process in a small loess catchment. journal of hydrology, 212– 213(1–4), 304–317. https://doi.org/10.1016/s0022-1694(98)00213-3 mishra, s., suresh babu, p., & singh, v. p. (2007). scs-cn method revisited. water resources publications, colorado, 36 págs. montiel gonzaga, r., prado hernández, j. v., vázquez peña, m. a., ibáñez castillo, l. a. & pascual ramírez, f. (2019). evaluación del escurrimiento superficial de los modelos de curva numérica y green-ampt en la cuenca río chapingo, méxico. terra latinoamericana, 37(3) 291-301. https://doi.org/10.28940/terra.v37i3.484 nash, j. e., & sutcliffe, j. v. (1970). river flow forecasting through conceptual models part i-a discussion of principles. journal of hydrology, 10(3), 282–290. https://doi.org/10.1016/0022-1694(70)90255-6 neitsch, s., arnold, j., kiniry, j., & williams, j. (2011). soil & water assessment tool theoretical documentation version 2009. texas water resources institute, 647 págs. ocampo, o. l., velez, j. j., & marín, j. p. (2014). intensificación de eventos extremos de precipitación en cuencas andinas colombianas y su influencia en el régimen de caudales: caso del río chinchiná. xxv congreso latinoamericano de hidráulica, santiago, chile, agosto, 1-9. pacheco moya, r. m., quiala ortiz, e., & martinez hernandez, i. (2018). determinación del parámetro número de curva en la cuenca las coloradas en un entorno sig. ingeniería hidráulica y ambiental, 39(2), 60–71. paz-pellat, f. (2009). mitos y falacias del método hidrológico del número de curva del scs/nrcs. agrociencia, 43(5), 521–528. pérez nieto, s., ibáñez castillo, l. a., arellano monterrosas, j. l., fernández reynoso, d. s., & chávez morales, j. (2015). curvas de escurrimiento para tres microcuencas de la cuenca huixtla, chiapas, méxico. revista mexicana en ciencias agrícolas, 6(8), 1729–1742. plesca, i., timbe, e., exbrayat, j. f., windhorst, d., kraft, p., crespo, p., vaché, k. b., frede, h. g., & breuer, l. (2012). model intercomparison to explore catchment functioning: results from a remote montane tropical rainforest. ecological modelling, 239, 3–13. https://doi.org/10.1016/j. ecolmodel.2011.05.005 ponce, v. m., & hawkins, r. h. (1996). runoff curve number: has it reached maturity? journal of hydrologic engineering, 1(1), 11–19. https://doi. org/10.1061/(asce)1084-0699(1996)1:1(11) santikari, v. p., & murdoch, l. c. (2019). accounting for spatiotemporal variations of curve number using variable initial abstraction and antecedent moisture. water resources management, 33(2), 641–656. https:// doi.org/10.1007/s11269-018-2124-0 wischmeier, w., & smith, d. (1978). predicting rainfall erosion losses: a guide to conservation planning. united states departament of agriculture, agriculture handbook n° 537, 58 págs. yusop, z., chan, c. h., & katimon, a. (2007). runoff characteristics and application of hec-hms for modelling stormflow hydrograph in an oil palm catchment. water science and technology, 56(8), 41–48. https:// doi.org/10.2166/wst.2007.690 https://doi.org/10.2489/jswc.63.3.120 https://doi.org/10.2489/jswc.63.3.120 https://doi.org/10.1590/2318-0331.241920170199 https://doi.org/10.1590/2318-0331.241920170199 http://dx.doi.org/10.4236/ojmh.2014.44014 http://dx.doi.org/10.4236/ojmh.2014.44014 https://doi.org/10.1061/jrcea4.0000957 https://doi.org/10.1061/(asce)0733-9437(1993)119:2(334) https://doi.org/10.1016/j.catena.2018.09.002 https://doi.org/10.1016/j.catena.2018.09.002 https://doi.org/10.1016/j.catena.2015.06.001 https://doi.org/10.1007/s10040-016-1460-5 https://doi.org/10.1007/s10040-016-1460-5 http://hdl.handle.net/10459.1/46546 https://doi.org/10.18537/mskn.08.01.12 https://doi.org/10.1016/s0022-1694(98)00213-3 https://doi.org/10.28940/terra.v37i3.484 https://doi.org/10.1016/0022-1694(70)90255-6 https://doi.org/10.1016/j.ecolmodel.2011.05.005 https://doi.org/10.1016/j.ecolmodel.2011.05.005 https://doi.org/10.1061/(asce)1084-0699(1996)1:1(11) https://doi.org/10.1061/(asce)1084-0699(1996)1:1(11) https://doi.org/10.1007/s11269-018-2124-0 https://doi.org/10.1007/s11269-018-2124-0 https://doi.org/10.2166/wst.2007.690 https://doi.org/10.2166/wst.2007.690 the study discuss about the soil moisture estimation using dual polarimetric risat-1. the semiempirical approach of modified dubois model (mdm) derived by (srinivasarao 2013) is worked out using (σ˚hh) and (σ˚hv) for soil moisture estimation using dual polarimetric backscattering image. irs liss iv data have been used to analyze the site suitability of different land use/cover types. the retrieval of backscattering coefficient values (σ˚) from sar is the common principle factor for soil moisture estimation. the surface roughness were measured in the selected sample location, for which the same backscattering values derived from the sar is linearly correlated showing r2 = 0.93. the estimated surface roughness is used for retrieval of dielectric constant using mdm. the dielectric constant derived from mdm in combination with the topps model proposed by (topp 1980), is used to derive satellite based soil moisture estimation. linear regression analysis was performed and the soil moisture derived from sar are well correlated with the volumetric soil moisture showing value of r2 = 0.63. abstract keywords: soil moisture, surface roughness, modified dubois model, topp’s model, risat 1-sar. derivation of soil moisture using modified dubois model with field assisted surface roughness on risat-1 data issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n1.59972 p. thanabalan*, and r. vidhya institute of remote sensing, anna university, chennai – 600025, tamilnadu, india. *thanabalanp@gmail.com earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018): 13-18 este estudio discute la estimación de la húmedad del suelo medida con el radar de imágenes satelitales (risat-1) doble polarimétrico. la aproximación semiempírica del modelo modificado de dubois (mdm), derivada por srinivasa rao (2013), se resuelve a través de (σ˚hh) y (σ˚hv) para la estimación de la húmedad del suelo usando la imagen de retrodispersión doble polarimétrica. la información de las imágenes liss iv se utilizó para analizar la pertinencia de diferentes tipos de uso del suelo y coberturas. la recuperación de los valores de coeficientes retrodispersos (σ˚) del radar de apertura sintética (sar, del inglés synthetic aperture radar) es el factor común principal para la estimación de la húmedad del suelo. la desigualdad de la superficie se midió en la ubicación del ejemplo seleccionado, por lo que los mismos valores retrodispersos derivados del sar están correlacionados linealmente con r2 = 0.93. la rugosidad de la superficie estimada se utilizó para la recuperación de la constante dieléctrica a través del modelo modificado de dubois. la constante dielétrica derivada del mdm, combinada con el modelo de topps, se utilizó para obtener la estimación de la húmedad del suelo partir de datos satelitales. se realizó el análisis de regresión lineal y la húmedad del suelo derivada del sar está bien correlacionada con el índice volumétrico de húmedad del suelo con un valor r2 = 0.63 resumen palabras clave: : húmedad del suelo; desigualdad de la superficie; modelo modificado de dubois; modelo de topp; radar de apertura sintética; record manuscript received: 08/09/2016 accepted for publication: 17/11/2017 how to cite item thanabalan, p., & vidhya, r. (2018). derivation of soil moisture using modified dubois model with field assisted surface roughness on risat-1 data. earth sciences research journal, 22(1). 13-18. doi: http://dx.doi.org/10.15446/esrj.v22n1.59972 medición de la húmedad del suelo a través del modelo modificado de dubois con superficie desigual, basado en información del radar imaging satellite-1 r e m o t e s e n si n g 14 p. thanabalan and r. vidhya introduction in hydrological studies, soil moisture is a critical component that controls the infiltration and runoff rates. the wetness of the soil controls the evapotranspiration rate. although only a small percentage (~ 1%) of the total fresh water budget is contained in the soil layers of the earth surface, it acts as a key parameter in the exchange of mass and energy at the soil-atmospheric layer and also in hydrological processes. the soil surface conditions, soil moisture content, and roughness are important parameters in agriculture and vegetation growth monitoring. technological advancements in satellite remote sensing provide various techniques for measuring the soil moisture across a wider area (anguela 2010). remote sensing is an advanced technology that provides accurate and repetitive spatial data; microwave remote sensing synthetic aperture radar (sar) is particularly useful as they have a tendency to measure the soil parameters under any weather condition (mirmazloumi 2016; radar 2016). for bare agricultural soils, the backscattered radar signals depend strongly on the geometric characteristics (roughness) and dielectric properties (moisture content and soil composition) of the soil (singh 2007; tiwari 2014; gharechelou 2015). the dynamic relationship between the radar signal and surface roughness is stronger in the l, c, and x bands. moreover, sar data is more sensitive to soil roughness at high incidence angles (baghdadi 2006). since the microwave radiation can penetrate through clouds, it aids the acquisition of images in all weather conditions, and also penetrates into the soil for at least 1-5 cm and often upto 15 cm of the root zone level of soil moisture for volumetric analysis. the large variation in the dielectric constant, which depends on the radar backscattering coefficient, has a linear relation with the soil moisture if the roughness condition is absent (terrain 2007). previously, many attempts have been made for the development of semi-empirical algorithms (radar 2016; khabazan 2013; oh 2004; dubois 1995). semi-empirical models have the advantages of both empirical and theoretical models. according to the research carried out by many researchers in barely and sparsely vegetatedareas (capodici et al., 2013), soil moisture predicted by the dubois model is better than the integrated equation model (iem). till date, the dubois model is one of the most successful soil moisture prediction models, with some limitations because it requires quad pole data (hh, hv, vh, and vv). in this study, the risat crs image (spatial resolution of 36 m) acquired at hh & hv polarization have been used for the retrieval of soil moisture in and around the theni district of southern tamil nadu. risat-1 measures the volumetric soil moisture content in all weather conditions. the soil moisture content was derived using modified dubois model for risat-1 data. thus, the overall aim of the study explains the risat sar retrieval of soil moisture both spatially and temporally to improve our understanding of the hydrological behavior of soil. materials and methods a. study area the study area (theni district of tamil nadu) is geographically located from latitudes 9°39’n to 10°30’n and from longitudes 77°00’e to 78°30’e. it is a part of the western ghats, mostly consisting of hilly area with intermittent plains. in theni, agriculture is the primary occupation and 67% of the population lives in rural areas. the part of theni district that has been taken into consideration for this study lies within 10°1’38.22”n and 77°20’18.70”e 9°49’37.12”n and 77°38’43.79”e (fig.1) parts of villages, namely, andipatti, virapandi, odapatti, and chinamanur lie within the geographical extension being studied. it is a semi-arid region with an annual maximum temperature of 38.5°c and minimum temperature of 26.3°c. the average normal rainfall is about 829.8 mm (tamil nadu agriculture university). the soil, however, is sandy and clayey according to is 2720: part 4 (1985). the district falls within the southwest monsoon region, accounting for 21% of the total annual rainfall; northeast monsoon accounts for 46%, winter 6%, and summer 27% of the total annual rainfall. the district depends on the northeast monsoon rains, which are brought by the troughs of low pressure established in the southern part of the bay of bengal between october and december. the maximum amount of water is supplied by periyar river. figure 1. location of the study area (theni, tamil nadu) b. sar data risat-1 is india’s first microwave remote sensing satellite carrying a sar payload operating in the c-band (5.35 ghz). risat-1 was successfully launched by pslv-c19 on april 26, 2012. it started imaging on may 1, 2012. in risat-1, quad pole data is available only for high-resolution spot light (hrs) and fine-resolution stripmap (frs). however, hrs and frs imaging modes have very small swaths, 10 m and 25 m, respectively, which cover a very small area. the c-band frequency of operation and risat-1 sar capability of imaging in hh, vv, hv, vh, and circular polarization has ensured its wide applicability. in this study, risat -1 datasets were acquired with hh, hv polarization in a crs mode with a standard geo-referenced product. the product was purchased from nrsc data centre. the project system used for this image is wgs_1984_utm_zone_43n and datum: wgs_84. the acquisition data of pass was obtained on march 1st, 2016, with an incidence angle of 27.394. majority of the area covers terrain and agricultural lands. data analysis and processing were carried out using envi 4.8 software. land use/cover classes have been prepared by digitizing the image with resourcesat 2 liss iv satellite sensor with a resolution of 5.8 m to identify the different land use and land cover classes for field sample collections. figure 2. field sampling locations 15derivation of soil moisture using modified dubois model with field assisted surface roughness on risat-1 data. c. field measurements a field visit was carried out for this study from march 1 to 4, 2016. a total of 15 field sites were selected for sampling using land use and land cover classes (fig.2). soil samples were collected at 0-15 cm depths using the core cutter sampling method as per civil engineering standards (is 2720 part 29, 1975). the ground control points (gcp) for each of the 15 locations were collected using handheld gps. during the time of field data collection, the humidity of that area was very low. no rainfall was recorded in theni district from march 1 to 5, 2016. d. surface roughness derivation the surface roughness model helps to describe the surface height variations with respect to the ground surface level. surface roughness and soil moisture are two important factors that affect sar backscattering from a bare agricultural field (dobson 1981; ulaby 1979; casb 2000). in order to quantatively compare the profiles and measure the spatial characteristics the spatial wave number (k) and spatial scale (2π) is used for estimating the surface roughness from soil moisture data using backscatter sar (evans 1992; wang 2011; fresnel 2005). the radar return pulse at different times shows that the effect of surface roughness is considerable enough to study the soil moisture retrieval (baghdadi 2002; terrain 2007). thus, surface roughness is one of the most crucial and sensitive parameters in the analysis of hydrological parameters (y.oh 2002). meanwhile, the measurement of surface roughness was carried out in all 15 sampling locations using transecthull method. the roughness profiles taken across the direction of tillage and opposite to the look direction of the satellite pass for each sample training site. the statistical approach of measuring the surface roughness is derived by using the root mean square height (r.m.s -vertical variations). where, – estimated pixel; total pixels; – number of counts e. soil moisture measurement & descriptions gravimetric soil moisture (mg) value was obtained from laboratory analysis using civil engineering standards (is 2720 part 2, 1973), which was converted to volumetric soil moisture (mv) by multiplying the dry density(ɣd). soil is an important factor for vegetation and agriculture as well as for drought assessment. soil texture has been grouped into five types on the basis of analysis of collected samples from 15 sampling sites. the most abundant type of soil texture is sandy clay (69%) and clay (31%) as per is 1498-1970. f. modified dubois model the dubois model (1995) is a semi-empirical model, which is studied for soil moisture estimation in indian context (srinivasa rao 2013). the model was used to extract the dielectric constant from the backscattering coefficient of risat sar data. the parameters such as incidence angle, wavelength and frequency of sar data, and field surface roughness is used. the model consists of two equations (hh, hv) having independent parameter as surface roughness (say root mean square height). the model was applied for hh polarization to derive dielectric constant (ɛ) (1) (2) (3) f. conversion of dielectric constant to sar soil moisture the dielectric constant was derived using eq. (3). soil moisture can be derived from the dielectric constant using topp’s model (khedikar 2014). this model has been effective for soil moisture estimation for a very prolonged period (oh 1992; song 2009). this model does not require any prior knowledge about the soil texture, but needs the dielectric constant. the relative soil moisture (θv) related with the dielectric constant (ɛ) is shown in the following equation (4). the methods are clearly shown in (fig.3). figure 3. methodology chart results and discussion a. landuse/landcover map land use and land cover maps were prepared by digitizing the object using resourcesat 2 liss iv. simple first level of classification was used for the land use and cover maps. the classification is based on the nrsc lulc level 1 classification system. the classification results show plantation (10.40%), cropland (25.47%), agricultural land (43.23%), forest (0.75%), built-up area (2.20%), water bodies (1.94%), and wasteland (15.98%). the agricultural land with respect to the total geographical area is about 28%. figure 4 shows the land use/cover classification. b. surface roughness derivations a linear regression is performed by comparing the field measured roughness of 15 samples having roughness for smooth to very roughness (shaohua zhao, 2016). the ratio hv/hh derived polarization showing the trend analysis with positive effect in the study site. the results thus obtained are well correlated and the coefficient of determination (r2) = 0.93 is shown in (fig. 4 &5). the roughness measurement is relatively depending upon wavelength k = 2π/λ and frequency. in real field the propagation of the electromagnetic wave, the surface appears rougher than in the far field, which can be compared with reflection of the sunset (reflectivity 2005) of was done for the last image as it was assumed that there was no variation in the roughness. 16 p. thanabalan and r. vidhya figure 4. land use statistics of the study area figure 5. surface roughness (satellite-derived vs. observed) c. retrieval of soil moisture the soil moisture was derived from mdm in combination with topp’s model. the study site comprises land use and land cover classifications such as (cropland, plantation, wasteland, forest and buildup). the derived soil moisture extracted from risat-1 sar shows the cropland ranges from (0.4 to 0.67), plantation ranges from (0.69 to 1.87), wasteland ranges from (0.45 to 0.59). the cropland and wasteland shows similar range of db values extracted from risat-1 sar data. the date of pass is acquired on 1st march 2016. this is post-harvested period, therefore some field would have been harvested. this is mixing up with open lands. so, the agricultural area is masked to identify the changes in landuse classes. other than the ancillary parameters such as seasonal rainfall helps to understand the water availability during the crop sown period. however, this also can be taken as indicator pertaining to dryness of the agricultural area, measuring the indicator of the soil moisture condition within the agricultural area. when superimposed with the land use map having agri/open land, decision regarding agricultural management may be derived. this can be achieved by varying with hh, hv cross polarization combination, that can be an indirect method for irrigation management practice ii) crop condition assessment iii) also a direct indicator of drought. the incorrect estimation of backscattering amount and distribution will affect the retrieval of soil moisture. this effect is more severe in the cropland, as the date of pass comprises during rabi-zaid seasons. according to the indian cropping season are classified into two main seasons i) kharif for pre-monsoon period (june – september) and ii) rabi for post-monsoon period (october – february), when rabi sowing area is also over some of the irrigation practice follows during (march – june) are called as zaid seasons. the plantation is the annual crops, which is having very good soil moisture profile. the observed soil moisture derived from risat-1 sar is found to be well correlated with ground measured soil moisture are described in (fig: 6&7). this validation is mainly to compare the soil moisture derived from risat-1 sar data. the performance of the algorithms were evaluated on the basis of a standard evaluation criterion suggested mainly for rmse calculation (nagamani 2008; sikdar 2004) which states that coefficient of determination r2 = 0.63 was obtained from the mdm – topp’s model. the overestimation of predicted soil moisture values is due to the mixed landuse classes particularly the settlement/built-up area gives high back scattering values, which is reflect in comparison with barren field. otherwise the retrieval of soil moisture over agricultural field is highly correlated with other field measured values which can be clearly seen in (fig7) the previous research finding explains that the dielectric constant of the soil-water ratio is related to the microwave brightness, temperature, and emissivity or reflectivity of the soil (jackson 1981; mekonnen 2009; su 2015). the horizontal (hh) and vertical (hv) distributions of soil moisture control the partitioning of the incident radiant energy into latent and sensible heat fluxes via evapotranspiration. the near-surface soil moisture distribution helps to determine the status and trend in agricultural drought monitoring. moreover, it helps to understand the behavior of the top 5 cm of soil characterized by near-surface level soil moisture (engman 2000). the classification accuracy was better identified in the c band dual polarization and incidence angles (ponnurangam 2011). hence risat-1 sar data gives better results on soil moisture studies in the study site. the comparison and validation of soil moisture with ground measured data is clearly positive. figure 6. soil moisture estimation (mdm &topp’s model) 17derivation of soil moisture using modified dubois model with field assisted surface roughness on risat-1 data. figure 7. soil moisture estimation (observed vs predicted)conclusions the irrespective of the field measurements, a number of inversion models have been developed to retrieve the soil moisture data using active and passive microwave remote sensing. sar has proved its high accuracy in soil moisture estimation at a regional scale. even though the backscattering coefficient was determined with respect to the surface features, the semi-empirical model, i.e., mdm and topp’s model were subsequently approached for more simple and realistic efforts to estimate the soil moisture. similarly, the spectral reflectance of different land use and cover types are classified to derive the optical remote sensing parameters, such as land surface temperature (lst) and normalized difference vegetation index (ndvi). the result shows changes in the vegetative biomass at a spatial and temporal scale. the study found that the cross polarization (hv/hh) is sensitive to the surface roughness. thus, the roughness influence can be removed inform the soil moisture estimation (2133(4). the complex permissivity of wet soil and dry soil ratio with different soil textures depends on the variations in dielectric constant and dielectric loss of various types of soil in effective measurements of sensitive hydrological parameters. the derived model (mdm and topp’s model) is efficient in soil moisture estimation with the present availability of risat data in hh, hv with medium-resolution scansar (mrs) and coarse-resolution scansar (crs). hence, the synergistic approach of using both optical and microwave remote sensing helps to strengthen and to assess the validation of the sigma naught (σ°) against the vegetative biomass with low and high incidence angles. the surface roughness and thermal variations are normalized for agricultural drought assessment. furthermore, to assess the spatial and temporal patterns of soil moisture, its physical and hydrological properties need to be studied using suitable techniques of very high-resolution microwave remote sensing. references anguela, t., zribi, m. (2010). analysis of local variation of soil surface parameters with terrasar-x radar data over bare agricultural fields. ieee trans. geosci. remote sens. 48(2), 874–881. baghdadi, n., gaultier, s.., king, c., (2002). retrieving surface roughness and soil moisture from synthetic aperture radar (sar) data using neural networks,” can. j. remote sens. 28(5), 701–711. baghdadi, n., holah, n.., zribi, m. (2006) soil moisture estimation using multi-incidence and multi-polarization asar data. int. j. remote sens. 27(10), 1907–1920. c, a. s. b.., ers-, s., (2000). estimation of surface roughness in a semiarid region from c-band ers-1 synthetic aperture radar data (1). 903–908. dobson, m. c.., ulaby, f. (1981). microwave backscatter dependence on surface roughness, soil moisture, and soil texture: part iii???soil tension,” ieee trans. geosci. remote sens. ge-19(1), 51–6. dubois, p. c., vanzyl, j.., engman, t. (1995). corrections to ‘measuring soil moisture with imaging radars. ieee trans. geosci. remote sens. 33(6), 1340. engman, e. t., (2000). soil moisture, pp (5-10). evans, d. l., farr, t. g.., zyl, j. j. van., (1992). estimates of surface roughness derived, 382–389. gharechelou, s., tateishi, r.., sumantyo, j. (2015). interrelationship analysis of l-band backscattering intensity and soil dielectric constant for soil moisture retrieval using palsar data. adv. remote sens. 4(march), 15–24. shaohua zhao., chengming zhand., yujju xjong., wencheng xjong., daian you., ying li., (2016) soil moisture mapping using two scenes sar imagery without knowing information on surface parameters. journal of the indian society of remote sensing, 44 (4), 651-656. jackson, t. j., schmugge, t. j., nicks, a. d., coleman, g. a.., engman, e. t., (1981). soil moisture updating and microwave remote sensing for hydrological simulation, hydrol. sci. bull. 26(3), 305–319. khabazan, s., motagh, m.., hosseini, m. (2013). evaluation of radar backscattering models iem , oh , and dubois using l and c-bands sar data over different vegetation canopy covers and soil depths. int. arch. photogramm. remote sens. spat. inf. sci. vol. xl-1/w3, 2013 smpr xl(october), 5–8. khedikar, s. y., geethalakshmi, v.., jagadeeswaran, r., (2014). seasonal variability in soil moisture retrieval from risat-1 satellite images by using emperical model, 2977–2985. mekonnen, d. f., (2009). satellite remote sensing for soil moisture estimation: gumara catchment, ethiopia satellite remote sensing for soil moisture estimation : gumara catchment, ethiopia. mirmazloumi, s. m., sahebi, m. r. (2016). assessment of different backscattering models for bare soil surface parameters estimation from sar data in band c , l and p. 261–278. nagamani, p. v., chauhan, p.., dwivedi, r. m., (2008). development of chlorophylla algorithm for ocean colour monitor onboard oceansat-2 satellite, 527–531. oh, y., (2004). quantitative retrieval of soil moisture content and surface roughness from multipolarized radar observations of bare soil surfaces. ieee trans. geosci. remote sens. 42(3), 596–601. oh, y., sarabandi, k.., ulaby, f. t., (1992). an empirical model and an inversion technique for radar scattering from bare soil surfaces. ponnurangam, g.g., rao, y.s., (2011). soil moisture mapping using alos palsar and envisat asar data over india, 606–609. radar, s. a. (2016). 2 . 2 analysis methods for sar data. 2–5. fresnel, j., augustin, p., fresnel, t. et al. (2005). 2. rough surface scattering models, 1–11. sikdar, m.., cumming, i., (2004). a modified empirical model for soil moisture estimation in vegetated areas using sar data, 4–7. singh, d., kathpalia, a. (2007). an efficient modeling with ga approach to acknowledgement the authors gratefully acknowledge prof. s. s. ramakrishnan, director, institute of remote sensing, anna university, chennai, india. the authors want to thank national data center (ndc) and national remote sensing centre (nrsc), hyderabad, for providing the risat satellite data used during the analysis. sincere thanks to dr. v. kumar, professor and head, department of agricultural engineering, agricultural college & research institute, madurai campus, and to all those who helped during the ground truth survey and analysis of satellite data. 18 p. thanabalan and r. vidhya retrieve soil texture, moisture and roughness from ers-2 sar data. prog. electromagn. res. 77, 121–136. song, k., zhou, x.., fan, y., (2009). empirically adopted iem for retrieval of soil moisture from radar backscattering coefficients. ieee trans. geosci. remote sens. 47(6), 1662–1672. srinivasa rao, s., dinesh kumar, s., das, s. n., nagaraju, m. s. s., venugopal, m. v., rajankar, p., et al. (2013). modified dubois model for estimating soil moisture with dual polarized sar data. j. indian soc. remote sens. 41(4), 865–872. capodici, f., maltese, a., ciraola, g., loggia, l., urso, g. d., et al., (2013). coupling two radar backscattering model to assess surface roughness and surface water content at farm scale. j. hydrological sciences. 58(8), 1677–1689. su, z. b., (2015). microwave remote sensing of soil moisture, pp (6-18). terrain, d., house, m., (2007). simulation of radar backscattering coefficients using jem a tool for surface soil moisture retrieval, 383–388. tiwari, r., singh, r. k., chauhan, d. s., singh, o. p., prakash, r.., singh, d. (2014) microwave scattering for soil texture at x-band and its retrieval using genetic algorithm, 120–127. topp, g. c., davis, j. l.., annan, a. p. (1980). electromagnetic determination of soil water content: measurements in coaxial transmission lines. water resour. res. 16(3), 574–582. ulaby, f. t. g. a. b. and mc. d. (1979). microwave backscatter dependence on surface roughness, soil moisture and soil texture: part ii-bare soil. ieee trans. geosci. rem. sens. 16(4), 286–295. wang, s. g., li, x., han, x. j.., jin, r., (2011). estimation of surface soil moisture and roughness from multi-angular asar imagery in the watershed allied telemetry experimental research (water). hydrol. earth syst. sci. 15(5), 1415–1426. y. oh., k. sarabandi., f. t. ulaby., (2002), semi-empirical modelof the ensemble-averaged differential mueller matrix for microwave backscattering from bare soil surfaces. ieee trans. geosci. remote sensing, 40, 1348 –1355. 119 earth sciences research journal earth sci. res. sj. vol. 17, no. 2 (december, 2013): 119 126 nonlinear forecasting of stream flows using a chaotic approach and artificial neural networks hakan tongal engineering faculty, civil engineering department, süleyman demirel university, turkey fluid dynamics this paper evaluates the forecasting performance of two nonlinear models, k-nearest neighbor (knn) and feed-forward neural networks (ffnn), using stream flow data of the kızılırmak river, the longest river in turkey. for the knn model, the required parameters are delay time, number of nearest neighbors and embedding dimension. the optimal delay time was obtained with the mutual information function; the number of nearest neighbors was obtained with the optimization process that minimizes rmse as a function of the neighbor number and the embedding dimension was obtained with the correlation dimension method. the correlation dimension of the kızılırmak river was 2.702d = , which was used in forming the input structure of the ffnn. the nearest integer above the correlation dimension (i.e., 3) provided the minimal number of required variables to characterize the system, and the maximum number of required variables was obtained with the nearest integer above the value 2 1d + (takens, 1981) (i.e., 7). two ffnn models were developed that incorporate 3 and 7 lagged discharge values and the predicted performance compared to that of the knn model. the results showed that the knn model was superior to the ffnn model in stream flow forecasting. however, as a result from the knn model structure, the model failed in the prediction of peak values. additionally, it was found that the correlation dimension (if it existed) could successfully be used in time series where the determination of the input structure is difficult because of high inter-dependency, as in stream flow time series. este trabajo evalúa el desempeño de pronóstico de dos modelos no lineares, de método de clasificación no paramétrico knn y de redes neuronales con alimentación avanzada (fnnn), usando datos de flujo del río kizilirmak, el mayor de turquía. para el modelo knn, los parámetros requeridos son tiempo de retraso, número de vecindarios cercanos y dimensión de encrustamiento. el tiempo óptimo de retraso fue obtenido con la función de información mutua; el número de vecindarios cercanos fue obtenido con la optimización de procesos que minimizan el rmse como una función del número de vecindarios y la dimensión de incrustación fue obtenida con el método de dimensión correlativa. la dimensión de correlación del río kizilirmak fue utilizado en la formación de la estructura de ingreso de las redes ffnn. la integración cercana sobre la dimensión de correlación proveyó el número mínimo de variables requeridas para caracterizar el sistema y el número máximo de variables requeridas fue obtenido con el número entero por encima del valor (takens, 1981). se desarrollaron dos modelos de redes fnnn que incorporan 3 y 7 valores de descargas retrasadas y el desempeño de predicción comparado con el modelo knn. los resultados muestran que el modelo knn fue superior al modelo de redes ffnn en el flujo de pronósticos. sin embargo, como un resultado del modelo de estructura knn, el modelo falla en los valores pico. adicionalmente, se encontró que la dimensión de correlación (de existir) podría ser usada eficientemente en series temporales donde la determinación de estructura de ingreso es difícil por la gran interdependencia, como en las series temporales de flujo. palabras clave: kizilirmak, método de clasificación no paramétrico knn, redes neuronales con alimentación avanzada, funciones de información mutua, dimensión de correlación. record manuscript received: 11/02/2013 accepted for publication: 01/11/2013 abstract resumen key words: kızılırmak, k-nearest neighbor, feedforward neural networks, mutual information function, correlation dimension 1. introduction reliable and accurate stream flow forecasting is essential for water resources management. stream flow simulations and forecasts are also important for optimization of water resources planning and allocation. therefore, in addition to many other reasons, understanding stream flow dynamics constitutes one of the most important problems in hydrology and water resources. for this purpose, many data-driven models have been developed, including linear, nonlinear, parametric and nonparametric models for hydrologic time series prediction in the past decades (marques et 120 al., 2006). generally, in regard to system dynamics, there are two basic assumptions that underlie different modeling techniques, stochastic and chaotic dynamics. regarding the former assumption, the observed hydrologic time series originated from a stochastic process with an infinite number of degrees of freedom. to this assumption, the mean behavior of a time series could be captured with linear models such as autoregressive, autoregressive moving-average (al-awadhi and jolliffe, 1998; toth et al., 2000), autoregressive integrated moving-average (chang et al., 2002; lisi and villi, 2001) and seasonal autoregressive integrated moving-average (modarres, 2007; ooms and franses, 2001; pekárová et al., 2009), from which great results have been obtained. however, to utilize these models, a priori assumptions are required, such as stationarity and gaussian distribution. as most hydrological time series involve non-stationarity and non-gaussian conditions, the data should be transformed before it can be used in stochastic models. however, after this transformation, nonlinearity still exists, and time series are processed with a linearity assumption (chen et al., 2008). recent studies have shown that the complex nonlinear behavior of stream flow dynamics should not necessarily be the outcome of a stochastic process. with the advent of deterministic chaos theory, a few hydrologic modeling studies have suggested irregular behavior could be the outcome of simple deterministic systems influenced by a few nonlinear interdependent variables (khokhlov et al., 2008; sivakumar, 2003; yu et al., 2011). numerous researchers have tried to reveal chaotic behavior in hydrologic time series, such as river flow (ng et al., 2007; sivakumar, 2007; sivakumar et al., 2001), lake level (frison et al., 1999; khokhlov et al., 2008) and rainfall (dhanya and kumar, 2011; sivakumar et al., 2006). because the studies revealed possible chaotic behavior in stream flow dynamics, the requisition for chaotic prediction methods in stream flow was obvious. the most employed prediction method for chaotic hydrological time series is the k-nearest neighbor (k-nn), which was used in this study (elshorbagy et al., 2002; liu et al., 1998; sivakumar, 2003; wu et al., 2009). however, in the hydrology literature, the studies that compare the modeling capability of the k-nn approach with other nonlinear modeling techniques are limited. to obtain more insight into the modeling capability of k-nn, another widely used nonlinear modeling technique the feed-forward neural network (ffnn) (araghinejad et al., 2011; deka et al., 2012; kuo-lin, 2011; vafakhah, 2012; wu and chau, 2010) was employed. additionally, in determining the number of input parameters for ffnn, to the author’s knowledge, the chaotic procedure in this study is the first proposed. the values for the dominant variables obtained from the chaos analysis were used as the minimum and maximum input parameters in the ffnn. generally, the prediction techniques for a dynamic system can generally be divided into two approaches: local and global (wu and chau, 2010). because the local approach uses only nearby states to make predictions, the k-nearest neighbor can be included in this class, and ffnn can be classified as the latter one. therefore, in this study, we preliminarily evaluate which nonlinear approach, local or global, is more efficient in forecasting stream flow time series. the two nonlinear approaches’ modeling capabilities are compared for the kızılırmak river, the longest river in turkey. the paper is organized as follows. section 2 introduces the study area and the stream flow time series that are used. in section 3, the basic principles of the k-nn and the feed-forward neural network as a sub-class of ann are described. following this, in section 4, the obtained results are given with a detailed discussion. the conclusions of the paper are presented in section 5. 2. study area and data the kızılırmak river basin is located between 37°58'-41°44' north latitudes and 32°48'-38°22' east longitudes. the kızılırmak river flows through a 1,355 km long course, the longest in turkey. in the basin, the continental climate is dominant, for which summers are characterized with moderate precipitation and winters are characterized with severe cold. the annual mean precipitation and temperature are 446.1 mm and 13.7°c, respectively. the flow regime of the river is irregular resulting from rainfall and snow-melt. the lowest discharges are observed between july and february, and the river starts to rise in the beginning of march and reaches its highest level in april (çakmak et al., 2007). the data that were used cover the period of january 1960 to september 2004 (figure 1), and the main statistical parameters of the stream flow time series are given in table 1. figure 1. daily discharge data at the 15-1501 (yamula) gauge station of the kızılırmak river 3. methodology 3.1. k-nearest neighbor approach the temporal evolution of a system can be described by a multi-dimensional phase space. the most frequently used reconstruction method for a univariate or multivariate time series is the delay time method which was developed by packard et al. (1980) and takens (1981). the main idea behind phase-space reconstruction is that the system is characterized by selfinteraction, and the observed time series can hold the information about the dynamics of the entire system (sivakumar et al., 2002). the past observations can be embedded in an m-dimensional state space according to: where , m is the embedding dimension of the vector and is the delay time. to characterize a dynamic system with an attractor dimension d, ( )-dimensional space is required (takens, table 1. the statistics of the kızılırmak river discharge data station no/name obs. period drainage area (km2) elevation above sea (m) mean (m3/s) std. dev. (m3/s) median (m3/s) coeff. of variation skewness kurtosis 151501/yamula 19602004 15182 990 66.47 87.66 31.20 1.32 2.94 14.59 hakan tongal (1) 121 1981). however, abarbanel et al. (1990) proposed that m d> would be sufficient. as it is seen in eq. (1), the main parameters that should be determined in the phase-space reconstruction method are the delay and the embedding dimension parameters. the most commonly used methods for determining these parameters are the mutual information (mi) and the correlation dimension methods, respectively. another commonly used method for determining the delay parameter is the autocorrelation function (acf). however, because the acf does not measure the nonlinear dependence, in this study, we employed a mi function that was based on joint probabilities that enabled us to measure the nonlinear dependency beyond linear correlation. the mutual information was computed according to: where and are successive values, and are the individual probabilities of and , respectively and is the joint probability density. to determine the optimal delay time parameter, it is advised to use the local minimum of the mi (frazer and swinney, 1986), i.e., no increase or decrease in the mutual information function’s successive values for a specific lag time. however, in some mi functions, it is rather difficult to determine a local minimum. therefore, to determine the optimal delay time, we recommend using the following formula to amplify the successive differences between the calculated mi function values (tongal et al., 2013): where currentmi is the mutual information value of which relative change will be calculated, nextmi is the successive value of currentmi the first local minimum could be taken where the relative change of mutual information as a function of delay time started to become constant with lag time. with a properly selected time delay, the considered time series can be reconstructed in the m-dimensional phase space by calculating the correlation exponent from the correlation integral ( ( )c r ) as follows: where h is the heaviside step function with ( ) 1h u = for 0u > , and ( ) 0h u = for , , r is the radius of a sphere centered on iy or jy , is the euclidean norm, and n is the number of data points. ( )c r gives the probability of two randomly selected vectors that lie within a certain distance (ng et al., 2007). the dimension d of the state space is related to the correlation integral as: additionally, the logarithm of both sides of eq. (5) gives a linear relationship where the slope equals to the correlation exponent d. if the correlation exponents increase as a function of embedding dimension, then the considered system can be thought of as stochastic, otherwise it is chaotic. in the latter situation, the correlation exponents reach a saturation value, which provides the correlation dimension of the system. the correlation dimension is a parameter that gives the number of dominant processes that are acting in the system dynamics (sivakumar and jayawardena, 2002). with the determination of the optimal delay time and embedding dimension, the state space of the system can be constructed, from which interpreting the dynamics of an m-dimensional map is possible using the following equation: where te is a noise term and and are vectors of dimension m that describe the state of the system at times t (current state) and t t+ (future state), respectively. if the function is known, it is possible to predict the future trajectory of a system. the function can be found with the k-nearest neighbor (k-nn) approach which was proposed by farmer and sidorowich (1987). in this approach, the nearby states are used to obtain future forecasts. one of the most commonly used functions for k-nn is the weighted form: where ( )ia t are the nearest neighbors of the last observed value (i.e., prediction starting point) and are the weights to be adjusted using the information from the n-nearest points. l denotes that these weights are different for each forecasted point. for more details, please refer to wu and chau (2010), ng et al. (2007), laio et al., (2003) and liu et al. (1998). 3.2. feed-forward neural networks artificial neural networks (ann) use a multilayered approach that approximates complex mathematical functions to process data. an ann is a system that is composed of discrete layers with each layer including at least one neuron. with a connection weight, each node of a layer is connected to the node(s) of the preceding layer but not to nodes of the same layer (sahoo et al., 2009). considering the feed forward neural network architecture in figure 2, if the layers and the neurons in the layers increase, the architecture of the ann becomes more complex, which complicates the solution process. therefore, it is important to select an optimal ann architecture. in this study, the numbers of neurons in the layers were determined with a trial and error process that minimized the root mean square error calculated from the differences between the predicted and observed values. figure 2. a typical feed forward neural network architecture in figure 2, the input signal propagates from layer to layer through the network in a forward direction. the weight vectors between the layers ( ijw and jkw ) were randomly generated in the range between -1 and 1. the total output of a jth hidden neuron was computed according to: where iq is the value of the ith input parameter to the hidden layer neurons, jb is the bias for the jth hidden layer neuron and n is the total number of input neurons. the calculated total input signal, js , that received the jth hidden layer neuron was converted to an output signal using an activation function . the output signal of the jth hidden layer neuron was for the n th pattern of the training data set. nonlinear forecasting of stream flows using a chaotic approach and artificial neural networks (2) (3) (4) (5) (6) (7) (8) 122 the output neuron received signals from all hidden-layer neurons and converted them to a single signal as output using an activation function . thus, the input and output signal of the kth output neuron were, respectively: where kb is the bias and n is the total number of neurons in the hidden layer (sahoo et al., 2009). by comparing the estimated output with the desired output, the weights were adjusted with well-known error propagation algorithms including levenberg-marquardt (lm), bayesian regularization (br) and gradient descent with momentum and adaptive learning rate back-propagation algorithm (gdx). in this study, we employed the lm algorithm as a learning algorithm. detailed information about the lm algorithm can be found elsewhere (aqil et al., 2007; daliakopoulos et al., 2005). in this study, the tangent sigmoid transfer function, of which the validity has been proven in hydrological applications, was used: where u is the js and ks in the hidden layer and output layer, respectively. 3.3. performance indices to evaluate the model performances, the following performance indices were used; nash-sutcliffe coefficient of efficiency (ce), mean absolute error (mae), persistence index (pi), root mean square error (rmse), relative volume error (rve) and coefficient of determination (r2). nash and sutcliffe (1970) proposed the coefficient of efficiency in the following form: where io and ip are the observed and predicted values, respectively, and o is the mean value of the observed values. mean absolute error is a measure that evaluates the absolute deviation of the predicted values from the observed ones. it is calculated as: the persistence index (pi) proposed by kitanidis and bras (1980), compares the predictions of a model with the best estimate for the future, which is given by the last observation (randrianasolo et al., 2011). the index has the following form: where is the last observed value at time i minus the lead time l. rmse is one of the most widely used criterions to assess model efficiency, and it evaluates the forecast errors with the following form: because this criterion is sensitive to large forecast errors (i.e., the errors are amplified by squaring), it provides a good measure for the goodness-of-fit at high flows. additionally, it has the same units as the observed values; thus, it enables the interpretation of the magnitude of error. the rve criterion shows the total relative error resulting from the model predictions as: r2 describes the proportion of the total variance in the observed data that can be explained by the model and is formulated as below: r2 changes between 0 (no relation) and 1 (perfect fit), which describes how much of the observed dispersion is explained by the model. 4. results and discussion to determine the chaotic dynamics within the stream flows, the correlation dimension method was applied. to calculate the correlation integrals, the delay time ( ) was computed using the mutual information function and its relative change with lag time (figure 3). figure 3. mutual information function and its relative change with lag hakan tongal (9) (10) (11) (12) (13) (14) (15) (16) (17) 123 in figure 3, it was difficult to select the first local minimum from the original mi function. the amplified differences (i.e., the relative change of the mi) between the successive values were more informative about where the first local minimum could be found. in the mi relative change, the mutual information function was stable and fluctuated around a constant value after 30 days. by determining the delay value, the correlation integrals were computed by the grassberger-procaccia algorithm for different embedding dimensions (m) from 1m = to 25m = (figure 4a). (a) (b) figure 4. (a) correlation integrals as a function of different embedding dimensions (m) (b) correlation exponents obtained from correlation integrals as a function of embedding dimensions from the scaling regions (i.e., nearly linear sections of each integral plots) of the calculated correlation integrals, the correlation exponents were calculated with the least squares estimation method and each calculated correlation exponent was plotted against its embedding dimension as seen in figure 4b. because the correlation exponent values increased with the embedding dimension up to a certain value (i.e. 2.702d = ) and then fluctuated around this value, chaotic behavior was indicated. this value was the correlation dimension calculated for the kızılırmak river and the nearest integer above this value provided the minimum embedding dimension for reconstructing the phase-space or the number of variables (i.e., the number of dominant variables) necessary to model the dynamics of the system (khokhlov et al., 2008; sivakumar and jayawardena, 2002; stehlik, 1999). thus, the results from these analyses showed that the required minimum number of variables to model the system dynamics was 3 (3 2.702> ) and the maximum number of variables to model the system dynamics was 7 ( ). in the k-nn model development, the embedding dimension was taken as 3 by considering the required minimum embedding dimension obtained from the above analysis. the required nearest neighbor numbers for k-nn analysis was determined with a trial and error process that minimized rmse as a function of nearest neighbor number. figure 5 shows that the rmse decreased as a function of nearest neighbor number until nearest neighbor number equaled 33 days and after this value, rmse started to increase. therefore, the optimal nearest neighbor number was selected as 33 days. figure 5. determination of the nearest neighbor number as a function of rmse criterion with these results, the required parameters for the k-nn model were obtained. to determine whether the obtained correlation dimension could be used as the lag value for the discharges, the following model structures were built. to the author’ knowledge, this study was the first to take the correlation dimension as the of required lag value for the discharges. by considering the correlation dimension value as the required number of variables that characterize the system, the following ffnn model structures were constructed that incorporate a minimum of 3 (3 2.702> ) and maximum of 7 ( ) lagged values. typically, the training data set is selected as 70%-80% of a time series and the remaining part is used as the calibration and test period (banerjee et al., 2011; daliakopoulos et al., 2005; riad et al., 2004). in this study, approximately 78% (36 years) of the entire data set was selected as the training period, and the remaining part, approximately 22% (10 years), was selected for the test period. table 2. the model structures that considered minimum and maximum lagged discharges ffnn model structures with minimum and maximum variables variable numbers ffnn-i model 3 ffnn-ii model 7 nonlinear forecasting of stream flows using a chaotic approach and artificial neural networks 124 (a) (b) figure 6. determination of the number of neurons in the hidden layers for the (a) ffnn-i and (b) ffnn-ii models as a function of rmse the ffnn model structures contained one input layer, two hidden layers and one output layer. the number of neurons in the hidden layers was determined using an optimization process that minimized rmse as a function of the number of neurons (figure 6). the results from these models are given in figure 7. from figure 7, the inadequacy of the k-nn model is obvious; failing to capture the peak flows. the reason for this is that the k-nn model predicted the next value by considering the past observed values as much as the number of nearest neighbors. to obtain accurate peak flow prediction, the number of peak values in the past observed period should be as much as the nearest neighbor number. if the number of observations of peak flow in the training period is smaller than the nearest neighbor number, than the model fails in the accurate prediction of peak flows. this is also valid for low flow predictions. from figure 7, there is a section at the end of the test period in which no flow was observed. the k-nn model constantly over-predicted this period. however, the ffnn models performed better in peak flow predictions than the k-nn model. to acquire more insight into the models’ performance, the performance indices were calculated and are given in table 3. (a) (b) (c) figure 7. the observed river flow and predicted river flow for the entire test period for the (a) k-nn (b) ffnn-i and (c) ffnn-ii models by means of the performance indices, the best model was selected as the k-nn model and the worst model was selected as the ffnn-i model. the highest pi, ce and r2 and the lowest rmse, rve and mae values were obtained with the k-nn model. as the results showed, there was not much difference between the ce and r2 values, and stand-alone evaluation of these performance indices does not give much insight into the model comparison. in addition to these criteria, rmse, rve, mae and pi demonstrated the clear superiority (nearly twice as much) of the k-nn model over the ffnn model. therefore, in the model comparison, it was important to take into account other performance indices that emphasized different features of the predicted values. table 3. forecasting performance of the nonlinear models kızılırmak river performance indices rmse rve mae pi ce r2 k-nn model 4.4776 10,634 1.6686 0.8803 0.9965 0.9970 ffnn-i model (3-7-3-1) 13.2685 23,571 4.3124 -0.0511 0.9689 0.9699 ffnn-ii model (7-6-7-1) 9.8451 20,838 3.4546 0.4219 0.9829 0.9829 hakan tongal 125 these results showed that the correlation dimension could be used in the determination of the ffnn model structure by taking the lagged values as the minimum and maximum dominant variable number. to the authors’ knowledge, this study is the first to show that the correlation dimension could be used in determining of number of lagged values of the discharges. this is important where the autocorrelation function remains high for higher lags such as in this case (figure 8). figure 8. autocorrelation function of the kızılırmak river it is difficult to determine the optimal lag values in the model structure where the autocorrelation function values start to become statistically insignificant. for instance, in figure 8, the autocorrelation function becomes statistically insignificant on the 90th day. to determine the optimal lag values that will be considered in the model structure, various models should be considered that incorporate combinations of lag values up to 90. obviously, this is quite time consuming. however, by considering the correlation dimension of the examined system, it is possible to construct two models that incorporate minimum and maximum lagged values that are determined from the correlation dimension. this result is evidence that chaos theory could be used in simplifying the modeling procedure, in which the determination of the input structure is rather difficult. 5. conclusions hydrological systems are complex and dynamic in nature as their current and future states depend on numerous variables (tongal et al., 2013). therefore, it is important to determine the number of dominant variables acting within the system dynamics. in regards to this, the methods from chaos theory provided us a proper framework. in this study, one of the chaotic forecasting methods, the k-nn method, was employed for the kızılırmak river, the longest river in turkey. the necessary parameters for this method are the delay time, the embedding dimension and the nearest neighbor number. the optimal delay time was determined from the mutual information function and the nearest neighbor number was determined from the optimization process that minimized rmse as a function of the nearest neighbor number. in determining the optimal delay time from the mi function, we calculated the relative differences between the successive values of the mi function. the optimal delay time was selected as 30 days. to determine the embedding dimension for the k-nn method, the correlation integrals were calculated for various embedding dimensions, i.e., 1m = to 25m = . when the obtained correlation exponents from the correlation integrals were plotted as a function of the embedding dimension, the correlation exponents reached a value (correlation dimension, 2.702d = ), which gave us the dimension of the system. the dimension of the system shows the number of dominant variables that are acting within the system dynamics. for this system, the minimal number of dominant variables was 3 (3 2.702> ) and the maximum was 7 ( ). to the author’ knowledge, this study is the first to examine whether the obtained correlation dimension could be used in the model development phase. in feed-forward neural network input parameter determination, the two models (ffnn-i and ffnn-ii) that were described incorporated lagged discharge values of 3 and 7, respectively. the number of hidden layer neurons was determined from a trial and error process that minimized rmse. the predictions obtained from the models showed that the k-nn model, which is one of the most commonly used chaotic prediction approaches, was superior to the ffnn models, which are a sub-class of another nonlinear prediction approach, the artificial neural networks. however, the k-nn model failed to predict the peak flows, in which the ffnn demonstrated better performance. therefore, the k-nn model (with averaging method) should not be used, when peak flow forecasting is important. additionally, the results showed that the correlation dimension method can successfully be used instead of the time-consuming trial and error process to determine input parameters for ann, where the interdependency of time series is high. references abarbanel, h.d.i., brown, r. and kadtke, j.b., (1990). prediction in chaotic nonlinear systems: methods for time series with broadband fourier spectra. physical review a, 41: 1782-1807. al-awadhi, s. and jolliffe, i., (1998). time series modelling of surface pressure data. international journal of climatology, 18: 443-455. aqil, m., kita, i., yano, a. and nishiyama, s., (2007). a comparative study of artificial neural networks and neuro-fuzzy in continuous modeling of the daily and hourly behaviour of runoff. journal of hydrology, 337(1–2): 22-34. araghinejad, s., azmi, m. and kholghi, m., (2011). application of artificial neural network ensembles in probabilistic hydrological forecasting. journal of hydrology, 407(1-4): 94-104. banerjee, p., singh, v.s., chatttopadhyay, k., chandra, p.c. and singh, b., (2011). artificial neural network model as a potential alternative for groundwater salinity forecasting. journal of hydrology, 398(3–4): 212-220. chang, f.j., chang, l.-c. and huang, h.-l., (2002). real-time recurrent learning neural network for stream-flow forecasting. hydrological processes, 16(13): 2577-2588. chen, c.s., liu, c.h. and su, h.c., (2008). a nonlinear time series analysis using two-stage genetic algorithms for streamflow forecasting. hydrological processes, 22: 3697-3711. çakmak, b., kendirli, b. and ucar, y., (2007). evaluation of agricultural water use: a case study for kizilirmak. journal of tekirdag agricultural faculty, 4(2): 175-185. daliakopoulos, i.n., coulibaly, p. and tsanis, i.k., (2005). groundwater level forecasting using artificial neural networks. journal of hydrology, 309(1–4): 229-240. deka, p.c., haque, l. and banhatti, a.g., (2012). discrete wavelet-ann approach in time series flow forecasting-a case study of brahmaputra river. international journal of earth sciences and engineering, 5(4): 673-685. dhanya, c.t. and kumar, d.n., (2011). multivariate nonlinear ensemble prediction of daily chaotic rainfall with climate inputs. journal of hydrology, 403: 292-306. elshorbagy, a., simonovic, s.p. and panu, u.s., (2002). estimation of missing streamflow data using principles of chaos theory. journal of hydrology, 255(1–4): 123-133. farmer, d.j. and sidorowich, j.j., (1987). predicting chaotic time series. physical review letters, 59: 845-848. frazer, a.m. and swinney, h.l., (1986). independent coordinates for strange attractors from mutual information. physical review a, 33(2): 1134–1140. frison, t., abarbanel, h., earle, m., schultz, j. and sheerer, w., (1999). chaos and predictability in ocean water level measurements. . journal nonlinear forecasting of stream flows using a chaotic approach and artificial neural networks 126 of geophysical research oceans, 104: 7935-7951. khokhlov, v., glushkov, a., loboda, n., serbov, n. and zhurbenko, k., (2008). signatures of low-dimensional chaos in hourly water level measurements at coastal site of mariupol, ukraine. stochastic environmental research and risk assessment 22: 777-787. kitanidis, p.k. and bras, r.l., (1980). real-time forecasting with a conceptual hydrologic model 2. applications and results,. water resources research, 16: 1034-1044. kuo-lin, h., (2011). hydrologic forecasting using artificial neural networks: a bayesian sequential monte carlo approach. journal of hydroinformatics, 13(1): 25-35. laio, f., porporato, a., revelli, r. and ridolfi, l., (2003). a comparison of nonlinear flood forecasting methods. water resources research, 39(5): 1129. lisi, f. and villi, v., (2001). chaotic forecasting of discharge time series: a case study. journal of the american water resources association, 37(2): 271-279. liu, q., islam, s., rodriguez-iturbe, i. and le, y., (1998). phase-space analysis of daily streamflow: characterization and prediction. advances in water resources, 21(6): 463-475. marques, c.a.f. et al., (2006). singular spectrum analysis and forecasting of hydrological time series. physics and chemistry of the earth, parts a/b/c, 31(18): 1172-1179. modarres, r., (2007). streamflow drought time series forecasting. stochastic environmental research and risk assessment, 21(3): 223-233. nash, j.e. and sutcliffe, j.v., (1970). river flow forecasting through conceptual models, part ia discussion of principles. journal of hydrology, 10: 282-290. ng, w.w., panu, u.s. and lennox, w.c., (2007). chaos based analytical techniques for daily extreme hydrological observations. journal of hydrology, 342: 17-41. ooms, m. and franses, p.h., (2001). a seasonal periodic long memory model for monthly river flows. environmental modelling & software, 16: 559-569. packard, n.h., crutchfield, j.p., farmer, j.d. and r.s., s., (1980). geometry from a time series. physical review letters 45(9): 712–716. pekárová, p., onderka, m., pekár, j., rončák, p. and miklánek, p., (2009). prediction of water quality in the danube river under extreme hydrological and temperature conditions. journal of hydrology and hydromechanics, 57(1): 3-15. randrianasolo, a., ramos, m.h. and andréassian, v., (2011). hydrological ensemble forecasting at ungauged basins: using neighbour catchments for model setup and updating. adv. geosci., 29: 1-11. riad, s., mania, j., bouchaou, l. and najjar, y., (2004). rainfall-runoff model using an artificial neural network approach. mathematical and computer modelling 40: 839–846. sahoo, g.b., schladow, s.g. and reuter, j.e., (2009). forecasting stream water temperature using regression analysis, artificial neural network, and chaotic non-linear dynamic models. journal of hydrology, 378(3–4): 325-342. sivakumar, b., (2003). forecasting monthly streamflow dynamics in the western united states: a nonlinear dynamical approach. environmental modelling & software, 18(8-9): 721-728. sivakumar, b., (2007). nonlinear determinism in river flow: prediction as a possible indicator. earth surface processes and landforms(32): 969-979. sivakumar, b., berndtsson, r. and persson, m., (2001). monthly runoff prediction using phase space reconstruction. hydrological sciences journal, 46(3): 377-388. sivakumar, b. and jayawardena, a.w., (2002). an investigation of the presence of low-dimensional chaotic behaviour in the sediment transport phenomenon. hydrological sciences journal, 47(3): 405-416. sivakumar, b., jayawardena, a.w. and fernando, t.m.k.g., (2002). river flow forecasting: use of phase-space reconstruction and artificial neural networks approaches. journal of hydrology, 265(1-4): 225-245. sivakumar, b. et al., (2006). nonlinear analysis of rainfall dynamics in california’s sacramento valley. hydrological processes, 20: 17231736. stehlik, j., (1999). deterministic chaos in runoff series. journal of hydrology and hydromechanics, 47(4): 271-287. takens, f., (1981). detecting strange attractors in turbulence. in: d.a. rand, jung, l.s. (editor), dynamical systems and turbulence, lecture notes in mathematics. springer-verlag, berlin, pp. 366-381. tongal, h., demirel, m.c. and booij, m.j., (2013). seasonality of low flows and dominant processes in the rhine river. stochastic environmental research and risk assessment, 27(2): 489-503. toth, e., brath, a. and montanari, a., (2000). comparison of short-term rainfall prediction models for real-time flood forecasting. journal of hydrology, 239(1–4): 132-147. vafakhah, m., (2012). application of artificial neural networks and adaptive neuro-fuzzy inference system models to short-term streamflow forecasting. canadian journal of civil engineering, 39(4): 402-414. wu, c.l. and chau, k.w., (2010). data-driven models for monthly streamflow time series prediction. engineering applications of artificial intelligence, 23: 1350-1367. wu, c.l., chau, k.w. and li, y.s., (2009). predicting monthly streamflow using data-driven models coupled with data-preprocessing techniques. water resour. res., 45(8): w08432. yu, b., huang, c., liu, z., wang, h. and wang, l., (2011). a chaotic analysis on air pollution index change over past 10 years in lanzhou, northwest china. stochastic environmental research and risk assessment, 25: 643-653. hakan tongal the yangtze river crossing project, which is part of the chinese gas pipeline trunk network and aims to transmit the natural gas to the yangtze river delta area, sets up a record for the length of horizontal directional drilling(hdd) in the world. the term of pipeline installed is 3300m, and the diameter is 40in. many technical difficulties have been inevitably encountered in a large diameter hole reaming in fine sand, the calculation of section scour depth which determines the minimum buried depth of the pipe, protection of outside coating and the method of long distance pilot hole drilling, such other challenges. in view of this,it was recommended that two pipelines of 711mm should be installed to replace the existing one of 1016mm in diameter and 1% flooding scour depth obtained by reliable statistical analysis. in period of installation, application glass fiber reinforced epoxy resin as a protective coating, executing the pilot hole by intersect method.as a result, this super project is finished successfully on may 21.2013. the experience can be applied to any pipeline engineering the installed by horizontal directional drilling or similar works. earth sciences research journal earth sci. res. j. vol.20, no. 1 (march, 2016): p1 p5 abstract keywords: engineering innovation, long distance, large diameter, gas pipeline, horizontal directional drilling (hdd), starting velocity. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.54504 li guohui1, ma xiaocheng2, yang chunling3 1. doctoral students/vice chief engineer, southwest jiaotong university/china petroleum pipeline engineering corp. email cppe_ligh@cnpc.com.cn 2.manager of crossing department, china petroleum pipeline engineering corp. email: maxiaocheng@cnpc.com.cn 3. engineer, china petroleum pipeline engineering corp. email: yang-chl@cnpc.com.cn. engineering innovation of a length of nearly 3300m large diameter pipeline installed by hdd record manuscript received: 02/12/2015 accepted for publication: 22/03/2016 how to cite item guohui, l., xiaocheng, ma, & chunling, y. (2016). engineering innovation of a length of nearly 3300mlarge diameter pipeline installed by hdd. earth sciences research journal, 20(1), p1-p5. doi: http://dx.doi.org/10.15446/esrj.v20n1.54504 introduction to improve the energy instruction in the yangtze river delta area by using clean fuels instead of coal, china national petroleum corporation(cnpc)executes jiangdu-rudong gas pipeline project. this pipeline aims to transmit the natural gas of jiangsu lng terminal to the delta area where is home to one-third of china’s population and generates an estimated 20% of the country’s annual gdp (carlin,etc.2014). yangtze river crossing is an essential part of the whole project,with the total length of 3300m,design pressure of 10.0mpa,the diameter of 1016 mm,pipe material of api 5l x70 and external coating of 3lpe. the crossing point is located at the downstream of yangtze river, and the topography is flat, which the feature land unit is over bank alluvial plain and new delta alluvial plain. the width of this location is about 3000m with active water transportation,and the average of water depth ranges between 20m to 30m..the stratumunder quaternary overburden is devonian, carboniferous, permian, triassic, jurassic and cretaceous along with the time sequence. 33 bores were drilled on both sides of crossing axis to make clear the geological condition. relying on properties like geological time, source, lithology, physical and mechanical properties, the field stratum was divided into four parts from top to the bottom. main properties are as follows: layer 1 is muddy, silty clay which contains a single thin layer of silt with a thickness of 2-5 mm. the silty clay was interbedded with silt locally. layer 2 is silty clay and the soil is uniform,containing little calcareous concretions with a diameter of 5-50mm, partially. this layer was exposed by all the borehole and it was found stable at the crossing site. layer 3is silty (dense). this layer was exposed by all the borehole,and it is evidenced stableat the crossing site. the buried depth on the roof is 10.70-43.70m and 28.40m on average. layer4 is medium sand, containinglittle like-round gravel particle and pebble. the overall diameter is 2-10mm and 20-40mm, individually.this layer was exposed to most of the borehole and distributed widely at the crossing site. the geotechnical characterization profile (gcp) is shown in figure 1. e n g in e e r in g g e o l o g y p2 li guohui, ma xiaocheng and yang chunling al t i t ude( m) 51. 00 39. 00 27. 00 15. 00 3. 00 61. 00 sout h bank 51. 00 39. 00 27. 00 61. 00 15. 00 3. 00 1 2 3 4 ri ver bed figure 1. geotechnical characterization profile cppe (china petroleum pipeline engineering corp.), which is a branch company of cnpc, served as a design contractor. engineers selected the scheme of installing by hdd rather than shield tunnel. however, it is a significant challenge: the crossing length is nearly 3300m, and the pipeline diameter is 40 inches. the critical problems of this project are 1) large diameter hole reaming in the fine sand; 2) the calculation of section scour depth, which determines the minimum buried depth of the pipe; 3) protection of outside coating, and 4) the method of long distance pilot hole drilling. 1. two small diameter pipeline replace a large one avoiding to reach large diameter hole yangtze river is the longest river in china.there are rare hdd crossings in yangtze river before this project.for example,lanzhouzhengzhou-changshan production oil pipeline project uses hdd crossing in yangtze river, with a pipe diameter of 610mm, and a length of 2100m (li guohui,etc.2012).the crossing construction confronted with many difficulties like bore collapse,drillpipefracture,and finally it was finished after two difficult years..other projects that adopt big diameter and long distance hdd crossing are also rare. in 2007, hangzhou-yongning pipelineprojectcrossed qiantang river,with a diameter of 813mm,crossing a distance of 2454m(tan mingxing,etc.2008).in2008, zhuhai-zhongshan gas pipeline project crossed the modaomen river,with a diameter of 660mmand a length of 2630m(yin gangqian,etc.2008).in the recent two years,pipe diameter enlarges continually.in 2011,china-myanmar gas pipeline crossed at irrawaddy river,using thehdd crossing method,with a pipe diameter of 1016mm, and a crossing length of 1760m. a final example is the second west-east natural gas pipeline crossingpengjiawan sluice,with a pipe diameter of 1219mm, and a length of 1397m. in others countries, the record-setting bore came during the drilling of a second 10,000-foot bore and pull-in of a 30-inch pipe near al jabayl for the bern causeway pipeline project. a new global horizontal directional drill (hdd) undersea crossing record has been set regarding pipe diameter, pipe weight and length of crossing, in saudi arabia, in 2007 (share, 2009). southern gas networks (sgn) is currently reinforcing the gassupply infrastructure on the isle of wight, located off the south coast of england. part of this process required the replacement of the pipelines supplying gas from the uk mainland to the island. two 324mm-diameter steel pipelines were installed through separate holes some 50m below the bed of the solent using horizontal directional drilling (hdd). each 3,930m-long crossing is believed to be some 20% longer than any previously completed using hdd(seamans,2011). analyzing the cases mentioned regarding pipe diameter, pipe weight and length of the crossing, the case of utilizing a 1016mm diameter,with 3300 lengths and a weight reaching 2096tons pipe cannot complete pullback.take into consideration adopting two pipes of the same diameter as a replacement.comparing different diameters ranging from 610mm to 914mm,the sectional area sum for two pipe are as followsin table 1. table 1. different diameters pipe sectional area. concluding from the table 1,there is little difference between two pipes with a diameter of 711mm and one pipe with a diameterof1016mmconcerning the medium through capacity.as a consequence, adopting hdd crossing will reduce the difficulty in comparison with the use of a previous scheme. meanwhile,set two pigging station at both sides, named jingjiang station and jiangyin station.it is required that the outlet temperature is no less than 20°cand outlet pressure is no less than 8.8 mpa in jiangyin station to satisfy the transmission to downstream.we should analyze processing influence for the new scheme, which includes two stations and two 711mm diameter pipelines.the analysis of the fundamental dataconsists oftwo pipes with a diameter of 711mm and pipe thickness of 19mm, the inlet pressure of 9.48mpa, the rate of flow 139.85×104nm3/h, 3pe reinforced protecting coating thickness of 3.8mm,no thermal insulation,no inner coating, earth temperature of 25°c. the total length of the pipe is 4600m. the two pipes processing result is shown in table 2. table 2. two pipes processing result through technology analysis,the pressure of transmission pigging station is 9.18 mpa, and the temperature is 21°c, having no influence on the feasibility ofdownstream and pipe.hence,this scheme does not affect the whole design. 2. 1% flooding scour depth obtained by reliable statistical analysis the yangtze river scour primarily controls the depth of penetration because of the flood peak. according to code gb50423(china petroleum pipeline engineering co., ltd., 2007), the minimum depth of penetration should not be less than 6m from the design flood scour line.the project is defined on a large scale, and we should choose 1% (100-year return period) flood frequency. the chosenrule is shown in table 3. the design flood frequency of channel erosion calculation is 1%,and the peak flood is 108800 m3/s. table 3. engineering scale for hdd project through water area p3engineering innovation of a length of nearly 3300mlarge diameter pipeline installed by hdd since 1960, the crossing region varies relatively small.one main reason is that this project section is influenced by the dual role of runoff and tidal current, so the riverbed maintains a certain amplitude balance of erosion and deposition.the frontier waters riverbed of proposed construction suffers interactive effects of upstream inflow and reciprocating tidal current. the shoreline varies little and keeps relatively stable for years. measures have to be taken to protect the shoreline of proposed project so that the scheme has strong anti-erosion features.on condition that yangtze river regime does not change greatly,the riverbed condition can keep stable over the long run. the change of riverbed is shown in figure 2. 20 10 0 10 20 30 al t i t ude ( m) 0 500 1000 1500 2000 2500 3000 f r om nor t h bank t o s out h( m) 1966. 04 1976. 05 1986. 12 1993. 12 1998. 11 2001. 09 2003. 01 2006. 05 2007. 08 2009. 01 r i ver bed figure 2. change of riverbed from 1960 to 2009 observing the 50 years scour condition,the river regime of pipe crossing region is stable;a deep groove(-15m) in the north bank varies a lot over the years, and the south bank keeps slightly stagnant.also,the axis of the deep groove in the south bank and the deep groove itself(-20m)remains stable. the envelope line for depth scour of 50 years is shown in figure 3. figure3. envelope line for depth scour of 50 years the commonly used methods for river scour depth calculations are samov formula,xie jianheng formula,zhang ruijin formula and sha yuqing formula(zhang fengliang,etc.2013).consider samov formula and xie jianheng formula are only adapted to coarse granular sediment which satisfied the condition d>0.15~0.20mm, and they are not suited to fine sand and silty clay particles.as a consequence,scour calculation in this article uses zhang ruijin formula and sha yuqing formula which are adapted to fine cohesive sediment to figure out the starting velocity. zhang ruijin starting velocity formula professor zhang ruijin from water conservancy and electric power institute wuhan university started from rolling mode of sediment and gave the equilibrium equations of force,then derived the zhang ruijin starting velocity formula. where the units of depth h,diameter d are both m.zhang ruijin starting velocity formula is not only adapted to fine cohesive sediment but also the cohesionless coarse sediment.it is a general formula. sha yuqing starting velocity formula where ε means porosity,stable data is 0.4; the units of d and h are mm and m,respectively;ucmeans starting velocity, and the unit is m/s. the maximum scour depth always arises in the main channel.when the main channel velocity is larger than sand starting velocity,the riverbed stops scouring until the main channel velocity is smaller than starting velocity, and then the scouring process stops. the maximum rate of flow per unit width multiples the maximum vertical average velocity can be derived according to the result of 2-d tidal flow mathematical model. then by the riverbed property,use the formula below to calculate the maximum depth after scouring. if the riverbed consists of cohesive soil,the highest depth after scouring is: where uc’ means the non-scouring velocity of cohesive soil in corresponding depth. if the riverbed consists cohesionless soil,use the formula below to calculate the maximum depth: where vc is the sediment starting velocity,which can be calculated by corresponding velocity formula. when the diameter satisfies the condition d<0.15mm,the sediment particle possesses certain viscosity,the starting velocity increases according to the d decreasing,the fine particle size should be replaced by the equivalent diameter in the same starting velocity or utilizing cohesive sediment starting velocity formula to calculate. the calculated result of scour depth by sha ruiqing formula and zhang ruijin formula is shown intable 4.the scour curve by calculation is shown in figure4. p4 li guohui, ma xiaocheng and yang chunling table 4.calculation of scour depth (m) figure 4. scour curve by calculation superimpose previous monitoring envelop curve for scour line and scour line by calculation, obtaining scour line for engineering. the scour curve for engineering is shown in figure 5. figure 5.scour curve for engineering figure 5 implies that we should lay the pipeline in layer 3 fine sand. 3. application a new type protective coating 3lpe is the most commonly used as outside coating for long distance pipelines(j.d.hair,etc.2000). strengthen 3lpe are coated on hdd pipes and dedicated filed joint coating system are used for 3lpe which consists of heat shrinkable sleeve (hss) and another protective sleeve installed in the front end of the drilling direction, with 100mm overlap on hss and 3lpe, respectively. an uneven seam, or step, would be inevitably formed between hss and 3lpe after the completion of field joint coating system installation, resulting hss dragged off from the pipe during hdd pulling, even extensive damage of 3lpe coating in case of harsh geological conditions. this condition would consequently affect the integrity performance of coating due to lots of hole or damages. in a worst-case scenario, the hss might slip off from the pipe causing sticking accidents and hence making the crossing fail. cppe’ engineers applied a modified liquid solvent-free, glass fiber reinforced epoxy resin (gfrer) coating to protect the anticorrosion coating on hdd pipes. as showed in figure 6, gfrer is coated outside completely encasing the 3lpe layer and its hss field joint coating system, obtaining 3lpe/hss/gfrer system with a smooth surface for the all hdd pipeline to favor pipe pullback. figure 6. 3lpe/hss/gfrer system gfrer protective coating was first applied to the hdd project. the coating thickness is 3.7mm. using a glass fiber reinforced epoxy resin as an armoring coating over the corrosion coating preserve the integrity of the coating and minimize damage when installation. figure 7 shows the glass fiber reinforced epoxy resin painting. figure 7. photos of gfrer protective coating on completion 4. hdd design entry and exit point. it is same geotechnical for drilling effectiveness at south and north bank. so we should select the availability of workspace for pull section fabrication at the exit point. the north bank side is flatness and has enough length for pipeline welding before pulling. it is properly to clarify that the north side is exit point, and the south is the entry. input and output angle. hdd input and output angle should be determined based on crossing length, burial pipeline depth, diameter, elastic laying conditions, terrain conditions. entry angle is limited by equipment capabilities and should be 8° ~ 20°; the exit angle should be 4° ~ 12°.choose lower exit angle when pipeline diameter is large. according to the condition of surface and subsurface, in this case, the selected entry angle is 10°, and the exit is 8°. design radius of curvature. the curvature radius of hdd elastic laying section should not be less than 1500 times the diameter of the pipe to be installed, and must not be less than the 1200d. in this case,the radius is 1067m,which is 1500d. the hdd crossing horizontal distance between yangtze river buried and unearthed point is 3290m and actual length is 3300m. the hdd curve is shown in figure 8. p5engineering innovation of a length of nearly 3300mlarge diameter pipeline installed by hdd figure 8. hdd curve 5.executing the pilot hole by intersect method magnetic signal submarine cable laying.in this project, prepare to lay armored cable used to provide artificial magnetic field strength in crossing area. rent dedicated laying barge to bury the submarine cable while laying.employ the dedicated submarine measurement company to measure the cable coordinate. parameters of measurement direction control.calibrate guide hole crossing paratrack ii direction control system and mark direction control parameter.to ensure the accuracy of data,measure at different locations across the center line to compare and make a record. intersect area.combine paratrack ii direction control system and magnetic signal cable,main rig drills along with the design crossing curve and the auxiliary rig drills at the same time.two sets of rigs coordinate each other while drilling,reaching the intersect area simultaneously. in the intersect area,start axial magnet which is installed in the assistant rig drill bit,leading drill bit of main rig drilling until two drill bits combined, that is drill bit of main rig drills into the guide hole of the auxiliary rig. auxiliary rig pulls back the drill pipe, the main rig pushes on the drill pipe and ultimately unearths along with guide hole dug by the auxiliary. mud sensor, which is installed behind the main rig drill bit, will supervise the mud pressure in holes during the entire processing in pilot hole construction procedure.if the mud pressure increases unusually,pull back the drill bit and go on drilling after hole flushing. pilot holeworks last 20 days only, from march 22, 2013, to april12, 2013. 6. result once the pilot hole was successfully completed, the following reaming operations were executed using new φ168mm s-135 drill pipe: stage one: φ610mm (24”) barrel reamer pre-reaming; stage two: φ762mm (30”) barrel reamer pre-reaming; stage three: φ924mm (36”) barrel reamer pre-reaming; stage four: φ1,118mm (44”) barrel reamer pre-reaming followed by washing of the hole. the pullback was the last step in the crossing process. the pullback was executed by φ168mm drill pipe, a φ1,067mm (42”) barrel reamer, and 3,300m of steel pipeline. all drill parameters and fluid parameters were closely monitored to ensure the smooth pullback and avoid damage to the outside coating (fig. 9). figure 9. pullback operations the original project included 274 working days. the actual work was completed within this time frame. overall, the drilling operations were successful at steering along the design profile. 7. conclusion the following conclusions were reached through the hdd project: 1. it is reasonable that two 711-mm-diameter pipelines replace one 1016-mm-diameter pipeline installation by hdd. this scheme is not only guaranteeing success installation but also reduce thedifficulty. no any contractor can finish the construction if we use one 1016-mm-diameter pipeline, and we should select the shield tunnel which is more expensive than hdd. 2. it is most important that confirm the depth of penetrationaccording to river scour depth analysis. in this project, we analyze the scour depthorganizing the 50 years river scour condition and calculation by zhang ruijin formula and sha yuqing formula. 3. the success of the crossing of yangtze river has established a new record of diameter * length in oil and gas pipeline installation by hdd. references carlin, m., ariaratnam, s., ma, b.s., and zeng, c., 2014. “record setting yangtze river hdd crossing of three parallel pipelines over a length of nearly 3300m,” proc. 2014 nastt no-dig conference, north american soc. for trenchless technology, ny. gb 50423-2007: code for design of oil and gas transportation pipeline. (2008). china plans to press, beijing, china. j.d. hair & associates, inc., 2000. “coating requirements for pipelines installed by horizontal directional drilling and slip boring,” pipeline research council international, inc., houston, united states. li guohui, wei.yang, feng.qian and xc. ma,2012. “high-risk, 610mm diameter, 2,090m hdd projecta case study”, icptt 2012, american society of civil engineers (asce), reston, united states seamans, j., 2011. “hdd in record solent crossing,” world tunnelling, 2011 (jun.):46-49. share, j., 2009. “challenging hdd project provides 10,000-foot recordsetting crossing,” pipeline and gas journal, 2009, 236(5):40-42. tan,mingxing.,liu,yufeng,yang fan,.zeng qiang, lei jie, xie qishuai, 2008. “design of hdd engineering in qiantang rover”, natural gas and oil, natural gas and oil press, chengdu, china yin gangqian.,and tang x.f.,2008. “the construction technology of horizontal directional drilling crossing in modaomen river,” petroleum engineering construction vol.34 (6), petroleum engineering construction press, tianjin, china. zhang fengliang and wang xintao,2013. “discussion on the calculation of the scour depth in the flood impact assessment”, china flood & drought management,2013(5), china flood & drought management press, beijing, china. earth sciences research journal earth sci. res. sj. vol. 17, no. 1 (june, 2013): 33 40 drastic-based methodology for assessing groundwater vulnerability in the gümüşhacıköy and merzifon basin (amasya, turkey) arzu firat ersoy* and fatma gültekin* *karadeniz technical university, engineering faculty, geological engineering department, 61080, trabzon, turkey corresponding author: arzu firat ersoy e-mail:arzufirat@gmail.com phone: + 90 462 377 20 63 fax: + 90 462 325 74 05 hydrogeology preparing aquifer vulnerability maps has become crucial during recent years for preventing adding new ones to aquifers which have been contaminated due to environmental effects and been out of use. gis techniques and drastic method were used when preparing vulnerability maps for the basin in which the gümüşhacıköy and merzifon aquifers are located. groundwater flow is approximately directed west-east and many villages are located across the aquifer in the basin which contains two sub-provinces and is characterised by intensive agricultural activity. drastic layers were created when preparing vulnerability map, using parameters such as groundwater level, recharge, aquifer environment, topography and hydraulic conductivity. the aquifer vulnerability map was prepared by overlapping the layers by means of gis. , three different vulnerability zones were determined in the gümüşhacıköy basin according to drastic scores low (<100), medium (100-140) and high (>140). based on the vulnerability map, it was found that the gümüşhacıköy basin had a low contamination potential. it was established that 16% of the basin had high vulnerability and 47% low vulnerability. areas having high vulnerability generally overlapped areas where the slope was gentle soil above the aquifer was permeable. la preparación de mapas de vulnerabilidad acuífera se ha convertido en una actividad crucial en los últimos años para prevenir la contaminación por efectos ambientales de un afluente y su posterior inutilización. técnicas gis y el método drastic fueron utilizados en la preparación de mapas de vulnerabilidad en la cuenca donde están localizados los acuíferos gümüşhacıköy y merzifon, en turquía. el flujo de las aguas subterráneas corre aproximadamente oeste-este y varias poblaciones están ubicadas al paso del acuífero por dos subprovincias que se caracterizan por la actividad agrícola. se crearon capas en el método drastic cuando se preparó el mapa de vulnerabilidad con parámetros como nivel, recarga, ambiente del acuífero, topografía y conductividad hidráulica. la representación de vulnerabilidad se logró al sobreponer estas capas a través de técnicas gis, lo que permitió determinar tres zonas diferentes de vulnerabilidad en la cuenca de gümüşhacıköy basado en los puntajes del método drastic: baja (<100), medio (100-140) y alta (>140). con base en este mapa, se concluye que la cuenca de gümüşhacıköy tiene un bajo potencial de contaminación. se estableció que el 16 % de la cuenca es altamente vulnerable y el 47 % de baja vulnerabilidad. en aquellos lugares identificados con alto potencial de contaminación se suelen sobreponer áreas donde la inclinación de tierra sobre el acuífero es permeable. key words: vulnerability mapping, drastic, geographic information system (gis), turkey palabras claves: mapa de vulnerabilidad, drastic, sistema de información geográfica (gis), turquía. record manuscript received: 23/10/2012 accepted for publication: 04/04/2013 abstract resumen introduction groundwater has been considered as an important source of water supply due to its relatively low susceptibility to pollution compare to surface water. groundwater quality is usually subject to contamination especially in agriculture-dominated areas having intensive activity involving the use of fertilisers and pesticides. vulnerability assessment has been recognised for its ability to delineate areas which are more easily contaminated than others as a result of anthropogenic activity on/or near the earth’s surface. vulnerability studies can thus provide valuable information for stakeholders working on preventing further deterioration of the environment (mendoza and barmen 2006). figure 1. location map of the gümüşhacıköy-merzifon basin (gmb) arzu firat ersoy and fatma gültekin the concept of groundwater vulnerability to contamination was introduced in the 1960s in france by margat (1968). several approaches for developing aquifer vulnerability assessment maps were adopted such as drastic (aller et al. 1987), god (foster 1987), avi (van stempvoort et al 1993), and sintacs (civita 1994). conventional methods (i.e. drastic, avi, god, sintacs) can distinguish degrees of vulnerability on a regional scale involving different lithology (vias et al 2005). drastic is a familiar method developed for the us environmental protection agency (epa) by aller et al (1987) it has been used in several regions (merchant 1994; melloul and collin 1998; cameron and peloso 2001; al-adamat et al 2003; baalousha 2006; jamrah et al 2007; sener et al 2009; massone et al 2010). the area studied in this research is located in amasya (mid black sea region), namely the gümüşhacıköy-merzifon basin (gmb) (figure 1). groundwater is the major source of irrigation in the amasya district in turkey. surface water has been the main source of water supply for irrigation during the last few decades (fırat ersoy and gültekin 2008). however, water demand has increased and groundwater is now used as a secondary source. annual groundwater exploitation yield was only 3.5x106m3 during the 1970s, and rose to 5.5x106 m3 in 2005 (fırat ersoy, 2007). due to the excessive exploitation of groundwater, water levels have significantly decreased. groundwater quality has also been affected by over exploitation. the town of, gümüşhacıköy is located in this basin. some well water’s nitrate concentration has reached 15.6 mg/l in the gümüşhacıköy basin. nitrite and ammonium concentration have reached 0.03 and 0.3 mg/l respectively, around the town (fırat ersoy et al, 2006). this paper was aimed at assessing groundwater vulnerability to pollution in a shallow aquifer using the drastic model and geographical information system (gis) techniques combined with hydro-geological data layers, i.e. depth of water, net recharge, aquifer media, soil media, topography, impact of vadose zone and hydraulic conductivity. a vulnerability map, showing high, medium and low vulnerability areas was produced for the mentioned basin. study area the gmb covers a 1,060 km2 area elevation ranging from 5501,873 m (figure 1). average annual rainfall is 458 mm average annual temperature is 13.6°c (url-1) and the average annual potential evaporation is 680 mm (fırat ersoy, 2007). the most important body of surface water flowing through the basin is the gümüşsuyu river, which discharges 8.5x106m3/ year (fırat ersoy, 2007). groundwater in the basin draws from both alluvium aquifers, one being confined, (the gümüşhacıköy aquifer) and the other unconfined (merzifon). agriculture is widespread in the basin, and fertiliser and pesticide application have caused groundwater contamination through leaching. 193 wells had been drilled in the gmb up to 2006, 173 wells were aimed at irrigation, and 20 well for domestic purposes. geology the paleozoic metamorphic rocks in the study area, represent the oldest formation. these rocks consist of clayey schist, chlorite schist and green schist, marble and re-crystallised limestone. upper jurassic-lower cretaceous limestone, in the area has fossils which are generally compact, thick-bedded, very hard and fissured lower cretaceous limestone is pink, very hard, thick bedded and micritic, overlying jurassic limestone. cretaceous limestone outcrops on the plain. the flysch series having mixed volcanic material composed of conglomerate, green and black sandstone, shale, marl, limestone, andesite, tuff and tuffite are deposited in the upper cretaceous limestone. cenozoic beds started with the middle eocene age flysch series in the study area. flysch consists of sandstone, shale, sandy limestone, marl, local conglomerate, tuff and agglomerate. the miocene series consist of thick blue claystone and marl and the pliocene beds of micro conglomerate, sandstone, sand, clay, gravel and a mixture of these drastic-based methodology for assessing groundwater vulnerability in the gümüşhacıköy and merzifon basin (amasya, turkey) layer thickness ranges from 10 to 50 cm in these series. the very loose layers are not continuous and change their lithology over short distances. this unit gradually become harder as one goes deeper and turns out to be conglomerate. some blocks of the gravels are 50 cm thick and about 5-10 cm-in diameter 95% of such gravel and blocks are usually rounded and are composed of volcanic material. the quaternary is characterised by alluvium and an alluvium cone consisting of detrical material that comes from north and south with the flood waters. alluvium and cone (10-60 m thick) take the form of gravel, sand, clay and a mixture of these, along the gümüşsuyu, köseler and salhan rivers. hydrogeology the gmb’s hydrogeological setting has been outlined by fırat ersoy (2007). the most impotant geological units for groundwater transport in the basin are quaternary alluvium and pliocene clay, sand, gravel and a mixture of these. unconsolidated quaternary and pliocene sediments are around 350 m thick. the other units underlying the alluvium do not bear significant amounts of groundwater. the gmb can be divided into discrete hydrogeological units, including permeable (alluvium), semi-permeable (weak cemented pebble and sandstone, silty clay and volcanic rocks) and impermeable (massive marble and limestone, silty clay and schist). alluvial materials and the pliocene units consisting loosely cemented pebbles, sand and clayey-silt materials outcrop in most parts of the basin. known as the gümüşhacıköy aquifer, this part is crucial to groundwater storage and transfer since it is characterised by high conductivity and storage capacity. deposits have a heterogeneous structure, being formed as alluvial cones at the end of tributary rivers. the alluvial cone formed by the alluvial unit of the paşa and yakacık river is called the merzifon aquifer. well logs show that the cone’s middle sections consist of clayey levels between pebble and sand layers and that level becomes thin along the eastern edge, dominated by clay and silt. since the section between the east of the gümüşhacıköy aquifer and the south of the merzifon aquifer consists of miocene clay and marl, it is not important in terms of groundwater. the late eocene volcanic rocks outcropping across the north, northwest and northeast of the basin form a catchment area with their fractured and fissured structure. natural discharge in the basin is provided by a stream flowing through it from west to east. the gümüşhacıköy aquifer naturally discharges into the gümüşsuyu river, located in the east of basin. the basin contains numerous springs discharging from geological units, faults and fractures. some are exploited as potable for drinking water and others are used for irrigation, consequently recharging the groundwater. the springs’ total flow rate is 720 l/sec. three ponds in the basin are used for irrigation. infiltration into the groundwater from these ponds has been estimated as being 41.5 m3/year on average (fırat ersoy, 2007). the basin contains 193 wells, 167 in the gümüşhacıköy aquifer and 26 in the merzifon aquifer (dsi̇, 2006). well depth varies from 39 to 290 m and pump flow-rates from 5 to 60 l/sec (dsi̇, 2006). as most of these wells are used for irrigation, pumps operate from may to october. the increased number of wells drilled in the aquifer during recent years and accordingly, the increased amount of water pumped from the aquifer has resulted in a decrease in groundwater level by 15 to 20 m (dsi, 2006). several pumping tests have been performed at existing wells in the gümüşhacıköy and merzifon aquifers. data interpretation has indicated 89.7 1727 m2/day transmissivity, 0.7619.17 m/day hydraulic conductivity and 1.5x10-57.9x10-3 storage coefficient for the basin (fırat ersoy, 2007). materials and methods drastic, proposed by the us environmental protection agency (aller et al., 1987) and its modification termed sintacs (civita, 1994), parameter range rating description relative weighting depth to water (d) (feet) 0-5 5-15 15-30 30-50 50-75 75-100 >100 10 9 7 5 3 2 1 refers to the depth to the water surface in an unconfined aquifer. deeper water table levels imply lesser chance for contamination to occur. depth to water is used to delineate the depth to the top of a confined aquifer. 5 net recharge (r) (in) 0-2 2-4 4-7 7-10 >10 1 3 6 8 9 indicates the amount of water per unit area of land which penetrates the ground surface and reaches the water table. recharge water is available to transport a contaminant vertically to the water table, horizontal with in an aquifer. 4 aquifer media (a) massive shale metamorphic/igneous weathered met./igneous bedded sandstone, limestone, shale sequences massive sandstone massive limestone sand and gravel basalt karst limestone 2 3 4 6 6 6 8 9 10 refers to the consolidated or unconsolidated medium which serves as an aquifer. the larger the grain size and more fractures or openings with in an aquifer, leads to higher permeability and lower attenuation capacity, hence greater the pollution potential. 3 soil media (s) soil thin or absent gravel sand peat shrinking and/or aggregated clay sandy loam loamsilty loam clay loam muck non-shrinking and non-aggregated clay 10 10 9 8 7 4 5 4 3 2 1 refers to the uppermost weathered portion of the vadose zone characterised by significant biological activity. soil has a significant impact on the amount of recharge which can infiltrate into the ground. 2 topography (t) (slope%) 0-2 2-6 6-12 12-18 >18 10 9 5 3 1 refers to the slope of the land surface. it helps a pollutant to runoff or remain on the surface in an area long enough to infiltrate it. 1 impact of vadose zone (i) silt/clay shale limestone sandstone bedded limestone, sandstone, shale sand and gravel with significant silt and clay metamorphic/igneous sand and gravel basalt karst limestone 1 3 6 6 6 6 4 8 9 10 is defined as unsaturated zone material. the significantly restrictive zone above an aquifer forming the confining layers is used in a confined aquifer, as the type of media having the most significant impact. 5 hydraulic conductivity (c) (gpd/ft2) 1-100 100-300 300-700 700-1,000 1000-2,000 >2,000 1 2 4 6 8 10 refers to the ability of an aquifer to transmit water, controlling the rate at which groundwater will flow under a given hydraulic gradient. material within the groundwater system 3 table 1 drastic weighting factors (aller et al. 1987) arzu firat ersoy and fatma gültekin drastic-based methodology for assessing groundwater vulnerability in the gümüşhacıköy and merzifon basin (amasya, turkey) are two methods for evaluating vertical vulnerability based on the following seven parameters: depth to water (d), net recharge (r), aquifer media (a), soil media (s), topography (t), vadose zone impact (i), and hydraulic conductivity (c) figure 2). each mapped factor is classified into ranges (continuous variables) or significant media types (thematic data) having an impact on pollution potential. weighting multipliers are then used for each factor to balance and enhance their importance, the typical rating ranging from 1 to 10 (table 1). the final vulnerability index is a weighted sum of the seven factors. the drastic index (di) can be computed using expression (1): di = dr dw + rrrw + araw + srsw + trtw + iriw + crcw (1) di drastic index for a mapping unit w weighting factor for each parameter r rating for each parameter d, r, a, s, t, i, and c the seven parameters figure 2. drastic method flowchart results depth to water (d) depth to water is defined as the distance (in meters) from the ground surface to the water table. groundwater table depth in the gmb has been measured since 1976. this present study, has used the 2005 values for groundwater table depth. the 167 wells’ location was digitised from the accompanying digital elevation model (dem).groundwater table depth changed between 9-40 m in the gmb the merzifon aquifer has the lowest groundwater depth in the gmb. the deepest groundwater occurred at the end of the impermeable layer over the aquifer media in the mostly confined gümüşhacıköy aquifer. the depth to water table map was then classified into ranges defined by the drastic model and assigned rates ranging from 1 (minimum impact on vulnerability) to 10 (maximum impact on vulnerability) (figure 3). figure 3. depth to water table map of the study area net recharge (r) net recharge is the total quantity of water infiltrating from ground surface to an aquifer on an annual basis. local recharge in the study area comes from inflow by the gümüşsuyu river and its branches, irrigation return flow and direct recharge. the main groundwater recharge source are the gümüşsuyu river, springs, located high in the basin, and irrigation leakage. the average direct annual volume of recharge into the aquifer from the surface of the basin and from the springs is about 11334316 m3 (fırat ersoy, 2007). irrigation pond canals contribute 41.5 m3 recharge in the area (fırat ersoy, 2007) the recharge map was then classified into ranges and assigned ratings from 1 to 8 (figure 4). figure 4. net recharge map of the study area aquifer media (a) aquifer media refers to consolidated or unconsolidated rock which serves as an aquifer. the main aquifer being exploited and that most vulnerable to contamination is partially confined, here called the gümüşhacıköy aquifer the central and northern parts of the gümüşhacıköy aquifer are confined. the clayey layer over high permeability uncemented sediments is 210 m thick from the drilling data. clayey layer thickness of gradually changes from the centre to the north of the basin. the merzifon aquifer is unconfined. the aquifer media was obtained using a subsurface geology map, geological sections, and drilling profiles of the gümüşhacıköy and merzifon aquifers. the main aquifer includes quaternary alluvium and pliocene gravel, sand and clayey levels. the merzifon aquifer’s alluvial fan is relatively thin, consisting of coarse-grained gravel and sand with silt and clay interbeds. the aquifer media were mapped as shown in figure 5. figure 5. aquifer media map of the study area soil media (s) soil media refers to the uppermost portion of the vadose zone characterised by significant biological activity. soil plays a significant role in the amount of recharge which can infiltrate into the ground and hence on a contaminant’s ability to move vertically into the vadose zone. a soil’s pollution potential is largely affected by the type of clay present, such clay’s shrink/swell potential, and soil grain size. soil media in the gmb was determined using drilling profiles. the gümüşhacıköy and merzifon aquifers are covered by clayey gravel and sand and alluvial plains. fractured volcanic rocks are located west and south-west of the gmb and limestone outcrops south of the basin. thickness over the volcanic rock and limestone gradually changes form bottom to top, the hills especially, having little or no soil. the soil media map was then classified into ranges and assigned ratings from 3 to 10 (figure 6). topography (t) topography refers to land surface slope variability. slope degree will determine the extent of pollutant runoff and settling long enough to infiltrate. a digital elevation model (dem) was used to extract the slope of the study area, while 27% of the gmb has a gentle slope, most of the basin has a steep slope. the areas in the extreme east and south consist of ridges which may reach 1050 m.. the topography has been mapped as shown in figure 7. figure 6. soil media map of the study area figure 7. topography map of the study area vadose zone impact (i) the vadose zone is defined as the zone above the water table which is unsaturated. unconsolidated clayey gravel and sand represents the vadose zone in the plain, volcanic rocks and limestone is the vadose zone in the mountain areas. the map of vadose zone impact is shown in figure 8. the aquifer’s hydraulic conductivity (c) hydraulic conductivity is important because it controls the rate of groundwater movement in the saturated zone, thereby controlling the degree and fate of contaminants. hydraulic conductivity values used in this study were derived from pumping test data. hydraulic conductivity varied from 8.79 × 10−6 to 2.21 × 10−4 m/s in alluvium (fırat ersoy, 2007). the hydraulic conductivity of the other rock in the basin was available from the pertinent literature. hydraulic conductivity values for various rock types have been proposed by domenico and schwartz, (1990). the hydraulic conductivity of the limestone and fractured volcanic rocks, located in the west and south part of the basin, were 10-3 m/s and 3x10-4 m/s, respectively. clayey unit permeability is 10−9 to 10−10 m/s. hydraulic conductivity rating distribution is shown in figure 9. arzu firat ersoy and fatma gültekin drastic-based methodology for assessing groundwater vulnerability in the gümüşhacıköy and merzifon basin (amasya, turkey) figure 8. vadose zone impact map of the study area figure 9. hydraulic conductivity map of the study area the aquifer vulnerability map the vulnerability map was obtained using the seven hydro-geological data layers in the arcview gis software environment. drastic scores ranged from 58 to 177, taking into consideration the determined ratings and weightings. these values were reclassified into three classes using the natural breaks (jenks) classification method. the study area’s vulnerability was classed as low (<100), medium (100-140) and high (>140) according to data obtained from hydrogeological investigations (figure 10). the gmb’s high groundwater vulnerability risk zones were mainly located in the centre of the basin where some villages are located and also in the northern and southern parts of the basin. these vulnerable zones covered around 16% of the studied area. four springs in the southern area had as high vulnerability risk. the gmb’s middle groundwater vulnerability risk zones were mainly located in the groundwater recharge area (figure 10), these vulnerable zones covered around 37% of the studied area. the gmb’s low groundwater vulnerability risk zones were mainly located in the west and south-east of the study area (figure 10), these vulnerable zones covered 47 % of the studied area. figure 10. vulnerability index map of the study area the resulting vulnerability map indicated that the highest potential areas for contamination were the central part of the basin where the slope is gentle. in the southern area where karstic limestone outcropped, the high drastic index probably represented the effects of aquifer media and hydraulic conductivity. impermeable volcanic rocks and clay, silty-clay units located in the west and east respectively had low drastic index. conclusions the merzifon-gümüşhacıköy (amasya-turkey) basin is an important agricultural centre for the central black sea section groundwater is a major water source for such activity. groundwater quality has deteriorated due to excessive abstraction of groundwater. this study involved using a gis model and the drastic method for determining the vulnerability of the groundwater in the basin. the aquifer vulnerability map was prepared using depth to water, net recharge, aquifer media, soil media, topography, vadose zone impact, and hydraulic conductivity. the study area was divided into three zones according to groundwater vulnerability assessment results: low (risk index <100); middle (risk index 100–140) and high groundwater vulnerability risk (risk index >140). the drastic method results should be useful in designing aquifer protection and management strategies. the drastic index map indicated that overall potential for groundwater becoming polluted was low for the gmb. low sensitivity areas lay outside the agricultural areas in the basin. the alluvium and most pliocene sediments were used for agriculture in the gmb. the town of gümüşhacıköy is located on an aquifer recharge area. areas determined by the drastic method should thus be given priority in research in terms of contamination. high nitrate concentrations were mainly near urban areas according to the the study area’s analysis (fırat ersoy et al, 2006). high nitrate concentration was likely to be related to wastewater leakage from industrial activities, urbanisation and agricultural practices. two towns and many villages were situated in the study area involving agricultural activities many wells were used for springs. the prevention of groundwater pollution caused by waste and wastewater in gümüşhacıköy’s recharge area was significant owing to groundwater flow being west to east in the basin. regarding urban planning and organisation of agricultural activities in the merzifon and gümüşhacıköy districts, the vulnerability risk map prepared in the study could be most important when considering protection off groundwater quality arzu firat ersoy and fatma gültekin acknowledgements this work was supported by the karadeniz technical university’s scientific research fund project number: 2008.112.005.14. the authors are grateful to geomatics engineer dr. r. nişancı and y. s. erbaş for training in gis mapping. references al-adamat. ran., foster. idl., baban. smj. (2003). groundwater vulnerability and risk mapping for the basaltic aquifer of the azraq basin of jordan using gis, remote sensing and drastic. appl geogr. 23:303–324. aller. l., bennett. t., lehr. jh., petty. rh., hackett. g. (1987). drastic: a standardized system for evaluating groundwater pollution potential using hydrogeologic settings. usepa report 600/287/035, robert s. kerr environmental research laboratory, ada, oklahoma. baalousha. h. (2006). vulnerability assessment for the gaza strip, palestine using drastic. env geol 50:405–414. cameron. e., peloso. gf. (2001). an application of fuzzy logic to the assessment of aquifers’ pollution potential. env geol 40:1305– 1315. civita. m (1994). le carte della vulnerabilit`a degli acquiferi all’inquinamiento: teoria e pratica [contamination vulnerability mapping of the aquifer: theory and practice]. quaderni di tecniche di protezione ambientale, pitagora, italy. domenico. p. a., and schwartz. f. w. (1990). physical and chemical hydrogeology, john wiley and sons, new york, 506 pp. dsi. (2006). yıl sonu faaliyet raporu, i̇şletme bakım şube müdürlüğü, samsun (in turkish). fırat ersoy a., ersoy h. and gültekin f. (2006). nitrate, nitrite and ammonia contamination in groundwater: a case study from gümüşhacıköy plain. turkey, asian journal of water, environment and pollution, vol. 4, no. 1, pp. 107-118. fırat ersoy a. (2007). gümüşhacıköy (amasya) akiferi’nin yeraltısuyu akım modeli. phd thesis, ktu graduate school of natural and applied sciences, trabzon, turkey (in turkish with english abstract). fırat ersoy a., gültekin f. (2008). modeling groundwater flow in the agricultural area of gümüşhacıköy (amasya, turkey). bulletin of engineering geology and the environment, vol. 67, no. 4, pp. 529-535. foster s. (1987). fundamental concepts in aquifer vulnerability, pollution risk and protection strategy. in: van duijvenbooden w, van waegeningh hg (eds) vulnerability of soil and groundwater to pollutants. committee on hydrological research, the hague, pp 69–86. jamrah a., futaisi aa., rajmohan n., al-yaroubi s. (2007). assessment of groundwater vulnerability in the coastal region of oman using drastic index method in gis environment. environ monit assess. doi:10.1007/s10661-007-0104-6. margat j. (1968). vulnerabilite des nappes d’eau souterraine a la pollution [groundwater vulnerability to contamination]. bases de al cartographie, (doc.) 68 sgc 198 hyd, brgm, orleans, france. massone h, mauricio mq, martínez d. (2010.) enhanced groundwater vulnerability assessment in geological homogeneous areas: a case study from the argentine pampas. hydrogeology journal, 18: 371–379. melloul aj., collin m. (1998). a proposed index for aquifer waterquality assessment: the case of israel’s sharon region. j environ manage 54(2):131–142. mendoza ja. barmen g. (2006). assessment of groundwater vulnerability in the río artiguas basin, nicaragua. env geol 50:569–580. merchant jw. (1994). gis-based groundwater pollution hazard assessment: a critical review of the drastic model. photogramm eng remote sens 60(9):1117–1127. sener e., sener s., davraz a. (2009). assessment of aquifer vulnerability based on gis and drastic methods: a case study of the senirkent-uluborlu basin (isparta, turkey). hydrogeology journal, 17: 2023–2035. url-1.,http://www.dmi.gov.tr/verideğerlendirme/yıllık-toplam-yagisverileri-amasya,15 march 2011. van stempvoort d., ewert l., wassenaar l. (1993). aquifer vulnerability index (avi): a gis compatible method for groundwater vulnerability mapping. can water resour j 18:25–37. vias jm., andreo b., perles mj., carrasco f. (2005). a comparative study of four schemes for groundwater vulnerability mapping in a diffuse flow carbonate aquifer under mediterranean climatic conditions. env geol 47:586–595. search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 using neuronal networks to determine site response from coda waves: application in armenia, colombia earth sciences research journal earth sci. res. j. vol. 8, no. 1 (dec. 2004): 19 24 manuscript received april 2004 19 paper accepted october 2004 temporal variation of coda-q at galeras volcano, colombia édgar moncayo1, carlos vargas2 and juan durán3 1universidad nacional de colombia. e-mail cavargasj@unal.edu.co 2universidad nacional de colombia. e-mail cavargasj@unal.edu.co 3instituto de investigaciones geológico minerasingeominas. abstract the temporal variation of coda-q was analyzed for galeras volcano, located in the andes of southwestern colombia. the q -1 value was calculated using the single back scattering (sbs) model working with seismograms filtered in frequency bands centered in 1.5, 3.0, 6.0 and 12.0 hz. the analysis was done in 1990-2002 period, using the urcr station and the frequency dependence q = qo f –n was calculated before and after some eruptions. we used volcano tectonic events with magnitude lower than 1.5 and rms, gap, vertical and horizontal errors lower than 0.1s, 180 and 1km. the temporal change of attenuation occurred in two stages, the first in 1989-1992 period, when the q -1 value increases from 0.015 to 0.070. the second is the 1992-2002 period that is characterized by a decreasing q -1 value. before july 1992 eruption the attenuation increase is associated with the extrusion of an andesitic dome, after this eruption the dome was destroyed and the q -1 value drops. starting from 1994 low attenuation values was associated with low galera’s activity. the analysis shows that q value increases before an eruption and decreases after it. key words: attenuation, coda-q, scattering, colombia, galeras volcano. resumen la variación temporal de coda q fue analizada en el volcán galeras ubicado en el sureste de los andes colombianos. el valor q -1 fue calculado usando el modelo “single back scattering” (sbs) trabajando con sismógrafos filtrados en bandas de frecuencia centradas en 1.5, 3.0, 6.0 y 12.0 hz. el análisis fue hecho durante el periodo 1990-2002, usando la estación urcr y las frecuencias dependientes q = qo f –n fueron calculadas antes y después de algunas erupciones. usamos eventos vulcano tectónicos con magnitud menor que 1.5 y rms, gap, vertical y horizontal con un margen de error menor que 0.1s, 180 y 1km. los cambios temporales en la atenuación se dieron en dos etapas, la primera en el periodo 1989-1992 donde el valor q -1 se incremento desde 0.015 a 0.070. el segundo en el periodo de 1992-2002, caracterizado por un decrecimiento del valor q -1. la atenuación se incrementa antes de la erupción de julio de 1992. esta erupción está asociada a la extrusión de la cúpula de los andes. después de esta erupción la cúpula se destruyó y cayo el valor q -1. iniciando en 1994, los bajos valores de atenuación fueron asociados a la baja actividad del volcán galeras. el análisis muestra que el valor q se incrementó antes de la erupción y decreció después. palabras clave: atenuación, coda-q, dispersión, colombia, volcán galeras. © 2004 esrj unibiblos. edgar moncayo et al. introduction galeras volcano (lat. 1°14’n, lon. 77°22’w) is 4200m-high andesitic volcano in the andes of southwestern colombia and is the most recent active center of galera’s volcanic complex (cepeda, 1986). the active cone has 320 m in diameter and lies within a caldera that is breached to the west. galeras is historically characterized by vulcanian type explosions (calvache, 1990). the most important one recorded, in 1936, generated a pyroclastic flow to the northeast of the volcanic structure. the volcano lies 8 km from pasto city and various figure 1. location of galeras volcanic complex area, and the earthquakes (circles) and stations (s) used for determinations. other towns locate on its flanks. in total, 400.000 people live close to the volcano, inside this hazard zone. since 1988, when galeras was reactivated, after 50 years of repose, it entered in a fumarolic and degassing stage, with seven eruptions between 1989 and 1993. an andesitic dome was extruded in september 1991 and destroyed during july 1992 eruption. since 1994 galeras is in calm stage with lesser explosions and seismic crises in 1995, 1982, 2000 and 2002.many studies have been done since cepeda (1986) and calvache (1990). the latter made a stratigraphic correlation based on c 14 in the 1990-1993 period galeras was declared the volcano of the decade and much research was done. on 1993 january 14 during an international workshop while a fieldwork was in progress in the crater, there was an eruption where several scientists died and others resulted injured. many works were published about galeras in different disciplines: calvache et al., (1997) worked the stratigraphy and evolution of the volcano, a model of degassing was proposed by stix et al., (1993) and geochemistry studies were done by zapata et al., (1997). in the seismology field, the works of gómez and torres (1997) and narvaéz et al., (1997) analyzed the tornillo-type signals whereas the different features seismic sources has been analyzed by ortega et al., (2002). any investigation about temporal variation of the attenuation hasn’t been done, so this is an innovative study that enlarges the knowledge about galeras volcano. geological setting galeras is located in southwestern colombia, a region with a metamorphic rocks basement of precambrian and paleozoic ages. the basement is overlaid by metamorphic rocks of cretacic age with low and medium grade associated with amigdular metabasalts. all of this is covered by volcano sedimentary units of tertiary age that made a plateau, over which the pleistocenic and holocenic volcanoes have emerged. the tectonic plate of this region is very complex, as a result of the collision between the nazca and southamerican plates. this causes the uplift of the andes and the volcanism in the region. the structural trend is n40°e and the principal tectonic feature is the romeral fault zone, which has been interpreted as the limit between continental crust to the east and the oceanic crust to the west (barrero, 1979). this system includes the silvia-pijao and buesaco faults, both of which cross under galeras, and are associated with many old caldera systems (figure 1). method we follow the single back scattering (sbs) model of aki and chouet (1975) for calculating q -1 from the ending portion of the records of local earthquakes, or coda waves. the q factor contains information about seismic energy for two reasons: intrinsic losses (qi) and radiation losses (qr). 20 temporal variation of coda-q at galeras volcano, colombia 21 the method assumes that local earthquake coda is composed by the sum of secondary s waves produced by heterogeneities inside a propagation medium. in this context, coda is the portion of seismogram corresponding to back-scattered swaves, whose travel times are assumed to be twice the s-wave travel time taken from the origin of the earthquake. we assume that the coda portion finishes when the signal-noise ratio is lower than approximately 2 (aki and chouet, 1975). the sbs model considers the back-scattered swaves produced by in homogeneities (scatterers) distributed randomly and uniformly in the lithosphere. the basic and simplified equation is: ln [ t2 aobs (f | r, t)] = c – q-1 2π f t (1) where c is a constant, aobs (f | r, t) is the squared amplitude of the record, f is the frequency, t is the time and q is the quality factor, which contains information about the attenuation of seismic energy. equation (1) has the form y = bx + c where: y = ln [ t2 aobs (f | r, t)] x = 2π f t b = q -1 (2) thus from this model it is possible to calculate q -1 from the slope b, using least-squares techniques. data and processing the seismic data used in this study covered the period between september 1989 and june 2002. the seismograph network installed at galeras is composed of 31 stations equipped with vertical component l-4c seismometers with 1second natural period. the ground motions were telemetric from an observatory located at pasto city and the information taken was signals sampled at a rate of 100s-1. however, seismograph stations have not operated continuously because of changes on network configuration, we used earthquakes with magnitude less than 1.5 to avoid clipping. a total of 435 earthquakes were selected for this study. for hypocenter determinations we used the hypo71 computer program (lee and lahr, 1975). table 1. velocity model used to locate the hypocenters depth at top of the layer (km) p-wave velocity (km/s) 0.0 3.50 2.0 3.70 4.0 4.00 8.0 6.00 26.0 6.80 44.0 8.00 vp / vs 1.78 defined to use regional seismic reflection profiles and based in the work by meissner et al., (1977). we carefully selected seismograms of located earthquakes with root mean square residuals of calculated travel times and observed times (rms) less than 0.1s, azimuthal separation between stations (gap) less than 180°, and formally computed horizontal and vertical errors in locations of less than 1 km. figure 1 shows the location of the seismograph stations and selected earthquakes. the coda-q was calculated for four frequency bands centered at 1.5, 3.0, 6.0 and 12.0 hz, (see figure 2), figure 2. data processing example a. raw seismogram b. bandpass filtered seismogram with centered band at 12 hz c. energy density d. q-1 values. we used the velocity model adopted by the pasto volcano observatory (table 1), using the sbs model with a variation of the in-house routine for matlab developed by vargas and mora (2000) with a 10 seconds window. because variations in q can be introduced by changes in the spatial location of the earthquakes and be misinterpreted as a temporal variation, we show in figure 3. that the hypocenters did not shift through time. edgar moncayo et al. figure 3. space-time and depth-time plots of the selected earthquakes. note that the hypocenters did not shift significantly with time, suggesting that any change in q-1 could be due to temporal changes rather than spatial changes associated with earthquakes’ locations. we used q-values at all stations because we note that during the entire 1989-2002 period the station were not operated, and finally find that these present a similar behavior. additionally we worked with the urcr station, because it has the best record and a wide number of records through the time. analysis and results the coda-q was analyzed for frequency bands centered in 6 and 12 hz. figure 4 shows that q -1 values for the 12 hz band show a clear change in time, maybe contribution of intrinsic absorption to the total attenuation is higher at 12 hz that at 6 hz for galeras volcano. the temporal variations of q -1 value obtained by the sbs model show clear tendencies that follow the changes in volcanic activity. to find the frequency dependence of q -1, we fit the values of q -1 for each frequency band (q = qo f -�). table 2 shows the results obtained from qo and for each eruption (pre and post-eruptive periods). there is a clear difference before and after each eruption. temporal changes can be separated in two stages, the first, before july 1992 eruption, shows a fast rise of q that begins in october 1991 and reaches a maximum value before the eruption. the second stage begins after this eruption, since this date, a long-lasting drop in q is registered, in a short time a little raise happening before june 1993 eruption, and from 22 figure 4. result of q -1 analysis a. 6, b. 12 hz. notice the temporal changes at 12 hz. the q-1 values for 6 hz show scatter. arrows mark the dates of eruptions and periods of unrest at galeras volcano. table 2. frequency dependence parameters before and after of some galeras eruptions period q -1 η before jul.-92 eruption 2.8290 ± 1.32 0.8229 ± 0.12 after jul.-92 eruption 3.2837 ± 1.31 1.4565 ± 0.44 before 16 apr.-93 eruption 5.4490 ± 1.43 0.8541 ± 0.15 after 16 apr.-93 eruption 6.0613 ± 2.60 0.9271 ± 0.15 before jun.-93 eruption 5.2673 ± 1.98 0.9840 ± 0.13 after jun.-93 eruption 6.0890 ± 1.67 1.2225 ± 0.10 1994 to 2002 there were not important changes. the frequency dependence parameters (table 2) show an increase before eruptions and a fall after these, a similar trend shows the q -1 value figure 5. temporal variation of coda-q at galeras volcano, colombia discussion and conclusion evident changes in q-1 occurred in 1990-2002 period. these changes cannot be ascribed to spatial changes in earthquakes hypocenters (figure 2). we concluded that these differences are attributed to changes in the volcano activity and discarded changes related to spatial variations. the values obtained in this work show two stages separated by july 1992 eruption. this date marks an important change at galeras activity, because before this date the attenuation had a clear upward tendency, and after it, the q -1 value decreased, with no similar values being recorded since. finally, the q-1 values show a slow recovery since 1995 to 2002 with a steep increase at the end. our results are according to the study by londoño et al., (1998) about nevado del ruiz volcano, colombia. they observed a systematic decrease in the q-1 values before and after two phreatomagmatic eruptions in 1985 and 1989. at galeras, the temporal variation shows that 12 hz is the best frequency value to discern changes in the activity. these variations in q-1 are clearly associated with changes in the volcanic activity as follows: the fall in the attenuation between 1990 and 1991 can be associated with a degassing process that begins after 1989 eruption. the rise between october 1991 and july 1992 is related to an important change in volcanic activity, such as the extrusion of an andesitic dome with accompanying shallow deformation, increase in the rates and amplitudes of long-period seismic events, and a decrease in the so2 flux. this suggest that the conduit was partially blocked because the viscous magma in the dome acted like a seal that stopped the gas flux, this caused an increase in pressure and attenuation that ended on july 1992 eruption. after this explosive eruption the dome was destroyed and the gas released. for this reasons the pressure drops and attenuation appears. before june 1993 eruption a little rise took place. we suggest this can be caused by small plugs formations that are destroyed in subsequent eruptions. for that reason we didn’t have enough data to get consistent results. attenuation parameters were analyzed on july 1992, april 1993, and june 1993 eruptions. although the error bars are superimposed, (figure 5), the frequency dependence of q shows a similar tendency because we observe three cases in which the attenuation (q-1) figure 5. changes in q -1 values before and after some eruptions. note the decrease of q-1 values after eruption. exhibits a decrease in value before and after each eruption. the parameters also show a long-term decreasing tendency that is according to the fall in the q -1 value during the 1992-1994 periods. in 1994-2002 period the q -1 value were stable. we suggest that it may be associated with repose of the volcano, because it reached an equilibrium state after 1993 eruptions. in this stage little rises in the q -1 value took place. these could be related to minor explosions and ash emissions possibly caused by small gas accumulations (i.e., during june 2002). the changes in the q -1 values show that this parameter can be used as an important tool at galeras volcanic surveillance, because is clearly related to the activity changes, especially before the eruptions. we suggest implementing an attenuation monitoring system. measure by q-1, can be a good complement to the information currently used, for instance, number of seismic events, deformation measurements, and fumaroles gas geochemistry thus helping improve the monitoring. finally we recommend more detailed studies that investigate a possible use of q -1 for forecasting. 23 edgar moncayo et al. 24 acknowledgements we thank ingeominas-corp for technical support in this study. we also thank the comments of our colleagues from volcanological and seismological pasto observatory, ovsp. specially thanks to r. torres for suggestions and comments. references aki, k., chouet, b., (1975). origin of coda waves: source, attenuation and scattering effects. journal of geophysical research, 80: 615-631. barrero, d. (1979). geology of central western cordillera west of buga and roldanillo, colombia. informe ingeominas, bogotá (colombia). calvache, m. (1990). geology and volcanology of the recent evolution of galeras volcano, colombia. m.s. thesis, louisiana state university. calvache, m., cortés, p., williams, s. (1997). stratigraphy and chronology of the galeras volcanic complex, colombia. journal of volcanology and geothermal research, 77: 5-19. cepeda, h. (1986). investigaciones petrológicas en el ámbito de las planchas 429-pasto y 410-la unión, con especial énfasis en el complejo volcánico del galeras. informe ingeominas, medellín (colombia). gomez, d., torrres, r. (1997). unusual lowfrequency volcanic seismic events with slowly decaying coda waves observed at galeras and other volcanoes. journal of volcanology and geothermal research, 77: 173-193. lee, w. h., lahr, j. c. (1975). hypo71: a computer program for determining hypocenter, magnitude, and first motion pattern of local earthquakes. u.s. geological survey. open-file report 75-311. londoño, j. m., sánchez, j. j., toro, l. e., gilcruz, f, and bohórquez, o. p. (1998). coda q before and alter the eruptions of 13 november 1985, and 1 september 1989, at nevado del ruiz volcano, colombia, bulleting of. volcanology, 59: 556-561. meissner, r., fluedh, e., stibane, f., y berg, e., (1977), dinámica del límite de placas activo en el sw de colombia, de acuerdo a recientes mediciones geofísicas. en: proyecto cooperativo internacional nariño "la transición oceáno-continente en el suroeste de colombia": 169-199. narváez, l., torres, r., gómez, d., cortés, p., cepeda, h., stix, j. (1997). “tornillo”-type seismic signals at galeras volcano, colombia, 1992–1993. journal of volcanology and geothermal research, 77: 159-171. ortega, a, torres, r., gómez, d. (2002). caracterización de fuentes sismogénicas de eventos volcano-tectónicos (vt) en el complejo volcánico galeras. memorias. primer simposio colombiano de sismología. stix, j., zapata, j., calvache, m., cortez, g., fischer, t., gómez, d., narváez, l., ordoñes, m., ortega, a., torres, c., williams, s. (1993). a model of degassing at galeras volcano, colombia, 1988-1993. geology, 21: 963-967. vargas, c, mora, p. (2000), conceptos sobre atenuación de ondas coda. red sismológica regional del eje cafetero–viejo caldas y tolima: 37-43. zapata, j., calvache, m., cortés, p., fischer, t., garzón, g., gómez, d., narváez l., ordóñez, m., ortega, a., stix, j., torres, r., williams, s. (1997). so2 fluxes from galeras volcano, colombia, 1989–1995: progressive degassing and conduit obstruction of a decade volcano. journal of volcanology and geothermal research. 77: 195-208. using neuronal networks to determine site response from coda waves: application in armenia, colombia earth sciences research journal earth sci. res. j. vol. 8, no. 1 (dec. 2004): 34 44 manuscript received june 2004 paper accepted november 2004 34 unexpected response of mt. wrangell volcano, alaska, to the shaking from a large regional earthquake: a puzzle for intermediate-term earthquake-volcanoes interactions. john j. sánchez and stephen r. mcnutt e-mail:jjalaska@giseis.alaska.edu geophysical institute, university of alaska, fairbanks 903 koyukuk dr. fairbanks, ak 99775, usa abstract on november 3, 2002 three segments of the denali fault in interior alaska ruptured during a mw 7.9 earthquake, offering a unique opportunity to study earthquake-volcano interactions. out of the 24 volcanoes that are seismically monitored by the alaska volcano observatory (avo) only mt. wrangell, the closest volcano to the epicenter (247 km), showed a clear response to the shaking in the intermediate-term (weeks to months) time scale. the response was unexpected because it consisted of a decline by at least 50% in the volcano’s seismicity rate (mostly lowfrequency events) that lasted for five months. because most well documented previous instances of short-term (minutes to days) responses of volcanic centers to the passing waves of distant earthquakes, have all been seismicity increases, the decline in seismicity at mt. wrangell poses a controversial puzzle. by using several independent methods to measure the seismicity rate at the volcano from before to after the main shock, and applying rigorous statistical testing, we conclude that the change in seismicity at the volcano was a real effect of the denali earthquake. we suggest that a depressurization of the volcanic plumbing system took place either as a result of sudden decompression (static stress changes) or because of creation of new pathways resulting from the strong shaking (dynamic stresses). at present we cannot distinguish between these two possibilities. key words: alaskan volcanoes, denali fault earthquake, mt. wrangell, seismicity decreases resumen en noviembre 3 de 2002 durante el terremoto de 7.9 mw, se quebraron tres segmentos de la falla denali en alaska, ofreciendo una única oportunidad de estudiar los terremotos por interacción volcánica. aparte de los 24 volcanes monitoreados sismicamente por el observatorio volcanológico de alaska (avo), sólo el mt. wrangell, el volcán más cercano al epicentro (247 km), mostró una respuesta clara ante el movimiento en un término de escala de tiempo intermedio de semanas a meses. la respuesta fue inesperada porque consiste en un declive de por lo menos 50% en la velocidad sismológica del volcán (frecuencia de eventos cada vez más lenta) que duró hasta cinco meses. puesto que muchos casos de corto término (minutos a días), muy bien documentados previamente, responden a olas de distantes terremotos centros volcánicos, todos han tenido incrementos en la sismisidad; por tanto, el declive sísmico en el volcán tuvo un efecto real sobre el terremoto denalie de mt. wrangell que ocasionó unexpected response of mt. wrangell volcano, alaska, to the shaking from a large regional earthquake: a puzzle for intermediate-term earthquakevolcanoes interactions. 35 un controvertido rompecabezas. usando muchos métodos independientes para medir la velocidad sísmica en el volcán desde antes hasta después del principal choque y aplicando rigurosas pruebas estadísticas, concluimos que los cambios sísmicos fueron un efecto real del terremoto denalie. sugerimos también que la despresurización del sistema del volcán tuvo efecto por la descompresión (drásticos cambios estáticos) o por la creación de nuevos caminos fue producto de un movimiento brusco (acentuada dinámica). hasta el momento no podemos diferenciar entre estas dos posibilidades. palabras clave: volcanes de alaska, terremoto de la falla delani, mt. wrangell, decrecimiento de la sismisidad. © 2004 esrj unibiblos. introduction the interactions between earthquakes and volcanoes have been a subject of scientific interest for a long time and, in recent years, a topic of research thanks to the increasing number of reports and the improvement of precise instrumentation. the link between tectonic and volcanic activity is better understood by considering the scales of time and space: at very long times (hundreds to millions of years) and large distances (tens of thousands of kilometers) the connection is explained under the theory of plate tectonics and one only needs to glance at a map of global tectonics to see that the distributions of earthquakes and volcanoes are not the result of mere coincidence. in the very short time scale of minutes to days and distances of kilometers to tens of kilometers, there is documentation of eruptions following the shaking from an earthquake, such interactions are exemplified by the 1975 summit eruption of kilauea volcano an hour and a half after the 7.5 kalapana earthquake (tilling et al. 1976) and the eruption of cordon caulle volcano two days following the magnitude 9.5 1960 chile earthquake (gerlach et al. 1988; gvn, 2004). in the intermediate-term scale of weeks to months or years, new evidence and research points towards a positive link, meaning that a regional earthquake with magnitude 6.5 or larger can trigger unrest at volcanoes located several hundreds of kilometers away (hill et al, 2002). in all instances of earthquake-volcano interactions, all observations indicate that a distant earthquake is able to cause changes in activity at volcanic (and nonvolcanic) areas in terms of increases in seismicity or eruptions. the opposite effect, however, has not been clearly observed and the literature, with only a few exceptions (alvarado, et al. mcnutt and sanchez, 2002; sanchez and mcnutt, submitted), does not include documentation of activity at volcanoes being inhibited by either the transient or permanent changes induced by a distant earthquake. figure 1. map of alaska showing the surface rupture from the dfe (thick solid line) and the epicenters of the october 23, 2002 nenana mountain earthquake (n) and the three sub events of the november 3, 2002 denali earthquake (s1, s2, s3) (surface rupture from eberhart-phillips et. al., 2003, epicenters of n and s1-s3 are from aeic, 2003); mt. wrangell volcano(star); other volcanoes seismically monitored by avo (dots); and main faults and tectonic structures (gray lines). rectangle encloses the area of figure 2. in november 2002, the mw 7.9 denali fault earthquake (dfe, figure 1) triggered short-lived seismicity at volcanic areas located along the rupture direction at distances as far as 3660 km from the epicenter (eberhart-phillips et al. 2003) and an apparent intermediate-term drop in the seismicity at mount wrangell volcano, located in the perpendicular direction with respect to the rupture and roughly 240 km away from the epicenter. this effect may have lasted for five months. in this paper we provide evidence, by means of statistical testing, that the decline in seismicity observed at mount wrangell volcano was real and can john j. sanchez and stephen r. mcnutt 36 be attributable to the dfe. to do so, we analyze the catalog of located earthquakes at mount wrangell, provided by the alaska volcano observatory and test it for the significance of the changes in seismicity following the dfe. data the alaska volcano observatory (avo) monitors the seismic activity at mount wrangell since july 2000, when two seismograph stations were installed. locations of earthquakes were possible beginning august 2001, when two additional stations were deployed (figure “wrangell area map”). since then and until the time of this writing (march, 2003) 5,451 events triggered the automatic detection system, but only 508 could be located (figure 2). the reason for this low detected/located ratio stems from a combination of low signal-to-noise ratio, limited number of seismograph stations and most important of all, the characteristics of the first arrivals of the earthquakes at the different stations. of all located earthquakes, 86.4% have been identified as lowfrequency seismic events (lf). lf events show emergent first arrivals to the seismograph stations, have not distinguishable pand s-arrivals, and have spectra with dominant frequencies between 1 and 5 hz (minakami, 1974). the remaining percentage of seismic events have been identified as high-frequency earthquakes, which may not different in their source mechanism from earthquakes that occur along tectonic areas such as the denali (alaska) , san andreas (california), or north anatolian faults (turkey). because seismic activity in the mount wrangell area is monitored and recorded in continuous and triggermode we can use three largely different datasets: digitally filtered waveform data and unfiltered analog seismograms of station wanc (located 3.2 kilometers west of the vent) and the catalog of located earthquakes. we will refer to each of these datasets as pseudohelicorder, helicorder, and located earthquake counts, respectively. the time period covered by our different data sets is september 1, 2002–december 31, 2002 for the pseudohelicorder and helicorder event counts and august 5, 2001–march 18, 2004 for the catalog of located earthquakes. method we measure the seismicity rates at mount wrangell based on daily event counts out of the three datasets described and the significance of the changes is estimated by applying the z (habermann, 1987) and β (mathews and reasenberg, 1988) statistical tests to the data in both, ordered and randomized time series. figure 2. detail of rectangular area showed in figure 1. red thick line: surface rupture from the dfe; fault plane solutions shown for the dfe sequence earthquakes in lower hemisphere projections with compressional areas in black and dilatational areas in white; green circles: some of the earthquakes located by aeic between october 23, 2002 and october 31, 2003, with depths above 30 km and magnitude 3 and above. (4,653 epicenters are shown); red squares: volcanic earthquakes (lf and hf) at mt. wrangell located by avo between august 5, 2001 and march 18, 2004; white triangles: seismograph stations on mt. wrangell. see other conventions as in figure 1. we define the seismicity rate changes at the volcano in terms of the rates during the periods before and after the dfe. the z test is a parametric test for estimating the difference between two means, in our case the two means being compared are simply the averages of daily earthquake counts for the period preceding and following the dfe. z-statistic formula is: 2 )2( 1 )1( )21( 22 n s n s mm z + − = (1) where m1, m2 mean rates during the two periods (before and after the dfe); s1, s2: standard deviations of the rates; n1, n2: numbers of earthquakes in each time period. the z value computed is interpreted in terms of significance as the number of standard deviations from the mean of a normal distribution (i.e., z = 1.64 represents 90% significance, z = 1.96 1represents 95% significance, and z = 2.57 represents 99% significance) (habermann, 1987). the sign of the z value indicates the polarity of the change (i.e., z < 0 indicates rates increases, z ≈ 0 indicates no change, and z > 0 indicates rate decreases). we also use a moving windows technique to compute the z values. this is done by finding all possible subdivisions of the data in two halves and computing the z value for all pairs of means, allowing us to estimate the significance of a difference in mean seismicity rates at any given point (time) in the curve. unexpected response of mt. wrangell volcano, alaska, to the shaking from a large regional earthquake: a puzzle for intermediate-term earthquakevolcanoes interactions. 37 we investigated the sensitivity of the z values versus time curve to the presence of abrupt changes in seismicity rates by computing the z values of several synthetic data sets with varying amounts of change (decreases) in the number of earthquakes per unit time. our synthetic datasets have the following features: (1) the seismicity rates from before to after the change are different but remain constant within each period; (2) the seismicity rate after the change is some normalized percentage of the seismicity rate before the change; (3) the value of the earthquake count at the middle point of the time series is zero, simulating a useless record the day of the dfe. to somewhat mimic our pseudohelicorder counts data, we created synthetic time series containing one break point and two segments, simulating drops in seismicity of 100% (total drop of seismicity), 95%, 75%, 50%, 25% , and 0% (no change in seismicity). we also investigated the effect of having a seismicity rate that decreases in two steps preceding the main drop. we did this by creating a data set that has two break points and three segments with constant rate each. in this way we simulate a constant rate, and a drop of 25% that precedes a second drop of 50%, after which the seismicity stays at that level. the length of these synthetic data sets is 101 data points. to mimic our data of located lf seismic events, we created a longer synthetic data set in which there are two break points and three segments. in this case the second segment contains a drop in seismicity of 80% and the third segment represents a recovery of the seismicity rates to background values (rate equal to first segment). the length of this synthetic data set was 487 data points. seismicity rates changes, can also be characterized using the -statistic as described mathews and reasenberg (1988): ttn ttnn n nen ab baba a aa / )/( )var( )( − = − =β (2) where: na=number of earthquakes in the region of interest after the perturbing main shock. ta=time period after. e(na)=expected number. e(na) = r.ta, and r = nb/tb is the ambient rate of earthquakes estimated from a background period, tb. nb= number during the background period. note that for a binomial distribution var(na) = npq where p = ta/t, q = 1 – p = tb/t, t = ta + tb and n = na + nb. because the observed value of na may be anomalous, e(na) is used instead, and n = nb(1 + ta/tb). the sign of β in this case indicates the polarity of the change in seismicity rate (i.e., β < o represents decrease in rates, β ≈ 0 means no change, and β > 0 represents rates increases). because there are various methods to compute β that differ in their underlying assumptions and the validity of those assumptions cannot readily be verified (gomberg et al. 2003) we do not ascribe absolute significance level to the rate variations computed with β statistic and only use to spatially assess the nature of the seismicity rate changes in the area of mt. wrangell. results pseudohelicorder and helicorder counts the plots of cumulative daily number of volcanic earthquakes estimated from pseudohelicorder plots and helicorder records are shown in figure 3. the absolute values in numbers of events are different because of the different criteria used to count earthquakes on each type of record and the different noise levels (usually less noise in pseudohelicorder plots) and sensitivities. the change in slope indicating the change in seismicity rate is more obvious in the pseudohelicorder counts curve. the z values at the date of the dfe are z = 6.7358 and z = 2.5531 for pseudohelicorder and helicorder counts, respectively, which correspond to confidence levels for the difference in mean rates before and after de dfe above 99.99% and 99% respectively. from the cumulative number and z values plot for pseudohelicorder counts, we note that the decrease in seismicity begins before the dfe, around the time of the mw 6.7 nenana mountain earthquake (october 23, 2002). the helicorder count data although somewhat noisier also shows the change in rates. we tested the hypothesis that the data showing the suggested decrease in seismicity at mount wrangell is no different from background noise, represented by many random series. to do this we randomly permute our data sets of pseudohelicorder and helicorder counts, plot the cumulative number of earthquakes versus time and compute the z values for the permuted series. this is done 100 times for each data set being analyzed and we always plot the randomly permuted series along with the original data. we observe less spread in the random series for pseudohelicorder counts; also the data from pseudohelicorder records equals the random noise towards both ends of the time series, but is different towards the middle of the series. the helicorder count data is no different from the random noise. john j. sanchez and stephen r. mcnutt 38 figure 3. seismicity rates and z values estimated from earthquake counts on pseudohelicorder and helicorder records from station wanc on mt. wrangell., between september 1, 2002 and december 31, 2002. top: cumulative number of volcanic earthquakes versus time; bottom: plot of z values versus time. squares and circles represent the cumulative sum and z values curves, respectively, for actual data; gray lines represent the result of 100 random permutations of the data in each plot. vertical dashed and solid lines mark the dates of the nenana mountain earthquake (10/23/2002) and the denali fault earthquake (11/3/2002), respectively. percentages indicate the confidence levels at which the two means, before and after the dfe, are different as evaluated by the z-test. analysis of synthetic counts datapseudohelicorder counts. figure 4 shows several cumulative numbers versus time and z values versus time plots for six synthetic data sets all representing a sharp decrease in seismicity, with varying amounts of change. the results show that the spread in the randomly permuted counts increases as the amount of change in seismicity rate increases and so does the separation between observations and random permutations. this means that the stronger the rate change, the more we should be able to see it above random noise. also we see that even in the case of small rate changes the z values versus time curve is sharply peaked near the time of the change. careful examination of the z values curves reveals that exactly one data point before the time where the rate change starts, the peak occurred. this is not the figure 4. cumulative sum and z values plots for six synthetic data sets with one break point in the middle and varying amounts of rate decrease. a vertical dashed line marks the time of the change, in which a value of zero in the counts has been assigned, simulating a useless record. the top plot of z values, for a drop in seismicity of 100%, has been cut off at z≈50 because the peak value is at infinity. other conventions as in figure 3. caused by the number of data points being either even or odd, or by the presence of a zero count at the time of the change, nor by the position of the time of change in the data set. the effect is expected because the z values curve is tracking the evolution of the cumulative sum in time. let n be the total number of data points and ni be the position of the breakpoint. the cumulative curve will be constant until exactly ni1 and then drops to a different rate. that is where the two segments being compared are the most different. an additional interesting result from our testing is the observation that the steepness of the rise in z values depends on the strength of the rate change. unexpected response of mt. wrangell volcano, alaska, to the shaking from a large regional earthquake: a puzzle for intermediate-term earthquakevolcanoes interactions. 39 the z values curve is sharply peaked for large changes in rate and rises more gradually for small rate changes (figure 5). figure 5. the data sets used in figure 4, plotted at the same scale to allow comparison of the z values curve. the inset is an enhanced view of the rectangular at the base of the z values plot. see other conventions as in figure 4. the analysis of a synthetic data set with two break points (figure 6) shows that when a minor decline in rate precedes a larger one, the z values curve is doublepeaked, but surprisingly, the first peak has a larger z value, regardless of the length of the two segments preceding the main break point. as the length of the second segment shortens, the peaks come together into one. similar effects are also present in the pseudohelicorder count data, with the two largest peaks in z values occurring several days before the date of the dfe. the first peak in z values occurs on october 15, 2002, and it is sensing a zero count value (unusable record) on october 16, after which the counts drop for a few days. we tested the influence of this zero value by replacing it with a value similar to its neighbors and the double peak persisted, although with slightly decreased significance. figure 6. cumulative number and z values plots for synthetic data with two break points representing an initial drop in rate by 25% followed after sometime by a larger drop of 50%. vertical arrows mark the time of the drops. the start time of the first (and smaller) drop varies, while the start time of the larger drop is constant. other conventions as in figure 3. the second peak in z values occurs on october 23, 2002 and it is sensing a general decrease in number of daily counts (mostly good days) leading to the strongest drop on november 3. we suggest that the relatively gradual increase in z values leading towards the peak on october 23 reflects the magnitude of the seismicity drop following the dfe, although we cannot reject the possibility that the decrease in seismicity actually started near october 23, 2002. john j. sanchez and stephen r. mcnutt 40 analysis of synthetic counts data-located earthquake counts. when we use a larger data set of 487 data points (equal to the number of located lf seismic events) simulating a time series with a final segment in which the rates recover to the background values, we observed that the z values curve has two peaks, the first one representing the drop in seismicity and a second one, with negative z value, indicating the recovery (increase) to background rates (figure 7). figure 7. z values and cumulative sum versus time for a synthetic data. this simulates a recovery to pre-perturbation rates following a drop in seismicity by 80%. vertical gray and black arrows mark the start times of the drop in rate and recovery, respectively. other conventions as in figure located volcanic earthquakes we examined the catalog of located earthquakes in the mount wrangell area from august 2001 through march 2004. in figure 8 we show a time-depth plot of earthquakes at mount wrangell. notable changes in seismicity rate include the apparent increase associated with the introduction of earthworm acquisition system (johnson and others, 1995; dixon et. al., 2003); the decrease following the dfe, and two other subsequent apparent decreases (july – september 2003 and december 2003) associated with temporary telemetry problems and data disruptions. we used the computer program zmap (wiemer, 2001) which is suited for analysis of located earthquakes and includes routines to estimate changes in seismicity rates. the catalog includes 508 events with magnitude from -0.2 to 2.7. an automatic estimate of the magnitude of completeness indicates the catalog to be complete for magnitudes > 0.9, in agreement for the value reported by dixon et al. (2003). to avoid artifacts introduced by artificial rate changes, we select a subset of earthquakes between early march 2002 and late june 2003 (dotted box in figure 8), also to ensure the completeness of the samples used, we select earthquakes with magnitudes > 0.9 for subsequent analysis. the other decreases in seismicity rates observed in the catalog of are unrelated to large earthquakes along the denali fault because the most recent event of note was the mw 5.7 earthquake on 22 october, 1996 (ratchkovski et al. 2003), but rather these decreases are caused by other factors such as temporary low signal-to-noise ratio or stations outages. we note that the decrease in seismicity at wrangell following the dfe was unrelated to “artificial” causes because the same analyst located all the volcanic earthquakes and the stations were operational during the time of the decrease. figure 8. comparison depth-time plot for located earthquakes between august 5, 2001 and march 18, 2004 at mt. wrangell. red squares and blue circles with gray error bars mark lf and hf earthquakes, respectively. only earthquakes with magnitude > 0.9 are shown. arrows indicated the two largest hf earthquakes and the time of change from iaspei to earthworm acquisition systems. a dotted rectangle encloses the most homogeneous part of the catalog, between march 1, 2002 and june 30, 2003 that is used for all analysis of located earthquakes. other conventions as in figure 3. the catalog of located earthquakes at wrangell from august 5, 2001 to march 18, 2004, includes at least three decreases of noticeable duration. the only unexpected response of mt. wrangell volcano, alaska, to the shaking from a large regional earthquake: a puzzle for intermediate-term earthquakevolcanoes interactions. 41 decrease in seismicity at mt. wrangell following a large earthquake along the denali fault is the one that followed the dfe. the other two decreases are artificial (data disruptions caused by temporary telemetry problems) and are not preceded by large earthquakes in the area or by moderate-sized (with magnitude 5.0 or larger) aftershocks of the dfe (aeic, 2003). thus although the seismicity rates at mt. wrangell fluctuate, the coincidence of the rate decrease onset time with the dfe sequence suggest a causative relation and not the result of chance. to test the hypothesis that the decrease following the dfe, as estimated from located earthquakes, may not be different from random noise, we follow our previous procedure of randomly permuting our data sets of located earthquakes counts, plot the cumulative number of earthquakes versus time and compute the z values for the permuted series. in this case we repeat the permutation 100 times and show the results along with the original data in figure 9. we do the analysis for lf and vt earthquakes separately. for lf events the cumulative number versus time plots shows several changes in slope that represent decreases in seismicity. the decrease associated with the dfe sequence, however, represents the breakpoint with the most significance, because the peak in z value happens near the time of the dfe and nowhere else. we found that for lf events the cumulative number and z values curves are different from their respective permuted series. we note that data of located lf seismic events does not show a negative peak in the z values curve that we observed by creating a synthetic data set in which there was a recovery towards background rates. for hf earthquakes the ordered data is no different from the randomly permuted series. the catalogs of located lf and vt earthquakes show that the decrease in seismicity initiated after the dfe, this is because some earthquakes could be located during the immediate few days following the dfe. spatial mapping of the amount of decrease using βstatistic to examine spatially the decrease in seismicity in the mt. wrangell area we divide the are of interest in a grid with nodes separated 3 km (average of horizontal location errors) and compute the β-statistic using equation (2) for the 30 nearest earthquakes to each node. the two periods being compared to compute the β-statistic are: march 1, 2002 – november 3, 2002 and november 3, 2002 – june 30, 2003. note that the two time periods are of roughly equal duration (eight months each). the map of the β-statistic for the mt. wrangell area is shown in figure 10 and it shows a general decrease in the seismicity rates everywhere on the volcano. a small area located sw of the vent experienced relatively little to no change, but this is a region with sparse seismicity. the catalog used is rather small, but the seismicity is distributed so that figure 9. cumulative sum and z values versus time plot for earthquake counts based on located events between march 1, 2002 and june 30, 2003 (dotted rectangle in figure 8). left: plots for lf seismic events. right: plots for hf earthquakes. conventions as in figure 3. the radius of circles used to enclose the 30 nearest earthquakes to each node varied between < 2 km for areas around and west of the vent to 10 km around the edges of the region mapped. discussion we examined the rate of occurrence of volcanic earthquakes at mt. wrangell from before to after the dfe, at different time scales and using different data sources, to find the amount and significance of an observed decrease in volcanic seismicity. we think we can reliably measure this decrease because mt. john j. sanchez and stephen r. mcnutt 42 wrangell is an area with high rates of background seismicity, and such places are optimal for observation of seismicity declines (toda and stein, 2002). the decrease that we see in the seismicity at mt. wrangell figure 10. map of the mt. wrangell area showing the seismicity rate changes, as evaluated by the β-statistic. the two time periods being compared are: pre-dfe: march 1, 2002 to november 3, 2002; postdfe: november 3, 2002 to june 30, 2003. triangle: vent. crosses: seismograph stations (wanc is the station immediately west of the vent). gray circles and black squares mark the epicentral locations during the pre-dfe and post-dfe periods, respectively. gray lines: topographic contours. following the dfe is not the decay from an aftershock sequence of a large earthquake rupturing along the denali fault. the seismicity along the denali fault is not collocated with the seismicity at mt. wrangell (figure 2) and because we carefully checked all seismograms at station wanc on mt. wrangell, we could easily identify non-local earthquakes and discard them from our counts (see for example figure 3 of sánchez and mcnutt, submitted). we used rigorous statistical testing of the earthquake data and found the rates of local volcanic earthquakes (lf and hf together) to have decreased by 40% when the pseudohelicorder record data is used in the time scale of two months before to two months after the dfe (figure 3). in some areas around the volcano the seismicity may have decreased by as much as 80% when the located earthquake data from eight months before to eight months after the dfe is used (figure 9). although visual inspection of the data suggests that the rates of both lf and hf earthquakes decreased (figures 8 and 9), it is statistically difficult to prove for hf earthquakes because of their low rate of occurrence. from the depth-time plot and the analysis of cumulative number of located earthquakes we found that the decrease lasted for five months after which the seismicity seems to recover to pre-dfe levels. we investigated the question of why the z values curve does not seem to peak exactly on november 3, 2002 (the time of the dfe), but in most cases, before that time. by computing the z values on six synthetic data sets with varying amounts of rate decrease (figures 4 and 5), we found that the z values curve peaks exactly one data point before the drop in seismicity begins, which is expected because the rate before the change is constant exactly until that data point. an obvious conclusion of this test is that if the rates of seismicity are not perfectly constant within each period, we should expect to find variations in the behavior of the z values curve and the presence of multiple peaks of varying significance, which is what we see in our data (figure 3). we also found that the z values curve is highly sensitive to sharp rate changes, regardless of the amount of change. we also tested the effect, in the z values curve, of having a decrease in seismicity in two steps before the main perturbation and the effect of a seismicity rate that fully recovers sometime after the perturbation. the motivation for these two tests is the fact that our pseudohelicorder count data shows an apparent decrease in rates beginning sometime before the dfe (figure 3) and because in the catalog of located earthquakes it appears as if the seismicity recovers to pre-dfe levels (figures 8 and 9). by simulating a two-step decrease in seismicity preceding the main rate change we observed that the z values curve is double-peaked and that the first peak indicates a more significant change (figure 6). this apparently puzzling result comes about because the z values are a combination of the difference in means, the standard deviations of the two samples, and the number of data points in each sample (equation. 1). examination of figure 6 indicates that at the position of the first peak: m1-m2= 0.4221, s1 =0, s2=0.1361, and n2=77; whereas at the second peak m 1 m2=0.3798, s1=0.1262, n1=50, s2=0.07, and n2=51. thus it is apparent that the relative values of the peaks depend mostly on the difference in means and the number of data points on each segment. in the z values curve for our pseudohelicorder counts data we see some of the features identified in the curves for synthetic data. the pseudohelicorder counts data shows a z values curve with “regional” maxima that happens before the dfe but we also observe that this maximum is double-peaked, indicating that there were unexpected response of mt. wrangell volcano, alaska, to the shaking from a large regional earthquake: a puzzle for intermediate-term earthquakevolcanoes interactions. 43 some step-wise changes in earthquake rates at the volcano. from the analysis of the located events at wrangell, on the other hand, we learned that the seismicity may have dropped after a few days following the dfe. we see this in figures 8 and 9. by separating lf events from hf events we infer that both types of events may have decreased in rates, but because the low number of hf events available it makes it statistically difficult to prove. the results of tests with synthetic data suggest that unless the rates of seismicity are fairly uniform before and after the perturbation, it is difficult to use the z values to pinpoint exactly the time of change. this does not mean the plot is useless, because the z values curve does indicate the presence of one or several important changes in rates within the data set under scrutiny, such as in the case of the data presented here. we also simulated a time series in which an 80% drop in rate is followed by a recovery to the original rates (figure 7). in this case we observed that the z values curve has two peaks, one positive and one negative. we made this synthetic data set based on some features of our located earthquakes counts data in which a drop of about 80% in seismicity is followed by an apparent recovery after five months of lowerthan-normal rates (figure 8 and 9). because we do not see the negative peak in the z values curve of our real data, we infer that the seismicity at mt. wrangell did not recover fully to pre-dfe rates, at least until the end time period analyzed (june 30, 2003). although seismicity rates at volcanoes vary in the absence of evident external factors, the coincidence of the decrease with the time of the dfe suggests that the drop in seismicity at wrangell was not the result of chance. the total duration of the catalog of located earthquakes, including those with m < 0.9, is 957 days, or roughly 32 months; we see in this time period (figure 8) three noticeable drops in reported seismicity (one of them following the dfe, the other two are artificial) and the average duration of these three periods of low seismicity is ~ 3 months. thus there are about 11 3-month-long periods during which we may or may not find a decrease. in the context of a binomial distribution, we have n =32/3 = 11 trials and three successes (decreases). the probability p of success in a given binomial trial is p=3/11=0.27. this means that the probability of not finding an anomalously low-seismicity period is q = 1p = 0.73, thus about 73% of the time we find that the seismicity is at background rates. if we consider a more homogeneous catalog, with magnitudes > 0.9, and only during march 1, 2002march 18, 2004 (since data acquisition with earthworm began) the number of available months and 3-month trials get reduced to 25 and 8, respectively, and the probability of finding a decrease rises to p = 0.36, making it a more common fact. but this may be a biased test, because we know that the two latest decreases are man-made. in our view, a more balanced test would only include the period march 1, 2002 – june 30, 2003 (dotted box in figure 8). in this case there are 487 days or roughly 16 months, which means we have only n = 16/5≈3 trials. the probability p of finding a decrease is p =1/3 = 0.33, thus q = 0.69. this means that about 70% of the time we would find the seismicity to be at background rates. because the only large earthquakes along the denali fault during the time period march 1, 2002 – june 30, 2003, were associated with the dfe sequence and because the decrease in seismicity began rather close in time to the dfe, we infer that it was not a chance occurrence but that the cause was a perturbation of the volcanic plumbing system caused by either the dynamic shaking or the sudden drop in pressure around the volcano (sánchez and mcnutt, submitted). a spatial mapping of the β-statistic (figure 10) shows a general decrease in seismicity around mt. wrangell with some patches of larger change located south, north and west of the vent and an area of little to no change southwest of the vent. the mapping should we interpreted with caution because the locations of earthquakes at mt. wrangell are preliminary. conclusions a decrease in volcanic seismicity occurred at mt. wrangell following the denali fault earthquake sequence that began with the mw 6.7 nenana mt. earthquake and ended with the nucleation of the mw 7.9 denali earthquake on november 3, 2002. our data indicates that the volcanic seismicity dropped by as much as 80% during the five months that followed the dfe sequence. our test with synthetic data showed that unless the seismicity rates are fairly constant during the pre and post-perturbation periods, we could not use the z values curve to pinpoint exactly the time of change. the presence of a broad peak, however, indicates a significant change in the rates and that is what we see in the data. because of the time coincidence of the decrease in the seismicity rate with the time of the dfe sequence we think this is not a change occurrence, but an effect attributable to the dfe sequence. a comparison of our data of located earthquakes at mt. wrangell with a synthetic data set reveals that the seismicity at wrangell during the period november 4, 2002 – june 30, 2003 did not recover to the pre-dfe levels. the mapping of the β john j. sanchez and stephen r. mcnutt 44 statistic indicates that the decrease in seismicity occurred around the general mt. wrangell area, with some patches showing a stronger change than others. acknowledgements the authors wish to thank j. grasso, e. brodsky, k. felzer and c. benson for useful comments and discussion. special thanks to s. stihler and j. dixon for their hard work in the location of earthquakes at alaskan volcanoes. the alaska volcano observatory and the u.s. geological survey (usgs), as part of their volcano hazards program, and, supported this work by additional funds by the state of alaska. references aeic-alaska earthquake information center. (2003) “earthquakes in alaska, periodic publications, 12 issues/year”. 1991-2003. alvarado, g. e., fernandez, m. barquero,r. flores, h. and soto g. j. (1992) “aborted eruptions triggered by earthquakes?”. eos, 73, 194. dixon, j. p., power, j. a. and stihler, s. d. (2003) “seismic monitoring of alaska’s volcanoes: a comparison of event detection with iaspei and earthworm acquisition systems”. eos 84-46. gerlach, d. c., frey, f. a. h. moreno and l. lópezescobar. (1988) “recent volcanism in the puyehue-cordon caulle region, southern andes, chile (40.5os): petrogenesis of evolved lavas”. journal of petrology, 29, 2, 333-382. gvn-global volcanism network (2004): “web page of the global volcanism network, smithsonian institution”. http://www.volcano.si.edu. gomberg, j., p. bodin and p. a. reasenberg. (2003) “observing earthquakes triggered in the near field by dynamic deformations”. bulletin of the seismological society of america, 93, 1, 118-138. habermann, r. e. (1987) “man-made changes of seismicity rates”. bulletin of the seismological. society of america, 77, 141-159. hill, d., f. pollitz and c. newhall. (2002) “earthquake-volcano interactions”. physics today, 55, 11, 41-47. johnson, c. e., a. bittenbinder, b. bogaert, l. dietz and w. kohler. (1995) “earthworm: a flexible approach to seismic network processing, incorporated research institutions for seismology newsletter”. 4, 2, 1-4. mcnutt, s. r., and sánchez j. j. (2002) “intermediateterm declines in seismicity at two volcanoes in alaska following the mw 7.9 denali fault earthquake”. eos 83, 47. matthews, m. and p. reasenberg. (1988) “statistical methods for investigating quiescence and other temporal seismicity patterns”. pageoph, 126, 24, 357-372. minakami, t. (1974) “seismology of volcanoes in japan”. in: civetta, l., p. gasparini, g. luongo and a. rapolla (editors), physical volcanology, elsevier, amsterdam-newyork, pp. 1-27. phillips, eberhartd., haeussler, p. j. t. freymueller, frankel,a. d. rubin, c. craw, m. p. ratchkovski, n. anderson, a. g. carver, g. a. crone, a. j. dawson, t. e. fletcher, h. hansen, r. harp,e. l. harris, r. a. hill, d. hreinsdóttir, p. s. jibson, r. w. jones, l. kayen, m. r. keefer, d. k. larsen, c. f. moran, s. c. personius, s. plafker, f.g. sherrod, sieh, b. k. sitar, n. wallace w. k. (2003): “the 2002 denali fault earthquake, alaska: a large magnitude, slip-partitioned event”. science 300, 5622, 1113-1118. ratchkovski, n. a., r. a. hansen, j. c. stachnik, t. cox, o. fox, l. rao, e. clark, m. lafevers, s. estes, j. b. maccormack, and t. williams. (2003) “aftershock sequence of the mw 7.9 denali fault, alaska, earthquake of 3 november 2002 from regional seismic network data. seismological research letters. 74,6, 743-752. sánchez, j. j. and s. r. mcnutt. (2004) “intermediateterm declines in seismicity at mt. wrangell and mt. veniaminof volcanoes, alaska, following the november 3, 2002, mw 7.9 denali fault earthquake”. submitted to bulletin of the seismological society of america. tilling r. i., r. y. koyanagi, p. w. lipman, j.p. lockwood, j.g. moore and d. a. swanson (1976) “earthquake and related catastrophic events, island of hawaii, november 29, 1975: a preliminary report”. u.s. geol. surv. circular 740: 33 pp. toda, s. and r. stein (2002) “response of the san andreas fault to the 1983 coalinga-nuñez earthquakes: an application of interaction-based probabilities for parkfield”. journal of geophysical research, 107, b6, ese 6, 1-16. wiemer, s. (2001) “a software package to analyze seismicity: zmap”. seismological research letters. 72,2, 374-383. keywords: exploration; biota; late cretaceous; bauru group; southern goiás. palabras clave: biota; cretácico tardío; grupo bauru; sur de goiás. how to cite item candeiro, c. r. a., brusatte, s. l., souza, g. k., q., carvalho, a. a., maia, d. s., dias, t. c., vidal, l. s., gomes, m. m. n., & barros, l. m. (2020). late cretaceous bauru group biota from southern goiás state, brazil: history and fossil content. earth sciences research journal, 24(4), 387-396. doi: https://doi. org/10.15446/esrj.v24n4.82831 the southern goiás state region of central brazil has an extensive and informative record of fossil biota. particularly over the last five years, there has been a great increase in fossil finds, which has enabled a greater understanding of this region's fauna and flora during the late cretaceous. in this article, we provide an updated synthesis of the biota from the cretaceous of the southern goiás state, the record of plants, gastropods, turtles, crocodilians, titanosaurs, and theropod dinosaurs. abstract late cretaceous bauru group biota from southern goiás state, brazil: history and fossil content biota del grupo bauru del cretácico tardío del estado de goiás sur, brasil: historia y contenido fósil issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n4.82831 la región del sur del estado de goiás, en el centro de brasil, tiene un registro extenso e informativo de la biota fósil. particularmente, en los últimos cinco años ha habido un gran aumento en los hallazgos de fósiles, lo que ha permitido una mayor comprensión de la fauna y la flora de esta región durante el cretácico tardío. en este artículo, proporcionamos una síntesis actualizada de la biota del cretáceo del sur del estado de goiás, basada en el registro de plantas, gastropodos, tortugas, cocodrilianos, titanosaurios y dinosaurios terópodos. resumen record manuscript received: 13/10/2019 accepted for publication: 17/07/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 4 (december, 2020): 387-396 *carlos roberto dos anjos candeiro1,2, stephen louis brusatte3, glayce kelly de queiroz souza1, adelino adilson de carvalho4, débora santos maia1, tamires do carmo dias1,2, luciano da silva vidal1,5, musa maria nogueira gomes1, lucas marques barros1 1universidade federal de goiás, curso de geologia, laboratório de paleontologia e evolução, brasil 2universidade federal de goiás, programa de pós-graduação em biodiversidade animal, brasil 3university of edinburgh, school of geoscience, scotland, united kingdom 4universidade federal de goiás, museu antropológico, brasil 5universidade federal do rio de janeiro, instituto de geociências, brasil *corresponding author: candeiro@ufg.br g e o l o g y https://doi.org/10.15446/esrj.v24n4.82831 https://doi.org/10.15446/esrj.v24n4.82831 https://doi.org/10.15446/esrj.v24n4.82831 mailto:candeiro@ufg.br 388 carlos roberto dos anjos candeiro, stephen louis brusatte, glayce kelly de queiroz souza, adelino adilson de carvalho, débora santos maia, tamires do carmo dias, luciano da silva vidal, musa maria nogueira gomes, lucas marques barros introduction the southern goiás state region is located in central brazil (fig. 1), on the borders of the triângulo mineiro region (minas gerais state), northern mato grosso do sul state, and eastern mato grosso state. some important federal highways cross the region, connecting it to other midwest states and the northern, southeastern, and southern regions of brazil. the state of knowledge of late cretaceous fossils from the bauru group in southern goiás state is limited when compared to the triângulo mineiro region and western são paulo state. nevertheless, in recent years, goiás state has started to demonstrate its potential as a region that has a diversity of fossil taxa such as invertebrates, ichnofossils, and vertebrates. in the literature that addresses fossil findings in the southern region of the state of goiás, little information is available on the upper cretaceous fossil records of the bauru group, which is the only major unit of the cretaceous period that crops out in the south of goiás (candeiro et al., 2006). in western são paulo state and the triângulo mineiro region, where many rocks from the bauru group outcrop, fossil specimens have been discovered for over 125 years (e.g., ihering, 1911; pacheco, 1913; huene, 1931, 1933; candeiro et al., 2006; martinelli & teixeira, 2015; candeiro & figueirôa, 2017; see review in brusatte et al., 2017). however, these authors have not often been cited in later studies, which have led to misconceptions about the knowledge of the local biota. the gastropod taxon viviparus souzai mezzalira, 1974, has been discovered in the municipality of mateira (currently paranaiguara municipality), but studies describing other materials such as invertebrates and vertebrates from southern goiás state were only published after 2013. the importance of correctly describing the location of deposits and their fossils in their stratigraphic contexts has been previously mentioned by some authors (resende, 2014; candeiro et al., 2018). since the publishing of these works, the existence of two fossil-bearing units have been confirmed and the listing of fauna from southern goiás state has been significantly updated. previous studies (e.g., simone & mezzalira, 1994; senra, 2002; basilici & dal´bo, 2010; dal´bo & basilici, 2011; candeiro et al., 2018) report the presence of fossil records from this region, and therefore, it seems necessary to update the full inventory of fossils. weishampel et al. (2004) proposed a summary list of the late cretaceous fauna of southern goiás state and, recently, basilici & dal´bo (2010) and candeiro et al. (2018) added other taxa to this inventory. yet, these lists did not consider the problematic history of the fossil record and uncertainties in correlating between formations, which can lead to errors when reconstructing the local biota. the aim of the present study is to present the history of the paleontology of the bauru group (upper cretaceous) of southern goiás state, a stratigraphic summary, and the fossil record of the local biota. material and methods the analysis of the documents using the hermeneutic method (i.e. investigating and interpreting the information in the original sources). the hermeneutic method can lead to discoveries, new contributions and critiques, thus expanding our horizon of possibilities. the data on records of fossiliferous localities used here were mainly based on bibliographic sources and on the direct observation of specimens: (1) the invertebrate specimen was housed at the instituto geográfico e geológico do estado de são paulo (currently instituto geológico de são paulo); and (2) specimens deposited under the acronym “paleontologia da universidade federal de goiás coleção de vertebrados / paleo-ufg/v” in the laboratory of paleontology and evolution (laboratório de paleontologia e evolução) of the universidade federal de goiás. for the stratigraphic units of the bauru group we followed the arrangement proposed by soares et al. (1980) and the ages suggested by dias brito et al. (2001). figure 1. map showing southern goiás state municipality fossil-bearing 389late cretaceous bauru group biota from southern goiás state, brazil: history and fossil content soares et al. (1980) proposed an updated stratigraphic system that still applies to the bauru group, including, in ascending order from oldest to youngest: the adamantina, uberaba, and marília formations. the bauru group rocks crop out in the states of goiás (south), minas gerais (triângulo mineiro region), mato grosso do sul (east), and são paulo (west) (sensu fernandes & coimbra, 1996). the bauru group outcrops irregularly in southern goiás state (fig. 2). some of the best exposures are in several municipalities (e.g., caçu, cachoeira alta, itajá, itarumã, mineiros, paranaiguara, paraúna, quirinópolis, rio verde, and turvelândia), where the rocks of the adamantina formation are superimposed on the lower cretaceous basalts of the serra geral formation (serra geral group, paraná basin) (sensu cprm — serviço geológico do brasil, 2004). the strata of the adamantina formation are composed of reddish fluviallacustrine sandstones that were deposited during warm and humid climatic events (soares et al., 1980; barcelos, 1984; barcelos & suguio, 1987; fúlfaro et al., 1994; candeiro et al., 2018). the fossils found in this formation include mainly vertebrates (santos-pereira et al., 2016; ferreira et al., 2016; gomes, 2017; simbras et al., 2013; resende et al., 2014; candeiro et al., 2016, 2018). the adamantina formation was deposited during the turonian-santonian intervals based in carophyte and ostracod content (dias-brito et al., 2001). almeida & barbosa (1953) described the marília formation in the western region of são paulo state and, later, in southern goiás state (soares deposition of specimens most materials reported from the bauru group of southern goiás state are part of the collection of the laboratory of paleontology and evolution, universidade federal de goiás, since most fossils deposits in this region were discovered by the laboratory’s research team. all collection numbers are included in table 1. the invertebrate reported by mezzalira (1974) from the municipality of paranaiguara – holotype 750-i – is housed at the instituto geológico e geográfico do estado de são paulo, currently scientific collection sérgio mezzalira of the instituto geológico da secretaria do meio ambiente do estado de são paulo. furthermore, the materials of fossil invertebrates described by basilici and dal´bo (2010) and dal´bo and basilici (2011) are still in situ and, therefore, do not have collection numbers. geological setting the southern part of goiás state is geologically located in the bauru supersequence (bauru group), one of the four supersequences that form the paraná basin (milani & zalan, 1999; milani et al., 2007; see review in brusatte et al., 2017). in this region, there is a significant sedimentary record of the upper cretaceous that is composed of the adamantina and marília formations (sensu soares et al., 1980). table 1. biota from late cretaceous bauru group recorded in southern goiás state. taxa geological unit age materials selected authors itajá plant indet. marília formation late maastrichtian leaf basilici & dal´bo (2010) itarumã ichnofossil marília formation late maastrichtian bioturbation basilici & dal´bo (2010) paranaiguara viviparus souzai adamantina formation turonian-santonian shells mezzalira (1974) paraúna theropoda indet. adamantina formation turonian-santonian tooth candeiro et al. (2020) quirinópolis vertebrata indet. adamantina and marília formations turonian-santonian/late maastrichtian indeterminated bones santos-pereira et al. (2016) tetrapoda indet. adamantina formation turonian-santonian indeterminated bones santos-pereira et al. (2016) crocodyliformes indet. marília formation late maastrichtian incomplete last sacral vertebra candeiro et al. (2018) rio verde vertebrata indet. marília formation late maastrichtian indeterminated bones santos-pereira et al. (2016) tetrapoda indet. marília formation late maastrichtian indeterminated bones santos-pereira et al. (2016) podocnemididae indet. marília formation late maastrichtian fragmentary right epiplastron candeiro et al. (2018) notosuchia indet. marília formation late maastrichtian proximal half of left femur candeiro et al. (2018) ?titanosauriformes indet. marília formation late maastrichtian partial rib candeiro et al. (2018) titanosauria indet. marília formation late maastrichtian partial tooth candeiro et al. (2018) cf. titanosauria indet. marília formation late maastrichtian fragment of the diaphysis of a right radius candeiro et al. (2018) 390 carlos roberto dos anjos candeiro, stephen louis brusatte, glayce kelly de queiroz souza, adelino adilson de carvalho, débora santos maia, tamires do carmo dias, luciano da silva vidal, musa maria nogueira gomes, lucas marques barros et al., 1980; barcelos, 1984; barcelos & suguio, 1987; fúlfaro etal., 1994). this geological unit is formed by fine to medium sandstones interbedded with levels of conglomerates. the sandstones are cemented and contain calcium carbonate concretions (fúlfaro & barcelos, 1991). according to barcelos & suguio (1987), the marília formation was deposited in an anastomosed river system that produced fine to medium sandstones and lacustrine limestone deposits. in the southern region of goiás state, this geological unit is the most richly fossiliferous, with records of invertebrates/ichnofossils and vertebrates (soares et al., 1980; barcelos & suguio, 1987; fúlfaro et al., 1994; basilici & dal´bo, 2010; dal´bo & basilici, 2011). dias-brito et al. (2001) assign the marília formation to the late maastrichtian interval based on the fossil record of ostracods from western são paulo state and in the triângulo mineiro region. results research history of the upper cretaceous of goiás state and current status of investigations although the bauru group outcrops occur across many areas of southern goiás state, there is comparatively few published information on the fossils of the adamantina and marília formations in this region. these formations are known for being richly fossiliferous in other states such as são paulo (western region) and minas gerais (triângulo mineiro), with records of invertebrates/ ichnofossils and vertebrates. the first paleontological references of cretaceous fossils from the bauru group in the literature were made by mezzalira (1974) and bonaparte (1978) (sensu weishampel, 1993; weishampel et al., 2004). weishampel (1993) and weishampel et al. (2004) reported a probable occurrence of fossil dinosaur remains in southern goiás state that could be assigned to titanosaurs and theropods. these materials were supposedly identified by llewellyn ivor price, 1905-1980 (paleontologist of the serviço geológico e mineralógico do brasil, currently departamento nacional da produção mineral, rio de janeiro, brazil) (paulo, 2009). although price was possibly the first to acknowledge dinosaur remains in goiás, weishampel (1990) mentioned them such as the record of bones referred as cf. titanosaurus sp., antarctosaurus brasiliensis, and theropoda indet for the first time. according to these authors, the localities where these records were found were mentioned in bonaparte (1978). later, weishampel et al. (2004) suggested that this fossil record was actually represented by “antarctosaurus brasiliensis”, lithostrotia indet. (= cf. titanosaurus sp.), and theropoda indet., also based on bonaparte (1978). however, bonaparte (1978) didn´t mention any dinosaur remains from the goiás state. based on the work of weishampel (1993, 2004), many subsequent studies (e.g., santucci, 2002; candeiro et al., 2004, 2006) have reported the existence of remains of “antarctosaurus brasiliensis”, sauropods, and theropods from the bauru group of goiás state. the previously mentioned references that indicate the possible existence of titanosaurs and other dinosaur records in goiás state do not show any information about what kind of fossil specimens are present or their origins, including their stratigraphic provenance. hence, our understanding of these specimens is mostly speculative, and we cannot verify any details associated with them. so, the references to these specimens do indicate that dinosaur remains might have been found decades ago in the bauru group rocks of southern goiás state, because during the last 120 years the outcrops of this geological unit in other areas of central brazil have revealed a rich dinosaur record. the gastropod viviparus souzai mezzalira, 1974 have been described from the rocks of the adamantina formation in the municipality of mateira (currently paranaiguara). this record is important because it correlates the strata of the bauru group in goiás state with other areas where it crops out, in minas figure 2. map showing areal distribution of the adamantina and marília formations in southern goiás state (adapted from cprm, 2014). 391late cretaceous bauru group biota from southern goiás state, brazil: history and fossil content gerais and são paulo states (simone & mezzalira, 1994; senra, 2002). some authors (e.g., senra, 2002; ghilardi et al., 2011a,b) reported that v. souzai came from “madeira”, goiás state, but its correct locality is mateira, toponym of the paranaiguara municipality. dal´bo & basilici (2009) described the first records of indeterminate trace fossils and bioturbation in the rocks of the marília formation, in the municipalities of itajá and quirinópolis. later, basilici & dal´bo (2011) described ichnofossils and other fossils (likely invertebrates) found in the same strata. according to the authors, these materials came from the levels of the adamantina and marília formations, i.e., the strata would be indistinguishable in the field, as they are characterized by the same lithofacies, architectural elements, and sequential organization. in 2013, the research team of the laboratory of geology from the geography course, pontal campus of the universidade federal de uberlândia, initiated research on the late cretaceous vertebrates from southern goiás state. in 2015, this research started to be conducted by the laboratory of paleontology and evolution of the geology course, aparecida de goiânia campus of the universidade federal de goiás. the first results were presented in scientific abstracts and in a book chapter (e.g. simbras et al., 2013; resende et al., 2014), which described indeterminate remains of sauropod dinosaurs (simbras et al., 2013). then, a fossil vertebrate-bearing stratigraphic column of the main deposits from the quirinópolis and rio verde municipalities was presented (candeiro et al., 2018). currently, studies and excavations are ongoing in the above mentioned localities where, today, the main research of our universidade federal de goiás team is concentrated. our studies have yielded the first formally described reptile fossils from these sites. furthermore, other materials recovered in these areas are still in preparation (e.g. indeterminate vertebrate remains, turtles, crocodyliforms, sauropods, and theropods). the results have recently published in more than eleven scientific works (candeiro et al., 2015, 2018, 2020; santos-pereira et al., 2016; souza et al., 2016; gil et al., 2018; lima et al., 2018; monteiro et al., 2018; gil, 2019; nogueira & candeiro, 2019; maia et al., 2020). the effort of our team over the last few years has made new information available and clarified the late cretaceous biota of southern goiás state. biota summary in order to facilitate understanding, the list of fossil records is divided into four parts: invertebrates and vertebrates from the adamantina and marília formations (fig. 3 and 4). both table 1 and figure 3 show fossil content cited here and the provenance stratigraphic units. the three dinosaur taxa from goiás state mentioned by weishampel (1993, 2004) –cf. titanosaurus sp./lithostrotia indet., “antarctosaurus brasiliensis”, and theropoda indet.– were not considered in the present study due to the lack of information regarding their provenance, the actual material preserved, and the collection in which they are (or were) housed. in other words, we are unable to verify that these taxa have been discovered in goiás state. some specimens assigned to tetrapoda indet., titanosaurifomes indet., and theropoda indet. are included in the list of taxa here, although they have been described only in papers that were recently submitted or that are in prep, because they are important records of the bauru group in goiás state. the turonian-santonian fossils from the adamantina formation the following vertebrate materials have been reported from the adamantina formation: figure 3. stratigraphic distribution of late cretaceous bauru group biota from southern goiás state, brazil. 392 carlos roberto dos anjos candeiro, stephen louis brusatte, glayce kelly de queiroz souza, adelino adilson de carvalho, débora santos maia, tamires do carmo dias, luciano da silva vidal, musa maria nogueira gomes, lucas marques barros figure 4. main faunal taxa reported from adamantina and marília formations of southern goiás state: (a) viviparus souzaishell; (b) podocnemidoidae fragmentary right epiplastron view. main faunal taxa reported from the adamantina and marília formations of southern goiás state: (c) cf. notosuchia indet. proximal half of a femur in lateral view; (d) crocodyliformes indet. last sacral vertebra view; (e) cf. titanosauria indet. diaphysis of a right radius; (f) ?titanosauriformes indet. – partial rib; (g) titanosauria indet. – fragmented tooth; (h) theropoda indet. – tooth (a-g from candeiro et al., 2018 and h from candeiro et al., 2020). 393late cretaceous bauru group biota from southern goiás state, brazil: history and fossil content locality 1: paranaiguara municipality. horizon: adamantina formation. age: turonian-santonian. fossil content: viviparus souzai mezzalira, 1974 (fig. 2). references: mezzalira (1974), simone & mezzalira (1994), senra (2002), ghilardi & maranhão (2010), ghilardi et al. (2011a, b). locality 2: ‘‘quiri3”, quirinópolis municipality. horizon: adamantina formation (one fossiliferous level). age: turonian-santonian. fossil content: vertebrata indet., tetrapoda indet., crododylomorpha – crocodyliformes indet. references: santos-pereira et al. (2016), candeiro et al., (2018, 2019). locality 3: ‘‘paraúna 1”, paraúna municipality. horizon: adamantina formation (one fossiliferous level). age: turonian-santonian. fossil content: dinosauria – theropoda indet. reference: candeiro et al. (2019). the late maastrichtian fossils from the marília formation today, the marília formation is the most fossiliferous geological unit in goiás state and its records include invertebrates, ichnofossils, and vertebrates. locality 4: ‘‘itajá 1”, itajá municipality. horizon: marília formation (one fossiliferous level). age: late maastrichtian. fossil content: invertebrata ichnofossils. reference: basilici & dal´bo (2010). locality 5: ‘‘itarumã 1”, itarumã municipality. horizon: marília formation (one fossiliferous level). age: late maastrichtian. fossil content: invertebrata ichnofossils. reference: basilici & dal´bo (2010). locality 6: ‘‘rio ver. 1”, rio verde municipality. horizon: marília formation (two fossiliferous levels). age: late maastrichtian. fossil content: vertebrata indet., thetrapoda indet.; testudines: podocnemididae; crododylomorpha indet., crocodyliformes, cf.; notosuchia; dinosauria; sauropoda; ?titanosauriformes indet. references: santos-pereira et al. (2016), resende et al. (2014), candeiro et al. (2018). locality 7: ‘‘quiri 1”, quirinópolis municipality. horizon: marília formation (two fossiliferous levels). age: late maastrichtian. fossil content: thetrapoda indet.; dinosauria: sauropoda: cf. titanosauria indet. references: santos-pereira et al. (2016), candeiro et al. (2018). locality 8: ‘‘quiri 2”, quirinópolis municipality. horizon: marília formation (one fossiliferous level). age: late maastrichtian. fossil content: vertebrata indet., thetrapoda indet., mesoeucrocodylia – crocodiliformes indet. references: santos-pereira et al. (2016), candeiro et al. (2018). discussion the present study on the history of the paleontology, stratigraphy, and biota of the late cretaceous of southern goiás state, aims to understand the paleontology of this important but little studied area where the bauru group crops out. in the municipalities where the fossil-bearing rocks of the adamantina and marília formations crop out in goiás, the exposures are moderately diverse compared to the outcrops from western são paulo state and the triângulo mineiro region, which are known for being rich in plants, invertebrates, and vertebrates (see fig. 5). figure 5. paleoenvironmental reconstruction of some taxa of the late cretaceous bauru group from southern goiás state showing mollusk (a), turtle (b), mesoeucrocodylians(c), titanosaurian (d) and theropod (e). drawing made by l. vidal. 394 carlos roberto dos anjos candeiro, stephen louis brusatte, glayce kelly de queiroz souza, adelino adilson de carvalho, débora santos maia, tamires do carmo dias, luciano da silva vidal, musa maria nogueira gomes, lucas marques barros up until recently, mezzalira (1974) described the only fossil specimen from the late cretaceous of the bauru group in goiás state. there is no previously published work that indicates in which geological unit of the bauru group viviparus souzai was discovered. here, this species was identified from the levels of the adamantina formation in the municipality of paranaiguara. this recognition was made during a field trip in 2015, when it was possible to verify that this is the only formation that crops out in paranaiguara, an observation corroborated by the mapping from cprm (2004). this fossil gastropod is important because it correlates the strata of the adamantina formation in goiás state with those of the triângulo mineiro region in minas gerais state (senra & silva, 1999). the representative tetrapod assemblages from the late cretaceous of southern goiás state have been known, apparently, since the 1940s. nevertheless, it has been only over the past five years that some of the fossiliferous layers in this region have been studied in detail. this has resulted in a considerable number of identifications of members of the flora and fauna of the adamantina and marília formations (see table 1; fig. 3). thus far, turtles, crocodyliforms, and dinosaurs are the most abundant groups recovered in these units. the presence of large terrestrial taxa such as mesoeucrocodylians is noteworthy, because these are among the most well-known fossil carnivores of the late cretaceous of south america. the sandstones and mudstones of the adamantina formation in the municipality of quirinópolis occur irregularly in southern goiás state. recently, this geological unit was, for the first time, described in detail by ferreira et al. (2016). the authors described the slopes where a few isolated crocodyliform, dinosaur and theropod remains have been found (candeiro et al., 2018). the late maastrichtian outcrops of the marília formation are the youngest, the most prolific, and most promising fossil-bearing sites from the late cretaceous of central brazil. a significant quantity of invertebrates and vertebrates have been found in these outcrops, although our work has only recently started. in southern goiás state, the marília formation is mainly formed by white and yellow sandstones with fine sandstone intervals and intraformational conglomerates (ferreira et al., 2016; candeiro et al., 2018; nogueira & candeiro, 2019; souza & candeiro, 2019). the localities 6 and 7 (quirinópolis, rio verde) of this geological unit are the fossil sites that contain the best dinosaurs. other associated flora and fauna are found in the northeast portion of the bauru group in goiás state. some of the vertebrate taxa reported from southern goiás state were found associated with other vertebrates that are known from other localities that are important for stratigraphic correlation: western são paulo state (presidente prudente, araçatuba, adamantina and marília formations), triângulo mineiro region (adamantina, uberaba, and marília formations), and even in upper cretaceous rocks of the neuquén group (candeleros, cerro lisandro, portezuelo, huincul and bajo de la carpa formations) and malargüe group (allen and loncoche formations), both in argentina (e.g., leanza et al., 2004; candeiro & rich, 2010; garrido, 2010). the materials of podocnemididae, notosuchia, titanosauria, and theropoda, found in the previously mentioned geological units, corroborate this connection with outrcrops of southern goiás state. turtles, crocodyliforms, and dinosaurs (especially titanosaurids) have also been found in other areas of the bauru group in minas gerais and são paulo states. regarding the titanosaurian records from the municipalities of quirinópolis and rio verde, these sauropods are widely known in the upper cretaceous geological units of argentina. the sphenodontids, ornithischians and mammals records from the late cretaceous of argentina is remarkable (e.g., leanza et al., 2004; de la fuente et al., 2007). some of these groups are also found in the bauru group of são paulo state but none have yet been found in the adamantina and marília formations of southern goiás state. the absence of these tetrapods in these areas may be due to the lack of extensive or less systematic prospecting being that the research history in the adamantina and marília formations in goiás state is still recent, but shows a great future potential. conclusions there are reports that cretaceous dinosaurs were found in southern goiás state since the 1940s, although these earlier discoveries cannot be verified. our recent work shows that there is indeed a diversity of fossils in southern goiás state, from two geological units: the adamantina (turonian-santonian) and marília (late maastrichtian) formations. the fossil fauna of southern goiás state is similar to that from similar aged deposits in the triângulo mineiro region, western são paulo state, and even argentina. southern goiás state is a promising area for future finds of cretaceous fossils, which can be crucialfor the evaluation of biodiversity and biogeographical patterns of tetrapods like dinosaurs and crocodylomorphs during the final part of the cretaceous, before the end-cretaceous mass extinction. in short, only a few seasons of fieldwork between the years of 2013 to 2018 in goiás state have resulted in a significant addition of new vertebrate records from the upper cretaceous of the bauru group, hinting that more discoveries are waiting to be made. our new finds increase the comparative data necessary to understand the faunal composition of central brazil and vertebrate distribution in this area during the cretaceous. acknowledgements crac was financially supported by conselho nacional de desenvolvimento científico e tecnológico (cnpq)/bolsista de produtividade em pesquisa. slb’s collaborative work with crac was funded by a grant from the fundação de amparo a pesquisa de goiás (fapeg) and newton fund to be researching in southern goiás state. references almeida, f. f. m. & barbosa, o. (1953). geologia das quadrículas piracicaba e rio claro, estado de são paulo. rio de janeiro (rj): sergraf do ibge. arid, f. m. & vizzoto, l. d. (1971). antarctosaurus brasiliensis, um novo saurópode do cretáceo superior do sul do brasil. in: proceedings of the xxv congresso brasileiro de geologia; set 1971; são paulo (sp): sociedade brasileira de geologia. p. 297-305. barcelos, j. h., & suguio, k. (1987). correlação e extensão das unidades litoestratigráficas do grupo bauru, definidas em território paulista e nos estados de minas gerais, goiás, mato grosso do sul e paraná. proceedings of the vi simpósio regional de geologia. agosto, 1987; rio claro (sp): universidade estadual de são paulo. p. 313-321. barcelos, j. h. (1984). reconstrução paleogeográfica da sedimentação do grupo bauru baseada na sua redefinição estratigráfica parcial em território paulista e no estudo preliminar fora do estado de são paulo. doctor’s thesis. rio claro: universidade estadual paulista. basilici, g. & dal’ bo, p. f. f. (2010). anatomy and controlling factors of a late cretaceous aeolian sand sheet depositional system the marília and the adamantina formations, nw of the bauru basin, brazil. sedimentary geology, 226, 71-93. basilici, g., dal’ bo, p. f. f. & ladeira, f. s. b. (2009). climate induced sediment-palaeosol cycles in a late cretaceous dry aeolian sand sheet marília formation (north-west bauru basin, brazil). sedimentology, 56, 1876-1904. bonaparte, j. f. (1978). el mesozoico de américa del sur y sus tetrápodos. opera lilloana, 26, 1-596. brusatte, s. l., candeiro, c. r. a. & simbras, f. m. (2017). the last dinosaurs of brazil: the bauru group and its implications for the end-cretaceous mass extinction. anais da academia brasileira de ciências, 89, 14651485. candeiro, c. r. a. & figueirôa, s.f.m. (2017). early twentieth-century paleontological research of freidrich von huene contributions to the knowledge of late cretaceous vertebrates of central brazil. historical biology, 29, 1-10. candeiro, c. r. a. & rich, t. (2010). an overview on the late cretaceous biota of the western são paulo state, brazil, bauru grup. journal of south american earth science, 29, 346-353. candeiro, c. r. a., bergqvist, l. p., oliveira, e. c., gomes, s. a. & carvalho, a. a. (2011). o impacto dos estudos do grupo de pesquisas do cnpq nas regiões do pontal do triângulo mineiro. in: candeiro, c. r. a., gomes, s. a. (eds). tetrápodes do cretáceo do pontal do triângulo mineiro e sul de goiás: atividades e reuniões 2009-2010 [tradução]. 1st ed. vol. 1. rio de janeiro (rj): letra capital, p. 109-115. http://lattes.cnpq.br/5419992057415168 http://lattes.cnpq.br/3162440187077806 http://lattes.cnpq.br/2584676055987613 http://lattes.cnpq.br/2584676055987613 http://lattes.cnpq.br/2584676055987613 http://lattes.cnpq.br/5940881576341018 http://lattes.cnpq.br/5763906339994881 http://lattes.cnpq.br/0359303667575014 http://lattes.cnpq.br/0359303667575014 http://lattes.cnpq.br/2584676055987613 395late cretaceous bauru group biota from southern goiás state, brazil: history and fossil content candeiro, c. r. a., brusatte, s. l. & simbras, f. m. (2018). occurrence of tetrapod fossils in goiás state: a rediscovery in central brazil. in: figueiredo, a. e. q., oliveira, p. v. (eds). proceedings of the xi simpósio brasileiro de paleontologia de vertebrados; jul 16-20; teresina (pi): boletim de resumos do xi simpósio brasileiro de paleontologia de vertebrados. p. 33. candeiro, c. r. a., brusatte, s. l., simbras, f. m., costa, r. g. f., ulian, m., ferreira, b. m., lima, c. v., pereira, p. v. & morais, e. m. (2020). first record of a theropod (dinosauria, theropoda) from the late cretaceous of the bauru group, southern goiás state, brazil. research, society and development, 9, e563985261. candeiro, c. r. a., marinho, t. s. & oliveira, e. c. (2004). distribuição geográfica dos dinossauros da bacia baurú (cretácio superior). sociedade & natureza, 16, 33-55. candeiro, c. r. a., martinelli, a. g., avilla, l. s. & rich, t. (2006). tetrapods from the upper cretaceous (turonian-maastrichtian) bauru group of brazil a reappraisal. cretaceous research, 27, 923-946. candeiro, c. r. a., martins, d. c., carvalho, a. a., simbras, f. m., santoro, a. c. m., barreto, r. f. b. n., moreira, s. a. g., peyerl, d., santos, d. c. b. & vidal, l. s. (2016). exposição: dinossauros do brasil central, um primeiro olhar sobre os dinossauros dos estados de goiás e mato grosso. in: grillo, o. n., romano, p. s. r., oliveira, g. r. (eds). sbp 2016. proceedings of the x simpósio brasileiro de paleontologia de vertebrados; out 3–7; rio de janeiro. boletim de resumos/ paleontologia em destaque: boletim informativo da sociedade brasileira de paleontologia. p.55. candeiro, c. r. a. & gomes, s. a. (2011). tetrápodes do cretáceo do pontal do triângulo mineiro e sul de goiás: atividades e reuniões 2009-2010. rio de janeiro (rj): letra capital. cprm companhia de pesquisa de recursos minerais (br). (2004). carta geológica do brasil ao milionésimo, folha se22, goiânia. brasília (df): secretaria de minas e metalurgia e ministério de minas e energia. dal’ bo, p. f. f. & basilici, g. (2011). interpretação paleoambiental da formação marília na porção noroeste da bacia bauru relações entre sedimentação e paleopedogênese em um antigo lençol de areia eólica. geociências, 30, 509-528. dias-brito, d., musacchio, e. a., castro, j. c., maranhão, m. s. a. s., suárez, j. m. & rodrigues, r. (2001). grupo bauru uma unidade continental do cretáceo do brasil e concepções baseadas em dados micropaleontológicos, isotópicos e estratigráficos. revue de paléobiologie, 20(1), 245-304. faria, c. c. j. (2012). levantamento da ocorrência de dinossauros (dinosauria: saurischia) no brasil e perspectivas para o estado de goiás. undergraduation work. curso de ciências biológicas, anápolis (go): universidade estadual de goiás. fernandes, l. a. & coimbra, a. m. (1996). a bacia bauru (cretáceo superior, brasil). anais da academia brasileira de ciências, 8(2), 195-205. ferreira, j. b. s., cavalcanti, r., pereira, c. s., carelli, d., souza, a. l., simbras, f. m., brusatte, s. l. & candeiro, c. r. a. (2016). a new vertebrate locality from southern goiás state, brasil. in: grillo, o. n., romano, p. s. r., oliveira, g. r. (eds). sbp 2016. proceedings of the x simpósio brasileiro de paleontologia de vertebrados; out 3-7; rio de janeiro. p. 164-164. fuente, m. s., salgado, l., albino, a., báez, a. m., bonaparte, j. f., calvo, j. o., chiappe, l. m., codorniú, l. s., coria, r. a., gasparini, z., gonzález-riga, b., novas, f. e. & pol, d. (2007). ameghiniana. publicación especial, 11, 137-153. fúlfaro, v. j. & barcelos, j. h. (1991). grupo bauru no triângulo mineiro: uma nova visão litoestratigráfica. proceedings of the ii simpósio de geologia do sudeste, boletim de resumos; august, 1991. rio claro. p. 59-66. fúlfaro, v. j., perinotto, j. a. j. & barcellos, j. h. (1994). a margem goiana do grupo bauru: implicações na litoestratigrafia e paleogeografia. in: dias-brito d. (ed.) proceedings of the iii simpósio do cretáceo do brasil; jul 24-31; rio claro: boletim de resumos. p. 81-84. garrido, a. (2010). estratigrafía del grupo neuquén, cretácico superior de la cuenca neuquina (argentina): nueva propuesta de ordenamiento litoestratigráfico. revista del museo argentino de ciencias naturales, 12(2), 121-177. ghilardi, r. p. & maranhão, m. s. a. s. (2010). anatomia interna de alguns moluscos bivalves fósseis do grupo bauru (cretáceo superior, bacia bauru), brasil implicações paleoecológicas. terr@ plural, 4(1), 113120. ghilardi, r. p., d´agosta, f. c. p., alves, k. & campos, a. c. a. (2011). tafonomia de moluscos fósseis do grupo bauru (cretáceo superior, bacia bauru), na região do município de monte alto, são paulo, brasil. boletim do museu paraense emílio goeldi. ciências naturais, 6, 67-72. ghilardi, r. p., rodrigues, s. c., simone, l. r. l., carbonaro, f. a. & nava, w. r. (2011). moluscos fósseis do grupo bauru. in: carvalho, i. s. (ed). paleontologia: cenários de vida [tradução]. 1st ed. vol. 1. bauru (sp): interciência; p. 197-208. gil, l. m. (2019). novos registros de titanosauria cretácicos do centro-oeste brasileiro (goiás e mato grosso). universidade federal de goiás, master dissertation. gil, l. m., candeiro, c. r. a. & brusatte, s. l. (2018). identificação e descrição de novosregistros de titanosauria do centro-oeste brasileiro (goiás e mato grosso). ii encontro de biodiversidade animal ufg, 2018, goiânia. boletim do ii encontro de biodiversidade animal/ufg, v. 1. p. 72-77. gomes, m. m. n. (2017). estratigrafia dos afloramentos da formação marília (grupo bauru, cretáceo superior) portadores de fósseis do município de quirinópolis/go. in: candeiro, c. r. a. (ed). proceedings of the ii seminário do laboratório de paleontologia e evolução/ufg boletim de resumos; nov 23-25; goiânia (go). p. 1-3. gomes, m. m. n., bampi, h., dias, t. c., souza, a. l., martins, d. c., carvalho, a. a. & candeiro, c. r. a. (2016). divulgação científica da paleontologia através da exposição itinerante ‘dinossauros do brasil central’. in: riff, a. c. s. (ed). proceedings of the caderno de resumos xiii reunião regional da sociedade brasileira de paleontologia em minas gerais; dez 2-3; uberlândia (mg) p. 1-34. huene, f. (1929). los saurísquios y ornitísquios del cretácico argentino. anales museo de la plata, 2(3), 1-196. huene, f. (1931). verschiedene mesozoische wirbeltierreste aus südamerika. neues jahrbuch für mineralogie, geologie und paläontologie, abteilung a, 66, 181-198. huene, f. (1933). ein versuchzur stammesgeschichte der krokodile. central blatt für mineralogie, geologie und paläontologie, abteilung b, 11, 577-585. ihering, f. (1911). fósseis de são josé do rio preto. revista do museu paulista, 8, 141-146. leanza, h. a., apesteguía, s., novas, f. e. & de la fuente, m. s. (2004). cretaceous terrestrial beds from the neuquén basin (argentina) and their tetrapod assemblages. cretaceous research, 25, 61-87. lima, m. v., monteiro, e. b., vidal, l. s., santos-pereira, c. & candeiro, c. r. a. (2018). importance of the collection of the laboratório de paleontologia e evolução of the ufg to the paleontological research in the goiás state. cadernos de geociências, 14, 73-77. maia, d. s., gil, l. m. & candeiro, c. r. a. (2020). preparação de restos de vertebrados do neocretáceo da formação marília (grupo bauru) do município de rio verde, goiás. revista sapiência, 9(3), 14-23. martinelli, a. g. & teixeira, v. p. a. (2015). the late cretaceous vertebrate record from the bauru group at the triângulo mineiro, southeastern brazil. boletin geologico y minero, 62, 129-158. mezzalira, s. (1974). contribuição ao conhecimento da estratigrafia e paleontologia do arenito bauru. boletim do instituto geográfico e geológico, 51, 1-163. milani, e. j. & zalan, p. v. (1999). an outline on the geology and petroleum systems of the paleozoic interior basins of south america. episodes, 22(3), 199-205. http://lattes.cnpq.br/2584676055987613 http://lattes.cnpq.br/7540333002746228 http://lattes.cnpq.br/5383557718345145 http://lattes.cnpq.br/2584676055987613 http://lattes.cnpq.br/3162440187077806 http://lattes.cnpq.br/3868502906886933 http://lattes.cnpq.br/7845280140203354 http://lattes.cnpq.br/3868502906886933 http://lattes.cnpq.br/7832485947283639 http://lattes.cnpq.br/8760110878574938 396 carlos roberto dos anjos candeiro, stephen louis brusatte, glayce kelly de queiroz souza, adelino adilson de carvalho, débora santos maia, tamires do carmo dias, luciano da silva vidal, musa maria nogueira gomes, lucas marques barros milani, e. j., melo, j. h. g., souza, p. a., fernandes, l. a. & frança, a. b. (2007). bacia do paraná. boletim de geociências da petrobras, 15(2), 265-287. monteiro, l., nogueira, m., ulian, m. & candeiro, c. r. a. (2018). curatorship of the vertebrate collection housed at laboratory of paleontology and evolution/universidade federal de goiás. xi simpósiobrasileiro de paleontologia de vertebrados, 2018, teresina. boletim de resumos do xi simpósiobrasileiro de paleontologia de vertebrados, p. 72-72. nogueira-gomes, m. m. & candeiro, c. r. a. (2019). estratigrafia dos afloramentos da formação marília (grupo bauru, cretáceo superior) portadores de fósseis do município de quirinópolis/go. revista interface (porto nacional), 17, 6-15. pacheco, j. a. (1913). notas sobre a geologia do vale do rio grande a partir da foz do rio pardo até sua confluência com o rio paranaíba. comissão geográfica geológica, 1, 33-38. pedro, o. p. (2009). vertebrados fósseis do estado de goiás, com ênfase em sua fauna de amniotas, compreendida entre o período permiano e a época pleistoceno. master’s thesis. rio claro: universidade estadual paulista. paulo, p. o. (2009). vertebrados fósseis do estado de goiás. master’s thesis. rio claro (sp): universidade estadual paulista. resende, i. l. m., souza, e. a. & candeiro, c. r. a. (2014). microrregião da confusão do rio preto, quirinópolis, goiás. in: urzedo, m. f. a. (ed). cultura e desenvolvimento regional mãos e olhares diferentes iii 1832-2014. 1ed. quirinópolis (go): kelps, p. 303-321. santos-pereira, c., figueiredo-lima, c. & candeiro, c. r. a. (2016). resultados preliminares del análisis de sedimentos del grupo bauru del sur de la provincia de goiás en la búsqueda de microvertebrados. in: novas, f. e. (ed). proceedings of the xxx jornadas argentinas de paleontología de vertebrados libro de resúmenes; may 17-20; buenos aires. p. 111. santucci, r. (2002). revisão dos titanosauridae (saurischia, sauropoda) do cretáceo superior continental da bacia bauru. master’s thesis. rio claro (sp): universidade estadual paulista. senra, m. c. e. (2002). malacofauna dulçaquicola do cretáceo do brasil. doctoral dissertation. rio de janeiro (rj): universidade federal do rio de janeiro. senra, m. c. e. & silva e silva, l. (1999). moluscos dulçaqüícolas e microfósseis vegetais associados da formação marilia, bacia bauru (cretáceo superior), minas gerais, brasil. in: dias-brito, d., castro, j. c., rohn, r., saad, a. r. (eds). proceedings of the v boletim do simpósio sobre o cretáceo do brasil; ago-set 29-02; serra negra (sp) p. 497-500. simbras, f. m., souza, l. c. a., machado, r., alves, m. c., lopes, w. h., santos, j. c. v. & muniz, f. p. (2013). bones out the cerrado: new dinosaur exploratory frontier in goiás state. in: candeiro, c. r. a., machado, e., alves, y. m. (eds). proceedings of the 1º simpósio brasileiro de dinossauros; abr 21-24; ituiutaba (mg). p. 75. simone, l. r. l. & mezzalira, s. (1994). fossil molluscs of brazil. boletim do instituto geológico, 11, 1-202. soares, p. c., landim, p. m. b., fúlfaro, v. j. & sobreiro, a. f. (1980). ensaio de caracterização estratigráfica do cretáceo no estado de são paulo grupo bauru. revista brasileira de geociências, 10, 177-185. souza, a. l., melo, n. t. p., bampi, h., carvalho, a. a., souza, f. i. & candeiro, c. r. a. (2016). a exposição ‘dinossauros do brasil central’: o público infantil e seu olhar como visitantes. proceedings of the 27º semana do instituto de ciências biológicas/universidade federal de goiás; nov 07-11; goiânia (go): p. 1-4. souza, a. l., soares, d. d. m., oliveira, j. f., & silva, l. f. (2015). distribuição dos registros fósseis de dinossauros (dinosauria: saurischia) no brasil central (estados de goiás e mato grosso). in: souza, a. l. jr. (ed). proceedings of the boletim de resumos i seminário de pesquisa do laboratório de paleontologia e evolução; nov 12-14; aparecida de goiânia (go). p. 37-38. souza, j. b. f., cavalcanti, r., pereira, c. s., carelli, d., souza, a. l., simbras, f. m., brusatte, s. l. & candeiro, c. r. a. (2016). new fossil vertebrate localities from southern goiás state, brazil. in: grillo on, romano psr, oliveira gr, editors proceedings of te x simposio brasileiro de paleontologia de vertebrados boletim de resumos/paleontologia em destaque: boletim informativo da sociedade brasileira de paleontologia; out 3–7; rio de janeiro. p.1-2. upchurch, p., barrett, p. m. & dodson, p. (2004). sauropoda. in: weishampel, d. b., dodson, p., osmólska, h. (eds). the dinosauria. 2nd ed. berkeley (ca): university of california press. p. 259-324. weishampel, d. b., dodson, p. & osmólska, h. (1990). the dinosauria. berkeley: university of california press. weishampel, d. b., dodson, p., osmólska, h. & richard, p. (2004). the dinosauria, 2nd ed. berkeley: university of california press. http://lattes.cnpq.br/1133817014664784 http://lattes.cnpq.br/1133817014664784 http://lattes.cnpq.br/1133817014664784 keywords: heavy metal pollution; soil; ecological compensation; internet of things. palabras clave: contaminación por metales pesados; suelo; compensación ecológica; internet de las cosas. how to cite item rojas barbosa, s., molano mendoza, j. shanshan, j., & yanqing, z. (2020). ecological compensation method for soil polluted by heavy metals based on internet of things. earth sciences research journal, 24(2), 153-161. doi: https://doi.org/10.15446/esrj. v24n2.87441 the traditional ecological compensation method for soil pollution by heavy metals is complicated. it consumes a lot of human and material resources and cannot quickly meet real-life needs. this study takes an area where heavy metals pollute the soil. increasing soil retention in this area is the target for ecological compensation, and the method is based on the internet of things. users' willingness to participate in the land heavy metal pollution control project was studied, and the pollution degree was calculated. the area was linked to the internet tools according to the agreement, and the information exchange was carried out through information-sensing equipment of radio frequency identification. this process provided the basis for the ecological compensation standard of soil heavy metal pollution area. according to the content, changing trend, and characteristics of pollutants in the study area, the best forest species were selected for this area. the soil retention was increased to the maximum extent to realize the ecological compensation to the area polluted by heavy metals. the final results show that the ecological compensation method for soil contaminated with heavy metals based on the internet of things is cost-effective and has excellent feasibility, which can be the essential means for sustainable development of ecological construction. abstract ecological compensation method for soil polluted by heavy metals based on internet of things método de compensación ecológica para suelos contaminados por metales pesados con base en el internet de las cosas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n2.87441 el método tradicional de compensación ecológica para el suelo contaminado por metales pesados es complicado, consume mucha mano de obra y recursos materiales, y no puede satisfacer las necesidades de la vida real. para este trabajo se tomó como ejemplo un área donde el suelo está contaminado por metales pesados. el aumento de la retención del suelo en esta área es el objetivo específico de la compensación ecológica, y este aumento se busca alcanzar a través de métodos relacionados con el internet de las cosas. se estudió la disposición de los usuarios a participar en el proyecto de control de la contaminación por metales pesados y se calculó el grado de contaminación. a través del equipo de detección de información de identificación por radiofrecuencia, el área de contaminación de metales pesados del suelo se conectó a herramientas de internet según el acuerdo, y el intercambio de información se llevó a cabo, a fin de proporcionar la base para el estándar de compensación ecológica del área de contaminación de metales pesados del suelo. según el contenido, la tendencia cambiante y las características de los contaminantes en el área de estudio, se seleccionaron las mejores especies forestales para esta área, y la retención del suelo se incrementó al máximo para obtener la compensación ecológica al área contaminada por metales pesados. los resultados finales muestran que el método de compensación ecológica para el suelo contaminado con metales pesados basado en el internet de las cosas es rentable y tiene una gran viabilidad, que puede ser el medio importante para el desarrollo sostenible de la construcción ecológica. resumen record manuscript received: 13/11/2019 accepted for publication: 27/03/2020 earth sciences research journal earth sci. res. j. vol. 24, no. 2 (june, 2020): 153-161 jia shanshan1*, zhou yanqing2 1school of inner mongolia vocational college of chemical engineering, department of computer and information engineering, hohhot, 010010, china; 2inner mongolia agricultural university, college of computer and information engineering, hohhot, 010018, china * corresponding author: jiashanzi@163.com; hellozhouyanqing@126.com so il s https://doi.org/10.15446/esrj.v24n2.87441 https://doi.org/10.15446/esrj.v24n2.87441 https://doi.org/10.15446/esrj.v24n2.87441 mailto:jiashanzi@163.com mailto:hellozhouyanqing@126.com 154 jia shanshan, zhou yanqing introduction the main sources of heavy metal pollution (gasiorek et al., 2017) include industrial pollution, metal mining, sewage irrigation, solid waste disposal, etc., and have received extensive attention from researchers in recent years. the accumulation of pollutants (diepens & koelmans, 2018) in the soil often presents a high degree of spatial variability due to the enhancement of human activities, which makes the soil heavy metal pollution more complicated, and often forms one or more trends in space. since heavy metals are not easily mobile and dissolved, they accumulate after entering the organism and have an adverse effect on the organism. for example, when the low concentration of hg is present in the soil, it can promote in the early stage of wheat germination. however, as the time increases, the respiration will decrease and eventually manifest as inhibition. it can be seen that the latent time of heavy metal pollution is long. heavy metals that are not easily degraded have adverse effects on the quality and yield of crops, and can even affect human health through the food chain. the construction of ecological compensation mechanism is an important means to improve the quality of the ecological environment and coordinate the contradiction between environmental protection and economic development. many experts and scholars at home and abroad have done a lot of research on this and achieved good results. various methods of ecological compensation standards and their application cases are analyzed to determine the ecological compensation method based on the internet of things (kokkonis et al., 2017), which provides the scientific basis for the determination of ecological compensation standards and the construction of mechanisms. (1) in order to solve the ecological environment loss caused by industrial pollution and determine reasonable ecological compensation standards, energy theory is used to calculate energy loss. taking the measured ecological environment loss as the ecological compensation standard, the research draws the isw pollution tax to balance the pollution control effect (liu et al., 2018). (2) the concept of theoretical deficit. based on the theoretical deficit, the calculation method of compensation cost between contaminated areas is improved. at the same time, compensation costs are predicted by combining population size, wealth and technology. finally, the compensation indicators for each contaminated area are determined (yang et al., 2019). (3) first-hand information obtained through public statistics and field surveys as the source of data. conditional valuation method (cvm) is used to estimate the source data and contaminated receptor data associated with the compensation area. according to the survey results, the compensation standard is determined, and the compensation amount is estimated, then the ecological compensation is finally realized (shu, 2018). (4) an area-based ecological compensation method is proposed. based on the traditional ellipsoidal area and 103 regions as reference, the biomass and ecological compensation are determined by surface area method. in this way, we will balance the interests of different regions and make reference for ecological sustainable development (he et al., 2018). due to the complexity of the process, it will consume a lot of human and material resources. therefore, an ecological compensation method based on the area where the soil is contaminated by heavy metals is proposed. by analyzing the living and production of the contaminated area, the cost of pollution compensation is calculated to establish a sustainable ecological compensation plan. the method has short calculation time, easy operation and high accuracy, and has excellent application promotion value. materials and methods the overall structure of the internet of things first, the concept, necessity, and positioning of the internet of things operating system are briefly analyzed. the basic functions of the internet of things operating system (gao et al., 2019) are similar to those of the mobile operating system in the field of mobile internet. first of all, the role of the mobile operating system is to achieve the separation of terminal hardware and software. the developer of any application does not have to consider the physical hardware configuration of the terminal. it can be run on all mobile operating system-based terminals simply by writing an application based on the programming interface provided by the phone. the function of the hardware is limited. if the software and hardware are tightly bundled and not separated, the functions of the entire system are limited. however, by stripping out the hardware functions, the functionality of the entire system can be improved by software changes. secondly, for the new internet of things, to achieve a benign development, the effect of software and hardware separation must be achieved. the internet of things has limited hardware capabilities. by writing the corresponding application by the specific developer, you can extend the capabilities of the hardware and make full use of the hardware. in addition, the internet of things needs to achieve another separation: the separation of hardware drivers from the operating system kernel. when there is the problem with the hardware of the internet of things used, only the problem hardware needs to be replaced, and the operating system automatically configures the corresponding driver. the installation process is transparent to the programs running on the operating system. the following diagram illustrates the internet of things ecosystem built on these two separations (application software (millard et al., 2018) is separated from the hardware; device drivers are separated from the operating system kernel),the work flow diagram of the internet of things is shown in figure 1: figure 1. work flow chart of the internet of things in order to obtain accurate spatial distribution information of pollutants in the soil by using the internet of things, the investigation of heavy metal pollution in the soil usually includes multiple stages such as preliminary investigation and detailed investigation of pollution. the main purpose of the preliminary investigation is to identify major pollutants and potentially contaminated areas of the soil (monaco et al., 2017), usually with a small sample size. the detailed investigation of pollution is based on the preliminary investigation, and the sample points in the potential pollution area are increased to determine the extent of the pollution area and the degree of pollution. the errors in the soil pollution survey program mainly include the underestimation of the contaminated area and the overestimation of the clean area. among them, the former will lead to the pollution control of the pollution area is not effectively 155ecological compensation method for soil polluted by heavy metals based on internet of things application for this hardware, while third party enterprises can use the java interface to develop applications that can be widely ported to rfid information sensing equipment to exchange information. (3) the framework of the driver of the internet of things operating system should be flexible enough and compatible enough to dynamically load and unload device drivers (myhre et al., 2017); (4) the version of the internet of things operating system has a variety of branches or variants. for each industry, even each type of hardware (such as cars, refrigerators, etc.) will have a corresponding version, which requires good support for compiling development tools. based on the “the internet of things + ecology” concept, the internet, gis and the principles and methods of image graphics, the ecological compensation of soil contaminated by heavy metals is discussed. based on the connotation and principles and theoretical basis of ecological planning, the basic procedures, contents and methods are elaborated, and the role of the internet of things is emphasized (li et al., 2017). from the perspectives of the diversity, comprehensiveness, systemicity and logic of ecological factors, the ecological compensation ideas are analyzed. the internet of things operating system kernel, peripheral modules, application development environment, etc., are software system platforms that support the upper-level industry application software. the industry application software is the software that finally releases the hardware function of the iot terminal. such the iot operating system is the basis for the rapid industrialization of industrial applications. software and hardware vendors and users can only benefit from the standard and secure internet of things operating system. the internet of things can also develop rapidly, laying a good foundation for ecological compensation. the ecological compensation structure based on the internet of things is shown in figure 3. ecological compensation assume that the factors affecting farmers’ compensation (kan et al., 2017) include age (fage), gender (fsex), cultural level (fedu), labor population controlled, and the latter will lead to unnecessary investment. in order to obtain accurate information on contaminated areas, it is often necessary to increase the sample size, but this will result in an increase in the cost of sampling analysis. the efficient sampling scheme minimizes the total cost of economic losses due to sampling survey costs and survey uncertainty. the goal of sampling scheme optimization is to seek the best sample information to reduce the uncertainty of pollution repair. the spatial distribution of soil pollutants is affected by factors such as pollution sources, environmental conditions, and pollutant properties. it exhibits varying degrees of spatial correlation and variability in space. the description of the spatial variability of soil contaminants is the key to the impact of the survey results. therefore, combined with the specific needs of soil pollution compensation, an encryption method based on pollution probability and local spatial variability of pollutants is proposed to improve the estimation accuracy of ecological compensation methods in terms of pollution range and pollution degree. it also provides methodological support for ecological compensation in soil contaminated areas. based on the architecture of this ecological environment (lin et al., 2018), the architecture of the internet of things operating system is shown in figure 2: hardware design (1) the overall image size of the internet of things operating system must be highly scalable to accommodate applications with limited hardware resources. it can not only adapt to the application scenarios where the hardwa re resources such as the wristband are relatively rich, but also adapt to applications such as environmental monitors that are not very intelligent. specifically, the size of the operating system should be scalable from 10k to 10 megabytes; (2) in addition to providing a java interface, the internet of things operating system should also provide a standard c language interface to deal with high efficiency and real-time applications. for example, the manufacturer of the internet of things terminal can use c interface to develop efficient figure 2. the general mode of ecological compensation based on the internet of things 156 jia shanshan, zhou yanqing figure 3. structure diagram of ecological compensation based on the internet of things (fwrf), family cultivated area (fsize), whether to waste (fabd), degree of farmland pollution (fdgr) and bid amount (fbid). these eight factors are used as independent variables to establish multiple linear regression equations: inw f f f f f f age sex edu wrf size a = + + + + + + α β β β β β β β 1 2 1 3 4 5 6 bbd dgr bidf f+ +β β7 8 (1) based on the internet of things, the detailed analysis is carried out to determine a set of relatively fixed methods for analyzing the preference data (chen et al., 2017). the utility level (u) of the farmer is as follows u j sy s j= ( ) +, ,  (2) in the above formula, v is an observable variable; ε is an unobservable variable. j = 0 indicates that the sample is willing to stop farming and participate in the project, j = 1 indicates that the sample continues to farm and refuses to participate in the project. y is the income level of the sample under different decision conditions; s is a series of potential influencing factors (including age, gender, etc.) of the sample’s willingness to pay. the basic information of the respondents is shown in table 1. assuming the given compensation criterion is a (he et al., 2017), when the given compensation criteria increase the utility level, the survey sample is likely to choose to participate in the project. at this time there are  0, , ,y a s( ) + ≥ (3) assume that the utility difference (δu) caused by farmers’ participation/ non-participation in the project is: u u u= 0 1 (4) table 1. basic information of respondents variable distribution ratio /% variable distribution ratio /% age cultural level 20~29 2.5 no 4.6 30~39 15.8 primary school 29.2 40~49 33.4 junior high school 51.7 50~59 33.3 high school 14.5 ≥ 60 15.0 university and above 0 gender cultivated land area/hm2 male 62.1 ≤ 0.33 22.9 female 37.9 > 0.33~0.67 48.4 working population > 0.67~1.00 19.5 0 2.1 > 1.00 9.2 1 7.1 whether to waste 2 56.3 yes 50.4 3 12.1 no 49.6 4 16.7 degree of pollution ≥ 5 5.7 1 12.0 2 47.8 3 40.2 157ecological compensation method for soil polluted by heavy metals based on internet of things when expressed as an observable variable, the utility difference is u = +α β (5) in the above formula, is a constant term;  is the coefficient of influence of the given compensation criterion on the willingness to participate. when multiple variables are involved, they can be written as u n= + + + … +α β β β1 2 (6) at the same time, let η ε −ε= 0 1 (7) it is concluded that the possibility of local farmers participating in the project p0 is p p v y a s v y s0 0 10 1= +( ) + ≥ ( ) + , , , ,  (8) it can also be written as following. p f u0 = ( )η ∆ (9) p is the probability value; f uη ∆( ) is the probability distribution function of u with respect to . according to the logit model, there is: f u e u η −∆ ∆( ) = + 1 1 (10) for the model integral, the upper limit of the score is the upper limit of the bidding interval, and the estimated value of the average lower limit of the willingness of compensation wmean can be obtained: w f u dmean = ( )∫ η ∆0 max (11) where a is expressed as: a e da u a = +∫ 1 10 max (12) cost of ecological compensation the use of market theory to calculate the conditions of ecological compensation is very strict, and the establishment of the ecosystem of the ecosystem (kaiserbunbury et al., 2017) is very difficult. semi-market theory can determine the criteria for ecological compensation if the market fails to function. the theory is to use the market supply (zabihollah, 2017) and demand to determine the compensation standard. according to the market theory, there is a corresponding relationship between the quantity and price of goods, and the supply curve and the demand curve are given from the supplier and the demander respectively. according to the theoretical amendments proposed by mankiw in the principles of economics, the formula is as follows: p f q qs d= ( ), (13) q f i p p a nd o o r d= ( ), , , , (14) q f i t a ns p s= ( ), , , (15) in the above formulas,p is the price; qs is the supply curve; qd is the demand curve; io is the income; po is the price of the relevant product; pr is a preference; a is expected; nd is the number of demanders; ip is the input; t is technology; a is expected; ns is the number of sellers. similar to the market, the determination of ecological compensation standards is also affected by various factors, summed up as formula: s f q qs d= ( ), (16) s f q f mac mics s s s= ( ) = ( ), (17) s f q f mac micd d d d= ( ) = ( ), (18) where,s is the compensation standard; ss is the compensated standard; sd is the compensator standard; mac is the macro factor; mic is the micro factor;  is the noise (other factors). due to the uncertainty and complexity of the macro factors (zhang et al., 2018) in the study area, there is currently little research on the impact of changes in macro factors on ecological compensation standards. most of the relevant research is based on microscopic factors (sun & ma, 2017). for providers of compensation, the micro-factors (micd) that affect their compensation criteria include income, preferences, and expectations, which can be reflected by the provider’s willingness. for the recipients of compensation, the main influencing factors are income, direct costs, expectations and opportunity costs (lomas et al., 2018). the first three can be reflected in a willing manner, and the opportunity cost can be used as a basis for determining the criteria. in summary, semi-market theory is based on market theory, and in fact it evaluates the unilateral criteria of providers and recipients of ecological compensation separately. in the end, find the influencing factors and determine the criteria for ecological compensation. in the investigation of pollution sources, it is known that there are large copper pipe plants, metal material plants and electroplating plants in the research area. many factories have heavy metals in the waste residue. the survey and analysis of the willingness to abandon contaminated land is shown in table 2. table 2. investigation and analysis of the willingness to abandon farming of contaminated land willingness to abandon farming due to serious soil pollution number percentage % proportion of the total number of interviewees the proportion of people who are unwilling to give up the total willing 223 58.37 unwilling 159 41.62 land pollution is not serious 9 2.35 5.66 replanting other crops 13 3.41 8.18 losing living security 78 20.41 49.06 insufficient compensation 54 14.14 33.96 polluted, but has no effect on your health 5 1.31 3.14 based on the above, the following research priorities are determined: first: the type of ecological compensation is indeed divided. for different types of ecological compensation, specific methods for determining ecological compensation standards were established. different ecological compensation objects are also classified. according to the different functions of ecosystem services, it can be divided into four types of ecological compensation for biodiversity, carbon emissions, water resources and landscape protection. there are also many projects that combine these ecosystem services to compensate. according to the scope of compensation, it can be divided into global scale, national scale, regional scale and watershed scale. there is no method for determining the corresponding ecological compensation standard for each type. second: based on the three theories (value theory, market theory and semi-market theory), the basic research on the method of determining ecological 158 jia shanshan, zhou yanqing compensation standards is further strengthened. among them, the breakthrough point of value theory is to find the appropriate functional relationship to transform the value of ecosystem service function into ecological compensation standard. the difficulty of market theory is how to effectively establish the market and make the standards of ecological compensation self-regulating. third, exploring and clarifying the key factors affecting the determination of ecological compensation standards. it is an inevitable trend of ecological compensation standards to determine the standard of compensation by finding one or several specific parameters. fourth, with the rapid development of information technology, the study of ecological compensation standards should further strengthen the connection with computers. the wta distribution feature of the interview area is shown in figure 4. figure 4. wta distribution characteristics in the interviewed area sampling in order to improve the accuracy of the ecological compensation method, the reliability of the survey sampling information increases, as shown in table 3. table 3. sources of data information data attribute date data sources the date of internet of things date of land use chinese academy of sciences resource and environmental data center date of soil socioeconomic data statistics and survey data digital elevation model (dem) spatial cloud of geographic data pollution erosive factor calculated basic data and others soil erodibility factor calculated vegetation cover and management factors based on models and related research results maximum parameter drawing on and adopting relevant research results analysis of ecological compensation data based on the internet of things calculation of new soil retention soil retention mainly includes erosion control and sediment retention (kim et al., 2018). the soil retention is calculated by the soil retention module of the invest model, which consists of two main components. the first is the amount of sediment retained by the plot itself, expressed as the product of the uphill sand content and the sediment interception rate. the second is the amount of soil erosion reduced by vegetation cover and soil and water conservation measures (cunha et al., 2017), expressed as the difference between the potential soil erosion and the actual soil erosion. sedret rklsx uslex sedrxxd = + (19) sedrx sex usley sez z y x y x = ( ) = += ∏∑ 11 1 1 1   (20) where, sedretxd is the amount of soil retention intercepted by the grid x itself´ is the amount of sediment retention intercepted by the grid x itself. rklsx is the amount of potential soil erosion, uslex is the actual amount of soul erosion. sex is the interception rate, usley is the actual soil erosion amount of the uphill grid y, sez is the sediment interception amount of the uphill grid z. the new soil holding amount is obtained by the internet of things data to know the unit soil holding amount after the land use type is changed. in the process of calculation, different forest species (deciduous broad-leaved forest, sparse shrub forest, coniferous and broad-leaved mixed forest, evergreen coniferous forest and deciduous broad-leaved shrub forest) will be produced during the conversion of cultivated land to forest land. e e ei fi ci=  (21) s s si fi ci=  (22) where, si is the unit new soil holding capacity, expressed as t/hm; sfi is the soil holding capacity of different forest species per unit area, expressed as t/ hm; i represents the soil holding capacity per unit area of cultivated land, t/hm. the income item sets the planting area, production and selling price of the main crop; the expenditure item takes into account the cost of inputs such as seeds, fertilizers and pesticides. the difference between the income and the expenditure is divided by the planting area, which is the opportunity cost per unit area of land. as shown in equations (23)-(25). c c soppi i i = (23) c c ci ti si=  (24) c c c c c csi z f n h y= + + + + (25) where coppi is the opportunity cost amount; si is a good planting area; ci is the amount of net income of farmers’ cultivated land; cti is the total income of the crop industry; csi is the total amount spent on agriculture; cz is the amount of seed purchase fee; cf is the amount of fertilizer purchase; cn is the amount of pesticide purchase; ch is the total harvesting amount of the main crop; cy is the total amount spent on refueling agricultural machinery. data processing according to the recommended range and value of heavy metals in gb15618-1995 china soil environmental quality standard, the pollution index of each heavy metal is calculated (oliveira et al., 2017). due to the different dimensions of the raw data for each indicator, the magnitude difference is also very different. to eliminate the original data dimension, merge the order of magnitude to make it comparable. the raw data is first preprocessed. assuming the sequence x x x x ni i i i= ( ) ( ) … ( )( )1 2, , , , the sequence after processing is ′ = ′ ( ) ′ ( ) … ′ ( )( )x x x x ni i i i1 2, , , . when analyzing the system, different data can be selected for pre-processing of the original data according to different situations. the main methods of raw data preprocessing are as follows. initial value transformation: 159ecological compensation method for soil polluted by heavy metals based on internet of things x k x xi i k i ( ) = ( ) ( ) 1 (26) mean transformation: x k x k x x x ki i k n ( ) = ( ) = ( ) = ∑, 1 (27) percentage transformation: x k x k k x k i i i ( ) = ( ) ( )max (28) multiple transformation: x k x k k x k i i i ( ) = ( ) ( )min (29) range transformation x k x k k x k k x k i i i i ( ) = ( ) ( ) ( )  min max (30) interval value transformation x k x k k x k k x k k x k i i i i i ( ) = ( ) ( ) ( ) ( )   min max min (31) the preprocessed data sequence is x x x x ni i i i= ( ) ( ) … ( )( )1 2, , , (32) where i = 0, 1, 2, …, m. assume that the compensation indicator based on the internet of things is: x x x x n0 0 0 01 2= ( ) ( ) … ( )( ), , , (33) then ∆ −0 0i ik x k x k( ) = ( ) ( ) (34) where, k = 0, 1, 2, …, n, m is the number of samples; n is the number of pollutions. then the difference matrix is ∆ ∆ ∆ ∆ = ( )… ( ) ⋅ ⋅ ⋅ ( )…                 01 0 1 1 0 m n m (35) find the maximum (major difference) and minimum (very small difference) in the difference matrix, expressed by max and min. the correlation coefficient is δ ∆ ε∆ ∆ ε∆ 0 0 i i k k ( ) = +( ) + min max max (36)  in the formula (36) is the resolution coefficient, δ∈ 0 1, . the degree of association is  0 0 1 i i n k= ( )∑ (37) based on the internet of things, the calculation of ecological compensation related data has lower requirements on data, and the calculation amount is small, which is convenient for popularization and application. accurate measurement of new soil holdings is a key issue in the design and evaluation of ecological compensation projects. compared with previous empirical models, it more accurately reflects the amount of soil retention in different spatial contexts. the internet of things data (liu et al., 2017) provides quick and timely access to soil and land cover information for the study area. results the ecological compensation method based on the internet of things is to change the regional land use type through ecological compensation. it changes the cultivated land to forest land, thereby changing the supply of various ecosystem services such as soil conservation, water supply and carbon sequestration, and achieving the goal of improving the ecological environment. the working schematic diagram of the ecological compensation system based on the internet of things is shown in figure 5. the regional background value of heavy metal contaminated soil and the statistical value of the study area are shown in table 4. the mass ratio of cd in soil sample bt-34 was 0169μg/g, and the single factor evaluation reached 2123, which exceeded the moderate pollution index in the national standard. due to the low mass ratio of the other six heavy metals such as as, the average value of the soil sample is low, and the nemero index is evaluated as mild pollution. the mass ratio of zn in soil sample bt-87 was as high as 572μg/g, and the single factor was evaluated as 3136, far exceeding the figure 5. working diagram of ecological compensation system based on the internet of things 160 jia shanshan, zhou yanqing table 4. regional background values of soil contaminated by heavy metals and statistical values of study areas background values minimum sampling value maximum sampling value average sample value intermediate sampling value as 15 5 76 7.34 6.00 cd 0.2 0.051 1.76 0.19 0.14 cr 90 41 180 63.68. 60.00 cu 35 11 104 20.63 19.00 hg 0.15 0.011 1.114 0.08 0.05 ni 40 5 41.00 25.46 25.00 pb 35 16 654 38.24 22.00 zn 100 35 3075 112.89 56.50 table 5. soil contamination results of association analysis number clean mild pollution moderately pollution severe pollution results bt-34 0.898 0.693 0.693 0.484 mild pollutio bt-37 0.758 0.660 0.714 0.552 severe pollution bt-111 0.949 0.69 0.619 0.441 clean bt-177 0.944 0.635 0.598 0.443 clean bt-178 0.942 0.640 0.592 0.434 clean bt-179 0.940 0.635 0.593 0.439 clean table 6. compensation standards for soils in the study area and the amount of soil added compensation area (a) (b) (c) (d) area / 10,000 hm2 0.11 0.14 0.66 0.27 compensation standard / 10,000 yuan 0.46 0.49 1.12 1.05 capital demand / 10,000 yuan 506.67 686.70 745.01 2803.60 new soil holding capacity / 10,000 144.00 63.04 574.20 245.50 forest soil retention / 10,000 1152.00 504.32 4593.60 1964.00 suitable forest plants evergreen coniferous forest evergreen coniferous forest deciduous broad-leaved forest deciduous broad-leaved forest figure 6. compensation results of the study area national standard for heavy pollution. however, the nemero evaluation index of the soil sample was 21,528, which was evaluated as moderate pollution. in the soil sample bt-111, the mass ratio of cd was at the critical point of light pollution, and the mass ratios of the other seven heavy metals were all within the clean range, which was evaluated as appropriate for cleaning. however, the nemero evaluation index at this point was 01857, which was evaluated as mild pollution. the sampling points of soil samples bt-177, bt-178, and bt-179 are less than 3km apart, and the soil pollution levels in the vicinity of more than ten kilometers are clean. however, since the cd in the soil sample bt-178 exceeded the light pollution threshold of 0104μg/g, it was evaluated as mild pollution, which resulted in a small lightly contaminated area in the spatial interpolation analysis. the degree of pollution is magnified, which affects the correct judgment of local pollution. through comparative analysis, it can be seen that the gray correlation analysis method (the corresponding soil sample analysis results are shown in table 5) can more appropriately assess the heavy metal pollution of the samples. the diversity and complexity of pollution factors in the soil determine that it is difficult to directly analyze the relationship between heavy metal pollution indicators and soil environmental quality. the standard of soil compensation and the amount of soil added in the study area are shown in table 6. as shown in figure 6, the four regions (a), (b), (c), and (d) are used as research areas, and the information of the internet of things is processed and extracted based on other information such as the internet of things, land use map, economic data, and basic data. the invest model was used to simulate the changes in soil conservation of different forest species before and after returning farmland to forest. the invest model was used to simulate the changes in soil conservation of different forest species before and after returning farmland to forest. the spatial distribution of the new soil retention curve for the best forest species was derived using the minimum data method. the relationship between the new soil conservation and the compensation price of the best forest species under different land use conversion ratios was calculated, and the impact of ecological compensation on regional soil conservation was analyzed. with the main goal of increasing soil conservation, the soil in the areas contaminated by heavy metals is all retreated. the best planting species in the research area (b) is evergreen coniferous forest, which needs to be compensated for 0.49 million yuan/ (hm2•a), and the soil holding capacity can be increased by 630,400 t/a. in 8 years, the soil holding capacity was increased by 5,043,200 t/a. the best forest species in the study area (c) is deciduous broad-leaved forest, which needs to be compensated for 112,000 yuan/ (hm2•a), and the newly added soil holding capacity is 5,742,200 t/a. the polluted area (d) needs to be compensated for 10,500 yuan/ (hm2•a), and the newly added soil holding capacity is 2,455,000 t/a, and then the soil holding capacity is increased by 9.64 million t/a in 8 years. in the research area (a), the best forest species is coniferous and broad-leaved mixed forest, which needs to be compensated by 4.6 thousands yuan/ (hm2•a), and the soil holding capacity can be increased by 7.151 million t/a. in general, the total compensation for the above polluted areas was 178.8319 million yuan / a, and the total new soil holding capacity was 17.4084 million tons / a. 161ecological compensation method for soil polluted by heavy metals based on internet of things conclusions the proposed method realizes the spatialization and dynamism of quantitative evaluation of ecosystem service functions, and can visually display the spatial distribution and difference of soil conservation in the study area, which is a good choice for ecological compensation research. after years of development, ecological compensation has become an important means of ecological environment protection and heavy metal pollution control in china, but the current ecological compensation model is still too extensive. the internet of things data, the invest model and the minimum data method are flexibly utilized. the reasonable ecological compensation project can effectively improve the efficiency of ecological compensation and enhance the sustainability of the project, which helps to achieve visual and refined ecological compensation. reference chen, i. a., markowitz, v. m., chu, k., palaniappan, k., szeto, e., pillay, m., ratner, a., huang, j., andersen, e., huntemann, m., varghese, n., hadjithomas, m., tennessen, k., nielsen, t., ivanova, n. n., & kyrpides, n. c. (2017). img/m: integrated genome and metagenome comparative data analysis system. nucleic acids research, 45, d507. cunha, e. r. d., bacani, v. m., & panachuki, e. (2017). modeling soil erosion using rusle and gis in a watershed occupied by rural settlement in the brazilian cerrado. natural hazards, 85, 1-18. diepens, n. j., & koelmans, a. a. (2018). accumulation of plastic debris and associated contaminants in aquatic food webs. environmental science & technology, 52, 8510-8520. gäsiorek, m., kowalska, j., mazurek, r., & pajak, m. (2017). comprehensive assessment of heavy metal pollution in topsoil of historical urban park on an example of the planty park in krakow. chemosphere, 179, 148-158. gao, w., ghanbari, b., & baskonus, h. m. (2019). new numerical simulations for some real world problems with atangana-baleanu fractional derivative. chaos, solitons & fractals, 128, 34-43. he, j., li, y. w., xu, d., liang, x., liang, b., & wang, c. (2017). deadbeat weighted average current control with corrective feed-forward compensation for microgrid converters with nonstandard lcl filter. ieee transactions on power electronics, 32, 2661-2674. he, q., zeng, c., xie, p., liu, y., & zhang, m. (2018). an assessment of forest biomass carbon storage and ecological compensation based on surface area: a case study of hubei province, china. ecological indicators, 90, 392-400. kaiserbunbury, c. n., mougal, j., whittington, a. e., valentin, t., gabriel, r., olesen, j. m., & blüthgen, n. (2017). ecosystem restoration strengthens pollination network resilience and function. nature, 542, 223. kan, t., nguyen, t. d., white, j. c., malhan, r. k., & mi, c. c. (2017). a new integration method for an electric vehicle wireless charging system using lcc compensation topology: analysis and design. ieee transactions on power electronics, 32, 1638-1650. kim, j., johnson, l. e., cifelli, r., choi, j., & chandrasekar, c. (2018). derivation of soil moisture recovery relation using soil conservation service (scs) curve number method. water, 10, 833. kokkonis, g., psannis, k. e., roumeliotis, m., & schonfeld, d. (2017). realtime wireless multisensory smart surveillance with 3d-hevc streams for internet-of-things. journal of supercomputing, 73, 1-19. li, y., qiu, j., zhao, b., pavao-zuckerman, m., bruns, a., qureshi, s., zhang, c., & li, y. (2017). quantifying urban ecological governance: a suite of indices characterizes the ecological planning implications of rapid coastal urbanization. ecological indicators, 72, 225-233. lin, y., chen, j., jin, z., quan, y., han, p., guan, s., & jiang, x. (2018). impacts of human activities on coastal ecological environment during the rapid urbanization process in shenzhen, china. ocean & coastal management, 154, 121-132. liu, c., wei, c., zhang, c., ma, m., rao, w., li, w., he, k., & gao, m. (2018). developing the ecological compensation criterion of industrial solid waste based on emergy for sustainable development. energy, 157, 940-948. liu, h., xu. r. z., & yu. h. l. (2017). simulation of agricultural greenhouse humidity data extraction based on internet of things. computer simulation, 34. lomas, j., claxton, k., martin, s., & soares, m. (2018). resolving the “cost-effective but unaffordable” paradox: estimating the health opportunity costs of nonmarginal budget impacts. value in health, s1098301517336136. millard, l., davies, n. m., gaunt, t. r., davey-smith, g., & tilling, k. (2018). software application profile: phesant: a tool for performing automated phenome scans in uk biobank. international journal of epidemiology, 47, 29-35. monaco, d., chianese, e., riccio, a., delgado-sanchez, a., & lacorte, s. (2017). spatial distribution of heavy hydrocarbons, pahs and metals in polluted areas (the case of “galicia”, spain). marine pollution bulletin, 121, 230. myhre, g., forster, p. m., samset, b. h., hodnebrog, o., sillmann, j., aalbergsjo, s. g., andrews, t., boucher, o., faluvegi, g., fläschner, d., iversen, t., kasoar, m., kharin, v., kirkevag, a., lamarque, j. f., olivié, d., richardson, t. b., shindell, d., shine, k. p., … zwiers, f. (2017). pdrmip: a precipitation driver and response model intercomparison project—protocol and preliminary results. bulletin of the american meteorological society, 98, 1185-1198. oliveira, l. m., maillard, p., & andrade pinto, e. j. (2017). application of a land cover pollution index to model non-point pollution sources in a brazilian watershed. catena, 150, 124-132. shu, l. (2018). games between stakeholders and the payment for ecological services: evidence from the wuxijiang river reservoir area in china. 6:e4475. sun, p. g., & ma, x. (2017). understanding the controllability of complex networks from the microcosmic to the macrocosmic. new journal of physics, 19, 013022. yang, g., shang, p., he, l., zhang, y., wang, y., zhang, f., zhu, l., & wang, y. (2019). interregional carbon compensation cost forecast and priority index calculation based on the theoretical carbon deficit: china as a case. science of the total environment, 654, 786-800. zabihollah, r. (2017). supply chain management and business sustainability synergy: a theoretical and integrated perspective. sustainability, 10, 275. zhang, t., li, b., yuan, y., gao, x., sun, q., xu, l., & jiang, y. (2018). spatial downscaling of trmm precipitation data considering the impacts of macro-geographical factors and local elevation in the three-river headwaters region. remote sensing of environment, 215, 109-127. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jsjfz201710058 about:blank palabras clave: roca fracturada; carga dinámica; vida a la fatiga; curva s-n. how to cite item li, r., jian, w., kang, r. (2018). study on fatigue life of fractured rock under constant amplitude cyclic loading. earth sciences research journal, 22(4), 275-280. doi: https://doi.org/10.15446/esrj.v22n4.77379 the electro-hydraulic servo loading system was applied to make cyclic loading tests for fractured rock-like specimens. according to the results of the fatigue test, the power function model was used to fit and analyze the test data, and the fitting degree of the curve was determined by the correlation coefficient, and the result was fitted well. the main factors influencing the fatigue life of specimens were analyzed. the upper limit stress was the most important factor affecting the fatigue life of specimens. under the same conditions, the specimen had a shorter fatigue life when being applied by higher upper limit stress; the rock of different types and strengths not only had different fatigue life, but also had the different sensitive degrees between the fatigue life and the upper limit stress; the fatigue life of specimens was significantly influenced due to different initial damages. abstract keywords: fractured rock; dynamic loading; fatigue life; s-n curve. study on fatigue life of fractured rock under constant amplitude cyclic loading estudio sobre la vida a la fatiga de rocas fracturadas bajo cargas cíclicas de amplitud constante issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n4.77379 el sistema electrohidráulico de carga servo se aplicó para realizar pruebas de carga cíclicas para muestras de roca fracturada. de acuerdo con los resultados de la prueba de fatiga (resistencia a la fatiga), el modelo de función de potencia se utilizó para ajustar y analizar los datos de la prueba; el grado de ajuste de la curva se determinó mediante el coeficiente de correlación, y se ajustó el resultado. se analizaron los principales factores que influyen en la resistencia a la fatiga de las muestras. la tensión límite superior fue el factor más importante que afectó la vida a la fatiga (resistencia de fatiga) de las muestras. en las mismas condiciones, el espécimen tuvo una vida a la fatiga más corta cuando se aplicó por un límite superior de tensión; la roca de diferentes tipos y fuerzas no solo tuvo una vida a la fatiga diferente, sino que también tuvo diferentes grados de sensibilidad entre la vida a la fatiga y la tensión límite superior; la vida a la fatiga de los especímenes fue significativamente influenciada debido a diferentes daños iniciales. resumen record manuscript received: 01/08/2018 accepted for publication: 08/10/2018 earth sciences research journal earth sci. res. j. vol. 22, no. 4 (december, 2018): 275-280 run li1*, wenbin jian2, rongtao kang3 1engineering college, fujian jiangxia university, fuzhou 350108, china; 2college of environment and resources, fuzhou university, fuzhou 350116, china; 3fuzhou architecture design institute, fuzhou 350011, china *corresponding author: lr52tkd25@126.com g e o t e c h n ic a l e n g in e e r in g 276 run li, wenbin jian, rongtao kang introduction many defects exist in the natural rock masses, such as: fractures, joints, etc. in the process of rock engineering construction and service, there are not only static loading, but also dynamic loading, especially the rock engineering along the highway, railway and bridge. due to the long-term impact from traffic fatigue loading, the internal stress field of rock is changing, the rock strength is continuously attenuated with the accumulation of fatigue damage, and long-term safety and stability of rock engineering are greatly impacted. with the improvement of people’s requirements on the safety and durability of rock engineering, the scholars at home and abroad conducted relevant researches on the fatigue of rock. xiurun (1987, 1992, 2003) researched the irreversible deformation and fatigue property of rock under cyclic loading, and put forward fatigue threshold value at the same time. yanbo, xing and guojie (2017)researched the stress-strain entire process curve, micro structure change, macro fracture characteristics and other fatigue damage characteristics of gypsum rocks under the cyclic loading by using acoustic emission system and environmental scanning electron microscope. erarslan and williams (2012) researched the fatigue damage mechanism of tuff under two different types of cyclic loading methods by using sem. momeni, karakus and khanlari (2015) researched the mechanical property of granite under cyclic loading. according to the existing researches, the strength, deformation, fracture characteristics and failure mechanism of rock materials under cyclic loading are obviously different from that of static loading. compared with other materials, the researches on rock fatigue are relatively limited at home and abroad, and rock fatigue is still one of the important lucubrate problems in rock engineering. in this paper, the constant amplitude fatigue test was applied to research the fatigue life of rocklike fractured specimens, and the suitable fatigue life models were selected. meanwhile, the main influencing factors of fatigue life were analyzed. the research results were helpful to the prediction of rock engineering life and design of fatigue resistance under fatigue loading. test overview rocklike specimens as the natural rock has defects without specific laws, there are great dispersion and uncertainty in the rock fatigue test, and thus, it is difficult to achieve the purpose of test. through the indoor similar tests, the way of researching the strength, deformation and failure mechanism of rocks under certain structural characteristics is very effective to study the fractured rocks (haeri, shahriar and marji, 2014a, b; cao, liu and pu, 2015; zhou, cheng and feng, 2014). in recent years, indoor simulation tests have been widely used (cheng-zhi and ping, 2012; jinjin, ping and chengzhi, 2014; shilei, junsheng and feng, 2016). in this test, barite powder, river sand (particle size was less than 0.5 mm), 32.5rcomposite portland cement, model gypsum, borax solution with concentration of 1% were used as similar materials, and the rocklike specimens were made according to the weight ratio, barite powder: river sand, cement, gypsum, borax solution = 42.0:21.5:8.5:12.7:15.3. the dimensions of specimens were 70mm×70mm×140mm, while the thin metal plate was used to make pre-facture, the length of fracture was 20mm, and the angles of fracture were set to 0°, 30°, 45°, 60° and 90°. the fractured specimens were shown in figure 1. test methods the electro-hydraulic servo fatigue testing machine was applied to load the fractured specimens, and constant amplitude loading control method was used in the test. the loading frequency was 5hz, and sine wave was used as loading waveform. the characteristic parameters of loading waveform were shown in figure 2. of which, was the upper limit stress of cyclic loading; was lower limit stress of cyclic loading; , was the loading amplitude; t was cycle time, , was the frequency. the amplitude range of loading was 1.9 ~ 2.4 mpa according to the uniaxial compressive strength of specimens. fatigue curve analysis s-n curve fatigue life s-n curve (also called as wohler curve) refers to the relationship curve between stress level sand material fatigue life n, which is mainly used to estimate and evaluate the fatigue life or fatigue strength of materials. in practical engineering, the stress level should be the upper limit of stress smax, n represents fatigue life, the absolute fatigue life log n or logarithmic fatigue life log n could be chosen. the fatigue life mentioned here refers to the required stress cycle-index n when the materials reach the fatigue failure under the cyclic loading. according to the characteristics of the relationship between stress level and material fatigue life of s-n curve, the domestic and oversea researchers put forward many empirical models on the basis of experiments, and there are many commonly used forms as follows: exponential function model: n e cs⋅ = (1) figure 1. fractured specimens 277study on fatigue life of fractured rock under constant amplitude cyclic loading in equation,  and c are material constants. both sides of equation (1) take the logarithm, then: lg lgn s cα+ = (2) a c= lg / , b = l/  are set, the equation (2) can be simplified as: s a b n=  lg (3) therefore, s-n exponential function model represented by equation (3) is a straight line in semi-logarithmic coordinate system. power function model: s n c = (4) in equation,  and c are material constants. both sides of equation (4) take the logarithm, then: α +lg lg lgs n c= (5) a c= lg / , b = l/  are set, the equation (5) can be simplified as: lg lgs a b n=  (6) therefore, s-n power function model represented by equation (6) is a straight line in double logarithmic coordinate system. (3) power function model of three parameters: s s n cf− α 1( ) = (7) in equation,  and c are material constants. s f 1 represents the normalized value of rock fatigue limit, and it is obtained according to the ratio between the fatigue limit and the static strength of rock. s f 1 is the material constant related to the rock property. both sides of equation (7) take the logarithm, then: α − +lg lg lgs s n cf1( ) = (8) a c= lg / , b = l/  are set, the equation (8) can be simplified as: lg lgs s a b nf 1( ) = (9) in practical engineering, the s-n curve is generally fitted to the indoor fatigue test data according to the corresponding model. fatigue curve fitting and analysis of rocklike specimens in this paper, the test data was fitted according to equation (6), the fitting degree of curve was judged by the correlation coefficient r2 of the fitting curve. the fitting results are shown in figure 3, and the curve data is shown in table 1. figure 3 and table 1 show: (1) under the fatigue loading, if the upper limit stress of complete and fractured rocklike specimens was higher, the fatigue life of specimens was shorter; if the upper limit stress was lower, the fatigue life of specimens was longer. (2) in terms of the slope of the s-n fitted curve, when the slope was smaller, the specimen fatigue life affected by the loading stress was more sensitive, the significant change of fatigue life was caused by the slight change of loading stress. therefore, if the slope of s-n fitted curve was smaller, the deviation range in the prediction of fatigue life should be larger; if the slope was larger, the deviation range in the prediction of fatigue life should be smaller. (3) in the process of specimens preparation and test, errors were inevitable. therefore, under the same upper limit stress conditions, the fatigue life of rocklike specimens with same fracture dip was not exactly the same. the fatigue life discreteness of some rocklike specimens was relatively large, but from the fitting results of the test data, the correlation coefficient r2 was 0.85 above. it indicated that the fitting degree of s-n model and test data was relatively good. in practical engineering, s-n fitted curves are widely used in the researches of material fatigue life due to its clear concept, simple form and convenient measurement. under the condition of constant amplitude loading, the fatigue life of rock can be estimated by using interpolation method in s-n curves. influence factors of fatigue life on the one hand, the fatigue damage of the rock under fatigue loading is related with rock itself (rock type, strength, defects, etc.); on the other hand, it is affected by test conditions (stress amplitude, waveform, frequency, figure 2. schematic diagram of periodic loading waveform table 1. fatigue life s-n curve table of rocklike specimens fractured dip curve parameters s-n curve correlation coefficient r2 a b complete 0.75798 0.09615 lg s = 0.75798 0.096151g n 0.90071 0° 0.73887 0.09489 lg s = 0.73887 0.094891g n 0.91066 30° 0.63917 0.07931 lg s = 0.63917 0.079311g n 0.85416 45° 0.55511 0.06348 lg s = 0.55511 0.063481g n 0.90898 60° 0.62014 0.07422 lg s = 0.62014 0.074221g n 0.85370 90° 0.76277 0.10267 lg s = 0.76277 0.102671g n 0.91967 278 run li, wenbin jian, rongtao kang etc.). due to the complex geological conditions and different environment of natural rock mass, the influence factors of fatigue life are more complicated, and there is relatively large discreteness in fatigue life. in this paper, only the main influencing factors involved in the test were analyzed. influence of upper limit stress the fatigue life of specimens is greatly affected by the upper limit stress. the previous studies have shown that the upper limit stress is the first factor influencing the fatigue life (xiufeng, 2008). the cyclic loadings of different upper limit stress on the rocklike specimens were made, and the fatigue life of typical specimens was shown in table 2. it was indicated in table 2 that under the same conditions, if the upper limit stress of fatigue test was larger, the fatigue life of specimens was shorter; conversely, if the upper limit stress was smaller, the fatigue life of specimens was longer. generally speaking, if the strength of the material is higher, the external force is lower and the fatigue life of specimens is longer (yu, 2003). here, the fatigue threshold value of the material should be considered, the fatigue threshold value refers to the stress threshold of the fatigue failure of rock. ge xiurun put forward, whether the rock was damaged or not, it was related to the stress threshold value, and he also thought, rock fatigue threshold value was close to conventional yield value (xiurun, 1987). in the fatigue test, if the upper limit stress is lower than the threshold of fatigue, the specimens could not reach the fatigue failure. if the upper limit stress is too close to the peak stress and the properties of (a) complete specimen (b)  = 0° (c)  = 30° (d)  = 45° (e)  = 60° (f)  = 90° figure 3. s-n fitted curve of typical complete specimens and fractured specimens. 279study on fatigue life of fractured rock under constant amplitude cyclic loading the specimen are unstable, the failure rate of test will be increased. in this test, the upper limit stress ratio 0.76 ~ 0.90 was selected, and after the test proved, this range was relatively suitable. influence of rock type and strength due to different formation causes of rocks, the mineral composition, content and other aspects are different, so that different types of rocks have differences in their natures. even the same type of rocks, their material composition, structure and construction are different under different internal and external dynamic geological processes, and then, the rock strengths are different. xiufeng (2008)made fatigue test researches on grayish yellow sandstone, and concluded the s-n relationship of grayish yellow sandstone fatigue strength under cyclic loading: smax = 1.326 0.109 log nf r = -0.997 in equation, smax = max/fc was upper limit stress ratio; nf was fatigue life. the fitted curve s-n was shown in figure 4. yu (2003) conducted a fatigue test study on red sandstone and concluded the regression equation: smax = 1.04333 0.05978 log nf r = -0.931 in equation, s = max/c was upper limit stress ratio; n was fatigue life. the fitted curve s-n was shown in figure 5. figure 4. grayish yellow sandstone fatigue s-n curve under cyclic loading (xiufeng, 2008) figure 5. red sandstone fatigue s-n curve under cyclic loading (yu, 2003) figure 4 and figure 5 showed both grayish yellow sandstone and red sandstone were called as sandstone; however, due to different internal mineral composition, content and strength of rocks, the slopes of fatigue life s-n curves were different. the slope of fatigue life s-n curves reflected the sensitive relationship between the fatigue life of specimens and upper limit stress ratio. it was indicated, if the slope was smaller, the change of fatigue life to upper limit stress ratio was more sensitive; on the contrary, it was less sensitive. figure 4 indicated that the slope of grayish yellow sandstone s-n curve was relatively large, but the figure 5 illustrated that the slope of red sandstone was relatively small. it was indicated, with the change of stress, the sensitive degree of fatigue life of grayish yellow sandstone was lower than that of red sandstone. it could be seen that different types and strengths of rocks not only had different fatigue life, but also had different sensitive degree between fatigue life and maximum stress. impact of initial damage the initial damage indicates the loading history of the rock, and it is the damage state of the rock before fatigue loading. even the same type of rock, if the environment is different, the initial damage is also different. in this test, the preparation process and curing conditions of each specimen were consistent, the small discrepancies in the specimen preparation were not considered as the key points; therefore, pre-fracture was used as the initial damage of rocklike specimen. it was found in the test, different dips of pre-factures had the significant effect on the fatigue life of rocklike specimens. figure 6 illustrated the fatigue life changes of rocklike specimens with different dips when the loading amplitude, the curves were the average fatigue life line of rocklike specimens with different dips. table 2. constant amplitude fatigue life of typical specimens facture dip upper limit stress smax / mpa rmax fatigue life nf / time average life nf /time 30° 2.0 0.81 13100 15500 18200 19000 20800 17320 2.1 0.85 8100 8600 9100 10300 13100 9840 2.2 0.89 5100 5200 5800 6600 8000 6140 45° 1.9 0.79 14300 19100 21200 24400 27000 21200 2.0 0.83 8900 9500 10200 11800 12300 10540 2.1 0.88 3700 4500 4900 5600 6100 4960 60° 2.0 0.80 14600 16800 17000 20700 21100 18040 2.1 0.84 8500 9500 11600 12000 13100 10940 2.2 0.88 3900 5800 6300 6600 7500 6020 note: rmaxwas upper limit stress ratio, equal to the upper limit stress to the uniaxial compressive strength of the specimens, rmax = smax/c 280 run li, wenbin jian, rongtao kang figure 6. curves of rocklike specimen fatigue life under same loading amplitude with the change of fracture dips figure 6 illustrated: according to the fatigue life data of specimens tested in this paper, under the same loading amplitude, the fatigue life of complete specimens was largest. when the dip of fracture  < 45º, the fatigue life of rocklike specimens was reduced with the increase of fracture dip; when  < 45º, the fatigue life was increased with the increase of fracture dip. conclusion (1) according to the fatigue test results of the rocklike fracture test, the power function model was adopted to fit the test data, and the fitting degree of the curve was determined by the correlation coefficient , and the result was fitted well. (2) s-n fitted curves had clear concept, simple form and convenient measurement; under the constant amplitude loading condition, the interpolation method was applied in the s-n curves to estimate the fatigue life of rock. (3) the upper limit stress was the most important factor affecting the fatigue life of specimens. under the same conditions, the specimen had a shorter fatigue life when being applied by higher upper limit stress, whereas the specimen had a longer fatigue life when being applied by lower upper limit stress. (4) the slope of fatigue life s-n curves reflected the sensitive relationship between the fatigue life of specimens and upper limit stress ratio. it was indicated, if the slope was smaller, the change of fatigue life to upper limit stress ratio was more sensitive; on the contrary, it was less sensitive. the rock of different types and strengths not only had different fatigue life, but also had the different sensitive degrees between the fatigue life and the maximum cycling stress. (5) the initial damage reflected the loading history of the rock, and different dips of pre-factures (initial damage) had the significant effect on the fatigue life of rocklike specimens. acknowledgements this work was kindly supported by natural science foundation of fujian province (no. 2015j05077), young and middle-aged education scientific research project of fujian province (ja15533), china scholarship council (csc, no. 201608350057), and national natural science foundation of china (no. 41072232). references cao, p., liu, t. & pu, c. (2015). crack propagation and coalescence of brittle rock-like specimens with pre-existing cracks in compression. engineering geology, 187, 113-121. erarslan, n. & williams, d. (2012). the damage mechanism of rock fatigue and its relationship to the fracture toughness of rocks. international journal of rock mechanics and mining sciences, 56, 15-26. haeri, h., shahriar, k. & marji, m. f. (2014a). experimental and numerical study of crack propagation and coalescence in pre-cracked rocklike disks. international journal of rock mechanics and mining sciences, 67, 20-28. haeri, h., shahriar, k. & marji, m. f. (2014b). on the strength and crack propagation process of the pre-cracked rock-like specimens under uniaxial compression. strength of materials, 46, 140-152. jinjin, cao, p. & pu. c. (2014). influence of flaw parameters on damage mode and strength of rock-like materials (in chinese). changsha, china: journal of central south university (science and technology), 45(2), 529-535. momeni, a., karakus, m., & khanlari, g. r. (2015). effects of cyclic loading on the mechanical properties of a granite. international journal of rock mechanics and mining sciences, 77, 89-96. pu, c., & cao, p. (2012). failure characteristics and its influencing factors of rock-like material with multi-fissures under uniaxial compression. transactions of nonferrous metals society of china, 22, 185-191. shilei, k., junsheng, y., & feng, y., (2016). macro-and-meso analysis of failure behavior of rock-like samples with flaws using a numerical method (in chinese). journal of yangtze river scientific research institute, 33(4), 71-77. xiufeng, f. (2008). study on the damage evolution of rock fatigue and the response of rock slope under cyclic loading (in chinese). fuzhou, china: fuzhou university. xiurun, g. (1987). study on deformation and strength behaviour of the large-sized triaxial rock samples under cyclic loading (in chinese). rock and soil mechanics, 8(2), 11-19. xiurun, g. & yingfa, l. (1992). discussion on the problem of rock fatigue failure and irreversible deformation under cyclic loading (in chinese). chinese journal of geotechnical engineering, 14(3), 56-60. xiurun, g., yu, j. & yunde, l. (2003). testing study on fatigue deformation law of rock under cyclic loading (in chinese). chinese journal of rock mechanics and engineering, 22(10), 1581-1585. yanbo, z., xing, h. & guo, j. (2017). experimental study of fatigue characteristics of gypsum rock under cyclic loading (in chinese). chinese journal of rock mechanics and engineering, 36(4), 940-952. yu, j. (2003). fatigue failure and deformation development law of rock under cyclic load (in chinese). shanghai, china: shanghai jiaotong university. zhou, x. p., cheng, h. & feng y. f. (2014). an experimental study of crack coalescence behaviour in rock-like materials containing multiple flaws under uniaxial compression. rock mechanics and rock engineering, 47, 1961-1986. palabras clave: fracturas tectónicas; área zhongguai; rocas volcánicas. how to cite item fan, c. h., qin, q. r., liang, f., fan, z. h., & li, z. (2018). fractures in volcanic reservoirs: a case study in zhongguai area at the northwestern margin of junggar basin (china). earth sciences research journal, 22(3), 169-174. doi: https://doi.org/10.15446/esrj.v22n3.75426 fractures in carboniferous volcanic rocks located at zhongguai area (china) highly influence the accumulation and productivity of oil and gas. as such, the study of development periods and genetic mechanisms of tectonic fractures could throw useful information regarding the evaluation and development of that reservoir. their tectonic origins caused high-angle and oblique shear fractures. the primary orientation of those fractures appears close to ew (270°±10°), nw (300°±15°), ne (45°±15°), and sn (0°±10°). at least four fracture generations can be found in carboniferous volcanic rocks at zhongguai area. combined with a tectonic evolution, they are based on the segmentation relationship of the fracture fillings, the thermometry measurement of the fracture filling inclusion, and the acoustic emission, as well. affected by a new horizontal principal stress, the opening and permeability of nearly ew fractures are the best. in this way, a priority in the development of well's patterns should be considered close to ew fractures. the pressure change in the process of exploitation may damage the reservoir permeability of fractured volcano rocks severely. accordingly, well patterns should be adjusted to dynamic changes of permeability happened during the oilfield development since some differences have been detected in distinct fracture sets. abstract keywords: tectonic fractures; zhongguai area; volcanic rocks. fractures in volcanic reservoirs: a case study in zhongguai area at the northwestern margin of junggar basin (china) fracturas en yacimientos volcánicos: caso de estudio del levantamiento de zhongguai en la margen noreste de la cuenca de junggar, china issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n3.75426 las fracturas en las rocas carboníferas volcánicas del área de zhongguai tienen una influencia determinante en la acumulación y producción de petróleo y gas. el estudio de los períodos de desarrollo y el mecanismo genético de las fracturas tectónicas proveen información importante para la evaluación del yacimiento y su desarrollo. en esta zona se encontraron fracturas con un ángulo alto de cizallamiento y fracturas oblicuas causadas por origen tectónico. la orientación principal de las fracturas es cercana a ew (270°±10°), nw (300°±15°), ne (45°±15°) y sn (0°±10°). se encontraron por lo menos cuatro generaciones de fracturas en las rocas volcánicas del área de zhongguai. combinadas con la evolución tectónica, las fracturas están basadas en la relación de segmentación del relleno de las fracturas, la termometría de las inclusiones, y la emisión acústica. afectadas por esfuerzos horizontales recientes, las fracturas de apertura y permeabilidad con dirección ew son las mejores. aquellas fracturas cercanas a la dirección ew deben ser consideradas como una prioridad en el desarrollo de los patrones del pozo. el cambio de presión en los procesos de explotación podría causar daños severos en la permeabilidad de las rocas volcánicas fracturadas en el yacimiento. durante el desarrollo del campo petrolífero, los patrones del pozo deben ser ajustados de acuerdo con los cambios en las dinámicas de permeabilidad ya que las configuraciones de fracturas son diferentes. resumen record manuscript received: 02/02/2018 accepted for publication: 17/07/2018 earth sciences research journal earth sci. res. j. vol. 22, no. 3 (september, 2018): 169-174 cunhui fan1,2,3, qirong qin1,3*, feng liang4, zenghui fan5, zhi li1,3 1state key laboratory of oil and gas reservoir geology and exploitation, southwest petroleum university; chengdu 610500, china 2state key laboratory of oil and gas reservoir geology and exploitation, chengdu university of technology, chengdu 610059, china 3school of geoscience and technology, southwest petroleum university; chengdu 610059, china. 4geological research institute of central sichuan district of petrochina southwest oil&gas field company, suining 629000, china 5cnpc bureau of geophysical prospecting inc, southwest geophysical research institute, chengdu 610213, china corresponding author’s e-mail: 986336692@qq.com g e o l o g y 170 cunhui fan, qirong qin, feng liang, zenghui fan & zhi li introduction volcanic reservoirs present a firm heterogeneity and poor physical properties. tectonic fractures originated from tectonism constitute the main seepage paths and reservoir spaces in the volcanic reservoirs. the study of the characteristics, controlling factors, and formation mechanism of tectonic fractures in volcanic rocks play a crucial role in oil and natural gas exploration and development. as several authors point out, the study regarding the growth characteristics and the division of the development phase in tectonic fractures constitutes a problem to tackle with (nelson, 2001; larsen, grunnaleite & gudmundsson, 2010). the zhongguai area is located at the northwestern margin of the junggar basin. the carboniferous system is one of the primary hydrocarbonbearing sources in the region with the main reservoir rocks being volcano ones. several years of exploration at the zhongguai area were needed to achieve a breakthrough series of oil and gas exploration in the carboniferous volcano rock reservoir. commercial oil flow was obtained in many testing wells, in which h019 well emitted a daily oil mean of 105.96 t/d and a (daily) gas mean of 2. 14 × 103 m3 in the carboniferous oil test, a reasonable prospect for a gas and oil exploration in the area of volcanic rocks. however, from current tests of oil and gas production, the variation of the single well productivity becomes larger (a unique well oil production from 0.1 to 105 t/d), displaying the typical properties of a fractured reservoir, with the fracture effect in the production control being significant. the frequent tectonic alteration experienced in the area contribute to relatively developed fractures. it is worth mentioning that fractures at different stages play a crucial role in the regulation and controlling of hydrocarbon migration and accumulation. also, fractures become essential in high yields (fan & qin, 2016). in this paper, the authors have analyzed the characteristics of the fracture's development and carried out a preliminary qualitative division of fracture formation times by combining structural analysis and core observation in the study area with imaging logging interpretation. additionally, they have contributed a quantitative division of fracture development phases by an acoustic emission experiment of rocks and a fracture filling inclusion (warren et al., 2017; yang et al., 2018; fu & liu, 2017). geological characteristics zhongguai area is located at the conversion parts of the fault belts of ke-wu and hongche, in the northwestern margin of junggar basin. also, it plunges to that basin from the hongche one to the southeast. three sides of the zhongguai area face the concave (adjacent to mahu sag, pcn1well west sag, and shawan sag), and it is a favorable directional zone for oil and gas migration and accumulation (figure 1). its main form is a wide flat nose paleo-uplift, which is formed under the action of extruding stress field in the carboniferous to early permian. it started to hide at the end of permian; overlap was deposited on a raised in wuerhe formation (upper permian) and angular unconformable contact with the jiamuhe formation of the permian; the triassic-jurassic draped on top of a raise was a smaller bulge. yanshan movement caused an uplift further, so the tectonic pattern formed to single southeast tilting, and upwelling area tilted eventually in the himalaya movement period. volcanic rock in the research area buries deeply. composing the basal part of the basin, which is mainly developed by the interaction between continental-oceanic volcanic rock and environment, primarily includes volcanic lava (andesite and basalt), pyroclastic rocks (volcanic breccia and tuff), and few intrusive rocks (granite), with tuff and andesite being widely distributed. in that area, the volcanic rock reservoir physical property is reduced. the average porosity range fluctuates from 5 to 7 %. its matrix permeability value is generally less than 1×10-3 μm2, with a mean value being only equal to 0.61×10-3 μm2. as such, it corresponds to a low reservoir not only in porosity but also in permeability. since carboniferous, the tectonic fracture of such a reservoir has been growing more and more due to frequent tectonic movements leading to advantageous accumulation places. also, the infiltration channel of the reservoir in this area exhibits a significant influence on oil and gas enrichment and development effect. therefore, research on layer fracture characteristics, formation stage, and flow characteristics of the volcano rock reservoir in that area becomes crucial for volcanic rock reservoir exploration and efficient development. figure 1. tectonic map in zhongguai area. development characteristics of structural fractures the core observation regarding 596, h56a, h019, and 16 wells suggests that carboniferous volcano rock reservoir in zhongguai area could be divided into four types of fractures, namely, diagenetic, structural, weathering, and solution ones. first, tectonic fractures are mostly developed and widely distributed in various types of volcano rocks. according to the statistics of tectonic fracture dips, it holds that fractures can be divided into two categories: a class of high angle of inclination > 60° seam and straight seam for about 70 %. another kind of dip is at 30° 60° oblique fractures, accounted for about 20 %. the high angle fractures are mainly shear ones, characterizing the joint surface flat rules, stable occurrence, visible scratches, mineral filling, and so on (figure 2-a, figure 2-b). oblique fractures mainly include shear and tensile ones. the different characteristics among the fractures of an intersection intersperse with others, leading to a connected seepage system. in this way, the development and distribution of the volcano rock in carboniferous reservoirs play a crucial role in controlling. according to some hydraulic fracturing experiments and the analysis of the fracture azimuth imaging logging after treatment, the fractures mainly developed are located at ne (45°± 15°), nw (300°± 15°), ew (85°± 10°), and sn (0°± 10°), with ne (45°± 15°) and ew (85°± 10°) being the ones mostly developed, followed by nw (300°± 15°). on the other hand, the worst is close to sn (0°± 10°). it is worth noting that fracture filling material mainly consists of calcite, chlorite, and gypsum; fracture filling degree is high; around 60 % of fractures are wholly filled; half-filled fractures and not filled ones accounted for 40 %. 171fractures in volcanic reservoirs: a case study in zhongguai area at the northwestern margin of junggar basin (china) the imaging logging analysis results point out that fractures are concentrated in the 2-5 strips/m, followed by 0.4-0.6 strips/m, with the overwhelming majority fracture density corresponding to a medium value < four strips/m. the scope of the fracture length change is broad. the long range between 10-30 cm accounts for more than 70%, and the maximum cutting depth can reach 180 cm. fracture width is in the range of 1-5 mm; in the local core exist open fractures whose fracture width is > 1 cm. based on the field investigation and statistics of fracture spacing, it holds that the fracture spacing range goes from 0.3 to 0.7 m, and the average spacing equals 0.5 m. discussion on fracture growth stages in recent years, the major technologies in the research on fracture development periods involve (1) geological methods, including analysis of regional construction features, investigation of fractures in outcrop area, outcrop fractures, and fracture characteristic analysis of survey underground (lee, 2001; espinosa, 1998); (2) lab test methods, including filling isotope analysis, inclusion thermometry, and rock acoustic emission testing (lopezpuente et al., 2005; tasdemirci et al., 2012; lankford, 2004; lausund et al., 2015), as well. this article study is mainly focused on the core of fracture filling composition and mutual cut features, fracture filling, the homogenization temperature of hydrodynamic inclusions, and rock acoustic emission methods to explore the volcanic fracture formation period of carboniferous in zhongguai area. the fracture character of core reflects the fracture stage the primary approach to explore the fracture stage involves the core fracture azimuth, the filling ingredients, and the interaction cutting relationship among the fractures to analyse the fracture formation stage. results from the fracture filling material composition and the interaction cutting relationships (figure 3a and 3b) combined with an image logging analysis show that the carboniferous volcanic rock fracture at the zhongguai area are mainly divided into four stages. as such, the first stage of fractures is close to both ew (85°± 10°) and nw west (300°± 15°), with the fillings mainly consisting of calcite and chlorite. the second fracture is ne (45°± 15°). its fillings are primarily gypsum. the third one appears at both ew (85°± 10°) and nw west (300°± 15°), whose fillings mainly consist of calcite. finally, the fourth phase primarily corresponds to both sn (0°± 10°) and nw west (300°± 15°) to fracture, with no filling practically. homogenization temperatures of fracture filling inclusion in recent years, the analysis of fracture filling inclusion lies in a mature technology allowing researching features of fracture filling inclusion and measuring the homogenization temperature. it is worth mentioning that the temperature filling formation can be obtained to determine both the formation period and the mineral filling stage (johnson & cook, 1985). the fourth stage is almost not filled. the present study is focused on 8 samples and 46 test points of fracture fillings from the first three different cutting stages to test the homogenization temperature, according to the various stages of the temperature distribution of fracture fillings and their morphological characteristics. the three filling times of tectonic fractures in that area can be distinguished in the following terms. the homogenization temperature of the first phase ranges from 75 to 82 °c. the second homogenization temperature lies in the range 92-97 °c, with the third homogenization temperature ranging from 106 to 112 °c. there are, at least, four stages of tectonic fractures under the assumption of fractures from the last period with almost no fillings (figure 4). acoustic emission experiment of rocks reflects the fracture formation period the formation of rocks experienced stress effects in the long run with a memory function of stress. in the early sixties, it was confirmed by goodman that rocks also exhibit a "kaiser" effect. the acoustic emission test is supported by the fact that there exists a micro griffith fracture generated by the ancient tectonic stress field in the underground rock stratum. only (a) 596 well, 2398. 5m, high-angle shear fracture (b) k007 well, 3072. 2 m, half-filled high-angle figure 2. characteristics of volcano rock reservoir fractures in the zhongguai uplift carboniferous development. (a) 598 well, 3129. 4 m, fractures cutting cross each other (b) 596 well, 3128. 8 m, fractures cutting cross each other figure 3. core fracture filling characteristics and cutting relationships. 172 cunhui fan, qirong qin, feng liang, zenghui fan & zhi li significant influence in the improvement of both the carboniferous volcanic reservoir property and hydrocarbon accumulation. as such, drilling in the fracture development areas will result in better oil testing results. fractures on the longitudinal direction are often more extended, though the fracture interval on the lateral is larger and the drilling rate of the vertical well is low. therefore, directional drilling is typically used in the drilling process to improve the fracture rate of the drilling. moreover, both the fracture system and the period in that area have been analyzed. it holds that the development situation and the lithology (tectonic stress, tectonic position, and thickness) are closely related. the fracture opening at different periods and distinct ranges in the underground reservoir are inconsistent, the difference of permeability is significant, and it is closely associated with the field direction of the current stress. the present principal stress orientation is close to ew (270°± 10°) with the degree of open fractures being parallel to the predominant stress orientation closely to ew (85°±10°), that is the highest underground. the fracture permeability of that group is the largest, with an effective permeability of 43. 2×10-3 μm2. secondly, the average permeability value of fractures that present a sharp angle with the principal stress orientation of nw (300°±15°) and ne (45°±15°) is 18-24×10-3 μm2. the opening and permeability of sn (0°±10°) are the lowest, with a permeability value equal to 2-10×10-3 μm2. further, the hydrodynamic seepage velocity in the fracture closely to ew (85°±10°) in the early development is the fastest, whereas the hydrodynamic seepage velocity in the fracture closely to sn (0°±10°) is the slowest. notice that both the hydrodynamic seepage velocity and the dynamic response in the nw (300°±15°) and ne (45°±15°) fractures lie between the corresponding values of ew (85°±10°) and sn (0°±10°) fractures. thus, the fracture permeability anisotropy of different groups in the carboniferous volcanic rock reservoir is evident. accordingly, the effect of fractures closely to ew (285°±10°) should be considered firstly in the deployment of development scheme of volcano rock reservoir in this area with the aim to systematically figure 4. temperature distribution chart of fracture filling inclusion at the zhongguai area. if the force exceeds the stress intensity, the fracture can extend and lead to a plurality of different kaiser effect points. more recently, the continuous evolution of technology expands its applications to fault evolution history, fracture stage by matching and ensuring the intensity of fracture formation stress fields (kresse et al., 2013; meyer & bazan, 2011). the present study consists of an acoustic emission experiment regarding the kaiser effect involving 12 pieces of carboniferous volcanic rocks from the zhongguai area. the experimental results suggest that our sample generally presents five obvious kaiser effect points. one of them stands as a result of the tectonic stress effect, whereas the others highlight that rock experienced four micro-fracture events along its geological history. combined with regional tectonic evolution characteristics and research results, it was observed that the first fracture development stage, a product of the middle hercynian tectonic movement, with a paleotectonic stress of 49.6 mpa. the second fracture development stage is a product of the late hercynian tectonic movement and presents a paleotectonic stress of 39.9mpa. the third fracture development stage is a product of the indosinian tectonic movement, whose paleotectonic stress is 34.1 mpa. finally, the fourth fracture development stage is a product of the yanshan tectonic movement, and its paleotectonic stress is 26.5 mpa (figure 5). seepage characteristics of structural fractures fractures in low porosity and low permeability reservoirs are crucial to reforming the reservoir physical property and also concerning the accumulation of oil and gas. fractures tend to control the hydrodynamic seepage network of reservoirs (wu, 2015). from the carboniferous porosity-permeability relationship diagram in the zhongguai area, it can be observed that the correlation between porosity and permeability is not good. the correlation coefficient is only 0.4257 (figure 6), with the difference between the porosity of the development samples and the porosity of the fissureless samples being quite poor. however, the permeability of the fracture development samples stands from 3 to 5 orders of magnitude higher than the permeability of the fissureless samples. for instance, the permeability of the vast majority of the fissureless sample at the work area lies in the range going from 0.01×10-3 μm2 to 1×10-3 μm2. the fracture development samples, such as andesite cores at 3366 m, developed fractures in 598 wells. the penetration rate in the tested samples is as high as 354.14×10-3 μm2, illustrating that the presence of fractures has a figure 5. carboniferous volcano rock acoustic emission ae evolution at the zhongguai area. figure 6. reservoir porosity and permeability correlations of the carboniferous volcanic reservoir at the zhongguai area. 173fractures in volcanic reservoirs: a case study in zhongguai area at the northwestern margin of junggar basin (china) evaluate distinct tectonic fracture seepage characteristics. in various stages of oil and gas field development, it holds that the dynamic changes of fracture permeability in different groups are separate from one to others due to the fracture permeability and seepage velocity. in this way, both the fracture opening and permeability closely to ew (285°±15°) is the greatest. drop speed of hydrodynamic pressure and fracture opening within the reservoir is faster in the process of oil well production. such a group of fracture seepage effect on oil and gas weakened gradually, while descent speeds of the hydrodynamic pressure and fracture opening of fractures in nw (300°±15°), ne (45°±15°), and sn (0°±10°) fall slower than the one closest to ew (285°±15°). with the oil field development process, the dynamic response of fractures in nw (300°±15°), ne (45°±15°), and sn (0°±10°) will be more and more evident, allowing to adjust the well's pattern. the fractured volcanic reservoir exhibits less sensitive mineral and fractures developments. to confine pressure changes of fractures in underground reservoirs will arouse fracture opening and permeability changes. also, its reservoir sensitivity presents typical characteristics of stress sensitivity, being this one of the critical factors leading the volcano rock fractured reservoir yield decrease rapidly. the stress sensitivity refers to the degree of change of the rock permeability caused by the effective stress change in the rock skeleton. volcano rocks have the priority since their body deformations belong to the dense medium. the calculation of the effective stress can be performed throughout the following expression from the double effective stress theory: σ σ ϕeff p p= − in the formula above, seff p denotes the effective primary stress,  is the overburden pressure, p is fluid pressure, and  denotes the porosity. due to the different rates of fracture occurrence and degree of development, the fractured reservoir stress sensitivity changes. in this way, the considerations that follow should be stated. a total amount of 12 different fracture occurrences of andesite and tuff samples in the area were selected, taking advantage of an automatic core displacement system to test stress sensitivity. in both figures 7 and 8, a high-angle fracture stress sensitivity experiment is developed for oblique fractures with a different fracture occurrence and a nonidentical degree of stress change sensitivity. using 20 mpa as the standard, the permeability damage rate of the oblique fractures was equal to 60 % (figure 7). the permeability damage rate of high-angle fracture is less than 40% (figure 8). the obtained results illustrate that the fractures from the oblique fracture samples are easily closed when the confining pressure drops. also, the higher the core loses its permeability, the stronger it is stress sensitivity. on the other hand, the fractures from the high-angle samples are hard closed when the confining pressure drops. as such, the permeability of the core loss becomes smaller as long as the stress sensitivity is relatively weak. visibly, for fractured volcano rock reservoir, the pressure change in the process of exploitation may cause severe damage to the reservoir permeability. accordingly, to be focused on targeted measures of exploitation results entirely necessary in the process of reservoir development in fractured volcano rocks. conclusions the main conclusions in this paper are stated below. (1) regarding the structural fractures of carboniferous volcano rock reservoir being developed at the zhongguai area. fractures constitute a crucial factor regarding space and seepage channels in volcanic rock reservoirs. they are mainly of oblique and high-angle-shear types. the locations of such a fracture development are close to ew (270°± 10°), nw (300°± 15°), ne (45°± 15°), and sn (0°± 10°). it is worth mentioning that these intertwine with each other leading to a fracture network system, which is favorable for oil and gas accumulation and high production. (2) the carboniferous volcanic rock fracture at the zhongguai area could be divided into four main stages according to the core observation and the temperature of homogenization of the fracture filling inclusion together with the regional tectonic evolution features. (3) structural fractures play a crucial role regarding volcanic rock reservoir seepages and accumulation of oil and gas. fracture system more abounds in that area. it has been observed that the opening degree and the fracture seepage effect differ from one area to another for different groups. to properly arrange the development of well's patterns, the impact of fracture seepages should be thoroughly considered. in various stages of the development of oil and gas fields, the dynamic changes of permeability in distinct groups are nonidentical. as such, the well's pattern should be timely adjusted. figure 7. effective stress permeability curve for oblique fractures. figure 8. effective stress permeability curve for high-angle fractures. 174 cunhui fan, qirong qin, feng liang, zenghui fan & zhi li acknowledgments this research was supported by open fund (plc 20180404) of state key laboratory of oil and gas reservoir geology and exploitation (chengdu university of technology) and open fund (pln 201718) of state key laboratory of oil and gas reservoir geology and exploitation (southwest petroleum university). references espinosa, h., dwivedi, s., zavattieri, p., & yuan, g. (1998). a numerical investigation of penetration in multilayered material/structure systems. international journal of solids and structures, 35, 2975–3001. fan, c. & qin, q. (2016). characteristics and formation stages of the reservoir fractures of the xujiahe formation in the fault-fold belt of the central yuanba area of the sichuan basin. journal of the balkan tribological association, 22, 1634-1647. johnson, g. r., & cook, w. h. (1985). fracture characteristics of three metals subjected to various strain, strain rates, temperatures and pressures. engineering fracture mechanics, 21, 31–48. kresse, o., weng, x., & gu, h. (2013). numerical modeling of hydraulic fracture interaction in complex naturally fractured formations. rock mechanics and rock engineering, 46, 555–568. lankford, j. (2004). the role of dynamic material properties in the performance of ceramic armor. international journal of applied ceramic technology, 1, 205–10. larsen, b., grunnaleite, i., gudmundsson, a. (2010). how fracture systems affect permeability development in shallow-water carbonate rocks; an example from the gargano peninsula, italy. journal of structural geology, 32, 1212-1230. lausund, k. b., johnsen, b. b., rahbek, d. b., & hansen, f. k. (2015). surface treatment of alumina ceramic for improved adhesion to a glass fibrereinforced polyester composite. international journal of adhesion and adhesives, 63, 34–45. lee, m., & yoo, y. (2001). analysis of ceramic/metal armour systems. international journal of impact engineering, 25, 819–29. lopez-puente, j., arias, a., zaera, r., & navarro c. (2005). the effect of the thickness of the adhesive layer on the ballistic limit of ceramic/metal armours. an experimental and numerical study. international journal of impact engineering, 32, 321–326. meyer, b. r. & bazan, l. w. (2011). a discrete fracture network model for hydraulically induced fractures: theory, parametric and case studies. presented at spe hydraulic fracturing technology conference, the woodlands, texas, usa, 24–26. spe-140514-ms. nelson, r. a. (2001). geologic analysis of naturally fractured reservoirs. boston: gulf professional publishing, 89-94. tasdemirci, a., tunusoglu, g., & guden, m. (2012). the effect of the interlayer on the ballistic performance of ceramic/composite armors: experimental and numerical study. international journal of impact engineering. 44, 1–9. wu, k. & olson, j. e. (2015). a simplified three-dimensional displacement discontinuity method for multiple fracture simulations. international journal of fracture, 193, 191–204. in this study, a statistical analysis is carried out to reveal the current earthquake potential in the eastern anatolia region of turkey. for this purpose, a space-time assessing based on the recent changes in gutenbergrichter b-value, seismic quiescence z-value, cumulative moment, annual probability and recurrence time of earthquakes is achieved. temporal variation of b-value indicates that average b-value decreased from 1.25±0.02 between 2002 and 2010 to 1.08±0.02 between 2011 and 2015. a clear quiescence in the seismic moment in recent years may be a clue foretelling the occurrence of an intense event. annual probability of the earthquakes between 3.5 and 4.0 magnitude levels exhibits a value equal to and larger than 30. recurrence time of the earthquakes has a value between 2.0 and 3.0 years for magnitude level of 6.0, and a value between 6.0 and 7.0 years for magnitude level of 7.0. a significant decrease in b-value and clear quiescence anomalies in z-value at the beginning of 2015 are observed in the several same regions. these areas cover the central anatolia fault zone, malatya and ovacık faults, the southeastern part of the eastern anatolia fault zone, and the junction of the eastern anatolia and dead sea fault zones. in these anomaly regions, a decreasing trend in b-value may be an indicator of stress increases and an increasing trend in z-value may show the quiescence areas before the next earthquake occurrences. as a remarkable fact, these areas in the eastern anatolia region may be interpreted as one of the most likely locations for the major earthquakes in the future. abstract keywords: eastern anatolia region; b-value; z-value; annual probability; recurrence time; earthquake potential; space-time assessing of the earthquake potential in recent years in the eastern anatolia region of turkey issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n2.50889 earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017): 67 75 s. öztürk1,* 1gümüşhane university, department of geophysics, tr-29100, gümüşhane, turkey serkanozturk@gumushane.edu.tr en este estudio se realizó un análisis estadístico para evidenciar el potencial actual de terremotos en la región de anatolia, al oriente de turquía. con este objetivo, se obtuvieron una evaluación espacio-temporal basada en los cambios recientes del valor b en la ley gutenberg-richter, el valor z de inactividad sísmica, el momento acumulado, la probabilidad anual y el tiempo recurrente de los terremotos. la variación temporal del valor b indica que este factor cayo desde 1.25±0.02, entre 2002 y 2010, a 1.08±0.02 entre 2011 y 2015. la clara inactividad en el momento sísmico de los años recientes podría ser una clave predictiva de la ocurrencia de un terremoto intenso. la probabilidad anual de terremotos de magnitud entre 3.5 y 4.0 muestra un valor igual o mayor a 30. el tiempo de recurrencia de terremotos tiene un valor de 2.0 y 3.0 años en la magnitud de nivel 6.0 y un valor entre 6.0 y 7.0 años para la magnitud de nivel 7.0. a comienzos de 2015 se observó una caída significante del valor b y claras anomalías en el valor z de inactividad en varias regiones. estas áreas cubren la zona de la falla de anatolia central, las fallas malatya y ovacık, la parte sureste de la falla este de anatolia, y las zonas de intersección de la falla este de anatolia con la falla del mar muerto. en estas regiones de anomalías, una tendencia a la baja en el valor b podría ser el indicador de un aumento de tensión, y una tendencia al alza del valor z podría mostrar las zonas de inactividad antes de la ocurrencia de próximos terremotos. como un hecho remarcable, estas áreas en la región este de anatolia pueden ser calificadas como unas de las más propensas a sufrir terremotos mayores en el futuro. resumen palabras clave: región este de anatolia; valor b; valor z; probabilidad anual; tiempo de recurrencia; potencial de terremoto. record manuscript received: 28/05/2017 accepted for publication: 11/05/2017 how to cite item ozturk, s. (2017). space-time assessing of the earthquake potential in recent years in the eastern anatolia region of turkey. earth sciences research journal, 21(2). 67 75 doi:http://dx.doi.org/10.15446/esrj.v21n2.50889 evaluación espacio-temporal del potencial de terremotos en los años recientes para la región de anatolia, en el este de turquía se is m o l o g y 68 s. öztürk 1. introduction the space-time assessment of earthquake activity is an important tool in seismicity studies and some valuable results have been provided to the literature by different authors. many researchers prefer to use different tools in practical applications of seismicity analyses, such as physical models, scaling laws and several different seismic parameters such as seismic b-value, fractal dimension, precursory seismic quiescence, annual probability, recurrence time, moment and energy releases. the magnitude-frequency distribution, known as the b-value of gutenberg-richter relation (gutenberg and richter, 1944), is one of the most commonly used as the power-law distribution of earthquakes. estimating the variations of b-value may supply some useful relations between the frequency of earthquakes, seismic moment or energy. the b-value reflects the relative numbers of both large and small earthquakes, and is related to the properties of the seismotectonic structures, and stress distributions in time and space (scholz, 1968). when b-value shows a decrease for a given region, one can make an assessment that there is a possibility of an earthquake occurrence. in addition to the b-value, the evaluation of the seismicity rate changes by considering the precursory seismic quiescence may supply remarkable outcomes. wyss and habermann (1988) made a definition of seismic quiescence phenomenon in the following way: “the seismic quiescence means that an important decrease in the average earthquake activity rate as compared to declustered background activity rate in the same crustal volume may be observed before the occurrences of some main shocks. this decrease in earthquake activity may occur within part, or all of the source volume of the subsequent main shock. also, this decreasing trend may continue to the main shock time, or may be separated from it by a relatively short period of increasing trend in earthquake activity”. since the duration of quiescence depends strongly on tectonic structure and probably on the loading rate, the measurement of earthquake activity rate may be important to reveal the future earthquake potential. these types of parameters have been used in many seismicity studies for different parts of the world and also turkey (e.g., katsumata and kasahara, 1999; cao and gao, 2002; öztürk et al. 2008; polat et al. 2008; joseph et al. 2011; öztürk, 2011; 2015a; öztürk and bayrak, 2012; rehman et al. 2015; negi and paul, 2015; yadav et al. 2015; singh, 2016; tsapanos et al. 2014; 2016). 2. seismic and tectonic structures of the eastern anatolia region it is well-known that the eastern anatolia (ea) area is one of the most seismically active regions in turkey, since many strong and destructive earthquakes have occurred in the past, such as on december, 26 1939 in erzincan (m8.0), august 17, 1949 in elmalıdere-bingöl (m7.1), march 13, 1992 in erzincan (m6.8), january 27, 2003 in tunceli (m6.1), may 1, 2003 in bingöl (m6.4), august 11, 2004 in elazığ (m5.7), january 25, 2005 in hakkari (m5.9), march 12, and 14, 2005 in bingöl (m5.7 and m5.9). in addition, because of the occurrence of three recent earthquakes “mw = 6.0 in elazığ, march 8, 2010; mw = 5.5 in erzincan, september 22, 2011; and mw = 7.2 in lake van, october 23, 2011” -, studies of earthquake hazard potential for this region have become more important. some details of these mentioned earthquakes are also given in table 1. the ea part of turkey has tectonically a very complex structure. the main tectonic regimes in this part of turkey are the eastern anatolia fault zone (eafz), the dead sea fault zone (dsfz), the bitlis thrust zone (btz) and the karlıova triple junction (ktj). the eafz is a transform fault forming parts of the boundaries between the arabian and african plates, and between the anatolian and the eurasian plates (westeway, 1994). this zone is a sinistral strike-slip fault zone and is about 550 km long. the eafz is thought to be a conjugate structure to the north anatolian fault zone (nafz). it meets and forms triple junctions with the nafz and the dead sea fault zone (dsfz), respectively (bozkurt 2001). the dsfz is a sinistral intraplate strike-slip fault zone, approximately north-south direction, and measures approximately 1000 km in length. it separates the arabian plate to the east and the african plate to the west, and it is considered as a plate boundary of transform type (şengör and yılmaz, 1981). the african plate moves northward slower than the arabian plate and the dsfz takes up this dissimilar motion between these plates. thus, the dsfz and eafz meet in a triple junction between the anatolian, african and arabian plates. the bztz is a complex collisional boundary of continent-ocean and continent-continent. the eurasian and arabian plates come into collision along the bztz and this clash results in the uplift of mountains along the suture zone. this boundary reaches to the north of fold-and-thrust belt of the arabian platform and it extends from southeastern turkey to the zagros mountains in iran (şengör and yılmaz, 1981). a north-south compressional tectonic regime is dominant in the area to the east of the ktj. this region is characterized by two combined strike-slip faults of sinistral and dextral character which is parallel to the nafz and eafz (bozkurt, 2001). compressional basins with east-west trending give shape to the most spectacular structures in the region. however, the major tectonics of ea are formed by the conjugate strike-slip fault systems. table 1. a list of some great and destructive earthquakes occurred in the eastern anatolia region of turkey. figure 1. a) main tectonic structures from şaroğlu et al. (1992) and bozkurt (2001). names of the faults: eafz: eastern anatolia fault zone, neafz: north east anatolia fault zone, çfz: çobandede fault zone, akf: aşkale fault, ezf: erzurum fault, kf: kağızman fault, af: ağrı fault, if: iğdır fault, dfz: doğubeyazıt fault zone, bfz: balıklıgölü fault zone, çf: çaldıran fault, erf: erciş fault, htf: hasan-timur fault, bf: başkale fault, ysfz: yüksekova-şemdinli fault zone, sf: süphan fault, mf: malazgirt fault, sufz: sancak-uzunpınar fault zone, gfz: göynük fault zone, pf: pülümür fault, bkkf: bingöl-karakoçan fault zone, gf: genç fault, mtz: muş thrust zone, kbf: kavakbaşı fault, kez: karacadağ extension zone, bzf: bozova fault, srf: sürgü fault, mlf: malatya fault, of: ovacık fault, cafz: central anatolia fault zone. b) epicenter distributions of whole catalog including 33,865 shallow earthquakes with md≥1.0 between 1970 and 2015 for the eastern anatolia part of turkey. magnitude distribution of earthquakes are represented with different symbols. 69space-time assessing of the earthquake potential in recent years in the eastern anatolia region of turkey figure 1a shows the principal faults in the ea. a large detail on the tectonic structure of the ea can be found in some studies such as şengör and yılmaz (1981), hempton (1987) and bozkurt (2001). also, there are several studies (e.g., sayıl 2005; bayrak et al. 2008; öztürk, 2009; öztürk and bayrak, 2012; tsapanos et al. 2014; öztürk 2015b; maden and öztürk, 2015) for turkey including the ea. these studies generally consider the regional and temporal variations of seismotectonic b-value, seismic quiescence z-value, and some other statistical parameters for the analysis of earthquake behaviors, and their possible usage as precursors. for this reason, seismic and tectonic assessments of temporal changes in b and z-values, cumulative seismic moment, annual probability and recurrence time of the earthquakes are evaluated in this study and it is aimed to present some useful clues for the future earthquake potential in the ea. 3. earthquake database and brief description of methods of analysis the earthquake database used in the scope of this analysis is obtained from bayrak et al. (2009) for the time interval of january 6, 1970 to december 31, 2005. in addition, the catalog from january 1, 2006 to december 31, 2014 is prepared using the database of bogazici university and kandilli observatory and research institute (koeri). in total, a catalog of 33,865 earthquakes (original catalog) whose magnitude is equal to or greater than 1.0 with a depth <70 km is prepared between the coordinates 36°n and 42°n latitude, and 36°e and 45°e longitude. duration magnitude, md, is used in analysis and epicenter distributions of all 33,865 earthquakes with main tectonics drawn from şaroğlu et al., (1992) and bozkurt (2001) are plotted in figure 1b. in order to decluster the earthquake catalog into main and secondary events, reasenberg’s (1985) algorithm is used in this study because the exclusion of dependent events from the catalog is an important part of an earthquake hazard analysis. all dependent events are separated from independent ones, and these dependent events are substituted with a unique event by removing each cluster. in all, 9,515 events (nearly 28 %) were removed by the declustering method, and the declustered catalog of 24,350 earthquakes is obtained. also, completeness magnitude, mc, is also a very important parameter for many seismicity studies, and temporal changes in mc-value can affect the results of the seismicity parameters, especially in b-value. thus, the maximum number of earthquakes in the catalog is tried to be used for high-quality results for the analysis of all statistical parameters. average mc-value for the earthquake catalog from 1970 to 2015 is estimated as 2.9. there are totally 33,865 shallow earthquakes with magnitudes larger than or equal to 1.0 in the catalog. as stated above, the reasenberg’s declustering algorithm eliminated 9515 earthquakes. after the estimation of mc-value as 2.9, the earthquakes with md<2.9 are removed from the original catalog and the number of earthquakes with md<2.9 is 13,793. totally, about 68.83% of all events are subtracted from the original catalog and the number of events for z-test is reduced to 10,557. 3.1. gutenberg-richter relation, b-value gutenberg and richter (1944) proposed an empirical relation for the frequency-magnitude distribution of earthquake occurrences. this power-law for the size distribution of earthquakes was given in the following form: log 10 n (m) = a bm (1) where n(m) is the expected number of events in a specific time interval with magnitudes greater than or equal to m, and a and b are constants. the a-value is proportional to the seismic activity rate, and variations in a-value depend on the length of the study area, observation period, and also number of events. the b-value is the slope of the magnitude-frequency distribution and b-value can be thought as one of the most important parameters for geotectonic and rheological characteristics in a given region. according to the results of laboratory experiments for rock fracture, a decrease in the confining pressure and an increase in the applied shear stress cause a reduction in b-value (scholz, 1968). however, an increase in the thermal gradient, fracture density, or material heterogeneity in the geological complexity (mogi, 1962) is related to high b-values. many studies show that b-value of g-r is a scale invariant and it is related to the clustering and spacing properties of epicenters or distribution of fault segments. one of the most fundamental observations in earthquake physics is the gutenberg-richter (g-r) relation. ishimoto and iida (1939) and gutenberg and richter (1944) suggested the linear equation (1) and this equation is also called as frequency-magnitude relation. it plays an important role in the formulation of realistic design earthquakes for a given region. in many cases, g-r relation for one and the same region or time interval follows the different regimes for small and large events (olsson, 1999). an estimation of the maximum b-value in g-r relation was carried out by olsson (1999) and the deduced maximum b-value was given as 1.5 (or with the error limits, between 1.30 and 1.64). however, utsu (1971) suggested that b-value varies from 0.3 to 2.0 from place to place in each seismic region in the world. the choice of the statistical method used for to estimate the b-value is an important and a critical point. several algorithms can be used to estimate the b-value, but the maximum likelihood method by aki (1965) is the most widely used and the most robust method to estimate the b-value. traditionally, the b-value was estimated graphically, but more efficient estimation is carried out by the maximum likelihood method. it was derived from the quality of the first order moment, and later aki (1965) showed that the maximum likelihood estimate maximize the likelihood function. thus, it is given as in the following form: (2) where mmean is the mean magnitude of the earthquakes, and mmin is the minimum magnitude of completeness in the earthquake catalog. a numerical value of 0.05 is a correction constant. correct estimates of local variations in mmin can be done if relatively large numbers (100 or so) of local observations are available for analysis. confidence limits of 95 % on the calculations of b-value are , where n is the number of events used to make the estimation. this calculation gives the confidence limits of ±0.1-0.2 concerning b-value for a characteristic sample of n=100 earthquakes. thus, the number of earthquakes in this work was 24,350 and the value of n=100 events is an ordinary sample for calculations and represent a specific calculation. 3.2. precursory seismic quiescence, z-value the phenomenon of precursory seismic quiescence was originally committed by wyss and habermann (1988). wiemer and wyss (1994) introduced a methodology that can be performed in zmap software (wiemer, 2001) and the recognizing of the seismic quiescence for different parts of the world has been reported in many papers. so, a brief description of the method will be given here. a continuous image of rate changes in the earthquake activity and plotting the areas showing seismic quiescence in space and time can be provided by zmap by plotting in geographical coordinates. the standard normal deviate z-test is one of the most common technique generally used for detecting of precursory seismic quiescence. in this study, z-test is applied to detect the anomaly regions of precursory quiescence and log term average (lta) function is generated for the statistical assessment of confidence level in standard deviation units: (3) where r1 is the mean rate in the overall background period, r2 is the average earthquake activity rate in the foreground window, s1 and s2 are the standard deviations in these periods, and n1 and n2 are the number of 70 s. öztürk samples with a measured earthquake activity rate. z-value is estimated as a function of time and named lta. 4. results the cumulative number of earthquakes with time both the original catalog including all shallow earthquakes with md≥1.0 (33,865 events) and declustered data including the earthquakes with md≥2.9 (10,557 events) is shown in figure 2. any significant seismic activity is not observed from 1970 to 1995 and there is a little variation between 1995 and 2000. conversely, there is a significant earthquake activity after 2000. however, it can be said that the database is separately homogeneous between 1970 and 1995, between 1995 and 2002, and between 2002 and 2015. koeri and the other observatories have provided the real time data, especially after the construction of many stations in this area in recent years. since this analysis focuses on the recent earthquake activity, completeness magnitude can be taken as 2.9 for all catalog as given in öztürk and bayrak (2012). some authors such as katsumata and kasahara (1999), and joseph et al. (2011) made detailed analysis on earthquake behaviors and they pointed out that the selection of the completeness magnitude and declustering of the catalog are an important stage for the analysis of earthquake occurrences. katsumata and kasahara (1999) made a seismic quiescence analysis for the 1994 kurile earthquake. they made a completeness analysis and they used the declustered the catalog for their statistical estimations. joseph et al. (2011) stated that just the main earthquakes must be considered in the recurrence analysis of earthquakes and all dependent events such as foreshocks and aftershocks must be excluded from the catalog before starting any calculations. as shown in figure 2, the slope of cumulative number curve of the declustered data is smoother than that of the original catalog. it is a remarkable fact that these two processes removed the dependent events from the original data set. as an important result, after completing of the declustering process and completeness analysis, a more reliable, homogeneous and robust earthquake data is provided for the statistical estimations. magnitude and time histograms of the study region are given in figure 3. as stated above, magnitudes of the earthquakes are between 1.0 and 6.6 and show an exponential decay in their numbers from the smaller to larger magnitudes. as shown in figure 3a, the size of the many earthquakes varies from 2.0 to 4.0 and a maximum is observed for md=2.8. the number of earthquakes with 2.0≤md<4.0 are 30,591. however, there are 2119 evens with 1.0≤md<2.0, 20,149 events with 2.0≤md<3.0, 10,442 events with 3.0≤md<4.0, 1074 events with 4.0≤md<5.0, and 81 events with 5.0≤md. thus, the earthquake occurrences with magnitudes between 2.0 and 4.0 are more dominant than those of the others in the study region. it is clear that this increasing tendency in the number of relatively small events may be a sign of increasing stress in recent years in the ea. also, time histogram from 1970 to 2015 is given in figure 3b. there is a little change in the earthquake activity between 1970 and 2002, and the number of earthquakes including all magnitude levels in this time interval is 2191. there is also a stableness between 2003 and 2009, and the number of earthquakes in this period is 8879. however, an increasing trend in the number of earthquakes have started after 2010 and it has continued until 2012. the number of earthquakes between 2010 and 2012 is 15,128 and the maximum increasing is reported in 2012. the systematic increase in seismic activity shows a decreasing trend after 2012, and the number of earthquakes between 2012 and 2015 is 7667. as a remarkable fact, such kind of analyses can supply a useful perspective for the evaluation of seismicity rate changes which is associated with the temporal variations of precursory seismic quiescence in the ea. figure 4 shows the annual probabilities and recurrence times for different magnitude levels. annual probabilities of earthquake occurrences for different sizes show relatively higher (30 and above) values between 3.5 and 4.0 magnitude levels, intermediate values (between 1 and 10) from 4.5 to 5.5 magnitude levels, and quite smaller values (<1.0) between 5.5 and 7.0 magnitude levels. recurrence times of earthquake occurrences for different magnitude levels are also shown in figure 4b. it is observed quite smaller (<1.0) years for magnitudes from 3.5 to 5.5. for magnitude level of 6.0, the values between 2.0 and 3.0 years are estimated while the values between 6.0 and 7.0 years are estimated for magnitude level of 6.5. it can be concluded that the occurrences of earthquakes between 3.5 and 4.0 magnitude level are more likely than those of the others, and an earthquake occurrence for a magnitude size larger than 6.0 can be expected in every seven years. these results can also be seen from table 1 and the results of the recurrence times analysis support the existing earthquake potential in the ea. consequently, such kind of analyses on the probabilities and recurrence times of earthquake occurrences for specific magnitude levels show that the ea has an earthquake risk potential for the possibility of strong earthquake occurrence. figure 2. temporal changes of the cumulative number of earthquakes for the original catalog including all md≥1.0 earthquakes (33,865 events) and for the declustered data set with md≥2.9 (10,557 events). figure 3. a) magnitude histogram and b) time histogram of the earthquake activity between 1970 and 2015. 71space-time assessing of the earthquake potential in recent years in the eastern anatolia region of turkey figure 4. a) annual probability and b) recurrence time of the earthquakes for different magnitude levels in the ea region. figure 5. gutenberg-richter relation and b-value with its standard deviation. declustered catalog is used for calculations and mc-value is also given. figure 6. changes in b-value and its standard deviation (dashed line, δb) as a function of time, from 1970 to 2015. thick horizontal lines show the average b-values, and the upper and lower thin horizontal lines shows their standard deviations. declustered catalog is used for analysis. as stated above, the maximum likelihood method is preferred to estimate the b-value since this algorithm gives a more robust estimate than the least-square regression method (aki, 1965). figure 5 shows the cumulative number of events versus the magnitude of earthquakes. as shown in figure 5, average mc for the whole catalog from 1970 to 2015 is taken as 2.9 and b-value is estimated as 1.02±0.01. on global scale, b-value varies from 0.3 to 2.0 depending on region (utsu, 1971) and is equal to 1.0 on average (frohlich and davis, 1993). as a result, both the mc and b-value estimations in this study are similar to the results of the analysis by öztürk and bayrak (2012), and the magnitude-frequency distribution of the earthquakes in the ea is well represented by the g-r power law distribution with the b-value close to 1.0. changes in b-value as a function of time are plotted in figure 6. temporal variations of b-value are estimated for overlapping samples of 350 earthquakes per window. a systematic decrease can be seen until 1996, and the average b-value from 1970 to 1995 (464 earthquakes) is calculated as 1.28±0.1 with mc=4.3. the b-value does not show any significant changes between 1996 and 2001 (1,168 earthquakes), and the average b-value is found to be 0.99±0.04, a lower value compared to that of the previous period, with mc=3.4 for this period. however, there are great fluctuations and a clear increase in b-value between 2002 and 2010 (9,067 earthquakes), and the average b-value is computed as 1.25±0.02 with mc=2.9 for this time interval, a larger value that that of the previous period. for the last time interval of 2011-2015 (13,651 earthquakes), there is a clear decrease in b-value and the average b-value is estimated as 1.08±0.01 with mc=2.5. as seen in figure 6, there is a clear increase from 2002 to 2011, whereas there is a clear decrease after 2011. in addition, one can see many fluctuations which are related to the period before the occurrence of large earthquakes between 2002 and 2015. there are 81 events whose magnitude is larger than or equal to 5.0 (figure 1b), and the b-value shows a clear tendency of decrease before some of these strong main shocks in this time interval. öztürk (2011) observed these types of decrease before some major earthquakes, such as the 1999 i̇zmit, 2003 tunceli and 2003 bingöl earthquakes. as mentioned above, many variables can affect the variations of b-value. it is thought that the decrease in b-value before the main shocks may be the result of increasing shear stress, and thus this tendency of decrease in b-value may be interpreted as an indicator of a possible earthquake in the ea. figure 7 shows the magnitude variations of earthquakes as a function of time. in order to see the fluctuations in magnitude levels in the ea from 1970 to 2015, temporal distribution of earthquakes with md≥4.0 is plotted. as shown in table 2, the average magnitude is 4.49±0.33 for the time interval between 1970 and 1995, 4.29±0.22 between 1996 and 2001, 4.35±0.31 from 2002 to 2010, and 4.34±0.32 from 2011 until 2015. there are 22 events whose magnitude is equal to or larger than 5.0 (the largest magnitude among them is 6.5) between 1970 and 1995; 9 events from 1996 to 2001 (the largest among them is 5.5); 28 events from 2002 until 2010 (the largest has a magnitude of 6.2); and 22 earthquakes between 2011 and 2015 (the largest magnitude is 6.6). magnitude-time histogram shows significant variations in the number of large events from 1995 but further on especially after 2002. the earthquake activity which is related to the clustering characteristics is obviously observable and it may correspond to a main earthquake in study region. also, temporal clustering properties of the earthquake activity related to the earthquakes with md≥5.5 is strong enough for many earthquakes which occurred in 1975, 1976, 1983, 1992, 1995, 1997, 2001, 2002, 2003, 2004, 2005, 2007, 2010, 2011. figure 7. temporal changes in earthquake magnitudes with md≥4.0 including 1155 events between 1970 and 2015. whole earthquake catalog is used for plotting. thick horizontal lines show the average magnitude and the upper and lower thin horizontal lines shows standard deviations for these periods. 72 s. öztürk figure 8. cumulative moment release as a function of time from 1970 to 2015. arrows indicate the occurrence of large earthquakes in study region. whole earthquake catalog is used for analysis. table 3. cumulative moment increases after some great earthquakes. figure 9. regional variations of temporal b-value changes between 2002-2010 and 2011-2015 in the eastern anatolia region of turkey. white dots show the declustered earthquakes used in the analysis. main tectonics are also given. figure 10. regional variations of z-value in the beginning of 2015 with t w (iwl)=5.5 years in the eastern anatolia region of turkey. white dots show the declustered earthquakes with md≥2.9 used in the analysis. major tectonics are also given. table 2. several statistics for the earthquakes with m≥4.0 between 1970 and 2015. cumulative moment variations (cmv) versus time from the declustered catalog are plotted in figure 8. cmv is calculated as about 1.36*1026 nm for the 1992 erzincan, cmv~1.86*1026 nm for the 2003 tunceli, cmv~2.26*1026 for the 2003 bingöl, cmv~2.43*1026 for the 2005 bingöl, cmv~3.06*1026 for the 2010 elazığ, and cmv~5.93*1026 for the 2011 lake van earthquakes (table 3). high values of moment release represent a high seismic energy release, and consequently the occurrence of strong earthquakes. it can be interpreted that temporal and regional changes in seismic moment are partly due to the variations in tectonic stress. the stress conditions in a given region after the occurrence of a strong earthquake will change, so the cumulative moment is expected to increase in proportion to the great energy released as the consequence of stress modification produced by the dynamic ruptures of the main shock. the total stress drop and cumulative moment depending on these conditions are associated with the seismicity of the focal region (marcellini, 1995). as shown in figure 8, sudden and high increases in cumulative moment are observed at the time when the main shocks occur. these large variations in cumulative moment are also in accordance with the temporal variations of magnitude as given in figure 7, and there are clear uniformities in moment variations before the occurrence of great earthquakes as seen in figure 8. this means that there are constant moment values in the period in which seismic activity is stationary, so it can be said that variations in cumulative moment are related to the changes of seismic activity. it can be seen from figure 8 that there is a clear quiescence in the temporal variation of seismic moment after 2011, and thus if these kinds of behavior in the seismic moment can be described as a reliable indicator of future earthquakes, then these variations can give us significant evidence about the forthcoming earthquake hazard potential in the ea. regional variations in b-value as a function of time in the ea are given in figure 9. this figure shows the changes of b-value between 2002-2010 and 2011-2015. in order to map these variations, the study region is divided into spatial grids of points with a size of 0.1° in latitude and longitude. there are six regions exhibiting large decrease (between -0.4 and -1.0 units) in b-value. these regions are centered at 39.03°n-37.19°e (region a, covering the cafz, mlf and of), at 39.46°n-41.80°e (region b, the southwest of af and kf), at 38.96°n-42.50°e (region c, the north of lake van, including mf), at 36.85°n-38.64°e (region d, the south of bzf), at 36.84°n-37.72°e (region e, the southeast part of the eafz), and at 36.35°n-36.29°e (region f, in the junction of the eafz and dsfz). regional variations in z-value in the ea are shown in figure 10. this demonstration supplies the spatial image of z-value estimation at the beginning of 2015. in order to establish the distribution of precursory seismic quiescence, the study region is divided into spatial grids of points with a size of 0.1° in longitude and latitude. there are several areas exhibiting seismic quiescence at the beginning of 2015. these anomaly areas are observed on the north western part of the nafz, between akf and ezf, on the pf, bkkf, sufz and gfz, in and around the gf, between bf and ysfz, on a part of the eafz, in the eastern and western parts of the mlf, in the southeastern part of the eafz, in the junction of the eafz and dsfz, and in the southern part of the cafz. as shown in figures 9 and 10, there are clear decreases in b-value and seismic activity rate z-value in the several same areas: the cafz, mf, of, the southeastern part of the eafz, and the junction of the eafz and dsfz. as a remarkable fact, such kind of combination of these seismotectonic parameters may supply significant clues to reveal the seismic potential in the ea and thus, special interest needs to be paid to these anomaly regions. 73space-time assessing of the earthquake potential in recent years in the eastern anatolia region of turkey table 4. return periods for some regions of ea part of turkey. these values are taken from öztürk et al. (2008). 5. discussions the ea has many important tectonic faults and fault systems and also is one of the most seismically active parts of turkey. many large and destructive main shocks have occurred in the last five years in the ea and so, the statistically analysis of earthquake behaviors would be important. in recent years, the space-time assessments of different statistical parameters have been made in order to put forth the earthquake hazard potential in the ea (öztürk et al. 2008; bayrak et al. 2008; öztürk, 2009; öztürk and bayrak, 2012; tsapanos et al. 2014; öztürk 2015b; maden and öztürk, 2015). also, several valuable investigations which are focused on the such kind of assessments, especially on b-value and seismic quiescence analysis, have been achieved by different authors for different parts of the world (e.g., cao and gao, 2002; polat et al. 2008; öztürk, 2011; 2015a; rehman et al. 2015; negi and paul, 2015; yadav et al. 2015; singh, 2016; tsapanos et al. 2016). polat et al. (2008) made a statistical analysis for the aegean extension part of turkey in order to evaluate the earthquake potential by using the seismotectonic b-value, fractal dimension dc-value and standard normal deviate z-value. their results show that the regions with smaller b-values and larger z-values may be interpreted as the possible areas for the large earthquakes. öztürk (2011) and öztürk (2015a) made a similar study on the space-time relationships between several seismotectonic parameters such as b-value, dc-value and z-value for the north anatolia fault zone and the western anatolia region of turkey, respectively. in these studies, it is suggested that the correlations between these three seismic and tectonic parameters (for example, the regions with the lower b-value, larger dc-value and higher z-value) may give significant evidences for the future earthquake potential. rehman et al. (2015) made an analysis on the spatial and temporal changes of b-value in and around the north pakistan. they used different histograms, time series distribution of earthquakes and b-value map in order to characterize the seismic behaviors of different parts of pakistan. their results on the temporal distribution of b-value for hazara region show a remarkable decrease before the occurrence of october 8, 2005 kashmir earthquake. negi and paul (2015) made a comprehensive study for the spatial and temporal analysis of earthquake activity for garhwal-kumaun himalaya region. they used b-value and dcvalue in order to explain the spatio-temporal characteristic of seismicity. they stated that the lowest b-value for the south east garhwal indicates high stress conditions and also means that the number of small earthquakes in this region decreases relatively due to the stress accumulation. as a remarkable fact that, they suggested a temporal observation of b-value is necessary for the possible earthquake risk assessment. yadav et al. (2015) achieved a probabilistic estimation for the most perceptible earthquake magnitudes in and around the nw himalaya. they used two earthquake recurrence models of kijko and sellevoll and gumbel’s third asymptotic distribution for different 28 subregions. they stated that spatial distribution of these parameters presents a good relation with high seismic hazard regions in previously reported hazard maps, the maximum expected magnitudes and acceleration. they suggested that such kinds of maps may give useful information in the selection of preparing earthquake criteria for the antiseismic design of non-critical structures. singh (2016) made a spatial analysis of seismic b-value for all nw himalaya segment between 1964 and 2010 in order to understand the tectonic implications. their results show that garhwal-kumaun segment with small b-value has a large seismic potential and indicates a high stress distribution. they suggested that b-value is a one of the most significant parameters for the seismic hazard assessment in a given region since the b-value is directly related to the changes of stress accumulation in the crust. tsapanos et al. (2016) made an assessment of the relative largest earthquake hazard level in 28 subregions in and around the nw himalaya. they used the gumbel’s third asymptotic distribution for the estimation of the largest earthquake in their study region. they suggested that spatial variations of their parameters such as b-value (upper bound magnitude) and k-index (relative seismic hazard level) may give useful clues for scientific goals along with the designation of priority areas for earthquake resistant design. as mentioned above, some authors used different parameters to make an earthquake risk assessment in the ea. an estimation of the earthquake hazard parameters for different parts of turkey is made based on gumbel’s i distribution by öztürk et al. (2008). in their study, a few parameters such as the mean return period, the probable maximum magnitude and the probability of a great earthquake occurrence for some magnitudes and periods are estimated for different 24 regions in and around turkey (table 4). their results show that the mean return period for ms≥5.0 is calculated as 10.23±0.47 years in their region 2 including af and kf in this study. the last earthquake greater than 5.0 was occurred in 2007 in this region. the mean return period for ms≥5.0 was obtained as 13.49±0.63 years for their region 3 covering the lake van, and mf in this study. the last earthquake larger than 5.0 in this area was occurred in 2011. the mean return period for ms≥5.0 was found as 45.71±4.22 years for their region 5 covering the bzf in this study. the last earthquake larger than 5.0 was occurred in 2012 in this zone. for their region 6, covering the eafz in this study, the mean return period for ms≥5.0 was estimated as 11.75±0.54 years. the last earthquake larger than 5.0 in this zone was occurred in 2012. the mean return period for ms≥5.0 was obtained as 10.23±0.71 years for their region 7 covering the junction of the ea and dsfz in this study. the last earthquake larger than 5.0 was occurred in 2002 in this region. the mean return period for ms≥5.0 was calculated as 8.91±0.41 years for their region 22 covering cafz in this study. for their region 23, including mlf and of in this study the mean return period for ms≥5.0 was estimated as 13.49±0.62 years. the last earthquake larger than 5.0 was occurred in 2012 in regions of 22 and 23. the mean return period for ms≥5.5 was found as 7.50±0.09 years for their region 24 including the eastern part of the nafz in this study. the last earthquake with magnitude 5.5 was occurred in 2011 in this zone. taking into consideration the mean return period values for strong events whose magnitudes are larger or equal to 5.0 from öztürk et al. (2008), the areas covering af and kf may have a next earthquake potential between 2017 and 2018. for the other anomaly regions, however, the future time of a strong earthquake can be expected after 2020. bayrak et al. (2008) made a seismicity analysis for different 24 seismogenic zones using gumbel first asymptotic distribution and gutenbergrichter relation. they stated that their estimations of b-values with the maximum likelihood method for the gutenberg-richter relation have a good consistency with the tectonics. they calculated similar b-values for their region 1 (b=0.74, in neafz), region 2 (b=0.76, in af and kf), region 3 (b=0.78, including lake van, and mf) and region 7 (b=0.79, in a part of dsfz). as given in bayrak et al. (2008), these regions include the earthquakes such as 1924 pasinler, 1983 horasan, 1976 çaldıran and 1903 patnos, and these authors stated that these low b-values are related to the low heterogeneity degree and great faults causing destructive earthquakes. they also calculated b-values smaller than 1.0 in their region 4 (b=0.96, in bitlis thrust zone), region 5 (b=0.94, including bzf and kez), region 6 (b=0.91, in eafz). according to the gps (global positioning system) data, there is a total shortening of 10±2 mm/yr between the strike slip faults in eastern turkey and thrusting along the caucasus (mcclusky et al. 2000). also, sinistral motion in the eafz is given as 9±1 mm/yr (mcclusky et al. 2000). because of the seismic activity rate and the fault slip rate, the b-values in regions 1, 2, 3 and 7 are smaller than those of regions 4, 5 and 6 (bayrak et al. 2008). the b-values are estimated as 0.62 for their region 22 (including cafz, 0.58 for their region 23 (covering mlf and of), and 0.68 for their region 24 (including the eastern part of the nafz), respectively. these resulting values show that these areas are not multifractal regions and have a large homogeneity degree (bayrak et al. 2008). also, many researchers such as öztürk (2011) and 74 s. öztürk öztürk (2015a) observed a decreasing trend in the b-value before the occurrence of a large earthquakes. consequently, as seen from the results, the regions with small b-values in their study and the regions showing great decrease in b-value in this study cover the same regions: cafz, mlf, of, af, kf, lake van, mf, bzf, eafz and the junction of the eafz and dsfz. as a remarkable fact that, the data, estimations and anomaly regions in this study are more up-to-date and the similar with those of bayrak et al. (2008). an earthquake hazard evaluation in the ea in terms of seismic quiescence z-value was made by öztürk (2009). in this study, he aimed to estimate the duration of precursory quiescence for the strong earthquakes occurred in ea and used a few strong earthquakes greater than 5.0 between 1970 and 2005. in öztürk (2009), average precursory quiescence duration before a strong earthquake in the ea is estimated about 5.0±1.5 years. then, regional distribution of precursory seismic quiescence in the ea at the beginning of 2009 is analyzed by öztürk and bayrak (2012). they observed four anomaly regions around aşkale (erzurum), ovacık (tunceli), elazığ and bingöl and, van lake. in two of these areas, 2010 elazığ and 2011 lake van earthquakes are occurred, but not aşkale and ovacık. however, a similar anomaly is observed in ovacık region. as a result, this common area in which both seismic quiescence and low b-value are observed may be considered as a potential earthquake zone. therefore, these kinds of seismic behaviors can be an important clue for the region of next earthquake occurrences. öztürk and bayrak (2012), tsapanos et al. (2014), maden and öztürk (2015), and öztürk (2015b) made a detailed evaluation of earthquake hazard potential for including the ea by using different statistical parameters. öztürk and bayrak (2012) achieved a statistical analysis for the eastern part of turkey by using the gutenberg-richter b-value and by applying the standard normal deviate z-test. the b-value is calculated as 1.07±0.09 with a completeness magnitude of 3.0. their results indicate four quiescence regions (with high z-value) at the beginning of 2009 and these anomaly regions are observed around erzurum (aşkale), tunceli (ovacık), elazığ, bingöl and lake van. the b-value estimated in this study (b=1.02 with mc=2.9) is smaller than that of öztürk and bayrak (2012). also, several seismic quiescence anomalies estimated in this study are the same regions with öztürk and bayrak (2012) and these regions cover erzurum, tunceli, elazığ, bingöl cities including akf, pf, bkkf, sufz, gfz, in and around the gf, between bf and ysfz. a significant result of öztürk and bayrak (2012) is that march 8, 2010 elazığ, september 22, 2011 erzincan and october 23, 2011 lake van earthquakes occurred in the seismic quiescence areas which are detected by them. when compared the results in both studies, it is clear that b-value is in a decreasing trend in recent years and this decrease can be also seen in figure 6 in this study. also, such kinds of assessments which is made on the seismicity rate changes can play an important role in the estimation of possible location of next earthquakes and in the earthquake hazard potential. tsapanos et al. (2014) made a study including regional changes of seismicity parameters in and around turkey. for this purpose, they used the gumbel’s third asymptotic distribution of extreme values in order to define the seismicity pattern. they suggested that there is a good relation between their seismicity parameters and the tectonic structures for turkey and surrounding area. they calculated a high value of upper bound magnitude w-value in the eastern part of turkey. according to their earthquake hazard maps (figures 2 and 3 in their study), the larger w-values greater than 7.0 can be seen from the geographical distribution of the parameter of the 50-years mode in the nafz, in the ea and in the western part of turkey. they related these large and very large values with the seismotectonically active zones and, they stated that large values in the eastern part of turkey are related to the btz. consequently, the higher w-values estimated by tsapanos et al. (2014) in the ea cover the anomaly areas of b-value and z-value found in this study. spatial distribution of b-value for the ea is also mapped by maden and öztürk (2015). they searched the relations between b-value, bouger gravity and heat flow data for tectonic implications. they calculated the b-value as 1.11±0.08 with a completeness magnitude of 2.8. one can clearly see that calculated b-value in this study (b=1.02 with mc=2.9) is smaller than that of maden and öztürk (2015). so, it can be said that b-value decreases at the beginning of 2015 (see figure 6). according to their spatial results, low b-values are shown in and around çfz, af, kf, the north, south and east of lake van, the west of akf, on the eafz and btz, between mlf-sf-bzf, between mtz and kez, in the south east part of the eafz, and in the junction of the eafz and dsfz. thus, the areas having small b-value estimated by maden and öztürk (2015) and the areas exhibiting great decrease in b-value in this study include several same regions: af, kf, the north of lake van, in the south east part of the eafz, and in the junction of the eafz and dsfz. such decreases can imply an increase in stress relaxation in the mentioned parts of the ea. by contrast, relatively small variations in b-value during the post-2011 epoch may be related to a relatively small stress distribution in which energy can be spent along with time during the earthquakes. thus, the earthquake dataset and anomaly regions of b-value in this study are more up-to-date and the similar with those of maden and öztürk (2015). öztürk (2015b) made a comprehensive statistical analysis for the ea. for this purpose, a modelling on the fractal dimension of earthquakes and intermediateterm forecasting for the location of expected earthquakes were achieved. the ea was divided into 19 seismogenic zones for a detailed evaluation. for each region, b-values and dc-values were estimated and a relationship between seismotectonic b-value and fractal dimension dc-value was determined by comparing different regression methods. it is also generated a composite forecast map based on the relative intensities (ri method) and pattern informatics (pi method). according to the results of öztürk (2015b), small b-values and large dc-values are observed in the neafz, af, kf, dfz, bfz, çf, bf, erf, sf, btz, bkkf, sufz and the eastern part of the nafz. according to the composite forecast map by öztürk (2015b), several hazardous regions were identified for the possible locations of the next earthquake occurrences: in and around akf, the west part of the lake van (between sf and erf), around ysfz, around and in the north part of of, on the eafz, on a part of nafz (between mtz and pf). when compared the results of this study and öztürk (2015b), it can be clearly seen that the results of spatial distributions of b-value and z-value in this study are suitable with the results of öztürk (2015b). 6. conclusions a space-time assessing in the eastern anatolia part of turkey is achieved by evaluating several seismotectonic parameters. seismic activity in this region show an important increase in recent years and 50 earthquakes with md³5.0 occurred after 2002. average value of completeness magnitude is calculated as 2.9 and the b-value is estimated as 1.02±0.01. important fluctuations and increases are observed in b-value (1.25±0.02) between 2002 and 2010. however, a clear decrease is observed in b-value from 2011 to 2015 and the b-value is calculated as 1.08±0.01. cumulative moment releases are estimated as about 1.36*1026 nm after the 1992 erzincan, ~1.86*1026 nm after the 2003 tunceli, ~2.26*1026 after the 2003 bingöl, ~2.43*1026 after the 2005 bingöl, ~3.06*1026 after the 2010 elazığ, and ~5.93*1026 after the 2011 lake van earthquakes. however, a quite clear quiescence in temporal changes of cumulative seismic moment is detected after 2011. an increasing trend in the number of earthquakes until 2012 tends to decrease after 2012. analyses of probability and recurrence time of the earthquakes suggested that the eastern anatolia region of turkey has an earthquake risk potential for the probability of large earthquake occurrence in the future. an important decrease is observed in temporal variations of b-values for the time interval of 2011 to 2015 in comparison with those for the time period of 2002 to 2010. also, there are several regions exhibiting seismic quiescence at the beginning of 2015. the regions having remarkable decreases in b-value and z-value at the beginning of 2015 cover the central anatolia fault zone, malatya and ovacık faults, the southeastern part of the eastern anatolia fault zone, and the junction of the eastern anatolia and dead sea fault zones. as a remarkable fact, the decreases in the number of earthquakes after 2012 support the temporal variations of b-value and seismic quiescence z-value. consequently, a decrease in the b-value, seismic quiescence anomalies, recurrence times of earthquakes and seismic moment variation may be an indicator of high stress release, and these changes may be interpreted as an evidence of strong earthquake occurrences in the next future in the eastern anatolia region of turkey. 75space-time assessing of the earthquake potential in recent years in the eastern anatolia region of turkey acknowledgements the author would like to thank to the reviewers for their useful and constructive suggestions in improving this paper. i thank to koeri for providing free earthquake database via internet. references aki, k. (1965). maximum likelihood estimate of b in the formula log n= a bm and its confidence limits. bulletin of the earthquake research institute, tokyo university, 43, 237-239. bayrak y., öztürk, s., çınar, h., koravos, g. ch., & tsapanos, t. m. (2008). seismicity assessment for the different regions in and around turkey based on instrumental data: gumbel first asymptotic distribution and gutenberg-richter cumulative frequency law. natural hazards and earth system sciences, 8, 109-122. bayrak, y., öztürk, s., çınar, h., kalafat, d., tsapanos, t. m., koravos, g. ch., & leventakis, g. a. (2009). estimating earthquake hazard parameters from instrumental data for different regions in and around turkey. engineering geology, 105, 200-210. bozkurt, e. (2001). neotectonics of turkey-a synthesis. geodinamica acta, 14, 3-30. cao, a., & gao, s. s. (2002). temporal variation of seismic b-values beneath northeastern japan island arc. geophysical research letters, 29(9), 10.1029/2001gl013775. frohlich, c., & davis, s. (1993). teleseismic b-values: or, much ado about 1.0. journal of geophysical research, 98 (b1), 631-644. gutenberg, b., & richter, c. f. (1944). frequency of earthquakes in california. bulletin of the seismological society of america, 34, 185-188. han, q., wang, l., xu, j., carpinteri, a., & lacidogna, g. (2015). a robust method to estimate the b-value of the magnitude-frequency distribution of earthquakes. chaos, solitons & fractals, 81, 103-110. hempton, m. r. (1987). constraints on arabian plate motion and extensional history of the red sea. tectonics, 6, 687–705. ishimoto, m., & iida, k. (1939). observations of earthquakes registered with the micro seismograph constructed recently. bulletin of the earthquake research institute, university of tokyo, 17, 443-478. joseph, j. d. r., rao, k. b., & anoop, m. b. (2011). a study on clustered and de-clustered world-wide earthquake data using g-r recurrence law. international journal of earth sciences and engineering, 4, 178-182. katsumata, k., & kasahara, m. (1999). precursory seismic quiescence before the 1994 kurile earthquake (mw = 8.3) revealed by three independent seismic catalogs. pure applied geophysics, 155(2-4), 443-470 maden, n., & öztürk, s. (2015). seismic b-values, bouguer gravity and heat flow data beneath eastern anatolia, turkey: tectonic implications. survey in geophysics, 36, 549-570. marcellini, a. (1995). arrhenius behavior of aftershock sequences. journal of geophysical research, 100(b4), 6463-6468. mcclusky, s., balassanian, s., barka, a., demir, c., ergintav, s., georgiev, i., gürkan, o., hamburger, m., hurst, k., kahle, h., kastens, k., kekelidze, g., king, r., kotzev, v., lenk, o., mahmoud, s., mishin, a., nadariya, m., ouzounis, a., paradissis, d., peter, y., prilepin, m., reilinger, r., s¸ anlı, i., seeger, h., tealeb, a., toksöz, m. n., & veis, g. (2000). global positioning system constraints on plate kinematics and dynamics in the eastern mediterranean and caucasus. journal of geophysical research, 105(b3), 5695-5719. mogi, k. (1962). magnitude-frequency relation for elastic shocks accompanying fractures of various materials and some related problems in earthquakes. bulletin of the earthquake research institute, tokyo university, 40, 831–853. negi, s. s., & paul, a. (2015). space time clustering properties of seismicity in the garhwal-kumaun himalaya, india. himalayan geology, 36 (1), 91-101. olsson, r. (1999). an estimation of the maximum b-value in the gutenbergrichter relation. geodynamics, 27, 547-552. öztürk, s., bayrak, y., çınar, h., koravos, g. ch., & tsapanos, t. m. (2008). a quantitative appraisal of earthquake hazard parameters computed from gumbel i method for different regions in and around turkey. natural hazards, 47, 471-495. öztürk, s. (2009). an application of the earthquake hazard and aftershock probability evaluation methods to turkey earthquakes. ph.d. thesis, karadeniz technical university, trabzon, 346 pp (in turkish with english abstract). öztürk, s. (2011). characteristics of seismic activity in the western, central and eastern parts of the north anatolian fault zone, turkey: temporal and spatial analysis. acta geophysica, 59 (2), 209-238. öztürk, s., & bayrak, y. (2012). spatial variations of precursory seismic quiescence observed in recent years in the eastern part of turkey. acta geophysica, 60 (1), 92-118. öztürk, s. (2015a). a study on the correlations between seismotectonic b-value and dc-value, and seismic quiescence z-value in the western anatolian region of turkey. austrian journal of earth sciences, 108 (2), 172-184. öztürk, s. (2015b). fractal dimension of seismicity and a modeling on the intermediate-term forecasting for the locations of expected strong earthquakes: eastern anatolian region, turkey. gümüşhane university journal of science and technology institute, 5(1), 1-23 (in turkish with english abstract). polat, o., gok, e., & yılmaz, d. (2008). earthquake hazard of the aegean extension region (west turkey). turkish journal of earth sciences, 17, 593-614. reasenberg, p. a. (1985). second-order moment of central california seismicity, 1969-1982. journal of geophysical research, 90 (b7), 5479-5495. rehman, k., ali, a., ahmed, s., ali, w., ali, a., & khan, m. y. (2015). spatiotemporal variations of b-value in and around north pakistan. journal of earth system science, 124(7), 1445-1456. sayıl, n. (2005). an application of the timeand magnitude-predictable model to long-term earthquake prediction in eastern anatolia. journal of seismology, 9, 367-379. scholz, c.h. (1968). the frequency-magnitude relation of microfracturing in rock and its relation to earthquakes. bulletin of the seismological society of america, 58, 399-415. singh, c. (2016). spatial variation of seismic b-values across the nw himalaya, geomatics. natural hazards and risk, 7(2), 522-530. şaroğlu, f., emre, o., & kuşcu, o. (1992). active fault map of turkey. general directorate of mineral research and exploration, ankara, turkey. şengör, a. m. c., & yılmaz, y. (1981). tethyan evolution of turkey: a plate tectonic approach. tectonophysics, 75, 181–241. tsapanos, t. m., bayrak, y., cinar, h., koravos, g.c h., bayrak, e., kalogirou, e. e., tsapanou, a. v., & vougiouka, g. e. (2014). analysis of largest earthquakes in turkey and its vicinity by application of the gumbel iii distribution. acta geophysica, 62(1) 59-82. tsapanos, t. m., yadav, r. b. s., olasoglou, s. m., & singh, m. (2016). assessment of the relative largest earthquake hazard level in the nw himalaya and its adjacent region. acta geophysica, 64(2), 362-378. utsu, t. (1971). aftershock and earthquake statistic (iii): analyses of the distribution of earthquakes in magnitude, time and space with special consideration to clustering characteristics of earthquake occurrence (1). journal of faculty of science, hokkaido university, series vii (geophysics), 3, 379-441. westeway, r. (1994). present-day kinematics of the middle east and eastern mediterranean. journal of geophysical research, 99, 12,071–12,090. wiemer, s., & wyss, m. (1994). seismic quiescence before the landers (m=7.5) and big bear (m = 6.5) 1992 earthquakes. bulletin of the seismological society of america, 84, 3, 900-916. wiemer, s. (2001). a software package to analyze seismicity: zmap. seismological research letters, 72, 3, 373-382. wyss, m., & habermann, r. e. (1988). precursory seismic quiescence. pure applied geophysics, 126, 2-4, 319-332. yadav, r. b. s., koravos, g. ch., tsapanos, t. m., & vougiouka, g. e. (2015). a probabilistic estimate of most perceptible earthquake magnitudes in the nw himalaya and adjoining regions. pure applied geophysics 172(2), 197-212. earth sciences research journal hydrogeology earth sci. res. sj. vol. 16, no. 2 (december, 2012): 27 30 the impacts of alkaline mine drainage on ba, cr, ni, pb and zn concentration in the water resources of the takht coal mine, iran behnaz dahrazma1* and mehdi kharghani2 1 environmental geology department, earth science faculty, shahrood university of technology, p.o. box 316, shahrood, 36199-95161, iran. e-mail: behnaz_dahrazma@yahoo.com. 2 mining department, islamic azad university shahrood branch, shahrood, iran. e-mail: kharghani.m@gmail.com. * corresponding author: behnaz dahrazma, environmental geology department, earth science faculty, shahrood university of technology, p.o. box 316, shahrood, 36199-95161, iran. phone: +98 (912) 373 9562 fax: +98 (273) 339 6007. e-mail: behnaz_dahrazma@yahoo.com abstract the release of heavy metals into the environment represents one of the most important environmental effects involved in extracting coal; it needs to be studied more fully. the present research investigated the effects of coal-mining in an alkaline environment and alkaline mine drainage in the takht coal mine regarding the distribution of selected heavy metals (zn, pb, ni, cr and ba) on the region’s surface and ground water . the mine is located 12 km southeast of minoodasht, in golestan province in northern iran. samples were collected from groundwater and surface water resources upstream and downstream of the mine. the elements’ concentrations were measured by the inductively-coupled mass spectrometry (icpms) method. the results showed that an alkaline environment was responsible for producing alkaline mine drainage due to the presence of limestone; this caused high ph (8.41) in the area’s groundwater resources. mining activities increased ba, cr, ni, pb and zn concentration in the groundwater from 3.39, 0.5, 0.2, 0.5, 9.2 ppb to 83.52, 2.2, 0.6, 2.6, 48.3 ppb and from 68.7, 0.5, 1.3, 0.8, 172.6 ppb to 91, 1.2, 4.5, 1.3, 27.6 ppb in surface water, respectively. due to the basic environment, heavy metal accumulation in the bed sediment for both tunnel effluents and runoffs was higher than during the soluble phase. ph was the main controlling factor in elements’ solubility and their distribution in the environment. increased ba concentration in water resources was due to high ba concentration in the coal, coal tailing and in quarry tailings. resumen la liberación de metales pesados en el medio ambiente representa uno de los efectos ambientales más importantes en la extracción de carbón, lo cual necesita ser estudiado con mayor profundidad. el presente estudio investigó los efectos de la minería del carbón en un ambiente alcalino y en el drenaje de la mina de carbón takht, con respecto a la distribución de determinados metales pesados (zn, pb, ni, cr y ba) en superficie y en las aguas subterráneas de la región. la mina está ubicada a 12 km al sureste de minoodasht, en la provincia de golestán en el norte de irán. se recogieron muestras de aguas subterráneas y aguas superficiales aguas arriba y aguas abajo de la mina. las concentraciones de los elementos se midieron mediante la espectrometría de masa de acoplamiento inductivo (icp-ms). los resultados indican que un ambiente alcalino produjo drenaje de minas alcalinas debido a la presencia de la piedra caliza; esto causo altos niveles de ph (8,41) en las fuentes de aguas subterráneas de la zona. la actividad minera incrementó los niveles de concentración de ba, cr, ni, pb y zn en las aguas subterráneas de 3,39, 0,5, 0,2, 0,5, 9,2 ppb a 83,52, 2,2, 0,6, 2,6, 48,3 ppb y de 68,7, 0,5, 1,3, 0,8, 172,6 ppb a 91, 1,2, 4,5, 1,3, 27,6 ppb en el agua de la superficie. debido al ambiente básico, la acumulación de metales pesados en el lecho sedimentario, tanto para los efluentes del tunel como para las aguas de escorrentia, fueron mayores que durante la fase soluble. el ph fue el principal determinante en la solubilidad de los elementos y su distribución en el medio ambiente. el aumento de la concentración de ba en los recursos hídricos se debió a la alta concentración de ba en el carbón, carbón de relaves y colas en cantera. palabras claves: metales pesados , aguas subterráneas, aguas superficiales, sedimentos, drenaje alcalino en minería, minería de carbón. keywords: heavy metals, groundwater, surface water, sediment, alkaline mine drainage, coal-mining. record manuscript received: 17/02/2012 accepted for publications: 06/05/2012 behnaz dahrazma and mehdi kharghani28 introduction coal is a heterogeneous material which is formed through plant decomposition in a variety of conditions regarding moisture, temperature and pressure. the diversity of elements in coal’s compositions as well as the complex process of the coal forming are responsible for the existence and accumulation of light elements (e.g. li, be), non-metallic elements (e.g. se, as, bi) and heavy metals (e.g. cu, pb, co, cr) in coal. these elements can be released into the environment during coal extraction, its preparation, processing, transportation and combustion (mainly in power plants) thereby causing environmental pollution (finkelman et al., 2002; swaine and goodarzi, 1995; yiwei et al., 2007). geochemical characteristics of coal in the alborz region of iran have been thoroughly investigated. the findings have shown that aluminium silicates and sulphide minerals are heavy metals’ main hosts. according to the chemical and physical weathering of excavated materials (coal and its wastes), elements are released from various complexes and then transported or precipitated due to their geochemical properties. after tunnel effluent transportation from a reduced to an oxidised environment, the elements’ oxidation state becomes changed and some heavy metals become precipitated (i.e. pb, ni and zn). reaction oxidation and reduction potential, ph, colloid materials in transporting media and transferred materials’ ion potential are the main factors controlling materials’ transportation and precipitation (northeast alborz mining co., 2009). the takht coal mine is located at 37o 2’ 10” n and 37o 10’ 35” n latitude and 55o 22’ 30” e and 55o 27’ 40” e longitude, lying 12 km southeast of minoodasht, golestan, in iran (figure 1). extraction started from 2001 onwards with horizontal tunnels being bored on the south side of the area (minoodasht water office, 2003a). summer in the area is warm and humid whilst the winter involves a large amount of precipitation, mostly in the form of snow in the mountainous areas. the temperature varies from -10oc to 35oc. the basin’s annual average rainfall is about 500 mm. runoff feeds the chehel-chay river which is located downstream of the mine (minoodasht water office, 2003b). this research investigated the effects of alkaline mine drainage on heavy metals’ distribution (namely ba, cr, ni, pb and zn) on groundwater and surface water in the vicinity of the takht coal mine in minoodasht, iran, regarding environmental concerns. materials and methods triplicate samples were collected from coal, coal and quarry tailings tunnel effluent, tailing leachate and surface and ground water as well as bed sediments from tunnel effluent and runoff to assess the effects of coal minparameter from to coal moisture 1.2% 1.8% ash content 11.2% 44.7% volatile compounds 34.2% 37.5% heat value 7.985 kcal/kg 8.132 kcal/kg carbon content 79.99% 84.24% hydrogen content 4.7% 5.77% oxygen + nitrogen content 10.72% 15.34% phosphorus content 0.04% 0.31%. sulfur content 0.56% 1.3% (mostly in the form of pyrite) table 1. coal sample characteristics (northeast alborz coal mining co. 2009). table 2. sampling locations and the samples’ ec and ph. no. place of sampling longitude latitude ph ec(μs/cm) 1 tunnel effluent 361608 4112469 8.41 1451 2 tailing leachate 361244 4113631 8.12 2620 3 upstream groundwater 361940 4111490 7.23 219 4 downstream groundwater 356526 4115781 7.58 726 5 upstream runoff 361494 4112509 8.21 536 6 downstream runoff 361451 4112539 8.29 828 7 downstream river 357480 4114509 8.31 654 8 tunnel effluents' bed sediments 361509 4112506 7.63 9 upstream runoff bed sediments 361494 4112509 7.53 10 downstream runoff bed sediments 361451 4112539 7.77 11 downstream river bed sediments 357480 4114509 7.37 the impacts of alkaline mine drainage on ba, cr, ni, pb and zn concentration in the water resources of the takht coal mine, iran 29 ing on heavy metals’ distribution in surface and ground water. sampling was done in october 2009 at the end of the dry season to find the worst conditions. table 1 gives the coal samples’ characteristics. sediments were air dried at room temperature and passed through #80 mesh (180 mm opening size) to measure the heavy metals’ concentrations. samples were analysed at the acme lab, vancouver, canada using inductively-coupled mass spectrometry (icp-ms) techniques. the water’s ph and ec were measured on site using a jenway ph meter (model 3510) and jenway ec meter, respectively. for the sediments, ph was measured referring to the epa sw-846 method 9045. table 2 shows sampling locations, ph and ec. results and discussion alkaline mine drainage has resulted from the neutralisation of acidic mine drainage (amd) by carbonate minerals and low grade pyrite rocks. the basic tunnel effluent (ph 8.4) indicated the presence of alkaline mine drainage (table 2) due to the highly alkaline environment. this alkalinity was responsible for the sedimentation of most elements in tunnel effluent bed sediments. ec tended to increase downstream of the mine which could have been due to the high concentration of heavy metals and ions. sample 4 (taken from groundwater 7 km downstream of the mine) had an ec which was 3.3 times higher than that of sample 3 (taken from groundwater upstream of the mine). this increase could have resulted from tunnel effluent and tailing leachate entering the environment (1,451 and 2,620 µs/cm ec, respectively) and causing an increased ec in surface water (from 536 µs/cm in upstream runoff to 828 µs/cm in downstream runoff). a rise in surface figure 1. overall view of the study area’s location. figure 3. heavy metals’ concentration in tunnel effluent bed sediment. figure 2. heavy metals’ concentrations in tunnel effluent and tailing leachate. table 3. the heavy metals’ concentration in coal and tailings (in mg/kg). figure 4. heavy metals’ concentration in bed sediment samples (sample 9: upstream runoff sediments, sample 10: downstream runoff sediments, sample 11: downstream river sediments) sample zn pb ni cr ba coal 33.8 21.76 28.4 18.1 118.8 coal tailing 128.4 46.66 70.8 43.3 54.8 quarry tailing 80.3 32.95 60.5 42.8 146.4 and ground water ph was also observed (table 2). the results indicated that sulphate concentration in surface water increased from zero to 48 mg/l due to pyrite oxidation and sulphate complex entering the environment which facilitated the retention of heavy metals in solution phase. the present research’s findings agreed with anderson et al., (2005) who noted that concentrations of sulphate, iron, manganese and other constituents can increase, even in the presence of alkaline mine drainage. coal, coal tailing and quarry tailing analysis revealed high concentrations of heavy metals (especially ba) in all three contaminant sources. comparing these three contaminant sources indicated that coal tailings had the maximum zn, pb, ni and cr concentrations while maximum ba concentration occurred in quarry tailings (table 3). washing and the consequent leaching of coal, coal tailings and quarry tailings were the main factors for contaminants entering water. tunnel effluent and drainage of tailing washing were the main sources of pollution, containing heavy metals ba, cr, ni, pb and zn. ba had the highest concentration due to its high solubility in an alkaline environment (figure 2). barium also had the maximum concentration in tunnel effluent behnaz dahrazma and mehdi kharghani30 bed sediment compared to the other heavy metals (figure 3). regarding ph 8.4, the heavy metals in the bed sediment had much higher concentrations than in the tunnel effluent. in other words, heavy metals’ concentrations in solid phase were higher than in the aqua phase when ph was 8.4 (figures 2 and 3, table 2). figure 4 shows that heavy metals’ concentration in bed sediments from upstream and downstream runoff had the following order: ba > zn > ni > cr > pb. ion exchange, adsorption and complexation are the main known mechanisms for heavy metals’ accumulation in sediment (dahrazma and mulligan, 2007). smith (1999) mentioned that increased ph enhances cation sorption and decreases anion sorption. the high ba concentration in upstream runoff bed sediment could have been due to a carbonated environment and ca exchange with ba; this finding agreed with that of hem (1985). concentrations of heavy metals in downstream river bed sediments (except for pb) became decreased, due to drained water becoming diluted in the river. contaminants entering the environment through mining activities (as mentioned above) changed the area’s heavy metals’ distribution in groundwater and surface water (table 4). ba concentration in downstream groundwater was 24.6 times greater than upstream due to high ba concentration in tunnel effluent (142.6 mg/l) and tailing leachate (90.8 mg/l). mining activities also caused increased ba concentration in downstream runoff compared to upstream (table 4). the results showed that cr, ni, pb and zn concentrations in groundwater increased 4.4, 2, 5.3, and 5.5 times, respectively, from upstream to downstream. in terms of runoff, cr, ni and pb concentrations from 0.5, 1.3 and 0.8 upstream reached 1.2, 4.5 and 1.3 (all in mg/l) downstream, respectively. zinc did not increase in downstream runoff (comparing figure 2 and table 4) since its concentration in tunnel effluent and tailing leachate (62.3 and 64.5 mg/l, respectively) was lower than its concentration in upstream runoff (172.6 mg/l). it should be noted that the concentration of heavy metals in the river was generally lower than in downstream runoff due to dilution. this was clearly an indication of the adverse effect of mining activities on the region’s groundwater and surface water. it can thus be concluded that alkaline mine drainage being released into the environment increased ba, zn, pb, cr and ni concentration (in descending order) in groundwater and ni, cr, pb and ba in surface water. conclusion alkaline mine drainage rarely occurs in coal mines throughout the world. the takht coal mine is one of the few coal mines where alkaline mine drainage occurs. the mining activities started in 2002 and have already had adverse effects on the quality of water resources in the area since the concentration of heavy metals in downstream groundwater is higher than in upstream water. tunnel effluent and tailing leachate entering the environment were the main causes of such increases. an alkaline environment was responsible for the heavy metals’ solubility and distribution in surrounding water resources. heavy metals’ concentration was much higher in tunnel effluent bed sediments and upstream and downstream runoff than in the aqua phase due to the alkaline environment reducing solubility. mining activities increased ba, cr, ni and pb concentration in the area’s surface water and groundwater. alkaline mine drainage has led to increased zn concentration in downstream groundwater compared to upstream groundwater (5.2 times greater) while it decreased zn concentration in surface water. the rate of coal-mining-related alkaline mine drainage effects on the environment should be investigated more fully to prevent further damage in the foreseeable future. references anderson, r. m., beer, k. m., buckwalter, t. f., clark, m. e., mcauley, s. d., sams, j. i., iii, williams, d. r., 2005. water quality in the allegheny and monongahela river basins pennsylvania, west virginia, new york, and maryland, 1996-98. u.s. geological survey circular 1202. dahrazma, b., mulligan, c. n. (2007). investigation of the removal of heavy metals from sediments using rhamnolipid in a continuous flow configuration, chemosphere, 69, 705-711. finkelman r. b., orem w., castranova v., tatu c. a., belkin h. e., zheng b., lerch h. e., maharaj s. v., bates a. l. (2002). health impacts of coal and coal use: possible solutions, international journal of coal geology, 50, 425-443. hem, j. d. (1985). study and interpretation of the chemical characteristics of natural water, 3rd ed. water supply, us geological survey, paper 2254, 263pp. minoodasht water office. (2003a). quality control report on minoodasht water samples (in persian). minoodasht water office. (2003b). takht basin report (in persian). northeast alborz mining co. (2009). detailed exploration of takht region, in annual exploration report (in persian). smith, k. s. (1999). sorption of trace elements by earth materials—an overview with examples relating to mineral deposits. in: plumlee, g. s., logsdon, m. j. (eds). the environmental geochemistry of mineral deposits. part a. processes, techniques, and health issues: society of economic geologists, reviews in economic geology 6a, 61–182. swaine, d. j., goodarzi, f. (1995). environmental aspects of trace elements in coal. kluwer academic publishers, dordrecht. yiwei, c., guijian, l., yanming, g., jianli, y., cuicui, q., lianfei, l. (2007). release and enrichment of 44 elements during coal pyrolysis of yima coal, china, journal of analytical and applied pyrolysis, 80(2), 283-288. sample place ba cr ni pb zn 3 upstream groundwater 3.4 0.5 0.2 0.5 9.2 4 downstream groundwater 83.5 2.2 0.6 2.6 48.3 5 upstream runoff 68.7 0.5 1.3 0.8 172.6 6 downstream runoff 91 1.2 4.5 1.3 27.6 7 downstream river 63.1 1.1 0.2 1 9.4 min 3.4 0.5 0.2 0.5 9.2 max 91,0 2.2 4.5 2.6 172.6 mean 61.94 1.1 1.36 1.24 53.42 standard deviation 34.6 0.7 1.8 0.8 68.5 table 4. the heavy metals’ concentration in the region’s water resources (in μg/l). keywords: wulingyuan; scenic area; peak forest landscape; spatial structure; characteristics; optimization. palabras clave: wulingyuan; pilares de arenisca; estructura espacial; características; mejoramiento. how to cite item zhang, q. (2020). study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area. earth sciences research journal, 24(1), 35-44. doi: https://doi.org/10.15446/esrj. v24n1.85206 to protect the ecological environment of peak forest landform and maintain its integrity and stability, the optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area was studied. by using gis and digital topographic analysis to study the basic features of sandstone peak forest landform, wulingyuan peak forest landform and karst peak forest have great differences in lithological composition, as the weathering resistance is better than cheltenham badland landform; by using tree theory to analyze the features of wulingyuan peak forest landform, according to area weight serialization of contour tree nodes, we can know the depression area. the karstification is stronger than that of the peak forest area, and the surface is relatively fragmented. based on different landscape indices, the landscape pattern of wulingyuan peak forest is analyzed. the fragmentation degree of vegetation is lower, and the fragmentation degree of building landscape is first increased, then decreased, and finally intensified. the proportion of artificial landscape decreases year by year and the trend of fragmentation is obvious. based on the present situation of spatial structure characteristics of wulingyuan peak forest landform, optimization methods such as combining centralization with decentralization and improving the quality of artificial landscape ecosystem were put forward. abstract study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area estudio sobre el método de optimización de las características de estructura espacial en los pilares de arenisca de wulingyuan issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.85206 para proteger el entorno ecológico de los pilares de arenisca en el área de wulingyuan y mantener su integridad y estabilidad se estudió el método de optimización de las características de la estructura espacial. al utilizar sistemas de información geográfica (gis) y el análisis topográfico digital para estudiar las características básicas de la forma de los pilares se estableció que la forma de relieve forestal del pico de wulingyuan y los pilares de arenisca tienen grandes diferencias en la composición litológica, mientras que el nivel de resistencia a la meteorización es mejor que en la zona de cheltenham badland; al utilizar la teoría de los árboles para analizar las características de los pilares de wulingyuan, de acuerdo con la serialización del peso del área de los nodos del árbol de contorno, se estableció el área de depresión. el índice de karstificación es mayor que el del área del bosque de pilares de arenisca, y la superficie está relativamente fragmentada. basado en diferentes índices de paisaje, se analizó el patrón de paisaje del los pilares del bosque wulingyuan. el grado de fragmentación de la vegetación es menor, y el grado de fragmentación de la construcción del paisaje primero aumenta, luego disminuye y finalmente se intensifica. la proporción de paisaje artificial disminuye año tras año y la tendencia de fragmentación es obvia. basado en la situación actual de las características de la estructura espacial de los pilares de wulingyuan se presentaron métodos de optimización como la combinación de la centralización con la descentralización y la mejora de la calidad del ecosistema del paisaje artificial. resumen record manuscript received: 27/07/2019 accepted for publication: 13/11/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 35-44 qian zhang* local cooperation office, jiaozuo teachers college, jiaozuo, 454000, china * corresponding author: jzszxqhzczq@126.com g e o m o r ph o l o g y 36 qian zhang introduction environment and development have become a common concern in today’s world. ecosystem is the link between environment and development. it not only meets the needs of material products and environmental services for human survival and development, but also plays an important role in maintaining global ecological balance and protecting biodiversity (sylvestre et al., 2017). wulingyuan world natural heritage situated in the northwest of hunan province is famous for its unique quartz sandstone peak forest landform. sandstone is a kind of rock formed by weathering, transporting, sedimentation and diagenesis. it has the characteristics of low strength, poor weathering resistance and little cementing material. it is the main part of the peak forest landform of wulingyuan scenic area. michael crosser, president of the international association of geomorphologists, said at the opening ceremony that similar geomorphological features exist in the world, but the geomorphology of zhangjiajie should be recognized by the world because it lacks such distinctive features and scale and integrates aesthetic feeling and scientific research value. zhangjiajie, a sandstone landform, is very rare in china and even in the world, and its formation conditions are very harsh (colgan et al., 2017). therefore, in order to protect the ecological environment of the peak forest landform in wulingyuan scenic area and maintain the integrity and stability of the peak forest landform, it is necessary to study the optimization method of the spatial structure characteristics of the peak forest landform in wulingyuan scenic area. at present, there are many studies on the peak forest landform of wulingyuan scenic area in zhangjiajie in china. qi chengjing and fang yingcai have visited zhangjiajie for plant investigation and specimen collection. zhang zhicai and others have carried out research on the types of forest resources, community structure and distribution characteristics, and the function of ecosystem in the peak forest landform of wulingyuan. based on the research and discussion, the countermeasures and suggestions are put forward in the development planning, evaluation of ecological benefits and construction and development ideas of peak-forest landform. liang mingzhong and others use spatial index and spatial statistical technology to describe the spatial attributes of peak-forest landform and study the ecological characteristics of the spatial structure of peak-forest landform. this paper studies the spatial structure characteristics of peak forest landform in wulingyuan scenic area from three aspects: basic landform characteristics, surface morphology characteristics and geomorphological spatial pattern, and puts forward rational optimization methods to provide basis and reference for the ecological environment protection of wulingyuan scenic area (jiang, 2017). research areas and research methods survey of research areas wulingyuan is located at eastern longitude 110°20’30”~110°41’15”, northern latitude29°16’25”~29°24’25”. it is the junction of zhangjiajie city in northwest hunan province with cili and sangzhi counties, with a square area of 369 km. it consists of four distinctive nature reserves: zhangjiajie, tianzishan, suoxiyu and yangjiajie. there are still scenic areas under development (cole et al., 2018) in its periphery, such as maoyan river, jiudian cave and tianmen mountain. zhangjiajie is the first national forest park established in china (1982), so it is often called wulingyuan scenic area by zhangjiajie. wulingyuan belongs to the national first-level protected area with an area of 53.5 square kilometers, the second-level protected area with an area of 55.5 square kilometers and the third-level protected area with an area of 203 square kilometers. there are fourteen 70 km tourist routes and more than 560 scenic areas (zhang et al., 2017). wulingyuan scenic area has ravines, high rock peaks, green trees and green vines, and flocks of animals and birds. it has the reputation of “the maze of nature” and “the world’s first strange mountain”. the landform of wulingyuan scenic area belongs to the quartz sandstone peak forest landform. there are as many as 15,000 peaks in wulingyuan. the peaks are distributed at an altitude of 500-1100 meters, ranging from tens of meters to 400 meters. this special landform is named “quartz sand pillar peak”. as shown in figure 1. the geomorphology of quartz sandstone peak forest is characterized by pure quality and stone thickness, quartz content is 75%-95%, and rock thickness is more than 520 meters. wulingyuan has a complex terrain, mild climate, abundant rainfall, lush forests and primitive sub-forest plant communities. its forest coverage rate is 88%. there are 4 species of first-class protected plants and 19 species of second-class protected plants in wulingyuan. there are 770 species of woody plants, of which pinus wuling is the only one in the region. in addition, there are tertiary relics such as davidia involucrata, bole tree, taxus chinensis, taxus chinensis and cephalotaxus angustifolia. wulingyuan has 28 kinds of state-level protected animals. there are also famous giant salamander (doll fish) precious animals growing in wulingyuan, as well as the first-level protected animals such as clouded leopard, leopard and other large beasts. in december 1992, wulingyuan scenic area was listed on the world natural heritage list (moreno et al., 2017) by unesco as a world famous tourist attraction. the formation factors of peak-forest landform are as follows: firstly, the rock is quartz sandstone with pure quality, mainly quartz, thick layer, hard and stable, which is beneficial to natural sculpture and image enhancement; secondly, the occurrence of rock stratum is gentle, which increases the stability of rock, which provides the precondition for the formation of peak-forest landform; thirdly, two groups of sags are developed in rock stratum. straight joints, generally 15 to 20 meters apart, provide the tectonic conditions for shaping the peak-forest landform. under the background of intermittent uplift of regional neotectonic movement, the effects of flow erosion cutting, physical weathering, chemical weathering and biological root splitting and other external forces begin with rock joints, which widen the fissures gradually and evolve over time to form peak forest landform. its characteristics are high, flat and steep (klongvessa et al., 2018). figure 1. geomorphology of quartz sandstone peak forest in wulingyuan scenic area experimental data this experiment chooses 30 m grid resolution srtm data, 1 m dom data of high resolution remote sensing image, fast bird image data and world view data as main information sources, supplemented by 1:10,000 scale topographic map and other related data (payo et al., 2017) from the completed national geographic database. in addition, the urban construction archives of the study area since 1990 were consulted, and a large number of data on social, economic and environmental changes (li et al., 2018) were collected. the characteristics of the experimental data are shown in table 1. extraction of topographic features of peak forest landforms based on tree node serialization in the process of extracting the topographic features of wulingyuan peakforest, based on the theory of karst geomorphology, hydrology and mathematical statistics, using gis spatial analysis and digital topographic analysis (davies et al., 2018), using the combination of qualitative and quantitative analysis indicators, macro and micro analysis, and theoretical analysis. combining 37study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area table 2. landscape indicators for research landscape indicators formula for calculating indicators indicator unit scope and meaning of indicators patch type area ca aij j n = = ∑ 1 hectare ca > 0, no upper limit, representing landscape composition patch density pd n ai= / number/ hectare pd > 0, no upper limit, characterizing the degree of fragmentation percentage of patch type area pland p a ai ij j n = = × = ∑ / 100 1 pland p a ai ij j n = = × = ∑ / 100 1 % 0 < pland < 100, characterizing the dominance of each type of plaque average plaque area mps a nij i j n = = ∑ / 1 hectare mps > 0, no upper limit, indicating plaque size maximum plaque area ratio max /p ai × % characterization of maximum patch dominance fractal dimension of plaque frac p aij ij= log / log 1 < frac < 2; frac = 1 is round; frac = 2, long strip landscape diversity index shdi p inpi i i m = ×( ) = ∑ 1 shdi > 0, no upper limit; shdi = 0, only one plaque results and optimization geomorphological characteristics of sandstone peak forest wulingyuan peak-forest landform is situated on the upper part of the mountain body above 1000 meters above sea level. the edge of the mountain body or the valley zone develops best. because of the cutting of the valley and the collapse of the rocks on the edge of the mountain body, the edge of the peak forest is generally a steep cliff. the rock wall is bounded by a clear unconformity of caledonian crustal movement, and the strata are mainly composed of the mesozoic and paleozoic strata. the upper devonian yuntaiguan formation and huangjiadun formation are composed of littoral clastic rocks (liang et al., 2017). the rocks are pure, thick, flat in base and well developed in vertical joints. the rocks are exposed in syncline outline, which reflects the special geological structure environment and basic conditions for the formation of sandstone peak forest landscapes. running water erosion and gravity collapse, biochemical and physical weathering effects of external geological activities are indispensable external conditions for shaping wulingyuan landform landscape. therefore, the formation of sandstone peak forest landform is the result of the long-term interaction of internal and external geological gravity in a specific geological environment, coupled with the intermittent uplift and tilt of regional neotectonic movement, flow erosion and cutting, gravity, physical weathering, biochemistry and root splitting and other external forces. peak forests are formed according to the complex natural evolution process, showing the characteristics of high, flat and steep walls. (catania et al., 2017). basic morphological characteristics sandstone peak forest landform is a group landform, its basic form is cliff, peak (column), cliff channel, cliff lane. with application validation, using multi-resolution dem data as the basic data source, exploratory study on the geomorphological characteristics of sandstone peak forest was carried out by using tree theory (hart, 2017). because the number of nodes in each layer of the contour tree is different, the size of the chain structure transformed by the weight of the nodes in each layer according to the area represented by the contour is also different (wang et al., 2019). therefore, in the specific analysis and application, the tree node serialization method based on area weight is used to intercept the standard table structure with the same growth degree from the chain structure of different lengths, so that the classical mathematical statistics method can be used for analysis. the principle of tree node serialization is shown in figure 2. determination of evaluation index for landscape pattern analysis landscape pattern analysis is the core content of landscape ecology. quantitative analysis of spatial distribution characteristics of landscape components is the basis for further study of landscape function and dynamics (goad & korista, 2018). the impact of human activities on the landscape spatial pattern of wulingyuan peak forest (changqing et al., 2017) was measured by landscape patch type area, patch density, patch type area ratio, average patch area, maximum patch area ratio, patch fractal dimension and landscape diversity index. the landscape of wulingyuan peak forest was studied separately. the landscape indices used in this study are shown in table 2. table 1. characteristics of experimental data type dem data remote sensing image quartz sandstone peak forest landform data sources nasa center website domestic and foreign websites data sales company scanning drawings analytical objectives core data for topographic feature extraction, spatial analysis, etc. basic data sources for various experiments correction of the peak position of peak-forest peaks and clusters auxiliary data for comparative analysis of terrain spatial differentiation characteristics figure 2. sequencing schematic diagrams of area-weighted nodes 38 qian zhang (1) cliff: cliff, which is the largest and most spectacular landscape in the peak forest, can be divided into two cliffs, edge and interior, according to their different locations. the edge cliff is located at the edge of the peak forest area, which extends tens of kilometers intermittently. the local area is thousands of meters to more than ten kilometers long, tens of meters to more than one hundred meters high and 530m high. the wall is steep, like a copper wall and iron wall. the trend of cliff wall in peak-forest is controlled by two groups of joints. they are nw and ne. they are generally more than 100 meters wide and tens of meters high. most of them are vertical cliffs with an inclination of more than 70°. all kinds of sedimentary phenomena on the wall are clearly visible. (2) peak (column): that is, the height of the peak forest is greater than the width of the stone peak and pillar, is the most basic morphological landscape of sandstone peak forest landform. the peak forest consists of thousands of single stone peaks and columns. peak shape has square column, flat column, cone, bamboo shoot, peak wall, tower, rod, etc. according to the morphological characteristics of the peak body, it can be divided into the following combinations. square columnar peak forest includes square columnar and flat columnar peak forest, as shown in figure 3. peak forests are usually composed of several or dozens of columns with diameters ranging from several meters to tens of meters and height ranging from tens of meters to more than 100 meters. after the rock is cut by two groups of conjugate vertical joints, some surrounding rocks collapse under the action of gravity. figure 3. columnar peak forests conical-columnar peak forest: the peak column is conical in shape with fine upper and thick lower, usually tens of meters in height and several meters to tens of meters in diameter at the bottom. its formation is related to lithology. sandstone in the lower part of the peak body is dominant, and mudstone in the upper part is increased. it is formed by differential weathering due to the different degree of rock hardness and softness. bamboo shoot-like peak forest: the top of the peak is thin and thick, the base is connected, and the top is bamboo shoot-like. generally, the height is more than several meters, and the scale is smaller than the conical peak forest. the rock is cut by dense vertical joints. peak wall peak forest: horizontal rock stratum is cut by vertical joints to form a cliff, side as a wall, is seen as a peak, wall direction is controlled by joint direction. hieroglyphic peaks: most of them are single or solitary peaks, varying in size and height, such as tower, rod and column shapes. they have various shapes, such as deer, pipe, lotus flower, general, duck head, snake head, and so on. they are similar to human beings. one scene at a time, they are formed by different lithological weathering and rock collapse. (3) cliff groove: refers to the vertical groove landform landscape in which the depth and height of the peak forest are greater than the width, also known as vertical groove, formed by the collapse of rocks in two or several vertical joints. the cliff trough is usually tens of centimeters wide to several meters, with a horizontal depth of 1 m to several meters and a height of 10 m to tens of meters. its width, height and orientation are related to the density, spacing and orientation of joints. it is a common micro-landform landscape in peak forest landform. (4) cliff alley: horizontal roadway with narrow and deep shape is the result of the development of the depth of the cliff groove toward the horizontal direction. its width is generally only 2-3 m, while its height and depth are tens of meters to more than 100 meters. the roadway wall is upright, very similar to a line of sky or a sky roadway. in addition to the above types of morphology, there are also some small landscapes in peak-forest area. they are combined with peak forest, valley, waterfall, forest vegetation and exotic flowers and trees to form a gorgeous landscape, which is described as a fairyland on earth. analysis and comparison peak forest refers to the landscape composed of separated stone peaks or pillars, most commonly in karst area, clastic rock area, granite area and cheltenham badland area are well developed, but far less developed than karst area. wulingyuan sandstone peak forest belongs to clastic rock peak forest. its characteristics are similar to karst peak forest and cheltenham badland landform, but there are also many obvious differences, as shown in table 3. (1) compared with karst peak forest. karst peak forests in china are widely distributed in guangxi, guizhou and eastern yunnan. they are mainly pure and hard carbonate rocks of devonian, carboniferous and permian with thickness of more than kilometers. they have been developed for a long time in hot and humid climates. the karst peak forest generally develops in the plain area or valley on the edge of peak cluster karst. it can be divided into three types: basin plain type, peak cluster valley type and edge slope valley type. stone peaks in peak forests stand apart from each other and vary in density. in plain areas, solitary peaks are dominant. they are pyramidal in shape, partly conical or saddle in shape and below 150 m in height. after being dissolved by water solution, caves on the peaks develop and are known as caves without caves. the main difference between wuling mountain sandstone peak forest and karst peak forest is that the lithological composition is completely different: the forming conditions of the two forests are different. the karst peak forest is mainly formed by water erosion and cave development, while the sandstone peak forest is mainly formed by flowing water erosion, physical weathering, chemical weathering and biological root splitting. (2) comparing with cheltenham badland landform. cheltenham badland landform in china is mainly distributed in renhua of guangdong, wuyimountain of fujian, longhu mountain of jiangxi, dajinhu of fujian, yuntaimountain of henan, yuntaimountain of guangxi, liangmountain of hunan, and the most typical landform of renhua of guangdong is named cheltenham badland landform. the cheltenham badland geomorphic diagenesis age in southern china is mainly the red rock series of the white seal period. its lithology is purple-red conglomerate, sandy conglomerate and coarse sandstone. the rock strata are nearly horizontally produced. the main bedding troughs, flat caves, frontal caves and rounded ridge peaks are formed by a variety of factors. compared with cheltenham badland landform, wulingyuan sandstone peak forest has similar lithology, color and danya landform. the difference is that the formation conditions and morphological types are different. sandstone peak forest rocks are fine clastic rocks with harder texture, strong weathering resistance and more developed danya landform. cheltenham badland landform rocks have coarse grain size and weak weathering resistance. there are obvious differences between bedding troughs (caves), rounded ridges and steamed bread ridges and sandstone peak forests. analysis of peak cluster shape characteristics based on tree node serialization taking the peak forest landform research area of wulingyuan scenic area as an example, the nodes of contour trees are serialized according to area weight. in the actual serialization process, the contour tree is constructed for the five geomorphic units of wulingyuan scenic area: depression, stone peak, compound depression, peak forest and peak-depression combination. only the first four nodes of each layer are intercepted, and the first node of each layer is recorded as the parent node. the first three nodes are sorted according to the size of the area. it is designated as the first child node, the second child node and the third child node. because the contour trees constructed by simple shape (stone peak and depression) are single-branch trees, and the structure is simple, four nodes are truncated downward from the root node, and the root node is 39study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area recorded as the parent node. according to the subordinate relationship between nodes, the first child node, the second child node and the third child node are recorded in turn. finally, based on the order of area serialization, the roundness coefficient c and the compactness coefficient t of the parent node and the child node are calculated respectively, and the corresponding confidence interval diagrams of roundness coefficient and compactness coefficient 95% are drawn. the results are shown in figures 4 and 5. from figure 4 and 5, we can see that the roundness coefficient c of wulingyuan peak forest landform gradually increases from parent node to child node, and the compactness coefficient t gradually decreases from parent node to child node. it shows that the shape of the sub-nodes is closer to the circle than that of the parent nodes in five different geomorphic units, whether roundness coefficient or compactness coefficient is used as a measure. for composite terrain (such as fung shek group, composite depression, combination of peaks and depressions), the shape of the geomorphic units nested step by step in its interior tends to become circular gradually. comparing the roundness and compactness coefficients of different geomorphological types, it can be found that the more complex the combination of landforms (such as the combination of peaks and depressions), the narrower the surface shape, the more complex the surface morphology. at the same time, because the outcrop of carbonate rocks is more obvious in the depression area, and most precipitation flows into the depression area, the karstification in the depression area is stronger than that in the fung shek area, and the surface of the depression area is more fragmented and fragmented, so the range of roundness coefficient and tightness coefficient in the depression area is larger than that in the fung shek area. it can be seen that the structure of contour tree can not only directly reflect the hierarchical nesting characteristics of peak cluster surface morphology, but also contain the related information of different levels of geomorphic units, so as to facilitate the analysis of the morphological characteristics of geomorphic units at different scales. geomorphological spatial pattern of sandstone peak forest change of vegetation landscape in the past 30 years, the vegetation in the study area has been growing and expanding. before tourism development, as in other areas of the country, under the influence of grain policy, deforestation and deforestation were very serious. except for the quartz sandstone peak forest landform, steep rock cliffs and some places with large slopes, most areas had reclaimed wasteland and few patches of vegetation around villages. vegetation coverage is low, the change of landscape characteristics of vegetation cover in wulingyuan scenic area is shown in table 4. table 3. comparison of geomorphological characteristics of different peak forests type lithology and age protolith formation environment peak forest morphology major forming factors geomorphology of sandstone peak forest in wulingyuan upper devonian sandstone the occurrence of rock layers is near horizontal and forms in river deltas on the land margin. the shape of peaks is varied, mainly columnar, with joints and vertical grooves developed. water erosion cutting, physical weathering, chemical weathering and biological root splitting, etc. karst peak forest in guangxi carbonate rocks from middle mupan to permian the occurrence of the strata is near the level and formed in the foothill, current and lake environment. the peak body is mainly pyramidal, and the peak body is large. the valleys, depressions and caves are developed. surface water and groundwater are mainly used for erosion and transportation. cheltenham badland landform peak forest in guangxi purple-red conglomerate and sandstone dominated by the bainig period the occurrence of rock strata is near the level and formed in foothills, rivers and lakes. the peak body is bread-like and rounded, the peak is gentle, the annular structure and horizontal groove, the pot hole is developed. epigenetic weathering and runoff erosion table 4. landscape characteristics of vegetation cover in wulingyuan scenic area from 1977 to 2007 particular year 1977 1987 1998 2004 2007 plaque area(ha) 19556.7 25849.1 27105.1 29516.1 32782.8 vegetation coverage(%) 50.37 66.22 69.38 74.47 83.69 compared with the average annual growth rate of the previous sample(%) 1.61 0.31 0.86 3.08 number of patches (number) 670 922 901 386 612 patch density(number/ ha) 0.0354 0.0357 0.0333 0.0131 0.0187 average plaque size(ha) 29.19 28.04 30.08 76.47 53.57 maximum plaque area(ha) 15876.4 22569.7 23010.7 28180.1 30804.8 percentage of maximum patch area(%) 81.18 87.31 84.89 95.47 93.97 maximum patch fractal dimension 2.2725 1.3473 1.3492 1.2444 1.257 40 qian zhang (a)depression(b)fung shek (c)composite depression(d)fung shekgroup (e)peak-depression assemblage figure 4. comparisons of 95% confidence intervals for roundness coefficients of different landforms after serialization 41study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area (a)depression(b)fung shek (c)composite depression (d)fung shekgroup (e)peak-depression assemblage figure 5. comparisons of 95% confidence intervals for compactness coefficients after serialization of different geomorphic units 42 qian zhang from the point of view of patch fragmentation degree, the degree of vegetation fragmentation is relatively low. there is little change from 1977 to 1987, the degree of vegetation fragmentation is slightly reduced from 1987 to 1998, the degree of fragmentation is obvious from 1998 to 2004, but the degree of vegetation fragmentation has increased from 2004 to 2007. from 1977 to 2007, the largest patch area increased steadily, but its proportion fluctuated, mainly due to the proportion of small vegetation patches in the core area and vegetation patches in the buffer area. the fractal dimension of the largest patch is higher, which is mainly due to the strong influence of human activities around the largest patch and the complex shape of the patch. the most notable one was 1998. architectural landscape change architectural landscape is the smallest landscape type in the study area, but it is the most profound and destructive landscape type of human activities. it is the disturbing landscape in the study area. the change of landscape characteristics is shown in table 5. in 1977, the largest area of architectural landscape was due to the fact that rural settlements were quite scattered and a large number of settlements were still in the core scenic area. by 1987, due to the improvement of production and living conditions, some residents moved spontaneously from the humble mountains to the relatively flat villages under the mountains. on the other hand, because wulingyuan was listed as a national forest park and a national key scenic area, the protection of mountain forests was strengthened, which prompted farmers to move out of the mountains. since 1987, the tourism industry has developed rapidly, bringing about the climax of reception construction. however, the government did not control the construction of reception facilities in an orderly way at that time, which led to blind development. a considerable part of the construction of reception facilities without any formalities is the fastest stage of the urbanization of heritage sites. in 1999, the local government demolished 2×104 m2 illegal buildings. in 2001, the demolition of scenic areas was initiated and nearly 3.4×105m2 of the core scenic buildings were demolished. since 2004, the relocation of residents in the core area has been initiated, and the area of architectural landscape in the core scenic area is less year by year. from the perspective of the number and density of landscape patches, the number of patches increased nearly twice from 1977 to 1998, decreased rapidly in 2004, and then increased; the density of patches increased from 1977 to 1998, and then decreased and increased. the degree of fragmentation of architectural landscape increased from 1977 to 1998, then decreased due to demolition, and then increased significantly due to the pressure of tourism and local residents. from 1977 to 2004, the proportion of the largest patches increased, indicating that the degree of urbanization agglomeration increased. however, after 2004, the patch was redesigned and the group development model was adopted. natural patches were introduced into the patch, which reduced the maximum patch area and proportion. artificial landscape change the artificial landscape in the study area is mainly land use which is strongly influenced by human activities rather than architectural forms, including paddy field, dry land, wasteland, artificial reconstructed waters, roads and other types. the overall characteristics of the artificial landscape in the study area are that the area decreases year by year, and the changes of landscape index characteristics are shown in table 6. from 1977 to 1987, the artificial landscape decreased rapidly, accounting for only 31.32% of the total area in 1987. the main reason for the decrease was that some of the dry land in the gentle slope of mountain forests was able to restore forest vegetation, while some of the artificial landscapes were occupied by the marginal lowland buildings. from 1987 to 1998, there were fewer artificial landscapes, accounting for only 27.67% of the total area in 1998. the main reason for the decrease was the conversion of farmland to forest in the core scenic area and the further reduction of the peripheral artificial landscape area. from 1998 to 2004, there were only a few patches in the artificial landscape of the core scenic area, the decrease was most obvious. in 2004, the artificial landscape only accounted for 24.42%. from 2004 to 2007, the number of artificial landscapes in the buffer zone decreased significantly. the main reason is that some areas with poor farming conditions in the buffer zone have been restored to vegetation cover. in 2007, only 15.12% of the total artificial landscapes were restored to vegetation cover. from the perspective of patch density and average patch size, the trend of fragmentation of artificial landscape is quite obvious, and only in 1998-2004, it has been reduced. the main reason is that the artificial landscape is gradually eroded by vegetation patches, and the number of patches is increasing, which has become a landscape type gradually eroded in the study area. spatial structural feature optimization method (1) based on the principle of combination of centralization and decentralization, a spatial development model (fang et al., 2018) for optimizing the overall pattern of peak forest landform in wulingyuan scenic area is constructed according to local conditions. the research shows that the focus of the construction of peak forest landscape spatial structure in wulingyuan scenic area is to optimize the overall spatial pattern of the landscape. there are fragmentation characteristics in the spatial structure of fenglin landform, which has the advantage of constructing an overall structure optimization model based on the principle of combining centralization with decentralization. (2) designing landscape corridors of wulingyuan scenic area at different scales and levels to give full play to the overall ecological function of the landscape. the holistic concept of “urban integration” and “suburban integration” is not only the basic concept of modern urban spatial structure planning, but also one of the basic principles of peak forest landscape spatial structure in wulingyuan scenic area. high fragmentation is one of the main characteristics of the spatial pattern of peak forest landform in wulingyuan table 5. architectural landscape characteristics of wulingyuan scenic area from 1977 to 2007 particular year 1977 1987 1998 2004 2007 plaque area(ha) 1486.3 1040.1 1113.5 1127.3 1048.1 building density(%) 3.75 2.63 2.81 2.85 2.65 compared with the average annual growth rate of the previous sample(%) -0.12 0.04 -0.02 -0.09 number of patches (number) 1024 643 624 356 543 patch density(number/ ha) 0.69 0.63 0.57 0.33 0.53 average plaque size(ha) 1.4515 1.6176 1.7845 3.1666 1.9302 maximum plaque area(ha) 8.62 24.24 82.31 196.14 122.21 percentage of maximum patch area(%) 0.58 2.33 7.39 17.40 11.67 maximum patch fractal dimension 1.0947 1.1413 1.1659 1.7223 1.1377 43study on optimization method of spatial structure characteristics of peak forest landform in wulingyuan scenic area scenic area, and it is also the main reason affecting the ecological function of peak forest landform. therefore, corridor construction in wulingyuan scenic area of different scales and grades should become the main task of landscape protection and construction. (3) strengthen the construction of landscape patches in key and sensitive areas. wulingyuan peak forest landform is a weak area in the construction and protection of spatial pattern. therefore, wulingyuan scenic area should not only increase the total amount of forest to balance the spatial distribution of forest landscape and improve the ecological quality of landscape, but also plan and build a large near-natural forest landscape patch in the peak forestland form area. (4) strictly protect artificial landscape patches and improve the quality of artificial landscape ecosystem. large artificial landscape patches with high naturalness not only play a key role in maintaining the ecological environment quality and protecting the biodiversity of scenic areas, but also play an irreplaceable role in constructing the overall development model of structural optimization. it is necessary to carry out natural, ecological, diversified and characteristic transformation of artificial landscape ecosystem. discussions the geomorphological characteristics of sandstone peak forest are mainly composed of thousands of stone peaks, pillars, vertical cliffs and grooves, and the main combination forms are columnar and conical peak forest. under the background of intermittent uplift of regional neotectonic movement, the effects of flow erosion cutting, physical weathering, chemical weathering and biological root splitting and other external forces begin with rock joints, which widen the fissures gradually and form the peak forest landform over time. the peak forest landform evolved from the devonian platform and experienced the platform. local mountains, valleys, peaks and bushes, peak forests and other stages of development and evolution, will continue to develop in the future (timbe et al., 2017). based on the topographic characteristics of the peak forest cluster with tree nodes, it is found that the peak forest cluster presents different degrees of self-similarity in different geomorphological types. based on the node sequence method of area weight, the semi-structured contour data are serialized into a chain structure with uniform growth. then the shape characteristics of peak clusters at different scales, angles and levels (liu et al., 2018) are analyzed. the results show that in different geomorphological types, whether measured by roundness coefficient or compactness coefficient, the shape of sub-nodes is closer to the circle than that of parent nodes, and the shape of complex geomorphological units gradually turns into a circle from the outside to the inside. the wulingyuan scenic area has had a profound impact on the landscape in the past 30 years. since the development of tourism in wulingyuan in 1979, tourism and human activities of local residents have become an important factor affecting the landscape of the study area. the remote sensing images in the five periods of 1977, 1987, 1998, 2004 and 2007 are representative of the landscape changes in the study area. 1977 reflects the landscape under the influence of strong farming activities before tourism development; 1987 represents the landscape under the influence of tourism industry just started; 1998 represents the inappropriate planning. under management, tourism activities and local human activities have a profound impact on the landscape of the study area (warned by the world heritage committee in 1998); 2004 is the landscape of the study area after the implementation of the large demolition; 2007 is the landscape of the study area under the pressure of tourism development and the strong influence of local residents. the spatial pattern of sandstone peak forest landform in wulingyuan scenic area from 1977 to 2007 was analyzed. the change of vegetation landscape showed that, despite the pressure of tourism, the vegetation coverage rate increased year by year, the fragmentation degree of patches decreased gradually, and the quality of vegetation patches improved gradually. architectural landscape is the smallest landscape type in the study area, but it is the most profoundly impacted and destroyed landscape type by human activities. it is the disturbing landscape in wulingyuan scenic area. artificial landscape is mainly influenced by human activities rather than land use in architectural form, including paddy field, dry land, wasteland and artificial land, revamped waters, roads and other types (kristensen et al., 2018). the overall feature of the artificial landscape in wulingyuan scenic area is that the area decreases year by year. conclusions sandstone peak forest in wulingyuan scenic area is located at the upper part of the mountain body at an elevation of 1000 meters. its edge is a high cliff, forming a dangerous and exotic landscape. it has the basic forms of cliff, peak (column), cliff trough, cliff lane and so on. the peak forest is composed of thousands of stone peaks, pillars, vertical cliff walls, grooves, etc. the main combination forms are column and conical peak forest, which has important scientific research, popular science and tourism development value. the structure of contour tree can not only visually reflect the hierarchical nesting characteristics of the surface morphology of peak forest clusters, but also contain the relevant information of different levels of geomorphic units, so as to analyze the morphological characteristics of geomorphic units at different scales. the spatial pattern of peak forest landscape in wulingyuan scenic area was analyzed by using various landscape indices. the vegetation landscape increased steadily, the artificial landscape decreased year by year, and the architectural landscape fluctuated, but tended to decrease as a whole. in order to provide better tourism resources and improve the ecological environment protection effect of wulingyuan scenic area, the principles of combining centralization with decentralization, the design of scenic corridors and the strengthening of landscape patch construction were adopted to optimize the spatial structure characteristics of peak forest landform. table 6. artificial landscape characteristics of wulingyuan scenic area from 1977 to 2007 particular year 1977 1987 1998 2004 2007 plaque area(ha) 19254.3 12431.1 10983.5 9697.2 6000.8 proportion of artificial landscape(%) 48.51 31.32 27.67 24.43 15.12 compared with the average annual growth rate of the previous sample(%) -1.73 -0.34 -0.55 -3.11 number of patches (number) 189 519 485 143 259 patch density(number/ ha) 0.0099 0.418 0.0442 0.0147 0.0431 average plaque size(ha) 101.875 23.952 22.646 67.813 23.167 maximum plaque area(ha) 6526.87 2238.57 1333.32 3516.83 2015.81 percentage of maximum patch area(%) 33.90 18.01 12.14 36.27 33.59 maximum patch fractal dimension 2.2725 1.3473 1.3492 1.2444 1.257 44 qian zhang references catania, g., hulbe, c., & conway, h. (2017). grounding-line basal melt rates determined using radar-derived internal stratigraphy. journal of glaciology 56(56), 545-554. changqing, d., luo, d., zhang, c., guo, d., & wang, y. h. (2017). study on screening method of lithium ion power battery. chinese journal of power sources, 41(7), 977–980. cole, s, weinberg, d. h., frenk, c. s., & ratra, b. (2018). large-scale structure in cobe-normalized cold dark matter cosmogonies. monthly notices of the royal astronomical society, 289(1), 37-51. colgan, p. m., amidon, w. h., & thurkettle, s. a. (2017). inland dunes on the abandoned bed of glacial lake chicago indicate eolian activity during the pleistocene-holocene transition, southwestern michigan, usa. quaternary research, 87(1), 66-81. davies, l. j. m., bremer, m. n., stanway, e. r., husband, k., lehnert, m. d., & mannering, e. j. a. (2018). identifying clustering at high redshift through actively star-forming galaxies. monthly notices of the royal astronomical society, 438(4), 2732-2752. fang, h., ji, b., deng, x., ying, j., zhou, g., shi, y., xu, l., tao, j., zhou, y., li, c., & zheng, h. (2018). effects of topographic factors and aboveground vegetation carbon stocks on soil organic carbon in moso bamboo forests. plant and soil, 433(5882), 363-376. goad, m. r., & korista, k. t. (2018). interpreting broad emission-line variations – i. factors influencing the emission-line response. monthly notices of the royal astronomical society, 444(1), 43-61. hart, j. k. (2017). identifying fast ice flow from landform assemblages in the geological record: a discussion. annals of glaciology, 28(1), 59-66. jiang, y. w. (2017). study on the characteristics of complex networks in network user behavior. journal of china academy of electronics and information technology, 12(5), 452-457. klongvessa, p., lu, m., & chotpantarat, s. (2018). response of the flood peak to the spatial distribution of rainfall in the yom river basin, thailand. stochastic environmental research and risk assessment, 32(10), 2871-2887. kristensen, n. p., johansson, j., jonzén, n., & smith, h. g. (2018). evolution of resident bird breeding phenology in a landscape with heterogeneous resource phenology and carryover effects. evolutionary ecology, 32(5), 509-528. liang, w., li, s., & hung, f. (2017). analysis of the contributions of topographic, soil, and vegetation features on the spatial distributions of surface soil moisture in a steep natural forested headwater catchment. hydrological processes, 31(22), 3796–3809. li, q. l., zhang, z. p., shi, w. s., jiang, z., zhao, j., & shi, w. (2018). brain tumor images retrieval method based on spatial pixel intensity. journal of jilin university(science edition), 56(3), 214-218. liu, y. c., huang, y., & li, x.. (2018). research on neutral-point potential of three-level npc inverter based on fuzzy virtual space vector modulation. journal of power supply, 16(1), 61-66. moreno, m., saco, p. m., merino-martin, l., espigares, t., & nicolau, j. m. (2017). landform-water-vegetation feedbacks: exploring ecosystem stability and restoration potential in semiarid hillslopes. indian journal of dermatology venereology & leprology, 78(3), 423-435. payo, a., favis-mortlock, d., dickson, m., hall, j. w., hurst, m. d., walkden, m. j. a., townend, i., ives, m. c., nicholls, r. j., & ellis, m. a. (2017). coastal modelling environment version 1.0: a framework for integrating landform-specific component models in order to simulate decadal to centennial morphological changes on complex coasts. geoscientific model development, 10(7), 1-45. sylvestre, t., copland, l., demuth, m. n., & sharp, m. (2017). spatial patterns of snow accumulation across belcher glacier, devon ice cap, nunavut, canada. journal of glaciology, 59(217), 255-264. timbe, e., feyen, j., timbe, l., crespo, p., célleri, r., windhorst, d., frede, h. g., & breuer, l. (2017). multicriteria assessment of water dynamics reveals subcatchment variability in a seemingly homogeneous tropical cloud forest catchment. hydrological processes, 31(7), 1456-1468. wang, z. w., jin, p., & pu, x. y. (2019). construction of power supply service risk identification model based on clustering analysis. automation & instrumentation, 232(2), 179-182. zhang, x., gang, w., su, q., guo, q., zhang, c., & chen, b. (2017). an improved fuzzy algorithm for image segmentation using peak detection, spatial information and reallocation. soft computing, 21(8), 2165-2173. http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201705003 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201705003 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dzkxjspl201705003 about:blank about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=jldxzrkxxb201803035 about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=dyxb201801010 about:blank http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 http://www.wanfangdata.com.cn/details/detail.do?_type=perio&id=zdhyyqyb201902049 about:blank about:blank issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n3.101477 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (september, 2022): 231 238 a t m o sp h e r ic s c ie n c e s role of ocean-atmosphere interface in annual and semiannual sst cycles in the indian ocean oldemar de oliveira carvalho junior research & project office/instituto ekko brasil/brazil cx postal 20, rua regina burttz clemer (estrada geral urussanga), 3207 – bairro india, aldeia cebb mendjila centro de estudos budistas bodisatva canelinha/sc. 88230-000 ocjunior@ekkobrasil.org.br abstract a long-term analysis of temperature can be used to describe the main mechanisms that operate at the surface of the ocean. the average sea surface temperature (sst) contour plots for the indian ocean are produced based on the world ocean atlas data set (1994). sst, together with the independent variables wind (wx, zonal component of pseudo-stress wind and wy, meridional component of pseudo-stress wind), net-down-fresh-water-flow (ndff), and ekman pumping, are included in a multiple regression analysis to define the relative importance of each one of these variables in the physical processes at the surface of the indian ocean. the ndff data set is based on coads (comprehensive ocean-atmosphere data set). the wind data is obtained from the florida state university (fsu). the harmonic terms of the variables are calculated, which is considered to be stationary and expressed by a fourier series as a cosine function. the harmonic terms are multiplied by the maximum amplitude of the variables and then added to their mean annual values. the isotherms are mainly meridional along the western boundary, but zonal in the southern indian ocean. the annual component is seen to have a maximum in july, summer monsoon (sw monsoon) and a minimum in january, during the winter monsoon (ne monsoon). the amplitude of the semiannual component is smaller, with two maxima in may and october and two minima in february and august. the small magnitude of these residuals errors is an indication that the temperature variability during this period and for this area can be explained reasonably well by the two harmonic terms. in the arabian sea, the final regression equations for sst variability show that it is mainly affected by the wx, ekman pumping and ndff. for most of the areas of the bay of bengal, as well as for most of the locations in the southern tropical indian ocean, the entered independent variables can explain sst. two components fit to observation can be used to predict sst together with the regression equations. although harmonic analysis can be used to study sst variability, a multiple regression analysis is required to identify and quantify the variables related to areas of large annual and semiannual variability. different techniques are therefore used together to provide more reliable results in sst configuration in the indian ocean. keywords: harmonic analysis, multiple regression, ocean circulation. papel de la interfaz océano-atmósfera en los ciclos anuales y semestrales de la tsm en el océano índico resumen este trabajo parte de la posiblidad de utilizar un análisis a largo plazo de la temperatura para describir los principales mecanismos que operan en la superficie del océano. los diagramas de contorno de la temperatura promedio de la superficie del mar (tsm) para el océano índico se produjeron con base en el world ocean atlas data set (1994). el tsm, junto con las variables independientes viento (wx, componente zonal del viento de pseudo-estrés y wy, componente meridional del viento de pseudo-estrés), flujo neto de agua dulce (ndff) y bombeo de ekman se incluyeron en un análisis de regresión múltiple para definir la importancia relativa de cada una de estas variables en los procesos físicos en la superficie del océano índico. el conjunto de datos ndff se basó en el coads (comprehensive ocean-atmosphere data set). los datos de viento se obtuvieron de la universidad estatal de florida (fsu). se calcularon los términos armónicos de las variables, que se consideran estacionarios y se expresaron mediante una serie de fourier como función coseno. los términos armónicos se multiplicaron por la amplitud máxima de las variables y luego se sumaron a sus valores medios anuales. las isotermas son principalmente meridionales a lo largo del límite occidental, pero zonales en el sur del océano índico. se considera que el componente anual tiene un máximo en julio, monzón de verano (monzón del sw) y un mínimo en enero, durante el monzón de invierno (monzón del ne). la amplitud del componente semestral es menor, con dos máximos en mayo y octubre y dos mínimos en febrero y agosto. la pequeña magnitud de estos errores residuales es una indicación de que la variabilidad de la temperatura durante este período y para esta área puede explicarse razonablemente bien por los dos términos armónicos. en el mar arábigo las ecuaciones de regresión finales para la variabilidad de la sst muestran que se ve afectada principalmente por el wx, el bombeo de ekman y el ndff. para la mayoría de las áreas de la bahía de bengala, así como para la mayoría de las ubicaciones en el sur del océano índico tropical, las variables independientes ingresadas pueden explicar la tsm. se pueden usar dos componentes ajustados a la observación para predecir la tsm junto con las ecuaciones de regresión. aunque el análisis armónico se puede utilizar para estudiar la variabilidad de la tsm se requiere un análisis de regresión múltiple para identificar y cuantificar las variables relacionadas con áreas de gran variabilidad anual y semestral. por lo tanto, se utilizan diferentes técnicas para proporcionar resultados más confiables en la configuración de la tsm en el océano índico. palabras clave: análisis armónico; regresión multiple; circulación oceánica; record manuscript received: 21/03/2022 accepted for publication: 16/09/2022 how to cite item: carvalho-junior, o. o. (2022). role of ocean-atmosphere interface in annual and semiannual sst cycles in the indian ocean. earth sciences research journal, 26(3), 231-238. https://doi.org/10.15446/ esrj.v26n3.101477 https://doi.org/10.15446/esrj.v26n3.101477 mailto:ocjunior@ekkobrasil.org.br https://doi.org/10.15446/esrj.v26n3.101477 https://doi.org/10.15446/esrj.v26n3.101477 232 oldemar de oliveira carvalho junior 1. introduction until recently, the oceans have been represented as a single large body with no large-scale temporal variability. indeed, temporal variability represents an additional difficulty when analyzing features of the ocean, such as eddies, currents, water masses, and primary production. sea-surface temperature (sst), for example, has been observed by satellites such as those of the series noaa (national oceanic and atmospheric administration). measuring sst is important as it is used in the study of oceanic physical processes such as frontal movements, water masses, currents, and upwelling zones (barret, 2013; khan et al., 2021). the present work intends to explore the sst of the tropical indian ocean as a major supplement to the more traditional procedures, and to examine the contribution made to an understanding of ocean circulation and water masses at the surface in this region. water masses also exist in the surface layer of the ocean. the definition of these water masses in terms of hydrographic properties is not an easy task because they undergo large seasonal and interannual variations. furthermore, surface water characteristics differ distinctly from one oceanic region to another. therefore, knowledge of surface ocean temperature can be important in helping to define water masses at the surface. a water mass which originates at the surface of the ocean is always a result of dynamic interactions between sea and atmosphere. since a sub-surface water mass usually originates from a well-defined formation region, it is important to know in detail which processes are involved in the water mass formation. physical and biological properties of surface waters, such as temperature, are constantly changing because of atmospheric and oceanic seasonal variations (khan et al., 2021). the proximity of land can also be an important factor affecting water masses characteristics, especially if there is a freshwater output from river runoff, such as the ganges-brahmaputra-meghna river catchment in the northern bay of bengal (wang et al., 2022). the identification and representation of the physical processes responsible for temperature distribution are important steps towards the understanding and management of marine ecosystems. the present study attempts to understand the relationship between regional circulation and the variability of temperature and its distribution at the surface of the indian ocean. long-term analysis of sst can be particularly useful to describe the main mechanisms that operate at the surface of the ocean, such as formation of water masses at the surface. the study of temperature in the oceans play a major role in the analysis of global warming. the interaction between atmosphere and sst is important to monitor and predict climate change (wang et al., 2022). the monsoon regime of the indian ocean, for example, is highly dependent on the sst (alory et al., 2007; huang et al., 2021; liao and wang, 2021; khan et al., 2021). khan et al (2021), define a four stages cycle for the sst in the tropical indian ocean, with a warming stage from february to may, a cooling from may to august, warming again from august to october and another cooling from october to january. two main regions can be distinguished in the indian ocean, the northern and the southern indian ocean (mcphaden, 1982; tomczak and godfrey, 1994). the northern indian ocean comprises three regions: the equatorial region (10° n to 10° s), bay of bengal and arabian sea. the great seasonal changes observed in the northern indian ocean are commonly attributed to the presence of the eurasian land mass and the absence of temperate and polar regions (shetye et al., 1994; singh and roxy, 2022). these result in a strong seasonal difference in the distribution of centers of pressure, which originates the monsoons. the boundary of the southern indian ocean to the south is not clear, and some researchers prefer to include the antarctic circumpolar current. three general circulation systems are defined to the indian ocean: the seasonal monsoon gyre, the subtropical anticyclonic gyre of the southern hemisphere, and the antarctic waters with the circumpolar current (wyrtki, 1973; reid, 2003; naqvi, 2008) (figure 1). the fact is that it is the seasonal reversal of the monsoonal gyre that gives the indian ocean its principal and unique characteristics. figure 1. surface currents in the indian ocean. (a) ne monsoon (dec-april); (b) sw monsoon (jun-oct). eac = east arabian coast; sjc = south java current; zc = zanzibar current; emc = east madagascar current; sc = somali currents; stf = subtropical front; saf = sub-antarctic front; pf = antarctic polar front; wgb = weddell gyre boundary. source: tomczak and godfrey (1994). during the ne monsoon, the main water flow in the equatorial region of the northern hemisphere is from east to west. north equatorial current turns southwards along the somali coast and, together with the zanzibar current, flows eastwards on both sides of the equator (equatorial counter current). some water from the equatorial counter current returns as the south equatorial current (zanzibar current) while the residue travels southeast as the java current (schott and mccreary, 2001; semba et al., 2019; jacobs et al., 2020). the sw monsoon starts in april with a northward flow along the somali coast. already in may, an eastward flow is well pronounced north of the equator. around july, the somali current becomes stronger and the counter current at the equator is replaced by the sw monsoon current. below the equator, the south equatorial current, coming from the east, turns north along the african 233 role of ocean-atmosphere interface in annual and semiannual sst cycles in the indian ocean coast and becomes the somali current (tomczak and godfrey, 1994). this current has an important effect on the local baroclinic adjustment. this work describes the temperature distribution in the surface layer of the tropical indian ocean (25° n to 30°s and 30° e to 120° e) based on analysis of semiannual and annual amplitudes of temperature and a multiple regression analysis between temperature, ekman pumping, components of pseudo-stress wind, wind magnitude and ndff. it is expected that this study can contribute to the definition of water masses on the surface of the indian ocean. a long-term data set is explored to a large-scale examination of the sst of the tropical indian ocean as a major supplement to the more traditional procedures, and to examine the contribution made to an understanding of ocean circulation in this region. the process aims to examine how the two components of the harmonic fit contribute and interact to produce the observed and modelled data cycle. 2. data and methodology a set of five degree by five-degree grid average monthly sst contour plots for the indian ocean is produced based on the world ocean atlas data set (1994). the aim is to use the avhrr mcsst/noaa satellite data to estimate a matrix, which will be then matched to another matrix originated for the world ocean atlas data set. all the temperature data are thus arranged on a fivedegree latitude-longitude grid, only for the surface of the oceans (zero meter) with a final spatial resolution of 500 km2. temperature, wind, ndff, and ekman pumping are included in a multiple regression analysis to define the relative importance of each one of these variables in the physical processes at the surface of the indian ocean. the ndff data set is based on coads (comprehensive ocean-atmosphere data set) and obtained from oberhuber (1988). it is originally organized in a 2° × 2° grid and averaged in 5° x 5° grid for this research. the ndff represents the balance between precipitation and evaporation, with a unit of mm. the wind data is obtained from the florida state university (fsu) wind stress data set for the tropical indian ocean in a 1° ×1° grid. the data is averaged in a 5° x 5° grid. the fsu data set is based on coads and ncei (national centers for environmental information). the wind stress is presented as pseudo-stress wind, which is the wind component multiplied by the wind speed, at a reference height of 10 m. for temperature, the twelve-monthly mean values are subjected to a harmonic analysis to determine the two harmonic components, one having a period of one year and the other a period of six months. the process aims to examine how the two components contribute and interact to produce the more complex cycle of the observed and modelled data. the harmonic terms are considered to be stationary, as the averages and corresponding autocorrelations are not expected to vary significantly in a long period of time. the fourier series for the stationary time series of temperature can then be expressed symbolically as a cosine function, in which the first and second harmonic terms are multiplied by the maximum amplitude of the variable and added to the mean annual parameter. the modelled data can thus be expressed as: o(t) = ō + a × cos (ωa ×t-a) + b × cos (ωs × t b) (katznelson. 2004), where: o(t) = observed value as a function of time when t is time in days, since the start of an arbitrary year, ō = mean value over the period analyzed, a = amplitude of the annual harmonic component, b = amplitude of the semiannual harmonic component, a = phase-lag of the annual component, b = phase-lag of the semiannual component, ωa = angular frequency (2π/365.25), which represents the increment in phase per unit time, t, of the annual harmonic component, ωs = ditto for the semiannual component but (2π/182.62), (ωa × t-a) = phase of the annual harmonic, (ωs ×t b) = phase of the semiannual harmonic. values of o, a, b, a and b are to be determined by analysis. they represent the average annual temperature and are a positive constant. the amplitude value (a or b) is also a positive constant that represents the amplitude of the motion and so its maximum departure from the mean value as the term oscillates within a range of + or -. phase-lag is expressed in degrees, with 30° per month for the first harmonic and 60° per month for the second. phase-zero implies that a maximum occurs at the mid of january. it expresses the average time when the maximum of sst occurs within the annual cycle. for example, the phase of 120° corresponds to the mid of april, 150° to the mid of may, and 270° to the mid of october. the sum of square residual is defined as: ssres = ∑ (y y’)2 and represents the sum of squared differences between observed (y) and the predicted values (y’), these differences representing the small inadequacies of the fit. r-squared describes how well the temperature is explained by the combination of the independent variables as a single unit and is expressed as percentage. r squared ranges between 0 and 1 and is represented by r2=1-ssres/ss total, where ssres is the sum of the squares of the residuals and ss total is the total sum of the errors. the closer to one the better. therefore, r-squared is the percent of variance explained by the temperature model. in other words, r-squared is the fraction by which the variance of the errors is less than the variance of the dependent variable. 3. results the 5° x 5° grid average sst contour plots during the sw monsoon (july, august, and september) can be seen in figure 2. in the northern indian ocean, the isotherms are mainly meridional along the western boundary, within the range of the somali current which runs from south to north during the sw monsoon. figure 2. average sst (°c) based on a 5° x 5° degrees grid for the sw monsoon (july, august, and september). data source: world ocean atlas data set (1994). a zonal distribution of temperature is observed in the southern indian ocean, and a tongue of high temperatures above 28 °c is present along the equator and east of 60° e. the influence of the tongue extends northwards into the bay of bengal and west of india, as can be seen from the 28 °c and 28.5 °c isotherms. high temperatures, close to 29 °c, are observed in the north bay of bengal. the upwelling areas along the somali and arabian coasts are marked by temperatures below 26 °c. the isotherms in the western region of the southern indian ocean, around madagascar, deflect slightly northwards and southward because of the south equatorial current, which runs from east to west and splits into two branches in this area. figure 3. superposition of the main currents and isotherms in sw monsoon, with reference to august. 234 oldemar de oliveira carvalho junior figure 3 is a graphic representation that relates figure 1 and figure 2, superimposing the regional circulation of the indian ocean with the distribution of isotherms. the objective is to provide a synoptic view of the study area, highlighting the importance of sst in the physical processes related to the monsoon regime. as can be seen, the isotherms reliably represent most of the oceanographic features in the study area, as the south equatorial current, which can be associated with the intensification of the zonal temperature gradient in the region. warm water from the pacific into the indian ocean is highlighted in yellow. the distribution of sst contours on a 5° x 5° grid average for january, february, and march, during the ne monsoon, can be observed in figure 4. during this time of the year, the circulation of the area is dominated by the north and south equatorial currents, as a consequence of the westerly winds blowing along the equatorial belt. since the northeast trades are strong because of the high-pressure system established in the asiatic region, temperature contours tend to be meridional (south-northeast direction) north of the equator. figure 4. average sst (°c) based on a 5° x 5° grid for the ne monsoon (january, february and march). data source: world ocean atlas data set (1994). figure 5, like figure 3, is a graphic representation that associates figure 1 and figure 4, superimposing the regional circulation of the indian ocean with the distribution of isotherms during the ne monsoon. it is also clear to see the warm water tongue coming from the pacific, shown in yellow. figure 5. superposition of the main currents and isotherms in ne monsoon, with reference to the month of february. comparing figures 3 and 5, it is clear that the main features, such as the south equatorial current and the meridional isotherms in the persian gulf, are present. the difference is in the intensity. temperature along the west coast of the indian ocean during the ne monsoon is more uniform, varying from 26 to 28 °c, and no strong temperature gradients are observed in the region along the somali and arabian coasts. below 10° s, the isotherms deflect slightly southward in the coastal areas of the southern indian ocean, madagascar, and the australian coast. there are two main currents in this region: the south equatorial current and the subtropical gyre. these result in an anticyclonic circulation with a mainly westward direction between 60° e and 100° e. to the west, the flow is southward and along the madagascar coast. in the eastern indian ocean, close to australia, the surface circulation is initially southward, before switching to a westward direction. the influence of these major circulation systems is seen in the curved shape of the temperature contours. the temperature variation through an average year, as expressed through the amplitude of the annual (first) and semiannual (second) harmonic, can be seen in figure 6. the smallest amplitude values of the annual and semiannual components (below 0.5 °c) are found along the equator, especially in the eastern part. here, the sst exceeds 28 °c throughout the year (figures 6a and 6c). the largest amplitude values are found in the western indian ocean, the arabian sea, the bay of bengal, and south of 20° s. figures 6b and 6d show the phase lags for the first and second harmonics of the sst. phase-lag is expressed in degrees and increments at 30° per month for the first harmonic and 60° per month for the second, where phase is zero it implies that a maximum occurs in mid-january. it expresses the average time when the maximum of sst occurs within the annual cycle. for example, phase of 120° corresponds to mid of april, 150° mid of may and 270° mid of october. they indicate the time of the year when the maximum amplitude occurs. figure 6. (a) distribution of sst amplitude for the first harmonic. (b) distribution of sst phase-lag for the first harmonic. (c) distribution of sst amplitude for the second harmonic. (d) distribution of sst phase-lag for the second harmonic. annual and semiannual amplitudes of sst are expressed in °c. amplitude and phase of temperature are calculated from the world ocean atlas data set (1994). figure 7a shows the observed data and the two harmonic components fit to the observations. figure 7b shows the two individual harmonic components and their interaction through the year in the arabian sea. the annual component is seen to have a maximum in july (sw monsoon) and a minimum in january (ne monsoon) which suggests a connection with a hot summer and a cold winter, likely due to proximity to a continental land mass in these latitudes. compared with the annual component, the amplitude of the semiannual component is smaller, with two maxima in may and october and two minima in february and august (figure 7b). a similar pattern is observed in the bay of bengal, but the two maxima (april and october) and the two minima (january and july) occur earlier than in the arabian sea. the overall fit is good, with a residual sum of squares of 0.39. in fact, the residual errors between observed and harmonic temperature are small, with a maximum of 0.32ºc during june (figure 7c). the small magnitude of these residuals suggests that the temperature variability during this period and for this particular area can be explained reasonably well by just the two harmonic terms. the sum of square residual can be defined as ssres = ∑ (y y’)2 and represents the sum of squared differences between observed (y) and the predicted values (y’). these differences represent the small inadequacies of the fit. mcphaden (1982) applied residual variance from regression techniques to analyze equatorial waves in the indian ocean. another study based on correlations between amplitudes and phases used a wind-driven model for the north pacific (white, 1978). 235 role of ocean-atmosphere interface in annual and semiannual sst cycles in the indian ocean figure 8 compares the observed and two components harmonic fit of the somali coast (52.5º e and 7.5º n) sea surface temperature (sst), the annual and semiannual components, and the residual errors. the highest annual component occurs in march/april. during this period, the area is affected by the north equatorial current and the anticyclonic flow, which are responsible for the input and re-circulation of warm equatorial waters into the arabian sea. the minimum of the annual component occurs in september/october, when the somali current represents the strong northward flow observed along the western arabian sea (figure 8a). the two peaks of the semiannual component occur in april/may and october/november, transition months between the monsoons, when wind stress is weaker and less evaporative cooling and mixing are expected (figure 8b). these result in the high temperature observed during april and may. the two minima occur in january/february and july/august. the first is associated with the peak of the ne monsoon and a slight winter cooling. the increase in wind speed, which results in large evaporative cooling and mixing, can explain the secondary low peak of temperature during this period. the second minimum that occurs during july/august can be explained by the strong wind stress and upwelling that occurs during this time of the year. since the annual and semiannual components are acting almost in conjunction, it is expected that the wind monsoon drive circulation is also contributing to this peak of low temperature. the semiannual (-1.4) and annual (-0.86) signals are both strong during august. the low residual errors (maximum of 0.38 °c in january) show a good agreement between observed and the two components harmonic fit, with a residual sum of squares of 0.4 (figure 8c). figure 9a compares the observed sst with the two components harmonic fit at a position on the equator, 62.5º e and 2.5º s. it is possible to see how the harmonic contribution due to the interaction between the annual and semiannual components affects the two harmonic components fit (figure 9b). the maximum of the annual component occurs in april and the minimum in october, which suggests that the variability of solar radiation is connected to this. the two maxima of the semiannual component are observed for april/may and october/november, the transition between the monsoons. the low residual errors indicate a good agreement between observed and the two components harmonic fit (figure 9c). figure 7: (a) observed and two components fit to observations (°c). (b) first and second components of the harmonic analysis. (c) residual plot between observed and harmonic temperature (°c). time (months) in which 13 is equal to 1 and corresponds to mid-january of the following year. figure 8. (a) observed and two components fit to observations (°c). (b) first and second components of the harmonic analysis. (c) residual plot between observed and harmonic temperature (°c). time (months) in which 13 is equal to 1 and corresponds to mid-january of the following year. figure 9. (a) observed and two components fit to observations (°c). (b) first and second components of the harmonic analysis. (c) residual plot between observed and harmonic temperature (°c). time (months) in which 13 is equal to 1 and corresponds to mid-january of the following year. 236 oldemar de oliveira carvalho junior the final regression equations for temperature variability in the arabian sea show that temperature is mainly affected by the wx, ekman pumping and ndff (table 1). components of wind stress, wind magnitude, ekman pumping and ndff are impacting the temperature, but the wx is a constant presence in all of the equations. table 1. predicted equations for the eastern arabian sea. location equation adjusted r2 67.5°e/12.5°n t=29.65+0.15ekman-0.026ndff0.037wmg+0.072wx+0.106wy 0.77 72.5°e/12.5°n t=30.54+0.16ekman0.03ndff+0.12wx+0.26wy 0.67 62.5°e/7.5°n t=27.65+0.05ekman-0.04ndff0.05wmg+0.08wx 0.74 wx = zonal component of pseudo-stress wind (m2s-2), wy = meridional component of pseudo-stress wind (m2s-2), ndff = net-down-fresh-waterflux (mm), ekman pumping = (-) downward/(+) upward (x10-3 kg.m2.s-1), t = temperature (°c), wmg = wind magnitude (ms-1). the final regression equations and their respective adjusted r-squared values for the bay of bengal are shown in table 2. for most of the areas of the bay of bengal, temperature can be explained by the entered independent variables. the meridional component of the wind stress (wy) is one of the main variables that affects temperature. all final regression equations have larger values of beta coefficients for this variable. it is followed by the zonal components of the wind stress (wx) and ndff, in order of importance. table 2. predicted equations for temperature at the bay of bengal. location equation adjusted r2 82.5°e/17.5°n t=26.85+0.013ndff-0.025wmg0.083wx+0.097wy 0.95 87.5°e/17.5°n t=27.24+0.005ndff0.096wx+0.116wy 0.83 92.5°e/17.5°n t=28.35-0.036wmg0.074wx+0.116wy 0.71 82.5°e/12.5°n t=25.890.07ekman+0.02ndff+0.06wmg0.13wx+0.17wy 0.95 92.5°e/12.5°n t=29.12-0.04wmg+0.04wy 0.61 wx = zonal component of pseudo-stress wind (m2s-2), wy = meridional component of pseudo-stress wind (m2s-2), ndff = net-down-fresh-water-flux (mm), ekman pumping = (-) downward/(+) upward (x10-3 kg.m2.s-1), t = temperature (°c), wmg = wind magnitude (ms-1). the final regression equations for temperature at the equatorial region, with an adjusted r-squared greater than 0.6, can be seen in table 3. the temperature in these areas can be explained by the independent variables. the magnitude of the wind and ndff are the main variables responsible for the temperature variability. table 3. predicted equations for the equatorial region (60° e to 100° e and 10° s to 5° n). location equation adjusted r2 77.5°e/2.5°s t=28.75-0.007ndff-0.04wmg 0.78 77.5°e/7.5°s t=28.16-0.02ekman-0.006ndff0.031wmg 0.81 92.5s/2.5s t=32.35-0.015ndff-0.069wmg 0.71 87.5°e/7.5°s t=28.23-0.02wmg 0.67 wx = zonal component of pseudo-stress wind (m2s-2), wy = meridional component of pseudo-stress wind (m2s-2), ndff = net-down-fresh-water-flux (mm), ekman pumping = (-) downward/(+) upward (x10-3 kg.m2.s-1), t = temperature (°c) and wmg = wind magnitude (ms-1). a set of final regression equations for temperature in the southern tropical indian ocean region with an adjusted r-squared larger than 0.6 is shown in table 4. these equations demonstrate that the variables affecting the temperature vary significantly from one location to another. temperature can be predicted for most of the locations, but the independent variables vary significantly from one location to another. for the locations at 57.5° e/17.5° s, 72.5° e/22.5° s, 87.5° e/17.5° s and 57.5° e/27.5° s, temperature can be predicted from only one independent variable at each location, ndff, ekman pumping, wy, and ekman pumping, respectively. the largest beta coefficients are obtained for wy, ekman pumping and ndff. table 4. predicted equations for the south tropical indian ocean (30° s to 10° s). location equation adjusted r2 57.5°e/17.5°s t=26.35+0.02ndff 0.73 72.5°e/22.5°s t=26.92+0.51ekman 0.79 87.5°e/17.5°s t=29.63-0.12wy 0.90 97.5e/22.5s t=28.45+0.24ekman0.13wx-0.28wy 0.71 57.5e/27.5s t=25.61+0.41ekman 0.76 82.5e/27.5s t=15.69-0.059ndff0.12wx-0.19wy 0.81 107.5°e/17.5°s t=34.00+0.047ndff0.08wy 0.75 wx = zonal component of pseudo-stress wind (m2s-2), wy = meridional component of pseudo-stress wind (m2s-2), ndff = net-down-fresh-waterflux (mm), ekman pumping = (-) downward/(+) upward (x10-3 kg.m2.s-1), t = temperature (°c) and wmg = wind magnitude (ms-1). 4. discussion the amplitudes of the annual component increase with latitude in the north indian ocean and reach maximum values to the north of the arabian sea and bay of bengal. this distribution reflects the response of the ocean to the solar radiation, in which amplitudes increase away from the equator. the reversal of the monsoons, associated with land mass distribution also results in a large contribution to these large amplitudes. the latest occurrence of the annual maximum temperature is observed in the northernmost parts of the arabian sea and bay of bengal, where the maximum of the annual occurs in july. the importance of the solar path is clear in the annual component of temperature, but it also seems that the boundaries have an influence on winter temperatures north of the arabian sea and bay of bengal. it is believed that larger amplitudes result from colder winters over adjacent continental land, due to the contrast between cold air and warmer sea surface water. the maximum semiannual amplitude observed for may and october are associated with the transition months between the sw monsoon and the ne monsoon. during may and october, wind intensity is expected to decrease, which would reduce evaporative cooling and water mixing. this could explain the raise in temperature during these periods. it also seems that the two minima, observed during february and august, can be linked to the peaks of the ne monsoon and sw monsoon, respectively. winter cooling (in the case of the ne monsoon) associated with maximum wind during these periods can result in large evaporative cooling and stirring of the water, which can decrease the sst. this is in line with khan et al. (2021) when studying the monthly cycle of sst in the north arabian sea for 2016/2017, from satellite and argo datasets. the interaction between the annual and the semiannual component of sst results in a depression of the temperature from mid-july to mid-september when the large annual peak of the annual component (summer leading 237 role of ocean-atmosphere interface in annual and semiannual sst cycles in the indian ocean to large solar radiation) act in opposition to the small negative peaks of the semiannual component (wind stress leading to large evaporation and mixing). the pronounced low temperature observed during february could be due to the annual component (low solar radiation during winter) working together with the semiannual component (strong wind, leading to large evaporation and mixing). in the indian ocean, the amplitude of the semiannual component can be comparable to the annual component. in some regions, such as the somali (47.5º e and 7.5º n) and arabian coast (52.5º e and 12.5º n), the semiannual component is considerable larger than the first harmonic with maximum amplitudes around 1.6 °c. the strong semiannual component along the somali and arabian coasts is thought to be associated with coastal upwelling derived from the strong monsoonal wind field acting upon a left-hand coastline from june to october. this semiannual variability of temperature is influenced by wind, since large values of semiannual periodicity also occurs in the western arabian sea (unnikrishnan et al., 1997; khan et al., 2021). previous works, such as that of levitus (1987) and khan et al. (2021) have already demonstrated semiannual cycles of temperature in the indian ocean. these cycles could be related to evaporative cooling and upwelling associated with the monsoons. the two maximum values of the semiannual component, seen in april/ may and october/november, are a result of the weaker wind stress during these periods, when less evaporative cooling and mixing are expected. these result in the high temperature observed during april and may. the first semiannual minima, observed in january/february and july/august, is associated with the peak of the ne monsoon and a slight winter cooling. the increase in wind magnitude, which results in large evaporative cooling and mixing, can explain the secondary low peak of temperature during this period. the second minimum that occurs during july and august can be explained by the strong wind stress and upwelling that occur during this time of the year. it is expected that the wind monsoon drive circulation is also contributing to this peak of low temperature, since the annual and semiannual components are acting almost in conjunction. one of the main characteristics of the equatorial indian ocean is its lack of upwelling. as a result of the dominance of the meridional orientation of the wind, ekman transport does not result in divergence along the equatorial region and, therefore, no upwelling occurs (tomczak and godfrey, 1994). however, the overall pattern with a secondary low during august remains, even where no upwelling is expected. the low will be located along the equatorial region, away from the coast and west of 70º e. the two maxima of the semiannual component, observed in april/ may and october/november, can be associated with weaker winds during these months, which result in less evaporative cooling and mixing. the larger temperature value for april can be attributed to the fact that the annual component acts with the semiannual component, and the opposite for october/ november. for example, during october and november, the annual (negative) and semiannual (positive) components act in opposition. the semiannual component (0.37 for october and november) is similar in magnitude to the annual but opposite in sign (-0.58 for october and -0.51 for november) and results in the secondary high-temperature peak observed during these months. the annual and semi-annual components, acting together from march to june, enhance the temperature during this period, with a maximum temperature peak during april. the distribution of the residual between the observed and the two harmonic fit reveals that the harmonic model is meaningful, as the maximum residuals are observed from february to may and are below 0.12 °c. in fact, only to the west of 70° e along the equator is the second harmonic temperature larger than 0.5 °c, with the two maxima occurring during april and november. the amplitude becomes weak eastward, along the equator, and, therefore, the seasonal variations are considered to be insignificant in this area. the low amplitude values east of 70º e are a result of an expected low rate of exchange between ocean and atmosphere near the equator, for example the lack of wind when compared to the higher latitudes. in the southern hemisphere, the maximum amplitude of temperature of the annual component occurs late january along the south coast of africa. the annual component increases eastwards, reaching a maximum during early march in australia. the annual component increases gradually towards the north, following the solar path, as shown by the phase-lag distribution. a gradual phase-lag shift (30° to 130°) from the southern to northern hemisphere is observed in the tropical region from 10° s. 5. conclusion large annual variations in temperature were observed in the north of the arabian sea and bay of bengal, in the western equatorial region and in the southern tropical indian ocean. large semiannual fluctuations in temperature were observed along the western boundary of the indian ocean, from 10° s to 25° n. based on multiple regression analysis, it was possible to identify and quantify the variability of the temperature. for example, the temperature within upwelling regions along the somali and arabian coasts could be explained by the variability of the components of the pseudostress wind, wind magnitude and ekman pumping, with an adjusted r-squared ranging from 0.64 to 0.84. the annual component, overall, is related to large-scale features of the seasonal solar radiation input and the seasonal monsoons. basic features associated with the monsoons, such as wind-driven advection, can have a significant influence on the annual term. the semiannual component is influenced by local and regional factors, which add to the basic pattern defined by the annual term. these local and regional influences are the result of different processes, such as upwelling and the non-directional effect of wind stress (e.g., evaporation, precipitation, and mixing). this study is expected to contribute to the understanding of the indian ocean, especially regarding the processes involved in the interface between the atmosphere and the ocean surface. this understanding may also contribute to the study of the water masses at the surface of the indian ocean and their expected impacts on climate change in the region. acknowledgments this study was undertaken with the support of the flinders university of south australia. i would like to express my deepest gratitude to matthias tomczak, who will always be a source of inspiration and support. references alory, g., wijffels, s., & meyers, g. (2007). observed temperature trends in the indian ocean over 1960–1999 and associated mechanisms. geophysical research letters, 34(2). https://doi.org/10.1029/2006gl028044 barret, e. c. (2013). introduction to environmental remote sensing (1a edição). routledge. carvalho-junior, o. (2022). chlorophyll distribution and variability at the surface of the indian ocean. journal of research in environmental and earth sciences, 8(3), 01–10. http://www.questjournals.org/jrees/current-issue.html huang, b., liu, c., banzon, v., freeman, e., graham, g., hankins, b., smith, t., & zhang, h.-m. (2021). improvements of the daily optimum interpolation sea surface temperature (doisst) version 2.1. journal of climate, 34(8), 2923–2939. https://doi.org/10.1175/jcli-d-20-0166.1 jacobs, z. l., jebri, f., raitsos, d. e., popova, e., srokosz, m., painter, s. c., nencioli, f., roberts, m., kamau, j., palmer, m., & wihsgott, j. (2020). shelf-break upwelling and productivity over the north kenya banks: the importance of large-scale ocean dynamics. journal of geophysical research: oceans, 125(1), e2019jc015519. https://doi. org/10.1029/2019jc015519 katznelson, y. (2004). an introduction to harmonic analysis (3rd revised ed. edição). cambridge university press. khan, s., piao, s., khan, i. u., xu, b., khan, s., ismail, m. a., & song, y. (2021). variability of sst and ild in the arabian sea and sea of oman in association with the monsoon cycle. mathematical problems in engineering, 2021, e9958257. https://doi.org/10.1155/2021/9958257 khan, s., piao, s., zheng, g., khan, i. u., bradley, d., khan, s., & song, y. (2021). sea surface temperature variability over the tropical indian ocean during the enso and iod events in 2016 and 2017. atmosphere, 12(5), 587. https://doi.org/10.3390/atmos12050587 https://doi.org/10.1029/2006gl028044 http://www.questjournals.org/jrees/current-issue.html http://www.questjournals.org/jrees/current-issue.html https://doi.org/10.1175/jcli-d-20-0166.1 https://doi.org/10.1029/2019jc015519 https://doi.org/10.1029/2019jc015519 https://doi.org/10.1155/2021/9958257 https://doi.org/10.3390/atmos12050587 238 oldemar de oliveira carvalho junior levitus, s. (1987). a comparison of the annual cycle of two sea surface temperature climatologies of the world ocean. journal of physical oceanography, 17(2), 197–214. https://doi.org/10.1175/1520-0485(1987)017<0197:acotac>2.0.co;2 liao, h., & wang, c. (2021). sea surface temperature anomalies in the western indian ocean as a trigger for atlantic niño events. geophysical research letters, 48(8), e2021gl092489. https://doi.org/10.1029/2021gl092489 mcphaden, m. j. (1982). variability in the central equatorial indian ocean. i. ocean dynamics. journal of marine research, 40(1), 157–176. naqvi, s. w. a. (2008). the indian ocean. elsevier. https://drs.nio.org/xmlui/ handle/2264/1498 oberhuber, j. m. (1988). an atlas based on the coads data set: the budgets of heat, buoyancy and turbulent kinetic energy at the surface of the global ocean. hamburg, max-planck-institut fur meteorologie. reid, j. l. (2003). on the total geostrophic circulation of the indian ocean: flow patterns, tracers, and transports. progress in oceanography, 56(1), 137– 186. https://doi.org/10.1016/s0079-6611(02)00141-6 schott, f. a., & mccreary, j. p. (2001). the monsoon circulation of the indian ocean. progress in oceanography, 51(1), 1–123. https://doi.org/10.1016/ s0079-6611(01)00083-0 semba, m., lumpkin, r., kimirei, i., shaghude, y., & nyandwi, n. (2019). seasonal and spatial variation of surface current in the pemba channel, tanzania. plos one, 14(1), e0210303. https://doi.org/10.1371/journal. pone.0210303 shetye, s. r., gouveia, a. d., & shenoi, s. s. c. (1994). circulation and water masses of the arabian sea. proceedings of the indian academy of sciences earth and planetary sciences, 103(2), 107–123. https://doi.org/10.1007/ bf02839532 singh, v. k., & roxy, m. k. (2022). a review of ocean-atmosphere interactions during tropical cyclones in the north indian ocean. earth-science reviews, 226, 103967. https://doi.org/10.1016/j.earscirev.2022.103967 tomczak, m., & godfrey, j. s. (1994). regional oceanography: an introduction. pergamon. unnikrishnan, a. s., prasanna kumar, s., & navelkar, g. s. (1997). large-scale processes in the upper layers of the indian ocean inferred from temperature climatology. journal of marine research, 55(1), 93–115. https:// doi.org/10.1357/0022240973224490 wang, y. v., larsen, t., lauterbach, s., andersen, n., blanz, t., krebs-kanzow, u., gierz, p., & schneider, r. r. (2022). higher sea surface temperature in the indian ocean during the last interglacial weakened the south asian monsoon. proceedings of the national academy of sciences, 119(10), e2107720119. https://doi.org/10.1073/pnas.2107720119 white, w. b. (1978). a wind-driven model experiment of the seasonal cycle of the main thermocline in the interior midlatitude north pacific. journal of physical oceanography 8, 818-824. wyrtki, k. (1973). physical oceanography of the indian ocean. in: b. zeitzschel & s. a. gerlach (eds.). the biology of the indian ocean (pp. 18–36). springer. https://doi.org/10.1007/978-3-642-65468-8_3 https://doi.org/10.1175/1520-0485(1987)017%3c0197:acotac%3e2.0.co;2 https://doi.org/10.1175/1520-0485(1987)017%3c0197:acotac%3e2.0.co;2 https://doi.org/10.1029/2021gl092489 https://drs.nio.org/xmlui/handle/2264/1498 https://drs.nio.org/xmlui/handle/2264/1498 https://doi.org/10.1016/s0079-6611(02)00141-6 https://doi.org/10.1016/s0079-6611(01)00083-0 https://doi.org/10.1016/s0079-6611(01)00083-0 https://doi.org/10.1371/journal.pone.0210303 https://doi.org/10.1371/journal.pone.0210303 https://doi.org/10.1007/bf02839532 https://doi.org/10.1007/bf02839532 https://doi.org/10.1016/j.earscirev.2022.103967 https://doi.org/10.1357/0022240973224490 https://doi.org/10.1357/0022240973224490 https://doi.org/10.1073/pnas.2107720119 https://doi.org/10.1007/978-3-642-65468-8_3 toc459375905 toc459375911 toc459375912 toc4285427272 toc428542730 toc428542733 toc428542736 _goback keywords: complex geological environment; gear and bearing; fault diagnosis; palabras clave: entorno geológico complejo; engranaje y cojinete; diagnóstico de fallas. how to cite item wang, s., liu, s. y., & meng, f. (2021). fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment. earth sciences research journal, 25(1), 115-122. doi: https://doi.org/10.15446/ esrj.v25n1.93184 the traditional research method of fault diagnosis mechanism has poor stability, which leads to the difference of fault diagnosis and location results. therefore, under the complex geological environment, a new research method of fault diagnosis mechanism of gear and bearing for coal mining equipment is proposed. this method calculates gears and bearings’ yield strength by analyzing coal mining equipment’s bearing capacity elasticity. according to the fitting degree, the equipment sample’s projection space is confirmed, the fault features of gear and bearing are extracted by segmentation algorithm, the optimal fitness is set by positioning algorithm, the location of fault center is obtained, and the fault mechanism diagnosis is studied. experimental results show that compared with the traditional method, the proposed method is more stable, and the difference in fault diagnosis results is minimal. it can be seen that this method is more suitable for fault diagnosis of coal mining equipment. abstract fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment diagnóstico de fallas de engranajes y cojinetes para equipos de minería de carbón en entornos geológicos complejos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n1.93184 el método de investigación tradicional del mecanismo de diagnóstico de fallas tiene poca estabilidad, y puede conllevar a diferencias entre el diagnóstico de fallas y los resultados de ubicación. por lo tanto, en un entorno geológico complejo se propone un nuevo método de investigación del mecanismo de diagnóstico de fallas de engranajes y cojinetes para equipos de minería de carbón. este método calcula el límite elástico de engranajes y cojinetes al analizar la elasticidad de la capacidad de carga de los equipos de minería del carbón. de acuerdo con el grado de ajuste, se confirma el espacio de proyección de la muestra del equipo, las características de falla del engranaje y el cojinete se extraen mediante un algoritmo de segmentación, la aptitud óptima se establece mediante el algoritmo de posicionamiento, se obtiene la ubicación del centro de falla y se estudia el mecanismo de diagnóstico de falla. los resultados experimentales muestran que, en comparación con el método tradicional, el método propuesto es más estable y la diferencia entre los resultados del diagnóstico de fallas es muy pequeña. se puede ver que este método es más adecuado para el diagnóstico de fallas de equipos de minería de carbón. resumen record manuscript received: 16/03/2020 accepted for publication: 28/11/2020 earth sciences research journal earth sci. res. j. vol. 25, no. 1 (march, 2021): 115-122 shuilin wang1,2*, songyong liu1, fanping meng3 1school of mechatronic engineering, china university of mining and technology, xuzhou, 221116, china 2xuhai college, china university of mining and technology, xuzhou, 221008, china 3yongcheng coal & electricity holding group co. ltd, shangqiu, 476600, china *corresponding author: wslxhxy2008@163.com m in in g e n g in e e r in g https://doi.org/10.15446/esrj.v25n1.93184 https://doi.org/10.15446/esrj.v25n1.93184 https://doi.org/10.15446/esrj.v25n1.93184 116 shuilin wang, songyong liu, fanping meng introduction with the rapid progress of science and technology and the continuous improvement of modernization, coal mining equipment is increasingly developing towards the direction of large-scale, integrated, high-speed, precise and intelligent. with the development trend of mechanical equipment, it is the complexity of equipment. the relationship between different equipment and different parts of the same equipment is more complex, forming a unified whole. when any one of the components fails, not only the operation of the equipment itself, but also the operation of other equipment associated with the equipment will be affected, which will cause a series of chain reactions, resulting in the interruption of the operation of the whole equipment, affecting normal production and product quality. on the other hand, affected by the geological conditions and hydrological environment of coal mining, under the complex natural environment, the mining difficulty of coal mining equipment is greatly increased. excessive operation makes the gears and bearings in the equipment prone to failure. therefore, through real-time monitoring of the working conditions of gears and bearings, the traditional research methods, obtain the operation parameters of parts, and statistics the fault location through the difference method, to find the nodes prone to failure and the time of failure. it is found that the signal of the fault location is unstable due to the poor ability to analyze the fault characteristics, which makes the accuracy of the fault mechanism decline. therefore, a new fault diagnosis mechanism research method is proposed. according to the relationship between gears and bearings, the method resets more stable fault feature extraction and location according to the acquisition method of cyclostationary signal and projection decomposition operation method to ensure the reliability of fault diagnosis mechanism (morris et al., 2018). this method not only solves the existing problems of the traditional methods, but also provides scientific and technical solutions for the fault diagnosis mechanism research of gears and bearings in other fields, and provides reasonable research methods for the development of national equipment and technological innovation. research method of fault diagnosis mechanism of gears and bearings for coal mining equipment in complex geological environment calculation of yield strength of gears and bearings in complex geological environment on the basis of defining the basic parameters of a complex geological environment, the yield strength of gears and bearings is calculated. as all gears and bearings will not be perfect, under the influence of a complex mining geological environment, the top of gears may buckle within a reasonable range due to the influence of bearing force during operation (balestro et al., 2018). therefore, to determine the directional elasticity of the stability of gears and bearings, the method of angular displacement can be used for directional analysis. under the complicated geological conditions, the equation of angular displacement of gears after bearing the axial force is as follows: β µ µ λ β µ µ λ a a b b b a f a b a b d f b a a b d = + − +( )          = + − +( )               (1) in the formula, f ei d= / represents the linear stiffness of the bearing, f ei d= / , f ei d= / represents the size of the gear, a and b represent the left and right ends of the gear respectively, a and b represent the angular displacement of the two ends of the gear, a and b represent the deformation angle of the two ends of the gear respectively,  represents the lateral displacement value of the bearing, a and b represent the stability function and the bending stiffness coefficient. figure 1 is a schematic diagram of the gear tooth deformation curve under the axial force. in the figure above, it is the displacement of a single gear tooth after deformation, and the critical load can be calculated only according to the rai sed section recessed section force ba force amplification gear tooth top cross section figure 1. schematic diagram of the elastic load capacity of the gear 117fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment balance equation. however, the fault situation shown in the figure is relatively simple. under the condition of bearing and gear tooth deformation, the tooth slot will also be deformed. therefore, the shear balance equation should be established to solve the elastic buckling load. figure 3 is the structural diagram of the gear after the deformation of the teeth and grooves. assuming that there is vertical load on the axis at the upper and lower ends of the structure, it is represented by p1 and p2 (wang et al., 2018). reverse rotation forward rotation support force contact force a b f2 f1 r1r2 figure 2. schematic diagram of the load structure of the gear under the load according to the above figure, in order to facilitate the analysis of the load elasticity of the gear, it is assumed that the a and b ends of gear mi buckle at the same time with the other gear mj in contact with each other, and that aa' and bb' are the contact ends. at this time, the end angles of the cross section in the same layer of gear are the same size and opposite direction, that is, a a a a =− ' , a a b b =− ' . it is known that the k value between the gears is the same, and the bending stiffness coefficients are both a b= and a b= . by default, the gear in the same layer buckle at the same time when losing stability. then, according to formula 1 and the balance condition p p 1 2 = −  , the elastic value of the bearing capacity of the gear slot under the tooth deformation is calculated by using the angle displacement method. f ei k a b a b f ei k a b a b aa m a b bb m b a i j ′ ′ = + − +( )   = + − +( ) 1 2 β β η β β η          (2) in the above formula, k1 represents the elastic value of bearing capacity of contact point aa', k2 represents the elastic value of bearing capacity of contact point bb', eim i represents the transfer force of target gear mi, eim j represents the bearing force of gear mj adjacent to mi,  represents the force balance coefficient, and  represents the force control coefficient. it is known that the sum of equilibrium conditions of node a and node b is 0, and there is f f faa bb' '=− +∆ , that is, f f faa bb' '+ −( )+∆ = 0 , where f is the force of other values. according to the equilibrium condition, let k k k= +( )1 2 2 , thus, the instability condition of gear under high strength operation is as follows: sk f f mhaa bb′ ′ +      −( ) = 2 02 (3) in the formula, s represents the linear stiffness of the gear under the action of external force, m represents a non-zero natural number, and h represents the constraint condition. the trigonometric function is substituted to the above formula, and j = f ei mh/ is made to obtain the buckling equation of gear teeth and bearings: sk f f k aa bb′ ′+ −       + = 2 02j j j tan (4) where  is the buckling evaluation parameter. the above equation is the elastic analysis equation of bearing capacity of gear and bearing. it should be noted that the bearing will also be affected by the horizontal load force, so under the condition of meeting the above formula, the elastic calculation expression of gear shear balance is: ′= × +( ) − +( ) + k ei d a b a b ab b c (5) in the formula, c represents the constant of change under the influence of bending stiffness coefficient a and b, and ei force modulus. when k' is greater than the specified elastic value, it means that the gear is distorted under the influence of load. under the influence of formula 5, the plastic value of the bearing capacity of the false gear is set as f ei d= /, and when the value of f ei d= / exceeds the maximum limit of the gear, the yield strength of the gear is calculated. according to the yield strength value, the critical value of functional failure of gear bearing capacity is confirmed. the basic material composition of the gear is known, and the amplification coefficient of the compressive strength of the material is calculated. the formula is: cdif p p e e = =           1 2 1 2 j (6) in the formula, e1 represents the strain rate and e1 ≥ 30s -1, e2 also represents the strain rate and e2 = 30  10 -6 s-1; p1 represents the dynamic compressive strength of the material under e1, p2 represents the static compressive strength of the material under e2; j= +      −3 4 3 1p u represents the force angle under impact, wherein p3 represents the compressive strength under the total volume of the material (jeong et al., 2018). the increase coefficient of tensile strength is: tdif q q = =           1 2 1 2 τ τ σ (7) in the formula,  1 represents the strain rate under the influence of the force,  1 ≤ 1s-1,  2 represents the strain rate under the influence of the force, and 2 = 10 -6 s-1; q q1 2= represents the dynamic tensile strength of the material composed of gears under  1 , q q1 2= represents the static tensile strength of the material under  2 ;  = 10-6 n-2 represents the force angle, where 118 shuilin wang, songyong liu, fanping meng n q q = +      − 1 8 13 0 1 , q q3 0= represents the standard static tensile strength that the material can bear, q q3 0= represents the static tensile strength of a variable material. by combining the formulas 6 and 7, the yield strength calculation formula is obtained z f m z ej j eff= +                     +( )1 1 0τ ω α ω (8) in the formula, z0 represents the initial yield stress,  represents the strain rate, weff represents the equivalent plastic strain, ej represents the plastic hardening modulus,  represents the hardening parameter, when  = 0 represents the follow-up strengthening, m and j represent the constants respectively. when z < 0.7, it means that the frame gear can bear the current working strength, the gear is in good condition, and the bearing capacity function has not been lost; when 0.7 ≤ z ≤ 0.8, it means that the gear can bear the current working strength, but affected by the geological conditions, at this time, the gear has the situation of tooth section depression or external convex, and part of the bearing capacity function; when e > 0.8, it means that the gear is seriously deformed or the gear teeth are directly broken, unable to withstand the current operation, and the bearing capacity function of the gear is completely lost (walker et al., 2018). extraction of gear and bearing fault features the real-time running state of the bearing is obtained by using the monitoring system. it is known that the monitoring data will be distorted due to salt and pepper noise, gaussian noise and reflected half point noise in the process of recording and transmission. therefore, the bearing monitoring data is estimated by using the geometric shape analysis method and digital analysis method of mathematical concept. through the analysis of the relationship between the elements, the bearing monitoring data is estimated, and the linear least square method is used to measure the impact of noise on the data, that is, according to the calculated minimum average variance, the monitoring background that can be fitted is determined. therefore, the calculation formula of the fitting degree is: f x k x k x k x k x( )= + + +1 2 3 4 (9) in the formula, k1, k2, k3 and k4 are respectively represented as the influence of background information on noise in each direction, and x is the mutation value of noise (spartalis et al., 2018). the average variance is calculated according to the formula 9: s z x k xi i i n 2 2 1 = − ( )   = ∑ (10) in the formula, xi represents the gray value of the i-th column of the monitored data. at this time, the calculation result is the noise factor of background information in four positive directions. due to the inaccurate fault location and calculation deviation, the information interference in other directions is corrected by information extraction, and the median filter is weighted and added. at this time, the corrected background noise estimate is: v u u w= +1 2 2g (11) in the formula, according to the 2 value obtained by the above fitting, w represents the median filter, u1 and u2 represent the weighting coefficient, and u1 + u2 = 1, so as to obtain a more accurate noise background. according to this value, the interference noise is proposed, and the complete bearing monitoring data is obtained. in order to detect the defects of the pre-processing monitoring data, it is necessary to determine the fault area and segment it. according to the monitoring data, roi is the region of interest. the bearing in this region is judged and segmented according to the result. according to the imaging principle, known bearings can be divided into three areas: the upper background area, the middle imaging area and the lower turntable area (ganson, 2018). according to the principle of uniform direction and gray projection algorithm, the distribution rule of bearing projection is analyzed to determine the scope of fault analysis. according to the above calculation results, the gray projection on the calculation axis is: g y v h x y i v ( )= ⋅ ( ) = ∑1 1 , (12) in the formula, v is the background value calculated by formula 11, h(x, y is the projection normal distribution function. according to the calculation results, the curve characteristics are analyzed, to determine the bearing monitoring data on the y-axis of the imaging boundary. figure 3 is the sample space schematic diagram of the projection curve. boundary conditions projection range figure 3. sample curve projection space the two curves in the image are the constraint conditions for generating projection curves for non damaged samples, scraped samples, bumped samples and over sharpened samples. it is known that no matter what type and specification of the bearing for coal mining equipment, its projection shape is similar, the upper background area has no detection objects, and there is no effective monitoring data; the bottom area has a certain degree of fluctuation because of the problem of reflection; the middle position is the main data monitoring area of the bearing, because the bearing itself is smooth, there will be reflection phenomenon, thus, the curve of this projection will fluctuate violently. according to the above analysis results, the region roi calculation algorithm is set up: the gray projection curve is filtered by means of mean value, and the curve is derived by smooth value, so that it passes through the zero point, and then the maximum value point and secondary value point of the curve are found out from the zero point, the abscissa is set as x1 and x2, and (0,c1 and (0,c2 are taken as control points to establish a rectangular region, which is roi (liu & zhu, 2018). because the image boundary of the bearing in the roi area at this time is relatively fuzzy, which will affect the classification of defect categories, the wave value transformation segmentation algorithm is adopted to accurately segment the image area of the bearing, as shown in figure 4. 119fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment background area transition area target area figure 4. schematic diagram of the segmentation process it can be seen from the figure that there may be errors in the boundary of the segmented region. one is to divide the background into the target region, and the other is to divide the target pixels into the background region. in order to avoid such a situation, morphological processing is carried out. morphological treatment is to expand and corrode. expansion is to merge all the contact points of the object’s contact background into the object, so that the boundary expands outwards, while corrosion is to eliminate the fictitious contact points, so that the boundary contracts inwards. the calculation process is as follows: u u s w w s u w s 2 1 1 1 2 2 1 1 3 − = − = − =      (13) in the formula, u1 and u2 represent the area obtained by two expansions, w1 and w2 represent the area obtained by two corrosions, s1 represents the part around the boundary that does not belong to the target range, s2 represents the area around the boundary that belongs to the target range, s3 represents the area that cannot be determined, so the area is divided into bundles. on the basis of region segmentation, the feature statistics and extraction of these data are carried out, and the rule subsets are established according to these features, so as to reduce the deposition data and improve the accuracy of fault diagnosis. the shape feature of gears and bearings in coal mining equipment is one of the important features of fault identification, so the common fault shape feature shown in table 1 below is listed (ali et al., 2018). table 1. basic shape characteristics when gears and bearings fail serial number characteristic shape feature description 1 area amount of data in the split area 2 perimeter amount of data in adjacent divided regions 3 diameter the horizontal distance between the two furthest data points on the boundary 4 center of gravity horizontal center of gravity, vertical center of gravity 5 aspect ratio aspect ratio 6 number of cracks easy to describe and quickly calculated values in the segmented area 7 thin length ratio of perimeter to area, ratio of diameter to area 8 compactness data to measure the complexity of the boundary 9 similarity ratio of vectors 10 symmetry symmetric distance after region rotation according to the data in table 1, the bearing deformation features are extracted. it is known that the color distribution of bearings and gears is divergent under the condition of light source, light and object reflection in machine vision. assuming that the hue is m, the saturation is d, and the brightness is r, the color distribution is calculated: µ= +      + − +       + = ∑e mm d d d d i n max min max min ln 1 1−− −      r 0 5 0 5 . . (14) in the formula, e represents the parameters of the segmented area, and m represents the hue resolution. through the color resolution model, the color features of gear and bearing images are obtained, and then the texture features are counted. because the change of texture image has certain regularity, which is embodied in the attributes of density, thickness, uniformity, regularity, linearity, frequency, etc., according to these attributes, the image area is set as w (x,y, the size is i ×j, and the calculation formula is as follows: i x f x j x i f x i i i n i i i n = ( ) = −  ( )      = = ∑ ∑ 1 1 (15) in the formula, xi represents the gray level series, f xi( ) represents the probability function of gray level series, which is the statistical result of texture features. in the process of feature extraction, it is necessary to make the selected features meet the separability constraints, which is to be representative, so that they can not only reflect the characteristics of all aspects, but also reflect the personality of the object; secondly, the graphic features should have very strong anti-interference, so as to ensure that the influence of different lighting, light source, shape and other factors is small and stable, the feature values of image content in the segmentation region are similar, and the selected features are independent of each other. in this paper, the scale feature extraction method is used to extract data based on the statistical results of features. the image is transformed to another space by function or matrix. the transformation methods include frequency domain transformation and geometric projection transformation. the data with different functional features in feature grouping are extracted and a new feature set is established. according to this feature, the calculation results are as follows: ε µζ κ κ= ⋅ ⋅ +( )n i j2 2 (16) in the formula, n represents n shape values, , i and j are the result statistics values of formula (2-7), so as to realize the extraction of gear and bearing fault feature data in coal mining equipment in a complex geological environment (sayers et al., 2019). fault location parameters setting of the location algorithm on the basis of defining the fault characteristics of gears and bearings, the fault location algorithm is developed according to the basic operation mode of coal mining equipment. generally speaking, the first step is to determine the fault range of gears and bearings. using cs technology to set up a search program, the target range of fault location is defined. the step factor is calculated, which is the key parameter to control the breadth of the search area. when the step factor is too large, the search program is controlled to start a full regional search; when the step factor is too small, the parameter can control the search program to start a local search in a local area. in order to control the search balance of the search program in the whole region and the local region, a balance adaptive adjustment parameter is added to make the search program distributed and synchronous search. the calculation formula of the step factor is: 120 shuilin wang, songyong liu, fanping meng r r k r r n f f fs = − −( ) × − max max min t 1 2 1 (17) in the formula, rs represents the search step of the s-th time, where s represents the current number of times; k represents the control parameter; rmax represents the maximum upper limit of step factor; rmin represents the minimum lower limit of step factor;  represents the adjustment parameter introduced; n represents the maximum number of iterations; f f f fi ri1 2= = = represents the optimal function under the initial fitness; f f f fi ri1 2= = = represents the optimal function suitable for this iteration. at the same time, under the search control of this parameter, the discovery probability of the approximate location of the equipment fault is calculated. in cs technology, the discovery probability is generally between 0.2-0.3, and the probability under this parameter is set to p. when the probability is always in a large value, the convergence speed of the fault search can be accelerated; when the value of the discovery probability p is small, the convergence rate is reduced. therefore, the optimal fitness is set by the ratio of the optimal fitness and the previous generation solution. f f f f f i ri i ri = > ≤     1 0 (18) in the above formula: f f f fi ri1 2= = = represents the optimal fitness of the previous generation; f f f fi ri1 2= = = represents the fitness of the i-th solution under the action of rs. according to the above calculation formula, the detection probability calculation results obtained by adjusting the adaptability of the detection probability are as follows: ′= = + −( )⋅ ′      = ∑p d f p p p p p i i n / min max min 1 (19) in the formula, p' represents the high-quality probability of the solution after iteration; di represents the optimal fitness for the i-th solution; n represents the total number of solutions; pmin and pmax represent the upper and lower limits of the discovery probability respectively. combining the above two sets of formulas, the search parameters are obtained, and the capture function is set by the parameters to extract the fault nodes of gears and bearings within the range of the target. according to the carrier frequency of the data, the capture function can get the code phase of the defect data and realize the acquisition of data nodes in any area. the capture times of the capture function are calculated as follows: n j d j b= × × +      1 0 2 (20) in the formula, n represents the number of acquisition times, ji represents the number of times of code phase, j0 represents the total step of wave frequency, d represents the step of wave frequency of defect data, and b represents a fixed constant, usually 1. the capture function is set according to the capture times, and the capture parameter f x( ) is introduced, as follows: dft f x q dx w s g t q dx a st t n f a st t = ( )   = ( ) ( )   −∫ ∫ 2 0 2 0 p p = ( ) ( )w s w sn f * (21) in the above calculation expressions, dft represents the fourier calculation result of the acquisition function under the condition of discrete transformation; q represents the degree of acquisition; x represents the independent variable of the acquisition function; q represents the exponential function; a represents the frequency of acquisition; s represents the conjugate value; t represents the target range space; g tf ( ) represents the real-time fault signal; wn(s represents the fixed value of the acquisition strength of the fault signal under n times of acquisition; w sf * ( ) indicates the actual acquisition strength of fault data when the acquisition strength is f ei d= /, * indicates the transformation flag. in the obtained defect nodes, two adjacent defect nodes are randomly selected as samples, to calculate the defect center point by distance scaling method, and the fault sample curve is selected, as shown in figure 5 (heffner et al., 2018). change time sub-peak maxim num peak d yn am ic p ea k ch an ge in di ca to r figure 5. fault peak curve the selected fault samples need to meet the basic requirements of two peaks. the fault data of gears and bearings are “flat”. assuming that the abscissa of high peak value and sub peak value in “flat area” is x1 and x2 , the center position c0 and peak width l of the area are calculated. the calculation results are shown in the following formula: c x x sh l x x 0 1 2 1 2 = − = −      h (22) in the formula, s is the number of peaks, and s ≥ 2; h is the error correction coefficient;  is the width measurement parameter. according to the above calculation results, the distance between the fault node and the fault center position is determined. according to the real-time changing fault intensity indicator, combining formulas 21 and 22, the random dynamic change parameter  is calculated. the formula is: γ ε ε α = + + × ⋅ d c l dft 0 0 1 (23) in the above formula,  is the adjustment control factor of defect peak; d0 is the adjustment factor of dynamic parameter;  is the peak active direction of the fault with different intensity; dft is the discrete transformation parameter. the linear regression model with activation function is as follows: 121fault diagnosis mechanism of gears and bearings for coal mining equipment in a complex geological environment y z q z h zi i i n ( )= ( ) + ( ) = ∑ w 1 (24) in the formula,n is the number of multi-peak defect nodes between gear and bearing; y(z is the expected output value; qi(z is the calculation regression factor in the i-th fault section of gear or bearing; i is the connection weight between fault nodes; h(z is the calculation residual. according to the model and the variation parameter , the defect center is obtained c y z y z = ( ) ( ) ∑ ∑ g 2 2 (25) according to the obtained central coordinate position, the hardware fault edge is determined, and the specific range of fault position is obtained (barrero et al., 2019). the boundary coordinates of the fault location are obtained by the least square method. according to the above calculation, the defect center c is obtained. supposing there are n anchor nodes from the center. within the determined fault target range, each coordinate can be represented by a1, a2,..., an. then, for each anchor node, there is an= {(x,y) | (x1,y1),(x2 ,y2),...,(xn ,yn)}, an unknown node is set as aj , the node coordinate is (xj , yj), and the node is assumed to be the true coordinate of the fault. the relationship between the unknown node and the anchor node is: x x y y d x x y y d x x y y j j j j j n j n −( ) + −( ) = −( ) + −( ) = −( ) + −( 1 2 1 2 1 2 2 2 2 2 2 2 2  )) =      2 2dn (26) d1, d2, ..., dn in the above formula represents the estimated distance between the unknown node and the anchor node. the coordinate boundary obtained by the formula is shown in figure 6. failure node tracking function lossless node figure 6. obtained coordinates of fault boundary it can be seen from the above figure that the gray area in the figure represents the true range of the fault, and the white area is the location of other nodes of the equipment affected by the fault. the boundary coordinates are obtained by the least square method. the formula (3-10) is rewritten into the form of ax = b, and the objective function min || ax = b ||2 is obtained by using the least square method x x y =         0 0 (27) in the formula, x0 and y0 represent the abscissa and ordinate of the fault boundary respectively. through the calculation process, the specific boundary coordinates of the fault in figure 6 above are obtained, the influence scope of the fault in the mining equipment is determined, and the diagnosis mechanism of gear and bearing fault for mining equipment is studied under complex geological conditions (vaziri et al., 2018). experimental study in this paper, a comparative simulation experiment is proposed to simulate the complex geological environment in the process of coal mining (yang et al., 2020). opening the simulation test software, the bearing fault effect in the coal mining equipment is simulated. table 2 below is the fault value to be simulated and the fault frequency parameters to be input (wu et al., 2020). table 2. simulated bearing failure frequencies item number fault frequency name failure frequency a1 rolling bearing outer ring failure frequency 162.178 a2 rolling bearing inner ring failure frequency 227.822 a3 rolling bearing rolling element failure frequency 86.593 a4 failure frequency of cage touching inner ring 12.475 a5 failure frequency of cage touching outer ring 17.525 different bearing speeds are set. when the bearing speed is too high, the gear working frequency will be too high. at this time, the bearing signal indicators that cause bearing failure and gear failure are shown in table 3 below. table 3. indicator values of bearing vibration signals at different speeds different speeds and signals segment indicators rotating speed signal m1 m2 m3 m4 m5 800r/min normal 1 0.6055 0.4791 0.3451 0.2477 0.0342 normal 2 0.6012 0.4792 0.3487 0.2921 0.0371 internal loss 1 0.5805 0.4788 0.3561 0.2858 0.0475 internal loss 2 0.5865 0.4874 0.3563 0.2687 0.0411 external loss 1 0.6345 0.4851 0.3522 0.2751 0.0480 external loss 2 0.6327 0.4792 0.3566 0.2741 0.0382 1200r/min normal 1 0.6281 0.4853 0.3485 0.2736 0.0567 normal 2 0.6204 0.4722 0.3669 0.2819 0.0552 internal loss 1 0.5977 0.4621 0.3851 0.2654 0.0523 internal loss 2 0.5947 0.4673 0.3554 0.2647 0.0604 external loss 1 0.6381 0.4615 0.3582 0.2708 0.0478 external loss 2 0.6296 0.4628 0.3479 0.2432 0.0455 according to the data in the table above, the fault parameters of gears and bearings for coal mining equipment is simulated in a complex geological environment under high-intensity work. two methods are used to analyze 122 shuilin wang, songyong liu, fanping meng the fault diagnosis mechanism. the proposed research method is used as the experimental group, and the traditional research method is used as the control group. the test results are shown in figure 7 (cheng et al., 2020). stationary curve a m pl itu de (v ) a m pl itu de (v ) difference curve testing time(s) (a) experimental group test result testing time(s) (b) control group test result expectation curve stationary curve difference curve expectation curve figure 7. experimental test comparison results according to the curve trend in the figure above, it can be seen that the stability curve of the proposed research method is similar to the expected curve, while the stability curve of the traditional research method fluctuates violently, which is quite different from the expected curve. it can be seen that the stability difference value of the proposed research method is far lower than the stability difference value between the traditional research method and the expected curve (gao et al., 2018). conclusions in the complex geological environment, the proposed research method of fault diagnosis mechanism is more stable, which improves the reliability of diagnosis results and provides more perfect technical means for the development of national mining technology (gao et al, 2019). references ali, e. e., teo, a. k. s., goh, s. x. l., chew, l., & yap, k. y. (2018). medadappq: a quality assessment tool for medication adherence apps on ios and android platforms. research in social & administrative pharmacy, 14(2018), 1125-1133. balestro, g., festa, a., borghi, a., castelli, d., gattiglio, m., & tartarotti, p. (2018). role of late jurassic intra-oceanic structural inheritance in the alpine tectonic evolution of the monviso meta-ophiolite complex (western alps). geological magazine, 155(2), 233-249. barrero, l. h., cifuentes, m., rodríguez, a. c., rey-becerra, e., johnson, p. w., marin, l. s., piedrahita, h., & dennerlein, j. t. (2019). whole-body vibration and back pain-related work absence among heavy equipment vehicle mining operators. occupational and environmental medicine, 76(8), 554-559. cheng, h. y., & liu, y. n. (2020). an improved rsu-based authentication scheme for vanet. journal of internet technology, 21(4). ganson, t. (2018) sensory malfunctions, limitations, and trade-offs. synthese, 195(4), 1705-1713. gao, n. s., cheng, b. z., hou, h., & zhang, r. h. (2018). mesophase pitch based carbon foams as sound absorbers. materials letters, 212, 243246. doi:10.1016/j.matlet.2017.10.074 gao, n. s., guo, x. y., cheng, b. z., zhang, y. n., wei, z. y., & hou, h. (2019). elastic wave modulation in hollow metamaterial beam with acoustic black hole. ieee access, 7, 124141-124146. doi:10.1109/ access.2019.2938250 jeong, g. y., kong, j. h., lee, s. j., & pang, s. j. (2018). comparisons of bearing properties for various oriented glulam using digital image correlation. journal of wood science, 64(3), 1-9. heffner, j. l., watson, n. l., mcclure, j. b., anthenelli, r. m., hohl, s., & bricker, j. b. (2018). “i smoke like this to suppress these issues that are flaws of my character”: challenges and facilitators of cessation among smokers with bipolar disorder. journal of dual diagnosis, 14(1), 32. liu, y., zhu, z. q. (2018). influence of gear ratio on the performance of fractional slot concentrated winding permanent magnet machines. ieee transactions on industrial electronics, pp(99), 1-1. morris, vachani, a., pass, h. i., rom, w. n., ryden, k., weiss, g. j., hogarth, d. k., runger, g., richards, d., shelton, t., & mallery, d. w. (2018). whole blood fpr1 mrna expression predicts both non-small cell and small cell lung cancer. international journal of cancer, 142(2), s34-s34. sayers, e., rich, j., rahman, m., kelly, b., & james, c. (2019). does help seeking behavior change over time following a workplace mental health intervention in the coal mining industry? journal of occupational and environmental medicine, 61(6), 1. spartalis, m., tzatzaki, e., & spartalis, e. (2018). letter by spartalis et al regarding article, “left ventricular assist device malfunctions: it is more than just the pump”. circulation, 137(21), 2298-2299. vaziri, v., hamidi, j. k., sayadi, a. r. (2018). an integrated gis-based approach for geohazards risk assessment in coal mines. environmental earth sciences, 77(1), 29. walker, t., zahn, n., bruckner, t., streit, m. r., mohr, g., aldinger, p. r., clarius, m., & gotterbarm, t. (2018). bruckner. mid-term results of lateral unicondylar mobile bearing knee arthroplasty: a multicentre study of 363 cases. bone & joint journal, 100-b(1), 42-49. wang, k. x., huang, h., liu x. d., & pan, j. y. (2018). origin and influence of a late mesozoic multistage iand a-type granitic complex in northern fujian province, south china. geological journal, 54(1), 39-61. wu, z., liu, y. n., & jia, x. x. (2020). a novel hierarchical secret image sharing scheme with multi-group joint management. mathematics, 8(3), 12. doi:10.3390/math8030448 yang, f., yang, f., wang, g. y., kong, t., wang, h., & zhang, c. s. (2020). effects of water temperature on tissue depletion of florfenicol and its metabolite florfenicol amine in crucian carp (carassius auratus gibelio) following multiple oral doses. aquaculture, 515, 9. doi:10.1016/j. aquaculture.2019.734542 earth sciences research journal earth sci. res. j. vol. 19, no. 2 (december, 2015): 95-96 letter from editor earth sciences research journal presents its latest issue, volume 19, number 2, with a positive balance in the eve of 2016 and with the challenges for the future editions. during the last two decades, we have published more than 200 papers in all the spectrum of earth sciences. we expect we will be here for many years more. esrj is now part of international databases, and we hope to improve our numbers thanks to the tireless word we do every day for our readers. we have accomplished the principal objectives for an international publication -punctuality in our volumes, respect for the authors and the spotlight on the scientific quality. for a fair editorial process, we are part of the open access and its policies of pair blind-review. the time of the operations has been reduced, and the international participation increased. for the coming years, esrj would reach a better visibility thanks to the publication of top-quality scientific articles, which allows us to be part of the central positions in international databases and providing researchers with fundamental elements of their work. to make possible this goal, we want to call authors to put lights on scientific quality and editorial quality. your help is essential for us. we want to encourage researchers to pick up esrj as a platform for their works, but also to check the authors' guidelines as a first step to saving efforts and time during processes. needles to say the importance of manuscript parts, english level and images readability to the acceptation of submissions. as a complementary strategy, we would like to invite our authors to create profiles in one of the available platforms, not only as an introduction to the esrj editorial team but also as a promotion of your work. several challenges we have gripped during the work of the last years, for that reason we have adopted the principles of ethic and best practices in our procedure protocols. this decision would be a tool giving papers’ authors a warranty of complete respect for their work and to readers a quality signature that our contents have been verified for their use. in this adoption of best-practices procedures, many submitted papers have been rejected for having a not proper citation; this fault needs to be eradicated from all journals promoting scientific contents. a glimpse of the numbers of esrj points the international collaboration as one of the biggest challenges for the future issues. international collaboration accounts for the articles that have been produced by researchers from several countries. it's not a call to improve the rates, but the scientific quality would be feed for the interaction of researchers facing different realities. early 2016, esrj will publish a special issue with the selected papers of the international conference on earth and environmental sciences by the university malaya. this special issue on frontiers in earth sciences will publish articles on the most outstanding discoveries across a broad research spectrum in the field. the mission of frontiers in earth sciences is to provide an introduction to some of the current advanced concepts, probb lems, and controversies that are at the forefront of earth sciences and to obtain a view of how the different parts of the earth behave and interact. as it has been a costume in this letter from the editor, the following paragraphs will introduce the works of the current issue. our first paper deals with the problem of recovering moho surface from satellite gradiometry data considering the non-isostatic corrections. the second manuscript presents a case study on the application of electromagnetic and gpr measurements to a suspicious metalcontaminated site. while subsurface pollution is very harmful to nature-beings and finally human, detection of its extent in surface and more in depth is considerably challenging. the following work analyzes the earthquake light phenomenon based in the christchurch earthquakes occurred in 2010 and 2011, with testimonies related from people who observed or recorded the lights. in the seismotectonic field is presented a paper showing the strain rate in the north anatolian fault zone (nafz), that is one of the most important fault zones of turkey. another turkish submission tried to determine seismic specifications of the bitlis province in lake van basin and used probabilistic seismic hazard analysis for the determination of local site-specific spectra. also from turkey, a paper examines the performance of a single base rtk station compared with similar systems and presenting a complete exposure of values and challenges of the system. a contribution from iran analyzes the magmatic evolution and compositional characteristics of tertiary volcanic rocks associated with the manganese mineralization in a mine located in central iran. also from in iran, a paper assesses the stability of the rock slopes with various rock mass qualities. the manuscript discusses the combination of rmr and smr systems to evaluate rock slope stability. from latin america, a paper makes a contribution to the knowledge of the tatui formation, in brazil, emphasized in paleontology and paleoenvironmental considerations. we hope you find useful these contents. carlos vargas editor in chief, earth sciences research journal keywords: underground mining; mining predict; probability integral model; geometric model; parameter inversion palabras clave: minería subterránea; predicción de minería; modelo de probabilidad integral; modelo geométrico; inversión de parámetros; how to cite item li, j., yu, x., chen, d., & fang, x. (2021). research on the establishment of a mining subsidence prediction model under thick loose layer and its parameter inversion method. earth sciences research journal, 25(2), 215-223. doi: https://doi.org/10.15446/esrj.v25n2.79537 most of the coal mining in china is underground, which will inevitably cause surface deformation and trigger a series of geological disasters. therefore, it is essential to find a suitable method to forecast the ground sinking caused by underground mining. the most commonly used prediction model in china is the probability integral model (pim). but when this model is used in the geological condition of mining under thick loose layers, the predicted edge of the sinking basin will converge faster than the actual measured sinking situation. a geometric model (gm) with a similar model shape as the pim but with a larger boundary value was established in this paper to solve this problem. then an improved cuckoo search algorithm (icsa) was proposed in this paper to calculate the gm parameters. the stability and reliability of the icsa were verified through a simulated working face. at last, the icsa, in combination with the gm and the pim, was used to fit 6 working faces with the geological mining condition of thick loose layers in the huainan mining area. the results prove that gm can solve the above-mentioned pim problem when it is used in geological mining conditions of thick loose layers. and it was obtained through comparative analysis that the gm and the pim parameters can take the same value except for the main influence radius. abstract research on the establishment of a mining subsidence prediction model under thick loose layer and its parameter inversion method investigación sobre el establecimiento de un modelo de predicción de subsidencia minera bajo una capa gruesa inestable y su método de inversión de parámetros issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n2.79537 la mayor parte de la minería del carbón en china es subterránea, lo que inevitablemente causa deformaciones en la superficie y desencadena desastres geológicos. por lo tanto, es necesario encontrar un método adecuado para pronosticar el hundimiento del suelo causado por la minería subterránea. el modelo de predicción más utilizado en china es el modelo integral de probabilidad (pim). pero cuando este modelo se utiliza en la condición geológica de la minería bajo capas gruesas inetables, el borde previsto de la cuenca de hundimiento converge más rápido que la situación de hundimiento medida. para resolver este problema, en este artículo se estableció un modelo geométrico (gm) que tiene una forma de modelo similar a la del pim pero que tiene un valor límite mayor. en este trabajo se propuso un algoritmo de búsqueda de cuco mejorado (icsa) para calcular los parámetros de gm, y se verificó la estabilidad y confiabilidad del icsa a través de una frente de trabajo simulado. por último, el icsa en combinación con el gm y el pim se utilizaron para ajustar 6 caras de trabajo con la condición de minería geológica de capas gruesas sueltas en el área minera de huainan. los resultados demuestran que el gm puede resolver el problema de pim mencionado anteriormente cuando se utiliza en las condiciones de minería geológica de capas gruesas inestables. y se obtuvo mediante análisis comparativo que los parámetros del gm y del pim pueden tomar el mismo valor excepto por el radio de influencia principal. resumen record manuscript received: 07/05/2019 accepted for publication: 13/12/2019 earth sciences research journal earth sci. res. j. vol. 25, no. 2 (june, 2021): 215-223 jingxian li1, xuexiang yu 1*, deshu chen2 , xinjian fang1 1. school of geomatics, anhui university of science and technology; huainan 232001, china 2. guo xuan high tech power energy co., ltd; hefei 232000, china email: xxy16321@163.com m in in g e n g in e e r in g https://doi.org/10.15446/esrj.v25n2.79537 https://doi.org/10.15446/esrj.v25n2.79537 216 jingxian li, xuexiang yu, deshu chen, xinjian fang introduction although the continuous mining of the coal resources provides a strong guarantee for the rapid development of china’s economy. however, on the one hand, with the continuous consumption of coal resources and the continuous acceleration of urbanization, the amount of the coal seams pressed under buildings, railways, and water is increasing. on the other hand, the surface subsidence and the massive structure damage caused by underground mining have been very serious in china (xuan & xu, 2014; zhao et al., 2016; li et al., 2019; xu et al., 2019). therefore, it is very important to predict and fully grasp the laws of the ground deformation and its deformation caused by underground mining, which can provide guidance for controlling the destruction of surface protection objects. however, the process of the surface subsidence is complicated and different geological mining conditions will cause different surface subsidence laws (xu et al., 2014). for example, the mining subsidence laws under the geological condition of thick loose layers have special characteristics compared with thinner loose layers and no loose layer. such as the maximum subsidence value may be greater than the normal thickness of the coal seam and it is apparently smaller of the predicted value than the measured data at the boundary of the subsidence basin (zhou et al., 2016; zhou et al., 2015). coal seams buried under thick loose layers are widely distributed in china. in recent years, most research on mining subsidence under the geological condition of thick domestic and foreign scholars has focused on the mechanism of rock formation movement. zhu et al. (2020) confirmed that the transition layers of the bedrock layers and the loose layers has a support effect on mining subsidence. liu et al. (2020) found that the thick loose layer formed a bending collapse zone, fracture development zone and caving zone during the mining process through simulation analysis, and derived a prediction model based on elastic mechanics. xu et al. (2020) studied the movement laws of the overlying bedrock while mining shallow buried coal seams under thick loose layers. yu et al. (2020) studied the rock-soil structure above the coal seam which is buried under thin bedrock layer and thick loose layer. wang et al. (2019) analyzed the stability of the loose layers and the bedrock layers during mining. ma et al. (2017) discussed the prediction methods for mining subsidence when coal pillars are solid backfilled under thick loose layers. yang & xia (2013) studied the ground deformation laws caused by mining under thick loose layers and thin bedrock layers by analyzing measured data. dai et al. (2011) researched the influence of the thickness of the loose layers on mining subsidence through numerical simulation. zhou et al. (2018) proposed a combined prediction model based on soil mechanics and random medium theory. chang et al. proposed a mining subsidence prediction model used in subcritical mining under thick loose layers with considering the different subsidence and movement mechanisms of the bedrock layers and the thick loose layers. but the model can only be suitable for subcritical mining under thick loose layers (chang et al., 2015). the advantage of the mining subsidence prediction model which is based on the mechanism of rock formation can fundamentally find the law of rock formation. it is not easy to establish the model and calculate the model parameters in actual application. so they all have not been widely used. the most commonly used mining subsidence prediction model is always the pim which is developed from the random medium theory (wang et al., 2020; zheng et al., 2019; diao et al., 2016; fan et al., 2015). however, due to the limitation of the shape of the model curve, the problem of small boundary value is much more obvious while the pim is applied to predict surface subsidence caused by mining under thick loose layers, and this problem cannot be solved by adjusting the model parameters. therefore, it is necessary to establish a model with simple parameters and suitable for surface subsidence prediction caused by mining under thick loose layers. an accurate prediction model is very important to mining subsidence prediction, a set of accurate parameters to the model is also important (sehn et al., 2015). at present, the mostly used parameters inversion methods in mining subsidence prediction area can be divided into direct optimization methods (shen et al., 2015; guo & wang, 2000) and intelligent optimization algorithms (zha et al., 2011; wang et al., 2018). the most commonly used of direct optimization methods is the modular vector method (mvm). the mvm has the advantages of easy programming and is suitable for working faces of any shape. but it needs a set of initial values of the parameters while using, and use them as a basis to follow the valley line or the ridge line to accelerate to the best values. the mvm has a strong dependence on the initial value of parameters. if the initial value is not accurate, the calculation result will deviate from the reasonable range. since the pim is an official forecasting method in china, many mining faces have a set of pim parameters. moreover, the subsidence prediction parameters of the working face with similar geological conditions are similar, so the pim parameters of the mined working face with similar geological conditions can be used as the initial value of the pim parameter of the working face to be solved. but on the one hand, the initial values of the parameters obtained in this way may be unreliable. on the other hand, it is difficult to obtain the initial values of the parameters of new prediction models. therefore, the use of the mvm was restricted. the intelligent optimization algorithm uses random global search within a certain range by simulating the behavioral characteristics of creatures such as foraging and reproduction instead of deterministic search of the modular vector method, so it can explore the global optimal value without initial value dependence. the most commonly used method of intelligent optimization algorithms is the genetic algorithm (ga) (du et al., 2014; xing et al., 2021; li et al., 2017). however, the accuracy of the ga to calculate the parameters of the mining subsidence prediction model is not ideal. at the beginning of this paper, the gm to predict the surface subsidence caused by limited coal seams is established. and it is found that by adjusting the parameters of the gm, the middle part of its model curve can be basically consistent with the pim, while the boundary value predicted is larger than the pim. therefore, it is inferred that the shape of the gm curve is more in line with the shape of the main section of the subsidence basin which is caused by mining under thick loose layers. then in order to calculate the gm parameters, an improved cuckoo search algorithm-icsa-is proposed, and a simulated working surface is used to verify the stability and accuracy of the icsa. finally, 6 working faces buried under thick loose layers were used to compare the accuracy of the gm and the pim. by comparing the fitting effects of the gm and the pim in 6 working faces, it can be concluded that the gm is more suitable than the pim in predicting surface subsidence caused by mining under thick loose layers. and the relationship between the gm parameters and the pim parameters is drawn. geometric model semi-infinite mining subsidence prediction model take the open-off cut of the coal seam as the origin, the direction from the origin to the inclination direction of the goaf as the positive direction of the y-axis and the direction from the origin to the strike of the goaf as the positive direction of the x-axis to establish the calculation coordinate system. then according to the pim, the sinking value of the point with abscissa x on the strike main section of the subsidence basin caused by a semi-infinite mining coal seam with the thickness of m and the dip angle of α can be calculated by equation 1. w x w r exp x s r dspim pim pim( ) = ( ) × −( )∫0 0 2/ / ) ∞ π (1) put erf x exp u du x ( ) = ( )× −( )∫2 0 2/ π into equation 1 to get equation 2: w x w erf x rpim pim( ) = × ×( ) ( / ) /0 2 π (2) where w0 represents the max subsidence value and can be calculated by w m cos qpim0 = × ×α , qpim is a parameter in the pim named subsidence coefficient whose value is mainly determined by the lithology of the overlying rock, and its value is generally less than 1. rpim is another parameter in the pim named main influence radius. it can be seen from equation 2 that the main section subsidence curve caused by a semi-infinite mining coal seam is in the shape of “s”. and the shape of function f x x x( ) = +/ 1 2 is also “s”. combined with the function 217research on the establishment of a mining subsidence prediction model under thick loose layer and its parameter inversion method f x x x( ) = +/ 1 2 and equation 2, the new formula to calculate the sinking value of the point with abscissa x on the strike main section of the subsidence basin caused by a semi-infinite mining coal seam with the thickness of m and the dip angle of α is established as equation 3. w x w x r x r gm gm gm gm ( ) = × +( ) +           ( / ) / / 0 2 2 1 1 (3) where w m cos q gm gm0 = × ×α , qgm is the subsidence coefficient of the gm. rgm is the main influence radius of the gm. limited mining subsidence prediction model the sinking value of the point with abscissa x on the strike main section of the subsidence basin caused by a limited mining coal seam with the length of d3 can be regarded as the superposition of two semi-infinite mining and calculated with equation 4. w x w w x s w x d spim pim pim pim pim ° ( ) = × −( )− − −( ) )0 3 3 4( ] (4) where s pim3 and s pim4 are parameters of the pim named inflection point offset to the left and inflection point offset to the right. from the analysis above, the gm to predict the sinking value of the point with abscissa x on the strike main section of the subsidence basin caused by a limited mining coal seam with the length of d3 is established as equation 5. w x w w x s w x d sgm gm gm gm gm ° ( ) = × −( )− − −( ) )0 3 3 4( ] (5)= × × × −( ) + −( )( ) − − − m q cos x s r x s r x d s r gm gm gm gm gm gm g α 3 3 2 3 4 1 / / ( ( )) / mm gm gmx d s r1 3 4 2 + − −( )( )          ( ) / where s gm3 and s gm4 are the inflection point offset to the left and the inflection point offset to the right of the gm. figure 1. the calculation principle of the movement and deformation of the inclined main section in limited mining according to figure 1, the formula to predict the sinking value of the point with coordinate y on the inclined main section of the subsidence basin caused by a limited mining coal seam with the width of d1 and depth of h is established as equation 6. w y w y s sin r y s r gm gm gm gm gm ( ) = × − × +( )( ) + −( )      0 1 1 2 1 sin θ α α/ / ++                 −( ) = × − −( )× +( ) 1 0 1 2w y l w y d s sin singm gm( / )θ α α // ( / ) / r y d s sin sin r w y gm gm gm gm1 1 1 2 2 + − −( )× +( )( ) +           ° θ α α (( ) = ( )− −( )w y w y l (6) where l represents the width for calculation of the coal seam and can be calculated by l d s s sin singm= − −( )× +( )1 1 2 θ α θ/ , gm , s gm1 and s gm2 are parameters of the gm named mining influence propagation angle, inflection point offsets in the downhill direction and inflection point offsets in the uphill direction. parameters analysis from the analysis above, it can be seen that there needs seven parameters to predict mining subsidence with the gm: qgm , rgm , gm , s gm1 , s gm2 , s gm3 and s gm4 . take a working face in huainan mining area which has the length of 770m, width of 205m, depth of 556m, thickness of 3m and dip angle of 5° as an example to compare the model curves of the gm and the pim. the pim parameters of this working face are as follows: qpim 1 01. , rpim 300 , θ pim = °88 , s pim1 5=− , s pim2 32=− , s pim3 19=− and s pim4 30 . the strike main section predicts the result of using the pim in combination with the above parameters is shown in curve 1 in figure 2. and the result of using the gm in combination with the above pim parameters is shown in curve 2 in figure 2. by comparing curve 1 and curve 2 in figure 2, it can be seen that the shapes and trends of the gm and pim model curves are similar under the same set of parameters, but the calculated specific subsidence values are different. then adjust the gm parameters to make r rgm pim / .2 5 and again use the gm to predict the subsidence of the strike main section to get curve 3. figure 2. comparison of the shape of the model curve of the gm and pim by comparing curve 1 and curve 3 in figure 2, it can be seen that by adjusting the parameter rgm, the middle part of the model curve of the gm can be roughly the same as the pim with the predicted value of the boundary part larger than the pim. which means the gm may meet the mining subsidence laws under the geological condition of thick loose layers better than the pim. so we can infer that the qgm , gm s gm1 , s gm2 , s gm3 , and s gm4 in the gm can take the same value with the qpim , pim , s pim1 , s pim2 , s pim3 and s pim4 in the pim except for r rgm pim / .2 5. improved cuckoo search algorithm in order to further study the characteristics of the gm and establish the relationship between the parameters of the gm and the pim, a suitable algorithm to solve the model parameters is needed. the cuckoo search algorithm (csa) is a new kind of heuristic bionic group intelligent optimization algorithm proposed 218 jingxian li, xuexiang yu, deshu chen, xinjian fang by professor xinshe yang and susan deb from the university of cambridge in 2009. the biological behaviors used include parasitic reproduction and lévy flights mechanism. the csa has the advantages of clear theory, few parameters, easy to expand, strong global search ability, easy to implement, and so on. it has been deeply studied and widely used by scholars in china and abroad. the csa parasitic reproduction behavior parasitic reproduction behavior means cuckoos do not build their own nests, they lay their eggs in communal nests or other birds’ nests. so mother cuckoos must face the risk that their eggs may be removed from the nests by other birds or the host birds. in the long-term evolution process, some cuckoos have learned some ways to improve the survival rate of their eggs such as pushing the host birds’ eggs out of the nests. lévy flights studies have shown that the foraging routes of many flying animals, including cuckoos, follow the lévy flights mechanism. the advantage of lévy flights mechanism in solving optimization problems is that it is combined with high frequency short step movements and occasionally long step movements. the high frequency short step movements make it possible to outcrop the local optimal solution and the occasionally long step movements ensure that the results will not fall into the local optimal solution. basic model according to the above biological background, make the following scenario hypothesis: • the total number of the nests is determined; • the cuckoos lay only one egg in each nest; • the one with the best fitness in the nest group will be retained directly to the next generation when the csa is used in solving optimization problems, the location of each nest represents a set of solutions. after several rounds of elimination, evolution and selection, the optimal individual can be selected out. the elimination strategy of the csa was random selection under fixed elimination rate. the selection strategy of the csa was greedy selection, that is, the individuals with better fitness will be retained after the new solutions were obtained and compared with current solutions. the evolution strategy of the csa adopts two modes: random walking mode and preferred walking mode. 1. random walking mode. use equation 7 when current nests’ positions need to be updated x x l i ni t i t+ = + ( ) =1 1 2α λ⊕ , , ,  (7) where xi t and xi t1 represent the location of the i-th nest of generation t and generation t+1 respectively; α represents the step size controller;  is point-to-point multiplication; n is the amount of the nests, that is, the amount of the feasible solutions; l λ( ) represents the random search path, l q u( ) < ≤−~ ( )λ λ1 3 , where  represents the random step size obtained by lévy flights. 2. preferred random walking mode. mimicking the behavior of a host bird perceives the cuckoo egg and throws it away, then a new nest is needed. the next updating operation is as equation 8. x x r x xi t i t j t k t+ = + −( )1 �� �� �� (8) where xj t and xk t represent two random solutions of generation t; r is a random number satisfying the uniform distribution principle between [0,1]. the icsa practice shows that the csa is not strong in local search ability. this is because each solution is independent of each other in the process of searching for the optimal solution, there is no experience learning from better solutions and no information exchange. therefore, the roulette selection strategy was adopted instead of the random elimination strategy in the csa and the average operator and crossover operator was added in the updating process to exchange the information between nests. the specific operations are as follows: the selection strategy. to select a better individual from the contemporary nests, it is first necessary to establish an individual fitness calculation function. according to the principle of least squares, for the purpose of making the fitting effect of the inverted parameters better, the fitness of solutions are evaluated based on the sum of the squares of the difference between the sinking value of predicted and observed. the fitness function is constructed as equation 9. the nest with high fitness value is of good quality. f w w i n j ni j p ri j j = −( ) = =∑1 1 2 1 2 2 / , , , , ; , , , ,   (9) where � f xi( ) represents the fitness of the i-th group of parameters, wpi j, represents the predicted subsidence value of the j-th observation point under the i-th group of parameters, and wrj represents the actual subsidence value of the j-th observation point. n is the amount of the nests in one generation, n is the amount of the observation points. the roulette algorithm is used to select individuals with better fitness and eliminate those with poor fitness. its specific operation process is as follows: 1. calculate the fitness of each nest as f xi( ) (i = 1,2... ,n), n is the total number of the nests; 2. calculate the probability of being passed on to the next generation of each individual as p x f x f xi i j m j( ) = ( ) ( ) = ∑/ 1 ; 3. calculate the cumulative probability of each individual as q p xi j i j= ( ) = ∑ 1 ; 4. generates a uniformly distributed pseudorandom rand in the interval [0,1]; 5. if rand q 1, then choose the first individual; otherwise, choose individual k to make q rand qk k− < ≤1 6. repeat step 4 and 5 for m times. save the selected individuals and eliminate the unselected the evolution strategy. declare a temporary nest xk � for communicating information between current nests. for any individual xt and xt1 , if x xt t= +1, calculate the step length by α =1 2/ t to generate a new nest xk. if x xt t≠ +1, then deal them with three different operators: average operator (take the average of the parameters in the two nests); crossover operator (randomly exchange the parameters in the two nests); and mutation operator (update the nests with eqs. (8)). the nest xk which has the best fitness can be selected out. then randomly select a nest from all nests and compare it with xk to keep the one with the better fitness. set the total number of the nests as n=200 and each nest represents a set of parameters, which is [ q, tan, , s1, s2, s3, s4] t in the gm. the specific steps of calculating mining subsidence prediction parameters with the icsa are shown in figure 3. test by simulated working face a simulated working face is designed so as to test the stability and reliability of the icsa and compare its accuracy with the csa and the ga,. 219research on the establishment of a mining subsidence prediction model under thick loose layer and its parameter inversion method its length, width, depth, dip angle and thickness are designed as 900m, 300m, 400m, 0° and 3m. its gm parameters are designed as is shown in table 1. there set up an observation line above both the strike main section and inclined main section as is shown in figure 4. the spacing and total number of the observation points are 30m and 88. take the value calculated by the gm with designed parameters as true subsidence value. then add random errors with median error of 5m, 10mm and 15mm into the true subsidence value to generate three sets of designed observations. in order to test the stability of the icsa, calculate 100 sets of the gm parameters of the designed working face with the above designed observations and the comparison of the parameter inversion effect of the icsa while the designed observation data has different degrees of median error are shown in table 2. in order to test the precision of the icsa, use the ga, the csa and the icsa to calculate 50 sets of the gm parameters of the designed working face with the above designed observations. take “q” as an example, the error comparison between true value and values inversed by the ga, csa and icsa are shown in table 3. table 1. designed gm parameters of the simulated working face parameters qgm tanβgm θgm(°) s1gm (m) s2gm (m) s3gm (m) s4gm (m) value 0.7 5.0 88 20 20 20 20 1 tan h rgm gmβ = / , where h represents the depth of the coal seam . table 2. comparison of the parameter inversion effect of the icsa while the designed observation data has different degrees of median error 5mm 10mm 15mm mean |∆| mean |∆|/l0 (%) mean |∆| mean |∆|/l0 (%) mean |∆| mean |∆|/l0 (%) qgm 0.028 4.00 0.035 5.00 0.037 5.29 tan 0.075 1.2 0.084 1.7 0.096 2.0 gm 0.315 0.36 0.332 0.38 0.328 0.37 s gm1 7.12 35.60 8.95 44.75 7.98 39.90 s gm2 6.37 31.85 7.82 39.10 8.15 40.75 s gm3 9.05 45.25 9.22 46.10 9.64 48.20 s gm4 9.12 45.60 8.99 44.95 9.42 47.10 δ 14.4 22.5 30.9 1 mean l l t∆ = − 0 / , where  represents the absolute error between the true value l0 and the inversed value l; t represents the inversion times, which equals 100 in this paper; 2 mean l l l l n n ∆ / / / %0 0 0 100= −( ) ×∑ ; 3 δ =  ( )sqrt n∆2 / , where n represents the total number of the observation points, which equals 88 in this paper. table 3. precision comparison of the ga, csa and icsa mean|∆q | mean|∆q |/l0(%) δ ga 0.037 5.3 31.2 csa 0.108 15.4 40.2 icsa 0.035 5.0 22.5 we can draw the following conclusions from table 2 and table 3: 1. the relative errors of qgm , tan gm and gm inversed by the icsa are all less than 6%, and the errors of s gm1 , s gm2 , s gm3 and s gm4 are all less than 10mm. indicating that the errors of the parameters inversed by the icsa are all within the normal range. the average values of the errors of qgm , tan gm and gm are all less than 1, indicating that the results calculated by the icsa are stable and reliable. 2. as the medium error of the random error in the observation data increases, the medium error of the inversed parameters has a tendency to increase. indicating that the quality of the observation value will directly affect the accuracy of the parameters inversed by the icsa. 3. the average error of qgm inversed by the icsa is 0.035, and its relative error is 5.0%, which is similar to the ga and significantly higher than the csa, indicating that the stability of the icsa is higher than the csa and figure 3. flow chart of the icsa figure 4. schematic diagram of the designed observation points of the simulated working face 220 jingxian li, xuexiang yu, deshu chen, xinjian fang comparable to the ga. compared with the other two methods, the error in the fitting of the parameters calculated by the icsa is the smallest, indicating that the reliability of the parameters calculated by the icsa is the highest. case analysis (1) the working face 1111 (1) of the zhujidong coal mine in huainan panji mining area has a length of 1584m, width of 220m, average depth of 920m, average thickness of 1.8m, and average dip angle of 3°. it is buried under the loose layer with an average thickness of 278m. its coal mining method and roof management method are the comprehensive mechanized method with a full height mining at one time and the total collapse method respectively. using the icsa combined with the last observation data of the working face to calculate the pim parameters and the gm parameters. the calculated model parameters are shown in table 10. the comparison between the observations and the subsidence value fitted by the gm and the pim are shown in figure 5 and table 4. table 4. fitting effect comparison of the gm and the pim for the 1111(1) working face median error overall median error of edge area relative error overall(%) relative error of edge area (%) gm 9.1 9.2 19.0 25.2 pim 17.9 19.8 42.0 56.5 figure 5. comparison of the fitting effect of the gm and pim for 1111(1) working face (2) the working face 12326 of the gubei coal mine in huainan xieqiao mining area has the length of 770m, width of 205m, average depth of 537m, average thickness of 2.55m, and average dip angle of 5°. it is buried under the loose layer with an average thickness of 448m. its coal mining method and roof management method are the comprehensive mechanized method with a full height mining at one time and the total collapse method respectively. using the icsa combined with the last observation data of the working face to calculate the pim parameters and the gm parameters. the calculated model parameters are shown in table 10. the comparison between the observations and the subsidence value fitted by the gm and the pim is shown in figure 6 and table 5. table 5. fitting effect comparison of the gm and the pim for the 12326 working face median error overall median error of edge area relative error overall(%) relative error of edge area (%) gm 64.8 42.1 37.0 50.8 pim 140.1 125.2 46.7 63.0 figure 6. comparison of the fitting effect of the gm and pim for 12326 working face (3) the working face 1613 (1) of the guqiao coal mine in huainan xieqiao mining area has a length of 1528m, width of 251m, average depth of 668m, average thickness of 2.8m, and average dip angle of 3°. it is buried under the loose layer with an average thickness of 420m. its coal mining method and roof management method are the comprehensive mechanized method with a full height mining at one time and the total collapse method respectively. using the icsa combined with the last observation data of the working face to calculate the pim parameters and the gm parameters. the calculated model parameters are shown in table 10. the comparison between the observations and the subsidence value fitted by the gm and the pim is shown in figure 7 and table 6. table 6. fitting effect comparison of the gm and the pim for the 1613(1) working face median error overall median error of edge area relative error overall(%) relative error of edge area (%) gm 28.4 29.4 17.5 23.2 pim 57.1 63.7 42.9 57.1 figure 7. comparison of the fitting effect of the gm and pim for 1613(1) working face (4) the working face 1414(1) of the guqiao coal mine in huainan xieqiao mining area has a length of 2120m, width of 251m, average depth of 735m, average thickness of 2.7m, and average dip angle of 5°. it is buried under the loose layer with an average thickness of 411m. its coal mining method and roof management method are the comprehensive mechanized method with a full height mining at one time and the total collapse method respectively. using the icsa combined with the last observation data of the working face to calculate the pim parameters and the gm parameters. the calculated model parameters are shown in table 10. the comparison between the 221research on the establishment of a mining subsidence prediction model under thick loose layer and its parameter inversion method observations and the subsidence value fitted by the gm and the pim is shown in figure 8 and table 7. table 7. fitting effect comparison of the gm and the pim for the 1414(1) working face median error overall median error of edge area relative error overall(%) relative error of edge area (%) gm 12.5 9.1 12.0 14.8 pim 48.6 53.6 68.6 86.2 figure 8. comparison of the fitting effect of the gm and pim for 1414(1) working face (5) the working face 1252(1) of the panyidong coal mine in huainan panji mining area has the length of 1148m, width of 262m, average depth of 802m, average thickness of 2.7m, and average dip angle of 6°. it is buried under the loose layer with an average thickness of 165m. its coal mining method and roof management method are the comprehensive mechanized method with a full height mining at one time and the total collapse method respectively. using the icsa combined with the last observation data of the working face to calculate the pim parameters and the gm parameters. the calculated model parameters are shown in table 10. the comparison between the observations and the subsidence value fitted by the gm and the pim is shown in figure 9 and table 8. table 8. fitting effect comparison of the gm and the pim for the 1252 (1) working face median error overall median error of edge area relative error overall(%) relative error of edge area (%) gm 22.6 22.7 20.1 21.3 pim 48.4 45.8 51.4 54.7 figure 9. comparison of the fitting effect of the gm and pim for 1252(1) working face (6) the working face 1312 (1) of gubei coal mine in huainan xieqiao mining area has the length of 629m, width of 205m, average depth of 528m, average thickness of 3.6m, and average dip angle of 5°. it is buried under the loose layer with an average thickness of 439.7m. its coal mining method and roof management method are the comprehensive mechanized method with a full height mining at one time and the total collapse method respectively. using the icsa combined with the last observation data of the working face to calculate the pim parameters and the gm parameters. the calculated model parameters are shown in table 10. the comparison between the observations and the subsidence value fitted by the gm and the pim is shown in figure 10 and table 9. table 9. fitting effect comparison of the gm and the pim for the 1312(1) working face median error overall median error of edge area relative error overall(%) relative error of edge area (%) gm 114.8 32.5 15.5 23.7 pim 123.1 122.6 41.5 77.8 figure 10. comparison of the fitting effect of the gm and pim for 1312(1) working face table 10. summary of the model parameters of the gm and pim calculated by the icsa q r  s1 / s2 / s3 / s4 1111(1) gm 0.43 242.1 87.1 5.5/-18.7/100.5/-19.5 pim 0.40 575.0 88.9 4.5/3.5/108.0/-11.7 12326 gm 1.20 131.0 89.5 -9.5/-11.5/46.5/70.5 pim 1.08 537/1.6 89.9 -1.7/-17.8/-68.5/-10.5 1613(1) gm 1.13 335.6 84.8 35.5/-5.5/-52.3/-26.3 pim 1.01 392.9 84.7 33.0/-4.3/59.0/35.3 1414(1) gm 1.15 183.8 88.0 37.0/-5.5/69.0/48.0 pim 0.90 459.4 87.5 26.0/75.5/120.4/109.5 1252(1) gm 0.90 222.8 88.0 29.5/90.5/123.5/114.5 pim 0.82 534.7 87.3 -30.7/7.0/43.5/34.5 1312(1) gm 1.36 125.7 87.6 -27.5/-5.5/73.0/19.5 pim 1.28 277.9 88.0 -42.5/-13.6/59.5/11.5 the six working faces selected in this article all buried under a thick loose layer. we can draw the conclusion from the above calculation results that the gm fitted better than the pim at the boundary of the subsidence basin. taking 1613 (1) as an example, the median error at the boundary of the subsidence basin fitted by the pim is 63.7, while fitted by the gm is 29.4, which is a reduction of 34.3; the relative error at the boundary of the subsidence basin fitted by the pim is 57.1%, while fitted by the gm is 23.2 %, which is a reduction of 33.9%. the fitting effect of the gm at the boundary 222 jingxian li, xuexiang yu, deshu chen, xinjian fang of the subsidence basin is significantly optimized compared with the pim in predicting the subsidence value caused by mining under thick loose layers. by comparing the gm parameters and the pim parameters in table 9, it is confirmed that the values of the two models parameters are basically the same except for � rgm and rpim . by comparing their values in table 9, it can be seen that rpim is about 2.5 times that of �rgm . results and discussion (1) through comparison, it is proved that the value at the boundary of the gm’s function curve is larger than that of the pim. which makes the gm more suitable than the pim for predicting surface subsidence caused by mining under thick loose layers. (2) in this paper, the icsa was established by adding operations to increase the information exchange between individuals on the basis of the csa. and through a simulated working face, it is proved that the stability and accuracy of the icsa can meet the requirements of coal mining subsidence prediction parameters inversion. (3) 6 working faces buried under thick loose layers in huainan mining area were selected in this paper to compare the fitting effect of the pim and the gm using the icsa. the results show that the gm has a better fitting result than the pim at the boundary of the subsidence basin. further confirmed that the gm is more suitable than the pim for predicting surface subsidence caused by mining under thick loose layers. and the parameters of the gm and the pim can take the same value except for r rgm pim / .2 5 . this result helps the gm to be used in actual engineering acknowledgement the research is supported by youth project of natural science foundation of anhui province (no.2008085qd178) references chang, z., wang, j., chen, m., ao, z., & yao, q. (2015). a novel ground surface subsidence prediction model for sub-critical mining in the geological condition of a thick alluvium layer. frontiers of earth science, 9(2), 330-341. https://doi.org/10.1007/s11707-014-0467-2 dai, h. y., li, w. c., liu, y. x., & jian, y. d. (2011). numerical simulation of surface movement laws under different unconsolidated layers thickness. transactions of nonferrous metals society of china, 21, s599-s603. https://doi.org/10.1016/s1003-6326(12)61647-1 diao, x., wu, k., zhou, d., li, l. (2016). integrating the probability integral method for subsidence prediction and differential synthetic aperture radar interferometry for monitoring mining subsidence in fengfeng, china. journal of applied remote sensing, 10(1). https://doi. org/10.1117/1.jrs.10.016028 du, s., zhang, j., deng, z., & li, j. (2014). a new approach of geological disasters forecasting using meteorological factors based on genetic algorithm optimized bp neural network. elektronika ir elektrotechnika, 20(4). doi: https://doi.org/10.5755/j01.eee.20.4.4238 fan, h. d. cheng, d., deng, k. z., chen, b. q., & zhu, c. g. (2015). subsidence monitoring using d-insar and probability integral prediction modelling in deep mining areas. survey review, 47(345). doi:10.1179/1752270614y.0000000153 guo, g. l., & wang, y. j. (2000). study of robust determining parameters model for probability-integral method and its application. acta geodaetica et cartographica sinica. li, m., zhang, j., sun, k., wu, z., & zhou, n. (2019). reducing surface subsidence risk using solid waste backfill technique: a case study under buildings. polish journal of environmental studies, 28(5), 3333-3341. doi: https://doi.org/10.15244/pjoes/94814 li, p., peng, d., tan, z., & deng, k. (2017). study of probability integration method parameter inversion by the genetic algorithm. international journal of mining science and technology, 1073-1079. https://doi. org/10.1016/j.ijmst.2017.06.006 liu, w., pang, l., xu, b., & sun, x. (2020). study on overburden failure characteristics in deep thick loose seam and thick coal seam mining. geomatics, natural hazards and risk, 11(1), 632-653. https://doi.org/ 10.1080/19475705.2020.1737584 shen, z., xu, l. j., liu, z., & qin, c. c. (2015). calculating on the prediction parameters of mining subsidence with probability integral method based on matlab. metal mine, 9, 10-174. ma, c., li, h., & zhang, p. (2017). subsidence prediction method of solid backfilling mining with different filling ratios under thick unconsolidated layers. arabian journal of geosciences, 10(23). doi: 10.1007/s12517-017-3303-7 wang, f., jiang, b., chen, s., & ren, m. (2019). surface collapse control under thick unconsolidated layers by backfilling strip mining in coal mines. international journal of rock mechanics and mining sciences, 113, 268-277. https://doi.org/10.1016/j.ijrmms.2018.11.006 wang, l., deng, k., & zheng, m. (2020). research on ground deformation monitoring method in mining areas using the probability integral model fusion d-insar, sub-band insar and offset-tracking. international journal of applied earth observations and geoinformation, 85. https://doi.org/10.1016/j.jag.2019.101981 wang, l., wei, t., li, n., chi, s. s., jiang, c., fang, s. y. (2018). research on probability integration parameter inversion of mining-induced surface subsidence based on quantum annealing. environmental geology, 77(21), 740.1-740.13. https://doi.org/10.1007/s12665-018-7927-z xing, x., zhu, y., yuan, z., xiao, l., liu, x., chen, l., xia, q., & liu, b. (2021). predicting mining-induced dynamic deformations for drilling solution rock salt mine based on probability integral method and weibull temporal function. international journal of remote sensing, 42(2), 639-671. https://doi.org/10.1080/01431161.2020.1813345 xuan, d., & xu, j. (2014). grout injection into bed separation to control surface subsidence during longwall mining under villages: case study of liudian coal mine, china. natural hazards, 73(2), 883-906. https:// doi.org/10.1007/s11069-014-1113-8 xu, h., liu, b., & fang, z. (2014). new grey prediction model and its application in forecasting land subsidence in coal mine. natural hazards, 71(2), 1181-1194. https://doi.org/10.1007/s11069-013-0656-4 xu, y., wu, k., li, l., zhou, d., & hu, z. (2019). ground cracks development and characteristics of strata movement under fast excavation: a case study at bulianta coal mine, china. bulletin of engineering geology and the environment, 78(1), 325-340. https://doi.org/10.1007/s10064017-1047-y xu, z., li, q., li, x., & caggiano, a. (2020). overburden migration and failure characteristics in mining shallow buried coal seam with thick loose layer. advances in materials science and engineering. https://doi.org/10.1155/2020/9024751 yang, w., & xia, x. (2013). prediction of mining subsidence under thin bedrocks and thick unconsolidated layers based on field measurement and artificial neural networks. computers and geosciences, 52, 199203. https://doi.org/10.1016/j.cageo.2012.10.017 yu, z., zhu, s., wu, y., & yu, h. (2020). study on the structural characteristics of the overburden under thick loose layer and thin-bed rock for safety of mining coal seam. environmental earth sciences, 79(1). zha, j. f., feng, w. k., & zhu, x. j. (2011). research on parameters inversion in probability integral method by genetic algorithm. journal of mining & safety engineering, 28(4), 655-659. zhao, k., xu, n., mei, g., & tian, h. (2016). predicting the distribution of ground fissures and water-conducted fissures induced by coal mining: a case study. springerplus, 5(1), 977. doi:10.1186/s40064-016-2609-3 zheng, m., deng, k., fan, h., & huang, j. (2019). monitoring and analysis of mining 3d deformation by multi-platform sar images with the probability integral method. frontiers of earth science, 13(1). https:// doi.org/10.1007/s11707-018-0703-2 zhou, d., wu, k., li, l., diao, x., & kong, x. (2016). a new methodology for studying the spreading process of mining subsidence in rock mass and alluvial soil: an example from the huainan coal mine, china. bulletin of engineering geology and the environment, 75(3). doi:10.1007/ s10064-016-0877-3 https://doi.org/10.1007/s11707-014-0467-2 https://doi.org/10.1016/s1003-6326(12)61647-1 https://doi.org/10.1117/1.jrs.10.016028 https://doi.org/10.1117/1.jrs.10.016028 https://doi.org/10.5755/j01.eee.20.4.4238 https://doi.org/10.15244/pjoes/94814 https://doi.org/10.1016/j.ijmst.2017.06.006 https://doi.org/10.1016/j.ijmst.2017.06.006 https://doi.org/10.1080/19475705.2020.1737584 https://doi.org/10.1080/19475705.2020.1737584 https://doi.org/10.1016/j.ijrmms.2018.11.006 https://doi.org/10.1016/j.jag.2019.101981 https://doi.org/10.1007/s12665-018-7927-z https://doi.org/10.1080/01431161.2020.1813345 https://doi.org/10.1007/s11069-014-1113-8 https://doi.org/10.1007/s11069-014-1113-8 https://doi.org/10.1007/s11069-013-0656-4 https://doi.org/10.1007/s10064-017-1047-y https://doi.org/10.1007/s10064-017-1047-y https://doi.org/10.1155/2020/9024751 https://doi.org/10.1016/j.cageo.2012.10.017 https://doi.org/10.1007/s11707-018-0703-2 https://doi.org/10.1007/s11707-018-0703-2 223research on the establishment of a mining subsidence prediction model under thick loose layer and its parameter inversion method zhou, d. w., wu, k., cheng, g. l., & li, l. (2015). mechanism of mining subsidence in coal mining area with thick alluvium soil in china. arabian journal of geosciences, 8(4). https://doi.org/10.1007/s12517014-1382-2 zhou, d. w., wu, k., miao, x. x., & li, l. (2018). combined prediction model for mining subsidence in coal mining areas covered with thick alluvial soil layer. bulletin of engineering geology and the environment, 77(1), 283-304. https://doi.org/10.1007/s10064-016-0961-8 zhu, x., zhang, w., wang, z., wang, c., li, w., & wang, c., & loupasakis, c. (2020). simulation analysis of influencing factors of subsidence based on mining under huge loose strata: a case study of heze mining area, china. geofluids. https://doi.org/10.1155/2020/6357683 https://doi.org/10.1007/s12517-014-1382-2 https://doi.org/10.1007/s12517-014-1382-2 https://doi.org/10.1007/s10064-016-0961-8 https://doi.org/10.1155/2020/6357683 _30j0zll _1fob9te _3znysh7 _3rdcrjn _aa8bucz8vd6w _26in1rg 15 many practical problems encountered in quantitative oriented disciplines entail finding the best approximate solution to an over determined system of linear equations. in this study, it is investigated the usage of different regression methods as a theoretical, practical and correct estimation tool in order to obtain the best empirical relationship between surface wave magnitude and rupture length for turkey earthquakes. for this purpose, a detailed comparison is made among four different regression norms: (1) least squares, (2) least sum of absolute deviations, (3) total least squares or orthogonal and, (4) robust regressions. in order to assess the quality of the fit in a linear regression and to select the best empirical relationship for data sets, the correlation coefficient as a quite simple and very practicable tool is used. a list of all earthquakes where the surface wave magnitude (ms) and surface rupture length (l) are available is compiled. in order to estimate the empirical relationships between these parameters for turkey earthquakes, loglinear fit is used and following equations are derived from different norms: ms = 1.02* logl + 5.18, for l2 norm regression (r2=0.71), ms = 1.15* logl + 4.98, for l1 norm regression (r2=0.92), ms = 1.04* logl + 5.15, for robust regression (r2=0.75), ms = 1.25* logl + 4.86, for orthogonal regression (r2=0.68), consequently, the empirical equation given by the least sum of absolute deviations regression as ms = 1.15* logl + 4.98 with a strong correlation coefficient (r2=0.92) can be thought as more suitable and more reliable for turkey earthquakes. also, local differences in rupture length for a given magnitude can be interpreted in terms of local variation in geologic and seismic efficiencies. furthermore, this result suggests that seismic efficiency in a region is dependent on rupture length or magnitude. muchos problemas prácticos encontrados en las disciplinas de orientación cuantitativa implican encontrar la mejor solución aproximada para un sistema determinado de ecuaciones lineales. en este estudio se investiga el uso de diferentes métodos de regresión tanto teóricos como prácticos y la herramienta de estimación correcta con el fin de obtener la mejor relación empírica entre la magnitud de onda superficial y la ruptura de longitud para los terremotos en turquía. para este propósito se hace una comparación detallada a partir de cuatro normas diferentes de regresión: (1) mínimos cuadrados, (2) mínimas desviaciones absolutas, (3) mínimos cuadrados totales y (4) regresiones robustas. con el fin de evaluar la regresión lineal adecuada y seleccionar la mejor relación empírica de grupos de datos empíricos, la correlación de coeficiente es una herramienta simple y muy práctica. se compiló una lista de todos los terremotos donde se cuenta con la magnitud de onda superficial (ms) y la ruptura de longitud superficial (l). con el fin de determinar las relaciones empíricas entre estos parámetros para los terremotos de turquía, se utiliza la regresión lineal adecuada y las siguientes ecuaciones se derivan de diferentes reglas. ms = 1.02* logl + 5.18, para la regresión normal l2 (r2=0.71), earth sciences research journal eart sci. res. j. vol. 18, no. 1 (june, 2014): 15 26 abstract resumen key words: turkey, correlation coefficient, l1, l2, orthogonal, robust regression palabras clave: turquía; coeficiente de correlación; l1, l2; regresión ortogonal; regresión robusta. a new empirical relation between surface wave magnitude and rupture length for turkey earthquakes serkan öztürk1,* 1gümüşhane university, department of geophysics, tr-29100, gümüşhane, turkey *corresponding author: serkan öztürk gümüşhane university, department of geophysics, tr-29100, gümüşhane / turkey email address: serkanozturk@gumushane.edu.tr phone: +90 456 233 74 25 / 26 fax: +90 456 233 10 75 seismology record manuscript received: 27/01/2014 accepted for publication: 01/06/2014 16 serkan öztürk introduction the regression is one of the most used tools in establishing the relationship between a response and an explanatory variable in applied statistics. an effective and accurate curve fitting method plays an important role and serves as a basic module in many scientific and engineering fields. there exists vast literature about the regression problem in mathematics, statistics and computer science and there are many statistical regression methods for this purpose. also, many software packages have allowed the practitioners to use the regression estimators to easily fit the data sets in order to determine whether or not there exists a relationship between hypothesized predictor or independent variables and some response or dependent variable (giloni and padberg, 2002). there are a number of regression methods in the literature such as l2 norm or least squares regression (cadzow, 2002), l1 norm or least sum of absolute deviations regression (giloni and padberg, 2002), total least squares or orthogonal regression (carroll and ruppert, 1996), robust regression (huber, 1964), principle components regression (pcr, maronna, 2005), geometric mean regression (gmr, leng et al. 2007), least trimmed squares regression (lts, rousseeuw and leroy, 1987) and covariate adjusted regression (car, şentürk and nguyen, 2006). the primary objective of this paper is to make an investigation on the usage of the best approximate regression solution to linear equations as a theoretical and practical estimation tool. for this purpose, a study on the comparison among the first four regression methods above is made in order to find the most suitable standard statistical model. the other methods such as pcr, gmr, lts or car are not discussed since such kind of methods are described for more specific fields and they have not common use in geophysical investigations. finally, these four different regression models are applied to turkey earthquakes in order to determine more reliable and an up to date empirical relationship between surface wave magnitude and rupture length. brief description of the algorithms for regression techniques there are different distance functions or metrics which can be utilized to perform a linear regression. therefore, the original problem is categorized under the class of mathematical problems. in order to solve these problems, a remarkably wide range of mathematical techniques can be found in the literature. however, the aim of this study is not to discuss the mathematical details of the regression methods. because the main goal of this work is to test different regression methods and to find the most suitable empirical relationship between surface wave magnitude and rupture length for turkey earthquakes, a brief and general reviews of four different regression approaches are given in this section as well as their fundamental statistical properties and qualities of fits. the linear regression problem is certainly one of the most important data analysis situations. in order to formulate a linear regression model, it is assumed that there are n measurements on the dependent variable y and some number p≥1 of independent variables χ1,……., χp of each one for which it is known n values as well. it is defined by giloni et al., (2006) as follows: (1) where y є r is a vector of n observations and x is an n x p matrix of real frequently referred to as the design matrix. furthermore, x1,…….,xp are column vectors with n components and x1,…….,xn are row vectors with p components corresponding to the columns and rows of x, respectively. the statistical (or hypothesized) linear regression model is given: y = xß + ε (2) where ßt = (ß1,........,ß2) is the vector of parameters of the linear model and ε t = (ε1,........,ε2) a vector of n random variables corresponding to the error terms in the asserted relationship. an upper index t denotes “transposition” of a vector or matrix. in the statistical model, the dependent variable y, thus, is a random variable for which we obtain measurements that contain some “noise” or measurement errors that are captured in the error terms e. on the other hand, for the numerical problem that we are facing, it can be written: y = xß + r (3) where given some arbitrarily fixed parameter vector β, the components ri of the vector r t = (r1,........,r2) are the residuals that result, given the observations y, a fixed design matrix x, and the chosen vector ß є rp . the residuals r are thus in terms of the statistical model, realizations of the random error terms e given the particular observations y and parameter settings β. given y and x, the general objective in linear regression is to find parameter settings ß є rp such that some appropriate measure of the dispersion of the resulting residuals r є rn is as small as possible (giloni and padberg, 2002). l2 norm or least squares regression least squares regression (l2 norm) is the most well-known and the most basic form of the least squares optimization problems. so, it has found many applications in mathematics and statistical data analyses as well as the other scientific fields. this statistical linear regression model has been studied intensively for well over 200 years now and under the assumption of a normal (or gaussian or exponential) distribution of the error terms e an impressive statistical apparatus has been created to assess the goodness of fit, the quality of individual and/or subsets of the regression coefficients, as well as other statistical properties of the linear regression model. if and when the error terms in the linear regression model do indeed follow a normal distribution, then the least squares regression estimates are “the best estimators” under most acceptable criteria. thus, this approach is ms = 1.15* logl + 4.98, para la regresión normal l1 (r2=0.92), ms = 1.04* logl + 5.15, para la regresión robusta (r2=0.75), ms = 1.25* logl + 4.86, para la regresión ortogonal (r2=0.68). por consiguiente, la ecuación empírica dada por la regresión de desviaciones mínimas absolutas es con un fuerte coeficiente de correlación (r2=0.92) que sería más apropiado y más confiable para los terremotos de turquía. también, las diferencias locales en la ruptura de longitud para una magnitud dada puede ser interpretada en términos de eficiencia sísmica y geológica en la variación local. además, el resultado sugiere que la eficiencia sísmicaen una región depende de la ruptura de longitud o de la magnitud. 17a new empirical relation between surface wave magnitude and rupture length for turkey earthquakes particularly helpful in all those situations involving the study of large data sets, handling large samples with a consistent numbers of outliers (e.g., cadzow, 2002; giloni and padberg, 2002). l1 norm or least sum of absolute deviations regression the least squares linear regression estimator is well-known to be highly sensitive to outliers in the data and as a result many more robust estimators have been proposed as alternatives. one of the earliest proposals was the leastsum of absolute deviations (l1) regression, where the regression coefficients are estimated through minimization of the sum of the absolute values of the residuals. l1 regression is called in the literature by a multitude of names that all mean the same: lad (least absolute deviation), lae (least absolute error), lav (least absolute value), lar (least absolute residual), lsad (least sum of absolute deviations), mad (minimum absolute deviation), msae (minimum sum of absolute errors), and so forth. an important advantage of l1 regression over l2 regression is its robustness. for the unconstrained problem it is well known that the l1 regression estimator can resist a few large errors in the data y (shi and lukas, 2005). the use of l1 norm criterion is appropriate when it is suspected that a small portion of the data being analyzed is unreliable (i.e., contains data outliers). the l1 norm criterion has the capability of effectively ignoring a few bad data points while emphasizing the majority of data points which more properly reflect the true nature of the data (cadzow, 2002). one can see huber (1987) for further discussion of the l1 approach in robust estimation. total least squares or orthogonal regression one of the most widely known techniques for errors-in-variables estimation in the simple linear regression model is orthogonal regression, also sometimes called as the functional maximum likelihood estimator under the constraint of known error variance ratio. in ordinary linear regression, the goal is to minimize the sum of the squared vertical distances between the y data values and the corresponding y values on the fitted line. in orthogonal regression the goal is to minimize the orthogonal (perpendicular) distances from the data points to the fitted line. this well-known orthogonal regression estimator is an old method and derived in many studies such as weisberg (1985), carroll and ruppert (1996) and leng et al., (2007). the use of orthogonal regression must include a careful assessment of equation error, and not merely the usual estimation of the ratio of measurement error variance. thus, when its assumptions hold, the orthogonal regression is a perfectly justifiable method of estimation. however it often lends itself to misuse by the unwary as a method, because orthogonal regression does not take equation error into account. robust regression the most serious problem with least squares regression is nonrobustness to outliers. in particular, if you have one extremely bad data point, it will have a strong influence on the solution because outliers can have a large influence on the regression parameters. a simple remedy is to iteratively discard the worst-fitting data point, and re-compute the least squares fit to the remaining data. another approach, termed robust regression, is to employ a fitting criterion that is not as vulnerable as least squares to unusual data. the most common general method of robust regression is m-estimation, introduced by huber (1964). nonlinear regression models play an important role in many fields. for estimating the parameters of a nonlinear model, the classical least squares method is commonly used in many cases. however, it is well known that these classical methods are often very sensitive to outliers and other departures from the underlying distribution. most robust developments on the estimation of regression models are based on the generalizations of least squares or maximum likelihood methods. robust regression procedure is less influenced by extreme values. in assessing the behavior of robust regression estimates, however, the techniques of small sample asymptotes can be very useful. the literature on the robust regression dates back to huber (1964). some of the robust techniques are also discussed in many studies such as huber (1981), sinha et al., (2003) and abdelkader et al., (2010). determination of correlation coefficient for regressions one of the most important problems in regression analyses is the selection of an appropriate probability distribution for a given data set, which can give reasonably accurate and robust estimates. as shown in the literature, there is no certain rule for selecting the appropriate distribution or parameter estimation techniques for a given data set and various distributions should be applied and the best-fitting model should be selected. thus, choosing the bestfitting model is quite an important procedure in frequency analysis. in most cases, the selection of an appropriate distribution is based on goodness-of-fit assessment. the goodness-of-fit technique can be described as the method for examining how well sample data agree with an assumed probability distribution as its population. several goodness-of-fit tests are developed and used in engineering decision-making. selection criteria among those methods, the determination of correlation coefficient (r2 or sometimes r is used) has been known as a powerful and conceptually simple method. although the r2 is solely based on the covariance penalty, it plays an important role in model fit assessment. it should certainly not be used as a unique model fit assessor, but can provide a reasonable and rapid model fit indication (heo et al. 2008). more traditionally, for the linear regression model, a very simple fit indicator is, however, given by the determination of correlation coefficient r2. if x is considered as random, it can be defined as the parameter that is the squared correlation between y and the best linear combination of the x. the r2 is usually presented as the quantity that estimates the percentage of variance of the response variable explained by linear relationship with the explanatory variables. it is computed by means of the ratio: (4) where ess, tss and rss are the explained, total and residual sum of squares, respectively. when there is an intercept term in the linear model, this determination of correlation coefficient is actually equal to the square of the correlation coefficient between yi and ŷi: (5) where y and ŷ denote the mean values of the observations yi and the fitted quantiles ŷi, respectively. equation (5) has a nice interpretation in that r2 measures the goodness of fit of the regression model by its ability to predict the response variable, ability measured by the correlation. the correlation coefficient is location and scale invariant, and statistically independent of the mean, y, and standard deviation. essentially, r2 measures the linearity of the probability plot, providing a quantitative assessment of fit and it is assumed that the observations could have been drawn from the fitted distribution if the value of r2 is close to 1.0 (heo et al., 2008). some characteristics of earthquakes used in the analysis and main geological environments around these earthquakes main goal of this study is to derive a relationship between the surface wave magnitude and rupture length for great earthquakes occurred in and around turkey. for this purpose, the earthquakes whose surface wave magnitude are greater than 5.5 and occurred between 1905 and 2005 were used. magnitude ranges vary from 5.5 to 8.0 and rupture length changes between 3 and 350 km for 63 earthquakes. these events and their detailed information can be found from different studies in the literature (e.g., ambreseys and jackson, 1998; nalbant et al. 1998; dewey, 1976). the epicenters of all earthquakes used in the analysis and whole tectonics are plotted in figure 1. in addition, some principle characteristics of several earthquakes and geological 18 serkan öztürk environments around these earthquakes are briefly discussed in this section. details of geological structure for whole epicenter regions are provided from the general directorate of mineral research & exploration (mta, http:// www.mta.gov.tr/v2.0/daire-baskanliklari/jed/index.php?id=500bas). figure 1. active fault map of turkey. the locations of earthquakes used in this study were also shown with different symbols. tectonic structures were modified from şaroğlu et al., (1992), bozkurt (2001), and ulusay et al., (2004). names of numbered faults or fault zones associated with the earthquake locations: 1-north east anatolian fault zone, 2-aşkale fault, 3-çaldıran fault, 4-başkale fault, 5-pülümür fault, 6-malatya fault, 7-ecemiş fault, 8-tuz lake fault, 9-ezinepazarı fault, 10-düzce fault, 11-eskişehir-i̇nönü-dodurga and kaymaz faults, 12-kütahya graben, 13-sultandağı fault, 14-beyşehir-tatarlı and kumdanlı faults,15-burdur fault zone, 16-gediz graben, 17-aliağa and dumlupınar faults, 18alaşehir graben, 19-simav graben, 20-soma graben, 21-zeytindağ-bergama faults, 22-etili fault, 23-yenice-gönen and sarıköy faults, 24-saros-gaziköy fault. the 1912 marmara (mürefte şarköy) earthquake was located between the gulf of saros and the sea of marmara. this earthquake was followed by several large aftershocks (august 10, 1912, ms = 6.3; september 13, 1912, ms = 6.9; september 27, 1912, ms = 6.6) to the west of the main shock. the surface rupture pattern of this earthquake was reported as a complex with a substantial right-lateral strike-slip component. nalbant et al., (1998) defined three fault segments with a vertical dip: the eastern one and middle ones have a reverse component, while the western one, bounding the gulf of saros, is pure strike slip. geological formation around the related earthquakes is mainly composed of clastic and carbonate rocks. another earthquake which is occurred in the same zone is the 1935 marmara earthquake. this earthquake occurred on the southwest extremity of the sea of marmara, north of the marmara islands. nalbant et al., (1998) modeled this earthquake as eastwest striking normal faulting dipping to north. they adopted a dip of 45° and mean displacement of 0.85 m. geological structure around this earthquake epicenter includes granitoid, schist, phillite, marble, metabasic rocks etc. there are two earthquakes in burdur region. the first one occurred in 1914 and the other one in 1971. these two earthquakes were located approximately in the same coordinates. the first earthquake occurred on the northeast segment of burdur fault and the other occurred on the southwest of burdur lake. focal mechanisms of the 1914 and 1971 burdur earthquakes are very similar. fault plane solution of the 1914 earthquake indicates normal faulting with a dip of 35° and a mean displacement of 1.6 m. focal mechanism of the 1971 earthquake shows pure normal faulting with a dip of 35° and a displacement of 0.7 m (taymaz et al. 1991). main geological structure around these epicenters includes undifferentiated quaternary, continental clastic rocks and undifferentiated continental clastic rocks. the 1919 soma (manisa) earthquake, in the south of the north anatolian fault, occurred in the eastern part of the bakırçay normal fault zone which also ruptured during the 1939 bergama earthquake. as east-west normal faulting and right-lateral strike-slip faulting are found in the area, nalbant et al., (1998) modeled this earthquake as right lateral normal fault dipping to north. they assumed a dip of 45° and mean displacement of 1.4 m for this earthquake. geological structure around this earthquake epicenter is given as undifferentiated quaternary by mta. the second earthquake in the same region is the 1969 demirci (manisa) earthquake. it was located on the north anatolian fault at the easternmost part of the sea of marmara, mudurnu valley. its fault plane solution based on teleseismic body-waveform inversion shows a e-w dextral strike-slip faluting mechanism (taymaz et al. 1991). nalbant et al., (1998) modeled this earthquake with dipping south at 45° and a displacement of 0.35 m. geological formation around this epicenter includes continental clastic rocks, dacite, rhyolite, rhyodacite, migmatite and gneiss. the other earthquake in the same region is march 28, 1969, alaşehir earthquake. this event occurred in the gediz river valley with 30 km surface rupture and extending from nw through alaşehir to se. the fault plane solution of this earthquake shows a normal faulting with a dip of 32° and a mean displacement of 0.61 m (eyidogan and jackson, 1985). geological structure around this epicenter includes undifferentiated quaternary, alluvial fan, slope debris, etc. the 1924 altıntaş (kütahya) earthquake occurred in the western anatolia, east of the 1970 gediz (kütahya) earthquake. the east-west normal faulting is a dominant mechanism in this region and was associated with 1956 eskişehir and the 1970 gediz earthquakes. nalbant et al., (1998) assumed that this event occurred on an east-west, normal fault located near epicenter with a dip of 45° which is compatible with the known tectonics of the area and mean displacement of 0.2 m. geological environments near the epicenter of this earthquake consist of undifferentiated quaternary, gneiss, metagranite, schist, marble, amphibolite etc. another earthquake around this area is 1928 emet earthquake and they show a similar tectonic environment. in this region, nnw-sse striking normal faults dipping to the ene were reactivated during the earthquakes in 1970 and 1944 (nalbant et al. 1998). different from 1924 altıntaş earthquake region, geological structure includes continental carbonate rocks. the third earthquake in the same region is 1944 şaphane earthquake. this earthquake was located in the western anatolia, southwest of gediz and also ruptured during the 1970 gediz earthquake. (nalbant et al., 1998) modeled this earthquake as nnw-sse normal fault with a dip of 45° and a mean displacement of 0.2 m. geological formation around this epicenter is formed by continental clastic rocks and pyrolastic rocks. the fourth earthquake in this region is 1970 gediz earthquake which is occurred in the western anatolia, east of the simav fault system. ambraseys and tchalenko (1972) mapped about 45 km of complicated normal faulting trending both to the nnw-sse and east-west down thrown to the east and north. eyidogan and jackson (1985) modeled the seismograms of this earthquake using three main subevents. the first subevent occurred on 15 km long nnw-sse segment with a mean displacement of 1.6 m and a dip of 35°. it then triggers the second subevent which ruptures the 24 km long east-west segment with a mean displacement of 2.4 m and a dip of 35°. all the remaining complexity of seismograms can be explain by slip on a~15° dipping fault extending the second fault segment from a depth of 12.5 to 17.5 km. geological structure around this epicenter includes continental clastic rocks. the 1930 hakkari earthquake was located near the iran-turkey border. focal mechanism of this earthquake display a right-lateral strike-slip faulting with a normal fault component with a dip of 36° and a mean displacement of 4.5 m (ambraseys and jackson, 1998). geological formation around this epicenter includes pyroclastic rocks, basalt, ophiolitic melange. the bergama (i̇zmir) earthquake was located in 1939 near the coast in the western anatolia, at the western extremity of the bakırçay normal fault zone. the focal mechanism solution indicates east-west normal faulting (ritsema, 1974). nalbant et al., (1998) modeled the rupture of this event to the dip at 45° and a mean displacement of 0.75 m. geological formation around this epicenter consists of generally undifferentiated quaternary and undifferentiated volcanic rocks (generally andesitic). the other earthquake in the same region is the 2005 urla earthquake. this earthquake occurred in the gulf of sığacık, in a region dominated by 19a new empirical relation between surface wave magnitude and rupture length for turkey earthquakes n–s extension and bounded by well-documented graben structures. the focal mechanisms of the 17 october 2005 earthquake seismic sequence show pure strike-slip motions in the gulf of sığacık and normal faulting combined with considerable strike-slip motion in the north and in the south (benenatos et al. 2006). geological structure around this epicenter includes neritic and lacustrine limestone, marl, shale etc. the 1939 erzincan earthquake occurred on the north anatolian fault zone was one of the most active strike-slip faults in the world. this earthquake nucleated on the eastern portion of the rupture. main body of the rupture extended along the master strand of the naf between erzincan and niksar basins. however, a part of 76 km-long the westernmost portion of the rupture directed towards on the ezinepazarı fault. ambraseys and zapotek (1969) stated that fault plane for this earthquake shows as a right-lateral mechanism. burumbaugh and pınar (2001) stated that the first-motion fault plane solution obtained at the epicentral location was a fault plane parallel to the easternmost surface rupture segment with a dip of 80° and mean displacement of 3.7 m. the other large earthquake in this region is the 1992 erzincan earthquake. this earthquake is located within a key region of eastern anatolia, which displays both expulsion of the anatolian block towards the west and regional n-s convergence due to the arabian-eurasian collision. the main focal mechanism is strike-slip with a slight normal component a dip of 68° and mean displacement of 1.0 m. (fuenzalida et al. 1997). main geological environments around these epicenters consist of undifferentiated quaternary, peridotite, slope debris, cone of dejection etc. the 1942 bigadiç (balıkesir) earthquake was located in the western anatolia, south of balıkesir. in this region, normal faults strike east-west and dip toward the north (westeway, 1990). nalbant et al., (1998) modeled the rupture of this earthquake as a12.5 km long normal fault and assume a dip of 45° and a mean displacement of 0.35 m. geological structure around this earthquake epicenter includes undifferentiated quaternary, undifferentiated volcanic rocks, dacite, rhyolite, rhyodacite. another event in the same city is the 1944 ayvacık earthquake. this earthquake occurred near the edremit gulf. the southern branch of the north anatolian fault reaches the aegean sea through the edremit gulf. the poorly constrained focal mechanism indicates nearly pure strike-slip (ritsema, 1974). however, nalbant et al., (1998) stated that it could have a normal component with an oblique slip on a 60° south dipping fault and a mean displacement of 1.6 m. geological formation around this epicenter consists of continental clastic rocks. the 1942 niksar-erbaa (tokat) earthquake was located in the erbaaniksar area, in the north anatolian fault zone from niksar in the kelkit river valley to the yeşilırmak river west of erbaa. focal mechanism of this earthquake shows right-lateral strike-slip faulting with a dip of 90° and a mean displacement of 1.8 m (ambraseys and jackson, 1998). geological structure around this epicenter includes undifferentiated quaternary, undifferentiated continental clastic rocks, schist, phyllite, marble and metabasic rocks. the 1943 hendek (adapazarı) earthquake occurred in the east of the sea of marmara in the mudurnu valley, 15 km north of the 1967 rupture. the focal mechanism of this earthquake is almost pure strike-slip motion parallel to the north anatolian fault. nalbant et al., (1998) modeled this event as a 16 km long right-lateral faulting with a dip of 90° and with 0.75 m of displacement (ambraseys and jackson, 1998). geological structure around this earthquake epicenter consists of undifferentiated volcanic rocks, clastic and carbonates rocks. the 1943 tosya-ladik (samsun) earthquake occurred near tosya, kastamonu province, in northern turkey. surface faulting was observed in section of the north anatolian fault zone from the destek gorge west of erbaa to the filyos river. focal mechanism of this earthquake shows rightlateral strike-slip faulting with a dip of 80° and with a mean displacement of 2.0 m (wells and coppersmith, 1994). geological structure around this earthquake epicenter is formed by undifferentiated quaternary. the 1944 gerede (bolu) earthquake was located on the north anatolian fault. nalbant et al., (1998) modeled this earthquake with a right-lateral strikeslip fault and with slip distribution 3.5 m to the west and decreases to 1.5 m in the east. geological formation around this epicenter consists of clastic and carbonate rocks. another event in the same region is 1957 abant earthquake. this earthquake occurred on the north anatolian fault. the focal mechanism of this earthquake indicates strike-slip faulting (mckenzie, 1972). nalbant et al., (1998) modeled this event with an average of 2.5 m strike-slip motion. geological structure near this earthquake epicenter is formed by clastic and carbonate rocks, alluvial fan, slope debris, cone of dejection etc. the other event in near region is july 22, 1967, mudurnu earthquake. this earthquake occurred in the east of the sea marmara on the north anatolian fault. the focal mechanism is given as a pure, east-west, strike slip faulting and mean displacement of 2.5 m (nalbant et al. 1998). geological environments near this epicenter include undifferentiated quaternary, undifferentiated volcanic rocks, clastic and carbonate rocks. the 1946 varto-hınıs (muş) earthquake occurred in varto district, province of muş. focal mechanism of this earthquake shows right-lateral strike-slip faulting with a dip of 58° and a mean displacement of 0.3 m (ambraseys and jackson, 1998). the other earthquake in the same region, the varto-üstükran earthquake of august 19, 1966, was located near the east end of the north anatolian fault system in turkey, very near where an earlier, less intense earthquake had caused damage in may 31, 1946. fault plane solution displays right-lateral strike slip faulting that having an extension of normal fault component with a dip of 65° and a mean displacement of 0.3 m (ambraseys and jackson, 1998). geological formation around these epicenters includes continental clastic rocks, pyroclastic rocks, basalt, slope debris, cone of dejection etc. the 1951 kurşunlu (çankırı) earthquake was related with north anatolian fault and occurred in the eastern tails of the i̇smetpaşa segment. focal mechanism of this earthquake shows right-lateral strike-slip faulting with a dip of 60° and a mean displacement of 0.6 m (nalbant et al. 1998). the june 6, 2000 orta (çankırı) earthquake was located about 25 km south of the north anatolian fault zone. the epicenter of this earthquake is close to a restraining bend in the e-w striking right-lateral strike-slip fault that moved in the much larger earthquake of august 13, 1951 kurşunlu earthquake. fault plane solution of 2000 earthquake shows normal faulting with a left lateral component with a dip of 69° and a mean displacement of 0.82 m (irmak, 2000). geological structure around these epicenters includes undifferentiated volcanic rocks, basalt, lacustrine limestone, marl, shale etc. the 1953 yenice-gönen earthquake was located between the sea of marmara to the north and the edremit gulf to the south. this earthquake ruptured the southern branch of the north anatolian fault. focal mechanism of this earthquake indicates pure southwest-northeast right-lateral strike-slip faulting (ambraseys, 1978). nalbant et al., (1998) modeled this earthquake using the slip reaching 3.5 m in the eastern and dropping to both ends. geological structure around this epicenter is formed by undifferentiated volcanic rocks, neritic limestone, schist and clastic rocks. the 1956 eskişehir earthquake was located in the western anatolia, 30 km west of eskişehir and 100 km north of gediz. this earthquake occurred on the eskişehir normal fault system (mckenzie, 1972). fault plane solution indicates east-west normal faulting. nalbant et al., (1998) modeled this event as resulting from the rupture of an east-west normal fault dipping to the north with an angle of 45° and a displacement of 0.30 m. geological formation around this epicenter consist of undifferentiated ophiolitic rocks, schist, fillite, marble and metabasite. the 1963 yalova-çınarcık earthquake was located in the southeast sea of marmara, north of the yalova. the fault plane solution is given as pure northwest-southeast normal faulting (taymaz et al. 1991). nalbant et al., (1998) modeled this event as a normal fault dipping northward with a dip of 60° and displacement of 0.60 m. geological structure around this earthquake epicenter includes continental clastic rocks, carbonate and clastic rocks. the 1964 manyas earthquake was located on the south of the sea of marmara between the manyas lake and the ulubat lake on the southern branch of the north anatolian fault. the focal mechanism solution shows east-west normal faulting (taymaz et al. 1991). nalbant et al., (1998) modeled this earthquake with a dip of 45° and a mean displacement of 1.2 20 serkan öztürk m. geological environments around this epicenter includes undifferentiated quaternary and continental clastic rocks. the first aegean earthquake which occurred in 1965 was located in the aegean sea at the southwest extremity of the north aegean through. the focal mechanism indicates right lateral strike-slip faulting on a northeastsouthwest plane (taymaz et al. 1991). nalbant et al., (1998) modeled this event with a slip of 0.6 m. the second earthquake in the same region is the 1967 aegean earthquake which was located in the aegean sea at the northwest extremity of the skyros basin, south of the north aegean through. the predominantly normal faulting focal mechanism (taymaz et al. 1991) defines two possible fault planes: one striking east-west with a dip toward the south, the other northwest-southeast with a dip toward the northeast. the major normal faults in the area bounding the western edge of skyros basin are oriented like the second nodal plane. nalbant et al., (1998) modeled this earthquake as a northwest-southeast normal fault with a dip of 45° and a displacement of 0.70 m. the third earthquake is december 19, 1981, aegean earthquake which occurred in the aegean sea, on the southern edge of the skyros basin. the fault plane solution shows right-lateral strike-slip faulting striking northeast-southwest (taymaz et al. 1991). nalbant et al., (1998) modeled this event with a dip of 79° and a mean displacement of 3.2 m. the fourth aegean earthquake occurred in the southwest extremity of the skyros basin on december, 27, 1981. its focal mechanism also indicates right-lateral faulting striking northeast-southwest (taymaz et al. 1991). nalbant et al., (1998) modeled this event with a dip of 79° and a mean displacement of 0.75 m. the fifth earthquake which was located in the same region is january, 18, 1982, aegean earthquake. this earthquake was located in the aegean sea, in the central part of the north aegean through. its fault plane solution shows right-lateral strike-slip faulting on a northeast-southwest plane (taymaz et al. 1991). nalbant et al., (1998) modeled this event with a dip of 62° and a mean displacement of 2.25 m. the last event in this region is 1983 aegean earthquake. this earthquake was located in aegean sea, just east of the previous one. the northeast elongation of the aftershock zone and the strikeslip focal mechanism are similar to the 1982 earthquake (taymaz et al. 1991). nalbant et al., (1998) modeled this event as a northeast-southwest rightlateral fault on the southern edge of the north aegean trough with dip of 83° and a displacement of 2.25 m. there is no information about the geological formation around these epicenters because these events occurred in the sea. the 1967 tunceli earthquake occurred in the part which is between pülümür and kalıova of north anatolian fault. fault plane solution of this earthquake shows right-lateral strike-slip faulting with a dip of 88° and a mean displacement of 0.2 m (ambraseys and jackson, 1998). the 2003 pülümür (tunceli) earthquake occurred on the pülümür fault that has 45° with the eastern part of the north anatolian fault. the epicenter of the earthquake was located at about 4.5 km w-sw sağlamtaş village. focal mechanism of this earthquake shows predominantly left-lateral strike-slip faulting that displayed a normal fault component with a dip of 71° and a mean displacement of 0.36 m (ozener et al. 2010). geological structure around these epicenters includes clastic and carbonate rocks, continental clastic rocks and undifferentiated volcanic rocks. the bartın (zonguldak) earthquake of september 3, 1968 is the strongest instrumentally recorded earthquake to occur along the black sea margin in northwestern turkey. the epicenter of the main shock was located 10 km north of amasra, in the black sea. the fault plane solution of this earthquake shows a thrust faulting with a dip of 38° and a mean displacement of 0.2 m (alptekin et al. 1986). geological formation around this epicenter consists of continental clastic rocks, clastic and carbonates rocks, neritic limestone. the may 22, 1971, bingöl earthquake was centered near bingöl. this earthquake produced surface breaks mostly along the southwestern half of the northeastern segment of the east anatolian fault between the karlıova triple junctions. earthquake mechanism of this event reveals left-lateral strike slip geometry with a dip of 45° and a mean displacement of 0.70 m (taymaz et al. 1991). the second earthquake in the same region is may 01, 2003, bingöl earthquake. this earthquake was located approximately 60 km southwest of the triple junction near karlıova, where north anatolian fault and east anatolian fault intersect. fault plane solution of this earthquake shows rightlateral strike-slip fault with a dip of 88° and a mean displacement of 0.9 m (milkereit et al. 2004). geological environments around these epicenters consist of undifferentiated volcanic rocks, schist, quvarzite, marble, phillite, etc. the earthquake of march 27, 1975, saros (çanakkale) earthquake was located on the west of the sea of marmara in the gulf of saros, a pullapart basin associated with the northern part of the north anatolian fault. fault plane solution shows a strike-slip, normal fault (a dip of 55°), rupture with the right lateral plane ene-wsw (taymaz et al. 1991). nalbant et al., (1998) modeled this earthquake as an oblique fault with a length of 20 km and displacement of 0.95 m. there is no information about the geological structure around this epicenter because this event occurred in the sea. the lice (diyarbakır) earthquake of 1975 was located near the town of lice, in the western part of the bitlis thrust zone. the slip vector in this event is thus directed ne, and may be related to the arabia-turkey motion, rather than the arabia-eurasia motion (moazami-goudarzi and akasheh, 1977). the mechanism of faulting for the lice earthquake of september 6, 1975 is thrust fault with a dip of 50° and a mean displacement of 0.5 m (wells and coppersmith, 1994). geological structure around this epicenter includes clastic and carbonate rocks. the 1976 çaldıran (van) earthquake occurred about 100 km east of the junction between the north and east anatolian faults, extending from 3 km west of sarıkök in the west to just west of baydoğan in the east. focal mechanism of this earthquake shows right-lateral strike-slip faulting with very small thrust component with a dip of 78° and a mean displacement of 2.05 m (wells and coppersmith, 1994). geological formation around this epicenter includes undifferentiated quaternary, undifferentiated continental clastic rocks, alluvial fans, slope debris, moraine etc. the 1983 horosan-narman (erzurum) earthquake took place on the ne-sw trending horosan fault, between horosan and şenkaya about 60 km east of erzurum basin. its fault plane solution indicates left-lateral strike slip geometry with a small thrust component, with a dip of 64° and a mean displacement of 1.2 m (wells and coppersmith, 1994). geological structure around this earthquake epicenter consists of pyroclastic rocks, undifferentiated volcanic rocks, and undifferentiated basic and ultrabasic rocks. the kars earthquake of 1988 december 7 was located within the lesser caucasus, which is dominated by a north-south compressive tectonics resulting from the young continental collision between the arabian plate and russian platform. focal mechanism of this earthquake was obtained as rightlateral strike-slip fault with a dip of 29° and a mean displacement of 1.6 m (wells and coppersmith, 1994). geological formation around this epicenter includes undifferentiated continental clastic rocks. the 1995 dinar (afyon) earthquake took place in the west anatolian province, in the lakes region horst-graben system. its epicenter is located near the dinar-çivril fault. focal mechanism of this earthquake indicates a normal faulting with a small lateral component with a dip of 56° and an average displacement of 0.2–0.4 m (pınar, 1998). geological structure around this epicenter consists of undifferentiated quaternary and neritic limestone. the other earthquakes in the same region occurred in the southwestern turkey on december 15, 2000 (sultandağı) and february 3, 2002 (eber) earthquakes. this earthquake sequence took place on the sultandağ fault. the 2000 earthquake occurred on the sultandağı fault, near the city of afyon in southwestern turkey whereas the 2002 earthquake occurred on the northwestern segment of the sultandağı fault. focal mechanisms of the 2000 earthquake indicate normal faulting with a slight left-lateral component with a dip of 41° and a mean displacement of 0.25 m. focal mechanism of the 2002 earthquake shows normal faulting with a dip of 38° and a variable displacement between 0.6 and 1.0 m (aksarı et al. 2010). geological environments around these epicenters include undifferentiated quaternary, undifferentiated continental clastic rocks, marble and recrystallized limestone. the 1998 adana-ceyhan earthquake occurred in southern turkey with its epicenter close to the city of adana, characterized by the advance of the arabic plate towards the north and the consequent wedging of the eurasian plate. this earthquake resulted by a left-lateral strike-slip faulting along the ne-trending east anatolian fault system and other fault zones, parallel to it, in the west. parameters of earthquake fault were given as 80° 21a new empirical relation between surface wave magnitude and rupture length for turkey earthquakes for dip and 0.37 m for mean displacement (lekkas and vassilakis, 1999; aktar et al. 2000). geological structure around this epicenter consists of undifferentiated quaternary. the 1999 i̇zmit earthquake, one of the most destructive earthquakes in turkey, was located in the western part of the north anatolian fault, in northwestern turkey in the gulf of i̇zmit region. the i̇zmit main shock occurred as a right-lateral strike slip faulting on an ew trending with a dip of 83° and a mean displacement of 5.0 m (tibi et al. 2001). geological environments around this epicenter consist of undifferentiated quaternary, clastic and carbonate rocks, alluvial fan, slope debris etc. the 1999 düzce earthquake was located in bolu basin in the adjacent fault segment associated with the i̇zmit earthquake. the focal mechanism of this earthquake indicates a right-lateral strike slip faulting on ew strike with a dip of 62° and a mean displacement of 3.0 m (tibi et al. 2001). geological structure around this epicenter undifferentiated quaternary, clastic and carbonate rocks, alluvial fan, slope debris, cone of defection etc. after the occurrence of an earthquake, the rupture length is related with the geological structure around the epicenter of the earthquake. at first glance, a relationship between seismicity and geologic structure is obvious because most earthquakes occur in regions where active faults are recognized. as given in details of related earthquakes, some geological features are nearly the same for different earthquakes. in many regions, geological formation in and around the earthquake epicenters is mainly composed of undifferentiated quaternary, continental clastic rocks or undifferentiated continental clastic rocks, undifferentiated volcanic rocks, clastic and carbonate rocks. however, alluvial fan, slope debris, cone of dejection, marble, schist, limestone etc. are dominant in some regions. thus, these variations in the geological formation surrounding the epicenters of earthquakes have a strong effect on the extent of different surface rupture lengths. results and discussions in the scope of this study, it is aimed to derive a new and reliable relationship between the main shock magnitude (ms) and surface rupture length (l) for great earthquakes (ms≥5.5) occurred in and around turkey. for this purpose, the earthquakes whose surface wave magnitudes are greater than 5.5 and the earthquakes which are occurred between 1905 and 2005 are used. some details such as origin times, locations, magnitudes and rupture lengths for these earthquakes are given in table 1. ms-value of may 1, 2003 bingöl earthquake is taken from the website of bogazici university, kandilli observatory and research institute (koeri). ms magnitude is preferred for this analysis since such kind of relations in literature was generally given with ms. as stated in hanks and kanamori (1979), mw can be calculated rather similar to ms for a number of earthquakes with ms≤8.0. reported magnitude types of almost all earthquakes in the scope of this study are generally given as ms in the literature. however, mw magnitudes of three earthquakes (15 december 2000 afyon, 3 february 2002 afyon and 17 october 2005 urla) are used instead of ms by taking into consideration hanks and kanamori (1979). thus, ms magnitude is used since any earthquakes with ms≥8.0 have not reported for the instrumental period. observations of surface ruptures usually accompany most large earthquakes and the logarithms of their corresponding dimensions (primarily length) have been found to relate linearly to earthquake magnitude. the empirical relationships that were established could predict not only the fault dimensions for a given magnitude, but also the maximum magnitude based on known fault dimensions. these relationships proved extremely useful to geotechnical, seismic hazard assessment and seismotectonic applications. a few authors calculated different empirical relationships connecting rupture lengths and magnitudes for earthquakes occurred in and around turkey and the world (e.g., acharya, 1979; wells and coppersmith, 1994; ambraseys and jackson, 1998). in this study, 63 earthquakes whose magnitude ranges vary from 5.5 to 8.0 and rupture length changes between 3 and 350 km are used for the analysis. these events and their information given in table 1 can be found in detailed from different studies in the literature. figure 2 shows the graphical representations of all regression fits between magnitude (ms) and rupture length (l) for turkey earthquakes. using four different regression methods, four different empirical relationships are obtained. log-linear fit is used for all regressions and following equations are derived: ms = 1.02* logl + 5.18, for l2 norm regression (6) ms = 1.15* logl + 4.98, for l1 norm regression (7) ms = 1.04* logl + 5.15, for robust regression (8) ms = 1.25* logl + 4.86, for orthogonal regression (9) although the choice of confidence coefficient is somewhat arbitrary, in practice 90%, 95%, and 99% intervals are often used, with 95 % being the most commonly used. so, confidence interval limits of each regression relation used in this study were calculated for 90%, 95%, and 99% confidence interval and these values were given in table 2 in detailed. these confidence intervals are designed to estimate statistical characteristics of sampled data. however, confidence intervals are more flexible and can be used practically in more situations. confidence interval generates a lower and upper limit and these limits define the most probable concentration range in which the true parameter ought to lie. the interval estimate also gives an indication of how much uncertainty there is in the estimate of the true value. for this study, there are reliable surface wave magnitudes and rupture lengths in the instrumental period for 63 earthquakes as given in table 1. from the regression analyses, l2 norm and robust regressions between ms and log (l) give a standard deviation of 0.03 in ms, l1 norm and orthogonal regressions give a standard deviation of 0.04 in ms, the confidence interval limits computed from given data contains the mean values which is calculated from the regression relations in the equations 6 to 9. that is, for all lower and upper confidence interval limits (90%, 95% and 99%), as shown in table 2, these intervals cover the mean values for all regression. also, the number of earthquakes in 90% confidence limit of the regressions was calculated for each regression and it was found as 14 events for l2 norm, 21 events for l1 norm, 14 events for robust and, 17 events for orthogonal regression. for 99% confidence limit of the regressions, the number of earthquakes was calculated as 25 events for l2 norm, 25 events for l1 norm, 23 events for robust and, 29 events for orthogonal regression. 22 serkan öztürk table 1. a list of great turkey earthquakes between 5.5 and 8.0 occurred between 1905 and 2005 associated with surface phenomena. date (d.m.y) location magnitude (ms) rupture length (km) references 04.12.1905 çemişgezek 6.8 38 nalbant et al., (2002) 09.02.1909 ender 6.4 15 ambraseys and jackson (1998) 09.08.1912 marmara 7.4 90 nalbant et al., (1998) 13.09.1912 marmara 6.8 37 ambraseys and jackson (2000) 03.10.1914 burdur 7.0 23 ambraseys and jackson (1998) 18.11.1919 soma 6.9 30 nalbant et al., (1998) 13.09.1924 pasinler 6.8 15 eyidoğan et al., (1999) 20.11.1924 altıntaş 6.0 10 nalbant et al., (1998) 02.05.1928 emet 6.2 12.5 nalbant et al., (1998) 06.05.1930 hakkari 7.2 30 ambraseys and jackson (1998) 04.01.1935 marmara 6.4 20 nalbant et al., (1998) 19.04.1938 kırşehir 6.6 36 acharya (1979) 22.09.1939 bergama 6.5 17 nalbant et al., (1998) 26.12.1939 erzincan 8.0 350 ambraseys and zatopek (1969) 15.11.1942 bigadiç 6.2 12.5 nalbant et al., (1998) 20.12.1942 niksar-erbaa 7.0 50 acharya (1979) 20.06.1943 hendek 6.4 16 nalbant et al., (1998) 26.11.1943 tosya-ladik 7.3 280 dewey (1976) 01.02.1944 gerede 7.3 180 dewey (1976) 25.06.1944 şaphane 6.0 18 ambraseys and jackson (1998) 06.10.1944 ayvacık 6.8 37 ambraseys and jackson (2000) 31.05.1946 varto-hınıs 6.0 9 welss and coppersmith (1994) 17.08.1949 elmalıdere 7.1 76 hearn et al., (2002) 13.08.1951 kurşunlu 6.8 49 acharya (1979) 18.03.1953 yenice-gönen 7.2 58 wells and coppersmith (1994) 16.07.1955 aydın-söke 6.9 35 pavlides and caputo (2004) 20.02.1956 eskişehir 6.1 11 nalbant et al., (1998) 26.05.1957 abant 7.0 40 ambraseys and jackson (1998) 18.09.1963 yalova-çınarcık 6.3 35 turgut (2007) 06.10.1964 manyas 6.8 40 ambraseys and jackson (1998) 09.03.1965 aegean sea 6.3 14 nalbant et al., (1998) 19.08.1966 varto-üstükran 6.8 38 acharya (1979) 04.03.1967 aegean sea 6.5 18 nalbant et al., (1998) 22.07.1967 mudurnu 7.1 80 ambraseys and zatopek (1969) 26.07.1967 tunceli 6.0 4 ambraseys and jackson (1998) 30.07.1967 mudurnu 5.5 3 ambraseys and jackson (1998) 03.09.1968 bartın 6.5 10 turgut (2007) 23.03.1969 demirci 6.0 10 eyidoğan and jackson (1985) 25.03.1969 alaşehir 6.1 5 ambraseys and jackson (1998) 28.03.1969 alaşehir 6.4 30 dewey (1976) 28.03.1970 gediz 7.1 41 wells and coppersmith (1994) 12.05.1971 burdur 6.2 4 ambraseys and jackson (1998) 22.05.1971 bingöl 6.7 38 acharya (1979) 27.03.1975 saros 6.6 20 nalbant et al., (1998) 06.09.1975 lice 6.6 28 ambraseys and jackson (1998) 24.11.1976 çaldıran 7.3 55 wells and coppersmith (1994) 19.12.1981 aegean sea 7.2 60 nalbant et al., (1998) 27.12.1981 aegean sea 6.4 16 nalbant et al., (1998) 18.01.1982 aegean sea 6.9 30 nalbant et al., (1998) 06.08.1983 aegean sea 6.9 30 nalbant et al., (1998) 30.10.1983 horasan-narman 6.9 12 wells and coppersmith (1994) 07.12.1988 kars 6.7 33 ambraseys and jackson (1998) 13.03.1992 erzincan 6.8 30 grosser et al., (1998) 01.10.1995 dinar 6.1 25 pınar (1998) 27.06.1998 adana-ceyhan 6.2 30 yalcinkaya (2005) 17.08.1999 i̇zmit 7.4 145 barka et al., (2002) 23a new empirical relation between surface wave magnitude and rupture length for turkey earthquakes 12.11.1999 düzce 7.1 40 akyüz et al., (2002) 06.06.2000 orta 6.1 10 irmak (2000) 15.12.2000 afyon 6.0mw 22 aksarı et al., (2010) 03.02.2002 afyon 6.5mw 40 aksarı et al., (2010) 27.01.2003 pülümür 6.0 15 usgs/neic* 01.05.2003 bingöl 6.4k 20 milkereit et al., (2004) 17.10.2005 urla 5.8mw 10 benetatos et al., (2006) * usgs: united states geological survey, neic: national earthquake information center, 6.4k: taken from koeri (http://udim.koeri.boun.edu.tr/zeqdb/), mw: moment magnitude (used as ms magnitude taking into consideration hanks and kanamori, 1979), ms: surface wave magnitude figure 2a figure 2b figure 2c figure 2d figure 2. surface wave magnitude (ms) versus surface rupture length (l, km) for turkey earthquakes listed in table 1: a) l2 norm regression fit, b) l1 norm regression fit, c) robust regression fit, d) orthogonal regression fit. corresponding equations and their correlation coefficients were also given. solid lines indicate the regression fits to data and dashed line indicates 95% confidence interval for the regressions. table 2. confidence interval limits for each regressions. confidence interval limits l2 norm regression limits l1 norm regression limits robust regression limits orthogonal regression limits 90% 95% 99% 90% 95% 99% 90% 95% 99% 90% 95% 99% lower limits 5.61 5.60 5.58 5.47 5.46 5.44 5.59 5.58 5.56 5.40 5.38 5.36 upper limits 7.83 7.84 7.86 7.97 7.98 8.00 7.86 7.87 7.89 8.09 8.10 8.13 24 serkan öztürk as shown in figures 2a to 2d, correlation coefficients of these regression fits are nearly the same as l2 (r2=0.71) norm and orthogonal regression (r2=0.68). correlation coefficient for robust regression is calculated as 0.75. however, this value is found as r2=0.92 for l1 norm and seems quite good. ms versus l for whole regression methods with fit curves, corresponding equation and 95% confidence interval are given in figures 2a to 2d. also, the number of earthquakes for 95% confidence limit of the regressions is calculated in all methods and it is found as 16 events for l2 norm, 22 events for l1 norm, 20 events for robust and, 22 events for orthogonal regression. thus, the highest number of earthquakes in 95% confidence interval is observed for l1 norm and orthogonal regression. regression analysis is routinely used by researchers in many disciplines to fit mathematical models to observed data. linear least squares estimates can behave badly when the error distribution is not normal, particularly when the errors are heavy-tailed. the traditional estimation technique of least squares is efficient if the error terms are independent of the regressors and are identically and independently distributed as a normal. while the unobserved random disturbances in a regression model are often assumed to be normally distributed, real data are often replete with outliers that lie far from the pattern evidenced by a majority of the data. while these errors may be the result of measurement inaccuracies or human recording error, many outliers are generated by genuinely thick-tailed or asymmetric error distributions. in such cases, discarding outliers is inappropriate since they are representative of the true data generating process (boyer et al. 2003). the use of any regression criterion is appropriate when it is suspected that a small portion of the data being analyzed is unreliable (i.e., contains data outliers). regression criterions have the capability of effectively ignoring a few bad data points while emphasizing the majority of data points which more properly reflect the true nature of the data. an insightful and useful characterization of a solution to the problem is to minimize the sum of residual error magnitude. the sum of error magnitudes is of particular use in many applications where the data vector contains a small number of data outliers. in such cases, the sum of squared errors criterion is unduly influenced by these data outliers and thereby often leads to a poor selection of the coefficient vector. the sum of error magnitudes, however, tends to ignore the data outliers provided that they are relatively few in number. in often happens that these data points cluster about a line. as shown in figures 2a to 2d, four plots of data set provides a visual mechanism for determining the basic nature of the relationships between surface wave magnitude and rupture length. although the data concentration is between 10 and 100 km and the data does not include outliers, lots of the data remain outside of the confidence limits in all regression fits. however, there are the same values of rupture length for different magnitude values. for example, also can be seen in table 1, for a rupture length of 30 km there are four different magnitude values such as 6.2, 6.4, 6.8, 6.9 and 7.1 (there are another examples like these values in table 1). however, all these earthquakes (as seen in figures 2a to 2d) remained outside the confidence intervals. this means that obtained regressions cannot always reflect the characteristics of earthquakes. as stated above, geological structure has a strong effect on the rupture length. as a result, it is apparent that these regression fits conforms more accurately to the given data points and tends to ignore these data outliers. as shown in figures 2a to 2d, all data interval for rupture lengths is used in the analysis. in order to make a detailed assessment, the earthquakes were grouped into different three sub-groups according to their rupture lengths. all sub-groups and calculations are given in table 3. in order to obtain the regression relations, the analyses were done for 1 to 10 km, 11 to 50 km, and larger than 51 km. to make a sub-group of rupture lengths, earthquake magnitude level given in figure 1 were also taken into account. as shown in figure 1, the earthquakes were grouped into three broad categories: 5.5≤ms≤6.0 (9 events), 6.1≤ms≤7.0 (41 events), and ms≥7.1 (13 events). these magnitude intervals are accordance with the rupture length intervals given above. that is, the earthquakes having 0-10 km (10 events) rupture lengths have the magnitude interval which generally varies from 5.5 to 6.0. also, the earthquakes having 11-50 km (43 events) rupture lengths have the magnitude interval between 6.1 and 7.0, and the earthquakes having larger than 51 (10 events) km rupture lengths have the magnitude interval higher than 7.1. thus, the rupture lengths intervals in table 3 were prepared in properly to the magnitude groups in tectonic map. it can be clearly seen that correlation coefficients in all regressions are rather weak for all sub-groups. these calculated weak coefficients may be result of a small number of data or large scattering of data. also, these results indicate that subdividing the data set according to various rupture lengths does not greatly improve the statistical significance of the regressions and these regressions may provide more reliable results with the usage of all data set. thus, the results suggest that separating the earthquakes by their rupture lengths is insignificant for rupture length relationships. several empirical relationships between rupture length and magnitude from many different regions around the world including turkey are proposed by several authors. acharya (1979) found following equation for 11 earthquakes occurred in turkey: ms = 0.92* logl + 5.33 (with a correlation coefficient = 0.83) (10) wells and coppersmith (1994) compiled a large dataset consisting of 244 earthquakes from around the world, along with reliable estimates of their source dimensions and moment magnitudes. they developed a series of empirical relationships for fault length, width, area, maximum and average displacement for the whole of their dataset but also separately for each faulting-type group (normal, thrust, strike-slip). they calculated following relation for global earthquakes: ms = 1.16* logl + 5.08 (with a correlation coefficient = 0.89) (11) ambraseys and jackson (1998) proposed following empirical relationship for eastern mediterranean earthquakes covering turkey: ms = 1.04* logl + 5.27 (12) table 3. some analysis summing up the trials for different sub-groups of data. surface rupture length (km) l2 norm regression fit l1 norm regression fit robust regression fit orthogonal regression fit 0-10 ms=0.58*logl+5.57 (r2=0.25) ms=0.25*logl+5.84 (r2=0.05) ms=0.12*logl+5.96 (r2=0.02) ms=1.34*logl+4.93 (r2=0.18) 11-50 ms=1.04*logl+5.14 (r2=0.34) ms=1.15*logl+4.98 (r2=0.42) ms=1.15*logl+4.98 (r2=0.42) ms=2.47*logl+3.13 (r2=0.30) ≥51 ms=0.61*logl+6.08 (r2=0.48) ms=0.20*logl+6.84 (r2=0.14) ms=0.44*logl+6.40 (r2=0.26) ms=0.84*logl+5.62 (r2=0.41) 25a new empirical relation between surface wave magnitude and rupture length for turkey earthquakes it is well known that the length of rupture at the surface is related with earthquake magnitude and many studies in literature relate magnitude to rupture length or some another earthquake source parameters for regions of different geographic or tectonic setting (e.g., acharya, 1979; wells and coppersmith, 1994). with developing a regression relationship among magnitude and rupture length, it can be predicted an expected value for a dependent parameter from an observed independent parameter. independent or dependent parameters will depend on the application, either the expected magnitude for a given rupture length or the expected fault length for a given magnitude. the suggested empirical relationship in this study can be used to assess maximum earthquake magnitude for a particular fault zone or an earthquake source. the assumption that a given magnitude is a maximum value is valid only if the rupture length also is considered a maximum value. evaluating the segmentation of a fault zone provides a basis for assessing the maximum length of future ruptures. given that the length and magnitude are assessed to be maximum values, empirical relations between magnitude and rupture length will provide the expected maximum magnitudes. these are expected maximum magnitudes for the given maximum fault lengths. thus, it can be concluded that rupture length regressions presented here are appropriate for estimating magnitudes for expected ruptures along the related fault segments. applying rupture length relations to estimating magnitudes may help to overcome uncertainties associated with estimating the surface rupture length for some seismic sources (wells and coppersmith, 1994). also, for engineering estimation purposes, acharya (1979) stated that the use of mean magnitude in magnitude rupture-length relationship should be sufficient since there is considerable evidence that beyond certain earthquake magnitude. consequently, a comparison of the empirical relationships is made between magnitude and rupture length considering the correlation coefficients. as stated above, the correlation coefficient and the number of events in confidence limit of l1 norm regression is the highest among the others. this value is very close to 1.0 and this means that it provides more quantitative assessment of fit than the other l2, robust and orthogonal regression. as a result, it can be suggested that equation (7) is more suitable empirical relationship between surface wave magnitude and surface rupture length for earthquakes occurred in and around turkey. also, the predictive relationship between magnitude and rupture length for the instrumental period is almost identical to those derived by different authors for turkey earthquakes. conclusions in this study, an application of different statistical regression methods is made and it is discussed how can be decided for the selection of the best regression method for a given data set. then, a linear relationship between the surface wave magnitude (ms) and rupture lengths (l) is derived for turkey earthquakes. for this purpose, the earthquakes which are in the time interval between 1905 and 2005 and whose magnitudes are greater than 5.5 are used. in the analyses, there are 63 great earthquakes whose magnitude ranges vary from 5.5 to 8.0 and rupture length changes between 3 and 350 km. in order to estimate the best empirical relationship from different statistical models, four regression norms are used as (i) least squares, (ii) least sum of absolute deviations, (iii) total least squares or orthogonal and, (iv) robust regression. in order to identify the quality of the fit in a linear regression and to select the most suitable empirical relationship for data set, the correlation coefficient (r2) as a quite simple and very practicable tool is used. the present research has confirmed a clear and straightforward relationship between magnitude (ms), the surface rupture length (l) for turkey earthquakes. there is not a general equation between ms and l for turkey earthquakes and only a few authors suggested different relationships for different parts of turkey. in order to obtain more appropriate and reliable empirical relationship between magnitude and the surface rupture length of turkey earthquakes, log-linear scale is used in four different regressions and different equations are derived. as a result, the relationship calculated as ms = 1.15* logl + 4.98 with a strong correlation coefficient r2=0.92 given by l1 norm is suggested for turkey earthquakes. acknowledgments the author would like to thank to professor dr. hakan karsli for his helps in preparing the analysis codes and anonymous reviewers for their useful and constructive suggestions in improving this paper. also, the computer programs used in this study are partially covered by gumushane university (turkey) with project no 2012.02.1717.2. references abdelkader, g., ali, l., & rachida, r., (2010). robust nonparametric estimation for spatial regression, journal of statistical planning and inference, 140, 1656-1670. acharya, h.k., (1979). regional variations in the rupture-length magnitude relationships and their dynamical significance, bulletin of the seismological society of america, 69(6), 2063-2084. aksari, d., karabulut, h., & özalaybey, s., (2010). stress interactions of three moderate size earthquakes in afyon, southwestern turkey, tectonophysics, 485, 141-153. aktar, m., ergin, m., ozalaybey, s., tapirdamaz, c., yörük, a., & bicmen, f., (2000). a lower crustal event in the northeast mediterranean: the 1998 adana earthquake (mw=6.2) and its aftershocks., geophys. res. lett., 27, 2361–2364. akyüz, h.s., hartleb, r., barka, a., altunel, e., sunal, g., meyer, b.,& armijo, r., (2002). surface rupture and slip distribution of the 12 november 1999 düzce earthquake (m 7.1), north anatolian fault, bolu, turkey, bulletin of the seismological society of america 92(1), 61-66. alptekin, ö., nábĕlek, j.l., & toksöz, m.n., (1998). source mechanism of the bartin earthquake of september 3, 1968 in northwestern turkey: evidence for active thrust faulting at the southern black sea margin, tectonophysics, 122(1–2), 73–88. ambraseys, n.n., & zatopek, s. (1969). the mudurnu valley, west anatolia, turkey, earthquake of 22 july 1967, bulletin of the seismological society of america, 59(2), 521-589. ambraseys, n.n., & tchalenko, j.s., (1972). seismotectonic aspects of the gediz, turkey, earthquake of march 1970, geophys. j. r. astron. soc., 30, 229-252. ambraseys, n.n., (1978). middle east-a reappraisal of the seismicity, engineering geology, 11, 19-32. ambraseys, n.n., & jackson j.a., (1998). faulting associated with historical and recent earthquakes in the mediterranean region, geophysics journal international, 133, 390-406. ambraseys, n.n., & jackson, j.a., (2000). seismicity of the sea of marmara (turkey) since 1500, geophysics journal international, 141, f1-f6. barka, a., akyuz, h.s., altunel, e., sunal, g., cakir, z., dikbas, a., yerli, b., armijo, r., meyer, b., chabalier, j.b.de., rockwell, t., dolan, j.r., hartleb, r., dawson, t., christofferson, s., tucker, a., fumal, t., langridge, r., stenner, h., lettis, w., bachhuber, j., & page, w., (2002). the surface rupture and slip distribution of the 17 august 1999 i̇zmit earthquake (m 7.4), north anatolian fault, bulletin of the seismological society of america, 92(1), 43-60. benetatos, c., kiratzi, a., ganas, a., ziazia, m., plessa, a., & drakatos, g., (2006). strike-slip motions in the gulf of siğaçik (western turkey): properties of the 17 october 2005 earthquake seismic sequence, tectonophysics, 426, 263–279. boyer, b.h., mcdonald, j.b., & newey, w.k., (2003). a comparison of partially adaptive and reweighted least squares estimation, econometric reviews, 115–134. bozkurt, e., 2001. neotectonics of turkey-a synthesis, geodinamica acta, 14, 3-30. brunbaugh, d.s & pınar, a., (2001). preliminary results of study of the mechanism of the december 26,1939 erzincan earthquake (m=7.9), american geophysical union, fall meeting 2001, abstract #s52e-0690. cadzow, j.a., (2002). minimum ℓ1, ℓ2 and ℓ∞ norm approximate solutions to an overdetermined system of linear equations, digital signal processing, 12, 524-560. 26 serkan öztürk carrol, r.j., & ruppert, d., (1996). the use and misuse of orthogonal regression estimation in linear errors-in-variables models, the american statistician, 50, 1-6. dewey, j.w., (1976). seismicity of northern anatolia, bulletin of the seismological society of america, 66(3), 843-868. eyidoğan, h., & jackson, j.a., (1985). seismological study of normal faulting in the demirci, alaşehir and gediz earthquakes of 196970 in western turkey: implications for the nature and geometry of deformation in the continental crust, geophysical journal of the royal astronomical society, 81, 569-607. eyidoğan, h., nalbant, s.s., barka, a., & king, g.j.p., 1999. static stress changes induced by the 1924 pasinler (m=6.8) and 1983 horosannarman (m=6.8) earthquakes, northeastern turkey, terra nova, 11(1), 38-44. fuenzalida, h., dorbath, l., cisternas, a., eyidoğan, h., barka, a., rivera, l., haessler, h., philip, h., & lyberis, n., (1997). mechanism of the 1992 erzincan earthquake and its aftershocks, tectonics of the erzincan basin and decoupling on the north anatolian fault, geophys. j. int., 129, 1-28. giloni, a., & padberg, m., (2002). alternative methods of linear regression, mathematical and computer modeling, 35, 361-374. giloni, a, simonoff, j.s., & sengupta, b., (2006). robust weighted lad regression, computational statistics and data analysis, 50, 3124-3140. grosser, h., baumbach, m., berckhemer, h., baier, b., karahan, a., schelle, h., krüger, f., paulat, a., michel, g., demirtas, r., gencoglu, s., & yılmaz, r., (1998). the erzincan (turkey) earthquake (ms 6.8) of march 13, 1992, and its aftershock sequence, pure applied geophysics, 152, 465-505. hanks, t.c., & kanamori, h., (1979). a moment magnitude scale, journal of geophysical research, 85, 2348-2350. hearn, h.e., hager, b.h., & reilinger, r.e., 2002. viscoelastic deformation from north anatolian fault zone earthquakes and the eastern mediterranean gps velocity field, geophysical research letters, 29, 10.1029/2002gl014889, x-1, x-3. heo, j.h., kho, y.w., shin, h., kim, s., & kim, t., (2008). regression equations of probability plot correlation coefficient test statistics from several probability distributions, journal of hydrology, 355, 1-15. huber, p.j., (1964). robust estimation of a location parameter, annals of mathematical statistics, 35, 73-101. huber, p.j., 1981. robust statistics, 2nd edn, pp. 354, wiley, newyork. huber, p.j., (1987). the place of the l1 norm in robust estimation, in: dodge, y. (ed.), statistical data analysis based on the l1 norm and related methods, north-holland, amsterdam, 23-33 pp. irmak, t.s., (2000). the source-rupture processes of recent large turkey earthquakes, individual studies, international institute of seismology and earthquake engineering, 36, 131-143. lekkas, e., & vassilakis, e., (1999). adana earthquake: peculiar damage distribution and seismotectonic characteristics, advances in earthquake engineering, earthquake resistant engineering structures, 4, 785-798. leng, l., zhang, t., kleinman, l., & zhu, w., (2007). ordinary least square regression, orthogonal regression, geometric mean regression and their applications in aerosol science, journal of physics, conference series, 78, doi:10.1088/1742-6596/78/1/012084. maronna, r., (2005). principal components and orthogonal regression based on robust scales, technometrics, 47, 264-273. mckenzie, d.p., (1972). active tectonics of the mediterranean region, geophys. j. r. astron. soc., 30, 109-185. milkereit, c., grosser, h., wang, r., wetzel, h.u., woith, h., karakisa, s., zünbül, s., & zschau, j., (2004). implications of the 2003 bingöl earthquake for the interaction between the north and east anatolian faults, bulletin of the seismological society of america, 94(6), 2400-2406. moazami-goudarzi, k., & akasheh, b., (1977). the earthquake of september 6, 1975 in lice (eastern turkey), tectonophysics, 40, 361-368. nalbant, s.s., hubert, a., & king, g.j.p., (1998). stress coupling between earthquakes in northwest turkey and the north aegean sea, journal of geophysical research, 103, 24469-24486. nalbant, s., mcclokey, j., steacy, s., & barka, a., (2002). stress accumulation and increased seismic risk in eastern turkey, earth and planet science letters, 195, 291-298. ozener, h., arpat, e., ergintav, s., dogru, a., cakmak., r., turgut, b., dogan, u., (2010). kinematics of the eastern part of the north anatolian fault zone, journal of geodynamics, 49, 141–150. pavlides, s., & caputo, r., (2004). magnitude versus faults’ surface parameters: quantitative relationships from the aegean region, tectonophysics 380, 159-188. pınar, a., (1998). source inversion of the october 1, 1995 dinar earthquake (ms=6.1): a rupture model with implications for seismotectonics in sw turkey, tectonophysics, 292, 255-266. ritsema, a.r., (1974). the earthquake mechanism of the balkan region, sci. rep. 74-4, pp. 1-36, r. neth. meteorol. inst., de bilt. rousseeuw, r.j., & leroy, a.m., (1987). robust regression and outlier detection, wiley, new york 329 pp. shi, m., & lukas, m.a., (2005). sensitivity analysis of constrained linear l1 regression: perturbations to response and predictor variables, computational statistics and data analysis, 48, 779-802. sinha, s.k., field, c.a., & smith, b., (2003). robust estimation of nonlinear regression with autoregressive errors, statistics and probability letters, 63, 49-59. şaroğlu, f., emre, o., & kuşcu, i., 1992. active fault map of turkey, general directorate of mineral research and exploration, ankara, turkey. şentürk, d., & nguyen, d.v., (2006). estimation in covariate-adjusted regression, computational statistics and data analysis, 50, 32943310. taymaz, t., jackson, j., & mckenzie, d., (1991). active tectonics of the north and central aegean sea, geophys. j. int., 106, 433-490. tibi, r., bock, g., xia, y., baumbach, m., grosser h., milkereit, c., karakisa, s., zünbül s., kind, r., & zschau, j., (2001). rupture processes of the 1999 august 17 izmit and november 12 düzce (turkey) earthquakes, geophys. j. int., 144, f1-f7. turgut, a., (2007). investigation of the relationship between depth of focus and overburden and the surface ruptures related with strike slip faults by mechanical models, msc thesis, hacettepe university, ankara, turkey (in turkish with english abstract). ulusay, r., tuncay, e., sönmez, e., & gökçeoğlu, c., 2004. an attenuation relationship based on turkish strong motion data and iso-acceleration map of turkey, engineering geology, 74, 264-291. weisberg, s., (1985). applied linear regression, 2nd edn, john wiley and sons, new york, 310 pp. wells, d., & coppersmith, j.k., (1994). new empirical relationships among magnitude, rupture length, rupture width, rupture area, and surface displacement, bulletin of the seismological society of america, 84(4), 974-1002. westaway, r., (1990). block rotation in western turkey, 1, observational evidence, j. geophys. res., 95, 19857-19884. yalcinkaya, e., (2005), stochastic finite-fault modeling of ground motions from the june 27, 1998 adana-ceyhan earthquake, earth planets space, 57, 107-115. earth sciences research journal geochemestry earth sci. res. sj. vol. 16, no. 2 (december, 2012): 39 56 investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria bassey edem ephraim department of geology, university of calabar, p. m. b. 1115 calabar – nigeria. phone: +2348035510115. e-mail: basifrem@yahoo.com abstract hitherto unknown metacarbonate deposits constitute parts of the cretaceous mamfe embayment which straddles the border between south-eastern nigeria and western cameroon. the rock is characterised by a high concentration of loi, cao and mgo and depleted content of various insoluble components. assuming all the cao and mgo content of the rock were related to calcite and dolomite phases, these two minerals would account for around 22.5 wt% and 76.3 wt% on average, respectively. among the trace elements investigated, only ba, cs, rb, sr, nb, pb, zr, cd, cu, ni, u, y and zn display concentrations beyond their detection limits. chondrite normalised rare earth element patterns show that the rock under investigation show moderate to strong fractionation of light rare earth elements (lree n) over heavy rare earth elements (hreen) and distinct negative eu anomaly. multivariate statistical treatment and variation plots revealed several geochemical interrelationships, among which are the sio 2 – al2o3 – k2o – tio2 –fe2o3 – ba – nb – rb – zr links which is associated with the rock’s silicate fraction. the carbonate fraction comprises cao, mgo, sr, pb and cu. the overall geochemical signatures support development of the metacarbonate deposit from sedimentary carbonate materials that was deposited in a saline, shallow-marine, low-energy seawater environment. the consistency of the rock’s chemical properties can be attributed to the relative stability experienced during the parent sedimentary materials’ deposition. resumen depósitos metacarbonatadas desconocidos hasta la fecha, parecen constituir partes de la bahía de mamfe, de edad cretácica, ubicada en la frontera entre el sudeste de nigeria y camerún occidental. la roca se caracteriza por una alta concentración de loi, cao y mgo, así como el contenido empobrecido de varios componentes insolubles. suponemos que todo el contenido de cao y mgo de la roca estaban relacionados con las fases calcita y dolomita, minerales que representan aproximadamente el 22,5% en peso y 76,3% en peso promedio, respectivamente. entre los elementos traza investigados, sólo las concentraciones de ba, cs, rb, sr, nb, pb, zr, cd, cu, ni, u, y y zn se encuentran por encima de los límites de detección. elementos de tierras raras normalizadops con patrones de condrita muestran que la roca bajo investigación presenta un moderado a fuerte fraccionamiento de elementos de tierras raras ligeros (lreen) sobre los elementos de tierras raras pesados (hreen) y una anomalía negativa distintiva de eu. el tratamiento estadístico multivariado y las proyecciones variacionales revelaron varias interrelaciones geoquímicos, entre ellos el sio 2 al2o3 k2o tio2 fe2o3 ba nb rb zr, asociados con la fracción de silicato de la roca. la fracción carbonatada comprende cao, mgo, sr, pb y cu. las firmas geoquímicas sugieren el desarrollo del depósito a partir de materiales sedimentarios carbonatados que se depositaron en una solución salina, de poca profundidad marina, y en un medio con agua de mar de baja energía. la consistencia de las propiedades químicas de la roca puede ser atribuida a la estabilidad relativa, experimentado durante la deposición de los materiales sedimentarias parentales. palabras claves: geoquímica, ambiente de deposito, bahía de mamfe. keywords: metacarbonate, geochemistry, depositional environment, mamfe embayment. record manuscript received: 16/01/2012 accepted for publications: 06/05/2012 bassey edem ephraim40 introduction nigeria lies within an area of the ancient african shield, in-between the archaean to early proterozoic west african craton and the congogabon craton (figure 1). the south-eastern region is characterised by several megastructural features, notably the calabar flank, the mamfe embayment, the anambra basin, the afikpo syncline, the abakaliki anticlinorium, the niger delta, the oban massif and the obudu plateau (figure 2). megastructural features, such as the mamfe embayment, straddle the border between south-eastern nigeria and western cameroon (figure 3). while the cameroon sector of the basin has benefited from diverse and repetitive studies (for instance, collignon, 1968; dumort 1968; eben, 1984; eyong, 2003; hell et al., 2000; kande, 2000; kangkolo, 2002; kangkolo and ojo, 1995; ndougsa-mbarga 2004; ndougsa-mbarga et al., 2004, 2007; ndougsa-mbarga and ntep–gweth, 2005; ngando et al., 2004; njieatih, 1997; nouayou, 2005; tabod, 2008; tokam et al., 2010), the nigerian segment is yet to be investigated in sufficient detail. the few available works on the nigerian segment of the mamfe basin, to the best of my knowledge, include those of fairhead and okereke (1987, 1988), fairhead (1991), olade (1975), petters (1987 and 2004) and reyment (1965). the paucity of geological information on the nigerian segment of the embayment has previously led to over-generalisation, some of which may subsequently be proven to be incorrect. a recent survey by the nigerian geological survey agency (ngsa) revealed mappable deposits of metacarbonate rock, which had been hitherto unknown in the south-eastern nigerian sector of the mamfe embayment. this discovery calls for re-examination of earlier paleo-environmental interpretations of the mamfe basin. of particular concern is an earlier viewpoint by petters et al. (1987) that apart from brine seepages which are common, there is no substantial evidence of marine influence during the deposition of the mamfe formation’s cretaceous fluviatile sequences. the present study which is focused on the investigation of the geochemical features of the newly-revealed metacarbonate deposits is intended to provide invaluable insight into the nature and processes which can be reasonably associated with the conditions of formation of the carbonate deposits. this work represents an effort towards revisitfigure 1. a generalised geotectonic map of africa, showing the location of nsofang (modified from affaton et al., 1991). –25˚e –25˚e –25˚n –25˚n 0˚n 0˚n 25˚n 25˚n 0˚e 0˚e 25˚e 25˚e investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 41 ing issues related to paleo-environmental interpretations for the basin. the present study also constitutes an important approach towards the in-depth exploration of the hitherto sparsely-known geology of the nigerian sector of the mamfe embayment. geological background the mamfe embayment which is located roughly between latitudes 5° n and 6° n and longitudes 8°45’ e and 10° e is a coastal sedimentary basin that straddles the federal republic of nigeria and the republic of cameroon (figure 3). despite the fact that the sedimentary infill of the aptian to albian (hell et al., 2000; ndougsa et al., 2004) mamfe basin is largely covered by dense forestation, the type locality on the bank of the cross river at mamfe town in the republic of cameroun reveals thickly folded and faulted series of massive arkosic sandstone and grit having intercalations of marl, arenaceous limestone and shale (reyment, 1965). metacarbonate deposits, which constitute the focus of the present study, are now revealed to be parts of the nigerian sector of the mamfe embayment (figure 3). exposure of mostly the clastic infillings in the embayment show fining upward cycles, which is characteristic of fluvial channel fill, with point bar deposits and over bank siltstone/mudrock. a sedimentary deposit comprising sandstone, mudstone, shale, limestone, microconglomerates and polygenic conglomerates having about 2,000 m thickness has been reported in the lower benue trough section (olade, 1975). these deposits narrow towards the east until it disappear under tertiary and recent rocks of the cameroon volcanic axis. thus, the cameroon figure 2. a geological sketch map of south-eastern nigeria, showing the various megastructural features characterising the region (modified from ofomata, 1973). figure 3. a geological map of south-eastern nigeria and western cameroon, showing the location of nsofang marble within the mamfe embayment of south-eastern nigeria bassey edem ephraim42 volcanic axis, comprising deposits such as basalt, trachyte and rhyolite, uncomformably overlies the mamfe formation (fitton, 1980). metamorphism has so far not been reported in the nigerian sector of the mamfe embayment, but the close association of sedimentary deposits with postdated volcanic rocks appear to provide a good setting for contact metamorphism in the region. field and petrographic aspects observation of the metacarbonate deposits (generally consisting of low-lying deposits having several slightly elevated portions) showed that they are distributed within low to very low grade metamorphosed and un-metamorphosed sedimentary sequences in very close association with the basement and volcanic rocks units. evidence of this association is provided by the close proximity of the units (figure 3) and the ubiquitous basement and volcanic rock fragments abounding in the vicinity of the deposits. generally, exposures of the metacarbonate have steep dips of about 26o–40o and consistent ne–sw strike orientation. joints and fractures having variable orientation are also common features of these outcrops. millimetre–scale laminations, small centimetre-sized voids/ vugs and caves of up to tens of centimetres wide and 1 to 2 metres high are also predominantly conspicuously displayed on the outcrops (figure 4). the fine laminations ran parallel to sub-parallel to the bedding in most cases (figure 4a and 4b) and the voids are irregular but occasionally occluded by mostly laminated buff coloured material (figure 4c). the caves are likewise irregular in shape, usually having solution sculpted walls (figure 4d). the metacarbonate rocks are dense and somehow uniform, having rugged surfaces that possibly indicate the relevance of earlier organic activity and/or weathering in the rocks’ evolutionary history. hand specimen investigation showed that the rocks are characterised by a fairly homogeneous texture, uniform hardness and good resistance to abrasion. also displayed are whitish or greyish colour, polymodal grainsized distribution pattern of mostly fine to medium grains with occasional porphyroblasts and distinct linear and planar fabric, highlighted by the laminar structures. the microscopic features (as observed by optical microscopy) show a predominantly carbonate mineralogy and the textural framework could best be described as heteroblastic. medium grained (~ 1.0–2.0 mm) carbonate minerals are frequently surrounded by finer (<1.0 mm) grain crystals of similar mineral. the boundaries of the medium crystals are often sutured to embayed and the grains are interlocked. unfortunately, the main carbonate minerals could not be differentiated during optical microscope examinations because the thin sections investigated were not stained with alizarin or any other relevant chemical. also, much of the silicate mineralogy is so fine-grained that identification by optical means is difficult to impossible, hence the need for using an x-ray diffraction (xrd) technique which is particularly suitable for discerning individual mineral components of the cryptocrystalline phases and for proper differentiation of carbonate minerals into calcite and dolomite. xrd analysis was carried out on powdered samples under the following conditions: cuk ∝1 graphite monochromatic radiation (1.54059 å) at 30 kv, 30 ma energy. table 1 gives the results of semi-quantitative xrd studies of representative samples of the metacarbonate rocks. a cursory appraisal of the results (table 1) indicates that the metacarbonates are (a) (b) (c) (d) figure 4. nsofang metacarbonates’ outcrop features. (a) distinct millimeter–scale laminations running parallel to the bedding. note cavities that are outlined with the laminates. (b) lamination comparable to algal laminates normally produced on tidal flats by colonies of blue green algae. (c) small centimeter-sized voids/vugs. (d) caves developed within the metacarbonates. investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 43 predominantly dolomitic with calcite as subordinate, while quartz, talc, phlogopite and probably muscovite constitute the accessory phases. geochemistry sampling and analytical procedure several unweathered rock samples weighing 1 to 2 kg were collected during fieldwork and traversing of various metacarbonate rock outcrops in nsofang and its environs in the ikom area of south-eastern nigeria. systematic sampling was hampered by the rocks’ irregular exposure and the thick forestation cover. the samples so collected were cleaned to remove evident allogenic material and/or observed weathered phases in situ before being transported to the laboratory for further investigation. eleven representative metacarbonate rock samples were used for the geochemical analysis which was performed at the acme analytical laboratories in vancouver bc, canada. the geochemical analysis involved measuring major, trace and rare earth element abundance as well as determining the total carbon, sulphur and loss on ignition. prior to the geochemical analysis, about 1 kg of representative rock sample was broken to thumbnail-sized pieces with a hardened-steel hammer. these pieces were crushed and pulverised to particle size as fine as –60 mesh with the aid of a “jawcrusher”. the samples were powdered in an agate mortar to –200 mesh after coning and quartering and thoroughly homogenised. every possible precaution was taken to minimise cross-contamination between samples, including cleaning all the crushing, grinding and homogenisation equipment with a brush, compressed-air, distilled water and acetone to remove possible remains from previously crushed samples. sample preparation and treatment was carried out at the thin–section workshop of department of geology, university of calabar, calabar, nigeria. loss on ignition (loi) was determined by igniting 400 mg of each sample split at 1,000°c and then measuring the weight loss. total carbon and total sulphur concentration was determined with the aid of a leco carbon–sulphur analyser, after sample ignition at >800oc. two instrumentation techniques were used for whole–rock geochemical analysis, namely inductively-coupled plasma-emission spectrophotometry (icpes) and inductively-coupled plasma-mass spectrophotometry (icp-ms). the lithium metaborate-tetraborate fusion digestion technique was found most appropriate for these instrumentation methods. each sample solution was analysed in duplicate in each analytical run and reproducibility was found to be within ± 2%. the detection limit for all the major and minor element oxides was 0.01%, the only exceptions being fe 2o3 and k2o (0.04%. detection limits). the trace elements’ detection limit came within the 0.01 to 1 ppm range. figure 5. comparison of the calcium and magnesium contents of the nsofang marble with the ratio characteristics of stoichiometric dolomite (modified from johnson, 2010). sample # dolomite calcite quartz talc phlogopite muscovite l 1 d sd ftr tr tr ftr l 2 cd cd ac ftr ftr ftr l 3 d sd tr ftr ftr ftr d = dominant (> 50%) cd = codominant (subequal abundance of major components) sd = subdominant (20%-50%) ac = accessory (5%-20%) tr = trace « 5%) ftr = faint trace «1%) l 1 = outcrop location 1, where l 11 , l 12 , l 13 geochemical samples were also collected l 2 = outcrop location 2, where l 21 , l 22 , l 23 geochemical samples were also collected l 3 = outcrop location 3, where l 31 , l 32 , l 41 , l 42 , l 43 geochemical samples were also collected table 1. semi-quantitative x-ray diffraction data of representative samples of metacarbonate rocks of nsofang, ikom area of southeastern nigeria. line depicting stoichiometric dolomite geochemical data and interpretation major element oxides and relevant data the concentrations of major element oxides and other related chemical data of the metacarbonate rocks are presented in table 2. a cursory appraisal of the data (table 2) revealed that loi, cao and mgo content frequently constituted more than 95 wt% of the rock composition and this corroborate mineralogical observations (table 1) that the carbonate phases are the predominating phases in the rock. the high loi values (table 2) possibly reflect the low silica composition of the rocks (table 1). qadhi (2008) observed that increase of silica and other silicate constituents in marble reduces the loi value. apart from loi, cao constitute the dominant constituent, with concentrations that ranges from 29.85 to 51.67 wt% (35.79 wt% mean value). this is closely followed by mgo which has concentration varying from 3.88 to 21.25 wt% and 16.70 wt% mean value (table 2). assuming that all cao and mgo content is related to calcite and dolomite phases, these two minerals would account for around 22.5 wt% and 76.3 wt % on average (table 2). however, the true picture may be slightly different as the non–correspondence of the metacarbonate rocks’ sample plots with the line depicting stoichiometric bassey edem ephraim44 dolomite on the ca cf mg biplot (figure 5), similar to that of johnson et al., (2010), is an indication that dolomite may not be the dominant host mineral for cao and mgo. it is possible that some mgo are also admixed in the calcite’s structural lattices, as observed in the jabal farasan marble of central-western saudi arabia (qadhi, 2008). both cao and mgo are also possibly bound in the structure of the small and probably insignificant silicate phases, which is represented by quartz, talc, phlogopite and muscovite that constitute parts of the rock’s modal mineralogy (table 1). the insoluble residues, notably, sio 2 (0.82–6.21 wt%), al2o3 (0.003–1.01 wt%) and k2o (0.01–0.49 wt%) have considerably low abundance. ignoring sample l11, fe2o3 is frequently less than 0.08 wt% while tio2, mno and na2o concentration are negligible. both k2o and p2o5 have 0.07 wt% mean concentration (table 2). the relatively lower abundance of fe, mn and p in the samples are probably the reflection of low detrital and organic effects relative to the inorganic chemical carbonate precipitate (tucker, 1983). the total carbon values are considerably high (11.31–12.97 wt%, av. 12.28 wt%) as expected for carbonate-bearing rocks, while the same could not be said for total sulphur concentration which are generally below the 0.02 wt% detection limit. trace element composition the trace elements geochemical data of the metacarbonate rocks are presented in table 3. as shown in table 3, only ba, cs, rb, sr, nb, pb, zr, cd, cu, ni, u, y and zn have concentrations beyond their detection limits since many trace elementsconcentrations were below the relevant detection limits (table 4). the rock’s trace element concentrations are not as low as expected and sr and ba values are highly variable (table 3), possibly suggesting a complex distribution of the element (georgieva, 2009). light ion lithophile element (lile) concentration, notably ba (55 ppm average), sr (145.1 ppm average) and probably rb (2.7 ppm average) are considered moderate. sr concentration particularly appears to be approtable 2. concentrations of the major element oxides and related chemical data of metacarbonate rocks of nsofang, ikom area of southeastern nigeria. l 11 l 12 l 13 l 21 l 22 l 23 l 31 l 32 l 41 l 42 l 43 major elements oxide, total carbon, total sulphurs and loi compositions (%) sio 2 6.21 4.08 3.26 1.08 2.74 0.82 1.6 1.43 1.03 1.37 1.82 tio 2 0.06 <0.01 <0.01 0.01 <0.01 0.01 <0.01 0.01 <0.01 <0.01 <0.01 al 2 o 3 1.01 0.03 0.03 0.24 0.15 0.15 0.3 0.25 0.14 0.06 0.06 fe 2 o 3 0.36 <0.04 <0.04 0.08 0.08 0.08 0.08 <0.04 <0.04 <0.04 <0.04 mno 0.01 0.01 <0.01 0.01 0.02 0.02 0.02 <0.01 0.01 <0.01 <0.01 mgo 20.42 21.25 20.56 12.6 21.09 20.14 18.43 11.43 3.88 16.35 17.52 cao 30.26 30.1 30.94 41.09 29.85 31.67 33.78 41.88 51.67 36.94 35.55 na 2 o <0.01 <0.01 <0.01 0.01 0.01 0.01 0.04 0.02 <0.01 <0.01 <0.01 k 2 o 0.49 <0.01 <0.01 0.03 0.06 0.06 0.03 0.02 0.02 0.03 0.03 p 2 o 5 0.05 0.05 0.06 0.02 0.15 0.14 0.14 0.13 0.02 0.04 0.02 loi 40.8 44.1 44.8 44.6 45.4 46.5 45.2 44.5 43.1 44.9 44.6 total 99.68 99.69 99.73 99.77 99.56 99.6 99.63 99.72 99.93 99.76 99.67 total carbon 11.31 12.16 12.17 12.46 12.18 12.97 12.37 12.61 12.14 12.45 12.21 total sulphur <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 <0.02 elemental abundance (ppm) si 29,028 19,072 15,239 5,048 12,808 3,833 7,479 6,684 4,815 6,404 8,507 ti 360 <60 <60 60 <60 60 <60 60 <60 <60 <60 al 5,345 159 159 1,270 794 794 1,588 1,323 741 318 318 fe 2,518 <280 <280 560 560 560 560 <280 <280 <280 <280 mn 77 77 <77 77 155 155 155 <77 77 <77 <77 mg 123,153 128,159 123,997 75,991 127,194 121464 111,151 68,934 23,400 98,607 105,663 ca 216,268 215,125 221,128 293,670 213,338 226,345 241,426 299,316 369,285 264,010 254,076 na <74 <74 <74 74 74 74 297 148 <74 <74 <74 k 4,068 <83 <83 249 498 498 249 166 166 249 249 p 218 218 262 87 655 611 611 567 87 175 87 computed abundance of carbonate minerals calcite 3 1 4 42 1 7 15 46 83 25 20 dolomite 93 97 94 58 96 92 84 52 18 75 80 investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 45 priate, given the fact that sr content of recent carbonates are expected to range from 30 ppm to 200 ppm (shearman and shirmohammadi, 1969). among the high-charged cations, zr has concentration ranging from 4.1 to 35.7 ppm (11.5 ppm mean value), nb concentration ranging from 0.1 to 1.5 ppm (0.3 ppm mean value) and u content varying from 0.3 and 3.0 ppm. y concentration ranged from 0.6 to 5.7 ppm; zn (5.6 ppm average.), cu (2.0 average) and ni (1.6 ppm average) display moderate concentration, while cd (0.3 ppm average), pb (0.6 ppm average) and cs concentration (0.6 ppm average) appear low. figure 6 illustrates the composition of the trace elements of the metacarbonate rock normalised to the average upper continental crust of taylor and mclennan (1981). rare earth element geochemistry table 5 gives the rare earth elements (la, ce, pr, nd, sm, eu, gd, tb, dy, ho, er, tm, yb and lu) abundance in the rock , together with other relevant data. the ree values shown in table 5 are quite low (∑ree = 1.33–6.12 ppm, excluding sample l 11 having an abnormal 15.41 ppm concentration) and the concen tration of most components from locations l22, l23, l42 and l43 are below detection limits (table 5). the tabulated data (table 5) clearly show that lree components dominate over hree, and that total lree and hree concentrations decrease from 12.53 to 1.1 ppm and from 2.88 to 0.27 ppm, respectively (table 5). identification of ree fractionation in the metacarbonate rocks was carried out by normalising (haskin et al, 1968) the concentration of the rare earth elements to average chondritic meteorites, and the result for samples having concentrations beyond detection limits is presented as ree pattern (figure 7). a cursory appraisal of the chondrite–normalised rare earth element (ree n) plot show that all samples exhibit similar reen patterns, moderate to strong fractionation of light rare earth elements (lree) over heavy ree (hree) (lan/ybn = 0.98 – 9.35) and distinct negative eu anomaly, eu/eu* ranging from 0.48 to 0.60. it is also obvious that la – nd – sm – eu appear to define an inclined straight line and the heavy rare earth elements (hree) exhibit uncoordinated zigzags patterns in the gd – dy – ho – er – yb – lu spans (figure 7). table 3. trace element composition and relevant ratios of metacarbonate rocks of nsofang, ikom area of southeastern nigeria. table 4. measured trace elements occurring below detection limits in metacarbonate rocks of nsofang, ikom area of southeastern nigeria l 11 l 12 l 13 l 21 l 22 l 23 l 31 l 32 l 41 l 42 l 43 ba 190 22 6 33 78 70 37 37 19 57 52 cs 3.9 0.2 <0.1 0.2 0.4 0.3 0.2 0.2 0.1 0.3 0.4 rb 17.8 0.5 0.3 0.8 2.7 2.2 0.9 0.7 0.6 1.2 1.5 sr 59.1 53.8 60.1 167.7 97.4 95.6 100.1 181.4 552.3 115.8 112.6 nb 1.5 0.2 0.1 0.3 0.2 0.3 0.2 0.2 0.1 <0.1 <0.1 pb 0.9 0.4 0.5 0.7 0.6 0.5 0.7 0.8 1.1 0.3 0.5 zr 35.7 4.1 6.7 7.7 14.9 15.1 7.8 12.4 3.8 5.4 12.8 cd 0.2 0.2 0.7 0.3 0.2 0.1 0.3 0.3 0.1 0.3 0.3 cu 0.9 0.7 1.6 0.6 1 0.6 2.7 1.5 12.1 0.2 0.2 ni 0.3 <0.1 0.8 3.9 1.2 1.7 1.3 3.5 1.6 1.1 1 u 0.5 1.3 3 0.8 0.3 0.3 0.6 0.6 2.3 2.3 1.8 y 5.7 2.7 2 1.8 0.6 0.7 1.4 2 1 0.7 0.6 zn 9 5 6 8 4 5 7 6 3 5 4 mn/sr 1.30 1.43 1.28 0.46 1.59 1.62 1.55 0.42 0.14 0.66 0.68 mg/ca 0.57 0.60 0.56 0.26 0.60 0.54 0.46 0.23 0.06 0.37 0.42 detection limits (in ppm except otherwise stated) comments ag <0.1 no sample revealed measureable amounts be <1 no sample revealed measureable amounts bi <0.1 no sample revealed measureable amounts co <0.2 2 samples showed traces of 0.3 – 0.4 ppm ga <0.5 1 sample showed traces of 1.4 ppm hf <0.1 2 sample showed traces of 0.2 – 0.8 ppm mo <0.1 3 samples showed traces of 1 – 2 ppm sb <0.1 no sample revealed measureable amounts sc <1 1 sample showed traces of 1ppm se <0.5 1 sample showed traces of 0.6 ppm sn <1 no sample revealed measureable amounts ta <0.1 1 sample showed traces of 0.1ppm th <0.2 4 samples showed traces of 0.2 – 0.6 ppm tl <0.1 no sample revealed measureable amounts v <8 5 samples showed traces of 9 22 ppm w <0.5 1 sample showed traces of 1.6ppm hg <0.01 (ppb) no sample revealed measureable amounts au <0.5 (ppb) no sample revealed measureable amounts bassey edem ephraim46 figure 6. selected trace element concentrations of nsofang metacarbonate rock normalized to the composition of average upper continental crust of taylor and mclennan (1981). l11 l12 l13 l21 l22 l23 l31 l32 l41 l42 l43 la 2.7 0.3 0.6 1 0.7 0.6 0.8 1.1 0.9 0.3 0.2 ce 5.9 1 1.7 2.2 1.2 1.2 1.9 2.6 1.8 0.5 0.5 pr 0.7 0.1 0.16 0.25 0.12 0.13 0.19 0.26 0.18 0.05 0.05 nd 2.6 0.5 0.7 1 0.6 0.7 1 1.1 0.7 <0.3 <0.3 sm 0.63 0.11 0.13 0.23 0.09 0.11 0.14 0.23 0.16 0.05 <0.05 eu 0.1 0.02 0.04 0.04 <0.02 <0.02 0.02 0.05 0.03 <0.02 <0.02 gd 0.66 0.15 0.13 0.18 0.08 0.09 0.12 0.2 0.13 <0.05 <0.05 tb 0.12 0.03 0.03 0.03 0.02 0.01 0.02 0.04 0.02 <0.01 <0.01 dy 0.73 0.21 0.14 0.21 0.08 0.08 0.12 0.22 0.14 0.05 <0.05 ho 0.15 0.05 0.05 0.05 <0.02 <0.02 0.03 0.04 0.02 <0.02 <0.02 er 0.48 0.21 0.16 0.14 0.07 <0.03 0.08 0.09 0.11 0.05 0.05 tm 0.08 0.03 0.02 0.02 0.01 <0.01 0.01 0.02 0.01 <0.01 <0.01 yb 0.49 0.19 0.15 0.14 <0.05 0.06 0.08 0.15 0.06 <0.05 <0.05 lu 0.07 0.03 0.03 0.02 <0.01 <0.01 0.01 0.02 <0.01 <0.01 <0.01 ∑ree 15.41 2.93 4.04 5.51 3.07 3.07 4.52 6.12 4.27 1.47 1.37 ∑lree 12.53 2.01 3.29 4.68 2.71 2.74 4.03 5.29 3.74 1.2 1.1 ∑hree 2.88 0.92 0.75 0.83 0.36 0.33 0.49 0.83 0.53 0.27 0.27 table 5. rare earth element composition of metacarbonate rocks of nsofang, ikom area of southeastern nigeria. investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 47 figure 7. chondrite normalised (haskin, 1968, 71) rare earth element patterns of the nsofang metacarbonate rock. table 6. correlation coefficients of major element components and selected trace elements of metacarbonate rocks of nsofang, ikom area of southeastern nigeria. sio 2 tio 2 al 2 o 3 fe 2 o 3 mno mgo cao na 2 o k 2 o p 2 o 5 loi total c ba nb ni pb rb sr zr sio 2 1 tio 2 0.78 1 al 2 o 3 0.62 0.94 1 fe 2 o 3 0.73 0.98 0.96 1 mno -0.23 -0.19 -0.05 -0.02 1 mgo 0.53 0.23 0.11 0.28 0.38 1 cao -0.57 -0.27 -0.16 -0.32 -0.38 -1.00 1 na 2 o -0.20 -0.13 0.11 -0.06 0.44 0.00 0.00 1 k 2 o 0.75 0.99 0.95 0.99 -0.10 0.26 -0.30 -0.14 1 p 2 o 5 -0.13 -0.15 0.02 -0.04 0.82 0.35 -0.37 0.51 -0.08 1 loi -0.69 -0.82 -0.74 -0.73 0.57 0.19 -0.15 0.18 -0.76 0.50 1 total c -0.85 -0.79 -0.66 -0.72 0.36 -0.15 0.18 0.16 -0.75 0.39 0.88 1 ba 0.62 0.90 0.87 0.93 0.09 0.34 -0.38 -0.15 0.94 0.07 -0.57 -0.59 1 nb 0.74 0.98 0.96 0.99 -0.11 0.26 -0.30 -0.11 0.99 -0.07 -0.75 -0.70 0.91 1 ni -0.63 -0.33 -0.09 -0.27 -0.05 -0.56 0.56 0.13 -0.32 0.10 0.27 0.54 -0.29 -0.24 1 pb 0.05 0.37 0.53 0.38 -0.10 -0.65 0.60 0.16 0.36 -0.05 -0.58 -0.36 0.21 0.37 0.29 1 rb 0.76 0.99 0.94 0.98 -0.09 0.28 -0.32 -0.15 1.00 -0.08 -0.76 -0.75 0.95 0.98 -0.34 0.34 1 sr -0.44 -0.20 -0.13 -0.24 -0.22 -0.92 0.91 -0.08 -0.22 -0.30 -0.21 0.05 -0.30 -0.25 0.30 0.69 -0.23 1 zr 0.63 0.89 0.88 0.91 0.08 0.35 -0.39 -0.12 0.92 0.15 -0.55 -0.56 0.95 0.91 -0.20 0.29 0.93 -0.33 1 bassey edem ephraim48 ponents. the tio2, fe2o3 and k2o concentration appear consistent with those of sio2 and al2o3 (table 5). furthermore, there appear to be no correlation between cao and na2o (table 6, figure 8). examination of variations of trace element in the rock reveals pb, ni, sr and cu enrichment with increasing cao values while ba, nb, rb and zr display weak negative correlation with cao (table 6, figure 9). it is also interesting to observe that most trace elements display positive interrelationships with the major insoluble residue components of the rock (table 6) and prominent among these are the sio 2 – al2o3 – k2o – tio2 –fe2o3 – ba – nb – rb – zr links (table 6, figure 10). in the rotated factor matrix (table 7), loading greater than ± 0.55 are highlighted and considered significant members of each of the 3 factor score groupings. as shown in table 7, factor 1 (comprising significant loadings of si, ti, al, fe, k, ba cs, rb, nb, zr, y and zn) account for 38.59% of total data variance while factor 2 (high loadings of ca, sr, pb and cu) account for 17.15% of total data variance. factor 3 (with high loadings of mn, ba, u and y) account for 12.68% of total data variance. interpretation and discussion evaluation of consistency of primary chemical signatures a number of geochemical parameters have been proposed for testing the degree of preservation of primary chemical signatures in carbonatebearing rocks (kaufman et al., 1993; narbonne et al., 1994; pandit et al, 2003; veizer, 1983). in particular, the mn/sr ratio has been accepted as a definitive indicator for the degree of preservation and/or post-depositional alteration (brand and veizer, 1980; derry et al., 1992; veizer et al., 1989, 1992). the mn/sr ratio of the nsofang marble varies between 0.14 and 1.62 with an average of 1.01 ppm. these values are considered to fall within acceptable limits for carbonate rocks that are well-preserved and unaffected by post-depositional alteration (derry et al., 1992; kaufman et al., 1992, 1993; kaufman and knoll, 1995). moreover, brand (1983) has suggested that trace element signatures can aid in interpretation of depositional environment conditions despite post-digenetic alteration. the chemical signatures of the marble can therefore be taken as representative of the precursor rock. distribution of chemical species the reduction in sio2 content compared to increasing cao concentration (figure 8) agree with the fact that the rock comprises distinct silicate and carbonate fractions, having contrasting variations. the carbonate fraction increases at the expense of the silicate fraction and vice versa. the positive relationship between sio 2 and mgo (table 6, figure 8) most likely confirm earlier observations that mgo constitutes parts of the silicate fraction. the positive relationship existing between sio2 and mgo (table 6, fig. 8) most likely confirm earlier observations that mgo also constitute parts of the silicate fraction. the positive relationship displayed between sio2 and al2o3 (fig. 8), together with the consistency observed in the variation of tio2, fe2o3 and k2o with those of sio2 and al2o3, (table 6) and the weak negative correlation existing between cao and the various insoluble residues (fig. 8) suggests that the non-carbonate or silicate fractions are mainly aluminosilicates. however, where al 2o3 is extremely low (as in sample l12, l13, l42 and l43), clearly the sio2 cannot have been introduced into the original sediment in aluminosilicate phases and was probably therefore either in traces of detrital quartz or in siliceous organisms. among the trace elements investigated, sr, pb and ni appear to be enriched in the carbonate fractions as implied by the consistency that these components display with cao, which is an important member of the carbonate fraction. the enrichment of sr in carbonates is often accounted for by the fact that sr2+ readily substitutes for ca2+ in calcium – bearing strucfactor 1 factor 2 factor 3 log si 0.58 -0.34 -0.53 log ti 0.94 0.01 -0.18 log al 0.75 0.51 0.36 log fe 0.95 0.08 0.17 log mn 0.03 0.01 0.78 log mg -0.09 0.00 0.00 log ca 0.00 0.71 -0.06 log na 0.00 0.41 0.18 log k 0.88 -0.04 0.36 log p 0.15 0.03 0.48 log ba 0.69 -0.21 0.57 log cs 0.91 -0.30 0.14 log rb 0.86 -0.20 0.38 log sr -0.42 0.67 0.22 log nb 0.95 0.11 0.08 log pb 0.36 0.82 0.05 log zr 0.79 -0.18 0.39 log cd -0.10 -0.25 -0.53 log cu -0.09 0.86 -0.10 log ni -0.31 0.47 0.48 log u -0.52 -0.10 -0.70 log y 0.64 0.27 -0.65 log zn 0.65 0.08 -0.25 eigenvalue 8.88 3.95 2.92 percent of total variance 38.59 17.15 12.68 cumulative percentage 38.59 55.75 68.43 table 7. varimax rotated r – mode factor analysis of bulk rock composition of metacarbonate rocks of nsofang, ikom area of southeastern nigeria. the highlighted loadings are greater than ±5.5 geochemical interrelationships the interrelationships between the major element oxides and trace elements of the metacarbonate rock have been investigated and the results presented as correlation coefficients (table 6) and co-variation plots (figures 8–11). also, following the procedure adopted by veizer et al (1992), r–mode factor analysis have also been employed to reveal any communality of chemical elements and, indeed, the most likely identity of the host minerals. the rotated factor matrix is presented in table 7, together with eigenvalues, total variance percentage and cumulative percentage. examination of changes of the major elements oxides with respect to cao concentration reveal that sio 2 and mgo are significantly decreased with increasing cao content in the rock (table 6, figure 8). cao composition also display an expected weak negative correlation with most of the insoluble residues, notably tio 2, al2o3, fe2o3, mno, k2o and p2o5 (table 6, figure 8), and both sio2 cf mgo and sio2 cf al2o3 cross-plots (figure 8) show significant positive correlation between the involved cominvestigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 49 ture. however, in the nsofang metacarbonates, both ca and sr display very strong positive correlation (figure 9) which rather suggests that both components are probably bound together in the same calcium – bearing lattice structures. this is consistent with the position of dissanayake (1981) that sr being the main trace elements in limestone and dolomites is often presume to be located in the lattices of carbonate minerals. the observed lack of significant positive correlation between sr and na (table 6; figure 11) is an indication that na 2o is not controlled by the carbonate phases, and consequently does not form parts of the lattices of the ca – mg carbonates nor occur as inclusions in the carbonate phases (fritz and katz, 1972; land and hoops, 1973). the lack of strong relationship between sr and na, together with the absence of significant positive correlation between na/ca versus mg/ca or sr/ca (figure 11), also precludes the consideration of an evaporitic hypersaline (sass and katz, 1982) conditions as existing during developments of the protoliths of the metacarbonate. other trace elements, notably ba, nb and rb, tend to concentrate in the silicate fractions as inferred from the weak negative correlation they displays with cao (figure 9) and the positive interrelationships they exude with major insoluble residue components of the rock (figure 10). arising from the marked correlation exhibited between zr and al 2o3 (figure 10), it thus seems reasonable to suggest that much of the rock’s zr composition was probably not introduced as detrital zircon but also formed parts of the rock’s silicate fraction. the strong positive interrelationships amongst the various trace elements and major insoluble residues (figure 10) underline the relevance figure 8. the degree of correspondence amongst the major element oxides. bassey edem ephraim50 of the minor aluminosilicate phase in controlling the distribution of both insoluble components and most trace elements. accordingly, the positive correlation among al2o3 and tio2, fe2o3, k2o, ba, nb, rb and zr (figure 10) is taken to imply that all these components are contained in different proportions of aluminosilicate phases, as appropriately reflected in the factor 1 loadings (table 7). the strong positive interrelationship amongst ba, rb and k 2o (table 6, figure 10) underline the relevance of k-bearing minerals in the aluminosilicate phases. it has long been known that ba and rb are especially retained in k-bearing minerals, and common k – bearing minerals are mostly the feldspars. however, the non–linear interrelationships involving na, pb, sr and al 2o3 (table 6, figure 11) and na2o and cao (figure 8) corroborate petrographic observations (table 1) that feldspar does not constitute a significant phase in the modal mineralogy of the rock, rather quartz, talc, and phlogopite or muscovite are the main aluminosilicate phases in the rock. the high loadings of ca, sr, pb and cu of factor 2 (table 7) most likely reflect the composition of the carbonate fraction, while the high loadings of mn, ba, u and y recorded for factor 3 could not be readily interpreted. the exclusion of mgo as a significant component in all three factor score groupings could be explained by considering diverse sourcing of the mgo. conditions of formation of metacarbonate deposit in an effort to gain insight into the pre-metamorphic conditions of the parent materials of the metacarbonate deposits, the chemical data of the rock were plotted on discrimination diagrams like garrels & mckfigure 9. the degree of correspondence of selected trace elements with cao composition of the nsofang metacarbonate rock. investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 51 enzie’s na2o/al2o3 cf k2o/al2o3 diagram (1971) and leyreloup et al’s mgo–cao–al2o3 discrimination diagram (1977). the samples dominantly plotted in the sedimentary field on the na2o/al2o3 cf k2o/ al2o3 diagram (figure 12), thereby confirming that the rock is largely of sedimentary origin. on the mgo–cao–al2o3 discrimination diagram (following leyreloup et al., 1977) (figure 13), the metacarbonate rock samples plotted outside the magmatic funnel, thereby supporting the sedimentary antecedents of the rock. the sedimentary origin of the precursor rocks is also clearly observed in the bedded nature of the metacarbonate rocks (figure 4) and the fact that sr and total ree abundance of the rock rarely exceed 200 ppm and 25 ppm, respectively, thereby agreeing with observations that sedimentary carbonates are usually characterised by low sr and total ree content (nothdurft et al., 2004; veizer et al., 1992). additionally, interrelationships observed earlier between major insoluble residue components and selected trace elements (notably the sio 2 – al2o3 – k2o – tio2 – fe2o3 – ba – nb – rb – zr links) (figure 10) points to the close association of the parent carbonate rocks with aluminosilicate phases, which is a diagnostic sedimentary feature. having confirmed the sedimentary nature of the protoliths, it become necessary to investigate the nature and characteristics of the depositional environment of the parent sedimentary carbonate, and for this, the trace element composition of the rock, especially the rare earth elements are involved. according to tanaka and kawabe (2006), unaltered marine sedimentary carbonates are expected to preserve the chemical characteristics (i.e. ree features) of the medium in which the carbonates precipitated. however, care must be taken when making such an interpretation figure 10. cross–plots showing the varying degrees of positive inter-relationships displayed between various trace elements and the insoluble components of the nsofang metacarbonate rock bassey edem ephraim52 in outlining the ree pattern differences between shallow coastal waters and deep waters, tanaka et al., (1990) observed that data points for la, nd, sm, and eu for coastal waters fall on a nearly straight line, and their ree patterns are characterised by marked discontinuity between eu and gd. similar observations have been made on the ree patterns of the nsofang metacarbonate rock (figure 7), and the indication is that the protoliths of the rock was deposited in a shallow, near-shore marine environment. besides, the observed uncoordinated zigzag patterns in the gd – dy – ho – er – yb – lu spans of the heavy rare earth elements (hree) of the nsofang metacarbonate (figure 7) require further investigation, as such features can also be associated with a surface or shallow water environment (masuda and ikeuche, 1979; tanaka et al., 1990). interpretation of a surface or shallow marine depositional environment for the parent carbonate material of the nsofang rocks is in agreement with the reported negligible na 2o, fe2o3, mno (table 3) and sr concentration (table 4) (clarke, 1911; frank 1975; turekiah and wedepohl, 1961). davou and ashano (2009) considered low concentration of na2o to imply a shallow, near-shore, low energy “clean water” marine environment of deposition for the parent carbonate materials of the marble deposit occurring east of the federal capital territory (fct) of nigeria. moreover, most sedimentary carbonates in nigeria’s cretaceous basins have been interpreted as product of shallow marine environment of deposition. enu and adegoke (1988) used microfacies and microfaunal evidence to interpret the ewekoro limestone occurring within the eastern dahomey basin of south-western nigeria as products of deposition in shallow marine environment. adekeye and akande (2002) integrated lithological, paleontological and sedimentary data to reveal that the sedimentary sequence of the yandev area in the benue trough (consisting of a limestone–shale sequence) were deposited in a shallow marine shelf lagoonal environment. figure 11. cross–plots showing vague and inconsistent inter-relationships displayed among various chemical components of nsofang metacarbonate rock figure 12. composition of the nsofang metacarbonate rock on the na2o/al2o3 cf k2o/al2o3 discrimination diagram of garrels and mackenzie (1971). because carbonate ree content could also be influenced by non-carbonate materials, such as fine-grained siliciclastic material in excess of 5% (banner et al., 1988). fortunately, in the case of the nsofang metacarbonate rocks, petrographic and geochemical observations have shown that noncarbonate phases are negligible, and this precludes the consideration of an additional influence on the ree chemistry of the rock by non-carbonates. investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 53 among the sedimentary cycles delineated for the mfamosong limestone of the calabar flank, is an initial transgressive cycle that occurred and resulted in a normal shallow marine environment (fayose, 1978). akande et al., (1988) concluded that shallow water epicontinental seas, flanked by fault–bounded basement uplands seems to have been the primary environments of deposition of carbonate materials in nigeria. following ikhane et al., (2009) the low al 2o3 concentration measured in nsofang rock is interpreted as being indicative of a low energy environment. when the relevant chemical data of the nsofang metacarbonate deposit are projected on the caco 3 – mgco3 – others (insoluble component) diagram (figure 14) adapted from harker (1939), carr and rooney (1985) and davou and ashano (2009), indication is that the metacarbonate rock is dolomitic. the occurrence of lamination, vug and cavernfigure 13. composition of the nsofang metacarbonate rock on the mgo – cao – al2o3 diagram by leyreloup et al., (1977). figure 14. composition of the nsofang metacarbonate rock on the caco3 – mgco3 – others (insoluble components) classification diagram, adapted from harker (1939), carr and rooney (1985) and davou and ashano (2009). ous structures as part of the structural features (figure 4) of the nsofang metacarbonates possibly reflects the relevance of biogenic activity at some point in the evolutionary history of the rock. the lamination probably represents relic organic imprints similar to the type produced on tidal flat environments by colonies of blue green algae, while the irregular shape of the caves and small cavities with their sculptured walls suggest formation through dissolution of lithified carbonate material. there is therefore every indication that the metacarbonate rocks most likely began as calcitic material before conversion to dolomitic prior to metamorphism. dolomitisation of sedimentary carbonate material is a widely-reported process for many nigerian carbonates and metacarbonates (e.g. davou and ashano, 2009; odigi and amajor, 2008; oti, 1983; petters, 1978). the prior demonstrated existence of precursor carbonate sedimentary material most likely provided the required setting for extensive dolomitisation of the precursor rock of the nsofang metacarbonate. however, as already known (e.g. deelman, 2008; purser et al., 1994; welch, 2001), the condition of formation of dolomite in the natural environment is frequently different from synthesis of ordered dolomite, accomplished at temperature of over 100°c and 20 atm pressure in the laboratory. the source of high mgo is of considerable importance in the conversion of the sedimentary carbonate to dolomitic rock. according to akcay et al., (2003), possible mgo sources for dolomitic carbonate–bearing rocks include pore-water, seawater and clay minerals. pore-water trapped in limestone may promote dolomitisation (kupecz and land, 1991) because these waters are typically ca-enriched and their molar ca/mg ratio increases with increasing temperature (kupecz and land, 1991). however, a pore-water mgo source for the nsofang metacarbonates is not likely because pore-water is usually characterised by moderate to high salinity (10–25 wt% nacl equiv.; gomez-fernandez et al., 2000; yao and domicco, 1997) and na2o and k 2o concentration in carbonate tends to decrease with increasing salinity but appears to increase with depth (clarke, 1911). the metacarbonate rocks of nsofang are characterised by low na 2o values and shallow depth of deposition, which is at variance with considbassey edem ephraim54 eration of the pore-water source. seawater, another major source of mgo for sedimentary dolomitic rocks, contains significant mg concentration (~3.7%) (taylor and sibley, 1986; tucker and wright, 1992). land (1985) concluded that seawater is the only widely-available fluid having sufficient magnesium to cause massive dolomitisation, and diagenetic dolomitisation of calcite-rich carbonates requires large volumes of water to supply the required mg (land, 1992). the low na 2o and k2o concentration in the metacarbonate, combined with the moderate sr content, support the seawater source of mgo and, with seawater as the major source of mgo, the low pb concentration in the rock (table 6) can be readily explained. wedepohl et al., (1974) advanced the view that calcite and dolomite cannot incorporate appreciable concentrations of pb since sea and interstitial water usually contains very little pb. clay minerals represent another important major source of mgo, which is frequently regarded as a source of mgo for dolomitisation, especially for late-stage dolomite formations (morteani et al., 1982; kupecz and land, 1991; warren, 2000). given that nsofang metacarbonate contains a quantity of quartz, talc, phlogopite and probably muscovite (table 1), it is highly probable that transformation of these materials may release additional (though negligible) mgo for the carbonates. detailed analysis of the textural and structural features preserved in the nsofang metacarbonate rocks reveal that the precursor sedimentary materials were affected by several widespread, laminae-parallel, diagenetic processes, most likely controlled by fine-scale layer-parallel differences in initial porosity and permeability. xu (2011) observed that fenestral fabrics such as lamination and open space structures like vug and cavernous structures are usually formed in shallow, near-coast supratidal and upper intertidal carbonate environments (flügel, 2004). the fenestral fabrics, formed in tidal flats, are generally attributable to diagenetic processes, associated with the degradation of microbial mats (hofmann et al., 1980). summary and conclusions the hitherto unknown metacarbonate deposits of the cretaceous mamfe embayment comprised distinct carbonate and minor silicate fractions. while the carbonate fraction include ca, mg, sr, pb and cu, the silicate fraction are associated with si, ti, al, fe, k, ba cs, rb, nb, zr, y and zn. geochemical data, integrated with field information, have been used to infer a metasedimentary petrogenetic affiliation for the rock. indication is that the metacarbonate deposits were formed from sedimentary carbonate materials that were deposited in a saline, shallow-marine environment, possibly in a supratidal and/or upper intertidal environments. seawater seemingly constituted a major source of mgo. generally, the consistency in the chemical properties of the rock most likely reflects relative stability during deposition of the initial sediments. future studies, possibly including isotopic and geochronological data, would shed more light on the evolutionary history of the metacarbonate deposit. however, the present study provides a basis for revisiting earlier paleo-environmental interpretations and intense exploration of the geology of the nigerian sector of the mamfe embayment as this could be very rewarding for the country. acknowledgements i would like to acknowledge the support and unparalleled hospitality received from the chiefs and people of nsofang community in present– day etung local government area in cross river state of nigeria. my heartfelt appreciation also goes to my graduate students, charles umagu, columbus edet and festus udumu, for the roles they played in the successful completion of this research. furthermore, i give special thanks to all the people who were involved in sample preparation and analysis stages of the work at both the department of geology, university of calabar, calabar, nigeria and acme analytical laboratories, vancouver bc, canada. last, but not least, is the anonymous peer reviewer whose review greatly added to the value of the final work. references adekeye o. a. and akande s. o. (2002). depositional environments of carbonates of the albian asu river group around yandev, middle benue trough, nigeria. j. min. geol. 38 (2): 91 – 101. affaton p., rahaman m. a., trompette r., and sougy j. (1991). the dahomeyide orogen: tectonothermal evolution and relationships with the volta basin. in: dallmeyer r. d. and lécorché j. p. (eds.) the west african orogen and circum-atlantic correlatives, springer verlag, berlin: pp.107 -122. akande s. o., horn e. and reutel c. (1988). mineralogy, fluid inclusion and genesis of the arufu and akwana pb – zn – f mineralization, middle benue trough. jour. afri. earth sci. 1: 167 – 180. akcay m., ozkan h. m., spiro b., wilson r. and hoskin p. w. o. (2003). geochemistry of a high-t hydrothermal dolostone from the emirli (odemis, western turkey) sb-au deposit. mineralogical magazine, august, 67(4), 671 – 688 banner j. l., hanson g. n. & meyers w. j. (1988). fluid – rock interaction history of regionally extensive dolomites of the burlington – keokuk formation (mississippian): isotopic evidence. in: shukla v. & baker p. a. (eds.), sedimentology and geochemistry of dolostones. spec. publ. soc. econ. paleont. miner. 43: 97 – 113. brand u. (1983). mineralogy and geochemistry of deep sea clay in the atlantic ocean and adjacent seas and ocean. geol. soc. amer. bull. 76: 803 – 832. brand v. and veizer j. (1980). chemical diagenesis of multicomponent carbonate system-1, trace elements. j. geol. petrol. 50: 1219-1250. carr d. d. and rooney l. f. (1985). limestone and dolomite. in: lefond s. y. (ed.). industrial minerals and rocks. 5th edn., american inst. met. and pet. engr inc, new york, pp. 833 – 868 clarke f. n. (1911). the data of geochemistry. 2nd edn., washington government printing press: 782p collignon f (1968). gravimétrie de reconnaissance de la république fédérale du cameroun orstom paris france 35p condie k. c., wilks m., rosen d. m. and zlobim v. l. (1991). geochemistry of metasediments from the precambrian hapschan series, eastern anabar shield, siberia. prec. res. 50: 37 – 47. davou d.d. and ashano e.c. (2009). the chemical characteristics of the marble deposits east of federal capital territory (fct), nigeria. global journal of geol. sciences, 7 (2): 189 – 198. deelman, j. c. (2008). low-temperature formation of dolomite and magnesite: a comprehensive revision. version 2.3, compact disc publications, eindhoven, the netherlands derry l.a., kaufman a.j. and jacobsen s.b. (1992). sedimentary recycling and environmental change in the late proterozoic: evidence from stable and radiogenic isotopes. geochim. cosmochim. acta 56: 1317 – 1329. dissanayake c. b. (1981). the strontium geochemistry of some precambrian carbonate rocks of sri lanka. j. natn. sci. coun. sri lanka 9 (2): 255 – 267. dumort j. c. (1968). notice explicative sur la feuille douala ouest: république fédérale du cameroun, brgm, 69p. eben, m. m. (1984). report of the geological expedition in the gulf of mamfe: archives of the department of mines & geology, ministry of mines & power, cameroon, 10p. enu e. i. and adegoke o. s. (1988). microfacies of shallow marine carbonates (paleocene) in the eastern dahomey basin, southwestern nigeria. j. min geol. 24 (1 & 2): 51 – 56 investigation of the geochemical signatures and conditions of formation of metacarbonate rocks occurring within the mamfe embayment of south-eastern nigeria 55 eyong j. t. (2003). litho-biostratigraphy of the mamfe cretaceous basin, s.w. province of cameroon – west africa. phd thesis, university of leeds., 265p. fairhead j. d. and okereke c. s. (1987). a regional gravity study of the west african rift system in nigeria and cameroon and its tectonic interpretation tectonophysics 213: 459 – 481. fairhead j d and okereke c s (1988) depth to major contrast beneath the west african rift system in nigeria and cameroon based on the spectral analysis of gravity data. jour. of african earth science 7(5-6): 769-777 fairhead j d, okereke c s and nnange j m (1991). crustal structure of the mamfé basin, west africa, based on gravity data. tectonophysics 186: 351-358. fayose e. a. (1978). depositional environments of carbonates calabar flank, southeastern nigeria. j. min geol. 15 (1): 1 – 13. fitton g. (1980). the benue through and the cameroon line: a migrating rift system in west africa. earth planet. sci. lett., 51: 132 – 138. flügel, e., 2004. microfacies of carbonate rocks-analysis, interpretation and application. springer press, germany, 190-203. frank w. (1975). sediment chemische and palolcologische asperkte stablier schewellen. ben sonderforschungsberoich. 48, univ. gottinygen, a: 31 – 40.. fritz p and katz a. (1972). the sodium distribution of dolomite crystals. chemical geology 72: 170 – 194. garrels r. m. and mackenzie f. t. (1971). evolution of metasedimentary rocks. new york: norton and company: 394p. georgieva m, cherneva z., hekimova s. and petrova a. (2009). petrology of marbles from the arda tectonic unit, central rhodope, bulgaria. abstracts of national conference, geosciences 2009. gomez-fernandez f., both r. a., mangas j. and arribas a. (2000). metallogenesis of zn – pb carbonate–hosted mineralization in the southeastern region of the picos de europa (central – northern spain) province: geologic, fluid inclusion, and stable isotope studies. economic geology 95: 19 – 39. harker a. (1939). metamorphism. 2nd edn., motheum and coy. ltd, london, 362p haskin l. a., haskin m. a., frey f. a. and wilderman t. r. (1968). relative and absolute abundances of the rare earths. in: ahrens l. h. (ed.), origin and distribution of the elements (889 – 911). pergamon, oxford. hell j. v., ngako v., béa v., olinga j. b. and eyong j. t. (2000). rapport des travaux sur l’étude de reconnaissance géologique du bassin sédimentaire de mamfé: irgm-nhc (national hydrocarbon corporation) unpublished report, 55 p. hofmann, h.j., pearson, d.a.b. and wilson, b.h., 1980. stromatolites and fenestral fabric in early proterozoic huronian supergroup, ontario. canadian journal of earth sciences 17, 1351-1357. ikhane p. r., folorunso a.f., nton m.e and oluwalaanu j.a. (2009). evaluations of turonian limestone formation exposed at nigercem – quarry, nkalagu, southeastern nigeria: a geochemical approach. the pacific journal of science and technology 10 (2): 763 – 771. johnson c. a., taylor c. d., leventhal j. s., and freitag k. (2010). geochemistry of metasedimentary rocks in the hanging wall of the greens creek massive sulfide deposit and of shales elsewhere on admiralty island. in: taylor c. d and johnson c. a.(eds.). geology, geochemistry, and genesis of the greens creek massive sulfide deposit, admiralty island, southeastern alaska. u. s. geological survey professional paper 1763, pp. 159 182 kande h. l. (2000). etude audiomagnétotellurique de la transition sédimentaire-métamorphique de la bordure orientale du bassin de mamfé, cameroun mémoire de maîtrise université de yaoundé i60p kangkolo r. (2002). aeromagnetic study of the mamfe basalts of southwestern cameroon. journal of the cameroon academy of sciences, 2(3), 173-180. kangkolo r. and ojo s. b. (1995). integration of aeromagnetic data over the mamfe basin of nigeria and cameroon: nigerian journal of physics 7, 53-56.njieatih, 1997; kaufman a. j. and knoll a. h. (1995). neoproterozoic variations in the c isotopic composition of seawater: stratigraphic and biogeochemical implications. precambrian res., v. 73, pp. 27-49. kaufman a. j., jacobsen s.b. and knoll a.h. (1993). the vendian record of sr and c isotopic variations in seawater: implications for tectonics and paleoclimate. earth planet. sci. lett., v. 120, pp. 409-430. kaufman a. j., knoll a. h. and awramik s. m. (1992). biostratigraphic and chemostratigraphic correlation of neoproterozoic sedimentary succession: upper tindiar group, northwestern canada, as a test case. geology, v. 20, pp. 181-185. kupecz j. a. and land l. s. (1991). late-stage dolomitization of the lower ordovician ellenburger group, west texas. journal of sedimentary petrology, 61, 551 –574. land l. s. (1985). the origin of massive dolomite. journal of geological education 33, 112ð125. land l. s. (1992). the dolomite problem: stable and radiogenic isotope clues. in: clauer, n. and chaudhary, s. (eds.), isotopic signatures of sedimentary records. lecture notes in earth sciences, v. 43, pp. 49-68. land l. s. and hoops g. k. (1973). sodium in carbonate sediments and rocks: a possible index to salinity of diagenetic solutions. journal of sedimentary petrology 43: 614 – 617. leyreloup a., dupuy c. and andriambololona r. (1977). chemical composition and consequences of the evolution of the french massif, central precambrian crust. contributions to mineralogy and petrology, 62: 283 300. masuda a. and lkeuchi y. (1979). lanthanide tetrad effect observed in marine environment. geochem. j. 1: 19 – 22. morteani g., moller p. and schley f. (1982). the rare earth element contents and the origin of the sparry magnesite mineralisations of tuxlanersbach, entachen alm, spiessnagel, and hochž lzen, austria, and the lacustrine magnesite deposits of aiani-kozani, greece, and bela stena, yugoslavia. economic geology 77: 617 –631. narbonne g. m., kaufman a. j. and knoll a.h. (1994). integrated chemostratigraphy and biostratigraphy of the indemere supergroup, northwestern canada: implications for neoproterozoic correlations and early evolution of animals. bull. geol. soc. amer. 106: 1281 – 1292. ndougsa–mbarga t. (2004). etude géophysique par méthode gravimétrique des structures profondes et superfi cielles de la region de mamfé: thèse doctorat/ phd, fac., sci., univ. youndé i, 255 p. ndougsa–mbarga t. and ntep–gweth p. (2005). report on the mamfe salt and sapphire reconnaissance expedition: internal report archives n°0058/2005 of the ministry of industry, mines and technological development, 12 pp. ndougsa–mbarga t., manguelle-dicoum e., bisso d.and njingti n. (2004). geophysical evaluation based on gravity data of the mamfe basin, southwest cameroon: segmite international. a journal of resource, industrial and environmental geology 1(1): 15 – 20. ndougsa–mbarga t., manguelle-dicoum e., campos-enriquez j. and yene atangana q (2007). gravity anomalies, sub-surface structure and oil and gas migration in the mamfe, cameroon-nigeria, sedimentary basin. geofísica internacional 46 (2): 129 – 139 ngando a. m.,. tabod c. t, manguelledicoum e., nouayou r., marcel j. and zakariaou a. (2004). structure géologique le long de deux profi ls audio magnétotelluriques dans le basin de mamfé. journal of cameroon academy of science 4(2): 149-162. bassey edem ephraim56 njieatih a. h. (1997). sedimentology and petroleum potentials of the middle cretaceous sediments of the mamfe embayement. southwestern cameroon. msc dissertation, university of ibadan, nigeria, 110 p. nothdurft l.d., webb g.e., kamber b.s. (2004). rare earth element geochemistry of late devonian reef carbonates, canning basin, western australia: confirmation of a seawater ree proxy in ancient limestones. geochimica et cosmochimica acta 68: 263 – 283. nouayou r (2005). contribution à l’étude géophysique du bassin sédimentaire de mamfe par prospections audio et hélio magnétotelluriques thèse de doctorat d’etat ès sciences, spécialité géophysique interne, université de yaoundé i, 184p ofomata g. e. k. (1973). aspects of geomorphology of the nsukka–okigwe cuesta, eastcentral state of nigeria. ifan bull. 35: 489 – 501. olade m. a., (1975). evolution of nigeria’s benue trough (aulacogen): a tectonic model: geological magazine 112(6): 93 – 103. oti, m. n. (1983). petrology, diagenesis and phosphate – mineralization of cretaceous limestonein the arochukwu/ohafia area, southeast nigeria. nig. j. min. geol. 20 (1 & 2): 95 – 103 pandit m. k., sial a. n., malhotra g., shekhawat l. s. and ferreira v. p. (2003). c-, oisotope and whole-rock geochemistry of proterozoic jahazpur carbonates, nw indian craton. gondwana research 6 ( 3): 513 – 522 petters s. w. (1978). dolomitization of the ewekoro limestone. j. min. geol. 15 (2): 78 – 83 petters s.w. (2004). southeastern benue trough and ikom – mamfe embayment. in: ekwueme b. n., nyong e. e. and petters s.w (eds). geological excursion guidebook to the oban massif, calabar flank and mamfe embayment, southeastern nigeria, soleprint (nig.) co., calabar. petters s. w., okereke c. s. and nwajide c. s. (1987). geology of the mamfe rift, southeastern nigeria. in: matheis, g. and schandelmerer, h. (eds) current research in african earth sciences, balkema, rotterdam: 299 – 302. purser, b., tucker, m., and zenger, d. (1994). “problems, progress, and future research ‘ concerning dolomites and dolomitization.” in: purser. b. tucker. m. and zenger. d. (eds.) dolomites: a volume in honour of dolomieu. international association of sedimentologists special publication, 21: pp. 3 20. qadhi t. m. (2008). testing jabal farasan marble deposit for multiple industrial applications. the arabian journal for science and engineering 33 (ic): 79 – 97. reymont r. a. (1965). aspects of the geology of nigeria. ibadan university press, ibadan, nigeria., 145pp. sass e. and katz a. (1982). the origin of platform dolomites: new evidence. american journal of science. 282: 1184-1213. shearman d. j. and shirmohammadi n. h. (1969). distribution of strontium from dedolomite from the french jura. nature, lond. 223: 606 – 608. tabod c.t. (2008). an audio-magnetotelluric investigation of the eastern margin of the mamfe basin, cameroon. united nations educational, scientific and cultural organization, and international atomic energy agency, the abdus salam international centre for theoretical physics, ic/2008/099, http://publications.ictp.it tanaka k. and kawab, i. (2006). ree abundances in ancient seawater inferred from marine limestone and experimental ree partition coefficients between calcite and aqueous solution. geochemical journal 40: 425 – 435. tanaka m., shimizu h. and masuda a. (1990). features of the heavy rareearth elements in seawater. geochemical journal 24: 39 – 46. taylor s. r. and mclennan s. m. (1981). the composition and evolution of the continental crust: rare earth elements evidence from sedimentary rocks. phil. trans. r. society of london a30: 381 – 399. taylor t. r. and sibley d.f. (1986). petrographic and geochemical characteristics of dolomite types and the origin of ferroan dolomite in the trenton formation, ordovician, michigan basin, u.s.a. sedimentology 33:61 – 86. tokam a. k., tabod c. t., nyblade a. a., julia j., wiens d. a. and pasyanos m. e. (2010). structure of the crust beneath cameroon, west africa, from the joint inversion of rayleigh wave group velocities and receiver functions, geophys. j. int. doi: 10.1111/j.1365246x.2010.04776.x: 1 – 16 tucker m. e. (1983). diagenesis, geochemistry and origin of a precambrian dolomite: the beck spring dolomite of eastern california. j. sed. petrol. 53: 1097 – 1119 tucker m. e. and wright v. p. (1992). carbonate sedimentology. blackwell scientific publications, oxford, uk. turekiah k. k. and wedepohl k. h. (1961). distribution of elements in some major units of the earth’s crust. geol. soc am. bull 72, 125p veizer j. (1983). chemical diagenesis of carbonates: theory and application of trace element technique. in: arthur m.a., anderson t.f., kaplan i.r., veizer j. and land l.s. (eds.) stable isotopes in sedimentary geology. soc. eco. petrol. mineral. short course notes 10, tulsa, pp. 3.1-3.100 veizer j. clayton r. n. and hinton r. w. (1992). geochemistry of precambrian carbonates: iv. early paleoproterozoic sea water. geochim. cosmochim. acta 56: 875 – 885. veizer j., hoefs j., lowe d.r. and thurston p.c. (1989). geochemistry of precambrian carbonates: ii. archean greenstone belts and archean seawater. geochim. cosmochim. acta 53: 859 – 871. warren j. (2000). dolomite: occurrence, evolution and economically important associations. earth science reviews 52: 1 – 81. wedepohl k. h., correns c. w., shav d. m., turekian k. k., zemann j. (1974). handbook of geochemistry, springer verlag, berlin welch, c. l. (2001). petrography and geochemistry of dolomites in the lower cretaceous edwards formation, taylor county, texas. m.sc. thesis, texas tech university, 129p. xu, b. v. (2011). microfacies, carbon and oxygen isotopes of the late archaean stromatolitic carbonate platform of the kaapvaal craton, south africa: implications for changes in paleo-environment. dissertation of the faculty of geosciences, ludwig-maximilians-university munich.munich yao q. j. and domicco r. v. (1997). dolomitization of the cambrian carbonate platform, southern canadian rocky mountains: dolomite from geometry, fluid inclusion geochemistry, isotopic signature, and hydrogeologic modeling studies. american journal of science 297: 892 – 938. logging observations of auger profiles (tarmizi, 2014) indicate a vertical, downwards, general decrease of peat humification levels with depth in a tropical lowland peat forest in the kota samarahan-asajaya área in the región of west sarawak (malaysia). based on pollen analyses and field observations, the studied peat profiles can be interpreted as part of a progradation deltaic succession. continued regression of sea levels, gave rise to the development of peat in a transitional mangrove to floodplain/floodbasin environment, followed by a shallow, topogenic peat depositional environment with riparian influence at approximately 2420 ± 30 years b.p. (until present time). the inferred peat vegetational succession reached phasic community i at approximately 2380 ± 30 years b.p. and followed by phasic community ii at approximately 1780 ± 30 years b.p., towards the upper part of the present, ombrogenic, peat profile. observations of the presence of large, hollow, shorea type trees, supports that successive vegetational zonation of the tropical lowland peat dome may have reached phasic community ii. some pollen types were found that are also known to occur in the inferred vegetational zonation of phasic community iii and iv or higher. pollen analyses indicates that estuarine and deltaic, brackish to saline water influence may have gradually ceased at approximately 0.5 m below the lithological boundary between peat and underlying soil (floodplain deposit) in the tropical lowland peat basin. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): o1 o10 abstract keywords: lowland peat swamp, phasic community, mangrove swamp, riparian environment, pollen diagram, vegetation succession. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.53670 e n g in e e r in g g e o l o g y mohamad tarmizi mohamad zulkifley1, ng tham fatt1, zainey konjing2, muhammad aqeel ashraf3,4* 1. department of geology, faculty of science, university of malaya, kuala lumpur, malaysia 2. biostratex sendirian berhad, batu caves, gombak, malaysia. 3. department of environmental science and engineering, school of environmental studies, china university of geosciences, 430074 wuhan, p. r. china 4. faculty of science & natural resources, university malaysia sabah88400 kota kinabalu, sabah, malaysia. corresponding authors: muhammad aqeel ashraffaculty of science & natural resources university malaysia sabah, 88400, kota kinabalu, sabah. (ashraf@ums.edu.my) development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia record manuscript received: 20/10/2015 accepted for publication: 12/02/2016 how to cite item mohamad, m. t., fatt, n. t., & ashraf, m. a. development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia. earth sciences research journal, (20)1, o1-o10. doi: http://dx.doi.org/10.15446/esrj.v20n1.53670 1. introduction organic materials only accumulate to form peat under certain conditions, whereby it is essential that the production of biomass (organic materials) be greater than its chemical breakdown to form peat (andriesse, 1988). anaerobic, muddy conditions, which prevent the microbiological activity needed for the chemical breakdown of organic materials, are assumed to be largely responsible for the accumulation of partly decomposed biomass/organic matter in the form of peat. the anaerobic conditions are created by a particular hydro-topography whether marsh, swamp, bog or mire. properties of such hydrotopographic units depend on many environmental factors, including climate, landform, local geology and hydrology (andriesse, 1988). the definition of peat also refers to the net accumulation of purely one hundred percent organic matter and the difference between soil and organic or vegetative accumulation varies (murtedza et al., 2002) probably due to varying definitions from the different academic fields associated with the practical study of peat and its’ properties (agriculture, botany, geology and engineering, for instance). ‘peat’ has been alternately referred to as ‘organic soils’ and histosols. peat is referred to as organic soils by mass composition (murtedza et al., 2002), i.e. soils that contain at least 65% organic matter or conversely, less than 35% mineral content. the soil division of sarawak (malaysia) has adopted a definition for organic soil that is based on profile partition, i.e. soils that have 50 cm or more organic soil matter within 100 cm or more than twice that of mineral soil materials overlying bedrock within 50 cm (murtedza et al., 2002). on the other hand, the united states department of agriculture (usda) has defined soil types as organic soils (or histosols) if more than half of the upper 80 cm of the soil is organic or if organic soil material of any thickness rests on rock or fragmental material having interstices filled with organic materials (united states department of agriculture, 1998; murtedza et al., 2002). peats, as we now know, are formed by the limited decomposition and hence, the accumulation of organic soil materials and these organic materials can further consist of undecomposed, partially decomposed and highly decomposed plant remains. tropical lowland peats usually have undecomposed and partly decomposed branches, logs or twigs. tropical lowland peats form a fragile ecosystem because of its’ domed shape and is almost purely organic (paramananthan, 2011). o2 mohamad tarmizi mohamad zulkifley, ng tham fatt, zainey konjing and muhammad aqeel ashraf 2. origin and characteristics of peat deposits the paludal deposits of sarawak occur in large basin swamps and in small interior valleys that have developed in mostly near coastal areas in relatively recent times (murtedza et al., 2002).14c dating from baram in northern sarawak indicates that the sea was at the inland margin of the peat swamp about 5400 years ago (wilford, 1960; murtedza et al., 2002). peat basin swamps are dome-shaped with ‘ombrogenous peat’ occupying the central portion of the peat dome (murtedza et al., 2002; tie, 1990). ‘ombrogenous peat’ or ombrogenic peats (paramananthan, 2011), comprises mainly of broken tree trunks, branches, leaves, roots and fruits (yogeswaran, 1995; huat, 2004; anderson and muller, 1975, anderson, 1976) and are relatively mineral or ash free. however, the marginal peats surrounding the base and along fringes or margins of the peat dome and lower bank of streams draining the peat swamps are dominated by ‘clayey or topogenic peat’, which consists of mainly slightly to moderately decomposed plant matter and fine clastic mineral-sediments (moderate to high ash content). 3. description of site studied the study area covers approximately 25 km2, is located between longitudes 01º 26’ 30’’n and 01º 29’46’’n (north and south of sungai tuang) and latitudes 110º 27’44’’e and 110º 30’58’’e (east and west of batang samarahan). it is located about 30 km from kuching city and is accessible by sealed road. within the study area (figure 1), accessibility is possible using gravel, bund roads or sealed roads. figure 1. location map of study area. physiography-topography, drainage, climate and vegetation the topography of the kota samarahan-asajaya area is flat and low-lying with isolated hills standing out of the floodplains of the batang samarahan and sungai tuang rivers. the general elevation of the flood basin is low and seldom higher than 5 m above sea level. the majority of these areas are prone to high tide flooding except few localities on the natural levees along rivers and man-made bunds. the major rivers draining the study area into the south china sea are the batang samarahan and its distributary, sungai tuang. the meandering batang samarahan flows in a northerly direction while the sungai tuang flows eastward to join the batang samarahan. the area experiences an equatorial climate which is characterized by hot and humid conditions with a drier period during the middle of the year and high rainfalls at the end of the year. 4. objective of study the initial purpose of this study was to identify, classify and map the distribution and natural occurrence of peat deposits in the kota samarahan-asajaya study area. a further objective was to determine the depth and thickness of peat encountered in the project area. the stratigraphy of a tropical lowland peat swamp and some of its related palynological characteristics were studied using samples collected from auger holes. the palynological investigation was performed to identify the peat swamp phasic communities’ vegetation succession that had occurred from basal clay to present-day tropical lowland peat swamp forest based on pollen analyses of augered core samples. also, this study aims to determine the initial age of peat deposition by 14c dating method. 5. materials and methods field methods augering was done using hand augers. augered holes and sampling points were located using a garmin gps device (models gps 76 and gpsmap 76csx). hand augered core samples were extracted, photographed and logged in the field. peat classification was carried out on a scale of h1 to h10 in the field (mohamad tarmizi, 2014) using the von post classification method (engineering geology working group, 2007; von post, 1922). peat sampler the peat or russian-type sampler was mainly used for peat sampling. it consists of an anchored fin and a movable sampling chamber. samples collected using the peat sampler is not disturbed compared to those collected with the gouge auger (guts auger). this is because the core sample remains enclosed within the sampling chamber throughout the process of sampling. field description peat and organic soils for field description, slightly organic silts or clays will appear as fine-grained inorganic soils, probably black or dark brown in colour, with an organic odour and possibly some visible organic remains. peat, on the other hand, may well appear to be completely organic, contains many recognizable plants remains, have a low density and also being black or dark brown in colour. the peat material was described using the von post squeeze test and results were provided according to the von post degree of humification with values between h1 and h10. soft soils classification and description in the field, the cores were logged and described by visual observation and feel . the samples were described according to the malaysian soil classification system for engineering purposes and field identification (engineering geology working group, 2007; jarret, 1995; bujang, 2004) which includes the factors of organic content and degree of humification for classification of peat. soils that have organic content from 3 to 20% are termed slightly organic soils, and soils with organic content in the range of 20 to 75% are classified as organic soils. finally, peats have more than 75% organic content. fibric or fibrous peats (ptf) have a humification range of h1 to h3 whereas hemic or moderately decomposed peats (pth) have a range of h4 to h6. sapric or amorphous peats range from h7 to h10 on the humification scale (jarret, 1995; bujang, 2004). general geology soft, quaternary sediments underlie the study area. two sedimentary facies observed to be exposed in the study area are the peat (paludal) o3development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia deposits and floodplain deposits. from logging observations and pollen analyses, these two facies were observed and concluded to be deposited over the estuarine/deltaic sediments (mohamad tarmizi, 2014 and mohamad tarmizi et al., 2013(a), 2013(b) and 2013(c)). peat (paludal) deposits the term paludal deposits has been introduced to define a holocene unit mainly made up of peat (lam, 1989). peat covers a small portion of the western part of the study area (also referred here as ‘plaie area’). the thickness of the peat in the western part/plaie area ranges from 0.2 to 2.0 m. the types of peat mapped and encountered in the study area are (a) topogenous peat: also known as clayey peat, composed of plant remains and clastic sediments (with ash content) and formed in topographic depressions by plants that extracted nutrients from the mineral subsoil. topogenous peat is influenced by flood waters which supply the mineral components (mainly clays and silts). (b) ombrogenous peat: plant remains formed by plants growing solely on nutrient cycled through vegetation and peat (no mineral subsoil or ash content) and is fed by rain water only. the type of peat produced is composed of plant remains without any clastic sediments from the mineral subsoil. it is usually formed beyond the flood levels and away from mineral deposition by flooding rivers. the topogenous peat is composed of slightly to moderately decomposed plant matters. topogenous peat with a humification range of h3 to h8 on the von post humification scale was augered, logged and classified at the peat forest, in the plaie area, on the western part of the study area (figure 5). ombrogenous peat is characterized by a high water content, an extremely low bulk density, a relatively higher decomposition rate and a relatively lower ph, and is composed of mainly loose trunks, branches, roots, fruits and leaves (paramananthan, 2011). ombrogenous peat encountered in the study area is moderate to highly decomposed and ranges from h4 (hemic) to h7 (sapric) on the von post scale of humification. the munsell colour value is 7.5yr 2.5/2 (very dark brown) depending on the degree of humification. peat in the western peat forest of study area (plaie area) peat in the plaie area is composed of slightly to moderately decomposed plant material ranging from h4 to h8 on the von post degree of humification scale. the peat that was encountered vertically at location ks.tp.10 can be classified as hemic to sapric, hemic and fibric to hemic peat. the thickness of the peat layers from the margin (ks.tp.02) towards the basin centre (ks.tp.10) ranges from 0.2 to 2.3-m thick, getting thicker westwards. the peat encountered at locations ks.tp.09 and ks.tp.10 vertically has a dark brown to very dark brown colour ranging from 7.5 yr 3/3 and 7.5 yr 3/2 (dark brown) to 7.5 yr 2.5/2 (very dark brown) according to the munsell soil colour chart (table 1). the groundwater levels in the western part of kota samarahan (plaie) peat area are approximately 0.3 to 0.5-m below ground surface (mohamad tarmizi, 2014) at augered locations ks.tp.09 and ks.tp.10 respectively (figure 2). field identification and classification (von post) of the top to the bottom peat layer of the tropical lowland peat dome shows that there is a vertical variation of peat humification levels (von post) with every 0.5-m interval of depth in the form of dominantly occurring hemic to sapric (h6 to h7), hemic (h5 to h6), hemic (h4 to h6) and fibric to hemic (h3 to h5) humification levels, respectively, at auger hole location ks.tp.10 , and sapric (h7 to h8), hemic to sapric (h6 to h7) and hemic (h5 to h6) humification levels, respectively, at auger location ks.tp.09 , with both profiles indicating a vertical decrease of peat humification levels with peat depth. floodplain deposits the exposed floodplain deposits cover the northern, eastern and southern parts of the study area, including the cross river asajaya areas clockwise from kg. sui, kg. tanjung parang, kg. tanjung tuang to the sg. paloh area and westwards. the deposits are observed to be overlain by peat deposits (mohamad tarmizi, 2014 and mohamad tarmizi et al., 2013) in the western plaie area and a portion of the kg. sui area (figure 2). figure 2. location of peat cores ks.tp.02, ks.tp.07, ks.tp.08, ks.tp.09 and ks.tp.10 within the peat basin/dome/deposit and depth or thickness of peat (lithological depth concluded from field observations and pollen analyses) table 1. thickness/depth of sampled peats occurring from the peat basin margin (near location ks.tp.02) towards the approximate centre of the peat dome (near location ks.tp.10) in the western peat forest of the kota samarahan-asajaya study area o4 mohamad tarmizi mohamad zulkifley, ng tham fatt, zainey konjing and muhammad aqeel ashraf floodplain deposits, also known as riverine deposits, were formed by the deposition of river sediments during floods in flood basins (backswamp deposits) or on levees along the river banks (levee deposits). these floodplain deposits are confined to river valleys upstream of the estuarine/deltaic system and away from tidal marine or estuarine influences. the levee floodplain deposits form narrow belts along the banks of the rivers and streams and consist mainly of silt and fine sand, rare gravel and minor to abundant plant remains. the backswamp floodplain deposits which are behind the levees consists predominantly of clays and silts with minor to abundant plant remains (lam 1989). lithology the lithology of the quaternary riverine or floodplain (backswamp) deposits observed to occur in the area consists mainly of clay, silt and fine sand with minor to abundant plant matters (wooden fragments 0.5 to 4-cm thick). plant matters are observed to occur as wood fragments ranging from less than 5% to 50 %. these deposits are usually pale brown in colour and sometimes greenish gray to dark bluish gray or bluish gray in colour with munsell colour values of 10yr 3/3 to 10yr 6/3, gley 2 6/1, gley 2 5/1, gley 2 4/1 and 10yr 4/2. typical auger log sections of the floodplain deposits are as shown in figure 2 (log no ks.tp.02 to ks.tp.10). for the profile section of ks.tp.10 and the related pollen diagram (figure 5), from the beginning of the interval 3.5 to 2.5 m, begins the rise in pollen abundance followed by peaking at 2.5 m for the elaeocarpus angustifolius pollen. the interval 2.5 to 2.0 m shows an increase for eugenia grandis type pollen and abundance peaks at 2.0 m. thus, it is interpreted here that an influx of pollen from riparian type vegetation had occurred due to increasing freshwater influence from 2.5 to 2.0 m and that this interval represents an increasingly riverine/floodplain depositional environment. from the profile section of ks.tp.09 (figure 4), at the interval 2.0 to 1.4 m, the presence of the riparian type pollen of both eugenia and elaeocarpus again supports the deposition of a layer of floodplain/floodbasin deposits (approximately 0.5m thick) immediately below the peat layer and overlying the estuarine and deltaic deposits. estuarine/deltaic deposits the estuarine/deltaic deposits are found to underlie the floodplain and peat deposits in the study area (tarmizi, 2014). from pollen analyses and the pollen diagram of the soil samples at interval 2.5 to 2.0 m of the profile section at ks. tp.10, the pollen abundance of rhizophora type pollen starts to decline from 2.5 to 2.0 m and disappears at 2.0 m, which marks the possible end or dominance of mangrove type vegetation and related mangrove depositional environment. from the beginning of the interval 3.5 to 2.5 m, begins the rise in pollen abundance followed by peaking at 2.5 m for the elaeocarpus pollen. the interval 2.5 to 2.0 m shows an increase for eugenia type pollen and abundance peaks at 2.0 m. hence, it is interpreted from pollen abundance of riparian type vegetation that freshwater influence had risen from 2.5 to 2.0 m and that this interval probably represents an increasing floodplain depositional environment with lesser mangrove/ estuarine or marine influence. thus, it is interpreted from this part of the profile section that the estuarine/mangrove type sediment depositional environment had decreased, and saline or brackish water influence has begun to lessen due to continued sea level regression from 2.5 to 2.0 m and, that the mangrove/estuarine or deltaic depositional environment is more predominant from 2.5 m and below the mentioned profile (ks.tp.10). likewise, from pollen analyses and related pollen diagram of the soil samples at interval 2.0 to 1.4 m of the profile section at ks.tp.09, the pollen abundance of rhizophora type pollen starts to decline from 2.0 to 1.4 m and disappears at 1.4 m, signifying the gradual cessation of mangrove type vegetation and related mangrove type depositional environment from 2.0 to 1.4 m. thus, estuarine and deltaic brackish to saline water influence may have gradually ceased at approximately 0.5 m below the peat-soil boundary as indicated by the core auger profiles of both ks.tp.10 and ks.tp.09. as a result, this approximately 0.5 m thick layer consisting of the floodplain, riverine or floodbasin deposits was deposited by flooding river waters after or with the gradual decline of mangrove influence due to receding sea levels. lithology the underlying estuarine/deltaic deposits are composed of mainly soft to very soft, marine clayey silts and silt, clay and sometimes thin layers of fine sand accompanied with shell debris/fragments as observed in augered holes ks.ts.05 and ks.ts.06. these deposits are greenish gray to dark bluish gray or bluish gray in colour with munsell colour values of gley 2, 6/1, gley 2, 5/1, gley 2, 4/1 and 10yr, 4/2. the estuarine and deltaic deposits are probably marine in origin and, in the field, these deposits can usually be differentiated from the overlying floodplain deposits because the former has a very soft consistency, has a gritty feel when rubbed between the fingers, greenish gray to dark bluish gray or bluish gray in colour, is saturated or very wet in appearance and is sometimes accompanied by marine shell fragments (lam, 1989). these deposits are observed, in the field, to have little or no wooden fragments when compared to the overlying floodplain deposits. they are usually sulphidic (acidic) and produce yellow coloured minerals known as jarosite after oxidation and exposure to the atmosphere (fitzpatrick et al., 1998). 6. results peat characterization characterization of logged peat samples, at augered holes ks.tp.09 and ks.tp.10 from locations near and towards the centre of the peat dome, in the western peat forest of the kota samarahan-asajaya study area are as described in table 2 (tarmizi, 2014; tarmizi et al., 2013(a), 2013(b), 2013(c). table 2. description of sampled peats occurring near the approximate centre of the peat dome (near locations ks.tp.09 and ks.tp.10), in the western peat forest of the kota samarahan-asajaya study area o5development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia the pollen diagram pollen analyses: methods a total of 8 samples were collected for every 0.5-m interval from the surface to a depth of 5.5 m for borehole ks.tp.10, and four samples were collected from the surface to a depth of 2.0 m for ks.tp.09 (mohamad tarmizi, 2014). these samples for pollen analyses were taken every 10 cm along each core for regular core sampling intervals such as shown in table 3 except for surface samples and samples for both radiocarbon dating and pollen analyses (mohamad tarmizi, 2014) which was taken at 180 to 200 cm (regular interval 2.0 to 1.5 cm) from borehole location ks.tp.10. figure 3 geological map of the kota samarahan-asajaya study area, west sarawak showing sampling location in the western plaie peat forest area. the enlarged inverted section (top left-hand corner) are the locations of the major peat cores selected for this study. the samples were treated in the laboratory with 10% koh, sieved to remove larger pieces of plant material, undergone acetolysis and heavy liquid separation in zinc bromide/hydrochloric acid for the samples in which some mineral residue was present especially in some clayey peat or organic soil samples (wood et al. 1996). the samples revealed considerable quantities of pollen grains and spores and counts varied from 26 to 266 grains per slide for ks.tp.10 and from 51 to 110 grains per slide for ks.tp.09. figure 4. pollen and spores assemblage (ks.tp.09). table 3. general and exact sampling intervals of auger cores from locations ks.tp.10 and ks.tp.09 o6 mohamad tarmizi mohamad zulkifley, ng tham fatt, zainey konjing and muhammad aqeel ashraf the results of the pollen analysis are as shown in the diagram figures 4 and 5 inclusive of all pollen and pteridophyte spores . pollen types which could not be identified with any certainty were excluded from the pollen sum. in the analyses of the peat borehole profile of ks.tp.10, no extreme over-representation was observed and hence all pollen, and spore types have been included in the pollen sum. however, the over-representation of some peat pollen types may occur (for e.g. the cephalomappa pollen) in peat swamps due low pollen dispersal when only a single prolific tree may produce the maximum pollen representation (anderson and muller, 1975). also, misrepresentation or misinterpretation due to the absence of shorea type pollen may also occur as shorea type trees were observed present in the field at the locations of the studied augered holes (ks. tp.09 and ks.tp.10). the absence of pollen from shorea type trees maybe due to the low pollen producing capability of these tree species. the affinity of the types recognized is indicated on the pollen diagram either by reference to a living taxon or, if the type occurs in more than one taxon, by adding “type”. the comparison was mainly limited to taxa known to occur in peat swamp forest, and in adjacent vegetational environments such as mangrove and riverine (or riparian) vegetation. description of the diagram from the pollen diagram, it is possible to interpret the inferred succession of some forest types present in the locality of the augered holes (ks.tp.09 and ks.tp.10), and that these can be used to assess the past and current depositional environments represented by different sampling intervals from the augered hole profiles mentioned. the inferred succession of various forest types is described and discussed in the following sections. description of pollen diagram for borehole ks.tp.09. a total of 4 samples were analyzed from the locality of borehole ks.tp.09 (figure 3 2) (tarmizi, 2014). based on the pollen diagram (figure 4) for the core profile from this augered hole and from the regular interval of 2.0 to 1.5 m, the rhizophora and oncosperma pollen is dominant along with a few grains of the excoecaria aggulocha pollen. this range is also well marked by the occurrence of pollen originating from riparian vegetation which is elaeocarpus and eugenia. the asplenium type spores with acrostichum aureum and stenochlaena palustris are also present, which represents non-arboreal pollen and spores. the sampling from 1.4 to 1.0 m shows a significant change in pollen assemblage, characterized by a decrease in the mangrove elements of oncosperma and rhizophora. this decrease continues with only 2 grains of oncosperma pollen occurring at 0.5 to 1.0 m and the marked disappearance of the rhizophora pollen from 1.4 to 1.0 m. from 2.0 to 1.4 m, there is a sharp increase of peat swamp pollen type campnosperma and a decrease of the mangrove elements. at 1.4 m and the actual beginning of initial peat deposition, there is a prominent peak for campnosperma. from 1.4 to 1.0 m, there is a marked decrease of campnosperma and is accompanied by other peat swamp species which, however, are only present in low numbers such as gonystylus, cyrtostachys, cephalomappa and zalacca. the upper section between 1.4 to 0 m shows a decreasing trend of campnosperma abundance but still shows dominance compared to other pollen types. the other types of peat swamp pollen present from 1.4 to 0 m in the peat section of the profile are alangium, calophyllum, dillenia, durio, parastemon, timonius, shorea, melanorrhea and stemonurus. the pollen types casuarina and podocarpus are also recorded in this interval. the non-arboreal pollen and spores which characterize this range are represented by asplenium type, figure 5. pollen and spores assemblage (ks.tp.10). o7development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia stenochlaena palustris, polypodium, pteris type and lygodium microphyllum. stenochlaena palustris has shown a significant trend as this spore is highly abundant from 1.4 to 1.0 m and shows a decreasing trend from 1.0 to 0 m. from 1.0 to 0.5 m, the pollen types found to include cephallomappa, callophyllum, campnosperma, palaquim, parastemon, zalacca followed by oncosperma, casuarina, magnolia, polypodium, stenochlaena palustris, durio and eugenia. from 0.5 to 0 m, the significant pollen types that make up the peat swamp catena, as reported by anderson and muller (1975) and observed to occur here, are stemonurus, campnosperma, palaquim and dactylocladus. these are followed by less significant types such as cephallomappa, callophylum and blumeodendron. the abundance of rhizophora and oncosperma along with minor amounts of excoecaria aggulocha between 2.0 to 1.5 m has revealed that the lower part of the inferred succession represents a former mangrove or estuarine swamp vegetational environment. the other important non-mangrove pollen types which are present in this zone are elaeocarpus and eugenia, which are locally subject to transportation by tidal action and are common species of riparian vegetation in sarawak (malaysia). these riparian type pollens may indicate and support the presence of a floodplain or a riverine depositional environment within this 2.0 to 1.4 m interval. a sharp floristic boundary is marked at 1.4 m. the zone between 2.0 to 1.4 m locally shows a sudden decrease in mangrove pollens such as oncosperma and rhizophora with the latter virtually disappearing at 1.4 m. this sharp decrease strongly indicates a transition zone from a mangrove to floodplain/riverine, to a peat swamp depositional environment. this zone may be comparable to the inferred succession of the peat swamp catena as studied and described by anderson and muller (1975) because it is characterized by the presence of significant pollen types of campnosperma, cyrtostachys and zalacca. hence, may strongly be comparable to the campnosperma-cyrtostachys-zalacca sub-association characteristic of transitional mangrove to shallow peats occurring before the vegetational succession of phasic community i (anderson and muller, 1975). this is further supported by the abundance of typical climbing ferns, of the spore type stenochlaena palustris especially at the lower part of the succession (1.5 m); this spore is common during the early development of peat swamps and is widely confined to the transitional mangrove-shallow peats (campnospermacyrtostachys-zalacca sub-association), which occurs prior to succession of phasic community i (anderson and muller, 1975). description of pollen diagram for borehole ks.tp.10. a total of 8 samples were analyzed from the locality of borehole ks.tp.10 (figures 2 and 3). from 5.5 to 3.5 m, the rhizophora pollen is present in relatively high percentages along with relatively high counts of the oncosperma pollen. the rhizophora pollens are still dominant from 3.5 to 2.0 m. however, this interval is marked by an increase of campnosperma, along with minor proportions of other mangrove pollens such as nypa fruticans, avicennia and excoecaria aggulocha. this range also records some prominent peaks from the riparian type pollens which are represented by elaeocarpus and eugenia. towards the top of the succession between 1.5 to 1.0 m, the mangrove elements show sharp decreasing percentages, especially the oncosperma and the rhizophora pollen with the latter virtually disappearing (at 2.0 m). this interval represents a transitional assemblage and is characterized by a significant change in pollen assemblage which is then dominated by peat swamp taxa. the campnosperma pollen is relatively dominant and is associated with typical peat swamp pollens such as blumeodendron, cyrtostachys, zalacca and parastemon. in the upper part of the section from 1.0 to 0 m, the campnosperma pollen is still dominant and is associated with other pollen types such as gonystylus, dillenia, cephalomappa and eugeissona. a peak from the dactylocladus pollen was recorded at 1.0 m. in the uppermost part of the assemblage, from 0.5 to 0 m, the dominant pollen types observed to occur are campnosperma followed by the presence of blumeodendron, gonostylus. parastemon and stemonurus the dominant non-arboreal pollens are represented by asplenium type spores throughout the succession especially from 0.5 to 0 m, with some minor occurrences of polypodium and stenochlaena palustris. the stenochlaena palustris is relatively more prominent from 2.0 to 1.0 m and shows an increasing trend before it virtually disappears in the interval 0.5 to 0 m. based on the analyses of the pollen assemblage from this locality, the assemblage can be divided to represent four main vegetational phases or environments of deposition. at the interval 5.5 to 3.5 m, the dominance of rhizophora and oncosperma clearly demonstrates a former mangrove or estuarine swamp environment . this range is locally overlaid by the interval 3.5 to 2.0 m that is characterized by the abundant occurrence of oncosperma, which is associated with the nypa fruticans pollen. the dominance of oncosperma and nypa fruticans are widely confined to the inland zone of mangrove or estuarine vegetation. the peak of elaeocarpus within this interval is probably related to tidal action or flooding which has brought an influx of this pollen type from riparian vegetational areas, and this relates to and supports a floodplain or riverine depositional environment. this zone may perhaps be comparable to the campnospermacyrtostachys-zalacca sub-association zone as studied by anderson and muller (1975) and is interpreted to occur within the 2.0 to 1.0 m interval. this zone which is locally defined as shallow, topogenic peat is interpreted and derived from the association of the campnosperma, cyrtostachys and zalacca pollen which is related to the initial development of peat or transitional phase from mangrove-floodplain to peat swamp environment. this is further supported by the abundance of stenochlaena palustris peaking at 1.0 m. at the interval of 1.0 to 0.5 m, approximately 2380 ± 30 years b.p.(mohamad tarmizi, 2014) , it is still interpreted as a combination of shallow peat, and probably begins to change to phasic community i (anderson and muller, 1975) as this zone is locally characterized by the peak of dactylocladus and the appearance of single pollen grain of gonystylus. from 0.5 to 0 m near the surface, or approximately 1780 ± 30 years b.p. until present time (mohamad tarmizi, 2014), the pollens of stemonurus and gonostylus are observed present indicating that vegetational successive zonation of phasic community ii has developed and may have been reached, but without the appearance of the shorea type pollen (however, the large, shorea type trees have been observed present in the field near to locations ks.tp.09 and ks.tp.10). 14c age of peat samples peat starts developing just above 200 cm from the profile at location ks.tp.10 which is near the centre of the peat basin and radiocarbon (14c) dating was attempted on a sample taken from 180 and 200 cm. an age of 2420 ± 30 years b.p. for the regular sampling interval 1.5 to 2 m (180 to 200 cm) was obtained for the age of initial deposition of peat into the tropical lowland peat basin studied (table 4). table 4. 14c age of peat sampling intervals (quoted in conventional years b.p. (before 1950 ad) o8 mohamad tarmizi mohamad zulkifley, ng tham fatt, zainey konjing and muhammad aqeel ashraf 7. discussion horizontal or lateral zonation in tropical lowland peat domes anderson (1961, 1964, 1976, 1983) and anderson and muller (1975) studied the domed topography of tropical peat deposits and the relationship between the concentric zonation of surface vegetation and increasing peat thickness, acidity and decreasing nutrient availability with horizontal distance between margin and centre of peat basin. variations that occur in peat type within the deposits reflect the succession and lateral migration of the surface vegetation and the associated environment concurrent with coastal progradation (murtedza et al. 2002, anderson and muller, 1975). buwalda (1940) in paramanathan (2011) earlier did work in sumatra, which supported the hypotheses/idea of horizontal zonality occurring in tropical lowland peats. he reported that different plant communities exist in the peat swamp forest depending on the thickness of the peat and the distance from the river (peat basin margin). in the central part of the peat forest, where the thickest peat deposits occurred, he observed that vegetation was poorly developed, with twisted and stunted trees and scattered pools of deep, acidic (ph value range of 3.0 to 3.5), brown water compared to vegetation occurring at shallower depths of peat (or peat deposited near or at the margin/boundary of peat dome/basin). however, on the peat deposits shallower than three meters deep, the soils had a ph of 3.5 to 4.5. buwalda (1940) in paramanathan (2011) reported six different vegetation types and zones occurring in the indragiri area in sumatra. anderson (1961, 1963, 1964) also described six vegetation zones occurring in the lowland peat forests of borneo and brunei. anderson (1961, 1963, 1983) did a comprehensive study of the ecology of the tropical lowland peat swamp forest. he observed and recorded 253 tree species that are mostly confined to the periphery of the peat swamp forest. according to anderson (1963), most of the plant species that grow in the forests at the centre of the peat domes are mostly those that are usually found on nutrient poorer soils, such as podzols of the heath forest (anderson, 1963). the tropical lowland peat swamp forests in sarawak, malaysia and adjacent brunei show lateral or horizontal changes in vegetation types from its periphery to the centre of the dome-shaped peat swamps and each of the six dominant lateral vegetation zone was designated “phasic community’’ by anderson (1961). six distinct phasic communities or successive lateral vegetation zones were recognized based on their floristic composition and structure of vegetation in each phasic zone (anderson, 1961; paramananthan, 2011) and were numbered phasic community i at the margin to phasic community vi in the centre of the peat swamp (table 5). table 5. characteristics of the six phasic communities occurring as horizontal or lateral zonation (adapted from anderson, 1961, 1963, 1983; paramananthan, 2011) o9development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia discussion on palynological analysis in general, palynological analysis has provided information regarding the locally inferred succession of a tropical lowland peat swamp vegetation. the pollen succession interpreted from the location at ks.tp.10 near the lowland tropical peat dome centre is comparable to a normal peat development which starts on mangrove clay. based on the pollen analysis and comparison to the inferred plant succession as studied and published by anderson and muller (1975), the peat deposit is concluded as local and autothochthonous and includes the phasic communities of i to ii. field observations have noted the presence of large shorea type species of trees with wide girths (1 to 3 m) along the traverse section from ks.tp.09 to ks.tp.10 which further supports that the vegetational zonation has reached phasic community ii (at an interval of 0.5 to 0 m) even though the pollen of shorea type is scarce and not observed in the studied profile of ks.tp.10. from 5.5 to 2.0 m, it is clear that mangrove or estuarine elements dominate. from 5.5 to 3.5 m, the dominance of rhizophora and oncosperma (plate 1) is evidence of a former mangrove swamp environment (anderson and muller, 1975). from 3.5 to 2.0 m there is the abundant occurrence and dominance of the oncosperma and nypa fruticans pollen which largely represents the inland zone of mangrove vegetation. the peak of elaeocarpus at 2.5 m in the 3.0 to 2.5 m interval is probably related to an influx of pollen from riparian type vegetation due to a freshwater influx or floodwaters, and the interval may thus represent and support a riverine floodplain depositional environment setting. overall, the analyses of the pollen succession at 2.0 to 1.5 (or at approximately 2420 ± 30 years b.p.) is indicative of a mangrove to shallow peat swamp transition zone interpreted to have developed near to the coast and which usually occurs before phasic community i (anderson and muller, 1975) and is indicated by the campnosperma-cyrtostachyszalacca sub-association (anderson and muller, 1975). from 1.5 to 1.0 m, the presence of peat pollen types that may support the vegetational successions or zones ranging from phasic community i and phasic community iv to vi is noted (anderson and muller, 1975). pollen types such as blumeodendron that are known to be abundant in phasic community i (anderson and muller, 1975), campnosperma (abundant in pc i), parastemon (abundant in pc i and pc iv to vi) and stenochlaena palustris (abundant in mangrove to shallow peat transition zones and pc i) are also present (anderson and muller, 1975). however, peak abundance of stenochlaena palustris occurs at this interval, thus signifying vegetation succession of phasic community i. from 1.0 to 0.5 m, at approximately 2380 ± 30 years b.p., the presence of peat pollen types that may support the vegetational succession of phasic community zones that range from phasic community i to vi is observed (anderson and muller, 1975). pollen types like dactylocladus (abundant in pc i to vi), pandanus (abundant in pc i to v) and parastemon (abundant in pc i and pc iv to vi) are also present in this interval as observed by anderson and muller (1975). however, stenochlaena palustris and zalacca (plate 1) is still present, which implies that the zonation may still be only in the phasic community i. from 0.5 to 0 m at location ks.tp.10, or approximately at 1780 ± 30 years b.p. until the present time, is noted the presence of peat pollen types that may support the occurrence of vegetational successions or zones ranging pc i to pc vi. pollen types like callophylum (abundant in pc i to pc iv), gonostylus (abundant in pc iv), blumeodendron (abundant in pc i), campnosperma (locally abundant near coastal margin of pc i), cephallomappa (abundant in pc ii to iii), eleocarpus (abundant in pc i to iv) and parastemon (abundant in pc i and pc iv to vi) are present (anderson and muller, 1975). in this interval, however, the stenochlaena palustris and zalacca pollen were observed not to be present, which implies that the zonation may no longer be phasic community i. hence, the zonation may have reached to be at least phasic community ii (or higher). as mentioned earlier, field observations support and have noted the presence of large, hollow, scattered shorea type (table 5) species of trees (abundant in pc ii to iv according to anderson and muller, 1975). 8. conclusion from field identification and von post classification as well as pollen analysis of the augered peat samples (ks.tp.09 and ks.tp.10) at the western peat forest of the study area, the findings could be summarized as follows: 1) field identification and classification (von post) of the vertical, top to the bottom peat layer on the tropical lowland peat dome in the western part of the study area shows that there is a vertical variation of peat humification levels for every 0.5 m interval, in the form of dominantly occurring decomposition levels of hemic to sapric (h6 to h7), hemic (h5 to h6), hemic (h4 to h6) and fibric to hemic (h3 to h5), respectively, at ks.tp.10; and sapric (h7 to h8), hemic to sapric (h6 to h7) and hemic (h5 to h6), respectively, at location ks.tp.09, with both profiles generally indicating a vertical decrease of humification levels with peat depth. 2) the proposed phasic communities inferred vegetational zonation that were observed to occur vertically, from the top to the bottom of the studied augered peat profiles near the centre of the tropical lowland peat dome are phasic community ii (alan swamp forest), phasic community i (mixed peat swamp forest) followed by the underlying transitional mangrove to peat swamp shallow/marginal peats, respectively. 3) based on pollen analyses and field observations, it can be concluded that most of the studied peat profiles from locations ks.tp.10 and ks.tp.09 were part of a progradation deltaic succession whereby a floodplain depositional environment is progressively replacing a mangrove swamp environment followed by a peat swamp environment. the peat swamp phasic communities inferred succession (bottom to top of peat profile) interpreted in this study is similar to that as studied by anderson and muller (1975) in the sense that it ranges from a transitional mangrove to peat swamp environment (or shallow/marginal peats) to a normal open swamp phasic community i (mixed peat swamp forest) that may have developed and extends up to at least phasic community ii (alan swamp forest) at present. also, field observations and the presence of large, hollow, scattered shorea type trees, supports that the surface, vegetational zonation of the studied tropical lowland peat dome may have reached phasic community ii, at the least. also, there are pollen types that are found to occur or are also abundant in pc iii and pc iv or more from pollen evidence and analysis, as previously reported by anderson and muller (1975). 4) it is interpreted that the depositional environment at the location of the profile seems to commence on a mangrove swamp environment which, due to a continued regression of sea levels, gave rise to a more inland mangrove swamp environment with an increased riparian influence. peat then starts to develop in a transitional mangrove to floodplain (or floodbasin), and followed by a shallow, topogenic peat depositional environment at approximately 2420 ± 30 years b.p. this shallow/marginal topogenic peat swamp assemblage with riparian influence then starts to develop with an inferred peat vegetational succession (anderson and muller, 1975) reaching phasic community i (at approximately 2380 ± 30 years b.p.) and phasic community ii (at approximately 1780 ± 30 years b.p. until present time), towards the upper part of the now ombrogenic peat profile. 5) it is concluded from pollen analyses of the augered core profiles of both ks.tp.10 and ks.tp.09, that estuarine and deltaic brackish to saline water influence may have gradually ceased at approximately 0.5 m below the lithological boundary between peat and soil. as a result, this approximately 0.5 m thick soil or slightly organic to organic soil layer consisting of floodplain or floodbasin deposits and laying between the peat and estuarine/deltaic deposits was increasingly deposited by flooding river waters after or with the gradual decline of mangrove or estuarine influence due to receding sea levels. acknowledgements the authors have no conflict of interests. the authors are thankful to all anonymous reviewers that have contributed constructive comments to improve this manuscript considerably. the authors also acknowledge the ippp grant pv018-(2011a) and bantuan khas penyelidikan (bkp) grant no o10 mohamad tarmizi mohamad zulkifley, ng tham fatt, zainey konjing and muhammad aqeel ashraf bk064-2015 for financial support from the university of malaya. the authors are grateful, appreciate and acknowledge the assistance and scholarship/ study leave (hlp) from the minerals and geoscience department of malaysia (jmg), public services department of malaysia, forestry department of malaysia and the ministry of natural resources and environment of malaysia. references anderson, j.a.r. (1961). the ecology and forest types of the peat swamp forests of sarawak and brunei in relation to their silviculture. ph.d. thesis, university of edinburgh, united kingdom. anderson, j.a.r. (1963). the flora of the peat swamp forests of sarawak and brunei, including a catalogue of all recorded species of flowering plants, ferns and fern allies. garden’s bulletin 20, 131-228. anderson, j.a.r. (1964). the structure and development of the peat swamps of sarawak and brunei. the journal of tropical geography, 18, 7-16. anderson, j.a.r. (1976). observations on the ecology of five peat swamp forests in sarawak and kalimantan. soil research institute bogor, 3, 45-55. anderson, j.a.r. (1983). the tropical peat swamps of western malesia. in: gore ajp (ed) ecosystems of the world, vol. 4b, mires: swamp, bog, fen and moor. elsevier scientific publishing company, amsterdam, 181-199. anderson, j.a.r. and muller, j. (1975). palynological study of a holocene peat and miocene coal deposit from nw borneo. review of palaeobotany and palynogy, , 19(4), 291-351. doi: 10.1016/0034-6667(75)90049-4andriesse, j.p. (1988). nature and management of tropical peat soils. fao soils bulletin 59. food and agriculture organisation of the united nations, rome. buwalda, p. (1940). bosverkenning in de indragirische bovenlanden. rep for res sta, bogor, indonesia (unpub) engineering geology working group. (2007). guideline for engineering geological investigation in peat and soft soils. minerals and geoscience department of malaysia. huat, b.b.k. (2004). organic and peat soils engineering. universiti putra malaysia press, serdang. fitzpatrick, r.w., merry, r.h., williams, j., white, i., bowman, g. and taylor, g. (1998). acid sulfate soil assessment: coastal, inland and minesite conditions. national land and water resources audit methods paper, australia. jarret, p.m. (1995). geoguide 6. site investigation for organic soils and peat. jkr document 20709-0341-95. institut kerja raya, malaysia. lam, s.k. (1989). quaternary geology of the kota samarahan area, sarawak. geological survey department of malaysia. murtedza, m., padmanabhan, e., mei, b.l.h., and siong, w.b. (2002). the peat soils of sarawak. in: strapeat (strategies for implementing sustainable management of peatlands in borneo) status report. university malaysia sarawak, sarawak. paramananthan, s. (2011). keys to the identification of malaysian soils according to parent materials (mimeo). param agricultural soil surveys, selangor, malaysia. mohamad tarmizi, m.z. (2014). peat stabilization, organic geochemistry and related palynological characteristics of a tropical lowland peat basin in the kota samarahan-asajaya area, west sarawak, malaysia (ph.d. thesiss). university of malaya, kuala lumpur. mohamad tarmizi m.z., ng t.f., kuna raj j., azman ghani, mustaffa k. s., hashim r., ashraf m.a., (2013a). definitions and engineering classifications of tropical lowland peats. bull eng geol environ. doi 10.1007/s10064-013-0520-5 mohamad tarmizi m.z., ng t.f., wan hasiah a., kuna raj j., paramananthan s., hashim r., ashraf m.a., (2013b). distribution, classification, petrological and related geochemical (sra) characteristics of a tropical lowland peat dome in the kota samarahan-asajaya area, west sarawak, malaysia. cent eur j geosci 5:285–314 mohamad tarmizi m.z., ng t.f., wan hasiah a., kuna raj j., azman ghani, mustaffa k. s., ashraf m.a., (2013c). geochemical characteristics of a tropical lowland peat dome in the kota samarahan-asajaya area, west sarawak, malaysia. environ earth sci doi 10.1007/s12665-014-3504-2 tie, y.l. (1990). studies of peat swamps in sarawak with particular reference to soil-forest relationships and development of domed shaped structures. ph.d. thesis, london metropolitan university, england. united states department of agriculture (1998). keys to soil taxonomy, 8th edition. natural resources conservation service, washington d.c. von post, l. (1922). sveriges geologiska undersoknings torvinventering och nagre av dess hittills vunna resultat, sr. mosskulturfor. tidskr, 1,1-27. wilford, g.e. (1960). radiocarbon age determinations of quaternary sediments in brunei and northeast sarawak. british borneo geological survey, annual report 1959. wood, g.r., gabriel, a.m., lawson, j.c. (1996). palynological techniquesprocessing and microscopy. in jansoniun j and mcgregor dc (eds), palynology: principle and application, american association of stratigraphic palynologist foundation, dallas, 29-50., yogeswaran, m. (1995). geological considerations in the development of kuching area dialogue session. geological and geotechnical considerations in civil works, geological survey of malaysia, kuala lumpur. earth sciences research journal earth sci. res. sj. vol. 17, no. 1 (june, 2013): 61 65 using conodont elements to distinguish permian-triassic boundary disconformity near haftad gholleh, central iran mostafa yousefirad1, somayeh ghanbari2 and mahnaz parvanehnejad shirazi2 1department of geology, faculty of earth sciences, payam-e-noor university, iran 2department of geology, faculty of earth sciences, payame noor university, shiraz, iran email: mostafa.yousefirad@yahoo.com mobile: 00989125700730 paleontology the present study focuses on the stratigraphy of the permian-triassic boundary in the haftad gholleh area in eastern arak and north of mahallat,located in central iran. this boundary of erosional unconformity or disconformity and as in other iranian sequences related to the permian period, contains dolomitic limestone and shale and sandstone disconformably located below the triassic sequence. a detailed measured stratigraphic section has provided conodonts from the permian–triassic boundary (ptb) sequences in the area being studied. three conodont areas have been recognied which place the ptb in this section by precise biostratigraphy. one of the most upper permian sequences belongs to the guadalupian period. the triassic sequence consists of vermiculate limestone layers with coloured shale inter beds belonging to the early triassic period; unconformity at the permian-triassic boundary therefore represents a hiatus of about 10 million years. el presente estudio se enfoca en la estatigrafía del límite pérmico-triásico en el área de haftad gholleh, al este de arak y al norte de mahallat, centro de irán. este límite de discordancia erosiva, al igual que en otras secuencias iraníes relacionadas al período pérmico, contiene calizas dolomías, esquistos y areniscas discordantes localizadas debajo de la secuencia triásica. en una medida estratigráfica detallada del área de estudio se encontraron conodontas del límite pérmico-triásico (ptb, por sus siglas en inglés). se reconocieron tres áreas de conodontas, lo que ubica esta sección del ptb como bioestatigráfica precisa. una de las secuencias superiores del pérmico pertenece al período guadalupiense. la secuencia triásica consiste en calizas con capas vermiculitas coloreadas de esquisto que pertenecen al período triásico temprano. la discordancia del límite pérmico triásico, entonces, representa un lapso cercano a los 10 millones de años. key words: conodont, haftad gholleh, permian-triassic, stratigraphy. palabras clave: conodont, haftad gholleh, permiantriassic, stratigraphy. record manuscript received: 08/10/2013 accepted for publication: 02/04/2013 abstract resumen introduction the end of the palaeozoic period and the beginning of the mesozoic saw the greatest of all extinctions in the phanerozoic causing mass mortality in the marine environment (up to 96% of species, 83% of genera and 55% of families) (sepkoski, 1990; raup, 1991) and a nearly equally large loss of life on land (erwin, 1993). important p/t boundary sections are located in iran (nw, ne and central iran, alborz region, belonging to the iranian plate, and the zagros fold belt, belonging to the arabian plate), oman, southern china, japan, pakistan, afghanistan, turkey, greenland, austria, italy, south africa and antarctica. the permian sea’s last regression period occurred at the end of this system, which is why the upper boundary of the permian strata in iran is commonly discontinuous but has a type of parallel unconformity (disconformity). nevertheless, evidence of continuous sedimentation from the permian to triassic been reported in many regions of iran (julfa, kandovan, amol, shahreza and abadeh). the recommended, but somewhat controversial, chronostratigraphical permian scale is summarised here, according to jin et al., (1997) and wardlaw (1999), but correlated with leven’s alternative classification (leven, 1992; kotlyar and pronina, 1995). the names midian (eventually abadehian) for capitanian, dzhulfian for wuchiapingian and dorashamian for changhsingian have been preferred concerning iran’s stratigraphy (fig. 1). due a lack of macroscopic and microscopic fossils, especially foraminifera, some uncertainties in conodonts’ biostratigraphy have not been studied and clarified. this research thus attempts to study conodonts, the figure 1. permian-triassic boundary in different areas of iran (aghanabati, 2004) mostafa yousefirad, somayeh ghanbari and mahnaz parvanehnejad shirazi disconformity interval and the permian-triassic boundary’s paleoecology in the area being studied. 2. geological setting the haftad gholleh area is a small part of central iran, located between volcanic belts at urmia kerman sanandaj sirjan. the soltanieh dolomites, zagun, lalun, mila and permian sequence deposits in the region are known as palaeozoic facies. soltanieh grey dolomite is overlain by a green to red shale formation containing little sand and mica, becoming converted locally into fine-grained sandstone. shale thickness varies from 100 to 200 metres (zagun formation). overlying the shale are about 50 metres of dense, hard lightcoloured quartzite sandstone which has been assigned to the lalun formation. this formation is overlain by light to dark dolomite containing coral, sand and crinoids, having less than 100 m thickness.the dolomite gradually changes from bottom to top from light grey to light green limestone, bearing trilobite fragments; such dolomite and limestone are assigned to the mila formation. this different kind of facies (permian) is located directly on older formations. no ordovician to carboniferous sequences thus occurred in the area being studied. permian deposits are mainly dolomite. in some locations, the base consists of lalun sandstone. the base of the permian deposits in some parts is calcareous shale associated with red sandstone and conglomerate. permian lithology mainly consists of thick bedded dolomite (light grey) (fig. 2). lithostratigraphy the upper permian sedimentary sequence in the area being studied included limestone continuing with dolomite; limestone has been assigned to the jamal formation. all deposits were isoclinal, indicating marine transgression in the studied area (permian). one sequence, having several meters of shale and sandstone, lacking fossils, had such adistinctive colour that it could be applied as a key bed. there were several metres of thin to medium-bedded claraia vermicular limestone inter-bedded with coloured shale above the key bed; this belonged to the early triassic, based on a study of conodont fragments. it has thus been concluded that early triassic deposits in the study area consisted of two parts: clastic (lower part) and carbonate (upper part), generally being equivalent to the shotori formation. the carbonate part included dolomite inter-bedded with coloured shale. there were few fossils in the sequence. based on the stratigraphic situation, lithological similarities and lithocorrelation, it has been assigned to the anisian age shotori formation (figs. 3 and 4). biostratigraphy apart from the studied section’s base part (related to the permian rock sequence) having few benthic foraminifera, other layers in the studied rock sequence were very poor in terms of fossil content (foraminifera). biological biozoning studies were thus based on conodont elements in the studied section’s complete rock sequence. elements in the studied section were biozoned according to standard kozur biozonation (1977). conodont elements known from permian time would be as follows: hindeodus minutes, streptognathodus barskovi, sweetognathus whitei, neostreptognathodus pequopensis, hindeodus excavates and merrillina divergens. conodont elements known to be related to the triassic age would be as follows: prioniodella prioniodellides, hindeodella multihamata, hindeodella nevadensis, lonchodina nevadensis, hindeodella triassica, ligonodina triassica, diplodella lautissima, pachycladina oblique, hadrodontina anceps, figure 2. the studied area’s geological setting ellisonia triassica, hibardella subsymmertica, parachirognathus geiseri, prioniodella decrescens, enantiognathus bitortus, chirognathus sp., xaniognathus sp., and anchiognathudos sp. (fig. 5). pachycladina oblique (staesch), posterior view, x90, early triassic (early triassic) 1 sweetognathus whitei (rhodes), lateral view, x80, middle permian (guadalupian) 2 hindeodella nevadensis (muller), lateral view, x80, early triassic (early triassic) 3 hindeodus minutes (ellison), lateral view, x80, upper permian 4 hindeodella multihamata (huckriede), lateral view, x95, middle triassic (anisian) 5 chirognathus sp., lateral view, x80, early triassic (scythian) 6 neostreptognathodus pequopensis (behnken), outer view, x80, upper permian 7 ligonodina triassica (muller), lateral view, x80, early triassic (early triassic) 8 enantiognathus bitortus (bender), outer view, x120, early triassic (early triassic) 9 10 hindeodella triassica (muller), lateral view, x80, middle permian lower triassic 11 hindeodus excavatus (behnken), lateral view, x150, middle permian (guadalupian) 12 streptognathus barskovi (kozur), lateral view, x80, middle-upper permian 13 xaniognathus sp., lateral view, x80, lower triassic 14 prioniodella prioniodellides (tatge), lateral view, x120, middle triassic (anisian) 15 lonchodina nevadensis (muller), lateral view, x110, early triassic (early triassic) 16 merrillina divergens (bender & stopple), lateral view, x80, (guadalupian) 17 parachirognathus geiseri (clarck), lateral view, x95, (early triassic anisian) 18 ellisonia triassica (muller), lateral view, x80, early triassic (early triassic). 19 following precise paleontological studies, including conodont recognition and taxonomy regarding samples from the studied section, each fossil’s biozone was identified on the table showing the extent of fossils, based on their presence in each rock sample. two assemblage zones and one acrozone for permian deposits and two assemblage zones for triassic deposits were identified based on the emergence, decline and expanding stratigraphic accumulation index for conodonts having known elements. 4.1. neostreptognathodus pequopensis hindeodus minutes assemblage zone the start of this base’s biozone was consistent with the occurrence of neostreptognathodus pequopensis conodont genus and species and its end with the occurrence of merrillina divergens conodont genus and species. conodont elements contained in this biozone were assigned to the following: hindeodus minutes, hideodus excavatus, sweetognathus whitei, hindeodella triassic and neostreptognathodus pequopensis.figure 3. studied section near haftad gholleh using conodont elements to distinguish permian triassic boundary disconformity near haftad gholleh, central iran figure 4. stratigraphic column of the section being studied near haftad gholleh biozone age was artin-skian, based on the standard kozur biozoning (1977), proposed as being equal to the neostreptognathodus pequopensis biozone. 4.2. merrillina divergens acrozone the base of this biozone started with merrillina divergens and terminated with m. divergens. conodont elements contained in this biozone were identified as follows: merrillina divergens, hindeodella triassic, hideodus excavates. this biozone’s age, according to conodonts, was guadalupian. according to standard kozur biozonation (1977) this zone was the biozone equivalent of merrillina divergens. 4.3. lonchodina nevadensis ellisonia triassica assemblage zone the base of this biozone index was the presence of lonchodina nevadensis and ellisonia triassica and the top of the biozone consisted of pachycladina oblique and hadrontina anceps. conodonts from elements contained in the biozone were assigned to the following: prioniodella prioniodellides, hindeodella multihamat, hindeodella nevadensis, lonchodina nevadensis, hindeodella triassica, ligonodina triassica, diplodella lautissima, pachycladina oblique, hadrodontina anceps, ellisonia triassica, hibardella subsymmertica, parachirognathus geiseri, prioniodella decrescens, enantiognathus bitortus, chirognathus sp. and xaniognathus sp. according to conodonts, this biozone’s age was scythian. based on kozur’s standard biozones (1997), this biozone was the equivalent of the isarcicella isarcica biozone. 4.4. hindeodella multihamata prioniodella prioniodellides assemblage zone this biozone included conodonts such as diplodella lautissima, prioniodella decrescens, hindeodella multihamata, prioniodella prioniodellides, hindeodella triassica and xaniognathus sp. the identified conodonts indicated that it was anisian (fig. 5). conclusions the permian-triassic boundary located near haftad gholleh is an erosional unconformity or disconformity, as in other parts of iran, where figure 5. conodont elements in the studied area mostafa yousefirad, somayeh ghanbari and mahnaz parvanehnejad shirazi figure 6. distribution and biozonation of conodonts from the permian-triassic boundary of the area being studied permian sequences contain dolomitic limestone, shale and sandstone situated below the triassic sequence at a disconformity. based on emergence, decline and expansion of index conodonts, one assemblage zone and one acrozone (permian) and two assemblage zones (triassic) were recognised as follows: neostreptognathodus pequopensis-hindeodus minutes assemblage zone (artinskian); merrillina divergens acrozone (guadalupian); lonchodina nevadensis and ellisonia triassica assemblage zone (early triassic); and hindeodella multihamata, prioniodella prioniodellides assemblage zone) anisian). the age of the uppermost permian sequence thus belonged to the guadalupian era. the triassic sequence consisted of vermiculate limestone layers with coloured shale inter-beds that belonging to the scythian era. the conodont fauna indicated that the permian-triassic sequence from the studied area was deposited in a warm shallow sea having normal salinity. most conodonts in the studied area were euryhaline (i.e. able to tolerate a wide range of salinity). red clastic sediments at the base of the lower triassic indicated that an erosion period started after sedimentation in the midpermian and has been dominant on land in the region. disconformity thus occurred at the permian-triassic boundary in the studied area. acknowledgements this research was supported by arak payame noor university. i would like to thank dr hadi ghaffari and dr mohammad hashem rezai for support and dr babazadeh for reviewing the manuscript and providing pertinent comments. references aghanabati, s.a. (2004). geology of iran, geological survey of iran publisher, tehran, iran, 183 pp. erwin, d.h. (1993). the great paleozoic crisis: life and death in the permian. the perspectives in paleobiolology and earth history, columbia university press, new york, usa, 18 pp. jin, y. wardlaw, b.r., glenister, b.f. kotlyar, g.v. (1997). permian chronostratigraphic subdivisions, episodes, 20, no. 1, 10–15. kodyar, g.v. pronina, g.p. (1995). murgabian and midian stages of the tethyan realm, permophiles, 27, no. 2, 23-26. kozur, h. (1977). beitrage zur stratigraphie des perm. teil i: probleme der abgrenzung und gliederung des perms, freiberger forschungsh, 319, no. 6, 79–121. leven, e.ya. (1992). problems of tethyan permian stratigraphy. international geology review, 34, no. 10, 976-985. raup, d.m. (1991). a kill curve for phanerozoic marine species, paleobiology, 17, no. 1, 37–48. sepkoski, j.jr. (1990). the taxonomic structure of periodic extinction, in: global catastrophes in earth history, sharptonv.l., ward p.d., (eds.), geological society of america special paper, 247, 33–44. using conodont elements to distinguish permian triassic boundary disconformity near haftad gholleh, central iran hua deng1,2,*, yan li1,2, yingchao zhang2, hou zhou1, peipei cheng1, jia wang3, muhammad aqeel ashraf4 1key laboratory of meteorological disaster of ministry of education, nanjing university of information science and technology, nanjing 210044, china 2collaborative innovation center on forecast and evaluation of meteorological disasters, nanjing 210044, china 3inner mongolia meteorological service center, huhehaote 010051, china 4international water, air & soil conservation society 59200 kuala lumpur, malaysia *email of corresponding author: denghua@nuist.edu.cn; dengnll@hotmail.com the forecast of wind energy is closely linked to the prediction of the variation of winds over very short time intervals. four wind towers located in the inner mongolia were selected to understand wind power resources in the compound plateau region. the mesoscale weather research and forecasting combining yonsei university scheme and noah land surface model (wrf/ysu/noah) with 1-km horizontal resolution and 10-min time resolution were used to be as the wind numerical weather prediction (nwp) model. three statistical techniques, persistence, back-propagation artificial neural network (bp-ann), and least square support vector machine (ls-svm) were used to improve the wind speed forecasts at a typical wind turbine hub height (70 m) along with the wrf/ysu/noah output. the current physicalstatistical forecasting techniques exhibit good skill in three different time scales: (1) short-term (dayahead); (2) immediate-short-term (6-h ahead); and (3) nowcasting (1-h ahead). the forecast method, which combined wrf/ysu/noah outputs, persistence, and ls-svm methods, increases the forecast skill by 26.3-49.4% compared to the direct outputs of numerical wrf/ysu/noah model. also, this approach captures well the diurnal cycle and seasonal variability of wind speeds, as well as wind direction. la estimación de la energía eólica está relacionada con la predicción en la variación de los vientos en pequeños intervalos de tiempo. se seleccionaron cuatro torres eólicas ubicadas al interior de mongolia para estudiar los recursos eólicos en la complejidad de un altiplano. se utilizó la investigación climática a mesoscala y la combinación del esquema de la universidad yonsei con el modelo de superficie terrestre noah (wrf/ysu/noah), con resolución de 1km horizontal y 10 minutos, como el modelo numérico de predicción meteorológica (nwp, del inglés numerical weather prediction). se utilizaron tres técnicas estadísticas, persistencia, propagación hacia atrás en redes neuronales artificiales y máquina de vectores de soporte-mínimos cuadrados (ls-svm, del inglés least square support vector machine), para mejorar la predicción de la velocidad del viento en una turbina con la altura del eje a 70 metros y se complementó con los resultados del wrf/ysu/noah. las técnicas de predicción físico-estadísticas actuales tienen un buen desempeo en tres escalas de tiempo: (1) corto plazo, un día en adelante; (2) mediano plazo, de seis días en adelante; (3) cercano, una hora en adelante. este método de predicción, que combina los resultados wrf/ysu/noah con los métodos de persistencia y ls-svm incrementa la precisión de predicción entre 26,3 y 49,4 por ciento, comparado con los resultados directos del modelo numérico wrf/ysu/noah. además, este método diferencia la variabilidad de las estaciones y el ciclo diurno en la velocidad y la dirección del viento. abstract resumen keywords: wind forecast, wrf/ysu/noah, bp-ann, ls-svm palabras clave: predicción del viento, esquema de la universidad yonsei combinado con el modelo de superficie terrestre noah (wrf/ ysu/noah), propagación hacia atrás en redes neuronales artificiales, máquina de vectores de soporte-mínimos cuadrados. prediction of hub height winds over the plateau terrain by using wrf /ysu/noah and statistical forecast record manuscript received: 01/03/2017 accepted for publication: 31/03/2017 how to cite item deng, h., li, y., zhang, y., zhou. h., cheng, p., wang, j., ashraf, m. a. (2017). prediction of hub height winds over the plateau terrain by using wrf /ysu/noah and statistical forecast. earth sciences research journal, 21(1). 37-43. doi:http://dx.doi.org/10.15446/esrj.v21n1.63004 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n1.63004 predicción de vientos en una altiplanicie a la altura del eje con el esquema de la universidad yonsei/modelo superficie terrestre noah y la predicción estadística earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017): 37 43 m e t e o r o l o g y 38 hua deng, yan li, yingchao zhang, hou zhou, peipei cheng, jia wang and muhammad aqeel ashraf 1. introduction the introduction of significant amounts of wind energy into power systems makes accurate wind power forecasting as a crucial element of modern electrical grids. power systems require forecasts with temporal scales of tens of minutes to a few days at wind farm locations. traditionally these forecasts predict the wind at turbine hub heights, and then they are converted into power output predictions at wind farms. since the available wind power is proportional to the wind speed cubed, even small wind forecast errors result in significant power prediction errors. accurate wind forecasts with high temporal scales and spatial resolution at the location of wind farms are significant and essential components in the integration of wind power into power systems. numerous efforts have been undertaken to predict the wind at a hub height of large wind farms using numerical weather prediction (nwp) models (asis et al., 2017; giebel et al. 2011) or statistical models. nwp models focus on the thermodynamic characteristics of the atmosphere coupling with the complex topography and land cover, which have the temporal scale of a few minutes to a day and horizontal spatial scale of a few kilometers to several hundred kilometers (lindang, 2017; pielke, 2002). however, nwp models systematic errors arise from errors in the initial conditions or parameterization. the statistical approach is based on training with measurement data and uses the difference between the predicted and the actual wind speeds in immediate past to tune model parameters, which are more economical regarding computer resources. usually, statistical methods forecast for lead times shorter than 6 hours and the mathematical relationships should be formulated from the past observation as long as possible. therefore, nwp model and statistical post-processing models can be combined to accurately forecast the hub-height winds at wind farms in complex terrain (landberg 2001). the physical models try to use environmental considerations as long as possible to reach the best possible estimate of the local wind speed. statistical models try to find the relationships between nwp results and observations to reduce the remaining error. the physical methods (nwp model) differ widely in their model formulations, spatial and temporal resolutions, and parameterizations. since wind turbines are situated in the planetary boundary layer (pbl), physics options including the pbl scheme and land surface model taking the variation of the wind with height into account are proposed. wind power statistical models, such as calman filter algorithm, artificial neural network (ann, carolin and fernandez 2007; li et al. 2009) and support vector machine (svm, yang et al. 2009; zhang and zeng 2010) models, have been shown to outperform the conventional statistical models as artificial intelligence methods. the combined statistical models, including 2 or more statistical models, overcomes the limitations of one statistical model and improve the accuracy effectively (peng et al. 2011). then, which mesoscale nwp model should be proposed to predict the hub-height winds, how to set up the nwp model, and which post-processing statistical modules could correct the errors, are significant problems and need more study. this study aims to accurately predict the hub heights (70 m) winds at a wind farm on a plateau located in the inner mongolia using a hybrid approach, which combines physical and statistical approach. the weather research and forecasting (wrf) model was used as the mesoscale nwp because wrf is evidenced to have the ability to simulate the near-surface wind even in complex terrain (lai et al., 2017; shimada et al. 2011; deppe et al. 2012). the most widely used planetary boundary layer (pbl) schemes, the yonsei university scheme (ysu), and the noah land surface model were combined with the wrf model. the statistical post-processing models were selected to accurate the nwp model output, including persistence (p), back-propagation artificial neural network (bp-ann) and least square support vector machine (ls-svm) models. the short-term forecasting (day-ahead), immediate-short-term forecasting (6-h ahead) and nowcasting (1-h ahead) were selected to verify the above dynamicalstatistical forecast. this work contributes to a better understanding of the wind conditions and the predictability of the hub-height winds for wind farms. in section 2, we describe the wind data, the wrf/ysu/noah, and the statistical models. section 3 presents the results, which are summarized and discussed in section 4. 2. data and methodology 2.1. data the wind observations are taken from the four meteorological towers (05009, 05010, 05011, 05012; see table a2) on a wind farm in inner mongolia at different height (10 m, 30 m, 50 m, 70 m, and 100 m) with a 10-min time step from 2 january to 31 december in 2010. based on the homogeneity test, the observations from 2 january to 22 march in 05010 and from 17 march to 22 march in 05011 are removed (li et al. 2015). table 1. the four meteorological towers used in this study. 2.2. wrf /ysu /noah the mesoscale model used in this study is the non-hydrostatic wrf model version 3.3.1 (skamarock et al. 2008). there are 34 vertical levels with higher resolution in pbl. the top of the model is 50 hpa, and the lowest is 10 m above ground level (agl). there are 12 levels in the lowest 1 km agl to improve the vertical resolution in pbl. wrf model is configured with four two-way nested domains with horizontal grid spacing of 27 km (d1), 9 km (d2), 3 km (d3), and 1 km (d4), respectively (fig. b1). the outermost domain covers the most of china, the south china sea and some parts of the pacific ocean, which provides the background circulation for the inner grid. the private domain covers the research region to simulate the local circulation system (yu & rahim, 2017). model physics options include the wrf single-moment 5-class (wsm5) scheme for microphysical parameterization (hong et al. 2004), the kain-fritsch cumulus scheme (kain 2004) only for the utmost two domains, the dudhia (1989) shortwave radiation scheme, and the rapid radioactive transfer model (rrtm) for radiation (mlawer et al. 1997). we found pbl schemes and land surface models are important for the reasonable prediction of the near-surface winds. then the ysu pbl parameterization with the monin-obukhow surface layer scheme (hong et al. 2006) and noah land surface model (ek et al. 2003) were selected after some examinations (li et al. 2015). table 2. root mean square error (rmse) for different prediction methods and different forecast lengths to 10 day test period from the 20th to 30th of each month in 2010. (unit: m·s-1) 39prediction of hub height winds over the plateau terrain by using wrf /ysu/noah and statistical forecast the national centers for environmental prediction (ncep) final operational global analysis (fnl) data on 1.0 ×1.0 grids at every six hours are used to initialize the wrf model, which are available on the surface and at 26 pressure levels from 1000 to 10 hpa. the moderate-resolution imaging spectroradiometer (modis) land data is adopted which is more close to the actual land surface features. the wrf output is archived at 10min intervals for analysis. d1, d2, d3, and d4 are activated at 0000 utc, 1200 utc, 1800 utc, and 0000 utc in the next day, respectively. then the model integral time of d1, d2, d3, and d4 are 48 h, 36 h, 30 h, and 24 h, respectively. the first 24 h of d1 runs as a spin-up period. the finest domain simulations are used to forecast the turbine hub winds. this forecast experiment covered the time from 2 january to 31 december in 2010. figure 1. the 3 km and 4 km model domains and topography (shaded, unit: m) used in this study, with a black dot showing the location of four meteorological towers. 2.3. statistical models (a) persistence method persistence is one of the simplest prediction models and most frequently used in wind energy forecasting. in this model, the forecast for all times ahead is set to the value it has now. the mean relative errors of persistence model for the four sites in 1-hour forecast horizon are shown as figure b2, which were calculated from the observed 70-m winds during 2 january to 31 december in 2010. the results indicate persistence has a good forecast in a few minutes to several hours ahead. this is because the dominant time scales of large synoptic scale changes in the atmosphere are in several days; then the pressure systems and winds would be changed on the same time scales. because the relative error of forecast winds is required to be smaller than 20%, the persistence model was used in immediate short-term forecasting and nowcasting. (b) bp-ann the back-propagation algorithm is the most common artificial neural network (ann), and the learning algorithm is the steepest descent algorithm that minimizes the errors between the produced output and the desired output by adjusting the weights. bp-ann is based on the error back-propagation learning algorithm, which has the distributed information storage and processing structure, and is suited for shortterm forecasting (carolin and fernandez 2007; li et al. 2009). threelayered bp-ann can approximate any nonlinear function and shows good performance on handling complex nonlinear problems. two normalized functions were used in this paper. one is premnmx, which is mainly used to normalize the training set, and another is tramnmx, which is used to normalize the prediction set. the anti-normalized function used in this paper is postmnmx. table 3. rmse of wind speeds intervals associated with different forecast methods and different forecast lengths to 10 day test period from the 20th to 30th of each month in 2010. (unit: m·s-1). (c) ls-svm support vector machine (svm) is a statistical learning theory system based on the structural risk minimization principle and vapnikchervonenkis dimension theory (vapnik 1995), which is to map the input data into a higher-dimensional feature space via a kernel function and to construct the optimal hyperplane for classification and regression. lssvm works with a sum squared error cost function, uses the equality constraints instead of inequality constraints in conventional svm and transforms the quadratic programming problem into linear equations, which simplifies and speeds up the calculation and enhances the accuracy of convergence. in this paper, intelligent search of genetic algorithm was employed in the process of choosing parameters of ls-svm, and to find the optimal model parameters of ls-svm for different data sets. compared with the exhaustive search, the performance is improved with less searching times in larger parameter space (rahim & usli, 2017). 2.4. hybrid structure bp-ann and ls-svm were used in this paper as statistical correction model for short-term forecasting. persistence combined with bp-ann and persistence combined with ls-svm were used for immediate short-term forecasting (referred to p6-bp and p6svm, respectively) and nowcasting (referred to p1-bp and p1-svm, respectively). to eliminate the seasonal variation of wind speeds, the statistical relationships are trained for each month. thus the each monthly wind speeds are divided into training (the first 2/3 time series) and validation sets (the last 1/3 time series). the statistical models are trained using the first 2/3 data to learn the relationship between the nwp output wind speeds, and the observations, then use the model parameters to forecast the 70 m winds for the last 10 or 11 days with a time step of 10 min (updates every 10 min). during the training processes, the training vectors are the wrf output wind speeds and the target vectors are the observations, which are both from the 1st to 20th days in each month with a time step of 10 min and the sample number is 2880. during the predicting processes, the testing and target vectors are from the wrf outputs and observations during the 20th to 30th/31th of each month in 1440 sample-numbers, respectively. the combined technique focuses on the variable weight coefficient, which is chosen based on the lowest root mean square error (rmse) of prediction. 40 hua deng, yan li, yingchao zhang, hou zhou, peipei cheng, jia wang and muhammad aqeel ashraf figure 2. the mean relative error (%) of persistence model for the four meteorological towers, which calculated from the observed 70-m winds during 2 january to 31 december in 2010. figure 3. a 10-day time series at 1-hour increments (first 20 days not show due to use as statistical training sets) of annual average wind speed (m·s-1) at four sites associated with seven forecast approaches compared to observational data. (a) 05009, (b) 05010, (c) 05011, (d) 05012 3. evaluation of 70 m wind forecasts to test the hybrid approach for 70 m wind forecasts, the wrf/ ysu/noah wind output, the short-term forecasting (including bp-ann, ls-svm), the immediate short-term forecasting (including p6-bp, p6svm), as well as the nowcasting (including p1-bp, p1-svm) were compared using the rmse as the error measure. four bias-correction approaches were examined: one based on wind speed only, one using wind speed along with the direction, one based on the diurnal cycle of wind speeds, and the last examine the seasonal variability of wind speeds. the test set using only the validation sets from the 20th to 30th of each month in 2010 was used to determine which bias-correction approach resulted in the lowest rmse. figure 4. a 24-hour time series of rmse (unit: m·s-1) for wind speed at 70 m agl associated with different forecast approaches, which calculated from the 20th to 30th of each month in 2010. (a) 05009, (b) 05010, (c) 05011, (d) 05012 3.1. wind speed evolution figure b3 shows the 10 day time series at 1 hour increments of annual average wind speed which were calculated from the 20th to 30th of each month in 2010 based on the wrf/ysu/noah, bp-ann, and ls-svm for short-term forecasting, p6-bp and p6-svm for immediate short-term forecasting, as well as p1-bp and p1-svm for nowcasting. the six forecast approaches were found to have underprediction of the wind speed, and the accuracy of wind speed prediction was decreasing along with prediction time. the wrf/ysu/noah outputs without any post-processing show the lowest average 70-m wind speeds, which indicates that the nwp model outputs combined with statistical techniques are necessary for wind speed forecast. the six hybrid approaches, bp-ann, ls-svm, p6-bp, p6-svm, p1-bp, and p1-svm, forecast the 70-m wind speeds in the rmse of 2.89, 2.46, 2.76, 2.43, 1.75, 1.67, respectively (table a2). the quality of short-term forecasting, including bp-ann and ls-svm, is better than the wrf/ysu/noah model outputs, which improved the day-ahead forecast in rmse of 12.3% and 26.3%, respectively. the immediate short-term forecasting shows a 16.3% and 26.9% improvement over nwp output in 70-m wind speed prediction using p6-bp and p6-svm methods respectively. the nowcasting approaches using p1-bp and p1-svm exhibits a significant degree of improvement in 70-m wind speed forecasting over the other approaches tested, and the lowest rmse is 1.75 and 1.67 m·s-1 respectively. it is noted that the ls-svm technique predicts much better than bp-ann after they were combined with persistence model for the three kinds of 70-m wind forecasting. the available wind power is proportional to the wind speed cubed, hence the power curve is non-linearity. the available wind speed for a turbine is between the cut-in and rated wind speed (3-25 m·s-1). when the wind speed is in the range of 3-8 m·s-1 the wind turbine is gradually running, and the production power rises slowly; when it reaches 8-14 m·s-1, the production power increases rapidly; and if it reaches 14-25 m·s-1, the wind energy is stable. therefore, to understand the different issues involved in 70-m wind forecasting it is useful to divide the wind speed into three distinct groups: (1) 3-8 m·s-1; (2) 8-14 m·s-1; (3) 14-25 m·s-1. table a3 shows the rmse of wind speed groups associated with different forecast methods and different forecast lengths, which indicates the rmse of wind speed prediction is increasing along with stronger wind speed. the improvement in rmse of the 70-m wind speed forecast is 41prediction of hub height winds over the plateau terrain by using wrf /ysu/noah and statistical forecast in the range of 9.4% to 32.6% in short-term forecasting, 12.8% to 33.9% in immediate short-term forecasting, and 39.7% to 53.7% in nowcasting. it should be noted that such forecast technique significantly improves the wind speed forecast in the range of 8-14 m·s-1, which is just associated with the fastest growth of turbine output along with the wind speed. figure 5. the monthly rmse (unit: m·s-1) for wind speeds at 70 m agl associated with different forecast approaches, which calculated from the 20th to 30th of each month in 2010. (a) 05009, (b) 05010, (c) 05011, (d) 05012 figure 6. wind roses of hourly winds (unit: m·s-1) at 70 m agl at four sites associated with different forecast approaches, which calculated from the 20th to 30th of each month in 2010. the wind speed interval is 5 m·s1. (a) 05009, (b) 05010, (c) 05011, (d) 05012 3.2. diurnal cycle of wind speeds figure b4 shows a 24-hour time series of rms for wind speed at 70 m agl associated with different forecast approaches in the training sets. the mountainous terrain of the inner mongolia plateau is complex with a strong turbulence, then the daily variation of the wind speed is different for different sites. but even that, all physical-statistical approaches exhibit a significant improvement in wind forecast and present the similar error in the diurnal cycle. the wrf/ysu/noah outputs have the largest error in rmse, and the nowcasting have the lowest. for the short-term forecasting, wrf/ysu/ noah coupling bp-ann and ls-svm has the similar bias-correction on the station 05009, and 05010, whose error curve in rmse is consistent with the cure of wrf/ysu/noah output. but for the station 05011 and 05012, the forecast ability of wrf/ysu/noah coupling ls-svm is better than coupling bp-ann obviously. the immediate short-term wind speed forecastings also develop the forecast error compared to the short-term forecasting, and the two error cures (p6-bp, p6-svm) are almost the same. 3.3. seasonal variability of wind speeds to test the forecast skill on seasonal variability, the monthly rmse for 70-m wind speeds of different forecast approaches is shown in figure b5. what the results also agree with the conclusion that the accuracy of wind speed prediction is decreasing along with prediction time and the now casting is better than other time-scale forecasts (shuib, et al., 2017). the three timescale wind speed prediction in rmse is between 1.2 to 4.9 m·s-1. also, the accuracy of wind speed prediction is smaller in autumn and larger in spring and summer. the relative largest rmse is in july and the smallest in october. furthermore, for the same time-scale forecasts, the method using ls-svm forecast wind speed is better than that using bp-ann. 42 hua deng, yan li, yingchao zhang, hou zhou, peipei cheng, jia wang and muhammad aqeel ashraf 3.4. wind speed rose diagram the wind direction is an important factor to be considered in wind power generation. the observation and forecast wind data were averaged over the 15day test period to generate the wind speed rose, plotted at a 22.5-degree angular resolution and in 5 m·s-1 intervals in wind speed, illustrated in figure b6. the value represents the average wind speed in the 16 different wind directions. as can be seen that all the wind speed roses illustrate a valley in the about northerly direction. the wrf/ysu/noah outputs exhibit the lower 70-m wind speed at the four sites than other forecasts (roslee, et al., 2017; ridzuan et al., 2017). the three time-scale forecasting successfully forecast the westerly (from ssw to nnw) speed at 05012, and the nowcasting shows a significant degree of improvement in wind speed forecasting at all four sites. the error of immediate short-term forecasting is between nowcasting and short-term forecasting. although the different approaches have different improvement on the different stations, the hybrid forecasting with physical forecasting and statistical postprocessing could predict the wind much better. 4. summary the wind energy forecasting problem is closely linked to the problem of forecasting the variation of winds over very short time intervals because the wind is variable and intermittent over various time-scales. to understand the different issues involved in 70-m wind forecasting, the four wind towers in the inner mongolia were used to represent the complex plateau terrain, and it was useful to divide the problem into three difference time scales: short-term, immediate-short-term, and nowcasting. the mesoscale wrf/noah/ysu as the nwp model performed 24-h wind forecast with 1-km horizontal resolution and 10-min time resolution. bp-ann and ls-svm were used to short-term forecasting. persistence method was combined with bp-ann and ls-svm to forecast the immediate short-term forecasting and nowcasting, respectively. the results indicate that the current forecasting techniques exhibit considerable skill in short-term, immediate-short-term, and nowcasting forecasting. short-term forecasts typically outperform an nwp forecast by 12% to 26%; nowcasting forecasts usually outperform nwp forecasts by 47% to 50%. this paper also demonstrates that the approaches using ls-svm model is more accurate than the approaches using bp-ann model over the different time-scales tested. the possible reasons are as follows: 1, svm is proposed for problems with limited samples to obtain the global optimum solution. 2, svm algorithm can transform the problem into a quadratic programming problem to obtain the global optimum solution which bpann cannot achieve. 3, svm uses the nonlinear transformation to map the original variables to the higher dimensional feature space to construct the linear classification function, thus ensuring the good generalization ability of the model, and then avoids the curse of dimensionality. acknowledgments this work is supported by national natural science foundation of china (40805039, 20110138, 41575056, and 41675156), key laboratory of meteorological disaster of ministry of education program (klme1303) in nanjing university of information science and technology, the research innovation program for college graduates of jiangsu province under grant no. sjzz16_0155 and sjzz16_0156, national public welfare industry (meteorological) scientific research, no. gyhy20110604, the open research fund of chongqing meteorological bureau, and the priority academic program development of jiangsu higher education institutions (papd). references asis, j., tahir, s. h., rahim, a. r., konjing, z., kob, r. c., & tjia, h.d. (2017). smaller benthic foraminifera analysis of kudat formation, kudat, sabah: preliminary interpretation. geological behavior, 1(1) 27-29. carolin, m. m., & fernandez, e. (2008). analysis of wind power generation and prediction using ann: a case study. renewable energy, 33(5), 986-992. doi: http://doi.org/10.1016/j.renene.2007.06.013 deppe, a. j., gallus, w. a., & takle, e. s. (2013). a wrf ensemble for improved wind speed forecasts at turbine height. weather and forecasting, 28, 212–228. doi: 10.1175/waf-d-11-00112.1 dudhia, j. (1989). numerical study of convection observed during the winter monsoon experiment using a mesoscale two-dimensional model. journal of the atmospheric sciences, 46(20), 3077–3107. ek, m. b., mitchell, k. e., lin, y., rogers, e., grunmann, p., koren, v., gayno, g., & tarpley, j. d. (2003). implementation of noah landsurface model advances in the national centers for environmental prediction operational mesoscale eta model. journal of geophysical research, 108, 8851. doi:10.1029/2002jd003296. giebel, g., brownsword, r., kariniotakis, g., & denhard, d. c. (2011). the state of the art in short-term prediction of wind power: a literature overview. 2nd edition. project report for the anemos and safewind projects, ris, dtu. hong, s. y., & noh, y. (2006). a new vertical diffusion package with an explicit treatment of entrainment processes. monthly weather review, 134, 2318–2341. doi: http://dx.doi.org/10.1175/mwr3199.1 hong, s. y., dudhia, j., chen, s. h. (2004). a revised approach to ice microphysical processes for the bulk parameterization of clouds and precipitation. monthly weather review, 132, 103–120. doi: http:// dx.doi.org/10.1175/1520-0493(2004)132<0103:aratim>2.0.co;2 kain, j.s. (2004). the kain-fritsch convective parameterization: an update. journal of applied meteorology, 43, 170–181. doi: 10.1029/2009wr007704. lai, g. t., razib, a. m., mazlan, n. a., rafek, a. g., serasa, a. s., … & mohamed, t. r. (2017). rock slope stability assessment of limestone hills in northern kinta valley, ipoh, perak, malaysia. geological behavior, 1(1), 22-26 landberg, l.short-term prediction of local wind conditions journal of wind engineering and industrial aerodynamics, 89(3-4), 235–245. doi: http://doi.org/10.1016/s0167-6105(00)00079-9 li, y., cheng, p., lu, y., & song, y. (2015). wind power forecasting over the gypical complex terrains (in chinese). plateau meteorology, 34(2), 413–425. doi: 10.7522/j.issn.1000-0534.2013.00181 li, x., liu, y., & xin, w. (2009). wind speed prediction based on genetic neural network. ieee conference on industrial electronics and applications, beijing-china. 2448–2451. doi: 10.1109/iciea.2009.5138642 lindang, h. u., tarmudi, z. h., & jawan, a. (2017). assessing water quality index in river basin: fuzzy inference system approach. malaysian journal of geosciences, 1(1) 27-31. mlawer, e. j., taubman, s. j., brown, p. d., iacono, m. j., & clough, s. a. (1997). radiative transfer for inhomogeneous atmospheres: rrtm, a validated correlated-k model for the longwave. journal of geophysical research, 102, 16663–16682. doi: 10.1029/97jd00237 peng, h., liu, f., & yang, x. (2011). short term wind speed forecast based on combined frediction. acta energiae solaris sinica, 32(4), 543–547. pielke, r.a.s. (2002). mesoscale meteorological modeling. volume 78 of international geophysics series. academic press, 2nd edition, 676 pags. rahim, i. a., & usli, m. n. r. (2017). slope stability study around kampung kuala abai, kota belud, sabah, malaysia. malaysian journal of geosciences, 1(1), 38-42. ridzuan, a. a., zahar, u. a., noor, n. a. (2017). association of evacuation dimensions towards risk perception of the malaysian students who studied at jakarta, medan, and acheh in indonesia. malaysian journal of geosciences, 1(1) (2017) 06-11 roslee, r., simon, n., tongkul, f., norhisham, m. n., & taharin, m. r. (2017). landslide susceptibility analysis (lsa) using deterministic model (infinite slope) (dessism) in the kota kinabalu area, sabah, malaysia. geological behavior, 1(1) 06-09. 43prediction of hub height winds over the plateau terrain by using wrf /ysu/noah and statistical forecast shimada, s., oshawa, t., chikaoka, s. (2011). accuracy of the wind speed profile in the lower pbl as simulated by the wrf model. sola, 7, 109–112. doi: http://doi.org/10.2151/sola.2011-028 shuib, m. k., manap, m. a., tongkul, f., rahim, i. b., jamaludin, t. a., surip, n., … ahmad, z. (2017). active faults in peninsular malaysia with emphasis on active geomorphic features of bukit tinggi region. malaysian journal of geosciences, 1(1), 12-24. skamarock, w. c., klemp, j. b., dudhia, j., gill, d. o., barker, m. g., duda, x. y., huang, d. m., wang, w., & powers, j. g. (2008). a description of the advanced research wrf version 3. technical report, ncar tech notes-475+str. vapnik, v.n. (1995). the nature of statistical learning theory. new york, springer-verlag, 1995, 138–170. yang, x., cui, y., zhang, h., & tang, n. (2009). research on modeling of wind turbine based on ls-svm. international conference on sustainable power generation and supply, nanjing, china, 1–6. doi: 10.1109/supergen.2009.5348180 yu., l. k., & rahim, i. a. (2017). application of gis system for slope stability studies on selected slopes of the crocker formation in kota kinabalu area, sabah. geological behavior, 1(1) 10-12 zhang, h., & zeng, j. (2010). wind speed forecasting model study based on support vector machine. acta energiae solaris sinica, 31(7), 928–932. mine water is a menace to coal mining. mine water quantity is detected by drilling in the coal mine roadway, and the disadvantage is the high workload and low efficiency. therefore, transient electromagnetic method (tem) was proposed, and tem data was processed to detect the water yield property of a targeted layer in the coal mine. through a mine-oriented 3d transient electromagnetic method observing system, the signal of induced voltage is obtainable. transient electromagnetic method 3d data volume can be calculated through the calculation of all-time resistivity and time-depth conversion. after an appropriate apparent resistivity value is set, the spatial distribution range of an aquifer can be determined. then, with the water-filling coefficient of the aquifer, its water quantity can be estimated. the water yield property detection results in the no.4 coal seam goaf of the no.80101 workface in jude mine of shanxi, china, demonstrates that the apparent resistivity of this goaf is less than 3 ω.m, and the projection area of low resistivity anomaly zone is 22,383 m2. by using the formula q=kms, we can estimate that the water volume is 33,574 m3. three boreholes have been constructed for the next dredging and drainage project, which results in a total water yield of 33,089 m3. the error percentage of the predicted water quantity is less than 1.5%. it can thus be concluded that it is feasible to predict aquifer water content with tem data. abstract keywords: transient electromagnetic method, apparent resistivity calculation, time-depth conversion, water quantity quantitative calculation of aquifer water quantity using tem data cálculo cuantitativo del volumen de aguas subterráneas a través del método electromagnético transitorio (tem) issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n1.63002 earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017): 51 56 lanying huang1, shengdong liu1, bo wang1*, fubao zhou2 1 state key laboratory of deep geomechanics & underground engineering and school of resource and earth science, china university of mining and technology, xuzhou 221116, china 2 school of safety engineering, china university of mining and technology, xuzhou 221116, china email of corresponding author: wbsyes@126.com el agua al interior de las minas es una amenaza para la minería de carbón. la cantidad del agua en las minas se detecta al perforar en las vías del socavón, con la desventaja de la carga de trabajo y la baja eficacia que significan. por esto, el presente trabajo propone el método electromagnético transitorio (tem, del inglés transient electromagnetic method) y utiliza la información generada para detectar las propiedades de producción de agua en una capa específica de una mina de carbón. a través de un sistema de observación tridimensional tem orientado a minas se puede obtener la señal del voltaje inducido. la capacidad de la información del sistema tem tridimensional se puede calcular al obtener la resistividad y la conversión tiempo-profundidad. tan pronto se establece el valor apropiado de la resistividad aparente se puede determinar el rango de distribución de un acuífero. luego, con el coeficiente de saturación del acuífero se puede calcular la cantidad de agua. los resultados de la detección de las propiedades de producción de agua en la mina abandonada número cuatro de frente de trabajo número 80101 de la mina jude, en shanxi, china, demostró que la resistividad aparente de esta excavación es de 3 ω.m, y el área de proyección de la zona anómala de baja resistividad es 22.383 m2. al usar la fórmula q=kms, se estimó que el volumen de agua es de 33,574 m3. tres pozos se han construido para el próximo proyecto de dragado y drenaje, cuyos resultados de producción de agua es de 33,089 m3. el porcentaje de error de la predicción de la cantidad de agua es menor al 1.5 %. se concluye, por ende, que es factible producir el contenido de agua de un acuífero con la información del tem. resumen palabras clave: método electromagnético transitorio, cálculo de la resistividad aparente, conversión tiempo-profundidad, cantidad de agua. record manuscript received: 01/03/2017 accepted for publication: 31/03/2017 how to cite item huang, l., liu, s., wang, b.,* zhou, f. (2017). quantitative calculation of aquifer water quantity using tem data. earth sciences research journal, 21(1). 51-56. doi:http://dx.doi.org/10.15446/esrj.v21n1.63002 h y d r o g e o l o g y 52 lanying huang, shengdong liu, bo wang and fubao zhou 1. introduction china is the largest producer of coal in the world because coal plays an essential role in the economic development of china (figure 1). coal mining accidents have become a frequent occurrence in recent years (zhang et al., 2014; wang et al., 2016). although the death rate resulting from these accidents in china has declined, it is still 70 times higher than that in the united states and 17 times higher than that in south africa. against this backdrop, it is urgent to study how to find effective ways to detect water quantity in the aquifer and reduce coal mining accidents (zhang et al., 2015). among some geophysical methods currently available, tem technology is one that has received much attention and has been widely used to reduce coal mining accidents. figure 1. global coal production structure and energy consumption ratio in china. in a tem application, an ungrounded loop-line first transmits downward a pulse-type primary field, which induces an eddy current. the spatial and temporal distribution of a secondary field caused by the eddy current can then be observed through the coil (yu, 2007; wang et al., 2011; tanaka and kunisada 2011). by measuring the variation rule of the secondary magnetic field over time during the timeoff period, the geoelectricity features at different depths can be obtained (guillemoteau et al., 2011; tuncer et al., 2014). from figure 2 it can be seen that tem receives induced signals during the time-off period, which will be attenuated over time. by using the tem method, we can solve some geological problems (cheng et al., 2013; hu et al., 2013). due to its features such as light equipment, little lateral influence, and high resolution, tem can be widely applied in water disaster prevention and control (zhang et al., 2010; mollidor et al., 2013; xue et al., 2013; tao et al., 2013). however, because of the focus on theory development and processing method restraints, past analyses have usually been qualitative, which cannot meet the standard of quantification, and, to certain extent, hinders the application of tem on a wider basis(xu et al., 2012; wang et al., 2014). hence, in this study, we will develop a theory and propose a method to calculate tem responses. to do so, we will use equivalent substitution, interpret the diffusion law of tem field from the perspective of physics, calculate physical parameters, and conduct 3d apparent resistivity imaging and quantitative analyses of detection results. figure 2 schematic diagram of tem survey. 2. tem data processing 2.1. apparent resistivity calculation the transmitting loop center induction electromotive force of horizontal round loop in uniformity whole space medium is: (1) (2) (3) where is the vacuum permeability, n is the turn number of transmitting loop, i is the respective emission current, s is the receiving loop similar area, 0r is the transmitting loop radius, t is the observation time and is the true resistivity of the medium. from equation (3), we can obtain an equation as below: (4) if the loop forms a square about b meters on a side, and , then, with equations (1) ~ (4), the all-time apparent resistivity can be calculated with the binary search algorithm. the calculation process is illustrated in figure 3. for any observation window t, u is set as [0,10], and the searching region of u can be seen as an absolute value, and among them, us is the initial value of the searching region, ue is final value, ue is a median value which is u value. u is calculated by a binary search algorithm. is actual measured magnetic value, and h(ti ) is: 2.2. depth conversion the detection depth d of the tem method is related to transmitting magnetic torque, all-time apparent resistivity, and the minimum resolutive voltage. the time-depth conversion equation for mine-oriented tem method is: (5) (7) (8) (9) (10) (6) 53quantitative calculation of aquifer water quantity using tem data figure 3. the solving procedure of apparent resistivity. where is the conversion coefficient, which is about the heterogeneous conductor in the vicinity of the roadway and usually values between 0.6 and 1.8; is the factor of proportionality and , where is the corresponding resistivity value of moment it . (11) 2.3. interpolation operation due to the unstable workload, the measured and computational data of experiment and field are always not accurate enough, so interpolation is introduced as a supplement. ( , , )k k kk x y z is an arbitrary point in the three-dimensional space. there is given point which is the vicinity of k.the inverse distance weighting method can be used to attribute the value xk with the equation: where id is the distance between the interpolating point and the given point in the vicinity. yk and zk can be calculated in the same way as xk . 2.3. 3d slice extraction 3d visualization is a technique to reveal and describe spatial data, receive the subsurface geological structure and features, which provides information for accurate description of the 3d geological structure, and facilitate the exploration and development of coal mine. according to the data received from different detection directions, it can be seen that 3d visualization can clearly display the spatial distribution of anomalies with different resistivity values that correspond to different threshold values, the scattered points, and the range of apparent resistivity. however, a cross-section map can be clearer in showing the anomaly areas at a particular depth on a profile. cross-sections can be extracted from different depths of plane xy, yz, and zx. apparent resistivity data can be extracted from any depth or vertical profile. 2.4. aquifer water quantity estimation through the cross-section map, the aquifer water volume can be calculated with the following formula: (12) in this equation, q is the aquifer water quantity, k is the water containing coefficient, m is the aquifer thickness, and s is the water-rich area in the aquifer. 3. field application 3.1. geological setting the no.80101 workface of jude mine in shanxi province is under l1 limestone in the middle and lower part of taiyuan formation, 60.6m beneath the 4# seam. the seam occurrence is stable with a thickness range of 1.8 to 3.2m, and a 2.8m thickness on average. part of the area contains a 0.05 to 0.1m dirt band of mudstone or carbonolite (carbonaceous mudstone). at the bottom of the seam is a pyrite seam in forms of lamella and nodule. the lithological column is shown in figure 4. 3.2. field detection to detect the spatial distribution of the goaf and estimate the water quantity, mtem was adopted with the ycs360 electromagnetic system (as shown in figure 5). the detection of the no.80101 workface starts at the intersection of the return airway and the concentrated rail roadway, then goes on from the return airway to the openoff cut, and ends at the intersection of intake airway and concentrated rail roadway. the total detection workload is 3470 m, including 1660 m along the return airway, 150 m in the openoff cut, and 1660 m along the intake airway. monitoring points are set every 10 meters, and the detection is conducted in 6 directions, including 60°, 45°, and 30° to the external wall roof, standard to the ceiling, and 30° and 45° to the interior wall roof. the tem detection direction diagram is shown figure 4. lithological column. 54 lanying huang, shengdong liu, bo wang and fubao zhou figure 4. lithological column. figure 5. ycs360 electromagnetic system. figure 7. arrangement diagram of mtem. table 1. drilling log table of anomaly yc7 in no.80101 workface figure 6. tem detection direction diagram. in figure 6, and the arrangement diagram of mtem is shown in figure 7. 3.3. data interpretation and result analysis the data is processed based on the standard flow chart (figure 8), and data interpretation is carried out from the perspective of geophysical characteristics. the features are shown in 3d images. water quantity is higher in areas where the fractures are concentrated, indicated by higher electrical conductivity and lower resistivity (i.e., high potentiallow resistivity) anomalies. this characteristic identifies aquifer areas as those where roof resistivity is less than 3ω.m. figure 9 is a 3d spatial distribution graph of anomalies at different depths above the no.80101 workface. 40 meters above the roof is a water-rich stratum, and key preventions should be focused on region yc1, yc7, and yc10. anomaly area yc7 is located between 1085m and 1850 m along the return airway (between monitoring point kf13 to kf15), 0-120m along the inclination slope (120m within return airway and workface), and 40m to 60m above the roof (more details in fig 9), with an anomaly-impacted area of 22,383 m2. the goaf water-filling coefficient is 0.25 to 0.35, and the detection result of the actual ratio is 0.3. it is indicated by the effect sketch that m is 5m, based on the equation q=kms. thus we can deduce that q=33,574m³. the verification of the calculated values based on field measurements is shown in table 1. h192, h19-3 and h19-s1 boreholes have been constructed for the later next and drainage project, which results in a total water quantity of 33,089m3, and the error percentage of the predicted water volume is less than 1.5%. 55quantitative calculation of aquifer water quantity using tem data figure 8. processing flow chart figure 9. the 3d spatial distribution of anomalies at different depths above the no.80101 workface. 56 lanying huang, shengdong liu, bo wang and fubao zhou 4. conclusions tem resistivity imaging boasts rapid imaging, high quantification level, and high resolution, especially when it comes to detecting low resistivity body with a high-resistivity overlayer. with the projection area s of low-resistivity anomaly zone, the water containing coefficient k and the aquifer thickness m, the aquifer water quantity can be estimated with an error percentage of the predicted water quantity below 2%. the tem 3d presentation and analysis has offered a new method to determine the water amount in an aquifer. acknowledgments this research has been performed under national natural science foundation project (grant no. 41604082, u1261202), and the fundamental research funds for the central universities (2014xt02), and a project funded by the priority academic program development of jiangsu higher education institutions. references cheng, j. l., qiu, h., ye, y. t., yan, g. c., zhou, j., cheng, f. b., & zhang, s. s. (2013). research on wave-field transformation and data processing of the mine transient electromagnetic method. journal of the china coal society, 38(9), 1646-1650. guillemoteau, j., sailhac, p. & béhaegel, m. (2011). regularization strategy for the layered inversion of airborne transient electromagnetic data: application to in-loop data acquired over the basin of franceville (gabon). geophysical prospecting. journal of immunoassay, 59(6), 1132-1143. hu, x. w., zhang, p. s., cheng, h., wu, r. x., & guo, l. q. (2013). quantitative assessment of interference induced by roof bolt during advanced detection with transient electromagnetic method in mine. journal of rock mechanics and engineering, 32, 3275-3282. lin, z., ge, s., li, d., & peng, w. (2015). structure characteristics of acidic pretreated fiber and self-bind bio-boards for public health. journal of pure and applied microbiology 9, 221-226. liu, z. l. (2014). the groundwater pollution and environmental protection in china. biotechnology: an indian journal, 10, 1883-14886. liu, z. x., liu, x. c., & liu, y.g. (2009). research on transient electromagnetic field of mine water-bearing structure by physical model experiment. journal of rock mechanics and engineering, 28(2), 259-266. mollidor, l., tezkan, b., bergers, r., & löhken, j. (2013). float-transient electromagnetic method: in-loop transient electromagnetic measurements on lake holzmaar, germany. geophysical prospecting, 61(5), 1056-1064. tao, f., li, w. g., wang, p., & an, s.p. (2013). research on fine interpretation for water containment of coal mine rock strata by mt imitated tem depth inversion method. journal of the china coal society, 38(z1), 129-135. tanaka, y., & kunisada, e. (2011). study on meshless method using rpim for transient electromagnetic field. ieee transactions on magnetics, 47(5), 1178-1181. tuncer, o., shanker, b., & kempel, l. c. (2014). a hybrid vector generalized finite-element method for transient electromagnetic simulations. electromagnetics, 34 (3-4), 286-297. wang, b., liu, s. d., liu, j., huang, l. y., & zhao, l. g. (2011). advanced prediction for multiple disaster sources of laneway under complicated geological conditions. international journal of mining science and technology, 21, 749-754. wang, b., liu, s. d., lu, t., & sun, h. l. (2014). coal seam thickness detection in mine roadway by using advanced prediction method. electronic journal of geotechnical engineering, 19, 4753-4762. wang, b., liu, s. d., zhou, f. b., lu, t., huang, l. y., & gao, y. j. (2016). polarization migration of three-component reflected waves under small migration aperture condition. acta geodynamica et geomaterialia, 13(1), 1-12. xue, g. q., cheng, j. l., zhou, n. n., chen, w. y., & li, h. (2013). detection and monitoring of water-filled voids using transient electromagnetic method: a case study in shanxi, china. environmental earth sciences, 70(5), 2263-2270. xu, j. p., liu, s. d., & wang, b. (2012). electrical monitoring criterion for water flow in faults activated by mining. mine water and the environment, 31(31), 172-179. yu, j. c. (2007). mine transient electromagnetism prospecting. china university of mining and technology press, xuzhou. zhang, s. f., meng, l. s., & du, x. j. (2010). transient electromagnetic method to investigating potential safety hazard of mine-out area in tailings pond of a gold mine. journal of jilin university (earth science edition), 40(5), 1177-1182. zhang, w., zhang, d. s., wang, h. z., & cheng, j. x. (2015). comprehensive technical support for high-quality anthracite production: a case study in the xinqiao coal mine, yongxia mining area, china. minerals, 5(4), 919-935. zhang, w., zhang, d. s., wu l. x., & wang h. z. (2014). on-site radon detection of mining-induced fractures from overlying strata to the surface: a case study of the baoshan coal mine in china. energies, 7(12), 8483-8507. search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 small fault ahead of the tunnel face in coal roadway is the important hidden hazard factor of coal and gas outburst accidents. the study of small fault prediction has important practical significance, which is the urgent demand of coal mine safety production. the diffraction of breakpoint can be used to identify the fault. however, unlike surface seismic exploration, the diffraction is with approximately horizontal incidence when the advanced detection is carried out in the roadway. the common advanced detection system is mainly as the reference of traffic tunnel, without considering the influence of low-velocity coal seam. considering the influence of an acoustic wave of the roadway cavity and channel wave of the coal seam, the advanced detection model of small fault ahead of tunnel face is established. diffraction advanced observation system in which sources located in front of tunnel face is constructed, and the numerical calculation of the high-order staggered-grid finite difference is carried out. the simulation results show that: compared with the data collected by reflection observation system, in seismic records acquired by diffraction observation system, the suppression effect of acoustic wave is appeared. the diffracted p-wave of the breakpoint of component x is clear with strong energy and short-wave group. multiple diffractions of the breakpoint are not found, but the multiple diffraction of tunnel face endpoint is obvious. the difference between breakpoint diffraction and multiple diffractions of the endpoint is clear, and diffracted p-wave of the breakpoint is easy to identify. the multiple reflected channel wave between the fault and the tunnel face is very obvious, and the reflected channel wave of small fault is so hard to identify. migration results show that the imaging resolution of diffracted p-wave of small fault is higher than the reflected channel wave, and breakpoint location of imaging is consistent with the actual model. abstract keywords: coal roadway; small fault; diffraction; reflected channel wave diffraction characteristics of small fault ahead of tunnel face in coal roadway issn 1794-6190 e-issn 2339-3459 earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017): 95 99 bo wang1, shengdong liu1*, fubao zhou2, jun zhang1, fangkun zheng1 1 state key laboratory of deep geomechanics & underground engineering and school of resource and earth science, china university of mining and technology, xuzhou 221116, china 2 school of safety engineering, china university of mining and technology, xuzhou 221116, china *email of corresponding author: wbsyes@126.com la falla pequeña en la calzada de carbón de un túnel es un factor importante no visible en los accidentes por explosión en minas de gas y carbón. el estudio de la predicción de falla pequeña tiene un importante sentido práctico: la demanda urgente de seguridad en la producción carbonífera. el punto de quiebre de la difracción puede utilizarse para identificar la falla. sin embargo, al contrario que la exploración sísmica superficial, la difracción se acerca a la prevalencia horizontal cuando se realiza la detección avanzada en la calzada. el sistema común de detección avanzada se usa principalmente para referenciar el tráfico del túnel, sin considerar la influencia de la baja velocidad en la veta de carbón. al valorar la respuesta de la onda acústica en la cavidad de la calzada y la onda de canal de la veta de carbón se establece el modelo de detección avanzada de pequeña falla para la calzada del túnel. se construyó el sistema de observación avanzada de difracción en el cual las fuentes se localizan en frente de la cara del túnel y se realizó el cálculo de diferencia finita en una red escalonada de orden alto. los resultados del modelo muestran que a diferencia de la información recolectada con el sistema de observación de reflexión, en los registros sísmicos adquiridos con el sistema de observación de difracción se puede ver el efecto de supresión de la onda acústica. el punto de quiebre de la onda p difractada para el componente x es claro, con energía fuerte y en el grupo de onda corta. no se encontró el punto de quiebre para difracciones múltiples pero es evidente la difracción múltiple para el punto final de la cara del túnel. es clara la diferencia entre el punto de ruptura de la difracción y las difracciones múltiples del punto final, mientras el punto de ruptura de la onda p difractada es fácil de identificar. los resultados de migración muestran que la resolución de imágenes de la onda p difractada de falla pequeña es mayor que la onda de canal reflejada y la ubicación del punto de quiebre de la imagen es consistente con el modelo actual. resumen palabras clave: calzada de carbón; pequeña falla; difracción; onda de canal reflejada. record manuscript received: 15/05/2017 accepted for publication: 30/06/2017 how to cite item: wang, b., liu, s., zhou, f., zhang, j., & zheng, f. (2017). diffraction characteristics of small fault ahead of tunnel face in coal roadways. earth sciences research journal, 21(2), 95 99. doi:http://dx.doi.org/10.15446/esrj.v21n2. 64938 características de difracción de falla pequeña en la calzada de carbón de un túnel http://dx.doi.org/10.15446/esrj.v21n2.64938 s e is m ic p r o c e ss in g 96 wbo wang, shengdong liu, fubao zhou, jun zhang, fangkun zheng 1 introduction in china, the deaths from 2012 to 2016 was 451 because of coal mine gas accidents (liu et al., 2016), so coal mine gas prevention and control are very grim, and gas control is the most important work of coal mine safety production. most accidents of coal and gas outburst in tunnel face of coal roadway occur near the geological structure (shepherd et al., 1981; lama et al., 1998), especially small fault (gao et al., 2015). this paper focuses on the research of small fault ahead of the tunnel face, which is the important hidden factor of coal and gas outburst accident. it is of great practical significance to the prevention and control of gas accidents. at present, there are a lot of advanced geophysical exploration methods adopted in the roadway, which can be divided into three classes: seismic class, electromagnetic class, and other class (roslee et al., 2017). at present, the seismic wave method is less affected by the detection environment, which is most suitable for the prediction of geological structure (lüth et al., 2006; jetschny et al., 2011). the seismic wave method of advanced detection mainly includes the reflected wave method, surface wave method, scattered wave method, channel wave method, diffraction method and so on (wang et al., 2016). the advanced detection by using reflected wave method is started early in the field of traffic tunnel. according to the launching order, seismic advanced detection technology includes: hsp, tvsp, tsp, trt, isis, tgp, tsd, usp, tswd, etc. (inazaki et al., 1999; otto et al., 2002; lüth, et al., 2005; jetschny et al., 2010). compared with the advanced tunnel prediction, the research on reflected wave in coal roadway is few, including rtsp, msp, etc. (wang et al., 2015). in the aspect of advanced detection of surface wave and scattered wave in underground coal mine, the advanced detection technique of rayleigh surface wave was carried out in fangezhuang mine (cheng et al., 2014). zhao et al. (2006) obtained higher image positioning accuracy based on inverse scattered imaging theory. cheng et al. (2013) studied the seismic scattered wave imaging in coal roadway by using numerical simulation method. in the aspect of reflected channel wave, essen et al. (2007) analyzed the response characteristics of rayleigh channel wave of split coal through numerical simulation; yang et al. (2012) carried out numerical simulation research with seismic wave field of small structure, and then pointed out that the rayleigh surface wave generated from the reflection of rayleigh channel wave can be used as the characteristic wave of the advanced detection of small structure. lu et al. (2013) obtained the reflected channel wave by using tunnel boring machine as the source to carry out the advanced detection survey with a vertical fault with fault throw of 8m. in the aspect of the diffraction, yang et al. (2010) analyzed the wavefield of the advanced detection model of the geological interface, and then pointed out that the diffraction was the powerful wave to identify the interface. deng (2012) pointed out that the diffraction of breakpoint due to the large-scale fault was the effective wave to detect the position of the coal seam. however, the studies on advanced detection in the underground coal roadway do not include the study of diffraction characteristics of small fault; therefore, the characteristics of diffraction wavefield of small fault are studied in this paper through the numerical simulation, so as to solve the fine imaging of small fault ahead of tunnel face. 2. principle when a seismic wave meets coal seam breakpoints, these discontinuity points can be regarded as the new sources, which can produce a kind of new disturbance to propagate around the elastic medium (lindang et al., 2017; kamsani et al., 2017). this wave of disturbance is called diffraction. in essence, the physical premise of diffraction generation is the wave impedance difference and minor geological structure. the diffraction follows a concept of full wavelength, and its propagation is in agreement with the “huygens fresnel” principle. the relation between the incident wave and the diffracted wave is one-to-many, while the relation between the incident wave and the reflected wave is that one-to-one, as shown in figure 1. the reflected wave is generated when the seismic wave encounters the medium of sudden change in the propagation process, which is the change of travel time and amplitude caused by large-scale heterogeneity lai et al., (2017). because the fault throw in coal roadway is less than 10m, it is hard to form the reflection of the fault. however, due to the small scale of the fault and the appear wave impedance difference between coal seam and surrounding rock (the average density and p-wave velocity of coal seam and surrounding rock are respectively as: ρ1=1400kg/m3, v1=2000m/s; ρ2=2800kg/m3, v2=3500m/s), the obvious diffraction could be formed in the process of seismic advanced detection in coal roadway, as shown in figure 2. figure 1. diagram of reflection and diffraction. figure 2. diffraction of advanced detection. 3. high-order staggered-grid finite difference in general, the structure of coal seam is layered. coal seam, layered roof, and floor can be regarded as horizontal isotropic media (lai et al., 2017). the equation of the acoustic wave propagating in two-dimensional isotropic media is as follows: (1) where p = p(x,z,t) represents acoustic wave field, v = v(x,z,t) represents velocity field. the difference scheme of time second-order and space 2nth-order difference accuracy is as follows: (2) the source is the spike pulse, and ricker wavelet is used as spike pulse source and time domain formula of ricker wavelet is as follows: (3) where a is the amplitude, f0 is the dominant frequency of wavelet, t0 is the beginning time? 97diffraction characteristics of small fault ahead of tunnel face in coal roadway when the forward modeling of 2nth-order difference accuracy in the condition of the regular grid is carried out, the stability condition is as follows: (4) taking into account the requirements of computational efficiency and simulation accuracy, time two-order and space tenth-order are adopted in this paper and c(l) is 0.541. the pml absorbing boundary is adopted as a model boundary to reduce the boundary disturbance caused by the artificial boundary. 4. wavefield characteristics analysis figure 3 shows the advanced detection model of small fault ahead of tunnel face with the fault throw of 10m. the dip angle of fault is 45 degrees, and the model size is 300 m × 300 m. the grid spacing is set as 0.25 m × 0.25 m, and sampling interval is δt=0.1 ms. taking into account the source features of field survey, the seismic frequency is set as 400 hz. the coordinate range of the tunnel face is x=130 m and z=149 ~ 152 m, while receivers are respectively located in the roadway roof and floor. the number of receivers is 40 in the roof and floor. (a) x-component receiver in roof (b) x-component receiver in floor figure 3. model of advanced detection. the receivers are located in the scope of x=91 ~ 169 m with trace intervals of 2 m. the depth of the receivers is 2 m in roof and floor. the two components x and z are used for geophone receiving, and x direction denotes the tunneling direction, and z direction denotes the direction being perpendicular to the floor (roslee et al., 2017). the coordinate of the source is (134, 149), and specific medium parameters of the model are shown in table 2. table 1. model parameter. (a) z-component receiver in roof (b) z-component receiver in floor figure 4. seismic data of diffraction advanced observation system. (a) x-component (b) z-component figure 5. seismic wavefield snapshots. 98 wbo wang, shengdong liu, fubao zhou, jun zhang, fangkun zheng figure 4 shows the seismic data of component x and component z with 0~80 ms, while figure 5 shows the wavefield snapshots of components x and z with 16 ms, 40 ms, and 57 ms. according to the seismic records and the wavefield snapshots of the model, with reference to kinematics and dynamics characteristics of various types of seismic wave, all kinds of waves can be recognized from the complex seismic wavefield, and they are as follows: a direct p-wave, this direct p-wave is with circular propagation from the source to the surrounding rock. b diffracted p-wave, is formed by the diffraction of the direct p-wave when propagating to breakpoint of the fault. c 1# reflected channel wave, is formed by the reflection of the direct channel wave when encountering fault. d 2# reflected channel wave, is formed by reflection of 1# reflection channel wave when encountering the tunnel face; e 3# reflected channel wave, is formed by the reflection of 2# reflection channel wave when encountering fault; f the roadway acoustic wave, is formed by the direct p-wave when propagating into the cavity of the roadway; g the diffracted p-waves at the endpoint of the tunnel face are the multiple diffractions when direct p-wave encounters the endpoint (kamsani et al., 2017). the fundamental difference b between g diffracted p-wave and diffracted p-wave is as follows: b diffracted p-wave can also be observed from the receivers ahead of tunnel face and behind tunnel face, while g diffraction can only be observed from the receivers behind tunnel face. the diffraction b can be clearly identified based on the above difference. it can be seen from figure 4 that the resolution of breakpoint diffraction b is highest in seismic data of component x, and the diffraction is recorded by the receivers in the floor. the waveform of the diffraction b is clear with strong energy and short-wave group. in the figure, the reflected channel wave of the fault is also obvious, but the wave group of the channel wave is wider than diffraction, which causes the low resolution when using the channel wave. due to the existence of the channel wave in the coal roadway and the roadway axial being parallel with the horizontal coal seam, the common observation system of the tunnel reflected wave should be with poor applicability when detecting an advanced fault. concerning to reflection observation system commonly used in traffic tunnel (liu et al., 2016), the sources are arranged in the rear of tunnel face, which results in figure 6 (kamsani et al., 2017) . from the figure, it is found that the acoustic wave in seismic data is specially developed and the diffraction of the breakpoint is not identified. it can be seen from figure 7, the lateral variation range of the fault location of reflected channel wave is about 20 m, and the distribution range of the channel wave migration imaging is large, and it is difficult to determine the position of small fault. the lateral variation range of breakpoint of diffraction is about 2 m, and the wave group of diffraction is short and high resolution, and imaging position of breakpoint of small fault is in accordance with the actual model, and the direction of breakpoint in full space is easily computed by polarization migration (wang et al., 2016). (b) diffracted p-wave figure 7. migration profile. 6. conclusion in this paper, considering the influence of the roadway cavity and coal seam with low velocity, the advanced detection model of small fault in front of heading face is established. the numerical simulation calculation is carried out by diffraction advanced observation system. the through comparative analysis between the diffraction and reflection observation system, and there are not acoustic wave and obvious diffracted p-wave with short wave group by diffraction observation system in seismic data of component x. compared with multiple diffractions of the endpoint, the diffracted p-wave is clear and easy to recognize and extract, and the reason is that: diffraction of small fault can also be observed from the receivers ahead of tunnel face, but diffraction of the breakpoint should not be observed from the receivers ahead of tunnel face. the reflected channel wave of small fault is difficult to identify because there are multiple reflected channel waves between the fault and the tunnel face. therefore, the diffraction of coal seam breakpoint is the naturalwave of small fault, and the results of the migration show that the imaging resolution of diffraction is high and breakpoint location of imaging is consistent with the actual model. acknowledgements this research has been performed by the national natural science foundation project (no. 41604082, 51323004, 41474122) and joint funding project of national natural science foundation and shenhua group corporation ltd (no.u1261202), and the fundamental research funds for the central universities (2014xt02). a project funded by the priority academic program development of jiangsu higher education institutions. (a) x-component receiver in roof (b) x-component receiver in floor figure 6. seismic data of reflection observation system. 5. advanced imaging of small fault based on the above analysis, the diffracted p-wave and the reflected channel wave are both the effective waves of advanced detection of a small fault in coal roadway. based on the principle of prestack diffraction migration, the contrastive study is carried out with diffraction of the breakpoint and the channel wave (roslee et al., 2017; lindang et al., 2017; kamsani et al., 2017; lai et al., 2017) . figure 7 shows the imaging results of b diffracted p-wave and c 1# reflected channel wave of component x which is received on the floor. the starting position of x-axis is 91m. (a) reflected channel wave 99diffraction characteristics of small fault ahead of tunnel face in coal roadway references cheng, j. l., li, f., peng, s. p., sun, x. y., “research progress and development direction on advanced detection in mine roadway working face using geophysical methods.” journal of china coal society, 39 (2014): 1742-1750. cheng, j. l., song, y. l., li, j. f., gao, f., xie, c., chen, q., “numerical simulation of scattered wave imaging of advanced detection in the roadway.” proceedings of the twenty-ninth annual meeting of the chinese geophysical society (2013). deng, s. q., study on numerical simulation of whole-space elastic wave and reverse time migration imaging method, (xuzhou: china university of mining and technology, 2012). essen, k., bohlen, t., friederich, w., meier, t., “modelling of rayleigh-type seam waves in disturbed coal seams and around a coal mine roadway.” geophysical journal international, 170 (2007): 511-526. gao, k., liu, z. g., liu, j., “effect of geostress on coal and gas outburst in the uncovering tectonic soft coal by cross-cut,” chinese journal of rock mechanics and engineering, 34 (2015): 305-312. inazaki, t., isahai, h., “stepwise application of horizontal seismic profiling for tunnel prediction ahead of the face.” leading edge, 18 (1999): 1429-1431. jetschny, s., bohlen, t., denise, d. n., “on the propagation characteristics of tunnel surface-waves for seismic prediction.” geophysical prospecting, 58 (2010): 245-256. jetschny, s., bohlen, t., kurzmann, a., “seismic prediction of geological structures ahead of the tunnel using tunnel surface waves.” geophysical prospecting, 59 (2011): 934-946. kamsani s.r., ibrahim n., ishak n.a., “psychological debriefing intervention: from the lens of disaster volunteers.’’ malaysian journal of geoscience, 1 (2017): 32-33. lai g.t., razib a.m.m., mazlan n.a., rafek a.g., serasa a.s., simon n., surip n., ern l.k., rusli t., mohamed., “rock slope stability assessment of limestone hills in northern kinta valley, ipoh, perak, malaysia.” geological behavior, 1 (2017): 16-18. lama, r. d., bodziony, j., “management of outburst in underground coal mines.” international journal of coal geology, 35 (1998): 83-115. lindang h.u., tarmudi z.h., jawan a., “assessing water quality index in river basin: fuzzy inference system approach.” malaysian journal of geoscience, 1 (2017): 27-31. liu, s. d., wang, b., zhang, j., mine seismic method and technology, (xuzhou: china university of mining and technology press, 2016). lu, b., “advanced detection of coal mine fault taking tunneling machine as the source” (proceedings of the twenty-ninth annual meeting of the chinese geophysical society, 2013). luth, s., buske, s., giese, r., goertz, a., “fresnel volume migration of multicomponent data.” geophysics, 70 (2005): 121-129. lüth, s., giese, r., otto, p., krüger, k., mielitz s., bohlen, t., dickmann, t., “seismic investigation of the piora basin using s-wave conversions at the tunnel face of the piora adit(gotthard base tunnel).” international journal of mining sciences, 45 (2008): 86-93. otto, r., button, e., bretterebner, h., schwab, p., “the application of trt at the unterwald tunnel.” felsbau, 20 (2002): 51-56. roslee r., bidin k., musta b., tahir s., tongkul f., norhisham m.n., “gis application for comprehensive spatial soil erosion analysis with musle model in sandakan town area, sabah, malaysia” geological behavior, 1 (2017): 01-05. shepherd, j., rixon, l. k., griffith, l., “outbursts and geological structures in coal mines: a review.” international journal of rock mechanics and mining sciences, 18 (1981): 267-283. wang, b., liu, s. d., lu, t., sun, h. l., “coal seam thickness detection in mine roadway by using advanced prediction method.” electronic journal of geotechnical engineering, 21 (2014): 4753-4762. wang, b., liu, s. d., zhou, f. b., lu, t., huang, l. y., gao, y. j., “polarization migration of three-component reflected waves under small migration aperture condition.” acta geodynamica et geomaterialia, 13 (2016): 47-58. yang, s. t., cheng, j. l., “numerical simulation of fore detecting with seismic in coal roadway and study of wave field characteristics.” journal of china coal society, 35 (2010): 1633-1637. yang, s. t., cheng, j. l., “the method of small structure prediction ahead with rayleigh channel wave in coal roadway and seismic wave field numerical simulation,” chinese journal of geophysics, 55 (2012): 655-662. zhao, y. g., jiang, h., zhao, x. p., “tunnel seismic tomography method for geological prediction and its application.” applied geophysics, 3 (2006): 69-74. earth sciences research journal earth sci. res. sj. vol. 16, no. 2 (dcember, 2012): 83 corrigendum presentamos excusas a nuestros lectores y al antropólogo javier andrés sandoval andrade puesto que en el vol 15 no 2 del 2011 de la revista earth science research journal (esrj) se publicó el articulo titulado "quaternary dating by electron spin resonance (esr) applied to human tooth enamel" en el cual no se hace un correcto pie de página de la figura 1, ya que no es tomada de (groot, 1992) como reza la misma, sino que es tomada de la tesis de pregrado titulada datación de restos humanos prehispánicos a través de esmalte dental usando resonancia paramagnética electrónica (epr) (2010). asimismo, no corresponde al sitio arqueológico aguazuque, sino que corresponde al sitio arqueológico checua. carlos vargas editor en jefe earth science research journal corrigendum we apologize to our readers and the antropology javier andres sandoval andrade because in the vol 15 issue 2 year 2011 in the earth science research journal (esrj) published the article entitled "quaternary dating by electron spin resonance (esr) applied to human tooth enamel" which there is an incorrect footnote in the figure 1 as it is not quoted from (groot, 1992), but it is taken from the undergraduate thesis entitled datacion de restos humanos prehispanicos a traves de esmalte dental usando resonancia paramagnetica electronica (epr) (2010). also, this figure do not correspond with the aguazuque arqueological site, but it is from the checua arquelogical site. carlos vargas editor-in-chief earth science research journal keywords: earthquake; prediction; prevention; risk reduction. palabras clave: benín; exponente de escalado; mfdfa; espectro de singularidad; temperatura; dinámica no lineal. how to cite item tapia-hernandez, e., reddy, e. a., & orosavilés, l., j. (2019). earthquake predictions and scientific forecast: dangers and opportunities for a technical and anthropological perspective. earth sciences research journal, 23(4), 309-315. doi: https://doi.org/10.15446/esrj.v23n4.77206 supporting earthquake risk management with clear seismic communication may necessitate encounters with various popular misapprehensions regarding earthquake prediction. drawing on technical data as well as insights from anthropology and economics, this paper addresses common and scientifically-unsupported ideas about earthquake prediction, as well as the state of science-based studies regarding statistical forecasting and physical precursors. the authors reflect on documented social and economic effects of unsubstantiated earthquake predictions, and argue that these may be dangerous but may also present certain opportunities for outreach and education in formal and informal settings. this paper is written in light of the importance that the united nations office for disaster risk reduction has placed on coordination and communication within and among diverse organizations and agencies as well as by the recent popularity of so-called earthquake prediction in mexico. abstract earthquake predictions and scientific forecast: dangers and opportunities for a technical and anthropological perspective predicciones y pronósticos científicos de sismos: peligros y oportunidades desde una perspectiva técnica y antropológica issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.77206 el apoyo de la gestión del riesgo de terremotos con una comunicación sísmica clara requiere incluir la atención de interpretaciones populares erróneas sobre la predicción de sismos. con base en datos técnicos, así como en percepciones de la antropología y la economía, este artículo discute ideas populares y sin sustento científico sobre la predicción de terremotos, así como el estado del arte de estudios científicos con bases estadísticas y con precursores físicos. se discuten los efectos sociales y económicos documentados de las predicciones de sismos sin sustento científico para argumentar su peligrosidad; pero también se hacen notar las oportunidades que representan su divulgación y educación en entornos formales e informales. el artículo discute a la luz de la importancia que la oficina de la organización de nacionales unidas para la reducción del riesgo de desastres ha otorgado a la coordinación y comunicación dentro y entre las diversas organizaciones y agencias, así como a la reciente popularidad de la llamada predicción de terremotos en méxico. resumen record manuscript received: 11/01/2019 accepted for publication: 05/07/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 309-315 edgar tapia-hernández1*, elizabeth a. reddy2 and laura j. oros-avilés3 1department of materials. universidad autónoma metropolitana azcapotzalco, 02200, mexico city. 2department of engineering, design and society, colorado school of mines, united states. 3department of economics. universidad autónoma metropolitana azcapotzalco, 02200, mexico city * corresponding author: etapiah@azc.uam.mx g eo lo g ic a l h a za r d s 310 edgar tapia-hernández, elizabeth a. reddy and laura j. oros-avilés introduction on march 20, 2012 a strong earthquake of magnitude 7.4 struck southern mexico. over the next days, other significant earthquakes around the world were registered: an earthquake of magnitude 6.6 in new guinea on march 21; an earthquake of magnitude 7.1 in maule, chile on march 25; earthquakes of magnitude 8.6 and 5.5 in sumatra on april 11; an earthquake of magnitude 6.5 in michoacán, mexico on april 11; and another of magnitude 6.9 in the gulf of california on april 12 (usgs 2015). this cluster of activity, though far from unusual, was interpreted by at least one public as empirical confirmation of earthquake predictions (e.g. univison 2013; latinospost 2013) which drew not on peer-reviewed scientific research but rather on methods which incorporated the position of the planets, dreams and even the so-called mayan apocalypse, slated to end this age and begin a new one on december 21 of that year.1 seismicity and related hazards have killed more people than all other natural hazards combined in the past 20 years (unisdr and cred, 2016) and the danger that they continue to pose requires ongoing action. the most recent united nation sendai framework for disaster risk reduction has drawn social and physical sciences together to indicate how crucial popular knowledge about hazards can be for effective disaster prevention. this is noting that not only can clear and coherent understanding of what can sometimes be very technical issues be important for publics, but it may also be crucial to facilitate coordination between institutions and stakeholders at all levels including community-based organizations, non-governmental organizations, and governmental organizations at local, national, and international scales (unisdr 2015)2. in this context, non-expert familiarity with basic information about earthquake hazards may be essential for risk management. scientifically unsubstantiated ideas about seismicity like those that we refer to above are not necessarily opposed to good earthquake risk reduction practices. the impetus of preparation need not be a scientifically verified threat to be potentially helpful; indeed, a marked rise in discussion of earthquake hazard and interest earthquake insurance interest was documented in the wake of a spurious prediction of a 1989 earthquake on the new madrid fault (shipman et al., 1993). an understanding that a scientifically viable threat is not necessary to encourage useful activity has informed a strategy by the centers for disease control to encourage emergency preparedness through discussion of a zombie outbreak (cdc 2015).3 although preparedness activities inspired by any threat can contribute to public safety, the spread of misinformation can be dangerous. misinformation can frighten people and facilitate miscommunication or mistrust, unpredictable social trends, or demands for resources. inaccurate seismic forecasts may be detrimental to public wellbeing in ways that drawing playfully on a horrormovie hazard are not. mistaken expectations about the degree to which seismic hazard can be predicted may create confusion and has been found in some circumstances to be related to activities that might have negative impact on potentially vulnerable populations and undermine the authority of scientific experts (stallings 1995). the formal and informal outreach and education about seismicity and earthquake risk mitigation strategies can be observed in the context of popular fascination with prediction, writing editorials in newspapers, giving talks, or simply engaging in ordinary conversation with friends, family, and colleagues. in light such activities, the state of the field of earthquake forecasting is a matter of some importance for earthquake engineers and other experts. knowledge of a hazard does not necessarily spur effective disaster prevention activities (landeros-mugica et al., 2016). however, with a clear understanding of the state of research on both seismic forecasting and the dangers and opportunities provided by predictions, technically-trained experts can better consider their intervention and outreach work. such labor, whether formal or informal, paid or unpaid, may substantially support the kind of general communication and coordination that we know can support prevention, save lives, and speed recovery. 1 see, also, theories circulated by brian t. johnstone, keith hunter, and those documented in austin (2016) 2 see particularly priority one, “understanding disaster risk.” 3 jason j. morrissette (2014) contextualizes the symbolic power of the figure of the zombie in critical security studies, while julia daisy fraustino and liang ma (2015) critique the effectiveness of the social media campaign. this paper has the following goals: first, to outline current mechanisms for forecasting earthquakes, and second, to describe documented dangers and opportunities afforded by popular interest in earthquake prediction. finally, we call for further systematic research on how the formal and informal communication about earthquakes undertaken by experts may contribute to public safety as well as on the effects of erroneous predictions for both disaster prevention works. although some commentators have described “a tendency on the part of officials to see disaster planning as a product, not a process” (wenger et al. 1980), authors are of the opinion that facilitating the ongoing discussion and the circulation of useful information regarding earthquakes is part of the necessary, but sometimes frustrating, process of effective disaster preparation. on “prediction” the prediction of an earthquake is defined as future seismic event may or may not occur in a region, a period of time, and magnitude with probability p (0.0 < p< 1.0) that the earthquake occurs in order to warn the population about the risk. the most memorable prediction occurred in 1975 in haicheng, china, where the changes in ground elevation, along with levels of groundwater, the behavior of animals and short quakes allowed an evacuation, just few hours before the earthquake. however, rather than the application of a methodology to provide a certain probability of occurrence, the prediction was based on coincidences and circumstantial facts. this illustrated by the fact that local experts failed to accurately predict another similar event. indeed, thirty incorrect predictions were made just in two years (1997 to 1999). this scenario dismissed the credibility and forced the chinese government to intend to stamp out wrong warnings and mass evacuations (saegura 1999). the interest of society in predictions is supported by the uncertainty of the scientific understanding of seismic activity and the threat that represents. in fact, they might have circulation in the social media and press. some of them, that apparently account a tested methodology, are occasionally even published in scientific venues (e.g. curiel 2010; straser et al. 2019). in particular, the earthquake predictions of gabriel curiel, which suggest that gravitational forces create a “season of earthquakes” between october and december, were presented in the 2010 mexican conference of earthquake engineering and, subsequently, published on curiel’s blog and referenced in countless newspaper articles (e.g. univision 2013; latinospost 2013), despite its lack of scientific support or even any attempt to offer convincing statistical evidence (curiel 2010). despite of the fact, that it is known that the available prediction methodologies have null scientific support, forecasting approaches continue to develop. after the 2009 l’aquila earthquake, the international commission on earthquake forecasting for civil protection developed a document, which reports the state of the art of short-term predictions and indicate guidelines for identification of possible precursors in order to warn the population (jordan et al. 2011). the main recommendations are as follow: a. it is desirable to continue to track the scientific evolution of probabilistic seismic predictions and develop the expertise to utilize these data for operational intentions. b. it is necessary to establish instrumented laboratories for studying seismic activity and their generation process. c. it should develop a research project focused on seismic activity as a part of a national program to develop operational prediction methods. d. it should develop a research time-independent and time-dependent of prediction methodologies in order to improve seismic hazard maps. e. it is necessary to develop an operational capability for the prediction of aftershocks. f. the validating the forecasting models should be developed from the international efforts to develop to test earthquake prediction methods. research on precursor physical phenomena gas measurement a strong earthquake (mw= 7.1) struck the philippines on november 15 in 1994. the epicenter was located 48 kilometers from the taal volcano, where a 311earthquake predictions and scientific forecast: dangers and opportunities for a technical and anthropological perspective stake and others noted an abnormal emanation of hydrogen in bubbles of mineral springs in five locations just one month before the may 26, 1983 earthquake in atotsugawa and ushikubi, japan (mw= 7.7) and before the september 14, 1984 earthquake in nagano (mw= 6.8) (igarishi and wakita, 1995). in contrast to the above, more than 150 studies have reported that measurements of gases might not be a proper reference of chemical concentration underground at considerable depths. this is due thermo-mineral waters contacts a considerable volume of rocks at various depths, so the records are not necessarily more representative of the environment at the tectonic plates. because of this, it is desirable to study the relationship between the gas emanations and the environmental disturbances such as atmospheric pressure, tide, temperature, rainfall, barometric pressure, soil moisture, wind temperature, volcanic disturbances, etc. (segovia et al. 1999). nevertheless, the study of gases emanations has been a typical effort in lessthan-scientific earthquake predictions and, indeed, they have appeared too often in public discussions of the seismic activity. for example, the prediction made by giampaolo giuliani before the 2009 l’aquila earthquake was established form gases measurements. after the event, giuliani became recognized in social and popular media for the apparent efficacy of the measurements, despite of he had been asked to desist from frightening the society after he had made some incorrect predictions (hall 2011; cartlidge 2012). geophysical observations the seismic activity has been associated also to electromagnetic phenomena. for instance, light flashes were related with earthquakes in derr in 1973 and komogawa and others in 2005. similarly, electromagnetism effects were reported by gokhberg and others in 1982 and nagao and others in 2002. atmospheric electric fields were reported by kondo in 1968 and vershinin and others in 1999. ionospheric perturbation was discussed by nagaov and hayakawa in 1998 and liu and others in 2000 (omori et al. 2007). the chemistry and isotope components of water were studied in 9 geothermal fields located around the anatolian fault in turkey from 2002 to 2004 in order to investigate their relationship with earthquakes. the study report changes in chemical and isotopic balance, which it seemed to be related with seismic activity of moderate magnitude (süer et al., 2008). additionally, the relationship between the december 2004 earthquake in sumatra and the f-region ionospheric parameters probed remotely by a digital ionsonde over new delhi have been also studied. according to the results, perturbations in fof2 and hmf2 were noted several hours before the seismic activity (dutta et al. 2007). a correlation was established between an anomalous cloud in geostationary satellite sensor above the active iran fault by guo and wang (2008). thick clouds had spread along the tectonic plate with high temperatures, 69 days after the seismic activity. similar clouds were also noted, 64 days after an earthquake of magnitude 6.0 on december 2005. with this scenario, despite the large number of possible precursors, no definitive results have been achieved. namely, in a probability space ( p), where  is a set of events, p is a set of probabilities, m is the precursor and n is an earthquake occurrence (m, n  ). if the probability of event occurrence exists p(m) > 0 and the probability that earthquakes in that geographic region exists p(n) > 0; then, the conditional probability of n in relation of m is defined as the equation 1. p |m n p m n p n ( ) = ( )( )  (1) research of gas emission had been developed since june 1993. the recollected data reported, 22 days before the seismic activity, an unexpected increase in the emanation of a gas called radon. radon is a colorless, odorless and tasteless gas. it is the most studied terrestrial gas for the purpose of earthquake prediction.4 earthquake forecast research involves measuring a rays associated with the radioactive disintegration (igarashi and wakita 1995). then, the observation reported by richon et al. (2003) and other similar researches indicated the possible identification that radon gas could be associated with the increased stress in the tectonic plates which may cause an earthquake. some other researchers have reported geochemical changes before an earthquake. for instance, two weeks before the magnitude 6.8 nagano prefecture earthquake, hirotaka and others noted variations of soil-gas radon recorded 65 km from where these measurements were taken in japan in 1988 (richon et al. 2003). igarashi et al. (1995) noted similar radon anomalies as the one reported by richon et al. (2003). abnormal emanations of soil gases (radon, helium and dihydrogen) were reported at several locations following the magnitude 8.0 sichuan earthquake of may 2008 (zhou et al., 2010; zheng et al., 2012). in addition, an anomalous radon 222rn decrease was detected (fig. 1) as a different precursor starting about 2 days before a magnitude 6.0 earthquake in japan in 1991. the anomalies were reported about 200 km from the epicenter by wakita and others (igarashi and wakita 1995). figure 1. anomalous radon 222rn decrease recorded before an earthquake (adapted from igarashi and wakita 1995) with this scenario, the hypothesis was that the near-failure compression with imminent movement produces emissions of gas radon 222rn, much as if it is being squeezed from the tectonic plates. thus, an increment of emanation might anticipate a seismic activity. in controlled conditions in laboratory tests, the increment of the stress in a rock sample under uniaxial compression had been related with before an increment of the emanations of gas radon 222rn (igarashi and wakita 1995). for this reason, observations of the emanations have been investigated at various places, and different methodologies have developed to relate the radon 222rn emanation and seismic activity (e.g. segovia et al., 1999; omori et al., 2007; richon et al., 2010). however, the results are not definitive and cannot be generalized, since the seismic activity continues happening without gas emanation in all cases (richon et al., 2010). and also, earthquakes have also occurred without abnormal emanation of gas radon 222rn (peña 2003). for instance, an anomalous radon decrease was reported after the seismic activity, not before the earthquake in measurements from january and may 1987 in fukushima prefecture at the northeast of japan (fig. 2). the emanation of other gases has been also studied. abnormal measurements of helium were noted in nagano, japan from one to three months before the september 14 earthquake (mw 6.8). the emanations were reported in mineral springs and gas fumaroles between 9.0 to 95.0 kilometers from the epicenter (nagamine and sugisaki 1991). hydrogen is produced in the cortex through a complex process. for this reason, hydrogen is not studied as an indicator of earthquakes as other gasses. kato and others demonstrated in laboratory tests that samples of granite under compression load might be related to the emanation of hydrogen. furthermore, 4 there are three radon isotopes: 222rn (t1/2= 3.825 days), 220rn (t1/2= 54.5 s) and 219rn (t1/2= 3.92 s). generally, radon refers to 222rn since it has the longest half-life and, therefore, is of most geochemical interest. figure 2. observed data on radon concentration (adapted from igarashi and wakita 1995). 312 edgar tapia-hernández, elizabeth a. reddy and laura j. oros-avilés p(m|n) is interpreted whereas if n is satisfied, the conditional portion of m must be met. in other words, if the identified precursor occurs, the conditional probability of an earthquake also exists. the disadvantage of the methods discussed above is that the space where the precursor event m exists in relation to natural phenomena unrelated to seismic activity is extremely high. similarly, there is a high probability that an earthquake will occur without a precursor n. precursor events have been occurring without an earthquake and seismic activity has been reported without precursor events. in general, the studies on physical precursors have demonstrated the need for simultaneous measurement of external parameters and the creation of networked monitoring stations. nowadays, the innate randomness and frequency of earthquakes are used to sustain predictions that might produce unwarranted confidence. for the real conditions, there is still much to do to establish relevant data. also, it is necessary to study the precursors in simulations in the laboratory as a function of the variety of cases and conditions. precursors based in statistical patterns statistical correlations from catalogs of the seismic activity and other techniques developed from probabilistic models are also under investigation. these methods are based on i) diminutions of earthquakes in a region (seismic gap), ii) the tendency of the epicenters motion, iii) increments of earthquakes in the periphery of a region (donut pattern) or iv) the increase of seismic motions at low magnitudes in a particular region (swarms). statistical methods are based on data collected with stations, records of the historical seismic activity and from collaborations between historical anthropologists and geophysicists (e.g. garcía and suárez, 1996 and zuñiga et al. 1997). it is supposed, according to the elastic-rebound theory, that the tectonic plates develop stress progressively and during an earthquake, the energy is released suddenly. and after the seismic activity, it starts another energy accumulation period that will produce a new earthquake and so on. then, a seismic gap is an active fault that has not slipped in a long time when compared with other regions along the same tectonic plate. for instance, at the state of guerrero in southern mexico, an earthquake has been anticipated for many years, based on historical records. supposedly, the accumulated energy is enough to produce a strong earthquake. the information about the disastrous magnitude was spread to the society after the 8.1 earthquake of september 19, 1985 by geophysicists and the potential danger of this gap (ugm, 1986). based on this, an earthquake early warning system was expanded along the region by policymakers in mexico (espinosa-aranda et al. 2009). locally, the seismic activity has been studied using models of the physical processes and bayesian statistics in order to make adequate use of the available information since then (e.g. singh and ordaz 1994). however, the predicted strong earthquake has not struck the country. it is thought two scenarios: i) the energy continues accumulating and will lead to rupture someday, producing an even major earthquake or ii) the release movement has already happened by way of “silent” earthquakes. the release of energy through the displacement of the plates for several days is called silent slip earthquakes. nowadays, they can be distinguished by gps monitoring. for instance, figure 3 shows the location of seven sites in southwestern british columbia, canada, where red arrows show the detected displacements by continuous gps, while black arrows show the common gps motions. namely, the studied sites reversed their direction of motion. the unusual displacements were estimated about 2 centimeters in the range from 6 to 15 days; so, the slip over a 50-kilometer by 300-kilometer area was equivalent to an event of a magnitude 6.7 (dragert et al. 2001). the probability that an earthquake might occur based on the identification of historical records is uncertain. for example, in parkfield, united states was predicted an earthquake based on previous events recorded in 1857, 1881, 1901, 1922, 1934 and 1966 (bakun et al. 1987; mileti and fitzpatrick 1992) as depicted in figure 4. the identification of a seismic gap and a return period of 21.9 years were estimated with a probability of occurrence at 95%. nevertheless, the seismic activity did not occur until september 2004, deviating from the trend. thus, the search for earthquake precursors useful in a short-term has had no success yet in any sense, but their possible effects might be powerful. prevention activities and earthquake prediction disaster is a complex of social and physical effect (o’keefe et al. 1976; garcía and rojas 1994; oliver-smith 1996). disaster is not the sure outcome of a large earthquake, but instead the product of impact on people. this means that we can mitigate disaster by taking steps to prepare for earthquakes. understanding that earthquakes are likely to happen in certain areas and at certain magnitudes can support such preparation; however, when preparation is anchored to a predicted event at a given time and place, there is significant concern that this, too, can be dangerous, though in oblique ways that can be difficult to prove in the complex welter of social action. the recent un sendai framework on disaster risk reduction (unisdr 2015) addresses social and structural issues as a part of vulnerability, highlighting the ways that vulnerable people are disproportionately affected by disaster. the sendai framework defines vulnerability, in reference of hyogo framework, as “the conditions determined by physical, social, economic, and environmental factors or processes, which increase the susceptibility of a community to the impact of hazards” (unisdr 2005). this means that factors such as knowledge, income, type of work, social support network and qualities of the built environment itself such as where homes or other important structures are sited, access to services like electricity or sewage or roads and local resource depletion or contamination can be considered to contribute to vulnerability of a community to disaster. strategies suggested by the sendai framework (unisdr 2015) are wide ranging, including understanding disaster risk; strengthening disaster figure 3. description of gps sites in canada in the summer of 1999 (from dragert et al., 2001) figure 4. statistical projection of expected earthquake in parkfield, united states 313earthquake predictions and scientific forecast: dangers and opportunities for a technical and anthropological perspective risk governance to manage disaster risk, investing in disaster risk reduction for resilience, and enhancing disaster preparedness for effective response, recovery, rehabilitation and reconstruction. some of this, falls under the purview of state actors, but especially as many bureaucracies find hazards challenging in the extreme (dynes et al. 1990; perrow 2008). indeed, fostering prevention activities broadly and facilitating clear communication and integration of efforts by state and non-state actors at local, national, and international levels has been identified by the united nations office for disaster risk reduction, as a strategy to diminish the risks that potentiallydisastrous hazards pose well in advance of any danger. there is nothing controversial here, nor is there much debate regarding the understanding that a general awareness of a hazard is highly desirable for earthquake risk reduction (fema-474, 2005), especially if the members of the population are represented by the possible actions that can be taken (farley et al. 1993; mileti and fitzpatrick 1993), and can be an excellent foundation upon which to build further interventions (gratton et al. 1987; mileti and fitzpatrick 1992). this kind of awareness is certainly most helpful when it resonates with informed understandings of how earthquakes happen and are likely to affect populations and can be used in multiple and even creative ways (as barrios 2014 argues). however, it can also be troublesome. not only can emergency planning itself face significant apathy and resistance, articulating the management of action often with the management of human emotion or affect5, but also the kinds of experience with hazards and knowledge about them are also sometimes related to avoidance or unwarrantedly optimistic assessment of safety (landeros-mugica et al. 2016). despite of experts hope that they might encourage planning and preparation. prevention does not necessarily depend on scientifically informed understandings of hazard, though multi-modal and detailed communication around a threat has been found to be a key strategy for emergency hazard communication (mileti and sorensen 1990; sorensen 2000). this is a fine line to parse, but an essential one. substantial anthropological evidence indicates that particularly well-established and pervasive belief systems, whether scientifically informed or non-informed, are likely to be efficacious in the face of hazards common in the contexts in which they emerged. the adaptability and creativity based on traditions represented significant support in the past (torry 1978; oliver-smith 1994) and might offer a powerful lesson for risk mitigation (turton 1977; oliver-smith 1986). indeed, the potential utility of unscientific belief systems is supported in law; for example, the experience of l’aquila residents. in the wake of a deadly 2009 quake there, a court case was brought against the experts. they, reacting against the spurious predictions of giampaolo giuliani (noted above), assured that it was unlikely that an earthquake was forthcoming. these assurances dissuaded the population from following local custom to sleep outside in the advent earthquake swarms, and for their part in 32 l’aquila deaths, the experts were convicted of manslaughter, since overturned (alexander 2014). policy frameworks including sendai (unisdr 2015) explicitly highlight traditional informed practices potentially useful or even models for the development of effective practices. if not necessarily dangerous in and of themselves, popular expectations about the predictability of earthquakes may become so if they interfere with potentially lifesaving actions, causing them to be put off or misapplied. although research in the social sciences demonstrates that panic is often less extreme than expected (mileti and sorensen 1990, tierney 2003), anxiety involved in earthquake predictions may still have effects on institutional resources and economic activity in significant ways (haas and mileti 1976; weisbecker et al. 1977). institutional & economic impacts of earthquake predictions there are clear mechanisms by which earthquake predictions might be understood to influence expectations and affect economic activity and formal political relations. they may do so in both positive and negative ways. on the one hand, earthquake damage, often calculated as damage to a year’s expectation of economic gain, can happen here even in the absence of an actual 5 on the topic of the historical and politically-situated promises and dangers of the management of fear in particular see daston (2008) and massumi (2005). earthquake (though not in a way that insurance can be drawn upon to defray) as people evacuate, fail to visit, and change their investment plans. these kinds of impacts in advance of a hazard that may not occur can be referred to as a matter of “psychological foreshocks” (olson et al. 1989). on the other, such predictions may happen in concert with increased economic activity, as people stock up on emergency supplies (as shipman et al. suggests they did in advance of a prediction regarding the new madrid fault in 1993). they might take action, further, upon their expectations regarding possible losses, expected profits, and market depreciations. indeed, earthquake predictions based on wellestablished systems and pervasive beliefs, could modify short term spending and institutional demands with potentially wider repercussions by means of longer-term plans and market activity. in economic theory, decisions are often discussed with an inter-temporal approach: an investment decision is determined by the relationship between understood conditions of the past, of the present, and those likely in the future. this can be considered as a matter of incentives, and the scale of such incentives relate extant possibilities to potential increases in production (mankiw 2007). business owners and investors’ forecasts alike, and their confidence about forecasts, can be critical to investment decisions. their decisions do not only pertain to their own endeavors but can come to involve investors and related industries through market mechanisms, which can spread impact of investment decisions widely. the assurance with which some earthquake predictions are deployed can be very persuasive, and may not only be expressed around single events, but also with respect to longer time frames or returning seasons (“earthquake season” for example). scientific forecasting can certainly influence economic decision making. however, predictions with strong assertions of regarding when an earthquake will happen, how large it will be, and where it will be felt, may appear more concrete than the general forecasts that current state-of-the-art forecasting can offer. these, then, can motivate economic action in the same way as they motivate preparatory action: in ways that are potentially beneficial and quite troubling, but may in the end be less extreme than experts worry (olson and olson 2001). the confidence on predictions can affect economic activity of consumers, business planning, and potentially markets; they can create challenges for public facing institutions of many kinds. olson et al. (1989) document the brady-spense prediction of an earthquake in 1981 and some of the scientific and institutional conflicts which surrounded it and emerged out of it. as news of a potentially credible earthquake prediction spread and became highly politicized in the us and in peru (the territory likely to experience its impacts) people with different goals, interests, and expertise became involved in responding to the threat. many organizations were unprepared for the demands that the prediction could inspire. while olsen et al. documented how the event catalyzed new peruvian assessments of risk and disaster response capabilities, the international assistance possibilities available, and how it raised public awareness of earthquake threat, they also demonstrate the kinds of institutional pressures that were created as business as usual in state institutions. for example, peruvian civil defense was overwhelmed with requests for trainings, and could only respond to a fraction of them. predictions can create unexpected conditions, interfering with planning; and while this may open the door for new opportunities, it may also create serious trouble for midand longrange planning. conclusions in newspapers, television, and social media, predictions about earthquakes circulate. sometimes, these predictions refer to religious or mystical explanations. they may seem scientific to people unfamiliar with seismic science, referencing studies or data that readers may find difficult to parse. this paper draws on engineering, anthropological and economic scholarship to provide an overview of the current state of the field in earthquake forecasting and discuss the social and institutional implications of earthquake predictions. it pretends to give earthquake engineers the tools they need to understand current research and engage with non-expert members of the public and facilitate the kinds of coordination between institutions and stakeholders at all levels including community-based organizations, non-governmental organizations, and governmental organizations at local, national, and international scales called for by the united nations office for disaster risk reduction. 314 edgar tapia-hernández, elizabeth a. reddy and laura j. oros-avilés engineers are highly trained members of professional and nonprofessional communities who may engage with their peers on the topic of predictions formally and informally. many correct misapprehensions and answering questions, even when they are not employed to do so, providing bases by which untrained members of their communities can better understand the sometimes-confusing information they may receive about scientific research on statistically-based earthquake forecasts and seismic precursors, as well as the predictions based on ideas about earthquake seasons or ancient mystical knowledge which sometimes circulate alongside other kinds of earthquake information in the popular press. this paper provides an overview of cuttingedge scientific research into physical precursors and statistical methods of earthquake forecasting and a discussion of the potential impacts of public education regarding seismic threat can have. unlike other natural phenomena, earthquakes are not, generally speaking, predictable. statistical and precursor research is nonetheless providing key insights into seismic effects in various local environments. in correcting misapprehensions, trained experts may misestimate the negative effects that predictions can have. frustrating and troubling as they might be, they also provide opportunities that, if seized, may be turned to useful effect for 1) public education and outreach regarding earthquake safety and 2) collaboration in and between government and non-governmental institutions at local, regional, and national scales. as discussed in this paper, substantial anthropological evidence indicates that technical knowledge about seismicity may not be significantly correlated to effective public practices for earthquake risk management. nonetheless, widespread confusion and poor communication around predictions can cause interruptions to normal activity that may be both deleterious and offer opportunities for outreach about effective preparation activities. further, as responses to prediction and public concerns may necessitate the involvement of professionals without significant technical knowledge about seismicity, the understanding that a general public has of earthquake science and that of experts in the field are not the only ones of import. professionals with various responsibilities may become involved in the case of an earthquake prediction. understanding not only the geoscientific context of predictive possibility but also the social science on the risks around it may facilitate the kinds of reasonable action necessary to make the most of predictions. as erroneous earthquake predictions continue to seize popular attention without scientific support, it would be very useful indeed to critically evaluate their effects --both positive and negative-on community disaster preparation. additionally, it was noticed a great potential utility for systematic research on the effects of formal and informal expert communication about earthquakes, the state of scientific knowledge regarding forecasting, and the bases of earthquake prediction efforts. we suspect that the outreach research that we and our colleagues do can facilitate the production of shared understanding about seismic conditions and hazards and contributes to coordination that the unisdr has recognized as crucial for supporting prevention, saving lives, and speeding recovery. despite of the complex social effects of predictions that have been discussed in this paper; authors maintain that if seismicity and related hazards continue to pose the same kinds of threats to human life, then any excuse for outreach and building effective networks of collaboration must be seized. references alexander, d. e. (2014). communicating earthquake risk to the public: the trial of the ‘l’aquila seven. natural hazards, 72(2), 1159–1173. doi: 10.1007/s11069-014-1062-2. bakun, w. h., breckenridge, k. s., bredehoeft, j., burford, r. o., ellsworth, w. k., johnston, m. j. s., jones, l., lindh, a. g., mortensen, c., mueller, r. j., poley, c. m., roeloffs, e., schulz, s., segall, p. & thatcher, w. (1987). parkfield california, earthquake prediction scenarios and response plans, u.s. geological survey. open file report 87-192. u.s.a. barrios, r. e. (2014). here, i’m not at ease: anthropological perspectives on community resilience. disasters, 38, 329–350. doi: 10.1111/ disa.12044. cartlidge, e. (2012). aftershocks in the courtroom. science, 338(6104), 184188. doi: 10.1126/science.338.6104.184. cdc. (2015). preparedness 101: zombie apocalypse, internet site for centers for disease control and prevention. http://www.cdc.gov/phpr/ zombies.htm. (last accessed february, 2017). curiel, g. (2010). teoría de las fuerzas gravitacionales. proceedings, xviii mexican conference of earthquake engineering, aguascalientes, mexico (in spanish). daston, l. (2008). life, chance and life chances, daedalus. the mit press, 137(1), 5–14. doi: 10.1162/daed.2008.137.1.5. dragert, h., wang, k. & james, t. s. (2001). a silent slip event on the deeper cascadia seduction interface. science, 292, 1525-1528. doi: 10.1126/ science.1060152. dutta, h. n., dabas, r. s., das, r. m., sharma, k. & singh, b. (2007). ionospheric perturbations over delhi caused by the 26 december 2004 sumatra earthquake. international journal of remote sensing, 28(1314), 3141-3151. doi: 10.1080/01431160601099590. dynes, r. r., quarentelli, e. l. & wenger, d. (1990). individual and organizational response to the 1985 earthquake in mexico city, disaster research center, technical report. university of delaware, u.s.a. espinosa-aranda, j. m., cuellar, a., garcia, a., ibarrola, g., islas, r., maldonado, s., & rodríguez, f. h. (2009). evolution of the mexican seismic alert system (sasmex). seismological research letters, 80(5), 694–709. doi: 10.1785/gssrl.80.5.694. farley, j. e., barlow, h. d., finkelstein, m. s. & riley, l. (1993). earthquake hysteria, before and after: a survey and follow-up on public response to the browning forecast. international journal of mass emergencies and disasters, 11(3), 305321. fema-474. (2005). promoting seismic safety: guidance for advocates. federal emergency management agency. washington. september. fraustino, j. d. & ma, l. (2015). cdc’s use of social media and humor in a risk campaign—‘preparedness 101: zombie apocalypse. journal of applied communication research, 43(2), 222–241. doi:10.1080/00 909882.2015.1019544. garcía, v. & rojas, t. (1994). los sismos como fenómeno social: una visión histórica. macrosismos: aspectos físicos, sociales, económicos y políticos. ed. centro de investigaciones y estudios superiores en antropología social. mexico (in spanish). garcía, v. & suárez, g. (1996). los sismos en la historia de méxico. vol. 1. universidad nacional autónoma de méxico press. méxico (in spanish). gratton, v. g., thier, h. d., arjonilla, e. & melgar, r. (1987). the recovery of schools from earthquake effects: lessons from mexico city. disasters, 11(4). doi: 10.1111/j.1467-7717.1987.tb00654.x. guo, g. & wang, b. (2008). cloud anomaly before iran earthquake. international journal of remote sensing, 29(7), 1921-1928. doi: 10.1080/01431160701373762. haas, j. e. & mileti, d. s. (1976). socioeconomic impact of earthquake prediction on government, business and community. institute of behavioral science, university of colorado. hall, s. s. (2011). scientists on trial: at fault? nature, 477, 264-269. doi: 10.1038/477264a. igarashi, g. & wakita, h. (1995). geochemical and hydrological observations for earthquake prediction in japan. journal of physics of the earth, 43(5), 585-598. doi: doi.org/10.4294/jpe1952.43.585. igarashi, g., saeki, s., takahata, n., sumikawa, k., tasaka, s., sasaki, y., takahashi, m., & sano, y. (1995). ground-water radon anomaly before the kobe earthquake in japan. science journal, 269(5220), 60-61. doi: 10.1126/science.269.5220.60. jordan, t., chen, y., t., gasparini, p., madariaga, r., main, i., marzocchi, w., papadopoulos, g., sobolev, g., yamaoka, k., & zschau, j. (2011). operational earthquake forecasting. state of knowledge and guidelines for utilization. annals of geophysics, 54(4). doi: 10.4401/ag-5350. landeros-mugica, k., urbina-soriaand, u. & alcántara-ayala, i. (2016). the good, the bad and the ugly: on the interactions among experience, exposure and commitment with reference to landslide risk perception in méxico. natural hazards, 80(3), 1515–1537. 315earthquake predictions and scientific forecast: dangers and opportunities for a technical and anthropological perspective latinospost. (2013). scientist claims that devastating earthquake will hit mexico city before year’s end. latinospost, august 26th. http:// www.latinospost.com/articles/26233/20130826/scientist-claimsdevastating-earthquake-will-hit-mexico-city-before-years.htm. last accessed february, 2017. mankiw, n. g. (2007). macroeconomics. 4th edition, worth publishers. massumi, b. (2005). fear (the spectrum said). positions, project muse, 13(1), 31-48. mileti, d. s. & fitzpatrick, c. (1992). the causal sequence of risk communication in the parkfield earthquake prediction experiment. risk analysis, 12(3), 393-400. doi: 10.1111/j.1539-6924.1992.tb00691.x. mileti, d. s. & fitzpatrick, c. (1993). the great earthquake experiment: risk communication and public action. boulder co: westview press. isbn-13 978-0813383699. mileti, d. s., & sorensen, j. h. (1990). communication of emergency public warnings: a social science perspective and state-of-the-art assessment. federal emergency management agency (fema), ornl-6609. washington dc. morrissette, j. j. (2014). zombies, international relations and the production of critical security studies versus the living dead. studies in popular culture, 36(2), 1-27. nagamine, k. & sugisaki, r. (1991). co-seismic changes of subsurface gas compositions disclosed by an improved seism-geochemical system. geophysical research letters, 18(12), 2221-2224. doi: 10.1029/91gl02710. o’keefe, p., westgate, k., & wisner, b. (1976). taking the naturalness out of natural disasters. nature, 260(5552), 566-567. doi: 10.1038/260566a0. oliver-smith, a. (1986). natural disasters and culture responses. studies in third world societies, eric institute of education sciences. pages 260. williamsburg, va: coll. william & mary. oliver-smith, a., varley, a. (1994). peru’s five-hundred-year earthquake: vulnerability in historical context. disasters, development, and environment. john wiley & sons. london. oliver-smith, a. (1996). anthropological research on hazards and disasters annual. review of anthropology, 25, 1-435. doi: 10.1146/annurev. anthro.25.1.303. olson, r. a., & olson, r. s. (2001). socioeconomic reverberations of earthquake prediction: snapshot in time, peru 1979–1981. natural hazards review, 2(3), 124-131. olson, r. s., podesta, b., & nigg, j. m. (1989). the politics of earthquake prediction. ed. princeton university press. princeton, n.j. omori, y., yasuoka, y., nagahama, h., kawada, y., ishikawa, t., tokonami, s., & shinogi, m. (2007). anomalous radon emanation linked to preseismic electromagnetic phenomena. natural hazards earth system sciences, 7, 629-635. doi: 10.5194/nhess-7-629-2007. peña, g. p. (2003). cambios de radón en suelo y elementos químicos y trazas en agua subterránea asociados a actividad sísmica. ph.d dissertation, universidad autónoma metropolitana, april (in spanish). perrow, c. b. (2008). complexity, catastrophe, and modularity. sociological inquiry, 78(2), 162–173. doi: 10.1111/j.1475-682x.2008.00231.x. richon, p., klinger, y., tapponnier, p., chen-xia, l., vann der woerd, j. & perrier, f. (2010). measuring radon flux across active faults: relevance of excavating and possibility of satellite discharges. radiation measurements, 45, 211-218. doi: 10.1016/j.radmeas.2010.01.019 richon, p., sabroux, j., & punongbayan, r. (2003). radon anomaly in the soil of taal volcano, the philippines: a likely precursor of the m 7.1 mindoro earthquake (1994). geophysical research letters, 30(9), 1481. doi: 10.1029/2003gl016902. saegusa, a. (1999). china clamps down on inaccurate warnings. nature, 397, 284. doi: 10.1038/16760. segovia, n., mena, m., peña, p., tamez, e., seidel, j. k., monnin, m., & valdés, c. (1999). soil radon time series: surveys in seismic and volcanic areas. radiation measures, 31(1-6), 307-312. doi: 10.1016/ s1350-4487(99)00142-0. shipman, m., fowler, g. & shain, r. (1993). whose fault was it? an analysis of newspaper coverage of iben browning’s new madrid fault earthquake prediction. international journal of mass emergencies and disaster, 11(3), 379-89. singh, s. & ordáz, m. (1994). sismicidad y movimientos fuertes en méxico: una visión actual. research report 18. centro nacional de desastres, cenapred. mexico (in spanish). sorensen, j. h. (2000). hazard warning systems: review of 20 years of progress. natural hazards review, 1(2), 119–125. stallings, r. a. (1995). promoting risk: constructing the earthquake threat. new jersey: transaction publishers. straser, v., wu, h-c, bapat, a., venkatanathan, n., shou, z., gregori, g., leybourne, b. & hissink, l. (2019). multi-parametric earthquake forecasting from electromagnetic coupling between solar corona and earth system precursors. general assembly of the european geoscience union. paper id. 5976. poster session. vienna, austria. april. süer, s., gülec, n., mutly, h., hilton, d. r., cifter, c., & sayin, m. (2008). geochemical monitoring of geothermal waters (2002-2004) along the north anatolian fault zone, turkey: spatial and temporal variations and relationship to seismic activity. pure and applied geophysics, 165, 17-43. doi: 10.1007/s00024-007-0294-4. tierney, k. (2003). disaster beliefs and institutional interests: recycling disaster myths in the aftermath of 9/11, terrorism and disaster: new threats, new ideas. research in social problems and public policy, 11, 33–51. doi: 10.1016/s0196-1152(03)11004-6. torry, w. i. (1978). natural disasters, social structure and change in traditional societies. journal of asian and african studies, 13(3-4), 167–183. doi: 10.1163/15685217-90007141. turton, d. (1977). response to drought: the mursi of southwestern ethiopia. disasters, 1(4), 275–287. doi: 10.1111/j.1467-7717.1977.tb00047.x. ugm (1986). declaration of morelia, mexican geophysical union. november 27, 1986. http://www.cires.org.mx/docs_info/cires_002.pdf, last accessed in february 2017 (in spanish). unisdr (2005). hyogo framework for action 2005-2015: building the resilience of nations and communities to disasters, united nations, kobe, hyogo, japan. january. unisdr (2015). sendai framework for disaster risk reduction 2015-2030, united nations, sendai, japan. march. unisdr & cred (2016). poverty and death: disaster mortality 1996-2015. switzerland and brussels: united nations. univision (2013). pronostican sismo de 8.5 richter en la ciudad de méxico, univision noticias. august 27th. http://www.univision.com/noticias/ noticias-de-mexico/pronostican-sismo-de-85-richter-en-la-ciudad-demexico. last accessed february, 2017. (in spanish). usgs (2015). historic world earthquakes, earthquake hazard program, united states geological survey, department of the interior. http// earthquake.usgs.gov/earthquakes/world/historical.php; (last accessed february 2017). weisbecker, l. w., stoneman, w. c., & ackerman, s. e. (1977). earthquake prediction, uncertainty, and policies for the future. final report stanford research institute. center for resource and environmental systems studies. menlo park, ca. wenger, d. e., faupel, c. & james, t. (1980). disaster beliefs and emergency planning. university of delaware disaster research center, newark. zheng, g., xu, s., liang, s., shiand, p., & zhao, j. (2012). gas emission from the qingzhu river after the 2008 wenchuan earthquake southwest china. chemical geology, 339, 187-193. doi: 10.1016/j. chemgeo.2012.10.032. zhou, x., du, j., chen, z., cheng, j., tang, y., yang, l., xie, c., cui, y., liu, l., yi, l., yangand, p. & li, y. (2010). geochemistry of soil gas in the seismic fault zone produced by the wenchuan ms. 8.0 earthquake, southwestern china. geochemical transactions journal, 11(5). doi: 10.1186/1467-4866-11-5. zúñiga, r., suárez, g., ordaz, m., & garcía, v. (1997). seismic hazard in latin-america and the caribbean project. international development research center. proyect 89-0190. september. ottawa, canada (in spanish). with the development of computer vision and high-precision 3d model reconstruction, used for the measurement and 3d reconstruction of the geological landslides, acquiring a high-precision relative orientation basing multiple images is crucial and the key point to ensuring and improving the accuracy of 3d model and space position. currently, the conventional relative orientation model includes five independent parameters. for the linear relative orientation model, there are nine parameters to construct the linear space geometric relationship between the imaging and space point. to eliminate the impact of more parameterization and improve the accuracy and stability of solved parameters for the conventional direct relative orientation model, a new relative orientation model with seven constraints is proposed and validated in this paper. the additional constraints are derived from the orthogonal property of the rotation matrix of a stereo imaging pair and associated with the least squares adjustment to obtain a high-precision result of the relative orientation. through the accuracy assessment using space position, it is revealed that the new proposed model is more advantage for the conventional type of direct relative orientation, especially at 3d model reconstruction and close range photogrammetric and applications for the geological landslides measurement. con el desarrollo del entorno computacional y la alta precisión del modelo de reconstrucción tridimensional, utilizados para la medida y reconstrucción de despredimientos geológicos, es crucial la obtención de la orientación relativa de alta precisión basada en imágenes múltiples y es el punto clave para asegurar y mejorar la exactitud del modelo 3d y la posición espacial. actualmente el modelo de orientación relativa incluye cinco parámetros independientes. en el modelo linear de orientación relativa hay nueve parámetros para construir la relación geométrica espacial linear entre el sondeo y la posición espacial. para eliminar el impacto de más parametrización y mejorar la exactitud y la estabilidad de los parámetros resueltos el modelo de orientación relativa convencional, este artículo propone y valida un nuevo modelo de orientación relativa con siete restricciones. las restricciones adicionales se derivan de la propiedad ortogonal de la matriz de rotación de la imagen estéreo y se asocian con el ajuste de los cuadrados mínimos para obtener un resultado de alta precisión de la orientación relativa. al medir la exactitud con la posición espacial se revela que el nuevo modelo propuesto tiene más ventajas que aquel de orientación relativa directa, especialmente en el modelo de reconstrucción 3d y en las aplicaciones fotográmetricas de rango cercano para la evaluación de desprendimientos geológicos. abstract resumen keywords: 3d, high-precision, relative orientation, space position, additional constraints, imaging pair palabras clave: 3d, alta precisión, orientación relativa, posición espacial, restricciones adicionales, sondeo estéreo. the model of direct relative orientation with seven constraints for geological landslides measurement and 3d reconstruction record manuscript received: 19/09/2016 accepted for publication: 06/01/2017 how to cite item chen, y., xie, z., qiu, z., zhang, m., & zhong, s. (2016). the model of direct relative orientation with seven constraints for geological landslides measurement and 3d reconstruction. earth sciences research journal, 20(4), f1-f6 doi:http://dx.doi.org/10.15446/esrj.v20n4.60211 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n4.60211 el modelo de orientación relativa directa con siete restricciones para la medida de deslizamientos de tierra y reconstrucción tridimensional g e o t e c h n ic s earth sciences research journal earth sci. res. j. vol. 20, no. 4 (december, 2016): f1 f6 yifu chen 1, 2, zhong xie1, 2, zhenge qiu 3*, min zhang2, saishang zhong 2 1 national engineering research center for geographic information system, china university of geosciences (wuhan), wuhan 430074, p. r. china, 2 facility of information engineering, china university of geosciences (wuhan), wuhan 430074, p. r. china 3 research center for ocean mapping and applications, shanghai ocean university, shanghai 201306, p. r. china author for correspondence: zhenge qiu (e-mail: zhengeqiu@sina.com) g2 yifu chen , zhong xie, zhenge qiu , min zhang and saishang zhong introduction relative orientation is the process of recovering the position and orientation of one image with the other in an individual image space coordinate system (läbe & förstner, 2006), which is a classic topic in satellite photogrammetry, computer vision, and 3d model reconstruction. with the development of the study fields, a higher accuracy of relative orientation model is demanded and is a crucial process for ensuring and improving the accuracy of relative orientation. for the 3d measurement and space model reconstruction in the geological landslides, basing on multiple images, the accuracy of relative orientation directly determined the accuracy of the reconstructed 3d model, whose accuracy can reach the accuracy with the laser scanner in precision, theoretically. therefore, the high-precision relative orientation model is a key point for the various applications basing on space relative orientation (horn, 1990; mikhail, bethel & mcglone, 2001). many attentions are concentrated on solving relative direction elementary with the linear methods, such as eight-point algorithm, five-point algorithm, and singular value decomposition (longuet-higgins, 1987; philip, 1996; faugeras, 1993; hartley & zisserman, 2003). longuet-higgins was the first to discover that the position and orientation of one image with the other is contained in the essential matrix (longuet-higgins, 1987). in that paper, an eight-point algorithm is proposed to estimate an essential model with the limitation that it will introduce significant error because of noises in the image. to get the more definite essential matrix, nistér proposed five-point algorithm (nistér, 2004). the five-point algorithm estimates essential matrix by calculating the root of polynomial ten times. there are also methods on singular value decomposition (taylor & kriegman, 1995; ma, soatto, kosecka & sastry, 2004; hartley & zisserman). although homogeneous algebraic representation and singular value decomposition strategy can calculate relative direction elementary efficiently, it is a significant error in the result because of unmatched conjugate points. when there are outliers among the conjugate points, the least squares iterative method is one of the best choices for more accuracy results of relative orientation elementary. the least squares iterative method will decrease the adverse influence of the outliers by the iterative process. in the calculation process for relative orientation parameters, there are two solving methods. one is using iterative calculation with the least square adjustment. the method needs to give the initial parameter value. however, the initial value cannot be provided in some case, and the range of the given initial value will affect the accuracy of the relative orientation (stewenius, engels & nistér, 2006). in the worst situation, the wrong initial value will cause the equation does not converge and then have no solution, especially at close range applications (li, liu & tang, 2007; pan, 1999; duane, 1971). for large rotation angle of close range images, more strict relative orientation method is adopted to restore their relative position (tang & heipke, 1994; werner, 2003; zhang, huang, hu, wan & lin, 2011). the other method is to solve the equation of relative orientation model with one or some accurate parameter value directly. for the case, it is also difficult to acquire the few specific value which constraints its application for some fields. in this paper, a new relative orientation model, direct relative orientation with seven constraints is proposed to overcome the problems in the models described above. in the new model, the initial value is calculated by direct relative orientation with a rough bx. then, the seven constraints are used to control and adjust the solved parameter result during the bundle block adjustment. six constraints are derived from the inherent orthogonal property of rotation matrix. the seventh constraint stems from the decomposition of baseline. in the experiment, the model correction is validated, and the space position assesses the accuracy. through the experimental result, it is revealed that the proposed new model can improve and acquire a higher accuracy of relative orientation, and is suitable for large oblique angles and translation using the multiple images to measure geological landslides and perform 3d reconstruction of the geological landslides. conventional model of relative orientation 1. coplanarity condition equation relative orientation is the process of recovering the position and orientation of one image with the other in an individual image space coordinate system. given two images (left and right images) with an overlap filed from different viewpoints, a stereo model can be created and reconstruct the original epipolar geometry, represented by figure 1. assuming a ground point m is imaged at two viewpoints s1 and s2, the two image points (m1 and m2) are acquired, and the corresponding imaged ray is r1 and r2 respectively. the vector distance between the two viewpoints is called baseline b. basing on the epipolar geometry, the vectors r1, r2, and b must be in one plane, coplane, whose mathematical expression is represented by equation 1. (1)b • (r1×r2) = 0 z y x b s2 bx bz by s1 m1 m2 m (x1 y1 z1) (x2 y2 z2) μ ν r1 r2 figure 1. the diagram of the epipolar geometry relationship in the stereo images. basing on the epipolar geometry, the mathematic model of relative orientation can be expressed as follow by coplanarity equation 3 and equation 17 according to equation 1. (2) where bx, by, bz are components of baseline of a stereo pair [x1 y1 z1]=[x y -f]t and [x2 y2 z2]=r[x’ y’ -f]t coordinates of conjugate points in the image space coordinate system, (x,y) and (x’,y’) represent conjugate points in the image space coordinates system, the rotation matrix 1 2 3 1 2 3 1 2 3 a a a r b b b c c c � � � �= � � � �� � composes of the three angles φ, ω, κ, and f the focal length of image, respectively. the bx, by, bz represents the component of baseline in x, y, z directions, respectively. 1 1 1 2 2 2 f= =0 x y zb b b x y z x y z g3the model of direct relative orientation with seven constraints for geological landslides measurement and 3d reconstruction 2. direct relative orientation model in aerial photogrammetry, the values of by, bz, φ, ω, κ are set to zero as the initial value, and the accurate value of bx is given to determine the scale of the stereo model; otherwise the model scale will be any size. however, the initial value of the baseline components (by, bz) and the rotation angles is unknown in some cases, which will result in a low accuracy or an error solution for relative orientation. for this case, direct relative orientation model can be used to calculate the orientation parameters and perform the relative orientation. for the model of direct relative ordination, the value bx needs to be fixed accurately, and then the model derives to a linear mathematic construct for calculating the other parameters of relative ordination. the direct relative orientation can avoid the iteration calculation in the least square adjustment owing to the linear mathematic construct. however, two sets of parameter result for the relative ordination can be acquired, and need to determine which one is true. according to the space image coordinate of the left and right image point, the coplanarity condition equation eq.(2) can be transformed into the following linear form of eq.(3). (3) in the eq. (3), the l1, l2, l3, l4, l5, l6, l7, l8, l9 are the polynomial coefficients which are described in eq. (4), respectively. (4) eq. (5) is obtained by eq. (3) dividing l5 on both sides. (5) where 0 5/i il l l= , 0 5 1l = . eq. (5) is the basic mathematic model of direct relative orientation. this model can directly calculate the eight unknowns 0 0 0 0 0 0 0 0 1 2 3 4 6 7 8 9, , , , , , ,l l l l l l l l with eight or more conjugate points without approximate values. the baseline parameter bx has no influence on building up a stereo model; it can be assumed to be known. therefore, parameters by and bz are obtained by the following equations: (6) the following equations can get nine elements of rotation matrix r: (7) finally, three rotation angles can be decomposed by the definition of r [3, 17]. (8) according to eq. (6), l5 can be positive or negative, which does not affect the values of by and bz. however, there are two sets of possible solutions for the rotation angles φ, ω, κ. one of the two solutions only can be used to acquire the real configuration of relative orientation. therefore, the degree of the two angles φ and ω are used to determine and select a true solution, which should satisfy the following range represented in eq. (9). (9) 3. model of direct relative orientation with constraints during the conventional model of direct relative orientation, the eight parameters 0 0 0 0 0 0 0 01 2 3 4 6 7 8 9, , , , , , ,l l l l l l l l can be calculated when the value of bx is fixed. it is evident that the eq. (3) is linear, so the values of eight parameters can be determined uniquely without the parameters correlation. for the model, the parameters of rotation r and baseline b in the relative orientation are only constrained by the model equation, but not constrained by the relationship among the parameters, which will decrease the accuracy and stability of the parameter result and then reduce the accuracy of relative orientation. in the computer vision, it is well known that epipolar equation x’tex=0, rank-two constraint det(e)=0, and trace constraint 2eete-trace(eet)e=0 are utilized to solve essential matrix e. in the epipolar equation, the x’ and x represent the normalized image coordinates of conjugate points from the left and right images, respectively. the essential matrix e can be calculated through the matrix multiplication between the translation t and rotation r, e=tr. for the solving method, the nine components in the rotation r is constrained by orthogonal property of rotation matrix included implicitly in these equations. however, the elements in the baseline b are not limited. also, the solving method in the computer vision yields nine homogeneous polynomial equations of degree 3, and the nine similar polynomial equations are correlated. different with the conventional model of direct relative orientation and the essential matrix in computer vision that normalized image points as observations, a new relative orientation model, direct relative orientation model with seven constraints, is proposed and validated. in the new model, the original image point coordinate of conjugate points and focal lengths f are used, and the constraint includes the components relationship in rotation r and baseline b for improving and ensuring the accuracy and stability of relative orientation. the seven constraints are represented by eq. (10). (10) the proposed new model of direct relative orientation with seven constraints consisted of eq. (2), (3) and (10). the model selects the initial nine parameters of direct relative orientation as unknowns, and seven constraints are considered to improving and ensuring the accuracy of the solved parameters. 4. solution of direct relative orientation with constraints in the process of calculation of parameters in direct relative orientation with constraints, assuming eq. (2) is not equal to zero and an error v is in the equation, thus the eq. (2) is derived to eq. (11) basing on the eq. (3). (11) / 2 ; / 2p j w p� < > 1 2 3 4 5 6 7 8 9' ' ' ' ' ' ' ' ' 0v l yx l yy l yf l fx l fy l ff l xx l xy l xf= + + + + + + + + � 2 2 2 1 2 3 2 2 2 1 2 3 2 2 2 1 2 3 1 2 1 2 1 2 1 3 1 3 1 3 2 3 2 3 2 3 2 2 2 2 1 1 1 0 0 0 + + =x y z a a a b b b c c c a a b b c c a a b b c c a a b b c c b b b b � + + = � + + =� � + + =�� × + × + × =� � × + × + × = � × + × + × =� � �� 1 2 3 4 5 6 7 8 9' ' ' ' ' ' ' ' ' 0l yx l yy l yf l fx l fy l ff l xx l xy l xf+ + + + + + + + = 1 1 1 2 2 2 3 3 3 4 1 1 5 2 2 6 3 3 7 1 1 8 2 2 9 3 3 x z x z x z y x y x y x z y z y z y l b c b a l b c b a l b c b a l b a b b l b a b b l b a b b l b b b c l b b b c l b b b c = × � × = × � × = × � × = × � × = × � × = × � × = × � × = × � × = × � × 0 0 0 0 0 0 0 0 0 1 2 3 4 5 6 7 8 9' ' ' ' ' ' ' ' ' 0l yx l yy l yf l fx l fy l ff l xx l xy l xf+ � + + � + + � = 2 2 2 2 2 2 2 2 22 2 0 0 0 0 0 0 0 0 0 5 1 2 3 4 5 6 7 8 9 0 5 2 1 7 2 8 3 9 2 4 7 5 8 6 9 2 / ( ) ( 1, 2, , 9) ( ) / ( ) / x i i y x z x l b l l l l l l l l l l l l i b l l l l l l b b l l l l l l b = + + + + + � � � = × = = � + + = + +  3 5 6 2 1 4 1 4 1 1 1 1 12 2 2 1 6 3 4 2 5 2 5 2 2 2 2 22 2 2 2 4 1 5 3 6 3 6 3 3 3 3 32 2 2 ; ; ; ; ; ; z y y z x y z x x z y y z x y z x x z y y z x y z x x l l l l b l b l b a l b a l a b c b b b b b l l l l b l b l b a l b a l a b c b b b b b l l l l b l b l b a l b a l a b c b b b b b � � � + + = = = + + � � � + + = = = + + � � � + + = = = + + 3 3 2 arcsin( ) arccos cos = arccos cos b c b w j w k w = � = g4 yifu chen , zhong xie, zhenge qiu , min zhang and saishang zhong then, differential form of the eq. (11) is represented by eq. (12). the f0 is the approximate value of f, and dbx, dby … dc2, dc3 is the variation of the 12 unknown parameters for relative orientation. thus, the error equation is described as eq. (13). (12) (13) where v, x, l represent the observation vector, parameter variation value, and the free vector, respectively. to improve and ensure the accuracy of the solved parameters, the seven independent constraints described above, are considered, and its equation is expressed by eq. (14). b • x + w = 0 (14) b and w are the coefficient matrices and the observation vector of constraint condition in the bundle block adjustment, which are represented by eq. (15) and eq. (16), respectively. (15) (16) combining the error equation of observation and the equation of constraint condition, the error equation with constraint condition is expressed in eq. (17). (17) according to the least square with restrictions, the 12 unknown parameters can be computed and express by the eq. (18). (18) in the eq. (18), the u, k, and wu can be acquired by the follow eqs. (19). (19) 5. experiments and results data and experiment the indoor close range images with large oblique angles are used for experiments to validate the correctness and accuracy of the proposed new model of direct relative orientation with seven constraints. the stereo image pair was acquired by a non-calibrated sony imx135 camera with a pixel size of 1.12 μm which is shown in figure (2). the numerous conjugate points from the left and right image were detected using surf match model, and the error match points were removed to obtain the reliable relative orientation results as is shown in figure 3. from the 62 high-precision matched points, the six conjugate points were selected and used to validate the model correctness and assess the result accuracy of the relative orientation. the baseline length was measured as 0.55m. figure 2. the stereo images are taken by the same camera. the first image above is taken from the left then the other is on the right. figure 3. the result of feature point detected on the stereo pair. the experiment includes the follow procedures. 1) the 62 conjugate points were used to calculate the parameters of the relative orientation of the stereo image pair. 2) basing on the image space coordinate of the left image, the space coordinate of the object points corresponding to the selected six conjugate points were calculated using the result of the relative orientation and the forward intersection in photogrammetry. 3) the six object points were linked to 15 line segments during permutation and combination, and the distances of the 15 line segments were accurately measured as a real value. 4) to validate and assess the accuracy of the result of relative orientation, the distances of the 15 line segments were calculated using the space coordinate of the six object points, and compared with the measured true value. then, the accuracy is assessed with the standard deviation. g5the model of direct relative orientation with seven constraints for geological landslides measurement and 3d reconstruction results and analysis for validating and assessing the accuracy of the result of relative orientation, the six object points (p1, p2, p3, p4, p5, p6) are selected and used to calculate the space coordinate basing on the image space coordinate of the left picture. the selected six object points are illustrated in figure 4. figure 4. six object points are selected for validating the accuracy of relative orientation. during the forward intersection, the space coordinates of the six object points was calculated with the result of relative orientation, the rotation matrix r, and translation matrix t. the space coordinates of the six object points represent in table 1 table 1. three-dimensional coordinates of 6 object points (unit: m) to validate the correctness and assess the accuracy of the proposed new model, direct relative orientation with seven constraints, the distances of the six object points calculated with their space coordinate were compared with the corresponding measured distances, and the error rates were calculated and represented in table 2. the first column shows the line segment composed of two points, such as (p1, p2), and the calculated distance and exact distance are in the second, third column, respectively. table 2. error rate of the observed distance respect to the distance in real (unit: m) in addition, the curve of the error rates is calculated and shown in figure 5. the blue square and curve represent the error rate of each line segment and the curve of error rates, respectively. figure 5. error rate of the distance. from table 2, it is evident that the minimum error of distances reaches 0.001 m and the maximum error is 0.0057 m. in the figure 5, the range of error rates is between the -3% and +3%. the standard deviation of the distance accuracy is calculated and reaches 0.0031 m. in the least squares adjustment, a wrong initial value will result in a very low precision of relative orientation, and even cause non-convergence or convergence to an error result. to overcome this problem, a reasonable initial value of parameters (by, bz, a1, a2, a3…c2, c2, c3) is calculated and acquired by direct relative orientation using a rough bx. in the procedure, the value bx can be an arbitrary value that is greater than zero. then, the initial value of parameters in the relative orientation are used in the model of direct relative orientation with seven constraints to calculate further and adjust the parameters with the least squares adjustment, improving and ensuring the accuracy and stability of relative orientation. owing to the seven constraints, the error rate of distance used to assess the accuracy of relative orientation ranges for from -3% to +3%, and the standard deviation of the distance reaches 0.0031 m that describes the range error composed of two points. therefore, the standard deviation of the point can be the half of distance error, 0.0015 m. furthermore, the point is three dimensions; thus the standard deviation in each direction is 0.0005 m for the relative orientation. for a non-calibration and general camera sensor, the accuracy of relative orientation is excellent basing on the model of direct relative orientation with seven constraints, and the residual error may include the image distortion, match error of conjugate point, measurement error of image point, calculation error and random error. the residual error is from various factors and needed to eliminate with other methods to improve the accuracy of relative orientation further. g6 yifu chen , zhong xie, zhenge qiu , min zhang and saishang zhong conclusion in this paper, a new relative orientation model, direct relative orientation with seven constraints is proposed and validated to improve the accuracy and stability of relative ordination and overcome the problems for the large oblique angle and translation and without the initial value. the constraints are derived from the inherent orthogonal property of rotation matrix and the vectors in space translation. also, the procedure for solving direct relative orientation with seven constraints consists of the acquirement of initial value with direct relative orientation, the least square adjustment with seven constraints, and calculation of six relative orientation decompositions. the rough error is detected with data snooping techniques and used to analyze the residues of observations and the variance and covariance matrix of parameters. during the experimental result and analysis, the model of direct relative orientation with seven can obtain a high accuracy than the conventional method in some cases, such as the large oblique angle and translation and without the initial value. in the cases, some result of the conventional method has very low accuracy or wholly wrong. however, the problems can overcome during the model of direct relative orientation with seven constraints. that means the proposed model can increase the applied field of relative orientation such as close range photography, low altitude oblique photography and 3d model reconstruction and measurement of the geological landslides with multiple images. acknowledgments this research was financially supported by the projects "the research of the trajectory model with weight based on kernal function" and "the key techniques of marine surveying and mapping based on the domestic high-resolution satellite". we also thank yuan le for their support during the entire course of this study. references duane, c. b. (1971). close-range camera calibration. photogrammetric engineering, 37, 855-866. faugeras, o. (1993). three-dimensional computer vision: a geometric viewpoint. the mit press, cambridge, ma, usa. hartley, r., & zisserman, a. (2003). multiple view geometry in computer vision. cambridge university press. new york, usa. horn, b. k. p. (1990). relative orientation. international journal of computer vision, 4(1): 59-78. läbe, t, & förstner, w. (2006). automatic relative orientation of images. proceedings of the 5th turkish-german joint geodetic days, 29, 31-39. li, z., liu, j., & tang, x. (2007). a closed-form solution to 3d reconstruction of piecewise planar objects from single images. ieee conference on computer vision and pattern recognition, 2007, 1-6. doi: 10.1109/cvpr.2007.383158 longuet-higgins, h. c. (1987). a computer algorithm for reconstructing a scene from two projections. in: fischler, m. a. & firschein, o. (eds.), readings in computer vision: issues, problems, principles, and paradigms (61-67). san francisco, usa. ma, y., soatto, s., kosecka, j., & sastry, s.s. (2004). an invitation to 3-d vision: from images to geometric models. springer-verlag, new york, usa. mikhail, e. m., bethel, j. s., & mcglone, j. c. (2001). introduction to modern photogrammetry. john wiley & sons inc. nistér, d. (2004). an efficient solution to the five-point relative pose problem. pattern analysis and machine intelligence, ieee transactions, 26(6): 756-770. pan, h. (1999). a direct closed-form solution to general relative orientation of two stereo views. digital signal processing, 9(3), 195-221. doi: 10.1006/dspr.1999.0344 philip, j. (1996). a non iterative algorithm for determining all essential matrices corresponding to five point pairs. the photogrammetric record, 15(88): 589-599. doi: 10.1111/0031-868x.00066 stewenius, h., engels, c., & nistér, d. (2006). recent developments on direct relative orientation. isprs journal of photogrammetry and remote sensing, 60(4), 284-294. doi: 10.1016/j.isprsjprs.2006.03.005 tang, l., & heipke, c. (1994). approach for automatic relative orientation. in: proceedings of optical 3d measurement techniques ii: applications in inspection, quality control, and robotics. international society for optics and photonics, 347-354. taylor, c. j., & kriegman, d. j. (1995). structure and motion from line segments in multiple images. ieee transactions on pattern analysis and machine intelligence, 17(11), 1021–1032. doi: 10.1109/34.473228 werner, t. (2003). constraint on five points in two images. computer vision and pattern recognition, 2. doi: 10.1109/cvpr.2003.1211471 zhang, y., huang, x., hu, x., wan, f., & lin, l. (2011). direct relative orientation with four independent constraints. isprs journal of photogrammetry and remote sensing, 66(6): 809-817. doi: http:// dx.doi.org/10.1016/j.isprsjprs.2011.09.006 earth sciences research journal geomorphology earth sci. res. sj. vol. 16, no. 2 (december, 2012): 87 94 the geomorphic history of the ainoura plain, kyushu, japan, based on excavation of the monzen ruins hirotsugu mori1, shoichi shimoyama1 and kazuaki soejima2 1 earth and planetary sciences department, kyushu university, fukuoka, fukuoka, japan. e-mail: hmori0622@gmail.com 2 sasebo property research office, education agency, nagasaki prefecture, sasebo, nagasaki, japan abstract the complex coastline of north-western kyushu, japan, consisting of cliffs, steep-sided inlets and archipelagos, has resulted from subsidence in mountainous terrain; the region’s populated river valleys, having narrow alluvial plains, often suffer rock avalanches. the ainoura plain is one such alluvial plain. excavation of the monzen ruins on the ainoura plain by the nagasaki prefectural board of education has revealed the plain’s geomorphic history regarding how the gravels and associated muds and sands were deposited during the last several thousand years. regarding alluvial succession, mud and sand first filled in the trough, possibly as part of the transgression caused by the holocene climatic optimum, and gravel later began to replace parts of the sand and mud layers by successive erosion and sedimentation, thereby raising riverbed level. it was confirmed that such conglomerate from the ainoura plain was derived from the hokusho landslide area in the hinterland because there is no conglomerate layer in the hino plain (a small plain southeast of the ainoura plain). the ainoura river’s thalweg shows that the monzen ruins are located at a sudden reduction in its gradient, thereby accounting for the thickness of coarse clastic material in that area. the reconstructed ainoura plain’s paleoshoreline during the holocene climatic optimum is also reported. the paleoshoreline did not move much until people began to reclaim land; it would thus be expected that the similar deposition would happened on the ainoura plain inside this paleoshoreline. resumen la compleja línea de costa en la región noroccidental de kyushu, japón,  está compuesta por acantilados, ensenadas de laderas inclinadas y archipiélagos, y es producto de la subsidencia de un cinturón montañoso. los valles aluviales y las estrechas llanuras de inundación de esta región se encuentran densamente pobladas y están sujetas a continuos movimientos en masa.  ainoura es un ejemplo de esta llanura de inundación. la excavación de monzen ruin realizada por  “nagasaki prefectural board of education” en la llanura de inundación ainoura revela la historia geomorfológica; la depositacion de gravas, arenas y lodos en los últimos milenios. en la sucesión aluvial, los primeros en llenar el canal fueron los depósitos de lodos y arenas, probablemente como resultado de la transgresión del óptimo climático que ocurrió durante el holoceno.  consecutivamente,  los depósitos de grava empezaron a reemplazar las capas de arena y lodo  mediante la erosión y depositación  sucesiva, elevando el lecho del rio. nosotros corroboramos que el conglomerado de la llanura de inundación ainoura proviene del deslizamiento de tierras de la zona interior de hokusho, ya que la llanura de inundación hino, ubicada al sureste de ainoura, no contiene capas de conglomerados. el lecho del rio ainoura muestra que el monzen ruin se localiza en la abrupta reducción del gradiente, lo cual explica el material clástico grueso en esta área. adicionalmente, nosotros reportamos la paleo-línea de costa de la llanura de inundación de ainoura durante el optimo climático que ocurrió durante el holoceno. la paleolínea de costa no se movió significativamente hasta que la población empezó a recuperar tierra. por consiguiente, nosotros predecimos que el mismo proceso de depositacion podría ocurrir en la llanura de inundación de ainoura al interior de la línea de costa. palabras claves: basalto de kitamatsuura, corrimiento de tierra de hokusyo, llanura de inundación, kyushu, transgreso keywords: alluvial plain, hokusyo landslide, kitamatsuura basalt, kyushu, transgression. record manuscript received: 03/10/2011 accepted for publications: 06/05/2012 hirotsugu mori, shoichi shimoyama and kazuaki soejima88 20 m 100 m 100 m 100 m 300 m 300 m 300 m 500 m 20 m kitamatsuura basalts ainoura river monzen ruin ainoura plain 129°38’e 129°42’e 129°46’e 33°10’n 33°12’n legend reclaimed land sand stone and conglomerate sand and conglomerate sand, mud, and tu� stone basalts fluvial sediment landslide sediment kitamatsuura basalt quaternary rough counter line remnants of landslides 1 km figure 1. topomap of the ainoura plain, sasebo, and its location in northern kyushu, japan. a detailed map around the monzen ruin is shown in figure 2. note that the higher portion of the mountain range is covered by basalts, and landslides have occurred around the boundary between the basalt layer and underlying sand, mud, and tuff layer. small islands other than kyushu island are not depicted. modified from hatano et al. (1974) and matusi et al. (1989). introduction the japanese archipelago is geologically active and therefore suffers from geological disasters; millions of people still live on the islands, coping with disasters. rock avalanches forming part of the process of sedimentation represent an example of such disasters. the hokusho landslide area is located in north-western kyushu (noda, 1957) and the region’s flood plains have suffered many rock avalanches. even so, the area is populated by nearly 400,000 people. understanding the region’s geomorphic history is thus important for predicting future disasters. the ainoura plain is one of the region’s small plains; several ruins have been excavated on this plain, indicating that people have been living on the ainoura plain for at least 6 ka. the monzen ruins represent one such site. the nagasaki prefectural board of education excavated the ruins the geomorphic history of the ainoura plain, kyushu, japan, based on excavation of the monzen ruins 89 figure 2. terrain classification in the ainoura plain showing the major rivers. the monzen ruins occupy a large portion of the narrow part of the floodplain surrounded by mountains and river terraces. mollusk shells were collected from the location indicated by the stars. and published a report (2006); this paper is partly based on their excavation work. their results were used to reveal the plain’s geomorphic history, including being struck by many rock avalanches. geological data the complicated shoreline of north-western kyushu, japan, has resulted from active subduction (shimoyama et al., 1999) combined with erosion. this area lacks flatlands, but there are occasional wide river valleys; the ainoura plain in western sasebo, in which monzen ruins are located, is one of these. figure 1 gives a geological map of the ainoura plain and vicinity. the most widespread stratigraphic units in the region are the sasebo group and the ainoura group consisting of paleogene sand and mud (matui et al., 1989). these groups lie beneath the 6-9 ma “kitamatsuura basalt” (which includes basalt and andesite; kaneoka and aramaki, 1971; matsui and uto, 1998). basalt and andesite cover mountain tops and the structure is unstable. erosion of fragile sand and mud layers causes landslides; accordingly, this type of landslide has been named a hokusho-type landslide by noda (1957) and a detailed landslide map was prepared by hatano et al. (1974a,b). the ainoura plain is an alluvial plain on the ainoura river; the hino river is a tributary of the ainoura river and these two rivers converge near ainoura bay. several ruins are present on the ainoura plain (figure 2). aso (1972) reported discoveries of houses, shell and animal bone middens and tombs from the early jomon era (c.a. 4,000 bc). 400 meters upstream of the shimomotoyama cave ruin, the shitanda ruins were built during the final jomon to middle yayoi era (c.a. 3,000 to 1,000 bc). sasebo municipal council (smc) excavated houses, paddy fields and stone tombs in this ruin from 1990 to 1995 (kumura, 1994). takebe c ruins were built on a hill and were excavated by the nagasaki prefectural board of education (npbe) (nagasaki prefectural board of education, 2008). this included the monzen ruins, located in the middle of the ainoura plain (n 33˚12’06’’, e129˚40’09’’, wgs84), where the ainoura river curves to the south (figure 2). npbe excavated the ruins and published a report in 2006 and 2008; they discovered archaeological remains from the initial jomon era (c.a. 5,000 bc) to the modern era, such as earthen vessels, shell middens, houses and tombs. the wooden tombs from the yayoi era (c.a. 300 bc) have special archeologically importance as they are some of the oldest wooden tombs found in japan. their excavation also unearthed clay pots, ceramic wares made in china and in the kinki region (near kyoto), as well as agricultural tools, such as a wooden spade. they also found at least five ancient channels from inside the monzen ruins and unearthed several archaeological items from these channels’ conglomerate deposits. most items from the ruins showed little evidence of abrasion. npbe and smc have conducted research on plant residues and phytoliths from the monzen and shitanda ruins. research on the monzen ruins indicates that the site was a moist environment surrounded by evergreen broadleaf forest. however, the amount of rice phytolith was not enough to affirm rice cultivation (npbe, unpublished). on the contrary, research on the shitanda ruins revealed rice phytoliths, indicating that rice might have been cultivated there (kumura, 1994). the non-abraded items and existence of the rice phytolith suggested that people were living on the plain. however, these ruins are characterised by having been built on an alluvial plain which has experienced several destructive avalanches. the chronology of japanese history, especially concerning when the yayoi era began, is an active area of research, currently lacking consensus (ozuka, 2000; onishi, 2005; yamamoto, 2007: nakahashi and iizuka, 2008). this research has been based on the conservative view that sets the start of yayoi era around 2,300 years ago (ozuka, 2000). methods a geological cross-section of the ainoura plain was first prepared along the ainoura river and the hino river from drill core data provided by smc, northern nagasaki prefecture development and promotion bureau and the fujinaga geological and construction corporation. a topographic cross-section was created based on a 50 m-mesh digital elevation model published by the geological information authority of japan (2002), using kashmir 3d software. hirotsugu mori, shoichi shimoyama and kazuaki soejima90 figure 3. geological and topographic sections along the ainoura (a) and hino rivers (b). drilling logs were used to draw the closed circles and triangles to draw a geological cross-section of figure 4; the former is along the transect ab and the latter is along the transect cd of the small topographic map. flood deposits in the ainoura plain could not be found in the hino river flood plain. the monzen ruins are located near the sudden reduction in the gradient. (a) the paleoshoreline’s approximate position was determined by using drill data: holocene deposits including marine shells and others which were continental. the location of the paleoshoreline was estimated from the intersection of the lowest marine deposits with the current terrain. the holocene marine layer’s maximum extension represents the paleoshoreline during the holocene climatic optimum (4000bc), when the sea level was at its highest. shell fossils were collected from the bank of the hino river (figure 2). the plain’s depositional history was investigated by plotting sediment layers’ elevation changes. the ages of layers determined by npbe (2008) was used mainly by means of seriation of potteries. when cultural objects were not found, 14c from fallen trees and tephra derived from the sakurajia volcano was also used. referring tephra helped to solve the problem of secondary deposition in some players. the elevations were adjusted by applying the modern gradient of the ainoura river near the ruins, (b) the geomorphic history of the ainoura plain, kyushu, japan, based on excavation of the monzen ruins 91 measured as 3.89/1,000 using an auto level. accordingly, 38.9 cm would be added to the elevation of a sediment layer 100 m downwards from the centre of the ruins along the paleoriver channel, for example. results a cross-section (figure 3) of the ainoura plain showed that the monzen ruins were located near a sudden reduction in the gradient and that the flood deposit layer was relatively thick here. figure 3a also represents the cross-section of the ainoura plain. from bottom to top, the plain consists of a paleocene unit, a non-marine pleistocene unit, a non-marine holocene unit, flood deposits after the holocene climatic optimum and human-made deposits. many basaltic conglomerates were included in this plain in the non-marine holocene unit. a geological cross-section of the hino river is shown in figure 3b. although layers similar to those of the ainoura river were found in this region, no flood deposits were observed. the following mollusk shells were collected from the bank of the hino river: granosa, macoma tokyoensis makiyama, anadotia stearnsina, pephia undulata, pistris capsoides, batillaria zonalis, cerithideopsilla cingulata, turbo cornatus coreensis, cryptonatica tigrina, cerithideopsilla djadjariensis, figure 4. the distribution of the drilling data and the estimated paleoshoreline of the ainoura plain. the paleoshoreline was estimated based on marine deposit distribution, shown as the intersection of the lowest marine deposits with the current terrain. the marine deposit was reported from locations marked with closed and open circles, but not from locations marked with closed and open triangles. the number in italics is the marine layer elevation at each location. hirotsugu mori, shoichi shimoyama and kazuaki soejima92 nassarius livescens and hemifusus ternatanus. most of these mollusks lived in the intertidal zone. the 4,000 bc paleoshoreline was estimated (figure 4). beside the drill data, a wave-cut notch found on a hillside was considered as evidence of the paleoshoreline. figure 4 shows that the monzen ruins were about 1 km away from the shoreline; the shoreline at the hino river was about 1 km away from the current shoreline. figure 5 shows how mud, sand and conglomerate layers developed on the plain. figure 5a show that sand and mud layer levels from the initial jomon era were 1 to 2 m lower than those of later periods. the levels of sand and mud layers rose suddenly about 4,000 bc, then stopped rising. figure 5b shows that levels of conglomerate layers have risen since the last 5,000 bc and four to five debris flows presumably brought the gravels striking the monzen ruins. discussion the ainoura plain’s paleogene unit consists of sand and mud layers having nearly horizontal stratification, covered by kitamatsuura basalts. landslides easily occur in such an area when underground water pressure rises. noda (1957) reported that the area has experienced many such landslides. comparing the cross-sections of the hino and ainoura rivers (figure 3) revealed differences in their depositional histories. considering the shell fossils from the hino river, it was inferred that the hino plain was originally an intertidal flat dominated by marginal marine sedimentation that subsequently became damp, subaerially exposed ground. by contrast, basaltic gravels from the ainoura plain must have either been derived from landslides near the ruins, or carried from farther upstream. since there is no evidence of landslides near the ruins (figure1), it was assumed that the basaltic gravels of the monzen ruins were carried by the ainoura river; fluvial influence is thus more pronounced than marine influence on the ainoura plain. a geological cross-section of the ainoura plain (figure 3) revealed that the monzen ruins were located near the sudden reduction in the gradient, and where the flood layer is thick. this suggested that such location was where the river’s flow rate suddenly drops and sedimentation occurs. these gravels did not begin to be deposited until 1,000 bc (figure 5b) at the location of the monzen ruins. gravel became successively deposited over the conglomerate layers after that, leading to the rising of the riverbed. such trend was dissimilar to that of mud and sand layers (figure 5a). mud and sand layers most likely responded to changes in river water level which was influenced by changes in sea level. yokoyama et al., (1999) reported sea level change in tsumizu, western kyushu, 50 km south-east of the ainoura plain. the sea level change there was characterised by a sudden rise in sea level around 5,000 to 3,000 bc and a subsequent gradual fall, similar to the sand-mud deposition pattern on the ainoura plain. since the geomorphic history of western kyushu is complicated (yokoyama et al., 1996; shomoyama et al., 1999), one cannot be certain whether the ainoura plain experienced the same change in sea level as at tsumizu, but it is possible that changes in ainoura plain elevation record/reflect changes in sea level in western kyushu. figure 4 shows that the paleoshoreline was located about 1-2 km seaward from the monzen ruins. after the holocene climatic optimum, the shoreline has been gradually prograding due to sedimentation; however, the cross-section (figure 3) showing locations between the paleoshoreline and the end of flood deposition is quite close. it was thus concluded that progradation did not alter the shoreline much, at least until the 17th century, when the japanese began to reclaim the land (nagata 1994). it can thus be expected that the deposition which happened at the location of the monzen ruins described above probably also happened in the region between the monzen ruins and the estimated paleoshoreline, accounting for almost half of the ainoura plain. conclusions this research was focused on the geomorphic history of the ainoura plain based on the monzen ruins. most of the strata of the ainoura plain consists of basaltic conglomerate (flood deposit) and sand. such basaltic gravels were derived from the kitamatsuura basalts in the hinterland; such gravel represents landslide debris carried by floods to the plain to be deposited. the geographical cross-section revealed that the monzen ruins were located at the sudden reduction in the river gradient, more flood sediments thus being deposited there. it was also found that mud and sand first filled in the trough, probably as part of the jomon transgression caused by the holocene climatic optimum. later, conglomerate began to replace sand and mud layers by successive erosion and sedimentation, thus raising the riverbed. around the final jomon era (c.a. 500 bc), depositional elevation levels of gravel surpassed those of sand and mud. considering the position of the paleoshoreline, such scenario is probably applicable to about half of the ainoura plain. acknowledgments data regarding the monzen ruins was kindly provided the nagasaki prefectural government’s excavation office, saseabo city municipal office and the northern nagasaki prefecture’s development and promotion bureau. the fujinaga geological and construction corporation provided the drilling core data. professor yoshiyuki tanaka from the social and cultural studies graduate school, kyushu university, suggested how to interpret wooden tombs. dr. ramues gallois from the british geological survey, dr. patrick druckenmiller and dr. rainer newberry proofread the paper. ms. susana salazar jaramillo prepared the spanish abstract. this research was financially assisted by emeritus professor tatsuro matsumoto takachiho scholarship, kyushu university. references aso, s. (1972): “shimomotoyama iwakage” otsuka kogei sya, tokyo, japan. hatano, s., okabe, f., and watanabe, y. (1974a): “compilation of 1:50,000 geomorphological classification map of landslide landforms in hokusho district, kyushu, japan”. notes of cooperative research for disaster prevention 32, 3-6. hatano, s., okabe, f., watanabe, y., and furukawa, t. (1974b): “a catalogue of large-scale landslide landforms formed in the past in hokusho (kitamatsuura) district, north-western kyushu, japan”. notes of cooperative research for disaster prevention 32, 7-23. kaneoka, i., and arakami, s. (1971): “k-ar age of plagioclase from a kitamatsuura basalt and the reliability of plagioclase k-ar age”. journal of the geological society of japan, 77. 6. 389-392. kumura, s. (1994): “excavation of shitanda ruin report” sasebo city, sasebo, japan. matsui, k., and utsu, k. (1998): “1:200.000 geological map “karatsu””. chishitsu news 521, 23-24. matsui, k., suda, t. and hiroshima, t. (1989): “1:200.000 geological map, nagasaki prefecture, ni-52-17”. geological survey of japan, tokyo, japan. nagasaki prefectural board of education, (2006): “excavation report of buried cultural properties involved in the construction of sasa-sasebo national route 497 vi: monzen ruin”. research report cultural property research office of sasebo, nagasaki prefecture, 190. nagasaki prefectural board of education, (2008): “excavation report of buried cultural properties involved in the construction of sasasasebo national route 497 v: takebe c ruin and takebe d ruin”. research report cultural property research office of sasebo, nagasaki prefecture, 3. the geomorphic history of the ainoura plain, kyushu, japan, based on excavation of the monzen ruins 93 figure 5. depositional history of the sand, mud (a) and gravel layers (b) on the ainoura plain. horizontal bars represent rough possible age ranges and vertical bars represent altitudes for layers connected along the river gradient. cross-section diagrams prepared during excavations were referred to for the elevations of the layers. the layers’ ages were determined by seriation, 14c dating and tephra. bars marked with an * are layers whose ages were determined by 14c and # by tephra. sand and mud layers’ depositional elevations rose around 5,000 to 4,000 bc, thereby agreeing with the change in sea level. the gravel layers began to be deposited around 1,000 bc, then already becoming replaced by deposited sand and mud layers. (a) (b) initial jomon early jomon middle jomon middle jomon final jomon kofunyayoi kodai mid-ageedo m od er n sea level legends mud sand conglomerate e le va tio n (m ) 4.5 4.0 4.0 5.0 5.5 6.0 6.5 7.0 7.5 8.0 e le va tio n (m ) age 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0 8.5 4 0 2 -2 -6 -8 -4 se a le ve l c ha ng e of w es te rn k yu sh u (m ) 5000 b.c. 4000 b.c. 3000 b.c. 2000 b.c. 2000 a.d.1000 b.c. 0 b.c. 1000 a.d. depositional trend of sands and muds depositional trend of gravels elevation range possible age range # # # * * hirotsugu mori, shoichi shimoyama and kazuaki soejima94 nagasaki prefectural board of education, (2008): “excavation report of buried cultural properties involved in the construction of sasasasebo national route 497 vi: monzen ruin ii”. research report cultural property research office of sasebo, nagasaki prefecture, 4. nagata, k. (1994): “evolution of land improvement districts in japan”. short report series on locally managed irrigation, 6. noda, m. (1957): “karatsu sasebo ryo tanden nai no jisuberi ni tuite (report on landslides in the coal field of karatsu and sasebo)”. kyushu kozan gakkai shi (journal of mining engineering of kyushu), 25, 11, 443-452. nakahashi, t.and iizuka, m. (2008): “anthropological study of the transition from the jomon to the yayoi periods in the northern kyushu using morphological and paleodemographical features (2)” anthropological science (japanese series), 116, 2, 131-143. onishi, s. (2005): “a review of the recent debate about the date of yayoi period”. anthropological science (japanese series), 112, 2, 95-107. ozuka, s. (2000): “desk-sized chronological table of japanese history augmented edition” iwanami books, tokyo, japan. shimoyama, s., kinoshita, h., miyahara, m., tanaka, y., ichihara, t. and takemura, k. (1999): “mode of vertical crustal movements during the late quaternary in kyushu, japan, deduced from heights of ancient shorelines”. tectophysics, 302, 1-2, 9-22. yamamoto, n. (2007): “archaeology integrates the humanities and science” takashi shoiun, tokyo, japan. yokoyama, y., nakada, m., maeda, y., nagaoka, s., okuno, j., matsumoto, e., sato, h. and matsushima, y. (1996): “holocene sea-level change and hydro-isostasy along the west coast of kyushu, japan”. palaeogeography, palaeoclimatology, palaeoecology, 123, 29-47. in this study, we re-examined the official hurricane database from the national hurricane center (hurdatnhc), an agency associated with noaa, for tropical cyclone activity from 1851 to 2012for the dominican republic on the island of hispaniola in the caribbean basin. we performed analyses at two different levels for the island (i.e., all of the storm tracks in the caribbean basin near to the study area that made landfall and all of the events that crossed the dominican republic from a radius of 300 km from the coastline). this study includes the statistical occurrence of these phenomena during the study period and the climatological analysis of all tropical cyclone tracks (112 total events) by decadal seasonal distribution, fifty-year seasonal distribution and monthly seasonal distribution to show the lowest and highest activities. we performed wavelet analysis on the continuous data over a long time series to determine the important frequencies. this analysis provided a general statistical conclusion resulting from the data collected. a landfall probability for the study area corresponding to the long time series of (it’s 162) years within a radius of ~100, ~185 and ~300 km, based on the historical climatology tropical cyclone tracks, reveals the likelihood of a strike for a major or a minor hurricane. we present a review of the tropical cyclone activities that passed the dominican republic, which also forms part of the author’s dissertation. abstract keywords: tropical storms, statistical analysis of tcs, dominican republic hurricanes, caribbean sea hurricanes. quantifying the statistical distribution of tropical cyclone activity in the historical record of the dominican republic cuantificación de la distribución estadística de la actividad de ciclones tropicales según el registro histórico para la república dominicana issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n4.63425 earth sciences research journal earth sci. res. j. vol. 21, no. 4(december, 2017): 159 167 santo trinidad alvarez ysabel1, gustavo adolfo agredo cardona2, david felipe rincon3 1 dr. sc. candidate at dehma, upc-barcelonatech santo.alvarez.ysabel@gmail.com 2 school of sustainability and urbnism, faculty of engineering and architecture, unal-manizales, colombia 3civil engineering department, unal-manizales, colombia para el presente estudio se ha tomado la base de datos oficial para huracanes del centro nacional de huracanes (hurdat-nhc), dependencia de la noaa, correspondiente al período de 1851-2012 relativos a ciclónicas tropicales para la república dominicana (isla hispaniola) en la cuenca del caribe. el análisis ha sido realizado de dos formas diferentes considerando la geografía de la isla (todos los puntos de posicionamiento de los eventos en la cuenca del caribe que han impactado el área de estudio al tocar tierra –impactado-y todos los eventos que han atravesado la república dominicana en un radio de ~300 km. se presenta un estudio estadístico de la ocurrencia de estos fenómenos durante el citado período, incluyendo un análisis climatológico de todos los ciclones tropicales que se han registrado (112 eventos en total) en distribuciones decanales, cincuentenarios y mensuales, indicando la de menor y mayor actividad. por igual se ha ejecutado un análisis de ondeletas sobre los datos continuos de la serie de tiempo para determinar importantes frecuencias. finalmente mostramos una probabilidad de ocurrencia de que uno de estos fenómenos toque tierra basado en una serie de 162 años para el área de estudio, dentro de un radio de ~100, ~185 y ~300 km, teniendo en cuenta la climatología de huracanes la cual revela la probabilidad de ser impactado tanto por un evento de menor o mayor categoría. el autor muestra una perspectiva de la actividad ciclónica para el área de la república dominicana la cual forma parte de su tesis doctoral. resumen palabras clave: tormentas tropicales, análisis estadístico de los cts, huracanes en república dominicana, huracanes en el mar caribe. record manuscript received: 22/03/2017 accepted for publication: 25/08/2017 how to cite item alvarez ysabel, s.t., agredo cardonag.a. &rincon, d.f. (2017).quantifying the statistical distribution of tropical cyclone activity in the historical record of the dominican republic. earth sciences research journal, 21(4). 159 167 doi:http://dx.doi.org/10.15446/esrj.v21n4.63425 h y d r o m e t e o r o l o g y 160 santo trinidad alvarez ysabel, gustavo adolfo agredo cardona, david felipe rincon 3. data and methodology 3.1. the hurdat2 historical record database historical records of hurricanes for the dr started at the beginning of the sixteenth century, shortly after the foundation of santo domingo´svillages, which had been moved and reconstructed on the west bank of the ozamariver after they weredestroyed by a hurricane figure 1. map of the study area. 1. introduction tropical cyclones are alsoknown as hurricanes or typhoons and are among the most deadlynatural disastersonthe earth. the caribbean countries are highlyvulnerable to these natural hazards. the latent vulnerability of the area exacerbates the impact of these hazards, which become natural disasters inthis geographic region andaffect the economic, social and environmental conditions. the north east atlantic (nea) and the caribbean sea basin (csb) are characterized as a high probability zone for the formation, development and impact of tropical cyclones (tc), according to the national ocean and atmospheric administration (noaa). millions of people are affected, and economic losses ofbillions of us dollarsare estimatedto occur in the caribbean regionover the past 30 years. certain island states in the region are susceptible to earthquakes (3% of the population), but they are more susceptible totropical cyclones (70% of the population) and floods (27% of the population) (nhc-noaa, 2012; cepal, 2010; gutierrez, m.e. &espinosa, t., 2010; pnud, 2008; unuwider, report paper 2008/61). the hurricane activity in the atlantic increased over the last decade of the xx and beginning of this century (xxi) and had significant consequences, such as heavy rain and theassociated flooding, deaths, damage and destruction ofcoastal and inland areas throughout the north atlantic basin (nab). the csb territories are well known as a tropical cyclone pathway because ofthe surrounding environmental flows associated with the location of the islands. this basin has experienced above normal tc activity, ranging from tropical depressions (td) to major hurricanes (mh) of category-3 to category-5 (h3-h5) on the saffir-simpson scale, as shownin table iii (karmalkar a. v. et al., 2013; ipcc, 2007 and 2012; steve graham and holliriebeek, 2006; pielke et al., 2005). these phenomena are the most lethal and costly natural disasters in the area. major events have occurred in the tropical cyclonecorridor(san zenón (1930), david (1979), georges (1998), jeanne (2004), noel and olga (2007), emily and irene (2011)) are among the hydro-meteorological events that have impacted the dominican territory and producedconsequences detrimental toall areas of society, which isevidence of the urgent need to assess their effects not only by the magnitude of thewinds but alsoby extreme precipitation, which can be properlyconfirmed by the recordings in thegrid-unep database (ofda/cred, 2012; campusano, m., 2009; cocco q., a., 2009; pnud, 2007 and 2008; cepal, 2008;pnud, 2007; unep-grid, 2003). a tc is a storm system (an atmospheric disturbance) that is characterized by a low-pressure center and numerous thunderstorms that produce strong winds and heavy rainfall. tcoriginates over tropical oceans and isdriven principally by heat transfer from the ocean. approximately 80 tropical cyclones develop globally each year over the ocean atsea surface temperatures(sst ≥26ºc) and very oftenmovefrom these regions into higher latitudes (equator-ward of 5º latitude north atlantic basin [nab]). once they are developed andorganized, the storms tend to translatewestward and slightly pole-ward.the commonly used categorized scale referred to in this articleis given in the appendix. the frequency and intensity of hurricanes and tropical storms vary significantly from one year to the next (graham and holliriebeek, 2006; emanuel, k.2003). hurricane climatological analysis can provide a useful tool forstudying the potential damage along the coast or landmass of a country, particularlyin populated areas. damage tends to increase in future years or along river valley floodplains,ifthe heavy rainfall associated with a cyclone passes over at least one of the five principal mountains ranges in the area and is intensified by these orographic systems (gutierrez, m.e. & espinosa, t., 2010; ministry of environment, 2010). hurricane georges in 1998 and the tropical storms noel and olga in2007hadpeakextreme heavy rains of ≥ 500, 330 and316 mm/day, respectively, which accumulated in some mountainous areas of the dominican territories andcaused more than 457 deaths (onamet, 1998; 2007; trmm, 1998; 2007; em-dat: ofda/ cred, 2012). we used an empirical study method to investigate the statisticalclimatology of tcactivityin the dominican republic (dr) territories.the tcsconsidered for this study were bothmajor (h5with awind speed >241 km/h) hurricanes and minor (≤ 63 km/h) tropical storms or depressions (ts/td) that made landfall or approachedwithin ~300 km of the study radius. a short overview of the study area is given to provide a better understanding. the article isorganized as follows:(a) the data and methodological method are presented and are followed by (b) the results and discussion derived from this work and(c) the conclusions. 2. study area the study area is the dominican republic (dr), a country on hispaniola island, which is shared with therepublic of haiti and is both a member ofthe so-called greaterantilles islands andpart of the tropical cyclone caribbean basin, whichforms the nab when the gulf of mexico is included. the dominican republic is located between 17º 36´14 and19º 55´57 north latitude and 68º 19´24 and72º 00´33 west longitude.the country is located on the edge of the tropical zone of the northern hemisphere and is bordered on the north by the atlantic ocean, on the south bythe caribbean sea,on the east by themona passage and on the west by the republic of haiti (fig. 1). the surface area is 48,670.82 square kilometers, with acoastline of1,228 kilometers(cia-the world factbook, 2012; ministry of environment, 2010;fao, 2000). the dominican republic hasa tropical maritime climate with little seasonal temperature and two main rainy seasons presented: the frontal-convective season (november-may) and the cyclonic season (june-october). areas of high precipitation are highly influenced by moisture-laden trade winds over the atlantic ocean that reaches the country from the northeast, producing so-called orographic rainfalls. the terrain consists of rugged highland mountains (five main hilly orographic systems) withinterspersed fertile valleys irrigated by natural,well-spread hydrographical canals and rivers (cia-the world factbook, 2012; ministry of environment, 2010; fao, 2000). 161quantifying the statistical distribution of tropical cyclone activity in the historical record of the dominican republic in 1502. several sources of information (e.g., historians, researchers and universities) indicate where storms and hurricanes have hit hispaniola island, although they have several omissions due to some events that followed the trajectories ofthe storm masses onthe ocean surface (cocco q. a., 2009). however, in the present study, we do not usethehistorical tropical cyclone tracks prior to the 1851 yearly recordsconcerning the pathways or approaches of those events that affectedthe drterritory. this is the case for noaa, who supplemented theirdata collection and with other sources (local or regional) to gather better data and information. ouranalysis is basedon the most recent official worldwide hurdat database, from the national oceanic and atmospheric administration (noaa´s coastal service center), for the atlantic tropical cyclone activitybasin, including the north atlantic basin, the gulf of mexico and the caribbean sea. this database has beenreanalyzed as part of the atlantic hurricane database re-analysis project, as documented bylansea et al. (2008). this study incorporates the reanalyzed data of the long time series from 1851 to 2012 (historicalhurricanes tracks: http://www.csc.noaa.gov/ hurricanes/#) as part of the climatological calculations for the tc activity (updated every year as the “six-hourly best track”(00:00, 06:00, 12:00, 18:00 utc)once the hurricane period from june to novemberhas passed andincludes theyear, month, day, hour, name, latitude, longitude, storm movement speed and direction, wind speed and central pressure, storms by location, name and basin. however, prior to the satellite era (1970), limitations exist for the determination of storm tracks, intensities, and landfall areas (table i, appendix section) the best track data fromthe atlantic hurricane database (hurdat2) can be downloaded in ascii format using the spreadsheet software contained in the office version of the microsoft office suite (ms excel v.2007).the geographical information system (gis) version of the atlantic tracks file database (available at: http://csc-s-maps-q.csc.noaa.gov/ hurricanes/download.jsp) wasused toperform the computational analysis and generatemapsusingthe gis-arcmap software from the environmental systems research institute (esri), a tool that hasmanyspatial calculation capabilities (klotzbach, 2011; chen, k. et al., 2009; landsea, 2004, 2007). more information is available in appendix i. as revised by landsea et al. (2007), the atlantic hurricane database (hurdat)contains estimatesofthe maximum sustained surface wind speed at the conventional altitudeof 10 m elevation. to recompile the historical tracks for all tropical cyclone storms, this analysis considers wind strength speed superior to 37 km/h (td/ds = tropical depression or depression storm) usingwind speed descriptions based on the saffir-simpson hurricane scale (table 2 in the appendix section). the chart is color-coded to show theintensity for each tropical cyclone by category based on the saffir-simpson scale. the graphs obtained as a result of the statistical analysis show classificationsaccording to the same criteria. the statistical computations from 1851 to 2012 forall tropical cyclones consider the multidecadal, fifty-year (pentads), monthly and general seasonal distributions and other statistical parameters (e.g., tendency, probability). this methodology (unless the contrary is indicated) considers aradius of approximately300 km from the landmass (santo domingo as a single point) to the open sea (the atlantic ocean to the north and the caribbean sea to the south), considering the upper radius limit of ~370 km permitted by the interactive webpage. this radius includes the tracks of all cyclones with a major effectscaused by both direct and indirect impacts(e.g., flooding, wind force, storm surge) during the study period for thedrland and coastal-sea territories.it is important to mention thatin some cases, the tropical cyclone tracks wererecorded in adjacent or nearby countries such as cuba, haiti, puerto rico (pr) and the turks & caicos islands. thus, the orientation of the track, the movement speed of the event, the degree of upslope, the intensity at landfall or approach and several otherparameters are consideredforthe open sea limits of the territories of these countries,as previously mentioned (figure 1). 3.2methodology forlandfall probability calculations we used the method proposed by philip klotzbach and william gray et al. (2009) at colorado state university (csu) forthe statistical occurrence analysis toinvestigatethe relationship between the frequency and probability of the events. statistical analysis wasalso used to determine how often a tc will occur near the study area at radii of ~100, ~185, ~300 kmfrom santo domingo as a single pointand the return period in years (t)or the occurrence for 1, 3, 5, 10, 20, 25 and 50 years based on the longer period of tc tracks in the 161-year dr hurricane climatological datafromhurdat2(klotzbach, p. and gray, w. et al., 2009).these distances are justified because events such as ts have produced effects asharmful and severe as hurricanes (the introduction highlights some of these), for which therainfall, not thewind strength, is responsible. philip klotzbach and william gray methodology display the probabilities of a tropical cyclone passing within a certain radius. it allows the probability of a tcs passing within a specified distance of an area in any particular year to be calculated. the methodology facilitates the awareness of the chances of the closest tropical cyclone occurrence that passes or impacts in a particular point in the study area. to approximate the future likelihood of storms, a poisson regression model is used because the analysis of the number of tropical cyclones that made landfall over the last century shows that the landfall frequency closely conformed to this distribution. further descriptions and details related to this topic arepresented in the appendix (section iii). 4. results and discussion the climatological analysis of all of the tropical cyclone tracks involved studying the 112 total landfall or open ocean approach events in the study area. the early historical hurdatrecord has been regularlyrevised (lansea et al., 2004, 2007). we wanted to address questions such as which parametersdetermine the variation in the number of tcs. is the variation due to sst and wind shear? if so, does combination of enso and noadrive it? what is the probability of ahit by a major hurricane? we must keep in mind that tropical cyclones were not formally named before 1950 and are referred to as “unnamed” prior to that date. nevertheless, it was common to name a storm for an important person inthe region based on any relevance the person had during the time of the storm. for example, inthe early xx century, some storms were called thefather ruiz hurricane, lily’s cyclone, saint ciriaco’s hurricane and saint zenon’s hurricane (cocco q., a., 2009). using the data related to the tropical cyclone tracks for the period of 1851 to 2012, we can subdivide the long time series into i) a decadal distribution, ii) a fifty-yeardistribution and iii) a monthly seasonal distribution. in the same vein, (i) a general summary of all tracked events is presented. 4.1. multi-decadal distribution for the first decade in the middle of the 19th century, the occurrence of tropical cyclones was low but the occurrencefor the country was4, which included 2hurricanesof category h1orh2 and 2ts. three made landfall, but the rest remained in the atlantic off the northern coast.tropical cyclone activity was rarer duringthe second decade,from1861 to 1870, when only one storm was recorded. a unique hurricane of category h1 hit the entire island (hispaniola), moving from east to west along the south coast of the caribbean sea. afterit passed thepr island, itendedinthe neighborhood of haiti as a weak tropical storm (ts). greatertc activitywas evidentfrom 1871 to 1880 than during the two preceding decades. the trajectories offive tcbrushedthe island’s south coast: 4ts and 1h1. the decade from 1881 to 1890hadmore frequenttcactivity, with 6 events; a minor h2, amajor h3, and 4tswererecorded.the final decade of the 19th century showeda considerable increase intc activity in the region, with eleven events impacting the study area, whichwas 162 santo trinidad alvarez ysabel, gustavo adolfo agredo cardona, david felipe rincon theperiod with the most events everrecorded in the history of thedr. the events included3minor storms between h2 and h1,2majorh3 events, 3tsand 1td. most of the stormtrajectorieswere east to northwest or east to west and had significant impacts overthe territories of pr and haiti. the first decade ofthe 20th century from 1901-1910 had similar characteristics to the last because ten events were registered: 4h1,1h3,4tsand 1td (figure3a). fivetcsoccurred from 1911 to 1920: 2h1, 1h2 and 2ts. two of these events made landfall in the country onthe easternedge of the island (figure3-b). the decade from 1920-1930 was the first time that all tc types were registered in the study area,ranging from an insignificant tropical storm to an h5 major hurricane that made landfall over the dr,an h4, and an h2that nearly made landfall. these events all hadtrajectories from east and to eastnorthwestonboth coastlines (cbs and nab). the trend during 1931-1940 for both coastlineswasquite opposite to the previous decade. a total of sevenminortropical storms (7-ts)and only one hurricane (1-h1)were observed.by the end of the 1941-1950 decade,six events had occurred: three hurricanes (2 h1 and 1h3) and three tropical storms ofall categories (2ts and 1td). it is important to note that 1950was the first year in which a namewasassigned to astorm, which occurred on august 28 (a weak depression). the nhc began to name each seasonal tropical cyclone in1949. the decadefrom1951 to 1960 showed seven total events; threewereh1, h2 and h3storms, andthe remaining four included3ts and 1td. the decade from 1961 to 1970 showed the highesttc activity in both quantity and intensity in the study area. seven major events were trackedin the historical record;fourwereclassified as major hurricanes, and another as a minor hurricane, with1h5, 2h4, 1h3, 1h1, and there were two minor storms classified as 1ts and 1td. six tropical cyclones occurred during the decade 1971-1980, and one was a major category h5hurricane. the rest were classified as minor stormsof intensity typests (3) and td (2). the most dangerous storm, due to its catastrophic damage all over the dr,was the remarkable hurricane david. david occurred on august31, 1979, which caused that name to be erased from the nhcassignments. this major hurricane reached the maximum intensity on the saffir-simpson scale and had an enormous impact in terms of the consciousness of the effects that a tc can have on a country. thereafter, the government, institutions, agencies and civilians put deep emphasis on warnings, preparedness and protection from the impact of any storms approaching the landmass or surrounding sea area. this event was a hallmark in the dominican republic tc historical record. little stormactivity was recordedfrom 1981 to 1990, with three events recorded, including one major h3hurricane named emily thatimpactedthe entireisland, including both thedr and haitionseptember 23, 1987. the other two tropical events were1tsand 1td (figure 4d). six major events were recordedat the end of the 20thcentury,1h3and 2h1, and three tropical storm events classified as ts (1) and td (2). the major hurricane georges occurred onseptember 22, 1998, andcaused great damage tothe country’seconomyand society, as measured by loss of life (figure 4e). the first decade of the 21stcentury plus two more years (2001-2012) make this period were very busyfor the dr. twenty events were well categorized and tracked from sea coastal services (scs-noaa´s database) and were classified as follows: 1h5, 1h4and 5h1 hurricanes, along with13tsevents that were weak storms. this decadehadthe greatestquantity ofnamed and mean events: 1.78, or approximately2 peryear. although some eventssuch asjeanne, dean, tomas or irene that impacted the countrywere hurricanes,the so-called tropical storms alpha (2005), noel and olga (2007),similar tothe previously mentioned hurricanes david and georges, shocked and shattered the confidence of society due tothe tremendous and devastating environmental damage caused by the extremely heavy rains that occurredover the mountainsandpoured out over the floodplain valleys. we developed a general summary to visualize the entirety of thetc tracks recorded from 1851 to 2012. it was sub-dived by decadeand tropical cyclone category, i.e.,major hurricanes (h5-h3), minor hurricanes (h2h1) and weak storms (ts-td)to allow better visual comparison and comprehension. consideringstatistical decades by cyclone classification, the periods of 1871-1880 and 1901-1910 showedhurricane activity of four (4) minor hurricanes (h1). the periods of 1891-1900 and 1961-1970 showedmajorh3 and h4hurricane activities, with two of each category. the decades of1931 to 1940 and 2001 to 2012 had the highest amount of tropical storm events (ts),withseven (7) and thirteen (13) respectively (figure 2a). figure 2. statistical analysis results for the decadal seasonal distribution of all tcevents from 1851 to 2012. figure 1. map of the study area. 4.2. fifty-year distribution a seasonal fifty-year distribution wasperformed, and the results obtained from the statistical analysis allowa comparisonoftropical cyclone behavior over162 years of activity. the results infigure 2bshow that the highestfifty-year activity wasthe period from 1901-1950,which had 36 total tropical cyclones with 5 major and 12 minor hurricanes. the fifty-year activitydistribution from 1951 to 2000 showed29 total tropical cyclones anda slight increase inmajor hurricane activity,with8events categorized as h5 (2), h4 (2) and h3 (4). the final period considered wasthe ongoing fifty-year activity distribution of at least twelve years and showed 20 tctracks,with seven (7) categorized as both major and minor hurricanes. the tc activity from 1851 to 2012, as summarized by the fifty-year activity distribution,containedall storm types,was characterized bytropical cyclones from h5 to tdand showed that 49 tswerethe most frequent events affecting the dr territories, followed by 27h1storms for the entire 161-year period for which storm tracks were recorded. when the50-year distributionwas consideredand applied toboth types of events,a considerable numberof recorded tracks (10h1 and17ts)areobtained. for the recent period from 2001 to 2012, ahighernumber of ts (13) was recorded in thisshorttime. statistics can indicate the overall amount of activity or theprobability that the dominican republic territories will be impacted directly or indirectlyby a major or minor hurricane. 4.3. monthly seasonaldistribution a distribution of all events in the historical tcsfrom 1851 to 2012 for theeventsduring the 162 years consideredwascalculated on a monthly basis. the hurricane season for this basin runsfrom june to november, and we wantedto determinate which months of the season were more or less active. september wasdetermined to bethe month with the most activity, with 48 total events, followed by august,with 33 events (figure3a). when the saffir-simpsoncategory of the events was consideredby month, september took first place, with 10 major hurricanes (between h5 to h3), 18 minor hurricanes (between h2andh1) and 20weak tropical storms (ts/ td). june hadthe lowestnumber of events,withzero eventsrecorded. july and november hadan equal numberof events(4ts/td).rare cases of nab tropical cyclone activity with tracked events occurred as early as may (2 events) and as late asdecember 3 events(figure 3a). 163quantifying the statistical distribution of tropical cyclone activity in the historical record of the dominican republic based on the climatological historical tropical cyclone tracks recorded from 1851 to 2012 (162 years),we performed a landfall probability analysis for the region in radii of approximately100, 185 and 300 km.the climatology is going to be biased low because of lack of satellite and other technology going back farther in time. it is suggested that the percentages are going to be biased low, because of the under sampling with time. after fixing the measured radii and the total quantity of events tracked for each distance, the previously mentioned csu methodology wasappliedto analyzethe computational statistical data results. the probability values for 1 or more named storms tracking for ~100-, ~185-, and ~300-kmradiiand their associatedoccurrence yearsare shown in table1. figure 3. (a) the seasonal monthly distribution chart and (b) cdf from 1851 to 2012. figure 4. (a) and (b) general historical tropical cyclones events greater than category h1and a statistical analysis of the period, type and area of study. (a) (b) figure3bshows the cumulative distribution-dates of the first tropical cyclone formed that passed through the study area. for the study area it shows that for an approximate 300 km radius, the first cyclones (hurricanes) formedprior tojune 15 with a probability of approximately20 to 25 (%)since 1851. for a half year, september shows the peaks with 60% of thehurricanes occurring. the distribution appears to bebi-modal (i.e., with two peaks), especially for tropical storms (ts), withpercentage values of95% and 100% for july and november respectively. 4.4. general distribution a general trend can be observed for thedata collected for the period of 1851-2012 in this analysis. the majorityof tropical cyclone events tracked duringthis long time wasts (49) and h1hurricanes (27) (figure 4a, b). hurricanes of type h3ledall major types of hurricanes, with 10 total events. the trajectories were in all directions, but moving from east-southeast to west-northwest isthe most frequent direction in this tropical cyclone pathway and could impact any geographical point in the dr territories. table1. storm landfall probabilities based on 1851-2012 tcclimatology for dr (162 years) based on the csu methodology (2009). 164 santo trinidad alvarez ysabel, gustavo adolfo agredo cardona, david felipe rincon figure 5. probabilities of mh occurrences (impact or landfall) directly or indirectly in the dr territories based on the time series from 1851 to 2012. figure 6. wavelet power spectrum of the time series of the total number of tcs over the region from 1851 to 2012 with relative 8-year cycles. the cross-hatched region is the cone of influence with a 10% significance level. the historical series generated a tc record of 162 years in length and the basic statistics and variability were evaluatedfor a total of 112 tcs over the region, witha mean of 0.7 events/yearand a standard deviation of 0.9.the relative 95% confidence interval for the mean is (0.9, 1.4).no landfalls within the radius considered from inner mass of the country were observed in 86 (53%) of the 162 years analized, whereasfour tcs madelandfall in76 years (47%) of the analyzed period. a basic spectral analysis of the time series resultedinthe wavelet power spectrum shown in figure 6 (a,b). this wavelet analysis indicates that throughout the time series,the power associated with these phenomena had a 2to 4-year period.however,most strinkingly,a large peak in power with aperiod between 8 and 16 yearswas evident,whichbegandeveloping towards the beginning of the 20th century. this analysis will give an opportunity to confirm the above conclusions from the literature examining the csb and nab becauseit has beendemonstrated that the sst anomalies in this region havesignificant globalimpacts on the climate (klotzbach, 2011; cited from barnston et al. 1997), thus addressing the variability in climatological tcsinthe study area. first, the tcsthat form in the study area haveat least impacted an area in this basin. second, the seasonal shift in the areas in which thetcsdevelop is quite distinct by month: at the beginning of the season (june and july), development is restricted to the western caribbean because the ssts are still cool; later, by august and september, the development shifts eastward to the central and eastern atlantic as the ssts become warmer and the vertical wind shear over the basin decreases.finally, development returns to the western part of the basin during october and november. the third aspect is the origin of the disturbances thatbecometcs. a large portion of tcs form from easterly waves, which are atmospherictroughs spawned over the african continent underunique set of circumstances. development in the region frequently occurs at latitude of 20º n, which is the only tc basin where this happens. the surrounding variation in the number of tcs is due to the sst and wind shear driven by a combination of enso and noa, which accounts for the nature of the climatological tc variability in the study area (jury et al., 2011; klotzbach, p.j., 2011; c. mcsweeney et al., 2010; christopher c. hennon, 2008). 5. conclusion based on how close the center of a storm came to the island, the so-called closest point of approach for the study area during the period of 1851-2012 (162-year period) of tcs tracks recorded was calculated. as a result, the following conclusions are presented: 1. the statistical analysis method successfully produced time series of the number of tcs that affected the dominican republic territories from the lowest (td/sd) to the highest (between h3-h5) categories (in the official star hurricane seasonal tcs yearly activity). although the drhas been exposed totcs prior to the development of scs/nhc-noaa database, the analysis showed thedecadewith the greatestactivity, the fiftyyearperiod with the greatest activity and a time series showing a monthly distribution for these events over the entire161 years. 2. the csu methodology revealedthe likelihood of an impact for a major or minor hurricane both ata distance and close to the study area. the population in these vulnerable areasmust be aware of and ready for the storms. this topic requires further study. 3. factor as tropical atlantic sst, the size of the atlantic warm pool, and tropical atlantic sst gradients and low-level trade wind strength, play a critical role in affecting the levels of the nab and csb hurricane activity (jury m.r., 2011; klotzbach, p.j., 2011; shieh, o.h. and colucci, s.j., 2010). on amulti-decadal time scale,due tothe rest of the atlantic multidecadal oscillation (amo) and alterations inboth the size of the atlantic warm pool and the phase of the meridional mode, the droscillation plays a significant role in the caribbean tropical cyclone activity (jury m.r., 2011; klotzbach, p.j., 2011). 4. quantifying the tc landfall probabilities in the region is important for both seasonal cyclone forecasting and disaster preparedness. thiswill allow governments, planners and emergency agencies to improve forecasting and preparedness to manage the relative risk for any given event ina tc season in the dr. a graph of these results was plotted (figure 5),andthe corresponding equationfor the trendline was calculated. we can infer thatastrongand statistically significant correlation exists betweenthe number of storms (probability) and time (occurrences) in years for both mh within approximately 300 km and 185 km, with a r2 value of approximately0.973. apotential for more storms exists when tropical cyclones appeardistantfrom the landfall mass or are inland,even in years when the storms are very close, the totalprobability of cyclones can be well below normal.in addition, when cyclones seem far away,at least one or two will pass through the study area during those 20 to 50 years. only one ortwo stormscanspoil a whole season. 165quantifying the statistical distribution of tropical cyclone activity in the historical record of the dominican republic studies by the ipcc (intergovernmentalpanel for climate change) indicated that there is low confidence that any reported long-term (centennial) changes in tropical cyclone characteristics are robust, after accounting for past changes in observing capabilities.the increased atlantic hurricane activity over the last decade of the 20thand the beginning of this century has hadsignificant consequences, such asa rise in sea level, heavy rain and associated flooding, the loss of human life, damage and destruction to infrastructure and ecosystems, and economic losses both in coastal and inland areas along the whole nab (ipcc, 2012 and 2007; bid, 2008; cepal, 2007; steve graham and holliriebeek, 2006).some researchers attribute the recent increase in tc activity to human actions caused by global warming, but others attribute the increased activity to the natural multi-decadal seasonal phenomena that occur in a similar manner to natural earth interaction cycles (geo-physical) activity, neglecting a consideration of the anthropogenicallyinduced rise inglobal ocean surface water temperatures (karmalkar a. v et al., 2013; ipcc, 2012, 2007; steve graham and holliriebeek, 2006; pielke et al., 2005). the csb territories are well-known tropical cyclone pathways becauseof the surrounding environmental flows associated with the location of the island of hispaniola. in general, thisbasin has experienced above normal tc activity from tropical depressions (td) to category-3 to category-5major hurricanes (mh) (h3-h5) on the saffir-simpson scale. although the probability of a major hurricane strike is lower within the approximately300 km-h/ fifty-year period (80% 100%), the probability of minor storms in the category of ts/td force winds is higher (80% within a decadeand approaching 100% for t = 20, 30, 40 and 50 years). although debate tends to concentrate on the reason for the recent increase intc activity to satisfy both the common people and the scientific community,atrulyinteresting taskremainsafter this statistical analysis of tcsfrom 1851-2012forthe dr.the opportunity for the planned development of the territory is a huge challenge.the environmental heterogeneity and human sustainability must be balancedin terms ofachanging climate and the vulnerability of the caribbean states (mcsweeney, m.n. and g. lizcano, 2010; unfccc, 2007 and 2008; pnud, 2008; cepal, 2007). acknowledgment the authors would like to acknowledge manuel gómez valentín, sánchez martí-juny and víctor o. rendón d. for their suggestions. special thanks also go to the reviewers for the helpful comments that contributed to improve the manuscript. references aoml (american oceanographic and meteorological laboratory), 2012. hurricane and tropical meteorology. discharge on: 2012-03-05 available from: http://www.aoml.noaa.gov/divisionobjective/ tropical_meteorology.pdf. http://www.imd.gov.in/section/nhac/ dynamic/faq/faqp.htm bid (banco interamericano de desarrollo), 2008. vulnerabilidad y adaptación al cambio climático. diagnóstico inicial, avances, vacíos y potenciales líneas de acción en mesoamérica. 2008. discharge on: 2012-04-15. available from: http://publications. iadb.org/handle/11319/1761?locale-attribute=en center intelligence agency (cia-the world-factbook), 2012. the dominican republic-country profile. discharge on: 2012-10-28. available from: https://www.cia.gov/library/publications/theworld-factbook/geos/dr.html cepal (comisión económica para américa latina), 2010. gráficos vitales del cambio climático para américa latina y el caribe. . discharge on: 2012-10-25. available from: http://www.grida.no/ files/publications/lac_web_esp_2011-01-04.pdf cepal (comisión económica para américa latina), 2010. analysis of extreme events in the caribbean 1990 – 2008. consulted: 201210-25. available from: http://www.eclac.org/ chen k.mcaneney j. and cheung k., 2009. quantifying changes of wind speed distributions in the historical record of atlantic tropical cyclones. nat. hazardsearthsyst. sci., 9, 1749–1757, 2009. cocco q. a., 2010. huracanes en la república dominicana. 2009. consulted: 2012-10-28. available from: http://www.acqweather. com/huracanes.htm#temporada. costal services center (csc-noaa), 2013. the interactive hurricane tracks database web page. consulted: 2012-10-25. available from: http://www.csc.noaa.gov/hurricanes/#. eo-nnasa (earth observatory), 2007. hurricane season 2007: noel (atlantic). tropical storm noel floods the dominican republic. consulted: 2012-09-18. available from: http://www.nasa.gov/ mission_pages/hurricanes/archives/2007/h2007_noel.htmlhttp:// earthobservatory.nasa.gov/naturalhazards/view.php?id=19279 em-dat: the ofda/cred international disaster database, 2012. universitécatholique de louvain, brussels, belgium. top 10 natural disasters in dominican rep. for the period 1900 to 2012. country profile. discharge on: 2012-09-18. available from: http:// www.emdat.be/result-country-profile. emanuel, k., 2003: tropical cyclones. ann rev. earth planet. sci., 31, 75-104. fao (food and agriculture organization), 2000. dominican republic country profile. consulted: 2012-10-28. available from: http:// www.fao.org/countryprofiles/index/en/?iso3=dom. steve graham steve and holliriebeek, 2006. hurricanes: the greatest storms on earth: hurricane climatology. 2006. consulted: 201210-28. available from: http://earthobservatory.nasa.gov/features/ hurricanes/hurricanes_3.php. gutiérrez m. e, & espinosa, t., 2010. vulnerabilidad y adaptación al cambio climático. diagnóstico inicial, avances, vacíos y potenciales líneas de acción en mesoamérica. notastécn. nº idb-tn-144. bid, 2010. consulted: 2012-04-15. available from: http://idbdocs. iadb.org/wsdocs/getdocument.aspx?docnum=35404402. hennon c. christopher, 2008. tropical meteorology. a first course. dept. of atmospheric sciences. unc asheville, usa. 2th edition, 2008. consulted: 2013-09-08 available from: http://www.osti. gov/eprints/topicpages/documents/record/006/2675529.html ipcc (intergubernament panel for climate change), 2012. gestión de los riesgos de fenómenos hidrometeorológicos extremos y desastres para mejorar la adaptación al cambio climático. srexspm spanish google docs. discharge on: 201211-02. available from: https://docs.google.com/file/ d/0b1gfp6ioo3aktvjyv0drynm4c0u/edit?pli=1. ipcc (intergubernament panel for climate change), 2007. impacts, adaptation and vulnerability. contribution of working group ii to the fourth assessment report of the intergovernmental panel on climate change. discharge on: 2012-11-02. available from: http:// www.ipcc.ch/pdf/assessmentreport/ar4/syr/ar4_syr_sp.pdf. jury, m.r., rios-berrios, r., and garcia, e., 2011.caribbean hurricanes: changes of intensity and track prediction. theory of applied climatology, 107, 297-311 karmalkar a. v., taylor m. a., campbell j., stephenson t., new m., centella a., benzanilla a., charlery j., (2013). a review of observed and projected changes in climate for the islands in the caribbean. atmósfera26 (2), 283-309. 2013 klotzbach, p.j. 2011.the influence of el niño–southern oscillation and the atlantic multidecadal oscillation on caribbean tropical cyclone activity. j. climate, 24, 721–731. landsea c., anderson c., charles n., clark g., dunion j., fernándezpartagas j., hungerford p., neumann c., zimmer m., 2004. the atlantic hurricane database re-analysis project: documentation for the 1851-1910 alterations and additions to the hurdat 166 santo trinidad alvarez ysabel, gustavo adolfo agredo cardona, david felipe rincon database. hurricanes and typhoons: past, present and future_, r. j. murname and k.-b. liu, eds., columbia university press, 177221. discharge on: 2012-03-05. available from: http://www.nhc. noaa.gov/pdf/tc_book_atl_1851-2006_lowres.pdf. landsea, c. w., feuer s., hagen a., glenn d. a., sims j., pérez r., chenoweth m. and anderson n., 2007. a reanalysis of the 1921–30. atlantic hurricane database. discharge on: 2012-0305. available from: http://www.nhc.noaa.gov/pdf/landsea-et-aljclimate2012.pdf onamet (oficina nacional de meteorología), 2009. desarrollo de un sistema de alerta temprana de fenómenos hidrometeorológicos para la república dominicana. discharge on: 2012-03-05. available from: http://www.cimhet.org/resumen_ejecutivo_cliber_ repdominicana_sep09.pdf pielke r. a. jr., landsea c., mayfield m., laver j. and pasch r., 2005. hurricanes and global warming. bull. amer. meteor. soc., 86: 1571–1575. philip klotzbach and william gray, 2009. . central america/caribbean landfall probability calculations. colorado state university (csu). discharged on: 2013-08-20. availablefrom:http://tropical. colostate.edu/ pnud (programa de las naciones unidas para el desarrollo), 2008. pnud en república dominicana. diagnóstico de la situación de la república dominicana en cuanto a la reducción de riesgo desastre. 2008. discharged on: 2013-08-20. available from: http://www.pnud.org.do/publicaciones/crisis. mcsweeney c., m. new and g. lizcano, 2012. dominican republic. undp-climate change country profiles. discharge on: 2013-0820. available from: http://country-profiles.geog.ox.ac.uk ministerio de ambiente y recursos naturales (ministro of envaronen), 2010. informe geo de la república dominicana. estado y perspectivas del medio ambiente. santo domingo, r. d., 2010. nacional huracane center (nhc-noaa), 2012. tropical cycloneclimatology. discharge on: 2012-01-11. available from: http://www.nhc.noaa.gov/climo/#overview; http://www.nhc.noaa. gov/data/ shieh, o.h. and colucci, s.j. 2010. local minimum of tropical cyclogenesis in the eastern caribbean. bull. amer. meteor. soc., 91, 185–196. steve graham steve and holliriebeek, 2006. hurricanes: the greatest storms on earth: hurricane climatology. 2006. consulted: 201210-28. available from: http://earthobservatory.nasa.gov/features/ hurricanes/hurricanes_3.php. torrence, c. and g. p. compo, 1998. a practical guide to wavelet analysis. bull. amer. meteor. soc., 79, 61-78. tropical rainfall measurement mission (trmm), 2012. rainfall totals over the period 26-30 october in the island of hispaniola. 2630/10/2007 (also others by dates). discharge on: 2012-01-11. available from: http://trmm.gsfc.nasa.gov/publications_dir/ extreme_events_2007.html united nations frameworks convention on climate (unfccc), 2007 and 2008. climate change, small island developing states. isbn 92–9219–012–1. , 2005. united nations frameworks convention on climate (unfccc), 2010. manual sobre evaluaciones de vulnerabilidad y adaptación, 2007. regional miaac/conferences/die 3 (12-agosto-2010)/presentation wetlands international/unfccc v-a. consulted: 2012-01-11. available from: http://www.pnuma.org/agua-miaac/curso united nation environment programmed (unep grid-geneva), 2010. global risk platform. country profile: dominican republic. 2010. consulted: 2012-01-11. available from: http unep (united nation environment programmed), 2008. climate change in the caribbean and the challenge of adaptation. unep-regional office for latin america and the caribbean, panama city, panama. printed in panama city, in october 2008 unu-wider. natural disasters and remittances. department of economics, university of the west indies, st. augustine. research paper no. 2008/61. isbn 978-92-9230-111-8. unu-wider. 2008. unu-wider, report paper 2008/61. natural disasters and remittances. exploring the linkages between poverty, gender, and disaster vulnerability in caribbean sids. helsinki, finland, 2008. isbn 978-92-9230-111-8 appendix table ii. list of tc categories within 160 nautical miles of the dr extracted fromhurdat-2-nhc database hurdat program: data collection and compilation http://www.aoml.noaa.gov/hrd/hurdat/hurdat2_1851_2012-jun2013.html tropical cyclones affecting the dominican territories for the mayor period of 1970 to 2012 167quantifying the statistical distribution of tropical cyclone activity in the historical record of the dominican republic satellite data era appendix i. the central america-caribbean landfall probability calculations methodology a methodology to display the probabilities of a tropical cyclone passing within a radius of approximately 100 and 185 kmfromvarious islands and landmasses in central america and the caribbean was developedat colorado state university (csu) and other institutions by the respective main and secondary authors.it allows the probability of a tcs passing within a specified distance of anarea in any particular year to be calculated. this method has been modified and adapted for the specificstudy area in this paper for distance (kilometers), the probability of landfall (t= 1, 3, 5, 10, 15, 20, 25 and 50years) occurrence and for an extended major period of time series for the years that the tcwere tracked(1851 to 2012). the data were taken from nhc-noaa webpage and the hurdat2 database, exported into the microsoft excel format and recalculated. the methodology facilitates the awareness of the chances of the closest tropical cyclone occurrence that passesor impactsin a particular point inthis area. figure 4a displays the tracks of all storms named thatpassed within anapproximate300-km radius,andappendix idisplays the maximum intensity of each storm by distance for the drfrom 1851 to 2012 but usesthe same criteria for the other distancesconsidered in this paper (~185 km). to approximate the future likelihood of storms, a poisson regression model was used becausethe analysis of the number of tropical cyclones that made landfall over the last century shows that the landfall frequency closely conformedto a poisson distribution, the formula for which is ep = (e-p) (px) / x! (1) where ep = expected probability p = annual average number of tcsoccurred in the past 161 years x = number of storms expected in the upcoming year based on the poisson formula x! = factorial e = 2.71828 for the probabilitycalculation,the fifty-year (50) probabilities of a major hurricane(mh) tracking within each km of distance fromthe country territories (inland or open-sea mass) wasprovided because most buildings were erected to last at least 50 years and because the construction criteria forthe cost of hurricane-protecting building materials should be based on alonger period. nevertheless, a descending and minor probability wasappliedtoobtain abetter viewpoint regarding this matter. for further information, see the citations or visit the website http://www.e-transit.org/ hurricane/welcome.html keywords: nested decomposition; gpu parallel; linear solution; cpr. palabras clave: descomposición anidada; gpu paralela; solución lineal; cpr. how to cite item how to cite item: shi, x. & di, y. (2019). gpu parallelization nested decomposition method for solving large linear systems in reservoir numerical simulation. earth sciences research journal, 23(3), 249-257. doi: https://doi.org/10.15446/esrj.v23n3.81669 this paper designs a highly parallel nested factorization (nf) to solve large linear equations generated in reservoir numerical simulation problems. the nf method is a traditional linear solution preprocessing method for reservoir numerical simulation problems and has regained attention in recent years due to its potential to extend to parallel architectures such as gpus (graphics processor units). the parallel algorithm of this paper is based on the mpnf (massively parallel nested factorization) framework proposed by appleya. the mpnf algorithm designed in this paper focuses on its efficient implementation on the gpu parallel architecture. its features include: using a custom matrix structure to achieve merge access, improving access bottlenecks, and improving the efficiency of the spmv algorithm. it is also applicable to the two-stage preprocessing method cpr (constrain pressure residual). cpr pressures solution and global preprocessing stage; the mpnf method is extended to the solution of the 2.5-dimensional unstructured grid problem. the parallel algorithm in this paper has been integrated into the reservoir numerical simulator. for the spe10 (million grid, highly heterogeneous) standard example, the gpu-based parallel nf algorithm is in the structured grid model and the equivalent 2.5-dimensional non-on the structured grid model. compared with the serial version of the nf method, the acceleration ratios of 19.8 and 17.0 times were obtained, respectively; compared with the mainstream serial solution method; the efficiency was also improved by 2 to 3 times. abstract gpu parallelization nested decomposition method for solving large linear systems in reservoir numerical simulation método de descomposición anidada a través de paralelización con unidades de procesamiento gráfico para resolver sistemas lineales grandes en la simulación numérica de yacimientos issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n3.81669 este artículo diseña una factorización anidada (nf) altamente paralela para resolver grandes ecuaciones lineales generadas en problemas de simulación numérica de yacimientos. el método nf es un método tradicional de preprocesamiento de solución lineal para problemas de simulación numérica de yacimientos, y ha recuperado atención en los últimos años debido a su potencial para extenderse a arquitecturas paralelas como las gpu (unidades de procesador de gráficos). el algoritmo paralelo de este artículo se basa en el marco de mpnf (factorización anidada masivamente paralela) propuesto por appleya. el algoritmo mpnf diseñado en esta contribución se enfoca en su implementación eficiente en la arquitectura paralela de gpu. sus características incluyen: usar una estructura de matriz personalizada para lograr un acceso combinado, mejorar los accesos de cuellos de botella y mejorar la eficiencia del algoritmo spmv. también es aplicable al método de preprocesamiento en dos etapas cpr (solución de presión residual de restricción) presión solución y etapa de preprocesamiento global; el método mpnf se extiende a la solución del problema de cuadrícula no estructurada de 2.5 dimensiones. el algoritmo paralelo en este artículo se ha integrado en el simulador numérico de yacimiento. para el ejemplo estándar de spe10 (millones de cuadrículas, altamente heterogéneo), el algoritmo nf paralelo basado en gpu está en el modelo de cuadrícula estructurada y el equivalente de 2.5-no dimensionalen el modelo de cuadrícula estructurada, en comparación con la versión en serie del método nf, los índices de aceleración de 19.8 y 17.0 veces se obtuvieron respectivamente; en comparación con el método de solución serial convencional, la eficiencia también se mejoró de 2 a 3 veces. resumen record manuscript received: 20/03/2019 accepted for publication: 15/07/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 3 (september, 2019): 249-257 xin shi1, yuan di1* 1college of engineering, peking university, beijing, 100871, china c o m pu te r s c ie n c e a pp li ed t o g eo lo g y 250 xin shi, yuan di gpu parallel architecture gpu multi-core parallel architecture has received more and more attention in the field of high-performance computing in recent years. gpu was originally used for graphics processing acceleration tasks. with the development of general-purpose gpu (general-purpose computing on gpu, gpgpu-related) software and hardware platforms, gpus are gradually applied to parallel acceleration in the field of scientific computing. one of them is nvidia. launched the compute unified device architecture (cuda) platform (appleyard, appleyard, wakefield, & desitter, 2011; nvidia, 2009). the algorithm in this paper is based on cuda implementation. the gpu achieves higher computing performance by integrating more computing units on the same chip area. take nvidia’s kepler architecture gk110 chip as an example. the most basic computing unit is the cuda core. a complete gk110 chip contains 2,880 cuda cores. the k10c gpu with gk110 chip has 5gb gddr5 (graphics double data rate, version 5) memory, and its single-precision and double-precision floating-point operation peak speeds are 3.52 and 1.17 tflops respectively (float operation per second, floating-point operations per second) ). the peak bandwidth has reached 249.6 gb/s. these metrics far exceed the peak data of the same level of cpu. gpu parallelism is often memory-constrained, so it must be welldesigned to accommodate the gpu’s multi-level memory structure to maximize gpu computing performance (nvidia, 2012a; nvidia, 2012b). linear system solution in reservoir numerical simulation control equation in reservoir numerical simulation, the basic governing equation is the mass mass conservation equation. for any discrete grid system, the mass conservation equation for a certain mesh i and component c can be written as: ∆ ∆ φ ρ ρ − − −γ − t v s x x q t i j p p j p p i p p g d j p p cp p cp w p p ∑ ∑( )  + ( ) , dd i pj neighbor                   =∑∑ 0 ∈ (1) which represents volume, represents porosity, represents phase saturation, represents phase density, represents component mass fraction, represents source term (well term), represents all meshes connected to i, represents a phase, between a connected mesh conductivity, which represents pressure, represents the mass density of the phase, represents the gravitational acceleration constant, and represents the mesh depth. additional constraints are required to solve the above system. for the most common case, in the constraint equation v represents volume,  indicates porosity, indicates phase saturation sp p ∑ = 1 (2) xcp c ∑ = 1, p (3) f p x t f p x tcp p cp cq p cq, , , ,( ) = ( ) (4) equation (4) represents the fugacity balance of the components between phases and . the above systems are usually solved using the newton iteration method. at each newton iteration step, the corresponding jacobian matrix and the rightend term are obtained, and the linear system is solved, and the newtonian update is performed until the nonlinear equation converges, and then the simulation of the next time step is performed. in the above process, the linear system solution takes up the most time. in the reservoir numerical simulation, the krylov subspace-based iterative solver is generally used to solve such large sparse linear systems. generalized minimum residual (saad, 2003; saad & schultz, 1986) (generalized minimum residual, gmres) is one of the most commonly used solvers. however, the performance of such solvers depends to a large extent on the pre-processing methods used with them. cpr two-stage pretreatment method the equation system of the reservoir problem has mixed characteristics, which are close to the part of the elliptic equation (pressure equation) and the part close to the hyperbolic equation (convection equation). in order to solve this mixed equation problem, wallis et al. (wallis, 1983; wallis, kendall & little, 1985) proposed a two-step pretreatment method, constrained pressure residual (cpr). in this method, the first step extracts the pressure matrix from the complete jacobi matrix, solves the pressure matrix to eliminate the low frequency error existing in the pressure variable, and the second step preprocesses the complete matrix to eliminate the high frequency error (or local error). ). the mathematical characterization and algorithm steps of the cpr method are as follows. the preprocessing matrix defining the cpr method is m1,2 m m i am m1,2 1 1,2 1 1 -1 1 -1=     + (5) where m1 and m2 are the preprocessing matrix for the first and second steps, respectively. it is the original matrix, which is the unit matrix. call cpr method to get x m v= 1,2 -1 . it can be broken down into two separate pre-processing processes to calculate: 1) call the first step of the preprocessing algorithm, calculate x m v= 1 -1 ; 2) calculate v v ax2 1=  ; 3) call the second step preprocessing method, calculate x m v2 2 1 2= − ; 4) calculate x x x= +1 2; reasonable selection of two separate preprocessing methods as cpr components, the cpr method proved to be very efficient in dealing with reservoir simulation problems, on a series of complex models including standard wells, multi-segment wells and unstructured grid models. can have very good performance (jiang, 2007; jiang & tchelepi, 2009). the pressure solution process in cpr generally uses the multigrid method (stüben, 1983), and the global preprocessing stage generally uses the ilu0 method. this paper retains the cpr solution framework and uses the mpnf method as a two-stage preprocessing method. nested decomposition traditional nf method the nf method (stüben, 1983) is a classical linear solution preprocessing algorithm for reservoir numerical simulation problems for structured grids. its basic idea is to use the nested three-diagonal matrix structure corresponding to the structured mesh to solve the problem quickly. for structured grid systems that are small (within hundreds of thousands of grids), the nf method has high efficiency. mpnf method nf is a completely serial algorithm, both in the decomposition phase and in the solution phase. in order to make it parallel on the gpu, the algorithm needs to have a large number of parallelizable units. mpnf is such a variant. mpnf achieves good fine-grained parallelism through dyeing and rearrangement. for example, a set of meshes extending in the z direction (other directions are also possible) is called a kernel; then coloring is performed at the core level (that is, the xy plane), and the dyeing operation refers to each unit is assigned a color, so that adjacent units always belong to different colors; finally, the grid system is rearranged by number, first in color order, and the same color is sorted by core (the order between cores can be arbitrary) ), the core internals are sorted from small to large in the z direction. 251gpu parallelization nested decomposition method for solving large linear systems in reservoir numerical simulation pre-processed gmres for the cpr first-stage solution, and introduced the cuda-based gpu algorithm implementation in detail, and this method applied to multi-gpu parallelism. this paper implements a complete gpu parallel linear algorithm based on cuda. the algorithm uses cpr as the preprocessor and gmres as the accelerator solution framework. the preprocessor selection of cpr two stages is mpnf method. consolidated access the global memory is the memory outside the gpu chip. it has the largest storage space and the largest access latency. when the algorithm works, it will store all the data that needs to be read and written from the host thus, the inner mesh of the core is a one-dimensional connection, and the numbers are adjacent, resulting in a three-diagonal structure. kernel d u l d u l u l d i i i i i i i nc i nc i nc i = − − 1 1 1 2 2 2 1 1 0 0 0 0 0 0 l o m o o m o o o l                     (6) where nc represents the number of grids in each core, representing the serial number of the core. it can be a single element or a dense matrix block, depending on the current matrix type (pressure matrix or main matrix) and the number of implicit variables in each grid block. the cores in the same color are not connected to each other, forming a block diagonal structure. color kernel kernel kernel i i i nk i =     1 2 0 0 0 0 0 0 0 0 0 0 0 l o o m o o m o o o l               (7) where nk represents the number of cores in a color (the number of cores in different colors is generally different), representing the serial number of the color. think of a grid within a color as a whole, as long as the dyeing algorithm ensures that the different colors are also one-dimensional connections, then the colors also form a block-like three-diagonal structure. a color u l color u l u l color = − − 1 1 1 2 2 2 1 1 0 0 0 0 0 0 l o m o o m o o o l nco nco nco ii                   (8) where nco representative dye number, l,u represents the non-diagonal elements produced by the connection between colors. compared with nf mehtod, mpnf through dyeing and rearrangement, a matrix of two-level three-diagonal nested structure is formed, and a similar factorization can be performed corresponding to the pre-processing matrix. since at the color level, the cores inside the same color are always not adjacent, the parallelism is obtained between the cores. in the decomposition and solution phase of the preprocessing, each core can be executed in parallel when it involves operations within each color. the simplest method of dyeing is the checkerboard dyeing of two colors, alternating the two colors to complete the staining on the plane. as shown in figure 1. in the figure, each core contains 4 grids, which are arranged in the longitudinal direction. after dyeing, the two colors each contain 10 cores, and the cores in the same color are not connected to each other. fewer staining results in higher parallelism, but at the same time the approximation of the preprocessing matrix is degraded, and convergence requires more iterations. appropriate increase in the number of stains may contribute to the parallel effect as a whole. a four-color oscillatory format staining method is shown in figure 2. the dyeing starts from one corner and is sequentially dyed according to the strip parallel to the diagonal, and the color is circulated in the order of 1→2→3→4→3→2→1. obviously, this dyeing method can still ensure that the cores in the same color are not connected, and the color is a one-dimensional connection. figure 3 is the corresponding matrix structure. efficient implementation and expansion of mpnf methods appleyard (2011) describes the principle of the mpnf method, but there is no specific implementation details; zhou (2012) used the mpnf figure 1. 5×4×4 regular sorting of the grid (left) and checkerboard sorting (right) figure 2. four-color oscillating format dye sorting 0 10 20 30 40 50 60 70 80 0 10 20 30 40 50 60 70 80 nz = 448 figure 3. matrix structure diagram corresponding to four-color oscillating format dyeing sorting 252 xin shi, yuan di rows is the same as the number of rows of the original matrix, and the number of columns is the same as the maximum number of non-zero elements in each row of the original matrix, and the position where there is a gap after compression by the row is filled with 0 elements. val is stored in column priority order. the col_idx storage format is the same as val, recording the original column index of each corresponding position element. when using cuda for spmv operations, the operation of each row in the ellpack format is handled by one thread, and the padded 0 element ensures that access to the matrix elements is a merged access. this mode allows it to perform better than the csr format when the number of non-zero elements in each row is not much different. in the reservoir numerical simulation problem, the maximum number of non-zero elements per row of the matrix depends on the number of connections of the grid (a grid is connected to at most 6 grids), which is relatively fixed, and after the grid number is determined, the maximum the number of non-zero elements can be calculated, and it is suitable to use ellpack format storage. combined with the characteristics of the matrix, this format can be further improved. one of the characteristics of such a matrix is that the first layer of the block structure is constructed in a core organization. in the matrix shown in fig. 3, the upper non-diagonal element representing the connection between color 1 and color 2 is taken as an example (see fig. 5), and the red box marked portion represents the connection relationship between the cores of the two colors, and constitutes a connection relationship. a layered structure, in which each small matrix block is a diagonal matrix, which is hereinafter referred to as a core block. another feature is that each diagonal element in the core block (ie, the black patch in the figure) is also a block structure. with these two layers of block structure, a more compact b-ellpack format can be constructed for spmv operations. figure 5. two-layer block structure diagram figure 6 is a schematic diagram of the storage of figure 5 as a b-ellpack. in this format, only the column subscript of the block block of the core block is stored in col_idx, and the column subscript of each non-zero element is not required to be recorded, and the required storage space and the access frequency thereof are greatly reduced, thereby improving the correlation. the efficiency of the operation. when performing spmv operation, each thread processes a block operation of a core block. the storage order of specific elements in the val array should be: firstly stored in the order of the block block of the core block; sorted by layer in the block column of the same core block (assume that the core is distributed along the z direction); the same layer is arranged in the order of the columns of the small matrix blocks; finally, the core blocks are sorted by the block rows. the matrix representing the connection between colors can be stored in the b-ellpack format. the spmv operation needs to be performed when solving the block-shaped three-diagonal matrix composed of colors, and the obtained result vector is used for the right-end term of the inner three-diagonal solution of the next color. therefore, in addition to the higher efficiency, this format can also make the order of the result vectors obtained by the matrix after the spmv operation coincide with the order of the right end items required for the inner tridiagonal solution of the color (ie, the order). for p-gmres and b-gmres, the spmv operation of the pressure matrix and the main matrix is required in the algorithm. a simple modification to the above format is applicable. this variant is hereinafter referred to as the b-diag-ellpack format. (host) memory. copy to the device (device, also known as gpu), stored in the global memory for thread access. access to global memory needs to satisfy coalesced access: if an instruction needs to access specific data in global memory, and the access locations of all threads (within the same thread bundle) are consecutive in memory, then the merge access is satisfied. . if the merge access condition is satisfied, only one read operation is required, and all threads in the warp can obtain their respective data; otherwise, each thread must perform a separate read operation to obtain the respective data, which greatly reduces the effective bandwidth. for the mpnf algorithm, the initial matrix data of the host does not meet the merge access requirements, and needs to be rearranged before being passed to the device for parallel algorithms. the data that the mpnf algorithm needs to use includes: matrix data, unknown vectors, and right-end vectors. the structure is similar and is illustrated by an example. after the mesh is dyed, the initial structure of the diagonal elements is first sorted by color number (that is, the elements whose subordinate color numbers are small are always in front of the elements with large numbers), the same color is internally sorted according to the core number, and the same core is internally meshed according to the grid. the number is sorted. finally, if the middle element is a matrix block (for b-mpnf), the matrix blocks are arranged in column priority order. since the parallelism of the mpnf algorithm exists between the cores in the same color, the thread responsible for processing different cores should access the contiguous memory under the same operation to satisfy the merge access, so it should be rearranged in the following order: firstly, the order is the same according to the color number. inside the same color, first arranged in layer order (assuming the core is in the z direction); if the middle element is a matrix block in the same layer, it is arranged in the order of the columns in the matrix block; the positions in the same matrix block are in the order of the core number arrangement. if the above arrangement order in the same color is denoted as in_ ker_order, in_blk_order, ker_order, the arrangement before and after rearrangement is changed from “ker_order” “in_ker_ order” “in_blk_order” to “in_ker_order” “in_blk_order” “ker_order”. in short, parallelism exists between cores, and only needs to be sorted by core number in the end, and the other order is unchanged to ensure merge access. in addition to using the same arrangement, the incoming right-end vectors should be rearranged in the same way, as is the calculated solution vectors. the intermediate variables calculated in this way also have the same order. block-based b-ellpack format the ellpack format is a sparse matrix storage format, and its storage mode is similar to the csr format. the difference is that after the matrix is compressed by row, the ellpack format fills 0 elements in rows with fewer nonzero elements so that all rows are compressed to the same length. see figure 4 for an example of its storage format. where val is a dense matrix, the number of figure 4. ellpack storage format example. 0 represents the filled 0 element, * represents the subscript corresponding to the filled 0 element, can be any value, will not be accessed when used 253gpu parallelization nested decomposition method for solving large linear systems in reservoir numerical simulation take the main matrix as an example. the main matrix differs from that of figure 5 in that the block diagonal elements are internally trigonal rather than diagonal. the diagonal block always exists, so only the diagonal block needs to be stored in the top position of each block row, and col_idx does not need to store the column index of the diagonal block, and the correlation operation is performed. the block column can be specially treated. the order of the special block columns can be various, and it is only necessary to ensure that the order of the last two steps is arranged in the order of in_blk_order ker_order. fig. 7 is a behavior example corresponding to the first color in fig. 3, showing the corresponding b-diag-ellpack storage format. correspondingly, the spmv results of the main matrix are also stored in the same order. when performing other vector parallel operations, they must first be converted back to the normal order. mpnf method for 2.5d unstructured grid the simulation of fractured reservoirs uses the discrete fracture model (dfm). the dfm mesh model can be a completely unstructured tetrahedral model or a semi-structured layered model (ie 2.5d). a layered model is a grid system that is still divided into layers in the longitudinal direction and is formed by triangulation (or other splitting format) on the plane. compared with the tetrahedral model, the layered model has limitations in applicability (requires cracks to be high-angle seams), but it has significant advantages in terms of sectioning difficulty, grid quality, formation information, and speed of solution. an important unstructured grid model. kuznetsova et al. (2007) introduced a method of applying the nf method to a layered model: dyeing on a plane to construct a two-layer one-dimensional joint structure (the grid in the same color is a one-dimensional connection, and the color is one between dimensional links, plus one-dimensional connections in the vertical direction, finally construct a three-dimensional one-dimensional structure in the unstructured grid. similarly, the mpnf method itself does not require the distribution of the mesh on the plane: as long as one direction can be selected as the direction of the core extension (requires the same number of core meshes, and is a onedimensional connection), and in the direction of the core extension a dyeing method can be given on a vertical plane to ensure a one-dimensional connection between colors, and the mpnf method can be applied. appleyard (2016) presented a method for staining unstructured meshes on a plane. first select the color number and the stimuli format of the dye sequence; select the starting grid, assign it the starting color; find the current unstained, but at least one of the grids connected to the dyed grid to form a candidate list; statistical candidate list the number of different colors connected to the grid is arranged in descending order of different color numbers; the next color in the color sequence is taken, and the grids satisfying the requirements (the same color cannot be connected) are sequentially given in the order of the candidate list until the traversal is completed. finally, look for a new candidate list and repeat the above process until all the grids have been dyed. this is a general dyeing method that works equally well for structured grids. when applied to a structured grid, the number of colors is at least 2; however, when applied to an unstructured grid, the number of colors required is generally large (eg, 6). figure 8 is a schematic diagram showing the results of staining on a plane using a dyeing method for the above dyeing method. the unstructured mesh is generated by triangle (shewchuk, 2002) (using the -q command to control the triangle element as close to the regular triangle as possible). it can be seen that the dye spreads outward from the layer at the initial grid, and the shock stimuli sequence is still well maintained. figure 9 is a layered model (divided into 41 triangular grids on the plane, divided into 3 layers in the longitudinal direction) and rearranged by 6 colors according to the above method, corresponding to the matrix structure diagram. the part marked by the red frame is a three-diagonal structure composed of connections between colors. it can be seen that for non-structural meshes, this method of dyeing does not completely guarantee a one-dimensional connection between colors, so there will be a small amount of elements that fall into the tri-diagonal off-band. when using mpnf preprocessing, these off-band elements can be ignored. although there is some loss in accuracy, a strict three-diagonal structure is guaranteed, so that the algorithm can be performed normally. but whether it is the gmres algorithm in the pressure solution process or the gmres algorithm used in conjunction with cpr, the spmv part still uses the exact matrix. figure 6. b-ellpack format diagram figure 7. b-diag-ellpack format diagram figure 8. 2.5d grid 6-color turbulence format staining result (staining start point at the center) 254 xin shi, yuan di study case test algorithm convergence performance and time-consuming analysis this section selects the first 10 layers of the spe10 (van der vorst, 1992) example, and performs performance test and time analysis on the cpr twostage mpnf algorithm. the grid size is 60×220×10, a total of 132,000 grids, and the fluid is oil-water two-phase. the mpnf method in the pressure phase is combined with gmres, that is, the one pressure matrix solution may need to call multiple mpnf solving processes, and the convergence error is set to 0.1; the second phase mpnf is used alone, that is, the second phase mpnf solving process is performed only once per cpr call. . the first is the test of convergence speed. ilu0 and amg were selected for comparison with p-mpnf, and bilu0 was selected for comparison with b-mpnf. the mpnf method also selects the number of stains as 3, 4, and 5 for comparison. if the mpnf method is not used in both stages of cpr, the grid number is in natural order. if mpnf is used in any stage, the grid number is the order after dyeing. when ilu0 is used in the pressure solution phase, it is also used in combination with gmres. the pressure solution convergence error is 0.1, and amg is used alone. the second phase of bilu0 is also used separately. the test platform hardware is configured for one xeon x5670 cpu and one k20c gpu. table 1 shows the test results. first, the number of newton iterations is almost the same in all the examples. this is because the selection of the specific components of the linear solution does not affect the newton convergence speed under the same linear solution strategy and convergence error. the number of linear iterations is the number of iterations of the outer gmres. cpr preprocessing is called once per iteration. the size reflects the validity of the cpr combination. the cpr combination of the example 1 achieves the best effect, which is consistent with the usual understanding. amg is best suited as the choice for the stress solution phase in cpr. in the comparison between the case 2 and the case 3, the average number of iterations required for each pressure solution is used as a standard, and the ilu0 is used for the pressure solution to be slightly weaker than the mpnf. considering that the calculation example 3 is based on the dyeing rearrangement, the rearrangement will make the quality of the matrix worse in the sense of solution, and thus the number of linear iterations is more than that of the second example. comparing example 3 with example 4, the number of linear iterations is very close, reflecting that mpnf as a second-stage preprocessing is similar to bilu0. a comparison of the examples 3, 4, and 5 shows that the increase in the number of stains is beneficial to the matrix solution, whether it is a pressure matrix or an outer iteration. but the difference between the three in this test case is not obvious enough. first, the mpnf decomposition phase time consumption is negligible compared to the solution phase. this is because for p-mpnf and b-mpnf, the decomposition phase is only executed once for each newton iteration; and the solution phase for b-mpnf is every time. the cpr call is executed once. for p-mpnf, it is executed once per pressure iteration in each cpr call; the number of times of pressure solution iteration is very large, and the number of executions in the decomposition phase and the solution phase are very different (for p-mpnf and bmpnf is 1:174 and 1:6) respectively. blas operation has good parallelism, but the computational density is very low (the ratio of the number of memory reads to the number of calculations), which is limited by the gpu bandwidth cannot achieve high performance; the reduction operation decreases rapidly with the parallelism of the algorithm. also can’t have a good performance. the number of these two operations is affected by the number of gmres iterations. as the number of iterations increases, the number of blas and reduction operations performed per iteration increases linearly. therefore, the two occupy a large proportion in the algorithm time composition. in order to reduce the general blas and reduction operations for parallel performance, the bicgstab (stabilized double conjugate gradient method) solver can be used instead of gmres, which is used in conjunction with the pressure matrix solution. bicgstab is also a subspace iterative method for general matrices. experience has shown that gmres is more stable because it guarantees that the norm of the residual is monotonically decreasing after each iteration. however, bicgstab is different from gmres in that the number of blas and reduction operations does not change with the number of iterations in each iteration. for the problem that the average number of iterations is close to 30, the ratio of these two operations is greatly reduced. on the other hand, the stress solving phase in cpr does not require high accuracy. convergence criteria for relaxation stress solving can be considered to reduce the number of iterations of excessive pressure system solutions. 0 20 40 60 80 100 120 0 20 40 60 80 100 120 nz = 611 figure 9. 2.5d grid 6 color oscillating format dyeing corresponding matrix structure table 1. comparison of iteration speeds of different cpr combinations (3c, 4c, 5c represent the number of dyes, respectively, 3, 4, 5) study number cpr combination newton iterations linear iterations pressure solution iteration number average number of iterations per iteration 1 amg+bilu0 147 667 667 1.0 2 ilu0+bilu0 145 786 25073 31.9 3 p-mpnf+bilu0(4c) 147 871 25770 29.6 4 p-mpnf+b-mpnf(4c) 147 869 25635 29.5 5 p-mpnf+b-mpnf(3c) 147 941 28412 30.2 6 p-mpnf+b-mpnf(5c) 147 840 24501 29.2 255gpu parallelization nested decomposition method for solving large linear systems in reservoir numerical simulation the pressure solution was changed to bicgstab and the pressure convergence criterion was relaxed to 0.5 and tested again. the comparison results are shown in table 2. the number of newton iterations is constant, the number of linear solutions is increased by 40%, but the average number of iterations of the solution is reduced to 11 times, and the total time consumption is reduced by more than 40%. the increase of the number of linear solutions is caused by the lower precision pressure solution, which increases the time consumption of the outer gmres and b-mpnf, but also reduces the average pressure solution iteration number, and finally reduces the total pressure solution iterations. . in addition, it should be noted that the total time consumption of mpnf has not decreased, but has increased. on the one hand, bicgstab will call p-mpnf solution twice in one iteration, so that the actual number of p-mpnf calls in two tests is close; the number of b-mpnf calls increases as the number of linear iterations increases, and the time spent on a single call is also higher than that of p-mpnf. thanks to the use of bicgstab, the proportion of time for reduction and blas operations decreased from 59% to 21%. parallel performance the general proportion of calculated components is reduced, which is the reason for the significant reduction in total time consumption. under the new setting, the mpnf solution stage time consumption is 44.7% (mainly p-mpnf solution stage, more than 90%), which occupies the main part. at the same time, the proportion of the remaining serial part in the linear solution algorithm is also more significant (close to 20 %). although this ratio is very high (resulting in an acceleration ratio of up to 5), the main time composition of the serial part is derived from the pressure matrix decomposition process. for more complex problems, a pressure matrix decomposition will correspond to more linear iterations, corresponding strings. the proportion of the line portion will also decrease. spe10 study test this section uses the complete spe10 model for parallel algorithm testing. the complete spe10 model consists of 60 x 220 x 85 with a total of 1.1 million grids. the model hole permeation field height is heterogeneous and is a standard example designed to evaluate the performance of linear solvers. the fluid uses two phases of oil and water. the model consists of four production wells located at the four corners of the model and one injection well located in the center of the model. all wells are shot through all formations. the test platform is the same as before. this paper implements a serial version of the two-stage mpnf algorithm as a reference to measure the acceleration of parallel implementation. at the same time, the gmres method preprocessed by cpr (amg+bilu0) under the serial algorithm is also added to compare the advantages of the parallel algorithm compared with the current mainstream serial algorithm. the comparison is classified according to the total time taken by the linear solution, the time-consuming of the pressure matrix, the second-stage pre-processing and the outer gmres time-consuming. the parallel version of mpnf achieved an acceleration ratio of 19.8 over the entire linear solution phase, and an acceleration ratio of 23.9 and 27.0 was obtained in the pressure solution phase and the second phase pretreatment and the outer gmres, respectively. since the total speedup includes additional overhead and other parts that cannot be paralleled, it is lower than the other two parts; the second stage preprocessing and outer gmres are similar in composition to the stress solution stage (average gmres iterations are 12 times) ), but thanks to the use of the block matrix, the access limit is smaller, so the acceleration ratio is higher than the pressure solution phase. the serial version of mpnf is significantly slower than the serial version of the mainstream solution combination. among them, the velocity matrix solution speed is very different, which verifies that amg is much better than mpnf in the serial hardware. the pressure matrix solution occupies most of the time (more than 70%) of the linear solution, so although the time consumption of the second stage preprocessing is not much different (about 2 times, the number of cpr calls caused by inaccurate pressure solution increases) cause), but the total linear solution time is still much higher. by comparing the latter two, it can be seen that the shortcomings of the mpnf preprocessing method itself can be compensated for on parallel hardware due to better parallelism. the parallel version of p-mpnf is more than twice as fast as amg, and the serial version of b-mpnf is more than 10 times faster than bilu0. the total linear solution speed is nearly three times faster. this proves the practicability of this parallel solution algorithm, and it can also bring about a real speed increase compared with the optimal strategy of the serial version. unstructured equivalent spe10 study test this section performs a crack-free layered model mpnf method test. the unstructured mesh (see figure 10) model is generated based on the spe10 example, keeping the number of layers unchanged, and the horizontal direction is re-divided into 12,680 meshes, and the attribute field is mapped according to the closest distance of the center point of the mesh. the unstructured model has 1,077,800 grids. in parallel mpnf vs. serial mpnf acceleration ratio, the structural model has decreased. this is because this study uses 6-color shading, and the number of parallel tasks drops, failing to take full advantage of hardware performance. the average number of iterations for the pressure solution is 20, which is more than the structured model, reflecting the consequences of ignoring elements outside the strip. but the parallel version of mpnf still has an advantage over the amg+bilu0 combination. table 2 . comparison of iterative speeds between different pressure solver selections and convergence errors study setting newton iterations linear iterations pressure solution iteration number average number of iterations per iteration total time consumption / s mpnf total time consumption / s p-gmres convp<0.1 147 869 25635 29.5 48.7 11.2 bicgstab convp<0.5 147 1233 13551 11.0 27.5 12.4 figure 10. layered model generated by spe10 study and dyeing results 256 xin shi, yuan di 2.5d non-structural example test with cracks next is a layered model test with cracks. the model is divided into 10 layers vertically, with a total of 26,418 grids in the horizontal direction, including 417 crack grids. the fracture permeability was 1000 md and the matrix permeability was 1.0 md on average. there is a production well in the two corners of the injection well. the fluid uses two phases of oil and water. the model is shown in figure 11. the model uses 6 color stains. the convergence speed information of the mpnf method is shown in table 3. although the elements outside the tridiagonal strip are rounded off, the overall convergence performance of the cracked example is still relatively stable. the number of linear iterations is not much higher; the average pressure matrix solution iterations is slightly higher than the regular grid, but it is also within the acceptance range. the speedup results are similar to the previous few examples. it is worth noting that the advantage of mpnf compared to amg+bilu0 combination is more obvious in this example, mainly because the total number of linear iterations is not much different between the two examples. conclusion the linear solution part is the most time-consuming part of reservoir numerical simulation. this paper mainly discusses how to use the gpu as a multi-core parallel architecture to parallelize the linear solution algorithm. parallel algorithm in the cpr two-step preprocessing solution framework, the mpnf method developed by the classical nf method is selected as the two-step preprocessor selection. the algorithm proposes a data rearrangement method for the two-stage mpnf preprocessing algorithm to ensure the merge access; the block-based b-ellpack structure is used to accelerate the spmv operation in the global gmres algorithm; bicgstab is used as the pressure stage solver. the weaker pressure is used to solve the convergence criterion to reduce the proportion of the less conservative reduction operation in the algorithm. through these optimization methods, the parallel solution algorithm achieves a good parallel effect. compared with the serial mpnf method, the complete spe10 algorithm obtains a speed ratio of more than ten times. compared with the mainstream linear algorithm, it also obtains 2-3. the speedup ratio is doubled. in addition, this paper uses the general dyeing algorithm to extend the mpnf method applicable to cpr two-stage preprocessing to 2.5 unstructured grid model, which expands the practicability of the mpnf method. in the 2.5d unstructured grid, the dyeing process requires more dyeing numbers, and there are a small number of elements located outside the three diagonal strips. these elements are ignored in mpnf preprocessing, and the parallel effect may be inferior to the structured net. grid. the algorithm was tested on a custom example containing discrete cracks and an example derived from spe10 mapping (both 2.5d unstructured grid studies). the results were shown on a 2.5d unstructured grid. the mpnf algorithm still has obvious advantages over the serial algorithm. references appleyard, j., appleyard, j., wakefield, m., & desitter, a. (2011). accelerating reservoir simulators using gpu technology. spe reservoir simulation symposium. society of petroleum engineers. appleyard, j. (2016). method and apparatus for estimating the state of a system: u.s. patent 9, 396, 162. 2016-7-19. jiang. y. (2007). techniques for modeling complex reservoirs and advanced wells. stanford university. jiang, y. & tchelepi, h. (2009). scalable multistage linear solver for coupled systems of multisegment wells and unstructured reservoir models. spe 119175, proceedings of the 20th spe reservoir simulation symposium. the woodlands, tx. kuznetsova, n. n., diyankov, o. v., kotegov, s. s., koshelev, s. v., krasnogorov, i. v., pravilnikov, v. y., & maliassov, s. (2007). the family of nested factorizations. russian journal of numerical analysis and mathematical modelling, 22(4), 393-412. nvidia. (2012a). cuda c best pratice guide, version 5.0. nvidia. (2012b). cuda c programming guide, version 5.0. nvidia. (2009). whitepaper-nvidia’s next generation cuda compute architecture. fermi. saad, y., & schultz, m. h. (1986). gmres: a generalized minimal residual algorithm for solving nonsymmetric linear systems. siam journal on scientific and statistical computing, 7(3), 856-869. saad, y. (2003). iterative methods for sparse linear systems. society for industrial and applied mathematics. shewchuk, j. r. (2002). delaunay refinement algorithms for triangular mesh generation. computational geometry, 22(1-3): 21-74. stüben, k. (1983). algebraic multigrid (amg): experiences and comparisons. applied mathematics and computation, 13(3): 419-451. table 3. comparison of iterative speeds of layered models with cracks solution strategy newton iterations linear iterations average number of iterations per iteration amg+bilu0 573 9166 1 p-mpnf+b-mpnf 571 9953 14.8 a) matrix part b) crack section figure 11. layered model with crack 257gpu parallelization nested decomposition method for solving large linear systems in reservoir numerical simulation van der vorst, h. a. (1992). bi-cgstab: a fast and smoothly converging variant of bi-cg for the solution of nonsymmetric linear systems. siam journal on scientific and statistical computing, 13, 631-644. wallis, j. r., kendall, r. p., & little, t. e. (1985). constrained residual acceleration of conjugate residual methods. spe reservoir simulation symposium. society of petroleum engineers. wallis, j. r. (1983). incomplete gaussian elimination as a preconditioning for generalized conjugate gradient acceleration. spe reservoir simulation symposium. society of petroleum engineers. zhou, y. (2012). parallel general-purpose reservoir simulation with coupled reservoir models and multisegment wells. stanford university. nota editorial la constancia y la perseverancia no se aprenden, se logran con el trabajo y en geoffsica este nos abruma.en nuestras instituciones no hay a veces suficientes alicientes; sin embargo, el ejemplo que personas 0 grupos nacionales e internacionales nos brindan nos incentiva a seguir. ei trabajo geoffsico continua y no podemos rezagarnos un momento, 0 la historia nos dejara. a nivel nacionalla actividad academica y los proyectos en geoffsica siguen avanzando. ei ingeominas, el igac, ecopetrol, el osso, entreotros, desarrollan proyectos relacionados con slsrnica, gravimetrfa, magnetometrfa y sismologfa principalmente; aplicados a la prospecci6n, al entendimiento de la tect6nica basica y a la prevenci6n de desastres. ei compromiso nacional de establecimiento de programas de posgrado en geoffsica y en general en las diversas areas de las ciencias de la tierra, se hace cada vez mas urgente, en este aspecto celebramos la apertura de la especializaci6n en meteorologfa en la facultad de ciencias de la universidad nacional. han surgido nuevas empresas de servicios geoffsicos y otras se han acabado. las aguas subterraneas, la soluci6n de problemas de ingenierfa y de minerfa e incluso ambientales, han aumentado la demanda de esfuerzos muchas veces personales que debemos estimular y abrirles una puerta para la discusi6n y anal isis de problemas especfficos. a nivel internacional la adquisici6n de informaci6n se ha desarrollado aceleradamente y se trabaja en problemas de interpretaci6n, especial mente en slsrnica 3-d, tomograffa, geof6nia y modelamiento nurnsrtco de problemas electromagneticos. ei mar, la atm6sfera y el interior profundo con sometidos permanentemente al anal isis cuidadoso de la geoffsica. nos complace colaborar una vez mas con el desarrollo de la geoffsica nacional, en esta oportunidad geoffsica colombiana presenta 10 artfculos para su discusi6n. ellos tienen que ver con diversos temas que van desde la teorfa en aspectos ffsico-matematicos en el trabajo de la ffsica dora bernal relacionado con un curso de mecanica newtoniana, hasta la presentaci6n de los ultirnos valores de gravedad absoluta medidos en el pais que presentan los ingenieros martfnez, florez y sanchez del igac. hay 2 trabajos sobre sismologfa y 2 mas sobre manejo de informacion meteorol6gica. se completa el nurnero con un trabajo sobre prospecci6n geoelectrica, otro sobre anornalias maqneticas, uno mas sobre un modele barico de geoide para colombia y otro sobre calculo de parametres elasticos a partir de datos sfsmicos. de este modo creemos presentar un corto pero amplio abanico de 10 que en colombia se esta trabajando en las areas geoffsicas. su colaboraci6n en la discusion y anallsis, sigue siendo para "geoffsica colombiana" un aporte y una necesidad. , issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n1.62063 earth sciences research journal earth sci. res. j. vol. 26, no. 1 (march, 2022): 83-89 g e o m o r p h o l o g y record manuscript received: 19/01/2019 accepted for publication: 21/01/2022 how to cite item: qian, f., wang, w., zhang, x., zhang, j., & bian, z. (2022). comprehensive landscape characteristics of the arable land red line in donggang, china. earth sciences research journal, 26(1), 83-89 https://doi. org/10.15446/esrj.v26n1.62063 comprehensive landscape characteristics of the arable land red line in donggang, china fengkui qian1*, weiwen wang 2, xuefeng zhang3, jingye zhang4, zhenxing bian1 1. college of land and environment, shenyang agriculture university, shenyang, china, 2 college of geography and environmental science, henan university, kaifeng, china 3 dalian tianshi real estate consultant co., ltd, dalian, china 4 liaoning longfeng land real estate and assets appraisal co., ltd, shenyang, china. *corresponding author: fkqian@syau.edu.cn abstract: in the face of endless idle and even abandoned farmland, it is trying to protect farmland resources and ensure food supply by determining the bottom line (red line) of arable land that cannot be changed in use. the red line has landscape characteristics and functions, but there remain limited understandings of how to judge the dominant landscape characteristics of cultivated land. therefore, this study divides landscape characteristics into aesthetic value and stability according to landscape pattern. a comprehensive evaluation system was developed to explore the dominant landscape features of the cultivated resources based on a multi-factor comprehensive evaluation. on this basis, this study conducts an empirical study in donggang, a coastal city in china. the results show that cultivated land resources with high aesthetic value are distributed near cities, major roads, and important natural resources. the cultivated land resources with greater continuity and compatibility with adjacent plots are relatively stable, mainly paddy fields. the red line of cultivated land should focus on the characteristics of cultivated land stability. at the same time, the aesthetic value of cultivated land resources is also an important factor (related degree is 0.852). according to this evaluation system, the dominant area of arable land protection in donggang city is 88,027.21hm2, which is the preferred area for the red line of arable land. keywords: cultivated land; red line; aesthetic landscape; landscape pattern; donggang análisis amplio de las características de la “línea roja” cultivable en donggang, china resumen: con el fin de resolver el alto índice de vacancia o incluso el abandono de tierras de cultivo, en este trabajo se busca proteger los recursos de tierras cultivables y asegurar el suministro de comida al determinar el margen inferior (línea roja) del terreno apto para sembrar al cual no se le debe cambiar el uso. para efectos de este estudio, se estableción la línea roja en 1800 millones de mu (cada mu corresponde a 1/15 de hectárea). la línea roja tiene funciones y características paisajísticas pero deben estar definidos los criterios de cómo juzgar las características dominantes paisajísticas de las tierras cultivadas. además, en este estudio se dividen las características paisajísticas entre valores estéticos y de estabilidad de acuerdo con los patrones del paisaje. se desarrolló un sistema amplio de evaluación para explorar las características dominantes del paisaje en los terrenos cultivados con base en la evaluación amplia multifactorial. con base en este enunciado, este trabajo consiste en un estudio empírico en la ciudad costera de donggang, china. los resultados empíricos muestran que las tierras cultivadas cerca de las ciudades, carreteras principales y reservas naturales tienen un valor mayor estético, cerca de 85440 hm2, y que corresponden a un 81.80 % de las tierras cultivadas. las tierras adyacentes cultivadas y las tierras altamente compatibles para el cultivo en las planicies tienen una alta estabilidad, y el tipo de paisaje dominante es el arrozal, con una área de unos 91682 hm2, que corresponde al 87.77 %. los análisis de las características del paisaje con las ventajas compuestas de las tierras cultivadas muestra que el área de los recursos cultivados tanto con las ventajas estéticas y de estabilidad es de cerca de 82.653hm2, que corresponde al 79,13%. este es el límite sugerido para establecer la línea roja de protección a la tierra cultivable. palabras clave: terreno cultivado; línea roja; paisaje no estético; patrón de paisaje; donggang https://doi.org/10.15446/esrj.v26n1.62063 https://doi.org/10.15446/esrj.v26n1.62063 https://doi.org/10.15446/esrj.v26n1.62063 mailto:fkqian@syau.edu.cn 84 fengkui qian, weiwen wang, xuefeng zhang, jingye zhang, zhenxing bian introduction arable land is a significant resource for the livelihood of communities (wang et al., 2021). for agriculture-based civilizations, arable land supplied basic human needs (hanioka et al., 2018, yang et al., 2021). during industrialization, arable land was a solid foundation for economic and social development as the carrier of production materials (lageras and fredh, 2020; zhou and lv, 2020). during ecology-centric construction, arable land can provide landscape value beyond basic resources (liang et al., 2020; schirpke et al., 2021). therefore, arable land as a complex system has been involved in an aesthetic landscape function and played an important role in providing social security. the term of arable land “red line” proposed by the chinese government, should be strictly used as the baseline for arable land protection (zhou and lv, 2020; qian et al., 2021). from an ecological landscape perspective, arable land can also contribute to an ornamental landscape (shifaw et al., 2019). arable land is an important factor in the rural landscape and can be used to enhance diversity with mountains and lakes (jiang et al., 2021). furthermore, improvements in arable land patterns can strengthen ecological protection, improve the connectivity of green spaces, and enhance the beautification of the rural landscape (liang et al., 2020). the concept of landscape ecology originated in the 1930s, was defined as the science and art of studying and improving the relationship between spatial patterns and ecological processes at multiple scales and organizational levels (gao and liu, 2021; mansourian, 2021). since landscape ecology was introduced in china regarded as the unity of ecology and geography, domestic scholars have achieved a lot of farmland landscape designs, aesthetics, and ecological planning incorporating chinese characteristics (mirghaed et al., 2020; isik and vessel, 2021). along with increased research into landscape patterns, the ornamental value of the farmland landscape has become a research focus (mirghaed et al., 2020). most foreign scholars have focused on landscape quality and on the aesthetic value of the landscape (schirpke et al., 2021; yazici and asur, 2021). traditional research on landscape quality has involved investigation and analysis of the visual characteristics of a target area or the visual landscape of a geographic area (gao et al, 2022; isik and vessel, 2021). keywords for this research have included visual landscape quality, landscape visual quality, scenic beauty, and so on, with research content focusing on the evaluation of visual aesthetic quality (krupowicz et al., 2020; isik and vessel, 2021; yazici and asur, 2021). compared with foreign research, environmental impact assessments of the visual landscape in china are mainly focused on aspects including landscape design for urban planning and road aesthetics, and research methods have prioritized theoretical analysis over quantitative or spatial evaluation (jiang et al., 2020a; gao and liu, 2021; qian et al., 2021). the visual landscape is an important index for assessing aesthetic value, which will be quantified in this paper. figure 1. research theoretical framework in china, arable land protection has been improved from quantitative protection to comprehensive protection of quality and landscape (yang et al., 2020; wang et al., 2021). the arable land red line represents the cordon of arable land protection, has ensured both stable patterns of food production and great aesthetic values for environmental protection (sakellariou et al., 2021). therefore, by using an evaluation index system, this study tentatively divided the landscape characteristics of the arable land red line into two categories: the aesthetic landscape characteristics and landscape patterns static characteristics (figure 1). the main objectives of this research are to (1) evaluate the aesthetic landscape characteristics of arable land; (2) quantify the landscape pattern characteristics of arable land. and (3) establish a framework to explore comprehensive features of the arable land using an empirical model and providing a basis for the determination of the arable land red line. materials and methods study region descriptions donggang city is located between 123°22′30″ and 124°22′30″ e and 39°45′ and 40°15′ n near the liaodong peninsula and the mountains of eastern liaoning in china (figure 2). its geographical conditions include the yellow sea in the south and the yalu river in the east. the terrain of the study area drops in elevation from north to south. the countryside is hilly in the north, low hilly areas are found in the central region, and coastal plains with flat topography and fertile soil ideal for planting high-quality rice are distributed in the south. donggang city is influenced by a humid monsoon climate with annual mean precipitation of 899 mm. the climate is suitable for agricultural production. donggang city is a major rice-producing area within liaoning province. according to the overall plan of land utilization of donggang city (2006–2020), arable land and primary arable land shall be less than 88,500 hm2 76,370 hm2 in 2020. the arable land resources are rich, which can not only ensure food security but also prompt a focus on ecological protection of the aesthetically pleasing scenic line in the construction and development of urban areas. data sources and methods the main data sources included: (1) land survey data including land use type (e.g. ecological land, traffic routes, the urban boundary); (2) digital evaluate map (dem) of donggang city, which provided slope attribute data, and (3) farmland classification data, from which the arable land data were extracted. this study analyzed the asymmetry of comprehensive landscape features, which can reveal the frequency distribution rule of each value (xiang and li, 1981), as listed in equation 1. 3 1 ][ )2)(1( ∑= -= n i i s xx nn n cso (1) where cso is the deviation coefficient, n is the number of samples, xi is the value of sample i, x is the median of the sample, and s is the standard deviation of the sample. this study analyzed the influence of aesthetic landscape characteristics and landscape patterns characteristics on comprehensive characteristics using the pearson correlation coefficient (zhang et al., 2005), as listed in equation 2. y i n i x i s yy s xx n r )()(1 1 -= σ = (2) where r is the pearson correlation coefficient, n is the number of samples, xi and yi are the values of x and y variables of sample i respectively, `x and`y are the mean values of x and y respectively, and sx and sy are the standard deviations of x and y respectively. evaluation processes the evaluation processes included five steps: (1) construction of the evaluation unit; (2) index selection for the aesthetic landscape characteristics; (3) index selection for landscape patterns characteristics; (4) calculation of indices score; and (5) building the evaluation model. d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); 85comprehensive landscape characteristics of the arable land red line in donggang, china construction of the evaluation unit the evaluation unit, which is the most basic evaluation object, has internal attribute uniformity. evaluation units have the same soil properties (e.g. organic content and surface soil texture), and the same land-use type and topography. in this study, the evaluation unit was formed by overlaying soil maps, topographic maps, and land use maps with the same spatial scale. according to this principle, all arable land plots were divided into 30,656 evaluation units using the second national land survey data of donggang city in 2012. aesthetic landscape characteristics of arable land landscape aesthetics, closely related to the landscape pattern, not only contains social and cultural values but also reflects ecological values. it should be paid attention to when carrying out resource protection and planning to give play to the aesthetic value of landscape utilization(kerebel et al., 2019; liu and schroth, 2019). landscape aesthetics can be studied quantitatively, and the commonly used methods are psychophysical (schirpke et al., 2019). the main idea is to understand the relationship between landscape and aesthetics as a stimulus-reflective relationship (tribot et al., 2018; ha and yang, 2019). therefore, aesthetic landscape characteristics were further divided into the number of views and ornamental value. the number of views is very important to the quality of life for current residents (a crucial factor in attracting future growth and development to the valley) (ha and yang, 2019). this factor was evaluated considering various viewpoints to provide a scientific basis for the aesthetic protection of the farmland landscape. this study used the access gallery and city boundary as principal viewing positions and for calculating land plot numbers. that is the number of points that can be seen from arable land by human activity sites including urban boundaries and roads, regardless of any angle. larger numbers indicated greater visual landscape features and a significant landscape aesthetic function. scores for the number of views were calculated using the cut-off value method. the ornamental value of arable land is mainly based on the minimum distance from the arable land patches to regions of important natural resources. reservoirs, seashores, and nature reserves are crucial natural resource regions and serve as important ecological barriers to land use. it is also important to protect the ecological function of arable land resources. this study calculated ornamental value scores using the cut-off value method. landscape patterns characteristics of arable land landscape pattern generally refers to its spatial pattern, that is, the spatial arrangement and combination of landscape elements of different sizes and shapes, including the type, number, and spatial distribution, and configuration of landscape components (jiang et al., 2018, jiang et al., 2019, jiang et al., 2020b). therefore, the quantitative indicators of landscape pattern characteristics in this study include compatibility with adjacent plots, degree of continuity, fractional dimension (fd), and shannon diversity index (shdi). compatibility with adjacent plots is an important index for assessing the stability of arable land resources. greater compatibility means a lower risk of development, greater stability, and more sustainable utilization of arable land resources. compatible plots included arable and ecological land (such as garden plots, woodlands, grassland, water areas, and beaches). compatibility value was based on a ratio of compatible length to the total length of every arable land plot. this study calculated the score of compatibility with adjacent plots using the cut-off value method. figure 2. location and land use status of the study area 86 fengkui qian, weiwen wang, xuefeng zhang, jingye zhang, zhenxing bian the degree of continuity is related to the scale of land use and aid in characterize agricultural landscape patterns. thus, the continuity of arable land should be a conservation priority. in this study, the degree of continuity is taken to be the covered area of intersecting arable land resources within a 10-m buffer area. the degree of continuity was calculated using the cut-off value method. fd is a landscape pattern index measuring the shape of landscape patches, and the fd values of each arable land patch were calculated according to equation 3. fd values theoretically range from 1.0 to 2.0. values close to 1.0 indicate stronger self-similarity and geometric rules, as well as greater effects of human disturbance (frohn, 1997). fd scores were calculated using the cutoff value method. csln fd += = )(ln)4/(l 2 k k (3) where fd is the fractional dimension, l is the perimeter, s is the area, k is the slope of the regression system, and c is the offset. the shdi is based on information theory and reveals landscape heterogeneity. it is especially sensitive to the unbalanced distribution of various types of patches. the shdi values of each arable land patch were calculated according to equation 4. the shdi emphasizes the contribution of rare patches. furthermore, shdi is a biochemical indicator for the comparison and analysis of changes in diversity and heterogeneity over different periods. when shdi equals 0, the whole landscape consists only of one plot. a larger shdi value indicates balance and diversity in the distribution of patch types (shannon and weaver, 1998). in this study, scores were calculated using the cut-off value method. )(ln i m i ppshdi σ-= (4) where shdi is the shannon diversity index, m is the number of patch types, and pi is the area of land type i. calculation of indices score this study divided the indices into ranges and calculated their scores using the cut-off value method (table 1). evaluation model construction the assessment model was based on a multi-factor comprehensive evaluation method (saaty, 1988), as summarized in equation 5. ∑ = = n j ijiji fwc 1 (5) where ci is the comprehensive score of evaluation unit i, fij is the score of evaluation unit i for evaluation factor j, and wij is the weight of evaluation unit i for evaluation factor j. the evaluation methods for index weight primarily included the analytical hierarchy process (ahp) and expert scoring method (table 2). table 2. index weight evaluation factors destination layer index weight aesthetic landscape characteristics number of views 0.25 ornamental value 0.25 landscape patterns characteristics compatibility with adjacent plots 0.15 degree of continuity 0.15 fd 0.1 shdi 0.1 results and discussion evaluation analysis of aesthetic landscape characteristics the aesthetic landscape is a significant perspective in landscape protection that can reflect the aesthetic value of arable land resources. in this study, aesthetic landscape characteristics were analyzed using the number of views and ornamental value. this study synthetically analyzed the aesthetic landscape characteristics (figure 3a) and further analyzed features of the number of views (figure 3b) and ornamental value (figure 3c). from figure 3a, aesthetic landscape characteristics scores were mainly distributed within the range of 76-88. the greatest-scoring areas were located around the city center, along both sides of arterial traffic, and adjacent to important natural resources. arable land in those regions had a greater number of views and ornamental value. areas with dynamic characteristic scores greater than 80 covered approximately 22,659 hm2, primarily distributed along both sides of china national highway 201, and near-coastal benched and other water areas located in the jurisdictions of haiyanghong farm, huangtukan farm, donggang city economic zone, and xinglong farm. the arable land in these regions was considered to have the best visual landscape with an aesthetic scenic line. this feature can also be incorporated into the red line to control the boundary of urban development. the visual landscape is beneficial to the groupand series type urban development, as it helps prevent a ‘street city’ condition in which all cities are connected directly without a buffer zone (e.g. farmland or forest). the number of views ranged from 0 to 606. in the north, the area with a view score equal to 100 covered approximately 3,980 hm2 and was distributed primarily around the danda highway. in the south, the area scoring 100, covered approximately 14,216 hm2 and was distributed in peri-urban areas, and along both sides of the roadway. scores in the north were significantly lower than those in the south owing to the more complex terrain in the northeast. in the north, low mountains, and hilly land block the line of sight. the south has an open field of vision, along with the main traffic corridors. the city center is also located in the southeast and relates to these roadways. thus, arable land resources in the south had greater view scores. the area of these high-scoring regions, distributed along the perimeter of the center and on both sides of arterial traffic, covered approximately 9,210 hm2. these regions also exhibited higher-quality natural conditions and location advantages. at the policy level, table 1. evaluation index calculation destination destination layer index score summary aesthetic landscape characteristics number of views 0 20 40 60 80 100 land plot numbers can be viewed in principle position0 [1-3] [4-8] [5-21] [22-53] >54 ornamental value >2000 [1001-2000] [701-1000] [401-700] [201-400] [0-200] distance from patches of arable land to important natural land landscape pattern characteristics compatibility with adjacent plots 0 (0-0.66] (0.66-0.82] (0.82-0.95] (0.95-0.99] >0.99 side proportion of compatible plots degree of continuit [0-0.32] [0.32-0.70] [0.70-1.37] [1.37-2.66] [2.66-5.54] >5.54 contiguous area fd 1 (1-1.06] (1.06-1.08] (1.08-1.11] (1.11-1.14] >1.14 self-similarity shdi —— 1.1194 1.5790 1.7039 1.8013 1.8846 balance degree of land utilization d:/youdao/dict/6.3.69.8341/resultui/frame/javascript:void(0); 87comprehensive landscape characteristics of the arable land red line in donggang, china the recommended arable land red line in this region satisfied the construction needs for permanent capital farmland. in terms of the visual landscape, arable land in this area is indispensable. thus, the aesthetic value and red-line function of arable land resources are consistent. the ornamental value of arable land was mainly based on the distance from patches of arable land to important natural resources. these distances ranged from 0 to 7,474 m, with a mean distance of approximately 247 m. the important natural resources of donggang city are mainly distributed in the north and along the southern coast. the areas with ornamental scores of 100 cover approximately 24,709 hm2. these regions should be classified as ecological conservation areas. arable land in this region, reliant upon the natural terrain, maintains the continuity of the natural landscape and, in conjunction with the adjacent landscape, can provide valuable ecological services. figure 3. aesthetic landscape characteristics(a) and explanatory variables: number of views (b) and ornamental value(c) of arable land in donggang city (source: farmland classification map of donggang city in 2012) evaluation analysis of landscape pattern characteristics the landscape pattern characteristics mainly reflect the diversity and disturbance degree of the landscape, which in turn indicates the stability and degree of development and utilization of the land. in this study, landscape pattern characteristics were analyzed by their compatibility with adjacent plots, degree of continuity, fd, and shdi. a higher score meant greater permanence and stability and prominent landscape characteristics (figure 4a), and further analyzed the features of compatibility with adjacent plots (figure 4b), fd (figure 4c) degree of continuity (figure 4d), and shdi (figure 4e). landscape pattern characteristics scores were mainly distributed between 70 and 82. the area with scores greater than 75 covered approximately 12,931 hm2, mostly distributed in the jurisdictions of wusi farm, xiaomianzi town, beijingzi town, and changshan town. arable land in the high-scoring region exhibited less conflict with adjacent plots, greater compatibility and continuity, a more regular shape, and a more stable ecosystem. the arable land of these regions should be well-maintained to reduce the impacts of human activity and should be incorporated into the permanent prime farmland area. in terms of compatibility with adjacent plots, proportions ranged from 0 to 1. there were approximately 14,186 plots with a value of 1. the arable lands in these areas had greater stability, continuity with the adjacent ecological land, protection of biodiversity, and overall environmental quality. the best-scoring jurisdiction was shifan farm, in which 97.26% of arable lands had a value of 1, followed by the jurisdictions of haiyanghong farm, xiaomianzi town, changshan town, and heigou town. the contiguous total area of arable land which the buffer intersects ranged between 0.06 and 242.94 hm2. in the northwest arid areas, arable land had low contiguity, and the mean contiguous area was approximately 2.22 hm2. in the southeastern coastal areas, the mean contiguous area was approximately 5.74 hm2. this area hosts abundant water resources, and the arable land comprises mostly paddy fields. the most contiguous jurisdiction was the developed zone, where the mean contiguous area was 6.79 hm2, followed by the jurisdictions of wusi farm, dadong office, beijingzi town, and changshan town. arable land with a greater degree of continuity has prominent landscape characteristics and greater stability and makes greater use of scale. fd values ranged from 1 to 1.39. lower scores indicated that arable land had greater self-similarity and stability and more geometric rules. regular patches of arable land also have greater aesthetic value than do irregular patches at large scales. the region with fd scores of 100 was mostly distributed around the city center and along both sides of arterial traffic in the southeast of donggang city and included the jurisdictions of xinglong farm, wusi farm, beijingzi town, and changshan town. shdi values ranged from 1.1194 to 1.8846. the number of patches with scores of 100 was approximately 97.26%. greater shdi values indicate balanced land utilization, a stable ecological system, and a highquality ecological environment. the areas with high shdi values included the jurisdictions of xincheng office, dadong office, the developed zone, huangtukan farm, shifan farm, wusi farm, xinxing office, and xinglong farm. these regions should be priorities in the protection of landscape pattern characteristics. figure 4. landscape pattern characteristics(a) and explanatory variables: compatibility with adjacent plots (b), fd (c), contiguous total area (d), and shdi (d) of arable land in donggang city (source: farmland classification map of donggang in 2012) 88 fengkui qian, weiwen wang, xuefeng zhang, jingye zhang, zhenxing bian evaluation analysis of comprehensive landscape features in terms of the comprehensive landscape features, uniform distribution ranged from 0 to 606, with a standard deviation of 16.835, and asymmetry was -0.073 (equation 1). the correlation with the aesthetic landscape characteristics was 0.852 (equation 2), and the correlation with landscape pattern characteristics was 0.609. thus, the effect of aesthetic landscape characteristics was relatively prominent, indicating that aesthetic features of arable land resources are important in determining the arable land red line. this research divides the dominant zones and non-dominant zones according to the arable land protection objectives issued by the higher-level government. the node is a comprehensive value of 42, which can be used as the best choice for permanent basic farmland. as shown in figure 5, the areas with comprehensive landscape features scores greater than 42 covered approximately 88,027.21 hm2. in the south, this area covered approximately 49,570.56 hm2. in the north, this area covered approximately 38,456.62 hm2. this region was located along the perimeter of the central city and on both sides of arterial traffic and adjacent to important natural resources. these areas, which can be classified as permanent basic farmland not only have a low-relief surface, broad visibility, and greater ornamental value, but also regular shape, greater continuity, greater stability, and greater use of scale. aesthetic landscape characteristics are prominent in the dominant landscape area, of which the value of the number of views ranged from 41 to 606, and the ornamental value ranged from 0 to 638.61. meanwhile, the index of the compatibility with adjacent plots and shdi are advantages in the landscape pattern characteristics, with values ranging from 0.87 to 0 and 1.81 to 1.88, respectively. therefore, in the recommendations for the arable land red line, not only the resource value of arable land should be protected but also the landscape value. conclusion the result showed that the arable land red line is not only a quantitative line for ensuring food security, but also has a landscape value. this study discussed the aesthetic value of arable land resources through assessing the aesthetic landscape characteristics and landscape pattern characteristics. finally, the comprehensive landscape features of arable land resources wereb discussed, and the distribution of dominant landscape features was confirmed. the result confirms that arable land with greater aesthetic landscape characteristics is distributed around the city center, along the sides of arterial traffic, and adjacent to important natural resources. this area covered approximately 81.80% of arable land in donggang city. arable land with stable landscape patterns is distributed on plain terrain, which covered approximately 87.77% of arable land in donggang city. the arable land in this region comprises mostly paddy fields and has greater continuity and compatibility with adjacent plots. the evaluation of dominant landscape features found that areas with better visual features and stable landscape patterns covered approximately 79.13% of arable land in donggang city. from these findings, the paper recommends that the arable land red line should consider the aesthetic value and landscape patterns of arable land. in this study, we evaluated the landscape value of the arable land red line, using limited indices of aesthetic and landscape patterns features in a typical region of china. further study will include a more comprehensive index that considers a larger scale. in addition, index differences will also be considered for different geomorphological regions. figure 5. the distribution map of comprehensive landscape features of arable land resources in donggang city (source: farmland classification map of donggang in 2012) 89comprehensive landscape characteristics of the arable land red line in donggang, china acknowledgments the authors sincerely thank all the students and staff who provided their input into this study. this research was financially supported by the national natural science foundation of china (grant no. 42077149). our acknowledgments also extended to the anonymous reviewers for their constructive reviews of the manuscript. references gao, j., gong, j., yang, j. x., li, j. y. & li, s.c., (2022). measuring spatial connectivity between patches of the heat source and sink (scss): a new index to quantify the heterogeneity impacts of landscape patterns on land surface temperature. landscape and urban planning, 217, 104260. https://doi.org/10.1016/j.landurbplan.2021.104260 gao, s. & liu, s. f. (2021). exploration and analysis of the aesthetic cognitive schema of contemporary western urban landscapes. international journal of environmental research and public health. 18(10), 51-52. https://doi.org/10.3390/ijerph18105152 ha, s. c. & yang, z. p. (2019). evaluation for landscape aesthetic value of the natural world heritage site. environmental monitoring and assessment, 191. https://doi.org/10.1007/s10661-019-7607-9 hanioka, m., yamaura, y., senzaki, m., yamanaka, s., kawamura, k. & nakamura, f. (2018). assessing the landscape-dependent restoration potential of abandoned farmland using a hierarchical model of bird communities. agriculture ecosystems & environment, 265, 217-225. https://doi.org/10.1016/j.agee.2018.06.014 isik, a. i. & vessel, e. a. (2021). from visual perception to aesthetic appeal: brain responses to aesthetically appealing natural landscape movies. frontiers in human neuroscience, 15. https://doi.org/10.3389/fnhum.2021.676032 jiang, p. h., chen, d. s. & li, m. c. (2021). farmland landscape fragmentation evolution and its driving mechanism from rural to urban: a case study of changzhou city. journal of rural studies. 82, 1-18. https://doi.org/10.1016/j.jrurstud.2021.01.004 jiang, p. h., cheng, q. w., zhuang, z. z., tang, h. q., li, m. c., cheng, l. & jin, x. l. (2018). the dynamic mechanism of landscape structure change of arable landscape system in china. agriculture ecosystems & environment, 251, 26-36. https://doi.org/10.1016/j.agee.2017.09.006 jiang, p. h., li, m. c. & cheng, l. (2020a). dynamic response of agricultural productivity to landscape structure changes and its policy implications of chinese farmland conservation. resources conservation and recycling, 156. https://doi.org/10.1016/j.resconrec.2020.104724 jiang, p. h., li, m. c. & lv, j. c. (2019). the causes of farmland landscape structural changes in different geographical environments. science of the total environment, 685, 667-680. https://doi.org/10.1016/j.scitotenv.2019.05.383 jiang, p. h., li, m. c. & sheng, y. (2020b). spatial regulation design of farmland landscape around cities in china: a case study of changzhou city. cities, 97. https://doi.org/10.1016/j.cities.2019.102504 kerebel, a., gelinas, n., dery, s., voigt, b. & munson, a. (2019). landscape aesthetic modelling using bayesian networks: conceptual framework and participatory indicator weighting. landscape and urban planning, 185, 258-271. https://doi.org/10.1016/j.landurbplan.2019.02.001 krupowicz, w., czarnecka, a. & grus, m. (2020). implementing crowdsourcing initiatives in land consolidation procedures in poland. land use policy, 99. https://doi.org/10.1016/j.landusepol.2020.105015 lageras, p. & fredh, e. d. (2020). long-term development of landscape openness and arable land use in an agricultural region of southern sweden: the potential of reveals estimates using pollen records from wells. vegetation history and archaeobotany, 29, 113-124. https://doi.org/10.1007/ s00334-019-00756-z liang, x. y., li, y. b., ran, c. h., li, m. z. & zhang, h. (2020). study on the transformed farmland landscape in rural areas of southwest china: a case study of chongqing. journal of rural studies, 76, 272-285. https:// doi.org/10.1016/j.jrurstud.2020.04.017 liu, m. & schroth, o. (2019). assessment of aesthetic preferences in relation to vegetation-created enclosure in chinese urban parks: a case study of shenzhen litchi park. sustainability, 11. https://doi.org/10.3390/ su11061809 mansourian, s. (2021). from landscape ecology to forest landscape restoration. landscape ecology, 36, 2443-2452. https://doi.org/10.1007/s10980-02001175-6 mirghaed, f. a., mohammadzadeh, m., salmanmahiny, a. & mirkarimi, s. h. (2020). assessing the interactions between landscape aesthetic quality and spatial indices in gharasoo watershed, north of iran. international journal of environmental science and technology, 17, 231-242. https:// doi.org/10.1007/s13762-019-02342-2 qian, f. k., wang, w. w., wang, q. b. & lal, r. (2021). implementing land evaluation and site assessment (lesa system) in farmland protection: a case-study in northeastern china. land degradation & development, 32, 2437-2452. doi: 10.1002/ldr.3922 sakellariou, m., psiloglou, b. e., giannakopoulos, c. & mylona, p. v. (2021). integration of abandoned lands in sustainable agriculture: the case of terraced landscape re-cultivation in mediterranean island conditions. land, 10. https://doi.org/10.3390/land10050457 schirpke, u., tappeiner, g., tasser, e. & tappeiner, u. (2019). using conjoint analysis to gain deeper insights into aesthetic landscape preferences. ecological indicators. 96, 202-212. https://doi.org/10.1016/j.ecolind.2018.09.001 schirpke, u., zoderer, b. m., tappeiner, u. & tasser, e. (2021). effects of past landscape changes on aesthetic landscape values in the european alps. landscape and urban planning, 212. https://doi.org/10.1016/j.landurbplan.2021.104109 shifaw, e., sha, j. m., li, x. m., bao, z. c., legass, a., belete, m., ji, j. w., su, y. c. & addis, a. k., (2019). farmland dynamics in pingtan, china: understanding its transition, landscape structure and driving factors. environmental earth sciences, 78. https://doi.org/10.1007/s12665-019-8537-0 tribot, a. s., deter, j. & mouquet, n. (2018). integrating the aesthetic value of landscapes and biological diversity. proceedings of the royal society b-biological sciences, 285. https://doi.org/10.1098/rspb.2018.0971 wang, w., gong, j., wang, y. & shen, y. (2021). exploring the effects of rural site conditions and household livelihood capitals on agricultural land transfers in china. land use policy, 108. https://doi.org/10.1016/j.landusepol.2021.105523 yang, b., chen, x., wang, z. q., li, w. d., zhang, c. r. & yao, x. w. (2020). analyzing land use structure efficiency with carbon emissions: a case study in the middle reaches of the yangtze river, china. journal of cleaner production, 274. https://doi.org/10.1016/j.jclepro.2020.123076 yang, b., wang, z. q., zou, l., zou, l. l. & zhang, h. w. (2021). exploring the eco-efficiency of cultivated land utilization and its influencing factors in china’s yangtze river economic belt, 2001-2018. journal of environmental management, 294. https://doi.org/10.1016/j.jenvman.2021.112939 yazici, k. & asur, f. (2021). assessment of landscape types and aesthetic qualities by visual preferences (tokat, turkey). journal of environmental protection and ecology, 22, 340-349. zhou, b. b. & lv, l. g. (2020). understanding the dynamics of farmland loss in a rapidly urbanizing region: a problem-driven, diagnostic approach to landscape sustainability. landscape ecology, 35, 2471-2486. https://doi. org/10.1007/s10980-020-01074-w https://doi.org/10.1016/j.landurbplan.2021.104260 seismic event characterization is often accomplished using algorithms based only on information received at seismological stations located closest to the particular event, while ignoring historical data received at those stations. these historical data are stored and unseen at this stage. this characterization process can delay the emergency response, costing valuable time in the mitigation of the adverse effects on the affected population. seismological stations have recorded data during many events that have been characterized by classical methods, and these data can be used as previous “knowledge” to train such stations to recognize patterns. this knowledge can be used to make faster characterizations using only one three-component broadband station by applying bio-inspired algorithms or recently developed stochastic methods, such as kernel methods. we trained a support vector machine (svm) algorithm with seismograph data recorded by ingeominas’s national seismological network at a threecomponent station located near bogota, colombia. as input model descriptors, we used the following: (1) the integral of the fourier transform/power spectrum for each component, divided into 7 windows of 2 seconds and beginning at the p onset time, and (2) the ratio between the calculated logarithm of magnitude (mb) and epicentral distance. we used 986 events with magnitudes greater than 3 recorded from late 2003 to 2008. the algorithm classifies events with magnitude-distance ratios (a measure of the severity of possible damage caused by an earthquake) greater than a background value. this value can be used to estimate the magnitude based on a known epicentral distance, which is calculated from the difference between p and s onset times. this rapid (< 20 seconds) magnitude estimate can be used for rapid response strategies. the results obtained in this work confirm that many hypocentral parameters and a rapid location of a seismic event can be obtained using a few seconds of signal registered at a single station. a cascade scheme of svms or other appropriate algorithms can be used to completely classify an event in a very short time with acceptable accuracy using data from only one station. los algoritmos de determinación de parámetros hipocentrales empleados en la actualidad, se basan específicamente en la información recibida en las estaciones de monitoreo mas cercanas al epicentro y no tienen en cuenta la valiosa información histórica registrada a lo largo del tiempo en dichas estaciones. es por esto que los procesos de caracterización toman varios minutos, tiempo precioso que podría ser de gran utilidad en la generación de alertas tempranas que permitan una oportuna reacción ante el evento. el registro de información, a lo largo el tiempo, de sismos ocurridos en los alrededores de la estación, puede ser empleada para dotarla de algún grado de experiencia que le permita, mediante detección y clasificación de patrones, realizar una caracterización previa mucho mas rápida, mediante el empleo de técnicas modernas las cuales pueden ser algoritmos bio-inspirados o métodos estocásticos mas recientes conocidos como métodos kernel. en el presente trabajo se emplea un método conocido como maquinas de soporte vectorial (msv), entrenando dicho algoritmo con información de la relación del área bajo la curva de la potencia de la transformada de fourier de las componentes n-s, e-w y vertical, calculada para 5 ventanas de 2 segundos, desde la onda p, de 123 sismos de magnitud superior a 3, desde 2004 hasta 2008, alrededor de la estación el rosal, de la red sismológica nacional de ingeominas. el algoritmo clasifica sismos que superen un umbral predeterminado de la relación entre el logaritmo de la magnitud y la distancia, que refleja, de alguna manera, la intensidad del sismo. con la obtención de este parámetro será posible conocer la magnitud del evento, debido a que la distancia puede ser calculada, con base en picado de la onda s, y de esta manera establecer una aproximación rápida de la magnitud en un tiempo aproximado de 20 segundos después del evento. los resultados obtenidos permiten confirmar que con poco tiempo se señal en una sola estación sismológica es posible obtener información confiable para ser empleada en alertas tempranas. earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014): 115 122 abstract resumen key words: machine learning, seismology, single station, magnitude, distance. palabras clave: aprendizaje de máquina, sismología, una estación, magnitud, distancia. record manuscript received: 06/12/2013 accepted for publication: 05/09/2014 severity classification of a seismic event based on the magnitude-distance ratio using only one seismological station luis h. ochoa¹, luis f. niño¹, carlos a.vargas¹ ¹ universidad nacional de colombia seismology issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.41083 118 luis h. ochoa, luis f. niño, carlos a.vargas 1 introduction seismic activity is a natural phenomenon with enormous and uncontrollable power, and damage in populated areas can impede appropriate responses when earthquakes occur. recent developments have allowed efficient and alternative solutions inspired by natural and biological processes that have been converted to stochastic models and have been used in seismology with acceptable results. when an earthquake is detected, seismologists determine its main features, such as magnitude, hypocenter coordinates, fault mechanisms, etc., to help authorities respond appropriately and allow scientists to study the earth internal structures. seismometers are distributed across an area of interest, many with three-component broadband sensors that detect, at the same point, movement in three mutual, orthogonal directions (north-south, east-west and up-down). these stations transmit information in near real time to seismological centers. the traditional characterization of earthquakes takes a few minutes, depending on the degree of automatization, network density and the number of stations involved in the process. earthquakes usually must be detected by at least five stations, and features such as compressive (p) and shear (s) wave arrival times are determined by expert analysis or an automatic algorithm. subsequently, the hypocenter coordinates can be calculated. in colombia, an institute called ingeominas can obtain seismic event localization in approximately 8 minutes due largely to the low-density network. in some seismological centers with many stations, this time can be reduced by half because earthquake energy requires additional time to arrive at stations farther away from the hypocenter. however, we are mostly interested in earthquakes occurring near population centers, where seismic energy can arrive in tenths of a second, making warning impossible. long processing times are associated with conventional methods that are based on information related only to the particular event and that ignore valuable previously collected historical information. therefore, new alternatives must be explored to reduce the processing time and allow rapid generation of warnings. the main purpose of this study is to use historical information from only one station to detect patterns correlated with known event features available from very accurate, traditional processing. 2 problem definition colombia has a central seismological network, managed by ingeominas, and local networks located in certain cities, such as armenia (quindio) and cali (valle). it is important to develop a system based on these valuable resources to discover patterns that will allow more rapid seismic early warnings. recent developments in computational intelligence (ci) techniques allow this goal to be achieved by solving certain seismological problems in near real time. the automatic detection and discrimination of different events recorded with a seismometer, including seismic and non-seismic signals, using ci have been achieved using artificial neural networks (anns) in many cases around the world. anns use a probe that is an excellent pattern recognition tool. anns were created to reproduce the mechanisms of human reasoning and allow discrimination of signals to simulate the brain processing of the operator. it is an efficient processing strategy for solving seismic signal discrimination problems, in which multiple hypotheses must be controlled in parallel, the number of input parameters is large and a well-defined solution is not always well known. the onset time determination is usually the first step to locating the hypocenter of an event. the first arrival of a p wave can be compressive or tensile, which is key to determining the focal mechanism. interactions between body waves and changes in the mechanical properties of rocks at strata interfaces generate diffraction and reflection phenomena, generating new body waves. these new onsets can provide clues to the distribution of strata in the subsurface, and these clues can then be used as seismic tomography data to determine the earth’s internal structure. the automatic detection of p waves can be accomplished using comparison algorithms, such as the one reported by allen 1978, who developed an application to determine p wave onset by comparing discrete signal parameters. the p onset is present where the short-term average (sta) of the signal increases abruptly with respect to the long-term average (lta). later, allen 1982 established that technological development will lead to more efficient s wave detection through the use of a signal-in-frequency domain. baer and kradolfer, 1987, developed an improved methodology based on signal processing. the complexity of the s wave onset requires alternative methods to achieve better solutions. cichowicz, 1993, developed an algorithm that identifies the s wave onset using three-component seismometers in local networks. recently, a neural network approach has been proposed to solve this problem because many simultaneous parameters can be included and the approach is easy to implement and flexible. it is also not necessary to be proficient with the interactions of the analyzed variables. wang and teng, 1997, applied a back propagation neural network (bpnn). dai and mcbeth, 1997, implemented a bpnn to identify and detect p and s waves from threecomponent recordings of local earthquake data. two years later, zhao and takano, 1999, used bpnns for p wave detection by combining three bpnn detectors (long, mid and short term). aldersons, 2004, developed a system (mannekenpix) that automatically detects the phase of a wave. this program is based on a weighting mechanism calibrated with reference detections and weights provided by the user; in other words, it is a learning process that uses examples and statistical techniques to separate groups by criteria similitude. this approach is a linear combination of predictors to maximize differences among groups, while minimizing differences within groups. di stefano et al. 2005 applied this algorithm to achieve precise phase detection using pg, pn and their polarities. these authors demonstrate that such data are suitable for high-precision earthquake location, focal mechanism determination and highresolution seismic tomography. once signal spectral characterization is complete and arrival times for p and s signals are determined, it is possible to generate the subsurface structure. this issue has been resolved using computational intelligence techniques, such as genetic algorithms (gas) and simulated annealing (sa). a good subsurface structural model is an important step for calculating the focal parameters of seismic events. the high model combination rates obtained with ga can perform a very wide exploration of the solution’s space, and this computation alternative is preferred when solving this kind of problem. bhattacharyya et al. 1999, analyzed regional seismograms using a ga to determine crustal structure in a four-layered crustal model of the eastern great basin and western colorado plateau, in which the optimizing variables were thickness and p-wave velocity. the resulting model agreed well with results reported in several geological publications. chang et al. 2004, implemented a ga and modeled crustal structure in southern korea using teleseismic p-wave receiver function data and rayleigh– wave phase velocity measurements. the genotype describes the distribution of wave velocities and is determined using 40 1-km-thick layers. lawrence and wiens, 2004, combined the inversion of body wave receiver functions and rayleigh wave phase velocities using a niching genetic algorithm (nga). using non-unique inversion models, it is possible to explore many solutions to problems associated with the determination of focal parameters. pezeshk and zarrabi, 2005, developed an inversion procedure for spectral analysis of surface waves (sasw) using a genetic algorithm. kim et al. 2006, proposed a method for determining hypocentral parameters for local earthquakes in 1-d using genetic algorithms and twopoint ray tracing, which calculates seismic ray trajectories that cross a known velocity structure. in this case, the velocity structure is not known, but the arrival times at different stations are available. with ga, it is possible to explore many subsurface velocity distributions, and a comparison between the calculated and observed arrival times can be a useful fitness selection function. assimaki and steidl, 2007, used a hybrid algorithm that combined a genetic algorithm with a local least-squares fit operator for inversion of weak and strong motion among an array of data obtained during the mw 7.0 sanriku-minami earthquake. this combination accelerates the convergence of the ga by improving the parental population with the local hill-climbing operator. many of the processes mentioned above can be described by computational intelligence tools, which must be designed, or trained, for 119severity classification of a seismic event based on the magnitude-distance ratio using only one seismological station particular geological, environmental and structural settings. additionally, these techniques can be applied to function in near real time by processing information at the seismological center or directly in the sensor using electronic devices with embedded models. these devices are faster and can transmit processed information via parallel link, requiring fewer resources for transmission. however, these processing tasks only use information transmitted from current seismic activity registered by the nearest stations. the processing takes approximately 5 to 10 minutes, which does not allow for early warnings and alerts. more stations can be used to determine seismic source magnitude and location, but more time is needed for processing; thus, less time is be available for reaction. the generation of early warnings depends on extracting the most accurate and reliable magnitude and location values using the fastest signal time. new data-driven pattern analysis models (e.g., kernel methods) with solid statistical foundations have been applied to classification and regression issues. these pattern analysis models apply historical data available in seismological databases to determine signal features (descriptors) and relationships between signal features and previously processed results. the descriptors and their relationships with previously processed results can be used to “train” certain simple algorithms, which can be used to process new data more rapidly. here, we present the first results obtained from a single three-component seismic station with our rapid and accurate characterization of a seismic event using only a few seconds of signal following the p wave onset. this station is near the source, where energy does not saturate the data, and uses support vector machine models to obtain a magnitude-distance ratio from 14 seconds of signal. 3 data set description the data set used in this research was obtained from ingeominas’s national seismological network. we have waveform files and calculated data from approximately 45,000 registered seismic events from 1993 to mid2008. we selected 130 of the events registered at the “el rosal” station that had magnitudes (mb) greater than 2.1 and epicentral distances less than 120 kilometers. this station has digital, three-component broadband seismometers and is located near bogota, which is vulnerable to mediumor high-magnitude events. this research is proposed as a first step to developing a reliable early warning system for the city. only events registered after late 2003 were used because the three-component station data were not available prior to this time. the location of selected events is presented in fig. 1, in which symbol size is proportional to event magnitude. 4 methodology one of the most important steps in kernel method processing is to determine the best “descriptors” to be used as input variables for the model. these descriptors are the key to obtaining good correlations with desired output, and the selection of descriptors depends on the type of information available for processing. in this case, we have time signals, which have certain characteristic features that can be described by signal transformations, such as fourier transform and others. • fourier transform was applied to each of the three components of the signal, starting at the p wave onset. • this transform was divided into 2-second windows (80 samples). • the integral of the squared frequency spectrum was calculated for each window, which generated one value for a given two-second partition. • finally, the ratios between the values of each component calculated in the step above are the input descriptors of the model and are calculated as follows: (1) (2) (3) where x, y and z are the values of the squared fourier transform integral at each x, y, and z component registered at the seismological station. these descriptors are used in our classification model as input variables. an output value is needed to separate the two classes that represent, in some way, the severity of a given earthquake’s effects at the station site. the effects of an earthquake at any place depend directly on magnitude and inversely on epicentral distance. a ratio between the logarithm of magnitude and epicentral distance is proposed here as a classifier based upon the following equation: , where m is body wave magnitude and d is epicentral distance, based on reports processed by ingeominas for each seismic event. the physical context of this classifier variable is related to the energy that an earthquake has when it reaches the seismic station; very near events with low magnitudes are perceived to be similar to distant events with high magnitudes. higher classifier values are obtained for high-magnitude earthquakes occurring close to the station. to achieve a balance between the number of events over and under a background limit, a value of 0.5 was determined to be the classification limit. with calculated descriptors and the information available for the classification variable, an svm was trained to find the optimal separator hyperplane that discriminates the events corresponding to classes a and b. appendix i includes the events used in the model and the calculated severity values from known magnitudes and distances, as well as the predicted svm model. 5 classifier evaluation a machine learning classifier performance depends entirely on the data used for training, validation and testing. therefore, the method’s evaluation must consider answers to certain questions about the data, such as ‘do we have enough training data to represent the model?’; ‘do we have balanced training data for each considered class?’; ‘what is the effect of re-using data in the training sets?’; etc. to answer these questions, the algorithms and data used in this research were evaluated using 10-fold cross validation techniques, and the results are summarized below. [complexity c=0.1] predicted real a b a 9 32 b 0 98 [complexity c=1] predicted real a b a 20 12 b 4 94 [complexity c=10] predicted real a b a 22 10 b 3 95 table 1: confusion matrix. 120 luis h. ochoa, luis f. niño, carlos a.vargas % correctly classified instances 87.69% % incorrectly classified instances 12.31% mean absolute error 0.12 mean square root error 0.35 relative mean absolute error 32.95 relative mean square root error 81.39 table 2: svm classifier with quadratic kernel results. a b true positives 62.5% 95.9% false negatives 4.0% 37.5% precision 0.830 0.887 recall 0.625 0.959 mean-f 0.714 0.922 roc curve area 0.792 0.792 table 3: estimated error of the svm quadratic kernel classifier. table 1 shows the results from the determination algorithm of the meta parameters, which determine the curve roc that was used to evaluate the value of c for which the classification model has the best results. figure 2 shows the curve roc for multiple values of c, achieving the best value as a compromise between hits and false alarms. similar curves were generated to locate other parameters, such as the degree of polynomial function used for mapping vectors in the feature space. using this approach, the quadratic kernel was selected as the best mapping function for this problem and dataset. figure 2 shows the relationship between the numbers of hits and false alarms, which are determined using the following equations: figure 2: roc curve for c estimation. wrongly classified events that exceeded the background limit of c, known as “false negatives”, are cases where no alarm was generated but a moderate event was imminent. this unique situation indicates that there is a local condition that must be analyzed carefully to determine the area’s unique pattern (figure 3). to increase accuracy and obtain better results, it is necessary to explore other descriptors. figure 4 plots the values of the confusion matrix. there is a very good correlation for a severity threshold of 0.5. this result can be used as an initial early warning parameter. figure 3: wrongly classified events (red points). figure 4: plotted confusion matrix. 6 results results obtained in this study using the classification methods described above are presented in tables 2 and 3. class b events were classified with high accuracy, but class a events were wrongly classified in 31% of the cases. future work will focus on the identification of new descriptors to better characterize the input signal using a priori knowledge. it is possible to implement new strategies using bagging, which combines many algorithms. the described algorithm proposes a new classification method for early warning purposes only. 7 conclusions • the structure of the input data is important for classifier behavior and determining relationships within the model. one of main challenges in this work was selecting the best signal descriptors and determining the relationships between them. • with a small signal interval processing from only one seismological station, it will be possible to estimate parameters such as magnitude and distance, which will allow the rapid classification of a seismic event based only on the effects at this station. • previous determinations of hypocentral parameters ignore historical data from previous seismic events and instead use only information recorded during current earthquakes. the proposed method uses historical data to calculate a relationship between magnitude and distance and allows rapid warning generation times. • the results suggest that it is possible to use the proposed method to calculate other hypocentral parameters, such as hypocenter location, azimuth, distance and others. 121severity classification of a seismic event based on the magnitude-distance ratio using only one seismological station • improvements to the technique can be accomplished by including a priori information in descriptors calculated from signals and spectral characteristics. these techniques have been used previously as input in some artificial intelligence approaches and resulted in better accuracy and faster processing times. • the adjustment results obtained using this approach cannot be compared with conventional strategies because they do not use the same information. this approach is based on a single station, whereas others are based on multiple stations (no fewer than 5). • early detection systems must consider combined strategies to optimize the identification of descriptors that can be used more efficiently in these types of approaches. references allen, r. v. (1978). automatic earthquake recognition and timing from single traces, bulletin of the seismological society of america. 68, 1521-1532. allen, r. v. (1982). automatic phase pickers: their present use and future prospects, bulletin of the seismological society of america. 72, s225-s242. assimaki, d. and steidl, j. (2007). inverse analysis of weak and strong motion downhole array data from the mw7.0 sanriku-minami earthquake, soil dynamics and earthquake engineering. 27, 73-92. baer, m. and kradolfer, u. (1987). an automatic phase picker for local and teleseismic events, bulletin of the seismological society of america. 77, 1437-1445. bhattacharyya, j., sheehan, a. f., tiampo, k. and rundle, j. (1999). using a genetic algorithm to model broadband regional waveforms for crustal structure in the western united states, bulletin of the seismological society of america. 89, 202-214. chang, s., baag, c. and langston, c. a. (2004). joint analysis of teleseismic receiver functions and surface wave dispersion using the genetic algorithm, bulletin of the seismological society of america. 94, 691-704. cichowicz, a. (1993). an automatic s-phase picker, bulletin of the seismological society of america. 86, 180-189. dai, h. and macbeth, c. (1997). application of back-propagation neural networks to identification of seismic arrival types, physics of the earth and planetary interiors. 101, 177-188. del pezzo, e., espossito, a., giudiceprieto, f., marinaro, m., martini, m. and scarpetta, s. (2003). discrimination of earthquakes and underwater explosions using neural networks, bulletin of the seismological society of america. 93, 215-223. di stefano, r., aldersons, f., kissling, e., baccheshi, p. and chiarabba, c. (2005). automatic seismic phase picking and consistent observation error assessment: application to the italian seismicity, geophysical journal international. 165, 121-134. dowla, f. u., taylor, s. r. and anderson, r. w. (1990). seismic discrimination with artificial neural networks: preliminary results with regional spectral data, bulletin of the seismological society of america. 80, 1346-1373. dysart, p. s. and pully, j. j. (1990). regional seismic event classification at the noress array: seismological measurements and the use of trained neural networks, bulletin of the seismological society of america. 80, 1910-1933. fedorenko, y. v., husebye, e. s. and ruud, b. o. (1999). explosion site recognition; neural net discriminator using single three-component stations, physics of the earth and planetary interiors. 113, 131-142. kim, w., hahm, i., ahn, s. j. and lim, d. h. (2006). determining hypocentral parameters for local earthquakes in 1-d using a genetic algorithm, geophysical journal international. 166, 590-600. lan, j., nahavandi, s., lan, t. and yin, y. (2005). recognition of moving ground targets by measuring and processing seismic signal, measurement. 37. 189-199. lawrence, j. f. and wiens, d. a. (2004). combined receiver-function and surface wave phase-velocity inversion using a niching genetic algorithm: application to patagonia, bulletin of the seismological society of america. 94, 977-987. muller, s., garda, p., muller, j. d., an cansi, y. (1999). seismic events discrimination by neuro-fuzzy merging of signal and catalogue features, physics and chemistry of the earth. 24, 201-206. pezeshk, s. and zarrabi, m. (2005). a new inversion procedure for spectral analysis of surface waves using a genetic algorithm, bulletin of the seismological society of america. 95, 1801-1808. romeo, g., mele, f. and morelli, a. (1995). neural networks and discrimination of seismic signals, computers & geosciences. 21, 279-288. tiira, t. (1996). discrimination of nuclear explosions and earthquakes from teleseismic distances with a local network of short period seismic stations using artificial neural networks, physics of the earth and planetary interiors. 97, 247-268. tiira, t. (1999). detecting teleseismic events using artificial neural networks, computers & geosciences. 25, 929-938. wang, j. and teng, t. (1995). artificial neural network-based seismic detector, bulletin of the seismological society of america. 85, 308-319. wang, j. and teng, t. (1997). identification and picking of s phase using an artificial neural network, bulletin of the seismological society of america. 85, 1140-1149. yamanaka, h. and ishida, h. (1996). application of genetic algorithms to an inversion of surface-wave dispersion data, bulletin of the seismological society of america. 86, 436-444. zadeh, m. a. and nassery, p. (1999). application of quadratic neural networks to seismic signal classification, physics of the earth and planetary interiors. 113, 103-110. zhao, y. and takano, k. (1999). an artificial neural network approach for broadband seismic phase picking. bulletin of the seismological society of america. 89. 670-680. zhou, r., tajima, f., and stoffa, p. l. (1995). application of genetic algorithms to constrain near-source velocity structure for the 1989 sichuan earthquakes. bulletin of the seismological society of america. 85. 590-605. 122 luis h. ochoa, luis f. niño, carlos a.vargas figure 1: selected earthquakes. appendix i: list of used events and parameters. date time lat lon dist mag depth hypo dist log mag dist hipo classification real (0.5) predicted 07/11/2002 22:22 4.008 -74.15 96 2.2 0 96.0 0.36 b b 13/11/2002 23:23 5.775 -73.906 112 2.5 115 160.5 0.25 b b 08/01/2003 09:09 4.492 -74.898 75 2.2 6 75.2 0.46 b b 14/04/2003 22:22 4.658 -74.535 32 2.5 3.1 32.1 1.24 a a 29/04/2003 01:01 4.05 -74.11 92 2.9 2.6 92.0 0.50 a b 08/06/2003 10:10 5.258 -73.649 88 2.4 156.2 179.3 0.21 b b 12/06/2003 11:11 4.636 -74.647 43 2.8 22.9 48.7 0.92 a a 20/06/2003 01:01 4.684 -74.785 54 2.7 21.9 58.3 0.74 a a 28/06/2003 23:23 5.59 -74.352 81 2.2 100.3 128.9 0.27 b b 01/07/2003 21:21 4.159 -75.146 119 2.8 0 119.0 0.38 b b 20/08/2003 23:23 5.519 -73.858 90 2.4 136 163.1 0.23 b b 01/09/2003 11:11 5.543 -74.389 76 2.6 104.3 129.1 0.32 b b 19/10/2003 02:02 5.618 -73.915 96 2.8 117.3 151.6 0.30 b b 20/10/2003 11:11 5.778 -74.675 109 2.2 6.8 109.2 0.31 b b 31/10/2003 00:00 5.213 -73.714 79 2.8 155.5 174.4 0.26 b b 03/11/2003 20:20 5.704 -74.413 94 2.2 0 94.0 0.36 b b 09/11/2003 14:14 5.375 -74.927 88 2.2 8.2 88.4 0.39 b b 20/12/2003 19:19 5.219 -74.546 47 3.8 2.2 47.1 1.23 a a 26/12/2003 12:12 5.779 -73.873 114 2.9 129.4 172.5 0.27 b a 02/01/2004 09:09 5.74 -74.311 98 2.5 1 98.0 0.41 b b 12/01/2004 17:17 5.589 -73.874 96 2.3 134 164.8 0.22 b b 24/01/2004 07:07 5.805 -73.878 116 2.8 122 168.3 0.27 b a 01/02/2004 10:10 5.431 -73.741 91 2.7 142.3 168.9 0.26 b b 10/03/2004 07:07 4.54 -75.06 88 2.3 26 91.8 0.39 b b 24/03/2004 01:01 5.868 -74.003 118 2.4 138.8 182.2 0.21 b b 123severity classification of a seismic event based on the magnitude-distance ratio using only one seismological station 01/04/2004 15:15 5.443 -73.781 89 2.2 140.6 166.4 0.21 b b 04/04/2004 16:16 4.347 -73.853 77 3.2 4 77.1 0.66 a b 17/04/2004 20:20 5.577 -73.884 94 2.6 135.2 164.7 0.25 b b 19/04/2004 00:00 4.579 -74.776 58 2.5 0 58.0 0.69 a a 24/04/2004 13:13 5.315 -73.774 80 2.6 140 161.2 0.26 b b 16/05/2004 01:01 5.629 -73.893 98 2.4 135.7 167.4 0.23 b b 25/05/2004 17:17 4.678 -74.679 43 2.6 14 45.2 0.92 a a 24/06/2004 23:23 4.686 -74.807 56 3.2 21.7 60.1 0.84 a a 28/06/2004 12:12 5.435 -73.769 89 2.7 140.7 166.5 0.26 b b 24/07/2004 22:22 5.215 -74.552 47 2.4 3.2 47.1 0.81 a a 22/08/2004 02:02 5.067 -74.761 53 2.5 13.6 54.7 0.73 a a 22/08/2004 17:17 5.596 -73.856 97 2.5 124 157.4 0.25 b b 26/08/2004 04:04 5.628 -73.986 93 2.3 111.9 145.5 0.25 b b 27/09/2004 14:14 5.785 -73.924 112 2.2 102 151.5 0.23 b b 07/10/2004 23:23 5.605 -74.381 83 2.2 96.4 127.2 0.27 b b 21/11/2004 14:14 5.361 -73.725 87 2.9 147.8 171.5 0.27 b b 22/11/2004 10:10 4.583 -75.138 95 3.1 0 95.0 0.52 a a 19/12/2004 14:14 4.053 -74.039 95 2.3 0 95.0 0.38 b b 01/01/2005 15:15 5.013 -74.844 60 2.4 18.3 62.7 0.61 a a 21/01/2005 07:07 5.419 -73.742 90 2.3 148 173.2 0.21 b b 23/01/2005 07:07 5.241 -73.686 83 2.3 153.1 174.2 0.21 b b 08/02/2005 09:09 5.367 -74.91 86 2.4 7.8 86.4 0.44 b b 11/02/2005 22:22 5.278 -73.777 77 2.4 155.4 173.4 0.22 b b 06/04/2005 19:19 5.791 -74.028 109 2.3 1.8 109.0 0.33 b b 16/04/2005 05:05 5.453 -73.817 87 2.2 134.4 160.1 0.21 b b 04/05/2005 12:12 5.291 -73.736 82 2.7 146 167.5 0.26 b b 26/05/2005 09:09 4.681 -74.863 62 2.9 22.2 65.9 0.70 a a 30/06/2005 03:03 5.326 -73.756 82 2.4 143.5 165.3 0.23 b b 10/07/2005 03:03 4.415 -75.05 94 2.3 8.7 94.4 0.38 b b 16/07/2005 12:12 5.594 -73.891 95 2.2 122 154.6 0.22 b b 16/07/2005 02:02 5.92 -74.499 119 2.2 28.7 122.4 0.28 b b 29/07/2005 22:22 5.632 -74.61 91 3 4 91.1 0.52 a a 05/08/2005 01:01 5.716 -74.38 95 2.2 58.7 111.7 0.31 b b 15/08/2005 18:18 5.608 -73.867 98 2.3 131.7 164.2 0.22 b b 19/08/2005 23:23 5.304 -73.751 81 2.4 150.5 170.9 0.22 b b 09/10/2005 04:04 5.325 -73.721 85 2.3 146.8 169.6 0.21 b b 15/10/2005 01:01 5.749 -73.973 106 2.2 101.5 146.8 0.23 b b 23/10/2005 13:13 5.744 -73.906 109 2.8 100.9 148.5 0.30 b b 12/11/2005 09:09 5.289 -73.751 80 2.2 144 164.7 0.21 b b 17/11/2005 03:03 4.728 -74.708 44 2.4 8.7 44.9 0.85 a a 21/11/2005 00:00 5.325 -73.761 82 2.2 150.1 171.0 0.20 b b 02/01/2006 00:00 5.45 -74.159 68 3.5 48 83.2 0.65 a b 07/02/2006 03:03 5.321 -73.76 81 2.7 143.6 164.9 0.26 b b 27/03/2006 02:02 5.286 -73.813 74 2.2 151.6 168.7 0.20 b b 04/06/2006 06:06 5.6 -73.869 97 2.2 131.5 163.4 0.21 b b 17/06/2006 05:05 5.287 -73.724 82 2.4 148.3 169.5 0.22 b b 20/06/2006 18:18 4.177 -74.54 79 2.6 0 79.0 0.53 a b 21/06/2006 03:03 5.435 -73.802 87 2.3 138 163.1 0.22 b b 10/07/2006 07:07 5.284 -73.738 81 3.4 155.7 175.5 0.30 b b 06/08/2006 04:04 5.615 -74.045 90 2.3 107.3 140.0 0.26 b b 16/08/2006 04:04 5.301 -73.754 81 2.3 155.2 175.1 0.21 b b 06/09/2006 00:00 5.641 -73.88 100 2.4 132 165.6 0.23 b b 06/09/2006 09:09 5.299 -73.735 82 2.4 149.5 170.5 0.22 b b 124 luis h. ochoa, luis f. niño, carlos a.vargas 18/09/2006 17:17 5.312 -73.745 82 2.5 154.9 175.3 0.23 b b 06/10/2006 03:03 5.281 -73.713 83 2.5 151.4 172.7 0.23 b b 23/10/2006 22:22 5.289 -73.709 84 3.4 148.9 171.0 0.31 b b 01/11/2006 12:12 5.206 -74.013 52 3.9 126.5 136.8 0.43 b b 07/11/2006 07:07 5.274 -73.708 83 3.5 151.8 173.0 0.31 b b 12/12/2006 07:07 5.09 -74.781 56 2.8 22 60.2 0.74 a a 18/01/2007 19:19 5.591 -74.086 86 2.5 103.3 134.4 0.30 b b 27/01/2007 05:05 5.515 -74.078 78 2.7 112.2 136.6 0.32 b b 31/03/2007 01:01 5.69 -73.865 106 2.4 131.7 169.1 0.22 b b 12/04/2007 16:16 5.485 -73.82 90 2.4 144.5 170.2 0.22 b b 16/04/2007 07:07 4.582 -74.754 56 3 28.3 62.7 0.76 a a 03/05/2007 07:07 5.737 -73.902 108 2.2 98.5 146.2 0.23 b b 19/05/2007 07:07 4.526 -75.309 115 2.6 0 115.0 0.36 b b 29/06/2007 17:17 5.303 -73.705 85 2.2 148.3 170.9 0.20 b b 10/07/2007 17:17 5.272 -73.664 87 2.4 153.6 176.5 0.22 b b 14/07/2007 10:10 5.238 -73.701 82 2.3 143.2 165.0 0.22 b b 16/10/2007 10:10 5.293 -73.669 88 2.5 142.7 167.7 0.24 b b 12/11/2007 09:09 5.459 -73.803 89 2.7 133.4 160.4 0.27 b b 17/11/2007 21:21 5.771 -74.26 102 2.9 62 119.4 0.39 b b 02/12/2007 11:11 4.696 -73.647 77.8 2.6 15.7 79.4 0.52 a b 17/12/2007 09:09 5.263 -73.673 85.7 3.1 150 172.8 0.28 b b 29/12/2007 06:06 5.265 -73.717 81.6 3.3 152 172.5 0.30 b b 11/01/2008 02:02 5.305 -73.739 82 2.3 146.3 167.7 0.22 b b 13/01/2008 04:04 5.244 -73.649 87 3 151.4 174.6 0.27 b b 14/03/2008 16:16 4.88 -74.821 55 3.8 4 55.1 1.05 a a 07/04/2008 04:04 5.345 -73.785 81 2.2 136.3 158.6 0.22 b b 13/04/2008 02:02 5.622 -73.941 95 2.2 110.2 145.5 0.24 b b 24/05/2008 19:19 4.422 -73.83 73 4.5 6.3 73.3 0.89 a a 24/05/2008 21:21 4.409 -73.762 80 3.4 2.9 80.1 0.66 a a 24/05/2008 19:19 4.395 -73.82 76 3.6 4 76.1 0.73 a b 24/05/2008 21:21 4.452 -73.766 77 2.9 0 77.0 0.60 a b 24/05/2008 17:17 4.453 -73.83 71 2.9 0 71.0 0.65 a b 24/05/2008 21:21 4.512 -73.657 84 2.5 22.8 87.0 0.46 b a 24/05/2008 23:23 4.58 -73.726 74 2.4 17.6 76.1 0.50 b b 24/05/2008 19:19 4.431 -73.797 76 2.3 4 76.1 0.48 b b 25/05/2008 11:11 4.452 -73.799 74 2.8 12 75.0 0.60 a a 25/05/2008 17:17 4.475 -73.729 79 2.6 15.9 80.6 0.51 a a 25/05/2008 00:00 4.483 -73.825 70 2.9 0 70.0 0.66 a b 25/05/2008 02:02 4.406 -73.698 86 2.5 0 86.0 0.46 b b 25/05/2008 06:06 4.439 -73.76 78 2.2 10.8 78.7 0.43 b b 26/05/2008 01:01 4.461 -73.775 75 2.3 3.4 75.1 0.48 b b 30/05/2008 19:19 4.439 -73.767 78 2.3 0 78.0 0.46 b b 02/06/2008 12:12 5.634 -74.382 86 2.9 89.6 124.2 0.37 b b 03/06/2008 02:02 4.446 -73.762 78 3.4 12.4 79.0 0.67 a a 07/06/2008 20:20 4.432 -73.83 73 2.6 0 73.0 0.57 a b 11/06/2008 18:18 5.399 -73.7 92 2.5 141.8 169.0 0.24 b b 02/07/2008 14:14 5.315 -73.747 82 2.2 149.4 170.4 0.20 b b 07/07/2008 05:05 4.429 -73.824 73 2.2 0 73.0 0.47 b b 09/08/2008 12:12 5.297 -73.741 82 2.7 142.5 164.4 0.26 b b 15/08/2008 01:01 5.311 -73.765 80 2.2 143.7 164.5 0.21 b b 02/09/2008 18:18 5.42 -73.757 89 2.3 139 165.1 0.22 b b 03/09/2008 02:02 5.517 -74.052 79 3.8 120.8 144.3 0.40 b b the ga’erqiong cu-au deposit, which sits on the north side of the coqên-xainzamagmatite belt, is a large-scale skarntype deposit, whose ore body has formed in the skarn zone in the contact part of quartz diorite and marble of duoai formation or the cracks of quartz diorite. its mineralization is closely related to quartz diorite. and granite porphyryrelated molybdenum ore still exists in its deep part. currently, there are disputes about the metallogenic dynamics background of this deposit. from previous studies, this paper carried out zircon la-lcpms u-pb dating and petrogeochemistry study for quartz diorite of ga’erqiong cu-au deposit. the testing result indicates: quartz diorite and granite porphyry were formed respectively in 88±2ma and 83±1ma, belonging to the magmatic activity of the early stage of upper cretaceous; quartz diorite and granite porphyry have geochemical characteristics similar to those of island arc rock of subduction zone and geochemical indexes similar to “adakite.” combining with the regional tectonic evolution, we think that quartz diorite and granite porphyry were all formed in the extension environment after the collision of lhasa block and qiangtang block. quartz diorite is the result of the migmatization of basic melt and acid melt evoked by asthenosphere material raise caused by lower crustal delamination; the formation of granite porphyry may be crust-mantle material’s partial melting results due to delaminated lower crustal. therefore, ga’erqiongskarn-type cu-au deposit belongs to the metallogenic response to the collisional orogeny in the closing process of meso-tethys. abstract keywords: tibet; zircon u-pb chronology; geochemistry; metallogenic dynamics; ga’erqiong cu-au deposit metallogenic dynamics background of ga’erqiong cu-au deposit in tibet, china origen de las dinámicas metalogénicas para el yacimiento de cobre y oro ga'erqiong en tibet, china issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n2.65192 earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017): 59 65 yuan ouyang1, 2, wunian yang*1, hanxiao huang2, hong liu2, jianlong zhang2, jianhua zhang2 1.key laboratory of geoscience spatial information technology, ministry of land and resources of the china, chengdu university of technology, chengdu, sichuan 610059, china 2. chengdu center, china geological survey, chengdu 610081, china *email of corresponding author: ywn@cdut.edu.cn el yacimiento de cobre y oro ga'erqiong, que se ubica en el lado norte del cinturón coqên-xainzamagmatite, es un depósito tipo skarn a gran escala cuyo cuerpo mineral se formó en la zona skarn, en la parte de contacto del cuarzo de diorita y mármol de la formación duoai y de las grietas de cuarzo de diorita. su mineralización está cercanamente relacionada a los cuarzos de diorita. la mena de molidbeno granítico relacionada a los pórfidos tiene presencia en estas zonas profundas. actualmente, se presentan varias discusiones sobre el origen de las dinámicas metalogénicas de este yacimiento. con base en trabajos previos, este estudio determinó la edad del circón uranioplomo con la técnica la-icpms y analizó la petrogeoquímica de cuarzos de diorita para el yacimiento ga'erqiong. los resultados del análisis indican que los cuarzos de diorita y los graníticos pórfidos se formaron en 88±2ma y 83±1ma, respectivamente, y pertenecen a la actividad magmática de la edad temprana del cretácico superior; los cuarzos de diorita y los graníticos pórfidos tienen características geoquímicas similares a aquellas de las rocas del arco insular en la zona de subducción e índice geoquímicos similares a la "adakita". en combinación con la evolución de la tectónica regional, se concluye que los cuarzos de diorita y los graníticos pórfidos se formaron en el ambiente extensivo tras la colisión de los bloques lhasa y qiantang. los cuarzos de diorita son el resultado de la migmatización de fundición básica y fundición ácida suscitada por el material elevado a la astenosfera gracias a un deslaminado menor de la corteza; la formación de los graníticos pórfidos podría ser el resultado de la fundición parcial de material en el manto de la corteza debido a un deslaminado menor en la corteza. además, el depósito ga'erqiong corresponde a la respuesta metalogénica de la orogénesis colisional en el proceso de cierre del mesotetis. resumen palabras clave: tibet; datación uranio-plomo; geoquímica; dinámicas metalogénicas; depósito de cobre y oro ga'erqiong. record manuscript received: 23/05/2017 accepted for publication: 30/06/2017 how to cite item ouyang, y., yang, w., huang, h., liu, h., zhang, j., & zhang, j. (2017). metallogenic dynamics background of ga’erqiong cuau deposit in tibet, china. earth sciences research journal, 21(2). 59-65 doi:http://dx.doi.org/10.15446/esrj.v21n1.63002 o r e d e p o si t s 60 yuan ouyang, wunian yang, hanxiao huang, hong liu, jianlong zhang, jianhua zhang 1. introduction bangong lake-nujiang river metallogenic belt includes not only ophiolite melange belt in suture zone but also the magmatite zone related to the tectonic evolution of meso tethys in the south and north side (geng, pan, wang, peng, & zhang, 2011). the porphyry cu-au deposits, like duobuza and bolong, skarntype cu-au molybdenum deposit like galale, tiangong nile andbaibuta and orogenic-type au deposits like dazha and shangxu discovered in recent years indicate that bangong lake-nujiang river is a giant polymetallic belt mainly of copper, gold, and iron. due to the limitation of the natural conditions of northern tibetan plateau, the prospecting degree of the discovered deposit of bangong lake-nujiang river metallogenic belt is relatively low in general (chen et al., 2013; tang et al., 2013; huang et al., 2012; huang et al., 2014; liu et al., 2015). only a few deposits reach the specific survey level and above. hence, deepening the understanding of the geologic feature and genetic mechanism of current deposits with relatively high working level plays an important enlightening role in determining the next range of reconnaissance and summarizing the regularity of ore formation of the metallogenic belt of bangong lake-nujiang river. ga’erqiong cu-au deposit is a large-scale skarn-type cu-au deposit which is located 20 km west to gyai county, ngari prefecture, tibet. chinese deposit scientists have conducted relatively detailed studies in aspects of the geologic feature, geochemistry of mineral deposits, metallogenetic chronology and diagenetic and metallogenic geodynamics of ga’erqiong cu-au deposit (tang et al., 2013; yao et al., 2013; lei et al., 2012). but consensus hasn’t been reached yet about the metallogenic dynamics background of ga’erqiong cu-au deposit. currently, there are two different viewpoints: first, the formation of ga’erqiong cu-au deposit is related to southward subduction of bangong lakenujiang river (lei et al., 2012; deng, tang, li, yao, & wang, 2011); and second, this deposit was formed in the extension environment after the collision of lhasa block and qiangtang block (tang et al., 2013; yao et al., 2013). given this, by detailed field investigations and previous studies, this paper conducted an analysis of chronology and geochemistry to granide which is closely related to mineralizing relation, deeply explored the cause of formation of this ore-related intermediateacid rock and analyzed metallogenic dynamics background of ga’erqiong cu-au deposit. our results provide a foundation for studies of the mineralization process and metallogenic prediction of ga’erqiong cu-au deposit. 2 metallogenic geological background bangong lake-nujiang river suture zone is an important tectonic line across central qinghai-tibet plateau. on the south and north sides are respectively gangdise block and qiangtang block (zhu et al., 2009). the suture zone is marked by interruptedly distributed ophiolite fragments, which has typical characteristics of accreting orogenic belt (pan et al., 2007). the main body of south qiangtang is jurassic marine basin. some sections of the southern rim have accreting mélange (geng et al., 2011); lhasa block may originate from the north side of australia. it has a distribution pattern of the old stratum in the middle and new stratum in two sides (zhu, zhao, & niu, 2001). the precambrian crystal old base in the middle has experienced multiphase metamorphism and presented unconformable contact with mesozoic volcanic, sedimentary layers. mesozoic and cenozoic marine and interactive marine and terrestrial stratum and volcanic-sedimentary stratum are the main stratum of south and north sides (geng et al., 2011). under the influence of tectonic evolution of meso and neo-tethys ocean, the main tectonic line of north and middle part of lhasa block runs nearly east-west. nearly east-west fracture apparently controlled the space distribution and form of magmatite. large-scale island arc volcanic activities happened in intermediate and late jurassic epoch, which belongs to a set of rock association dominated by andesite. the volcanic has geochemical characteristics of continental margin arc, whose formation is related to the south subduction of meso-tethys. the emplacement of intermediate-acidic rocks mainly happened during the period from early cretaceous epoch to the early stage of upper cretaceous, with stocks and batholiths in nearly east-west zonal distribution and the rock type transforming gradually from type i granite at the initial stage to s-type granite at the later stage (lv et al., 2011). therefore, according to the distribution feature of magmatite, considering shiquan river-lhaguo tso-yongzhu-namtso-lhari figure 1. schematic geological map of the ga’erqiong cu-au gold deposit 1quaternary; 2cinerite; 3volcanic breccia; 4volcanic agglomerate; 5limestone; 6marble; 7hornstone; 8skarn; 9-diorite prophyrite; 10-granodiorite; 11granodiorite porphyry; 12quartz diorite; 13granite porphyry; 14pegmatite; 15fault and inferred fault; 16ore body ga’erqiong cu-au deposit is tectonically located on the south rim of the west of north gangdise (fig. 1a), whose formation is related to limestone and marble of early cretaceous epoch duoai formation (k1d) and intermediateacid rock of upper cretaceous. it belongs to typical skarn-type deposit, but porphyry-type molybdenum ore bodies exist in deep part of the reservoir (tang et al., 2013). intermediate-acid rock and present stratum relation of remarkable intrusive contact. under the influence of emplacement, limestone and mudrock of duoai formation (k1d) are respectively marbleized and hornfelsed (zhang et al., 2013; li et al., 2011). copper (gold) ore body i and ii that have been found are controlled by both surrounding rock and intermediate-acid rock and are generated in the skarns near the contact zone of quartz diorites and marbles and the cracks of quartz diorite respectively. ore body i and ii are skarn-type.metallic minerals such as chalcopyrite, bornite and native gold, etc. were generated together with quartz stockwork, which indicates that the mineralization of copper and gold mainly happened in hydrothermal mineralization period (asis et al., 2017). ore body iii is distributed along structural fracture belt (f1) north to east, which belongs to ore body of iron, copper and gold under control of structural fracture belt and iron-oxide copper gold (iocg) ore body, whose formation may be related to the intrusion of aplite and granite porphyry and activities of north-east fault. also, skarn ore of ga’erqiong cu-au deposit contains rare intermetallic compounds of ni-cr-fe and cu-zn (xiao et al., 2012). fault zone as the boundary, lhasa block can be divided into north gangdise and middle gangdise, while the border of middle gangdise and the further south gangdise arc-ridge fault-uplift zone is ga’er-longgeer-zhari namco-cuomai fault zone (zhu et al., 2009; zhu et al., 2001). 61metallogenic dynamics background of ga’erqiong cu-au deposit in tibet, china 3 sample collection and petrology features the distribution of intrusive body in ga’erqiong cu-au deposit area is relatively extensive (fig. 1c). the intrusive body belongs to a set of intermediate-acid composite rock body, which mainly includes quartz diorite, granodiorite, quartz diorite, aplite, etc. and presents various forms such as stock, apophysis, and dike (tang et al., 2013). the ore body has distinct characteristics of multiperiodic magmatism. the intrusive contact relationship of rock masses in the field suggests that the formation of quartz diorite and granodiorite was relatively early, while quartz diorite and aplite were formed later. four pieces of quartz diorite and four pieces of granodiorite which are closely related to mineralization were taken as samples. the quartz diorite was taken from earth surface far away from tectonic alteration zone; the granodiorite was taken from the north of the mine. both of them are fresh and without alteration. quartz diorite: it is mainly distributed in the middle and south part of the mine. there are irregularly-shaped dark inclusions in the rock. boundaries of the listing and quartz diorite are clear with fine-grained rims and chilled borders. the rock is gray, off-white subhedral granular texture, and massive structure. it mainly consists of anorthose, hornblende, potassium feldspar, biotite, quartz and a few monoclinic pyroxenes and metallic minerals. anorthoses are mainly andesines with medium-fine grains and subhedralclintheriform and girdle structure, and the content is about 60%. hornblende is in the shape of short columns, which partly experienced metasomatism of biotite and chlorite and the content is about 15%. potassium feldspar has no or rare bicrystals, and the content is about 10%. the content of quartz is about 10%, whose fine grains are in irregular shape and filled among feldspars with an uneven distribution. biotite is about 5%. monoclinic pyroxene is relatively less and often in irregular form or short column (fig.2). granodiorite: it is light gray or off-white in fine-grained texture and massive structure. its primary minerals are anorthose, quartz, potassium feldspar, hornblende and a few iron pyrite and biotite, which are mostly subhedralanhedral crystals with a grain size of 0.3 to 3 mm. epidotization and chloritization can be seen in fractures. alteration inside the rock is uneven. figure 2. field photos and micrographs of granitoid in ga’erqiong area 4 analysis method and result 4.1 analysis method analysis of major element, rare earth element, and microelement is completed in southwest china supervision and inspection center of mineral resourced, ministry of land and resources (chengdu center, china geological survey, chengdu institute of geology and mineral resources). according to gb/t 14506.28-93 methods for chemical analysis of silicate rocks, the detected temperature is 18°c to 25°c, and the humidity is 35% to 65%. the major element is tested using axios-x fluorescence spectrophotometer, whose analytical precision is higher than 5%; microelement is tested by icpms method, using plasma mass spectrometry (x-series), whose precision is greater than 5%. the separation of zircon is finished in lang fang cheng xin geo-service co., ltd. backscattered electronic phase, image analysis of zircon cathodoluminescence and zircon u-pb dating are all conducted in china university of geosciences (wuhan) state key laboratory of geological process and mineral resources (gpmr). backscattered electronic phase and cathodoluminescence image analysis mainly aim at confirming zircon particle’s internal structure, determining the test point and avoiding cracks and inclusions. zircon u-pb isotope in situ analysis is conducted by use of agilent7500a inductively coupled plasma mass spectrometry of agilent company and geolas2005 excimer laser ablation on-line system of microlas company (ridzuan, zahar, & noor, 2017). helium is used as a carrier gas of ablation material. the spot diameter of laser beam is 32μm. revision is implemented with standard zircon 91500 used as the external standard and nist610 as the internal standard. data obtained from experiments is processed by icpmdatacal software. after common pb correction, isoplot 4.15 is used for the calculation of the weighted average age of zircon and the drawing of concordia diagram. for method and process, see document (ludwig, 2013). the obtained isotope ratio and age error are on the 1σ level. 4.2 geochemistry for sample analysis result of quartz diorite and granodiorite, see list 1. sio2 content of tested granide is 53.73% to 61.58%, among which the sio2 content of quartz diorite sample is between 53.73% and 56.16% (averagely 55.28%) and that of granodiorite is 60.48% to 61.58% (averagely 61.16%). the na2o content is 0.79% to 5.19% (averagely 3.18%), the k2o content is 2.08% to 2.92% (averagely 2.52%), and total alkali content w (na2o+k2o) is 7.13% to 7.84% with an average of 7.41% which is alkali-rich. the ratio of w (na2o/ k2o) is between 1.56 and 2.51 with an average of 1.98%. the rittman index (σ) is 1.80 to 4.49, which belongs to calc-alkaline rock and falls in high-k calcalkaline series rock area in sio2-k2o diagram (fig.3a); al2o3 content is between 15.10% and 17.39% with an average of 16.26%. aluminum saturation index a/ cnk is between 0.67 and 0.80 (averagely 0.76). a/nk is between 1.42 and 1.64 (averagely 1.51). in cipw standard mineral calculation, diopside (di) appeared, and corundum (c) didn’t, which belongs to quasi-aluminous series (fig.3b). the overall characteristics of samples suggest that granide in this area mainly belongs to quasi-aluminouscalc-alkaline granite series rock (rahim & usli, 2017). 62 yuan ouyang, wunian yang, hanxiao huang, hong liu, jianlong zhang, jianhua zhang the total amount of rare earth element of granodiorite and quartz diorite in this area is between 100 and 155×10-6 with an average of 127×10-6. the rare earth element content of quartz diorite is 100 to 115×10-6(averagely 105×10-6), which is obviously lower than that of granodiorite (σree: 140~155×10-6, averagely 150×10-6). lerr/herr is 5.8 to 1.1 (averagely 8.6) and (la/yb)n is 6.2 to 16.8 (averagely 11.4), which indicates that lree is enriched while hree is depleted, and the fractionation is clear. δeu is averagely 0.9 without obvious anomaly. δce is averagely 1.0, which suggests that the samples have no ce anomaly. the distribution mode pattern of rare earth of granodiorite and quartz diorite is in mild right deviation. the differentiation of two lithologies is evident. the rare earth distribution of quartz diorite is above that of granodiorite (fig. 4a). there is no obvious eu anomaly. tilt level of light rare earth’s curve of all samples is higher than that of the heavy rare earth, which indicates that fractionation degree of light rare earth element is remarkably greater than that of the heavy rare earth element, reflecting the clear differentiation of light and heavy rare earth during magmatism. figure 3. diagrams of sio2 -k2o and a/cnk-a/nk (a: base map from pecerillo & taylor, 1976; middlemost, 1985; b: base map from maniar, & piccoli, 1989.) on first mantle-normalized spider diagram (fig. 4b), the overall characteristics of granodiorite and quartz diorite of ga’erqiong area are similar. they are both “multi-peak” fluctuant curves inclining to the right. there is an obvious differentiation between high field –strength element (hfse) and large ion lithophile element (lile) (fig. 4b), which manifests as enriched in large ion lithophile element and depleted in high field -strength element, among which ta and nb are strongly depleted. figure 4. chondrite-normalized ree (a) and primitive-mantle-normalized trace element patterns (b) or the magmatic rocks in qingcaoshan region(chonodrite values from [21]; primitivem antle from[22] ) 4.3 zircon u-pb age zircons picked from granodiorite, and quartz diorite samples of ga’erqiong area have been tested. check points are all in parts with clear oscillatory zoning structure. for testing position and result, see fig. 5 and list 2. all tested zircons are transparent, colorless and with good crystalline form and typical magmatic zircon dense oscillation girdle. zircon th/u ratio (0.8 to 2.6) is within the magmatic zircon scope (normally metamorphic and hydrothermal origin zircon’s th/u<0.1, magmatic zircon th/u>0.1. zhao, 2010), which matches typical characteristics of magmatic zircon. therefore, zircon analyzed in this paper is magmatic zircon which crystallized with ga’erqiong granodiorite table.1 the major and trace elements characteristic values of granitoids in ga’erqiong area 63metallogenic dynamics background of ga’erqiong cu-au deposit in tibet, china at the same time. for young zircon, 206pb/238u age can provide more reliable test result (compston, williams, & kirschvink, 1992). the 206pb/238u surface ages of 14 analysis points of this test are between 87 and 89 ma, and the weighted average is 88±2ma (mswd=3.8) (fig.5), which represent the crystallization age of granodiorite. the weighted average u-pb age of granite porphyry zircon is 83±1ma, which is about 4 ma later than quartz diorite (yao et al., 2012). 5 results 5.1 petrogenesis geochemical compositions of granitic magmas without remarkable fractional crystallization can reflect the source rock characteristics of the magma. quartz diorite and granodiorite are both enriched in large ion lithophile elements. compared with depleted high-field-strength elements, they show notable characteristics of depleted nb and ta, which is similar to that of island arc magmatite in subduction zone; both of their mineral compositions are generally consistent with those of amphibole calc-alkaline granitoid (acg) under the mechanism of subduction (lei et. al, 2012). in the diagram for discriminating granite ta-yb tectonic setting (fig.6a), all the analyzed samples are classified into the scope of volcanic arc granite. these evidence all indicate that granite is formed in the island arc environment (qu, wang, xin, h. b., zhao, y. y., & fan, 2009; you & rahim, 2017). however, the lateral intraplate magmatism of the crust can also generate characteristics similar to those of island arc granite (plank, 2005). thus, only by combining the regional tectonic evolution can their tectonic settings of the formation be effectively discriminated. current research shows that the bangong lake-nujiang river section of meso-tethys started bidirectional subduction in late jurassic, and after collision and closure had occurred in middle stage of early cretaceous, it got into the intracontinental environment (zhu et al., 2009; wang, zeng, liu, xiao, & gao, 2013). the emplacement of quartz diorite and granite porphyry both happened in late cretaceous and notably after a collision between lhasa plate and qiangtang plate. this indicates that the ga’erqiong cu-au deposit is formed in bangong lake-nujiang river metallogenic belt crash environment, instead of island arc environment. figure 6. tectonic environment (a) and magma mixing trend diagrams (b) of the magmatic rocks in the gaerqiong area (a: base map from [30], b: base map from [31]) at the earliest, adakite refers to the intermediate-acidic igneous rock with unique geochemical characteristics, which is formed by young subducting oceanic crust melting under eclogite facies condition. its geochemical characteristics often manifest as containing sio2≥56%, high al content (al2o3≥15%), mgo content generally <3% (seldomly>6%), y lower than normal island arc (generally≤18×10-6) and hree (e.g. yb≤1.9×10-6), sr higher than normal island arc (seldomly<400×10-6), without eu anomaly or slight negative eu anomaly, and with low initial ratio of 87sr/86sr (<0.704) (defant & drummond, 1990). seeing from the geochemical characteristics, quartz diorite and granodiorite both have geochemical characteristics similar to those of adakite. the research suggests that formation of adakite has various reasons including partial melting of subducted oceanic crust, crystallization differentiation of mantle-derived basaltic magma, migmatization of magma, partial melting of delaminated lower crust and partial melting of thickened lower crust. that lhasa plate and qiangtang plate collided in middle stage of the early cretaceous, which caused shortening, thickening and rapid lifting of the crust, and the low k2o/na2o ratio formed by partial melting of ga’erqiong adakite and subducted oceanic crust does not conform to reality indicates that the ga’erqiong intermediate-acid rocks can’t be formed by partial melting of subducted oceanic crust. no same-period basalt is found in the mining area. the rocks are a lack of notable negative anomaly of sr and eu. in the diagram for feot~mgo (fig.6a), the sample also showed no evolution tendency of crystallization differentiation (you & rahim, 2017). all these indicate the ga’erqiong rock is not generated from crystallization differentiation of mantle-derived basaltic magma. in addition, the ga’erqiong adakite also can’t be the result of partial melting of the thickened lower crust, as ga’erqiong intermediate-acid and acid rocks notably contain more mg# than melt formed by partial melting of thickened lower crust or underplating basalt. there is higher content of cr and ni; zirconhf isotopes of the rocks show notable characteristics of mantle-derived or mantle-crust mixed source (yao et al., 2012). these indicate that there is mantle-derived material taking part in the diagenetic process. dark micro granular inclusions are generally considered as a product of halfway migmatization of mantle-derived basic magma, which is the most direct and powerful evidence for migmatization of magma. irregular dark micro granular inclusions are found in quartz diorite. the inclusions are dioritic type with distinctive boundary with its host rock and chilled border, which indicates the inclusions are not the residual after partial melting, but the product of magma crystallization. also, the zircon hf isotope system has very high closure temperature which won’t be changed along with partial melting or fractional crystallization. this can be used for identifying magma migmatization between different magma end members. zircon hf isotopes of quartz diorite have significant inhomogeneity (yao et al., 2012), which indicates the magma source of quartz diorite has occurred magma migmatization. in the sio2-k2o and sio2-la second variant diagram, quartz diorite showed both good linear variation trends. all these evidence show that, during formation of quartz diorite in ga’erqiong, the second end-member magma migmatization occurs. therefore, quartz diorite might be formed in the post-collision extension environment after the collision of lhasa block and qiangtang block; lower crust delamination figure 5. cathodolum inescence (cl) images of representative zircon (a) and concordia diagrams (b) for zircons form the magmatic rocks table 2. la-icp-ms zircon u-pb dating results in ga’erqiong region 64 yuan ouyang, wunian yang, hanxiao huang, hong liu, jianlong zhang, jianhua zhang caused uplifting of asthenosphere material, which further caused partial melting of lithospheric mantle and let basic melt mix with the acid melt. emplacement of granite porphyry is significantly later than quartz diorite, and there are no dark micro granular inclusions found in the rock, which implies that there might be different reasons for the formation of granite porphyry and quartz diorite. the granite porphyry is rich in k, which indicates it comes from thickened lower crust. the qinghai-tibet plateau started crust shortening and thickening in middle stage of the early cretaceous. thickening of the crust caused the formation of eclogite at the bottom of the lower crust. due to gravity differences, the eclogite-facies lower crust is easier to have delamination; what’s more, bimodal volcanic rocks appeared in nyixung region indicates that in the early stage of late cretaceous, the north-central area of lhasa block was in postcollision extension environment (qu, xin, xu, yang, & li, 2006). hence, partial melting of crust-mantle material caused by delamination of lower crust might be a major reason for the formation of ga’erqiong granite porphyry. 5.2 geodynamic background of mineralization there are still disputes about the geodynamic mechanism of the cretaceous volcanic rock in north and middle part of lhasa block. so far, the mainstream opinions include: it’s generated from neo-tethys crust’s northward subduction. it’s related to re-melting of thickened (delaminated) lower crust during or after collision of the qiangtang block and lhasa block. it’s related to southwards subduction of the meso-tethys oceanic crust and slab breaking off (zhu et al., 2009; harris, inger, & ronghua, 1990; coulon, maluski, & bollinger, 1986). in recent years, many late cretaceous magmatite are found in south part of the lhasa block. these geological phenomena can’t be explained by plate subduction model of the neo-tethys crust. thus, quartz diorite and granite porphyry in ga’erqiong can’t be the product of neo-tethys crust’s plate subduction. in the early stage of the early cretaceous, neo-tethys had occurred oceanic-ridge subduction (zhu et al., 2009). both quartz diorite and granite porphyry in ga’erqiong have characteristics similar to those of island arc magma, and they are probably the product of neotethys’ oceanic-ridge subduction. in the second place, bangong lake-nujiang river meso-tethys had been closed at that time, and the magmatism of crustmantle material caused by slab break-off during its southwards subduction also has the features of island arc magma. so, some scholars hold the opinion that late cretaceous magmatite in middle gangdese is the common outcome of the tectonic evolution of meso-tethys and neo-tethys (wang et al., 2013). as ga’erqiong is about 300km away from the yarlung-zangbo suture zone, it’s still lack of enough evidence so far to prove that the formation of intermediateacid rock in ga’erqiong is related to oceanic-ridge subduction of neo-tethys in early stage of late cretaceous. on the contrary, ga’erqiong is not so far from the bangong lake-nujiang river suture zone. therefore, we think the formation of intermediate-acid rock in ga’erqiong is more likely to be the magmatic response of meso-tethys’ tectonic evolution. judging from the geologic features, the formation of ore body i and ii in ga’erqiong cu-au deposit is closely related to magmatic emplacement of quartz diorite. there has been testing data to support the result of the field inspection. zircon u-pb age of the quartz diorite is 88±2ma, which is consistent with the re-os isotope age (86.89 ma) of molybdenite in skarn (li et al., 2011). it indicates that 1ma after the quartz diorite emplacement, the hydrothermal activity of magma formed the cu, au ore body. zircon u-pb age of the granite porphyry is 83±1ma, which is significantly later than that of quartz diorite, and its mineralization is likely to be related to the subsequent mineralization of porphyry molybdenum (yao et al., 2013). research suggests that, the bangong lake-nujiang river meso-tethys ocean basin opened in late triassic, and after ephemeral expansion, it started subduction in late jurassic and resulted in formation of shiquan river – gêrzê subduction zone and rutog subduction zone separately in the south and north, and cultivation of abundant island arc magmatite on both sides of the suture zones at the same time (qu et al., 2009). in the middle stage of early cretaceous (around 113ma), the lhasa block and qiangtang block had a collision, and after that, they entered into the arc-arc (continent) collisional evolution stage. the meso-tethys oceanic crust subducted southwards underneath the lhasa block. slab break-off caused upwelling of mantle-derived material and further resulted in upwelling of mantle material, which had mantle-derived magma underplating in different scales on both sides of the suture zones (zhu et al., 2009). after that, crust in north and middle part of lhasa block kept thickening under the compression of north-south stress and finished shift from intracontinental compression to a partial extension in the early stage of late cretaceous (rahim & usli, 2017; you & rahim, 2017). the thickened lower crust reached eclogite facies and started delamination, forming a series of magmation along with mineralization. it thus appears that the formation of ga’erqiong cu-au deposit is due to the metallogenic response of collision orogeny during the closing of meso-tethys. 6 conclusions (1) u-pb age of zircon from quartz diorite that is closely related to ga’erqiong cu-au deposit and mineralization is 88±2ma, and it’s product of magmatic activity in the early stage of late cretaceous. (2) quartz diorite and granite porphyry have geochemical characteristics similar to those of island arc granitoid in a subduction zone, and they are formed in the post-collision extension environment after the collision of lhasa block and qiangtang block. quartz diorite formed after asthenosphere material uplifting caused by lower crust delamination and is the mixture of basic melt formed by partial melting of lithospheric mantle and acid melt formed by partial re-melting of crustal materials. and granite porphyry is formed by partial melting of crustmantle material caused by lower crust delamination. (3) formation of ga’erqiong cu-au deposit is the metallogenic response during the collision of qiangtang block and lhasa block. acknowledgements: wunian yang was sponsored by the national natural science funds (no. 41071265 and no.41372340) and the specialized research fund for the doctoral program of higher education of china (no. 20105122110006) references: asis, j., tahir, s. h., rahim, a. r., konjing, z., kob, r. c., & tjia, h. d. (2017). smaller benthic foraminifera analysis of kudat formation, kudat, sabah: preliminary interpretation. geological behavior, 1(1), 27-29. boynton, w. v. (1984). geochemistry of the rare earth elements: meteorite studies. in: henderson, p. (ed.). rare earth element geochemistry. elsevier, 36(6), 323-338. chen, h. a., zhu, x. p., ma, d. f., huang, h. x., li, g. m., li y.b., ... liu, c. q. (2013). geochronology and geochemistry of the bolong porphyry cuau deposit, tibet and its mineralizing significance. acta geologica sinica, 87(10), 1593-1611. compston, w., williams, i. s., & kirschvink, j. l. (1992). zircon u-pb ages for the early cambrian time-scale. journal of the geological society, 149(2), 171-184. coulon, c., maluski, h., & bollinger, c. (1986). mesozoic and cenozoic volcanic rocks from central and southern tibet: 39 ar-40 ar dating, petrological characteristics and geodynamical significance. earth and planetary science letters, 79(3), 281-302. defant, m. j., & drummond, m. s. (1990). derivation of some modern arc magmas by melting of young subducted lithosphere. nature, 347, 662-665. deng, s. l., tang, j. x., li, z. j., yao, x. f., & wang, y. (2011). geochemical characteristics of rock mass in the gaerqiong cu-au deposit, tibet. journal of chengdu university of technology (natural science edition), 38 (1), 85-91. geng, q. r., pan, g. t., wang, l. q., peng, z. m., & zhang, z. (2011). tethyan evolution and metallogenic geological background of the bangong co-nujiang belt and the qiangtang massif in tibet. geological bulletin of china, 30(8), 1261-1274. 65metallogenic dynamics background of ga’erqiong cu-au deposit in tibet, china harris, n., inger, s., & ronghua, x. (1990). cretaceous plutonism in central tibet: an example of post-collision magmatism. journal of volcanology and geothermal research, 44(1), 21-32. huang, h. x., li, g. m., liu, b., dong, s. l., shi, h. z., zhang, z. l., & fan, a. h. (2012). zircon u-pb geochronology and geochemistry of the tiangongnile skarn-type cu-au deposit in zhongba county, tibet: their genetic and tectonic setting significance. acta geoscientica sinica, 33(4), 424-434. huang, h. x., li, g. m., liu, b., zhang, z. l., ma, d., qu, z., xiao, w. f., & liu, h. (2014). discovery of shangxu orogenic type gold deposit in northern tiber and its significance. mineral deposits, 33(3), 486-496. lei, c. y., li, z. j., zhang, z., hu, z. h., wang, h. x., & song, j. l. (2012). geochemical characteristics and geodynamic significance of the granites in the ga’erqiong cu-au deposit, tibet. acta geoscientica sinica, 33(4), 601-612. li, z. j., tang, j. x., yao, x. f., duo, j., liu, h. f., deng, s. l., ... hu, z. h. (2011). geological characteristics and prospecting potential of gaerqiong copper-gold polymetallic deposit in ali district, northern tibet. mineral deposits, 30(6), 1149-1153. liu, h., huang, h. x., li, g. m., xiao, w. f., zhang, z. l., liu, b., ... ma, d. (2015). factor analysis in geochemical survey of the shangxu gold deposit, northern tibet. geology in china, 42(4), 1126-1136. ludwig k. r. (2003). user's manual for isoplot 3.00: a geochronological toolkit for microsoft excel. berkeley geochronology center: special publication, 1-70. lv, l. n., cui, y. b., song, l., zhao, y. y., qu, x. m., & wang, j.p. (2011). geochemical characteristics and zircon laicp -ms u-pb dating of galae skarn gold (copper) deposit, tibet and its significance. earth science frontiers, 18 (5),224-242. maniar, p. d., & piccoli, p. m. (1989). tectonic discrimination of granitoids. geological society of america bulletin journal, 101, 635-643. middlemost, e. a. k. (1985). magmas and magmatic rocks. london: longman. pag. 266. pan, g. t., mo, x. x., hou, z. q., zhu, d. c., wang, l. q., li, g. m., ... liao, z. l. (2007). spatial-temporal framewoke of gangdese orogenic belt and its evolution. acta petrologica sinica, 22(3), 521-533. pearce, j. a., harris, n. b. w., & tindle, a. g. (1984). trace element discrimination diagrams for the tectonic interpretation of granitic rocks. journal of petrology, 25,956-983. peccerillo, r., & taylor, s. r. (1976). geochemistry of eocene calcalkaline volcanic rocks from the kastamonu area, northern turkey. contributions to mineralogy and petrology, 58, 63-81. plank, t. (2005). constraints from thorium/lanthanum on sediment recycling at subduction zones and the evolution of the continents. journal of petrology, 46(5), 921-944. qu, x. m., xin, h. b., xu, w. y., yang, z. s., & li, z. (2006). discovery and significance of copper-bearing bimodal rock series in coqin area of tibet. acta petrologica sinica, 22(3), 707-716. qu, x. m., wang, r. j., xin, h. b., zhao, y. y., & fan, x. t. (2009). geochronology and geochemistry of igneous rocks related to the subduction of the tethys oceanic plate along the bangong lake arc zone, the western tibetan plateau. geochimica, 38(6), 523-535. rahim, i. a., & usli, m. n. r. (2017). slope stability study around kampung kuala abai, kota belud, sabah, malaysia. malaysian journal of geoscience, 1(1), 38-42. ridzuan, a. a., zahar, u. a. u., & noor, n. a. m. (2017). association of evacuation dimensions towards risk perception of the malaysian students who studied at jakarta, medan, and acheh in indonesia. malaysian journal of geoscience, 1(1), 07-12. sun, s. s., & mchdonough, w. f. (1989). chemical and isotopic systematics of oceanic basalts: implication for mantle composition and processes. in: saunders, a. d. & norry, m. j., (eds.). magmatism in the ocean basins, geological society special pulication, 42, 303-345. tang, j. x., zhang, z., li, z. j., sun, y., yao, x. f., hu, z. h., ... he l. (2013). the metallogensis, deposit model and prospecting direction of the ga’erqiong-galale copper-gold ore field, tibet. acta geoscientica sinica, 34(4),385-394. wang, w., zeng, l. s., liu, j., xiao, p., & gao, l. (2013). the late cretaceous andesite shiying and determining the geochemical characteristics in cuoqin, tibet. chinese journal of geology, 48(2), 484-500. xiao, y. f., sun, y., wang, q., li, z. j., wang, y. l., zhang, s. m., ... he j. l. (2012). the discovery of rare intermetallic compounds (ni-crfe, cu-zn) in the garqiong copper-gold deposit of tibet. geology in china, 39(5), 1311-1317. yao, x. f., tang, j. x., li, z. j., deng, s. l., ding, s., hu, z. h., & zhang z. (2012). magma origin of two plutons from gaerqiong copper-gold deposit and it's geological significance, western bangonghu-nujiang metallogenic belt, tibet: implication from hf isotope characteristics. journal of jilin university(earth science edition), (s2): 188-197. yao, x. f., tang, j. x., li, z. j., deng, s. l., ding, s., hu, z. h., & zhang z. (2013). the redefinition of the ore-forming porphyrys age in ga’erqiong skarn-type gold—copper deposit, western bangong lake-nujiang river metallogenic belt, xizang(tibet). geological review, 59(1), 193-200. you, l. k., & rahim, i. a. (2017). application of gsi system for slope stability studies on selected slopes of the crocker formation in kota kinabalu area, sabah. geological behavior, 1(1), 10-12. zhang, z., tang, j. x., chen, y. c., li, z. j., song, j. l., yao, x. f., ... wang, h. x. (2013). skarn mineral characteristics of the gaerqiong cuau deposit in bangong co-nujiang river suture zone, tibet. acta petrologica et mineralogica, 32(3), 305-317. zhao, z. h. (2010). trace element geochemistry of accessory minerals and its applications in petrogenesis and metallogenesis. earth science frontiers, 17(1), 267-286. zhu, d. c., mo, x. x., zhao, z. d., niu, y. l., pan, g. t., wang, l. q., & liao z. l. (2009). permian and early cretaceous tectonomagmatism in southern tibet and tethyan evolution: new perspective. earth science frontiers, 16(2),001-020. zhu, d. c., zhao, z. d., & niu, y. l. (2001). the lhasa terrane: record of a microcontinent and its histories of drift and growth. earth and planetary science letters, 301(1), 241-255. zorpi, m. j., coulon, c., & orsini, j. b. (1991). hybridization between felsic and mafic magmas in calc-alkaline granitoids: a case study in northern sardinia, italy. chemical geology, 92, 45-86. using neuronal networks to determine site response from coda waves: application in armenia, colombia earth sciences research journal earth sci. res. j. vol 8, nº 1 (dec. 2004): 45 51 manuscript received april 2004 45 paper accepted august 2004 fine thermohaline structure of the colombian pacific ocean nancy villegas department of earth sciences, metheorology, universidad nacional de colombia bogotá, colombia e-mail: nlvillegasb@unal.edu.co abstract the present document shows strata classification of the colombian pacific ocean – columbian pacific ocean, done by first time according its fine thermohaline structure, based on temperature and salinity fields analysis. layers, where different mechanisms of fine structure predominate, were determined and everywhere in the area a stable stratification was observed, although conditions for not stability as a result of the double diffusion were present. key words: fine thermohaline structure, small-scale thermohaline structure, stability of the ocean, stratified ocean, salt finger stratification, differential-diffusion convection, and step structure resumen el presente documento muestra la clasificación de las capas de la cuenca del pacífico colombiano (cpc), hecha por primera vez de acuerdo con su estructura termohalina fina, basada en el análisis de los campos de temperatura y salinidad. las capas fueron determinadas con predominancia de diferentes mecanismos de estructura fina y se observó que en toda la región existe una estratificación estable, aunque se encontraron condiciones de inestabilidad como resultado de una doble difusión. palabras clave: estructura termohalina fina, estructura termohalina de baja escala, estabilidad en el océano, estratificación oceánica, estratificación de salinidad, convención de difusión diferencial, paso de estructura. © 2004 esrj unibiblos. nancy villegas 46 introduction on a fine structure scale, the profiles of hydrophysical fields show alternated sections with small and high gradient properties, the first one called quasi-homogeneous layers and the second one by layers. inside temperature (t) and salinity (s) profiles, sections with inverted properties distributions are frequently encountered. the fine structure is understood as a model of physical fields represented by a set of layers instead of homogeneous properties in strata ranging from 0.1 to 1 m, divided by even more thin layers with sharp gradients of temperature and salinity. the vertical gradients are 10 to 100 times higher and more, exceeding corresponding average gradients values (karlin, 1988). the fine stratification of water layer is testified by numerous measurements, done in different regions of the ocean with low-inertia sounding equipment, in particular ctdprobes. contrasting profiles obtained by standard hydrological instruments, these profiles contain many structural details, reproduced by repeated soundings and, therefore, which live long. the frequently curves of vertical distribution t and s take the form of correct steps or variables on the sign of deviations from the average profiles. there are fine-structure heterogeneities in the seasonal and main thermoclines on big depths and even in the upper quasi-homogeneous layer of the ocean (karlin et al., 1988). fedorov showed (fedorov, 1976) that below the upper quasi-homogeneous layer, a fine structure is formed in the majority of cases against the hydrostatic stability background on the vertical distribution density. on these conditions, two mechanism of fine structure formation are possible. the first one requires kinetic energy from an external source, used to increase the potential energy in the liquid layer. the mechanisms present are shift instability of currents, internal waves, etc. the second mechanisms do not require external energy sources. there, the structurization occurs by liberation of accessible potential energy, dissipated later as heat. the mechanism presents instability due to differential-diffusion convection or dual diffusion (zhubas and ozmidov, 1984, ozmidov, 1983). in shift instability of currents and internal waves, turbulence and elements of fine structure of hydrophysical fields arise under action of strong hydrodynamical instability in zones with high vertical gradients of speed current, like located areas observed as separate turbulent spots (benilov, 1985, maderich and nikishev, 1986). turbulent mixing regions are outlined on the vertical profiles of hydrophysical fields as the fine structure elements. the mixture process in the turbulent volume decreases temperature and salinity gradients inside it, being critical on its boundaries. spots formation due developed turbulence may also occur because instability or breaking of internal waves (fedorov, 1976, fedorov, 1991, konyaev and sabinin, 1992). this mechanism is observed when the richardson´s number achieve a value below some critical value, being the probability of formation of spots of turbulence at instability higher than at breaking of internal waves (ohotnikov and panteleis, 1985). the sizes of vertical instable areas of internal waves vary from decimeters to meters. the differential-diffusion instability is due to diffusion of two or more components of hydrodynamic system at different speeds. such instability forms a fine structure of temperature and salinity fields in water with a wide circulation at the ocean. because differences in the molecular thermal conductivity coefficients and salt diffusion in the system, the convective instability appears, producing intensive mixing and forming quasi-homogeneous layer divided by layers with sharp gradients properties (turner, 1985). a special attention is paid to the mechanism of differential-diffusion instability, it allows accelerated heat and salts transfers without energy from external sources (karlin et. al, 1988, fedorov et. al, 1986, karlin, 1988). this mechanism is observed, when temperature and salinity gradients produce opposite contributions to the vertical density gradient. here, the potential energy associated to vertical stratification is used, contributing to destabilize the density gradient. liberation of instability energy is due to differences in coefficients of molecular exchange of heat and salts. differential-diffusion instability is manifested in the regime of salt fingers and layered or diffuse convection. methodology initial natural data the calculation of stability of layers in waters of colombian pacific ocean (villegas, 2001) was made according to the fields of temperature and salinity, obtained in the scientific expedition during may 2000 (otero y pineda, 2000). the vertical distribution of water stability and the fine thermohaline structure of the colombian pacific ocean 47 vaisala-brunt’s frequency (villegas, 2002b, karlin and villegas, 2003) were made based on the example of 3 hydrometeorological stations obtained in this expedition: 14, 49 and 111 (figure 1). 84 8 3 82 81 80 79 78 77 1 2 3 4 5 6 7 8 9 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 n or th la tit ud e western longitude panama c olom bia ecuad or figure 1. hydrometeorological stations on the colombian pacific ocean. these stations, covering the 3 more interesting zones: coastal zone-station 14 (at 4°n/78°w), open sea-station 111 (at 3°n/84°w) and zone of mixing waters-station 49 (at 2°n/80°w), were selected on basis to the quasi-homogeneous zones on this study region (villegas, 2002a). diagnostic of possible fine structure activity in the ocean the temperature (et) and salinity (es) components of stability (e) are widely applied in the solution of different problems, including the typification of thermohaline conditions for stratification, diagnosing possible forms of mixing and processes of structurization, including the scales of fine structure (karlin et al, 1988; malinin, 1998). diagnostics of possible fine structure activity in the ocean is carried out usually along background ts-profiles with the aid of some parameters and criteria. one of them relates temperature and salinity components of stability, and also density relationship rp (karlin et al, 1988; filimonov, 1990, fedorov and pereskokov, 1991). based on these parameters and criteria, the diagnostics of possible fine-structure activity in colombian pacific ocean is represented below. according to data obtained on colombian pacific ocean during may 2000 (otero and pineda, 2000), and the help of the hesselberg sverdrup’s criterion (1) ([malinin, 1998; valerianova and zhukov, 1974; kamenkovich and monin, 1978), stability is calculated and contributions of temperature and salinity are located into general stability colombian pacific ocean (villegas, 2001). dz dt tdz dt tdz ds s e a ∂ ∂ + ∂ ∂ + ∂ ∂ = ρρρ (1) with: s∂ ∂ρ as density variation with a change in the salinity, kg/m3‰; dz ds as vertical gradient of salinity, ‰/m; t∂ ∂ρ as density gradient with change in temperature, kg/m3°cm; dz dt is the vertical gradient of temperature, °c/m; dz dta as vertical gradient of adiabatic temperature, °c/m; the temperature and salinity stability components, and the density relationship rp (2) (karlin et al., 1988, malinin, 1998, fedorov, 1991) have been used for carrying out of diagnostics of possible fine structure activity at ocean on background t-s profiles. s t e e dz ds dz d r −≈ ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ ⎟ ⎠ ⎞ ⎜ ⎝ ⎛ = β θ α ρ (2) with dz dθ as vertical gradient of potential temperature, °c/m; dz ds as vertical gradient of salinity, ‰/m; te as temperature component of stability, kg/m 4; se as salinity component of stability, kg/m 4. on the same data is calculated vaisala-brunt’s frequency (villegas, 2002b) according to the formula (malinin, 1998, konyaev and sabinin, 1992, kamenkovich and monin, 1978; valerianova and zhukov, 1974): nancy villegas 48 dz dg n ρ ρ = (3) where dz dρ is the vertical gradient of density. results the stability of layers in waters colombian pacific ocean n, according to temperature and salinity data of may 2000 expedition (villegas, 2001), showed a vertical distribution of stability and the vaisala-brunt’s frequency with a jump in the range 10 75 m. see station 14 (figures 2 and 3); station 49 (figures 4 and 5) and station 111 (figures 6 and 7) (villegas, 2002b; karlin and villegas, 2003). d ep th , m stability, кg/m4 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 -400 -350 -300 -250 -200 -150 -100 -50 0 e et es figure 2. vertical distribution of stability on colombian pacific ocean. may 2000, station 14 d ep th , m frequency, s-1 0.005 0.01 0.015 0.02 0.025 0.03 0.035 -400 -350 -300 -250 -200 -150 -100 -50 0 figure 3. vertical distribution of vaisala-brunt’s frequency on colombian pacific ocean. may 2000, station 14 besides that, colombian pacific ocean has a positive stability on all vertical and prevalence of temperature stability above salinity stability. the hydrological stations near the coast showed that, in a superficial layer, the stability in salinity prevailed over temperature stability. the contribution of salinity to stability decreases with removal from the coast. d ep th , m s tab ili ty, кg /m 4 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 -500 -450 -400 -350 -300 -250 -200 -150 -100 50 0 e et es figure 4. vertical distribution of stability on colombian pacific ocean. may 2000, station 49. fine thermohaline structure of the colombian pacific ocean 49 d ep th , m frequency, s-1 0.005 0.01 0.015 0.02 0.025 0.03 0.035 -500 -450 -400 -350 -300 -250 -200 -150 -100 -50 0 figure 5. vertical distribution of vaisala-brunt’s frequency on colombian pacific ocean. may 2000, station 49. vertical distribution vaisala-brunt’s frequency has similar behavior with vertical distribution of stability. vaisala-brunt’s frequency has values from 0.0025 to 0.036 s-1 which decrease with removal from coast. the greatest values took place in the center section of the coast, and minimum in the southeastern region colombian pacific ocean. d ep th , m stability, кg/m4 -0.1 0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 1.1 1.2 1.3 -400 -350 -300 -250 -200 -150 -100 -50 0 e et es figure 6. vertical distribution of stability on colombian pacific ocean. may 2000, station 111 d ep th , m f requency, s-1 0.005 0.01 0.015 0.02 0.025 0.03 0.035 -400 -350 -300 -250 -200 -150 -100 -50 0 figure 7. vertical distribution of vaisala-brunt’s frequency on colombian pacific ocean. may 2000, station 111 in 100 150 and 75 100 m layers, the greatest values of stability were located on the north along the coast and in the region from 78 to 80° w and from 5 to 7° n. in the 50 75 m layer an absolute stability on the north and southwest colombian pacific ocean is observed, while minimum values were observed around the gorgona island and southeast colombian pacific ocean. the layer 25-50 m shows absolute stability. the minimal values were located on the northeast and southwest colombian pacific ocean. in the 10-25 m layer negative values of stability are observed colombian pacific ocean . in general, in colombian pacific ocean area stable stratification is observed, but there are conditions for not stability as result of double diffusion (villegas, 2002b; villegas, 2003). table 1 shows the results of conducting diagnostics of the possible fine structure activity in colombian pacific ocean by the example of stations located on all researched area, with the help of vertical distribution of temperature and salinity, combinations of the contributions and to the general stability, and also criterion . in the given table four types of stratification are submitted as follows: te se ρr • full, or absolute stability (as): 0<∆t , 0>∆s , , , ; 0>te 0>se 0<ρr • salt fingers type (sf): , 0<∆t 0<∆s , , 0>te 0<se , ; 0>ρr nancy villegas 50 • type of layered convection (lc): 0>∆t , , , , ; 0>∆s 0<te 0>se 0>ρr • absolute instability (ai): , 0>∆t 0<∆s , , , . 0<te 0<se 0<ρr it is evident from table 1 that in colombian pacific ocean predominates the complete stability upper 150 m layer. below 200 m the salt the salt finger stratification is observed. in the layer 0-10 m only at separate stations 17, 42 and 114 absolute instability takes place. the type of salt finger stratification is at stations 49, 81 and 107, and the type of layered convection at the station 1. in the open colombian pacific ocean, in all layers, the alternation by stratification of the type pu and sp is evident. mixing here is irregular, i.e., in the form of spots, which forms steps in the ts-profiles. table 1. types of stability stratification of layers on columbian pacific ocean. may 2000 numbers of station and type of stability layers, m 1 5 14 17 29 33 42 43 47 49 79 81 107 111 114 0-10 lc as as ai as lc ai as as sf as sf sf as ai 10-25 as lc as as as as as as as as as as as as as 25-50 as as as as as as as as as as sf sf as as sf 50-75 as as as sf as as as as as as as as as as as 75-100 as as as as as sf sf as sf sf sf sf as sf as 100-150 as as as as as as as as sf as as sf sf as sf 150-200 as sf as sf sf sf as sf sf as sf as sf sf as 200-300 sf sf sf as sf sf sf sf sf sf sf sf sf sf sf 300-400 – sf sf sf sf sf sf sf sf sf – sf sf sf sf 400-500 – sf sf – – sf as sf – sf – as sf sf sf 500-600 – – – – – – – – – sf – – sf – sf conclusions layers with predominance of different mechanisms of fine structure were determined. in columbian pacific ocean, over the 150 meter layer, the total stability stratification predominates. below 200 m the salt fingers type stratification and the diffusion instability were observed on the southwest of columbian pacific ocean. in the vertical line on the open columbian pacific ocean the alternation of the total stability stratification type and the salt fingers stratification type were observed. the other remaining regions have stable stratification. in general, in columbian pacific ocean everywhere stable stratification is observed, but there are conditions for not stability as a result of the double diffusion. references benilov a.y. (1985) about the turbulent mechanism of formation to fine structure in the ocean //. izv as ussr. serie. fao. -. t.21, no. 9. p. 973-981. fedorov k.n. fine thermohaline structure of waters. l.: gidrometeoizdat. 1976. 184 p. fedorov k.n., and. pereskokov, a.i. (1991) typification of the thermohaline conditions of stratification in the world ocean: physical oceanology. l.: gidrometeoizdat. p. 127-135. fedorov k.n., pack v.t. et al. (1986) analysis of convective steps in the sea temperature inversion// izv. as ussr. serie. fao. t.22, no. 9 p. 969-978. fedorov, i.n. (1991) chosen transactions on physical oceanologyl.: gidrometeoizdat, p 309. filimonov, ag. (1990) typification of fine thermohaline structure in two hydrological sections in the subtropical part of the northeastern atlantic. hydrology of the southern ocean and north atlantic// sb.nauch.tr. l.: publ. lgmi, p. 145-149. kamenkovich, v.m., and monin, (1978) a.s. physics of ocean. vol. 1 hydrophysics of the ocean/ussr. m.: science, p 456. (ser. oceanology). fine thermohaline structure of the colombian pacific ocean 51 karlin l.n., kluykov y.y. and kutko, (1988) v.p. small-scale structure of hydrophysical fields on the upper layer of ocean. m.: gidrometeoizdat,. 164 p. karlin, l.n. and villegas, n.l. (2003) isolation of layers in the columbian pacific ocean (colombian pacific ocean) with different fine structure mechanisms of formation// materials of academic council session of on 27 28 january part.ii. sections of oceanology, ecology and physics. st. petersburg: rshu, c 26-28. karlin, l.n. (1988) processes of formation and interaction of the different-scale vertical thermohaline structure on the upper layer ocean and their numerical simulation. thesis to the competition of the scientific degree of dr. of phys. and math. l.: lgmi p. 289. konyaev, k.v. and sabinin, k.d. (1992) waves inside the ocean. st. petersburg: gidrometeoizdat, p 272. maderich, v.s. and nikishev, v.i. (1986) the diffusion-viscous stage of spreading variable spots in the stratified liquid// izv. as ussr. serie. fao. t. 22, no. 6. p 656-658. malinin, v.n. (1998) general oceanology, part of 1 physical processes. st. petersburg: publ. rshu, p 342. ohotnikov, i.n. and panteleis, n.a. (1985) the shearing instability of internal waves and vertical exchange in the ocean// sea hydrophysical periodical. no. 3. p 13-20. otero l., y pineda r. (2000) cruceros oceanográficos pacífico xxxii erfen xxx y pacífico xxxiii erfen xxxi 2000: reporte final.– tumaco.: cccp,.–200 p. ozmidov r.v. (1983) small-scale turbulence and the fine structure of hydrophysical fields on in the ocean// oceanology. 23, ed. 4. p. 533-537. turner, j.s. (1985) convection in multicomponent systems//natur wissenschaften. 72, no. 2. – p. 70–75. valerianova, n.a. and zhukov, l.a. (1974) practical works of general oceanology. l.: publ. lgmi. villegas, n. (2001) estructura vertical de las aguas en el océano pacífico colombiano: informe final.– tumaco.: centr. contr. cont. pac. villegas, n.l. (2002a) definition of uniform zones in the surface water of the columbian pacific ocean// materials of academic council session of on 30 31 january. st. petersburg: rshu,. – pp. 126-128. (a). villegas n. (2002 b)estabilidad de las aguas en el océano pacífico colombiano en mayo del 2000. boletín científico cccp.– tumaco.: centr. contr. cont. pp. 1-11 villegas n.l. laws governing the formation of the thermodynamic structure of the columbian pacific ocean. thesis to the competition of the scientific degree of dr. of phys. and math..: rshu. 2003. 189 p. zhubas, v.m. and ozmidov, r.v. (1984) stepped structures form of oceanological thermocline and the mechanisms of their generation// oceanology. 2a. ed., pp. 197-204. using neuronal networks to determine site response from coda waves: application in armenia, colombia earth sciences research journal earth sci. res. j. vol. 8, no. 1 (dec. 2004): 25-33 manuscript received july 2004 25 paper accepted october 2004 possible rapid strain accumulation rates near cali, colombia, determined from gps measurements (1996-2003) robert trenkamp1, héctor mora p.2, elkin salcedo h. 3 and james n. kellogg 1 1 university of south carolina, department of geological sciences 2 ingeominas, observatorio vulcanológico y sismológico, manizales, av. 12 de octubre 15 47 3 universidad del valle, facultad de humanidades abstract global positioning system (gps) data from southern central america and northwestern south america collected between 1991 and 1998 reveal wide plate margin deformation along a 1400 km length of the north andes. also associated with the oblique subduction of the nazca plate at the colombia-ecuador trench is the ‘escape’ of the north andes block (nab). the nab is delineated by the bocono-east andean fault systems and the dolores guayaquil megasheare to the east, the south caribbean deformed belt on the north and the colombia-ecuador trench and panama on the west. within the nab many damaging crustal earthquakes have occurred which is most recently exemplified on january 25, 1999 (mw = 6.1) armenia earthquake. preliminary analysis of recent occupations (2003 geored gps) of several previously observed (1996-2001) gps sites suggest shear strain accumulation rates in the cauca valley near cali of approximately 2.1 x 10-7 yr1 and 1.6 x 10-7 yr-1. these strain rates are measured within 2 delaunay triangles with common vertices at cali and restrepo, which encompass areas, located north and west of cali. seismicity has been monitored in the cauca valley for the last 17 years by the “observatorio sismológico del suroccidente” (osso) since 1987 and by the red sismológica nacional del ingeominas since 1993. their catalogs list numerous shallow earthquakes near cali but nothing larger than magnitude 5. historically, however, several large earthquakes are associated with the “falla cauca almaguer” in locations both to the south and north of cali in the cauca valley. preliminary calculations using the strain rates determined for these delaunay triangles and a simplified kostrov formula suggest possible decadal (30 – 90 years) recurrence intervals for mw = 6.0 – 6.3 earthquakes, centenary (90 – 900 years) recurrence intervals for mw = 6.4 – 6.9 earthquakes and millennial (900+ years) recurrence intervals for mw ≥ 7 earthquakes. key words: gps, kostrov formula, recurrence intervals, seismotectonics, strain rates resumen los datos del sistema de posición global registrado desde la zona meridional de centro américa y el noroccidente de sur américa, tomados en el periodo que comprende 1991 y 1998 revelan un amplio margen de deformación a lo largo de los 1400 kilómetros de longitud al norte de los andes. igualmente robert trenkamp et al. 26 asociado a la subducción oblicua de la placa de nazca en el corte de colombia y ecuador es el escape del bloque norandino (nab). el nab está delineado por el sistema andino de fallas bocono y el gran corte de dolores, guayaquil al este; el cinturón del caribe sur; el corte de colombia ecuador, al norte, y panamá, al oeste. dentro de esta zona han ocurrido muchos terremotos; como ejemplo está el más reciente, ocurrido el 25 de enero de 1999 en armenia (mw = 6.1). los análisis preliminares de recientes ocupaciones (2003 geored gps) de muchos gps previamente observados (1996-2001) sugieren que la tensión en la acumulación de los cortes en el valle del cauca, cerca de a cali, son de aproximadamente 2.1 x 10-7 yr-1 y 1.6 x 10-7 yr-1. estas velocidades de tensión son medidas dentro de los dos triángulos delaunay con vértices comunes en cali y restrepo, que comprometen áreas localizadas al norte y occidente de cali. desde 1987 la sismología del valle del cauca ha estado monitoreada por el observatorio sismológico del sur occidente (osso) y desde 1993 por la red sismológica nacional de ingeominas. se cataloga una lista de leves terremotos cerca a cali, pero ninguno mayor a una magnitud de 5. sin embargo históricamente muchos terremotos están asociados a la “falla cauca almaguer” en localidades cerca al sur y al norte de cali en el valle del cauca. cálculos preliminares realizados con velocidades de tensión determinadas por estos triángulos delaunay y la fórmula simplificada de kostroy que sugiere posibles intervalos recurrentes de algunas décadas (30 a 90 años) para terremotos mw = 6.0 – 6.3; intervalos centenarios (90 – 900 años) de terremotos mw. = 6.4 – 6.9 e intervalos milenarios (900 años o más) para terremotos de mw ≥ 7 palabras clave: formula kostrov, gps, intervalos de recurrencia, parámetros de tensión, sismotectónica, tasas de esfuerzo. © 2004 esrj -unibiblos. introduction the central and south america (casa) gps project was initiated in 1988 to study plate motions and crustal deformation in a tectonically active area of complex interaction among the nazca, cocos, caribbean and south american plates. data from the casa project collected between 1991 and 1998 reveal wide plate margin deformation along a 1400 km length of the north andes (trenkamp et al. 2002). associated with the oblique subduction of the nazca plate, the colombia-ecuador trench is the ‘escape’ of the north andes block (nab). the nab is delineated by the bocono-east andean fault systems and the dolores-guayaquil megasheare to the east, the south caribbean deformed belt to the north and the colombia-ecuador trench and panama on the west (figure 1). within the nab many damaging earthquakes have occurred. with this continuing threat of damaging earthquakes to major metropolitan centers, a large combined geological and geophysical study, microzonificacion sismica de la ciudad de santiago de cali, was proposed and performed by ingeominas (alvarado et al. 2003). in general, the project was a broad-based geological and geophysical approach toward understanding the stresses and deformation responsible for the neotectonics of the area. as a part of this study to assess the earthquake hazard potential near cali and areas adjacent to the cauca valley, a gps project, geodesia: red de estudios de deformacion 2003 (geored03) (mora and trenkamp, 2003) was coordinated and executed between july and september 2003 by the volcanological and seismological observatory of ingeominas in manizales. the objective of the field project was a reoccupation of a subset of casa stations and other previously occupied gps stations in the cauca region focusing on the city of cali. plate tectonic models for southwest colombia, based on regional mapping of the tertiary and quaternary deposits, show a series of graben shaped sub-basins located between the cauca patía and romeral fault systems. palinspastic reconstruction shows restraining bend, releasing bend and pull apart basins containing several tectonic blocks (valle and eduardo, 1999). possible rapid strain accumulation rates near cali, colombia determined from gps measurements (1996-2003) 27 figure 1. generalized tectonic map of northwestern south america and southern central america including the fault slip planes for the 1906, 1942, 1958 and 1979 earthquakes. slip planes are from kanamori and mcnally (1982). a gravity survey completed in 1964 shows two anomalies of interest on the west side of the upper cauca basin, the vijes and cali anomalies. these two anomalies are bounded on the west by the cauca fault and on the east by the candelaria fault and separated by transverse faults. seismic lines/sections confirm the gravity interpretations and show that the anomalies are not intrusions but are uplifted basement blocks (noel, 1996). also obvious from the seismic lines is an upper crust broken by numerous faults. seismic sources that affect cali and the surrounding cauca region are diverse. during the 20th century a four (4) large sequence and great subduction related earthquakes occurred. the first and largest event occurred in 1906 (mw=8.8) and ruptured a 500 km length of the subduction interface on the colombia-ecuador trench between manta, ecuador and buenaventura, colombia (kelleher, 1972, kanamori and mcnally, 1982). kanamori and mcnally (1982) report that three smaller events in 1942 (mw= 7.9), 1958 (mw=7.8) and 1979 (mw=8.2) re-ruptured most of the thrust fault plate boundary segment that ruptured during the 1906 event (figure 1). these earthquakes are shallow (< 50 km) on the subduction interface. other deeper earthquakes also occur that seem to be associated with deeper segments of the subduction interface. a third source of earthquakes is intraplate and occurs on faults in the shallow crust. these earthquakes have been most recently exemplified by the january 25, 1999 (mw=6.1), armenia earthquake and pose the greatest threat of damage to the populated areas of the cauca valley and environments. the seismicity in the cauca valley has been monitored for the past 17 years by the observatorio seismológico del suroccidente (osso) since 1987 and also by red sismológica nacional de ingeominas since 1993. their catalogs list numerous shallow earthquakes near cali but nothing larger than magnitude 5. page (1986) reports faulted alluvium in the vicinity of palmira, a city near cali, which suggests shallow seismicity during the quaternary and historically several large earthquakes are associated with the “almaguer fault” in locations to the north and south of cali. data and data analysis casa data in colombia consists of hundreds of sites, which have at least one epoch measurement. of these sites and sites previously occupied by ingeominas, a subset of 36 sites was chosen for the larger geological and geophysical project looking at zones of interest within and surrounding the cauca valley and the city of cali. data was collected using 3 trimble 4000 ssi receivers and dorne margolin choke ring antennas. site occupations consisted of a minimum of three 8-hour observation days and were continuous over the 3 days when security conditions permitted (table 1). all of the gps data presented in this analysis and the larger geored03 study was processed using the jpl/nasa developed gipsy/oasis-ii (gps inferred positioning system/orbit analysis and simulation software) software (release 5) (zumberge et al. 1997). loosely constrained solutions were obtained using jpl's fiducially free trenkamp robert, mora p. héctor salcedo h. elkin, kellogg james n 28 orbit (a ''non-fiducially'' orbit is one that is estimated without significant a-priori site position constraints), which were then transformed into the international terrestrial reference frame 2000 (itrf2000). in order to transform loosely constrained solutions into the itrf2000 reference frame, daily 7 parameter transformations were determined using all reference sites (igs tracking stations) used in our daily solutions and which are contained in the itrf2000 position and velocity model. these are then applied to the daily solution, which transforms them to the desired reference frame (itrf2000). horizontal and vertical velocities were determined using a combination of all 352 daily solutions and 174 stations from the 1994, 1996, 1998, 1999, 2001 and 2003 surveys. a least squares inversion was used to estimate site velocities and position at an arbitrary epoch from the daily itrf2000 coordinates weighted by the full covariance matrix of the coordinates. a scaling factor of 7.2 was applied to the input covariances so the reduced chi-square statistic equals 1.0. reduced chi-square statistics that equal one indicate that the formal errors agree with the scatter of the measurements. the scaling by 7.2 is equivalent to scaling the sigmas by 2.68. variance scaling assumes a gaussian (normal) error distribution, removal of all human errors (blunders) from the dataset and systematic underestimation of the true errors by gps software, which is generally accepted. table 1. the east, north and vertical velocities with one sigma errors relative to itrf2000 of the 30 stations with multiple year observations made during the project geored03. six sites were established and observed for the first time. these sites were aqu2, café, fres, letr, uval and vers and will be reoccupied in the near future in order to add their vectors to the growing sw colombia gps database. on. lat. e n u σe σn σu site deg deg mm mm mm mm mm mm 287.12 5.55 0.0 12.6 4.8 3.2 1.5 6.4 aqui 282.61 6.20 7.3 12.3 1.1 0.7 0.4 1.3 bhsl 285.92 4.64 1.3 12.1 -35.3 0.5 0.4 0.6 bogt 283.01 3.82 3.9 14.2 1.5 0.9 0.5 1.8 buen 284.04 4.75 9.0 14.0 -7.8 1.3 0.7 2.7 cago 283.64 3.50 4.0 13.9 -3.4 0.7 0.4 1.2 cali 284.33 5.92 6.2 13.3 -3.1 2.0 0.7 3.1 frda 284.42 6.26 2.2 14.3 3.0 1.4 0.7 2.7 mede 283.40 0.98 2.1 9.2 4.7 8.7 2.3 8.2 moco 285.11 5.20 1.0 15.5 1.9 1.7 0.8 3.5 mqta 284.53 5.03 5.0 14.2 -4.2 1.1 0.6 1.9 mzal 284.57 6.18 7.4 14.1 -0.8 1.8 0.7 2.7 negr 284.70 2.94 2.1 13.2 -0.3 1.4 0.6 2.5 neiv 284.87 4.47 1.4 14.3 0.2 1.5 0.6 2.7 ombl 284.54 5.53 5.7 15.2 -2.1 1.3 0.6 2.6 paco 282.74 1.22 1.5 12.2 -0.5 0.7 0.5 1.6 past 284.27 4.82 3.7 14.7 15.1 1.5 0.7 2.7 pere 286.63 3.27 -2.9 10.0 -0.3 1.6 0.6 2.7 plle 283.42 2.48 1.7 13.5 13.2 0.8 0.5 1.8 ppyn 283.58 3.22 1.2 12.8 2.7 1.4 0.7 2.7 ptej 283.46 3.81 1.3 14.0 29.8 1.3 0.7 3.0 rest 284.57 6.18 6.4 14.1 -0.5 0.7 0.5 1.3 rion 283.85 4.40 2.4 12.8 -1.3 1.8 0.8 3.5 roln 284.39 4.87 3.8 16.6 -11.9 1.6 0.7 2.7 srdc 283.78 4.09 5.6 13.1 1.2 1.5 0.7 3.1 tul2 281.25 1.81 13.3 9.8 6.4 0.8 0.5 1.7 tuma 286.62 4.07 -3.7 9.5 -1.0 0.8 0.5 1.7 vill 284.40 4.98 1.4 15.2 -9.1 2.8 1.0 4.0 chin 284.70 2.94 2.1 13.2 -0.3 1.4 0.6 2.5 neiv 284.34 4.56 2.5 13.9 -15.6 5.0 1.5 6.1 uniq trenkamp robert, et al. velocities, in this study, are determined relative to the itrf2000 reference frame (figures 2 & 3). due to an assumption that each daily solution is uncorrelated (white noise error model) uncertainties in the velocity estimates may be underestimated even though individual position uncertainties have been scaled to match the observed scatter. recent analysis (zhang et al. 1997; mao et al. 1999; dong et al. 2002) have shown that a better description of noise in a continuous gps time series can be obtained by incorporating both white noise and either ''flicker'' noise or ''random walk'' noise (both time related noise processes) into the error model. however, since every continuous gps site will have its own ''unique noise spectrum'' it is not certain how best to apply these results to campaign style occupations which are too infrequent for noise characteristic determination. it is also not clear what noise model and weighting between available noise model should be applied to relative position time series over baselines a few hundred kilometers in length (freymueller et al. 2000) much less baselines which are less than 100 kilometers. figure 2. vector field for sites observed during geored03. a few sites near pere are not shown to maintain clarity of the figure. one site chip was added. it was re-observed at the end of the project because it was near the ingeominas office and had been previously observed in 1999 after the armenia earthquake. the chip/mzal vectors and the rion/negr are interesting examples of the consistency of gps measurements. (see table 2 for the rion/negr vectors magnitudes). overlain on the vector field is the 31 block grid used for the project microzonificacion sismica de la ciudad de santiago de cali and a small sampling of the faults (light gray lines) in and around the cauca valley. results for this study, only four stations are analyzed which are the four nearest to the city of cali and which are also in a position favorable for the creation of delaunay triangles. these stations are tul2, rest, cali and ptej (figure 3). for these sites casa data collected prior to 1996 was removed from the strain analysis to minimize the effect of viscoelastic overprinting of the geodetic signal due to suggested ongoing viscoelastic effects from the 1979 (mw=8.2) earthquake as reported in white et al. (2003). in order to solve for the velocity gradient tensor components ),,,( 21 ωθεε &&& and to estimate the shear strain (γ& ) (table 2), the loosely constrained itrf 2000 daily solutions for all 174 stations were input as quasi observations into the quasi observations combination analysis (qoca) software developed at jpl (dong et al. 1998). the qoca software permits forward and backward filtering of the daily solutions and tightening and loosening of geodetic constraints on selected stations. this is done to achieve 29 trenkamp robert, mora p. héctor salcedo h. elkin, kellogg james n 30 consistency in results after all constraints are applied to the dataset. figure 3. map showing the topography of the region; the stations used in the strain analysis and the delaunay triangles for which the strain was calculated. for our purposes the data were processed forward and backward with tight and loose constraints on a subset of the global sites until the solutions were consistent. at all times the stations used in the strain analysis were completely unconstrained so that the results forthcoming were completely determined by the data and not artificially determined by tight constraints. at this step, the data prior to 1996 was removed from the input data to qoca for the four stations used in the strain analysis and the strain rates were determined. two separate data runs using both forward and backward filtering were processed and the average strain rates were calculated from a combination of these separate runs (table 2) the simplified kostrov formula was used to determine the time for release of accumulated strain for different moment energy release events these calculations were iterated using seismic moments for various synthetic moment magnitude events and the calculated shear strains until natural breaks (decadal, centenary and millennial) were discovered in the release time, see figure 4. table 2. separate processing results of the runs using the qoca software. listed is the eigenvalue parameterization ),,,( 21 ωθεε &&& along with theγ& evaluation and the average strain rates used in the kostrov recurrence rate calculations. delaunay vértices 1ε& 2ε& ω& θ γ& 10-8strain yr1 10-8 strain yr1 10-8strain yr1 degrees 10-8 rad yr-1 cali-resttul2 15.7 ± 5.60 -0.98 ± 1.20 -0.43 ± 1.80 11.13 ± 6.88 16.74 ± 6.88 cali-resttul2 14.8 ± 5.62 -0.99 ± 1.19 -0.49 ± 1.80 10.46 ± 7.22 15.84 ± 6.88 average used 16.29 ± 6.88 cali-restptej 24.5 ± 8.71 2.80 ± 1.41 -0.66 ± 2.69 -6.45 ± 6.59 21.65 ± 10.24 cali-restptej 23.2 ± 8.71 2.64 ± 1.41 -0.89 ± 2.69 -7.45 ± 6.94 20.59 ± 10.23 average used 21.12 ± 10.24 these strain rates were then input into a simplified kostrov formula γµ &v m o 2 ≅τ where mo is the moment energy release in dynecm, µ is the shear modulus taken to be 3x1011 dyne/cm2, t is time, v is the volume being strained in cm3 and γ& is the shear strain rate. possible rapid strain accumulation rates near cali, colombia determined from gps measurements (1996-2003) 31 figure 4. results of the recurrence rate calculations using the simplified kostrov formula. decadal recurrence intervals (30-90 years) include earthquakes of magnitude (mw = 6.0 – 6.3), centenary (90 -900years (mw = 6.4 -6.9) and millennial (900+ years) (mw ≥ 7.0+) discussion instantaneous velocity gradients within the continental lithosphere in wide plate margin deformation zones such as the nab are surely driven by plate tectonic forces. with the inception of space based geodesy, especially the global positioning system, the possibility of determining increasingly dense, instantaneous, three dimensional surface velocity fields has become a reality. the density and precision of these measurements has made possible the determination of strain rates which can be calculated to a certain degree of accuracy and the denser the measurements, the more accurately the straining areas can be mapped. pollitz (2003) points out that there is a distinction between instantaneous velocity and steady state velocity in that instantaneous velocity is typically represented by geodetic measurements obtained over short time periods while steady state velocity is best described by fault slip rates over geologic time. and so, if the instantaneous surface velocity measurements represent the interseismic velocity field and not the steady state velocity field, this implies that the procedure of relating geodetically determined velocity fields directly to rates of seismic moment release and, subsequently, moment recurrence rates is not altogether correct. in fact, if, as determined in pollitz (2003), the instantaneous velocities are potentially underdetermined steady state velocities the recurrence rates determined must be considered as maximal. the strain rates determined in this study for these two delaunay triangles near cali (table 2), while not high compared to subduction zone rates (10-5 – 10-6) (bilham et al. 1989), are consistent with the seismicity that has been observed in this region of the nab. the energy release rates were determined using conservative volumes, which increase the recurrence intervals but are believed to be valid for 1an area with a thin sedimentary cover similar to this section of the cauca valley. these data are, however, very preliminary and all of the vectors except the vector at cali are based on 2 epoch measurements. although two measurements are enough for obtaining a vector, are certainly not very robust vectors and all of the sites have measurements, which are within the effective viscoelastic window reported in white et al. (2003) and will also have an affect on the results. the historical record only goes back approximately 450 years and if the release is to be greater than mw=6.6 then it may be an event to be trenkamp robert, mora p. héctor salcedo h. elkin, kellogg james n 32 expected in the near future. this would seem unusual or unlikely in the area since the largest recent events near this area have been nearer the mw = 6.1 range. however recent paleoseismic work near the pereira-armenia region produced evidence of faulted offsets that may have generated an mw=6.6 for a ne-sw trending fault and ‘at least’ mw=6.9 for an e-w trending fault (lalinde et al. 2003). maximum seismic magnitudes had been believed to be between 6.2 and 6.5, but this paleoseismic evidence brings a better understanding of the seismicity of the area over the last 30,000 years. in other words, this work and the work of many other scientific investigators indicate a need to study further the geodynamics of the area and apply the findings to the seismic risk in a more informed manner. this area is an area in need of site densification and multiple epoch observations in order to improve the strain analysis and focus the strain search to a localized set of faults. this information would then aid other researchers in the search for paleoseismic study locations, multiple front geologic and geophysical projects and possible permanent gps site installation and observation. conclusions the cauca valley of colombia is a seismically active region of high potential risk to large population centers. a myriad of faults are being tectonically stressed which have repeatedly, in the past, slipped with devastating results. unfortunately, the number of observable points throughout the valley and surrounding areas and resources for expanding the observational database are, at present, small but the need for a minimum 100 station gps network throughout the valley surrounding the population centers of cali, armenia, pereira and medellín along with the two bounding cordilleras, occidental and central, is obvious. our results while not conclusive and rife with difficulties are important in context with several other studies and historical documentation of seismic activity, which has affected cali and other large population, centers in the region. acknowledgments the casa project was supported by nsf grants ear-8617485, ear-8904657 and nasa. the gps measurements were made with the assistance from ingeominas and the instituto geografico agustin codazzi in colombia. the project microzonificacion sismica de ciudad santiago de cali was supported by funding from dagma and performed by ingeominas. special consideration and gratitude to the facultad de ingeniería of the universidad de manizales for use of office space and sun workstation for data processing and to the andean geophysical laboratory at the university of south carolina for the use of two trimble ssi receivers and choke ring antennas. figures 1-3 were created using the generic mapping tool (gmt) (wessel and smith, 1995). references alvarado, c., et al. (2003) projecto microzonificación sismica de santiago de cali, reporte preliminar, ingeominas, dagma. 2003. bilham, r., yeats, r.s. and zerbini, s. (1989) space geodesy and global forecast of earthquakes, eos, transactions, american geophysical union, 70, 65, 73. dong, d., herring, t.a. and king, r.w. (1998) estimating regional deformation from a combination of space and terrestrial geodetic data, journal of geodesy, 72 (4), 200-214. dong, d., fang,p. bock,y. cheng, m.k. and miyazaki. s. (2002) anatomy of apparent seasonal variations from gps-derived site position time series. journal of geophysical research, b, solid earth and planets, 107 (4), 13. freymueller, j.t., cohen, s.c. and fletcher, h.j. (2000) spatial variations in present-day deformation, kenai peninsula, alaska, and their implications, journal of geophysical research, b, solid earth and planets, 105 (4), 8079-8101, 2000. kanamori, h., and mcnally, k.c.(1982) variable rupture mode of the subduction zone along the ecuador-colombia coast. bulletin of the seismological society of america, 72 (4), 12411253. kelleher, j.a. (1972) rupture zones of large south american earthquakes and some predictions, journal of geophysical research, 77 (11), 20872103. kostrov, v.v., (1974) seismic moment and energy of earthquakes, and seismic flow of rocks, lzv. acad. sci. ussr phys. solid earth, 1, eng trans l., 23-44. lalinde pulido, c.p., toro v., g.e. lópez, m.c. velásquez, a. and audemard, f.a. (2003) paleoseismic evidence at taller san miguel, pereira-armenia region, colombia south america, gsa conference, reno, nevada. mao, a., harrison, g.a. and dixon, t.h. (1999) noise in gps coordinate time series. journal of geophysical research, b, solid earth and planets, 104 (2), 2797-2816,.mora, h. y trenkamp, r., investigaciones geodesicas satelitales, projecto microzonificación sísmica de santiago de cali, reporte preliminar, ingeominas, 2003. noel, j.a.,(1996) landsat analysis of the upper cauca basin, colombia, annual meeting abstracts – american association of petroleum geologists and possible rapid strain accumulation rates near cali, colombia determined from gps measurements (1996-2003) 33 society of economic paleontologists and mineralogists, 5, p.106. page, w. geología sísmica y sismicidad del nw colombiano. interconexión eléctrica s.a. medellín colombia. pollitz, f.f., (2003) the relationship between the instantaneous velocity field and the rate of moment release in the lithosphere, geophysical journal international, 153, 595-608. trenkamp, r., kellogg, j.n. freymueller, j.t. and h. mora p., (2002) wide plate margin deformation, southern central america and northwestern south america, casa gps observations, journal of south american earth sciences, 15, 157-171. valle, p., and eduardo, l. (1999) evolución de subcuencas cenozoicas en el valle-cauca-patia; evolution of cenozoic basins at valle-cauca-patia, colombia, boletín de geología (bucaramanga), 21, 27-36. wessel, p., smith, w.h.t., (1995) new versión of the generic mapping tools released. eos trans. am. geophys. u. 76, 329. white, s.m., trenkamp, r. and kellogg, j.n. (2003) recent crustal deformation and the earthquake cycle along the ecuador-colombia subduction zone, earth and planetary science letters, 216, 231-242. zhang, j., bock, y. johnson, h. fang, p. williams, s. genrich, j. wdowinski, s. and behr, j. (1997) southern california permanent gps geodetic array; error analysis of daily position estimates and site velocities, journal of geophysical research, b, solid earth and planets, 102 (8), 18,035-18,055. zumberge, j.f., heflin, m.b. jefferson,d.c. watkins, m.m. and webb, f.h. (1997) precise point positioning for the efficient and robust analysis of gps data from large networks, journal of geophysical research, b, solid earth and planets, 102 (3), 5005-5017, hydrogeochemical studies were carried out in and around udayagiri area of andhra pradesh in order to assess the chemistry of the groundwater and to identify the dominant hydrogeochemical processes and mechanisms responsible for the evolution of the chemical composition of the groundwater. descriptive statistics, correlation matrices, principal component analysis (pca), together with cluster analysis (ca) were used to gain an understanding of the hydrogeochemical processes in the study area. pca has identified 4 main processes influencing the groundwater chemistry viz., mineral precipitation and dissolution, seawater intrusion, cation exchange, and carbonate balance. further, three clusters c1, c2 and c3 were obtained. samples from c1 contain high level of cl− and may be due to the intensive evaporation and contamination from landfill leachate. most of the samples from c2 are located closer to the sea and the high level of na++k+ in these samples may be attributed to seawater intrusion. the geochemistry of water samples in c3 are more likely to originate from rock weathering. this has been supported by gibbs diagram. the groundwater geochemistry in the study area is mostly of natural origin, but is influenced to some degree by human activity. se realizaron estudios hidrogeoquímicos en y alrededor del área udayagiri de andhra pradesh para evaluar la química del agua subterránea e identificar los procesos hidrogeoquímicos dominantes y los mecanismos responsables de la evolución en la composición química del agua subterránea. se utilizaron estadísticas descriptivas, matrices de correlación, análisis de componentes principales, al igual que análisis de grupos, para obtener y entender los procesos hidrogeoquímicos en el área de estudio. los análisis de componentes principales identificaron cuatro procesos determinantes que influenciaron la química del agua subterránea, estos son, la precipitación y disolución de minerales, l intrusión de agua marina, el intercambio de cationes y el equilibrio de carbonatos. de esta forma se obtuvieron tres grupos, c1, c2, y c3. las muestras del grupo c1 contienen un alto nivel de cl-, lo que podría deberse a la intensa evaporación y contaminación de los lixiviados de rellenos sanitarios. muchas de las muestras del grupo c2 se ubican cerca del mar y el alto nivel de na++k+ podría atriburise a la intrusión de agua marina. la geoquímica de las muestras de agua en el grupo c3 probablemente se desprende de la meteorización de rocas. el diagrama de gibbs valida estos resultados. la geoquímica del agua subterránea en el área de estudio es principalmente de origen natural pero también está influenciado por la actividad humana en algún grado. earth sciences research journal earth sci. res. j. vol. 20, no. 4 (december, 2016): e1-e7 abstract resumen keywords: hydrogeochemistry, multivariate analysis, pca, ca, udayagiri palabras clave: hidrogeoquímica, análisis multivariado, análisis de componentes principales, análisis de grupo, udayagiri. assessment of groundwater quality of udayagiri area, nellore district, andhra pradesh, south india using multivariate statistical techniques evaluación de la calidad del agua subterránea a través de técnicas estadísticas multivariadas en el área udayagiri, distrito nellore, andhra pradesh, en el sur de india record manuscript received: 04/12/2015 accepted for publication: 18/11/2016 how to cite item nagaraju, a., thejaswi a. and sreedhar y (2016). assessment of groundwater quality of udayagiri area, nellore district, andhra pradesh, south india using multivariate statistical techniques. earth sciences research journal, 20(4), e1-e7 http://dx.doi.org/10.15446/esrj.v20n4.54555 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n4.54555 h y d r o g e o l o g y arveti nagaraju, arveti thejaswi 2 and yenamala sreedhar 1 1sri venkateswara university, india. arveti9@gmail.com 2department of environmental sciences, sdlcku, kakatiya university, warangal – 506 009, a. p. india e2 arveti nagaraju, arveti thejaswi and yenamala sreedhar 1. introduction water is a solvent and dissolves minerals from the rocks with which it comes in contact. groundwater may contain dissolved minerals and gases that give the taste being enjoyed by many people. however, over 97% of the water on earth is salty and most of the remaining 3% is frozen in the polar ice caps (bhattacharya et al. 2012). the atmosphere, river, lakes, and underground stores hold less than 1% of all the fresh water needed to support the earth’s population (venkateswara rao, 2011). the investigation and assessment of groundwater chemistry and quality is a hot topic worldwide, triggering international attention from both researchers and policy makers due to its relevant link with environmental objectives and regional development. the quality of water is of vital concern for mankind since it is directly linked with human health, protection of the environment, plant growth and sustainable development.groundwater is one of the earth’s most important resources for human life. the quality of groundwater depends on various chemical constituents and their concentration, which are mostly derived from the geological data of the particular region (nagaraju et al. 2006; suresh et al. 2007; gupta and saharan, 2009; tripathi et al. 2012; gnanachandrasamy, et al.2015; nagaraju et al. 2015). groundwater chemistry has been extensively studied, and a great deal of literature can be found in various scientific journals and databases. geochemical analysis is an approach for defining groundwater baseline concentrations using geochemical and statistical tools (edmunds et al. 2003). it is of most importance for thorough examination of spatial and temporal groundwater chemistry in aquifer while interpreting other possible contributing sources and processes of groundwater. earlier workers have applied different methodologies to assess variations of groundwater geochemistry (panagopoulos et al. 2012). amongst these methods are multivariate statistical analysis are important (cloutier et al. 2008; razmkhah et al. 2010; nagaraju et al. 2016a, b, c). the multivariate statistical analysis such as principal component analysis (pca) and hierarchical cluster analysis (hca) is an approach to classify groundwater samples and make correlation between variables (hussain et al. 2008). pca attempts to identify underlying variables, or factors that explain the pattern of correlations within a set of observed variables. hca helps to identify relatively homogeneous groups of cases (or variables) based on the selected characteristics by an algorithm that starts with each case (or variable) in a separate cluster and combines clusters until only one is left (massart and kaufman, 1983). this technique can be used to group the commonly collected water quality data, in which each cluster indicates water with particular quality (belkhiri et al. 2010; nagaraju et al. 2014a, b, c). in the present study, principal component analysis (pca) and cluster analysis (ca) techniques have been applied to study the water quality characteristics of udayagiri area. to identify the rock–water interaction processes and groundwater redox conditions. all the hydrogeochemical studies mentioned in this study will show that multivariate statistical analyses significantly help to classify groundwater and identify major mechanisms influencing groundwater chemistry. hence, the main objective of this study is to enumerate the usage of factor score in identification of the hydrogeochemically active regimes represented by the major factors. relationship among dissolved ions present in water may vary with seasons and different lithologies. this study also illustrates the usefulness of statistical analysis to improve the understanding of groundwater systems. study area location and climate the study area is located between longitudes 79°17'00" and 79°26'30" e and latitudes 14°51'00" and 15°00'00" n (figure 1). the climate of the area enjoys a temperature ranging from the highest mean maximum of 44°c in april to lowest mean maximum of 32°c in january. this area receives major rainfall from north-east monsoon season starting from october and end by decemberjanuary months. this period forms the main cropping season receiving 66.7% of the annual rainfall through ne monsoon, while drought conditions generally prevail during south-west monsoon season with roughly 33.3% of the annual rainfall. farmers take up cropping if monsoon rainfall is good in kharif season (rainy season in india, which lasts between april and october). figure 1. geological and hydrogeological map of study area with water sample location geology the study area is underlain by variety of geological formations comprising from the oldest archaeans to recent alluvium (figure 1). the formations include mainly garnetiferous amphibolites and pelitic schist of nellore schist belt, granitic gneiss and cuddapahs (quartzites and shales) of pre-cambrian period. among these consolidated formations, schists (amphibolites and pelitic schists) and gneisses occupy most of the study area. hydrogeology. hydrogeologically this area is classified as consolidated (hard), semiconsolidated (soft) and unconsolidated (soft) formations. groundwater occurs in all most all geological formations and its potential depends upon the nature of geological formations, geographical setup, incidence of rainfall, recharge and other hydrogeological characters of the aquifer. among the consolidated formations gneisses are relatively good aquifers. schistose formations also form potential aquifers when the wells tapping contact zones with intrusives. quartzites and shales of cuddapah group are of little significance from the groundwater point of view as they are restricted to the hilly terrain. groundwater is developed in these formations by dug wells, dug cum bore wells and bore wells tapping weathered and fractured zones. the yield of the dug wells are in the range e3assessment of groundwater quality of udayagiri area, nellore district, andhra pradesh, south india using multivariate statistical techniques of 15 to 35 m3/day. the depth of dug wells ranges from 5 to 12 m, while the depth of filter point wells generally varies from 6 to 11 m. the yields generally range in this aquifer between 500 to 1000m3/day.hh materials and methods sample collection and analysis a total of 30 groundwater samples were collected from udayagiri area during september 2014. sampling locations were recorded using a potable gps device and they were shown in figure 1. the water samples were collected from bore wells after 10 minutes of pumping and transferred into precleaned polyethylene bottles. electrical conductivity and ph were measured in the field immediately after sampling, while the determination of major cations (na+, k+, ca2+, mg2+ and anions (f–, cl–,so42– hco3 -) were carried out in the laboratory using the standard analytical procedures recommended by apha (2005).minimum, maximum, average values and se values are presented in table 1. table 1. minimum, maximum, average and standard deviation values of different constituents of water samples sd: standard deviation; se: standard error multivariate techniques multivariate techniques such as principal component analysis (pca) and hierarchical cluster analysis (hca) were useful in analyzing hydrochemical data (wu et al. 2014). in this study, the ibm® spss v.20 software was used to analyze the dimensional groundwater quality data of the samples including ec, ph, tds, ca2+, mg2+, na+, k+, hco3 −, co3 2−, f−, so4 2−, cl−, hardness, alkalinity. pca can determine the principal factors corresponding to both natural and anthropogenic processes that may have significant influences on the chemical composition of groundwater, while hca can determine the number of groups that the hydrochemical data can be classified (voutsis et al. 2015). two types of hierarchical cluster analysis (hca) are usually used: q mode hca and r mode hca. q mode hca can be used to highlight the spatial relationships among the sample points, while r mode hca is usually applied to analyse the parameters by clustering them into groups or facies based on their similarity (wu et al. 2014). in this study, q mode hca was adopted. before clustering, raw data was standardized using the z score approach. a combination of the squared euclidean as a similarity measure and the ward’s agglomeration scheme to link clusters was determined to yield optimal results in the hca. determining the number of principal components is important in pca. it can be determined either by the total variance for which the total variance explained by the principal components should be higher than 85% or by the kaiser criterion for which only the components with eigenvalues greater than one can be retained(cloutier et al. 2006; wu et al. 2014). in this study, the kaiser criterion was adopted to determine the number of components. results and discussion from the geochemical analyses of groundwater samples as depicted in table 1, the ph is in range of 6.98 to 7.70 indicates its suitability for irrigation.the ec is ranging from 100 to 272 µmhos/cm. ca is varying from 14 to 144 mg/l and mg is ranging from 33 to 122 mg/l; na is between 31 to 312 mg/l and k is varying from 2 to 56 mg/l. the fluoride concentration is ranging from 0.09 to 1.05 mg/l. the total dissolved solids lie between 65 to 177 mg/l, in which most of the samples are within desirable limit. the concentration of sulphate varies between 11 to 120 mg/l. the hco3 and co3 concentrations in groundwater range from 168 to 827 mg/l and 39 to 181 mg/l, respectively. the chloride content varies from 28 to 228 mg/l/. the alkalinity varies from 567 to 1771 mg/l and the hardness varies from 148 to 540 mg/l. the abundance of the major ions in groundwater is in following order: na >mg >ca >k > and cl> hco3 − > co3 2− > so4 2− multivariate statistical analyses multivariate statistical analyses are useful tool solving multivariate problems (wu et al. 2014). in this study, pearson correlation analysis was first conducted to reveal the hidden relationship between each pair of the physiochemical parameters, and then principal component analysis (pca) and hierarchical cluster analysis (hca) were carried out in spss 20 to analyse the hydrogeochemical data for a better understanding of the origin and evolution of groundwater chemistry. pearson correlation analysis correlation analysis is useful in understanding the chemical reactions occurring in groundwater system (wu et al. 2015). in this study, pearson correlation coefficients were calculated using spss 20 and were shown in table 2. ec is positively correlation with tds (r=1.000, p<0.01), because both ec and tds represent the salinity of groundwater due to dissolved ions. furthermore, these two parameters are positively correlated with major ions except f− and co3 2−, indicating that the dissolved ions are the major factors influencing the values of ec and tds. both ca2+ and mg2+ show significantly negative correlations with ph (r=-0.518, p<0.01 and r=-0.668, p<0.01, respectively), which suggests that low ph value favours the occurrence of ca2+ and mg2+. considering the significantly positive correlation between ca2+ and mg2+, it is reasonable to conclude that the dissolution of dolomite is a possible origin of ca2+ and mg2+ (r1), as these two minerals are prevalent in shale and the quaternary formations. the positive correlation between mg2+ and hco3 − is an additional evidence of this conclusion. (r1) co32−is negatively correlated withca 2+ and hco3− (r=-0.403, p<0.05 and r=-0.419, p<0.05, respectively), which can be explained as follows: in a balanced co3 2−−hco3− system, the addition of ca 2+ in this system will incur the decrease of hco3− (the reverse reaction of r1), which will further promote the reaction of r2, resulting in the decrease of co3 2−. (r2) e4 arveti nagaraju, arveti thejaswi and yenamala sreedhar table 3. total variance explained by each pc and the loading matrix of pcs ** correlation is significant at the 0.01 level (2-tailed). * correlation is significant at the 0.05 level (2-tailed). hardness shows significantly positive correlations with ca2+ and mg2+ (r=0.823, p<0.01, and r=1.000, p<0.01, respectively), because hardness is an approximate measure of ca2+ and mg2+ (li et al. 2013). cl− and na+ usually show a positive correlation if they originate merely from the weathering of halite r3 (li et al. 2011, 2014a). but some other reactions such as cation exchange and weathering of silicate may change their relationship (li et al. 2014b). in the present study, the cation exchange and sea water intrusion may be reasonable explanations to this. as the study area is located near the sea, sea water intrusion will alter the na-ca balance in the fresh water system, promoting cation exchange (r4). the negative correlation between ca2+ and na+ (r=0.361, p<0.05) can be indicative of the cation exchange. r(3) r(4) correlation analysis is helpful to provide a preliminary understanding of the processes controlling groundwater chemistry. however, it should be noted that these reactions induced by correlation analysis are with great uncertainty (li et al. 2011). geological and hydrogeological conditions must be taken fully into consideration and other techniques must be applied to verify its results. principal component analysis according to the kaiser criterion, the first four principal components (pcs) which represent 84.531 % of the total variance in the hydrochemistry were extracted (table 3). the component loadings which represent the importance of the variables for the components (values greater than ±0.532 are in bold) (table 4). table 2. pearson correlation coefficients of physiochemical parameters e5assessment of groundwater quality of udayagiri area, nellore district, andhra pradesh, south india using multivariate statistical techniques pc1 explains the greatest of the variance (41.654%) and has significantly positive loadings by cl−, tds, ec, so 4 2−, hardness, mg2+, ca2+, and a negative loading by ph (table 3), which may represent the primary impacts from the dissolution/precipitation processes of minerals such as dolomite, calcite, silicates and gypsum. considering the origin of cl−and so4 2-, as well as the location of samples, it is reasonable to guess that intensive evaporation is also responsible for this component. therefore, pc1 represents a mixed origin of these parameters from natural processes including mineral weathering and groundwater evaporations, and it is the major factors influencing the characteristics of groundwater chemistry. pc2 explains 23.209% of the variance and has positive loadings in na+, alkalinity and k+, which is indicative of cation exchange triggered by salt water intrusion to a great degree. this is also an important process regulating the evolution of groundwater chemistry (wu et al. 2014, 2015). pc3 and pc4 explain 12.127 and 7.541% of the variance, respectively, which shows that they are relatively more local or specific effects than pc1 and pc2. pc3 is negatively correlated with co3 2− and has a positive loading in f−, indicating the dissolution of fluorite that introduces more f− into the groundwater but constrains the dissolution of carbonate minerals such as dolomite and figure 2. calculated illustration of factor scores of each sample calcite. pc4 has negative loading in hco3 −, which may be explained by the transformation between hco3 − and co3 2− (carbonate balance). factor scores of each sample have been calculated and illustrated in figure 2. from this diagram it is clear that samples v1, v2 and m3 are governed by factors represented by pc1, pc2 and pc3, i.e. influenced by natural water-rock interactions, seawater intrusion and some local fluorite dissolution. m3 is also influenced by carbonate balance represented by pc4. samples v3, v4 and v5 are influenced by pc2 and pc3, indicating the impacts from seawater intrusion and local fluorite dissolution. samples g8, g9, g10, m6, m7, m8 and m10 have positive loading on pc3 and pc4, indicating that these samples are governed mainly by local fluorite dissolution and carbonate balance. m2 is governed by pc1 and pc2, while v8 is influenced by pc2 and pc4. actually, most of the samples are associated with pc1, pc2 and/or pc3, indicating that natural processes such as mineral weathering and ion exchange are responsible for the groundwater chemsitry in the study area, although pc1 represents some effects of local groundwater evaporation. therefore, the results from correlation analysis can be regarded logical evidenced by pca analysis. table 4. mean concentrations of parameters from different clusters hierarchical cluster analysis (hca) figure 3 shows the dendrogram of q mode hca. three distinct clusters can be divided by hierarchical cluster analysis as shown in (figure 3).c1 includes 11 samples (m4, m5, m6, m7, m8, m9, m10, v9, v10, g2 and g3), c2 and c3 with a lower linkage distance, indicating that they have greater similarity.c2 includes 8 samples (v1, v2, v3, v4, v5, v8, m2 and m3) and c3 consists of 11 samples (g1, g4, g5, g6, g7, g8, g9, g10, m1, v6 and v7). the mean values of geochemical and physical data were presented in table 4 to describe the characteristics of each cluster of samples. the stiff diagrams (figure 3) and the data (table 4) indicate that the three clusters are geochemically independent groups.c1 is represented by high concentrations of hco3 −+co3 2− and cl− for anion and mg2+ for cation, c2 is characterized by high levels of hco3 −+co3 2− for anion and high concentrations of na++k+ and mg2+ for cation, and c3 is characterized by high concentrations of hco3−+co3 2− for anion and mg2+ for cation (figure 3). c1, indicated by table 3, is characterized by highest salinity and hardness and medium f− and alkalinity among the three clusters, while c3 possesses the lowest salinity, f− and alkalinity, and medium hardness, and c2 has medium salinity and hardness, highest alkalinity and f−. e6 arveti nagaraju, arveti thejaswi and yenamala sreedhar figure 3. dendrogram of q mode hca for water quality parameters the distinct geochemical characteristics of the clusters indicate that they have undergone geochemical evolution of different types and intensities. samples from the three clusters are all characterized by high concentration of hco3 −+co3 2−, but samples in c2 have more na++k+ than those from c1 and c3. considering that most of the samples in c2 are located closer to the sea than those from c1 and c3, it is possible that seawater intrusion may be a significant factor impacting the groundwater chemistry. the geochemistry of water samples in c3 are more likely to originate from rock weathering, but rock weathering is of different intensities for different samples in this cluster. samples in c1 are also characterized by high cl−, which may be due to intensive evaporation. mechanism controlling the groundwater geochemistry the gibbs diagram can evaluate the hydrochemistry of groundwater in the study area. mechanism controlling groundwater geochemistry a reaction between groundwater and aquifer minerals has a significant role in water quality which is useful to understand the genesis of water (gibbs 1970; subramani et al. 2009; vasanthavigar et al. 2012). the chemical data of water samples of the area are plotted in gibbs diagrams (figure 4).in the present study, the gibbs ratio 1 values vary from 0.11 to 0.48 and gibbs ratio 2 values ranges from 0.34 to 0.94. the majority of the samples irrespective of the formation are falling in the rock weathering region. this is due to the chemical weathering with the dissolution with rock forming minerals. figure 4. mechanism controlling the chemistry of groundwater (after gibbs 1970) conclusions the gibbs diagram revealed that the hydrochemistry of groundwater falls in the rock weathering region and is due to dissolution with rock forming minerals. pearson correlation analysis reveals that natural processes such as mineral dissolution/precipitation and cation exchange are dominant factors influencing the groundwater chemistry. four components are extracted by the pca. pc1 which explains the greatest of the variance represent a mixed influence of the dissolution/precipitation processes and salt water intrusion on the groundwater chemistry.pc2 is indicative of cation exchange triggered by salt water intrusion. pc3 and pc4 indicate relatively more local or specific effects such as the dissolution of fluorite and carbonate balance. three distinct clusters (c1-c3) are obtained through cluster analysis. samples from c1 contain high level of cl− that may be due to intensive evaporation and contamination from landfill leachate. most of the samples from c2 are located closer to the sea and the high level of na++k+ in these samples may be attributed to seawater intrusion. the geochemistry of water samples in c3 are more likely to originate from rock weathering. references apha (2005). standard methods for the examination of water and wastewater, 21st edn. american public health association, dc. bhattacharya, t., chakraborty, s. and tuck neha. (2012). physico chemical characterization of ground water of anand district, gujarat, india, i. res. j. environment sci., 1(1): 28-33. belkhiri, l, boudoukha, a, mouni, l, baouz, t (2010) application of multivariate statistical methods and inverse geochemical modeling for characterization of groundwater-a case study: ain azel plain (algeria). geoderma, 159: 390-398. cloutier v, lefebvre r, savard mm, bourque é, therrien r (2006) hydrogeochemistry and groundwater origin of the basses– laurentides sedimentary rock aquifer system, st. lawrence lowlands, québec, canada. hydrogeol j, 14:573–590 cloutier, v., lefebvre, r., therrien, r., savard, m.m. (2008).multivariate statistical analysis of geochemical data as indicative of the hydrogeochemical evolution of groundwater in a sedimentary rock aquifer system. j hydrol, 353:294–313. edmunds, w.m., shand, p., hart, p., ward, r.s. (2003). the natural (baseline) quality of groundwater: a uk pilot study. sci total environ, 310: 25-35 gibbs, rj. (1970). mechanisms controlling world water chemistry, science, 17: 1088–1090. gnanachandrasamy,g., ramkumar,t., venkatramanan, s., vasudevan, s., chung,s. y and bagyaraj, m.(2015). accessing groundwater quality in lower part of nagapattinam district, southern india: using hydrogeochemistry and gis interpolation techniques. applied water science, 5(1): 39-55. gupta, d . p, saharan, s. j. p., (2009). physiochemical analysis of ground water of selected area of kaithal city (haryana) india. researcher, 1(2): 1 5. hussain, m., ahmed, sm., abderrahman, w. (2008). cluster analysis and quality assessment of logged water at an irrigation project, eastern saudi arabia. j environ manag, 86: 297-307. li, p. y., qian, h., wu, j. h. (2011). hydrochemical characteristics and evolution laws of drinking groundwater in pengyang county, ningxia, northwest china. e-journal of chemistry, 8(2): 565-575. doi: 10.1155/2011/472085. li, p., qian, h., wu, j., zhang, y., zhang, h. (2013). major ion chemistry of shallow groundwater in the dongsheng coalfield, ordos basin, china. mine water environ, 32:195–206. doi: 10.1007/s10230-013-0234-8. li, p., wu, j., qian, h. (2014a). hydrogeochemistry and quality assessment of shallow groundwater in the southern part of the yellow river alluvial plain (zhongwei section), northwest china. earth sci res j, 18(1):27−38. li, p., wu, j., qian, h., lyu, x., liu, h. (2014b). origin and assessment of groundwater pollution and associated health risk: a case study in an industrial park, northwest china. environ geochem health, 36:693–712. doi:10.1007/s10653-013-9590-3. e7assessment of groundwater quality of udayagiri area, nellore district, andhra pradesh, south india using multivariate statistical techniques massart, d.l., kaufman, l. (1983). the interpretation of analytical chemical data by the use of cluster analysis. wiley, new york. nagaraju, a., suresh, s., killham, k. and hudson-edwards, k. (2006). hydrogeochemistry of waters of mangampeta a barite mining area, cuddapah basin, andhra pradesh, india. turkish j. eng. env. sci., 30: 203-219. nagaraju, a., sunil kumar, k., thejaswi, a., sharifi, z. (2014 a). statistical analysis of the hydrogeochemical evolution of groundwater in the rangampeta area, chittoor district, andhra pradesh, south india. american journal of water resources, 2(3): 63-70. nagaraju, a., sreedhar, y., sunil kumar, k., thejaswi, a.,sharifi, z. (2014 b). assessment of groundwater quality and evolution of hydrochemical facies around tummalapalle area, cuddapah district, andhra pradesh, south india. jour. env. analytical chemistry,1:1-6. nagaraju, a., sunil kumar, k. and thejaswi, a. (2014c). assessment of groundwater quality for irrigation: a case study from bandalamottu lead mining area, guntur district, andhra pradesh, south india. appl. water science, 4:385-396. nagaraju, a., .balaji, e., thejaswi, a. and sun, l. (2015). quality evaluation of groundwater in mulakalacheruvu area, chittoor district, andhra pradesh, south india based on hydrogeochemistry. fresenius env. bull., 24: 4496-4503. nagaraju, a., thejaswi, a., and sun, l. (2016 a). statistical analysis of high fluoride groundwater hydrochemistry in southern india: quality assessment and implications for source of fluoride. environ. engineering science, 33: 471-477. nagaraju, a., sharifi, z. and balaji, e . (2016b). statistical and analytical evaluation of groundwater quality of tirupati area, chittoor district, andhra pradesh, south india. jour. geol. society of india, 88: 222-234. nagaraju, a., balaji, e, and thejaswi, a. (2016c). hydrogeochemical evaluation of groundwater in certain parts of vinukonda area, guntur district, andhra pradesh, south india. fresenius env. bull., 25: 1519-1532. panagopoulos a., kassapi k.a., arampatzis g., perleros b., drakopoulou s., tziritis e., chrysafi a.a. and vrouhakis i. (2012) assessment of chemical and quantitative status of groundwater systems in pinios hydrological basin greece. proc. int. conf. protection and restoration of the environment xi, thessaloniki, pp. 511-517. razmkhah, h, abrishamchi, a, torkian, a. (2010). evaluation of spatial and temporal variation in water quality by pattern recognition techniques: a case study on jajrood river (tehran, iran). j environ manag, 91: 852-860 subramani, t., rajmohan, n., elango, l. (2009). groundwater geochemistry and identification of hydrogeochemical processes in a hard rock region, southern india. environ monit assess, doi:10.1007/s10661-009-0781-4. suresh, s., dinakar, n., prasad, t.n.v.k.v., nagajyothi, p.c., damodaram, t and nagaraju, a. (2007). effects of barite mine on groundwater quality in andhra pradesh, india. mine water and env., 26: 119-123. tripathi, a.k., mishra u.k., mishra ajay., dubey, p.(2012). assessment of groundwater quality gurh tehseel, rewa district madhya pradesh, india. international journal of scientific and engineering research, 3(9): 1-12. vasanthavigar, m., srinivasamoorthy. k., prasanna. m. v. (2012). evaluation of ground water suitability for domestic, irrigational, and industrial purposes: a case study from thirumanimuttar river basin, tamilnadu, india, journal of environmental monitoring assessment, 184: 405 420. voutsis, n.,kelepertzis, e.,tziritis, e.,kelepertsis, a. (2015). assessing the hydrogeochemistry of groundwaters in ophiolite areas of euboea island, greece, using multivariate statistical methods. journal of geochemical exploration, 159:79–92. wu, j., li, p., qian, h., duan, z., zhang, x. (2014). using correlation and multivariate statistical analysis to identify hydrogeochemical processes affecting the major ion chemistry of waters: a case study in laoheba phosphorite mine in sichuan, china. arab j geosci, 7:3973–3982. doi:10.1007/s12517-013-1057-4 wu, j., li, p., qian, h. (2015). hydrochemical characterization of drinking groundwater with special reference to fluoride in an arid area of china and the control of aquifer leakage on its concentrations. environ earth sci, 73:8575–8588. 127 earth sciences research journal earth sci. res. sj. vol. 17, no. 2 (december, 2013): 127 133 a methodology for the analysis of the rayleigh-wave ellipticity v. corchete higher polytechnic school, university of almeria, 04120 almeria, spain seismology in this paper, a comprehensive explanation of the analysis (filtering and inversion) of the rayleigh-wave ellipticity is performed, and the use of this analysis to determine the local crustal structure beneath the recording station is described. this inversion problem and the necessary assumptions to solve are described in this paper, along with the methodology used to invert the rayleigh-wave ellipticity. the results presented in this paper demostrate that the analysis (filtering and inversion) of the rayleigh-wave ellipticity is a powerful tool, for studying the local structure of the crust beneath the recording station. therefore, this type of analysis is very useful for studies of seismic risk and/or seismic design, because it provides local earth models in the range of crustal depths. en este artículo, será realizada una explicación fácil de comprender del análisis (filtrado e inversión) de la elipticidad de la onda rayleigh, así como, de su aplicación para determinar la estructura local bajo la estación de registro. este problema de inversión y los necesarios supuestos para resolverlo, serán descritos en el artículo, juntamente con la metodología utilizada para invertir la elipticidad de la onda rayleigh. los resultados presentados en este artículo mostrarán que el análisis (filtrado e inversión) de la elipticidad de la onda rayleigh es una poderosa herramienta, para estudiar la estructura local de la corteza terrestre bajo la estación de registro, siendo este tipo de análisis muy útil para estudios de riesgo y/o diseño sísmico, porque proporciona modelos de tierra locales en el rango de profundidades corticales. palabras clave: elipticidad; filtro técnico múltiple; filtrado temporal variable; inversión. record manuscript received: 23/10/2012 accepted for publication: 06/08/2013 abstract resumen key words: ellipticity, multiple filter technique (mft), time variable filtering (tvf), inversion. 1. introduction as it is well known, many surface-wave studies have been devoted to obtaining the structure of the earth (crust and upper mantle) from the surface-wave dispersion analysis (filtering and inversion). the surface-wave dispersion varies along the propagation path, in a manner related to the earth structure that is crossed by these waves. therefore, the surface-wave dispersion analysis is a good tool for studying the most important features of the earth structure. in particular, in the iberian region, several shear-velocity models (earth models) of the crust and upper mantle have been obtained by corchete et al. (1993, 1995), from the analysis of rayleigh-wave dispersion. this analysis consisted of filtering and inverting the rayleigh-wave dispersion, to obtain the variation in the shear-wave velocity versus depth. nevertheless, the rayleigh-wave dispersion is dependent upon the propagation path of these waves, and it does not depend (primarily) upon the local crustal structure beneath the recording station (seismic station). as a consequence of this, the rayleigh-wave dispersion analysis could not be used to determine earth models (crustal models) beneath local areas (small areas), such as the area in the vicinity of the seismic station (malischewsky et al., 2010; tuan et al., 2011). at this point, the analysis (filtering and inversion) of the rayleighwave ellipticity gain importance, because sexton et al. (1977) have shown that the rayleigh-wave ellipticity depends primarily upon the local crust structure and that it does not exhibit azimuthal dependence, for a period range of 10 to 50 s. thus, for this period range, the observed ellipticity is primarily controlled by the local crustal geology beneath the seismic station, it does not depend upon the propagation path of the rayleigh waves. this basic property of the ellipticity means that the ellipticity analysis is a very useful tool for obtaining local crustal models, which can be used to determine the site and/or local effects in studies of seismic risk and/or seismic design (lee et al., 2003; rastogi et al., 2011). for this reason, it is very desirable to develop a methodology for the analysis of the rayleigh-wave ellipticity that can be used to determine the local crustal structure beneath the recording station (seismic station). this analysis consists of filtering and inverting rayleigh-wave ellipticity curves to obtain s-velocity models (variations in shear-wave velocity versus depth). thus, the goal of this paper is to provide a comprehensive explanation of the analysis (filtering and inversion) of the rayleigh-wave ellipticity, devoting particular attention to the inversion problem and the assumptions necessary to solve it. 128 table 1. near events recorded at the tol station (latitude 39.881 ºn, longitude -4.049 ºe). event (nº) date (day month year) origin time (hour min. sec.) latitude (ºn) longitude (ºe) magnitude (mb) 1 09 05 80 09 21 36.7 35.43 1.10 6.0 2 13 10 80 14 33 48.2 36.28 1.29 5.2 3 14 10 80 17 35 02.1 36.22 1.33 7.1 4 15 10 80 03 17 24.7 36.19 1.29 6.0 5 19 10 80 21 47 24.2 35.51 1.20 5.8 6 22 10 80 16 23 15.8 36.22 1.28 5.4 7 23 10 80 09 57 56.7 36.35 1.32 5.4 8 30 10 80 23 38 14.9 36.18 1.34 5.3 9 01 02 81 23 00 42.0 36.37 1.64 5.8 10 05 02 81 09 12 33.2 36.35 1.44 5.4 11 14 02 81 13 15 18.0 35.85 1.22 5.3 12 20 02 81 20 41 02.9 35.40 1.18 5.0 13 05 03 81 01 21 52.3 38.47 0.00 5.5 14 11 09 82 21 35 51.9 36.32 1.60 5.6 15 11 09 82 04 48 41.6 35.56 1.31 5.5 16 15 11 82 20 07 52.4 35.59 1.25 5.7 17 31 03 84 03 31 56.0 36.41 1.52 4.9 18 12 02 89 12 02 21.6 36.33 2.60 5.3 19 09 05 80 09 21 36.7 35.43 1.10 5.0 20 13 10 80 14 33 48.2 36.28 1.29 5.4 21 14 10 80 17 35 02.1 36.22 1.33 5.2 22 15 10 80 03 17 24.7 36.19 1.29 5.1 figure 1. frequency response of the wwssn seismograph. v. corchete 129 2. data in this type of study, the primary data used are seismograms. the seismograms used in this study correspond to 22 earthquakes (table 1) that occurred in the neighboring region of the iberian peninsula. these earthquakes were registered by the wwssn station tol located in iberia (latitude 39.881 ºn, longitude -4.049 ºe). the records of this station are considered to be suitable for this study, because the period range of best registration for the wwssn seismograph is between 5 and 50 s (the period range in which the amplitude spectrum is maximized), as seen in figure 1 (kulhánek, 1990). this period range (from 5 to 50 s) is most suitable for exploring the elastic structure of the earth for a depth range from 0 to 50 km, which coincides with the goal of this study. only earthquake traces in which there exist well-developed rayleigh wave trains with a very clear dispersion are considered to ensure the reliability of the results. logically, instrumental correction must be taken into account to eliminate the time lag introduced by the seismograph system and all distortions produced by the instrument (bath, 1982; kulhánek, 1990). this correction recovers the true amplitude and phase of the ground motion, allowing the analysis of the true ground motion attributable to the rayleigh wave. for this reason, all traces considered in this study have been corrected for the instrument response. 3. computation of the observed ellipticity it is known that the observed ellipticity of a rayleigh wave can be calculated using the formula (aki and richards, 1980; ben-menahem and singh, 1981) where and are the instrument-corrected spectral amplitudes of the longitudinal (also called radial) and vertical seismograms for the period t (sexton et al., 1977). nevertheless, the use of equation (1) poses some practical difficulties. the first difficulty of formula (1) is that the ellipticity values can be calculated only for the periods generated by the fast fourier transform (fft). thus, many values can be computed for short periods, but only a few values can be computed for long periods. the second important difficulty of formula (1) is the presence of spectral holes in , the spectral amplitude of the vertical seismogram. a spectral hole is a small value (near zero) that can exist, for some period t, in the spectral amplitude. this small value, when introduced into the denominator of a spectral ratio such as the ratio given by formula (1), produces a very large value (near infinite) for this ratio ( (t) → ∞ for some period t in which → 0). in a spectral hole, the formula (1) has no physical meaning and it cannot be applied. spectral holes can be produced by noise or other such undesirable perturbations that typically are present in all seismograms. because of the above-mentioned difficulties, the use of equation (1) directly is not suitable for a systematic computation; instead, a time-domain method, based on the multiple filter technique (mft) of dziewonski et al. (1969), is more suitable for the systematic computation of the observed ellipticity. thus, the time-domain method developed by sexton et al. (1977) is used here for the computation of the observed ellipticity, instead of the spectral ratio given by formula (1). figure 2 shows the seismograms recorded at the tol station for event 3 (listed in table 1). these records are used to illustrate the time-domain method applied in this study for the computation of the observed ellipticity. figure 3 shows the results obtained by applying the method developed by sexton et al. (1977), to the seismograms shown in figure 2. it should be noted that noise and other undesirable perturbations, which are always present in the data (figure 3c), can introduce errors into the computation of the observed ellipticity. figure 2. observed (instrument-corrected) seismograms corresponding to event 3 of table 1 recorded at the tol station. the three components are: north-south (n), east-west (e) and vertical (z). figure 4. (a) observed instrument-corrected seismogram (vertical component) corresponding to event 3 of table 1 recorded at the tol station. (b) group velocity dispersion curve obtained after the application of the mft to the seismogram shown in figure 4a. (c) time-variable filtered seismogram. figure 3. (a) observed (instrument corrected) seismograms corresponding to event 3 of table 1 recorded at the tol station. the longitudinal (radial) seismogram (l component) is computed from the horizontal seismograms (north-south and east-west seismograms) shown in figure 2. (b) observed ellipticity computed using the timedomain method developed by sexton et al. (1977). (c) rayleigh-wave particle motion. a methodology for the analysis of the rayleigh-wave ellipticity (1) 130 the majority of this noise and undesirable perturbations can be removed from the data (seismograms) using the time variable filtering (tvf) technique. tvf is a filtering technique that produces a smooth seismogram (a time-variable-filtered seismogram), in which all effects of noise, higher modes and other undesirable perturbations have been removed (cara, 1973). this technique can be used in combination with the mft, as described by corchete et al. (2007), to obtain better results than those obtained using the mft only. the use of tvf in this study is illustrated in figure 4, in which the vertical seismogram shown in figure 2 has been time-variable filtered. this technique was applied to all seismograms used in this study to compute the observed ellipticity, as shown in figure 5. a comparison between figures 3 and 5 illustrates the improvement achieved by the combination of the mft and tvf (figure 5) versus the application of the mft only (figure 3). the particle motion (ground motion) attributable to the rayleigh wave is much clearer in figure 5c than in figure 3c, which confirms that the observed ellipticity presented in figure 5b is more reliable than the one presented in figure 3b. once the ellipticity curve (observed ellipticity) for the seismograms of each event listed in table 1 has been calculated using the above-described combination of filtering techniques (mft and tvf), an average ellipticity curve (figure 8c) can be obtained by computing the median of the 22 ellipticity curves calculated for the 22 events listed in table 1. an estimation of the error on this average ellipticity curve is obtained by computing the standard deviations (1-s error) of the values involved in the calculation of this median. 4. inversion of the observed ellipticity theoretical problem. once the rayleigh-wave ellipticity (observed ellipticity) has been obtained, as described in the section above, a relation between the ellipticity and the earth model (given by its characteristic parameters d: layer thickness, a: p-wave velocity, b: s-wave velocity and r: density) must be established. fortunately, this relation exists (and is well known) through the rayleigh-wave amplitudes in the form (ben-menahem and singh, 1981) = u(z = 0) /w(z = 0) where u(z) and w(z) are the horizontal (also called radial or longitudinal) and vertical amplitudes, respectively. these amplitudes are calculated from an earth model as functions of the depth z, for a given period t, using the forward modelling method described by abo-zena (1979) and aki and richards (1980). thus, the theoretical ellipticity (known as the surface ellipticity) defined at z = 0 by formula (2) can be related to an earth model through the general relation i = 1, 2, …, m where i is the ith value of the period t (in which the observed ellipticity is calculated), m is the number of values for the period t and n is the number of layers for the earth model under consideration. this layered model is defined by the following parameters: (layer thickness), (pwave velocity), (s-wave velocity) and (density), where the subscript j corresponds to the jth layer of this model. linearization of the problem. equation (3) is a set of m no-linear equations with 4*n unknowns that express the existing relation between the ellipticity and the earth model. nevertheless, the inverse relation of equations (3) is needed to obtain the values of the model parameters ( , , and for j varying from 1 to n) from the observed data ( i for i varying from 1 to m). this inverse relation can be very difficult to obtain from equations (3) in absence of any simplification. the simplification approach used in this study is the method developed by dorman and ewing (1962). this method allows the linearization of equations (3) using the partial derivatives of the ellipticity. these partial derivatives are introduced into the problem through the definition of the ellipticity differential, given by i = 1, 2, …, m where n (= 4*n) is the total numbers of model parameters and pj is a model parameter (with value ). then, if an initial model is prepared to produce a theoretical ellipticity curve that is as close as possible to the observed curve, the difference between the theoretical and observed ellipticities can be written as i = 1, 2, …, m where is the observed ellipticity obtained for the ith value of the period t, is the theoretical ellipticity calculated using formula (3) and dei is computed using formula (4). it should be noted that equations (4) and (5) provide an approximate linear relation between the ellipticity and figure 5. (a) time-variable-filtered seismograms computed from the seismograms shown in figure 2, as illustrated in figure 4. the longitudinal (radial) seismogram (l component) is computed from the time-variable filtered seismograms (horizontal seismograms). (b) observed ellipticity computed using the time-domain method developed by sexton et al. (1977). (c) rayleigh-wave particle motion. figure 6. (a) shear velocity distribution as a function of depth for the initial model described in table 2. (b) theoretical ellipticity calculated from the initial model (described in table 2), by means of forward modelling. v. corchete (2) (3) (4) (5) 131 the earth model, when only small perturbations ( ) of model parameters are considered. in this case, a linear relation between the values of the model parameters ( , , and for j varying from 1 to n) and the observed data (ei for i varying from 1 to m) can be written in the form i = 1, 2, …, m equations (6) is a set of linear equations that can be inverted to obtain the values of the model parameters from the observed data. nevertheless, when the total number of parameters n that are taken into consideration is very large, a precise solution for the inverse problem can be very difficult to obtain. for this reason, the total number of parameters must be reduced to the minimum necessary parameters. the total number of parameters can be reduced when only the s-wave velocities are considered in formula (6), i. e. and n = n. thus, the perturbations ( ) for the p-wave velocity and density are not considered in formula (6) because surface waves are not very sensitive to them (dorman and ewing, 1962). under this final assumption, equation (6) can be inverted if the partial derivatives are evaluated numerically. this numerical evaluation of the partial derivatives can be performed using the approximate relation given by where i is calculated using formula (3) and is a small perturbation ( << ). inversion scheme. once the forward problem has been linearized using (6) and (7), the inversion of (6) or, equivalently, the inversion of the matrix relation y = ax must be performed, where y = , a is given by (7) and x = figure 7. steps followed in the process of inverting the observed ellipticity to obtain the s-velocity model. the circles are used to denote computation steps, and the rectangles are used to denote the results obtained. the triple-outlined rectangles are used to indicate the initial data and the final result. a methodology for the analysis of the rayleigh-wave ellipticity (7)(6) (8) 132 . the inverse of the matrix a in relation (8) can be calculated using (dimri, 1992) where a is the damping parameter or damping factor and i is the identity matrix. the addition of this factor to the main diagonal of matrix ata increases the magnitude of its eigenvalues, thereby avoiding the problems caused by small or zero eigenvalues, which can cause the matrix ata to be singular or nearly singular. when the relation called the singular value decomposition (svd) of a a = uplpvp t is substituted into formula (9), b can be written as where the matrices and are described in detail by aki and richards (1980) and dimri (1992). the matrix b defined by formula (10) is called the damped generalized inverse. then, the linear equations given by (8) can be inverted to obtain x using the expression given in (10) as follows where it should be noted that the product of matrices b and a is called the resolution matrix, and it is obtained as follows if the resolution matrix given by (12) is the identity matrix, then the solution obtained in this manner is unique and the resolution is perfect. nevertheless, it should be noted that the addition of the factor a to the main diagonal ensures that the equality ba = i is never reached when the damping factor a is not zero. therefore, the particular solution that is obtained is never the true solution. the obtained solution is instead a smoothed solution because each model parameter is a weighted mean of the parameters x that define the true model. for this reason, the factor a must be as small as possible to yield the best possible solution, i.e., the particular solution that is as near as possible to the true solution. the error on the model parameters is given by the corresponding variance (assuming that the data are statistically independent and that the same is true for the model parameters) as follows where and are the variances of the model parameters and the data, respectively. the inversion scheme proposed in this paper is defined by (10) and it introduces the damping parameter, which degrades the resolution but stabilizes the obtained solution by reducing the variance, as seen in (12) and (13). thus, an increase in the value of a reduces the error in the determination of the model parameters at the expense of the degradation of the resolution. on the other hand, the inverse given by (10) uses the svd method instead of the least-squares method, which is used by dorman and ewing (1962), because the svd method provides higher computation accuracy (keilis-borok, 1989). inversion process. a necessary step that must precede the inversion process is the preparation of an initial model that contains all available information for the study region, including the s-wave, p-wave and density distributions as functions of depth. in this study, this information was obtained from the references provided and is described in section 1. all this information was considered in the preparation of the initial model used to invert the ellipticity curve shown in figure 8c, which was obtained as described in section 3. the initial model must be prepared to produce a theoretical ellipticity curve that is as close as possible to the observed curve, because the inversion scheme described above considers only small perturbations in the model parameters involved. this method simply conditions the initial model, which is chosen to be as similar as possible to the true earth structure, as represented by all geophysical information available for the study region. the initial model prepared here is described in table 2, and its shear-wave velocity distribution as a function of depth is plotted in figure 6a. the calculated theoretical ellipticity is shown in figure 6b. this curve was obtained using the forward-modelling method described by abo-zena (1979) and aki and richards (1980). table 2. initial earth model considered in this study (a: p-wave velocity, b: s-wave velocity, r: density). thickness (km) α (km/s) β (km/s) ρ (g/cm3) 2 3.30 2.50 2.28 3 5.60 3.19 2.74 5 6.10 3.48 2.79 5 5.60 3.18 2.79 10 6.40 3.58 2.80 5 6.85 3.90 3.05 60 8.00 4.50 3.20 80 8.15 4.40 3.40 130 8.49 4.77 3.53 8.81 4.89 3.60 once the initial model has been prepared, the inversion of the observed ellipticity curve (figure 8c) can be performed to obtain the corresponding s-velocity model. a computer program based on the above-described inversion scheme, which performs the computation steps shown in the flow chart displayed in figure 7, was used for this task. the results of this computer program are shown in figure 8. the s-velocity model shown in figure 8a is the final obtained model, which was improved iteratively. the iterative improvement in the model is assessed by comparison between the observed ellipticity, considered to be the observed data, and the theoretical ellipticity calculated from the actual model via forward modelling. if each successive curve is closer to the observed ellipticity curve than the previous iteration, the process converges. when the theoretical curve falls within the vertical error bars of the observed data, as seen in figure 8c, the inversion process is considered to be finished because the obtained shear-velocity model describes the observed curve within its experimental error. the resolving kernels shown in figure 8b indicate the reliability of the solution obtained for the considered inverse problem. the resolving kernels are the row vectors of the resolution matrix given by formula (12), and therefore, they are a measure of the exactness of the solution of the inverse problem. good agreement between the calculated solution and the true solution (which implies the reliability of the estimated solution) is obtained when the absolute maxima of these functions lie at the reference depths. these reference depths are defined as the median depths of each layer considered in the model. the reliability of the solution is also related to the widths of these absolute maxima. the solution identified for the inversion problem is more reliable when the maxima of these resolving kernels are narrower. to v. corchete (9) (10) (11) (12) (13) 133 achieve this goal and improve the resolution, the number of model layers considered for the inversion must be reduced to be the minimum number of layers that are necessary to obtain a detailed shear-velocity distribution as function of depth, with sufficient precision to describe the observed data within the experimental error (singh et al., 2011). in this case, the number of layer considered for the inversion (i. e., the total number of model parameters to be perturbed) is n = 9. if a large number of layers are considered, then the number of unknowns is increased but the amount of data remains the same, and the resolution is degraded as a result (corchete et al., 2007). this effect manifests as resolving kernels with wider absolute maxima and in the lack of coincidence between the maxima of these functions and the reference depths. another method of improving the resolution is to consider layers with thicknesses that increase with increasing depth, as the resolution worsens. this effect is apparent in figure 8b: the resolving kernels worsen as the depth increases. the layer thickness considered in our model increases with increasing depth to avoid this effect as much as possible. thus, the shallow layers are thinner than the deeper layers, which are thicker (figure 8a). figure 8. (a) shear-wave velocity model obtained after the inversion of the observed ellipticity curve. the horizontal bars represent the standard deviation computed for each layer considered in this inversion process (9 layers). (b) resolving kernels of the inversion problem posed. the reference depths are indicated with vertical bars at the median depth of each layer considered. (c) theoretical ellipticity curve calculated from the final model plotted in figure 8a (continuous line). the dots and line denotes the observed ellipticity, and the vertical bars represent the standard deviation of the ellipticity (1-s errors). 5. conclusions the results presented in this paper demonstrate that the analysis of rayleigh-wave ellipticity is a powerful tool for investigating the local structure of the crust and upper mantle directly beneath the recording station. this technique can reveal the principal structural features beneath the seismic station because the rayleigh-wave ellipticity depends primarily upon the local crust structure and does not exhibit azimuthal dependence in the period range of 10 to 50 s. thus, a general picture of the lithospheric structure beneath the station, in a depth range from 0 to 50 km, can be obtained using this type of analysis. therefore, the methodology presented in this paper, which consist of filtering and inverting rayleigh-wave ellipticity curves to obtain s-velocity models (variations in shear-wave velocity versus depth), can be a very useful tool in studies of seismic risk and/or seismic design because it provides local earth models in the range of crustal depths. acknowledgements the author is grateful to instituto geográfico nacional (madrid, spain), which provided the seismic data used in this study. references abo-zena a., 1979. dispersion function computations for unlimited frequency values. geophys. j. r. astr. soc., 58, 91-105. aki k. and richards p. g., 1980. quantitative seismology. theory and methods. w. h. freeman and company, san francisco. bath m., 1982. spectral analysis in geophysics. elsevier science publishers. amsterdam. ben-menahem a. and singh s. j., 1981. seismic waves and sources. springer-verlag. new york. cara m., 1973. filtering dispersed wavetrains. geophys. j. r. astr. soc., 33, 65-80. corchete v., badal j., pujades l. and canas j. a., 1993. shear velocity structure beneath the iberian massif from broadband rayleigh wave data. phys. earth planet. inter., 79, 349-365. corchete v., badal j., serón f. j. and soria a., 1995. tomographic images of the iberian subcrustal lithosphere and asthenosphere. journal of geophysical research, 100, 24133-24146. corchete v., chourak m. and hussein h. m., 2007. shear wave velocity structure of the sinai peninsula from rayleigh wave analysis. surveys in geophysics, 28, 299-324. dimri v., 1992. deconvolution and inverse theory. application to geophysical problems. methods in geochemistry and geophysics, 29. elsevier science publishers, amsterdam. dorman j. and ewing m., 1962. numerical inversion of seismic surface wave dispersion data and crust-mantle structure in the new yorkpennsylvania area. journal of geophysical research, 67, 5227-5241. dziewonski a., bloch s. and landisman m., 1969. a technique for the analysis of transient seismic signals. bull. seism. soc. am., 59, 427444. kulhánek o., 1990. anatomy of seismograms. elsevier science publishers. amsterdam. lee w., kanamori h., jennings p. and kisslinger c., 2003. international handbook of earthquake and engineering seismology. elsevier, new york. malischewsky p. g., zaslavsky y., gorstein m., pinsky v., tran t. t., scherbaum f. and flores-estrella h., 2010. some new theoretical considerations about the ellipticity of rayleigh waves in the light of site-effect studies in israel and mexico. geofísica internacional, 49, 141-152. rastogi b., k., a. p. singh, b. sairam, s. k. jain, f. kaneko, s. segawa and j matsuo, 2011. the possibility of site effects: the anjar case, following the past earthquakes in the gujarat, india. seismological research letters, 82(1): 692-701; doi 10.1785/ gssrl.82.1.692. sexton j. l., rudman a. j. and mead j., 1977. ellipticity of rayleigh waves recorded in the midwest. bull. seism. soc. am., 67, 369-382. singh, a., p., o. p. mishra, b. k. rastogi and dinesh kumar, 2011. 3-d seismic structure of the kachchh, gujarat and its implications for the earthquake hazard mitigation. natural hazards, 57(1): 83-105; doi 10.1007/s11069-010-9707-2. tuan t. t., scherbaum f. and malischewsky p. g., 2011. on the relationship of peaks and troughs of the ellipticity (h/v) of rayleigh waves and the transmission response of single layer over half-space models. geophys. j. int., 184, 793-800. a methodology for the analysis of the rayleigh-wave ellipticity there are enormous resources of unconventional gas in coal measures in ordos basin. in order to study the geological characteristics of unconventional gas in coal measures in ordos basin, we analyzed and summarized the results of previous studies. analysis results are found that, the unconventional gas in coal measures is mainly developed in upper paleozoic in eastern ordos basin, which including coalbed methane, shale gas and tight sandstone gas. the oil and gas show active in coal, shale and tight sandstone of upper paleozoic in ordos basin. coalbed methane reservoir and shale gas reservoir in coal measures belong to “self-generation and selfpreservation”, whereas the coal measures tight sandstone gas reservoir belongs to “allogenic and self-preservation”. the forming factors of the three different kinds of gasses reservoir are closely related and uniform. we have the concluded that it will be more scientific and reasonable that the geological reservoir-forming processes of three different kinds of unconventional gas of coal measures are studied as a whole in ordos basin, and at a later stage, the research on joint exploration and co-mining for the three types of gasses ought to be carried out. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): q1 q5 abstract keywords: ordos basin, unconventional gas in coal measure, coalbed methane, shale gas, tight sandstone gas, upper paleozoic. geological characteristics of unconventional gas in coal measure of upper paleozoic coal measures in ordos basin, china record manuscript received: 12/11/2015 accepted for publication: 19/02/2016 how to cite item he, j., zhan, x., ma, l., wu, h., & ashraf, m.a. (2016). geological characteristics of unconventional gas in coal measure of upper paleozoic coal measures in ordos basin, china. earth sciences research journal, 20(1), q1-q5. doi: http://dx.doi.org/10.15446/esrj.v20n1.54140 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.54140 p e t r o l e u m g e o l o g y jinxian he,1,2* xiaoli zhang,1,2 li ma,1,2 hongchen wu,1,2 muhammad aqeel ashraf3 1school of resources and geosciences, china university of mining and technology, xuzhou, jiangsu, 221116, china; 2 key laboratory of coalbed methane resource and reservoir formation process, ministry of education(china university of mining and technology), xuzhou 221008 china; 3 faculty of science and natural resources, university malaysia sabah 88400 kota kinabalu sabah, malaysia * corresponding author: jinxian he, email: jxhe@cumt.edu.cn introduction unconventional gas in coal measures is all natural gasses which are generated through evolution when the parent materials of source rocks underwent coalification of biochemistry and physiochemistry. it includes gasses such as coalbed methane, shale gas, tight sandstone gas, natural gas hydrate (cao et al., 2014; wang et al., 2014). unconventional gas resources of coal measures have a huge amount (table 1). according to the principle of “giving priority to coal, comprehensive exploration, comprehensive evaluation” of the geological survey of coal resources in china (deng et al., 2006), improving the research of the gas resources associated with coal in coal measures is of great significance. table 1. distributional situation of gas resources of coal measures in ordos basin and china (zhao et al., 1998; feng et al., 2002; liu et al., 2009; jie, 2010; zhang et al., 2012a; yang et al., 2015) in chinese coal measure strata, because the coal seam, dark mud shale, and tight sandstone interbedded deposit, the geological phenomenon which several kinds of coal measures gasses jointly preserve is relatively prevalent. scholars have agreed to take unconventional gas in coal measures as a system of research and development. they also agreed to carry out joint exploration and exploitation for multi-gasses (qin et al., 2014; cao et al., 2014; wang et al., 2014; gao et al., 2014). the comprehensive research of unconventional gas in coal measures is the trend in the future(qin et al.,2016; liang et al.,2016a).it may improve the production of the single well, reduce production cost and increase economic benefit that multigasses are jointly exploited from a well(qin et al.,2016; liang et al.,2016a). additionally, in coal mining, because the surrounding rock gas of coal seam is a primary source of the coal mine gas, the coal measures’ gas (not only coalbed methane) drainage helps reduce the hazard of coal mine gas and efficiently reduce greenhouse gas emissions. 1 regional geological characteristics as the second largest sedimentary basin in china, ordos basin is also an important coal-bearing and oil-bearing one. the whole basin may be divided into six first-order tectonic units. in the basin, the gas resources of q2 jinxian he, xiaoli zhang, li ma, hongchen wu and muhammad aqeel ashraf upper paleozoic coal measures mainly distribute in the east and principally includes the east of yanan area in the yishan slope and the deflection fold belt in western shanxi (fig. 1). in the eastern ordos basin, greater total thickness, multi-beds, and stable distribution of coal seams developed in the upper paleozoic coal measure strata; at the same time, some dark mud shale and tight sandstone interceded with these coal seams. it provides the abundant material basis for preservation and development of coalbed methane, shale gas, and tight sandstone gas. previous researches indicated that coalbed methane, shale gas, tight sandstone gas jointly developed in the upper paleozoic strata in the eastern ordos basin (including linxing area) (guo et al., 2012; guo and zhao, 2014), therefore providing a favorable geological condition for carrying out joint multi-gases exploration and exploitation. figure 1. sketch of regional tectonic division of ordos basin. the developed lithologies in upper paleozoic of ordos basin are mainly coal seam, sandstone, mud shale, and few carbonatite. benxi formation, taiyuan formation, shanxi formation, shihezi formation, and shiqianfeng formation, from bottom to top, developed in sequence. of them, taiyuan formation and shanxi formation are leading strata containing coal, followed by the development degree of benxi formation (table 2). therefore, taiyuan formation and shanxi formation are also the most important target beds for exploration and exploitation of coal measures gases. 2 coalbed methane coalbed methane resource in ordos basin reached 10.72 × 1012 m3 and accounted for about 1/3 of the country’s total volume. therefore, it played a significant role in the distribution of chinese coalbed methane resource. in the basin, about 1/2 of the coalbed methane resources distribute in its eastern margin (table 1), making the east margin of ordos basin one of the most principal explored and exploited areas for coalbed methane in china. at the same time, it is also the area which is most explored in ordos basin (feng et al., 2002). at present, commercial exploitations for coalbed methane have been launched. in eastern ordos basin, it was well developed that upper paleozoic carboniferous-permian coal measure strata, the layers of coal seams are more, and the distribution is relatively stable. coal seams concentratedly drawn up in taiyuan formation and shanxi formation. of which taiyuan formation contains 3~7 layers of coal seams, and 2~4 layers can be exploited; total thickness is 4.3~7.1 m. shanxi formation contains 4~8 layers of coal seams, and 2~3 layers can be exploited; total thickness is 3.1~21.8 m. burial depths of coal seams are less than 1500 m. in the eastern margin of ordos basin, main coal seams overall take on a tendency of “thick north and thin south” (zhang et al., 1996). the distributional ranges of coal are from long flame coal to meager coal, concentrating on medium rank coal. moreover, the vitrinite reflectance ro is between 0.5 and 2.2 % (jie, 2010). table 2. stratigraphic division and lithological developing characteristics of upper paleozoic in ordos basin (yan et al., 2013) table 3. geological characteristics data of main coal seam of upper paleozoic in ordos basin (jie, 2010) in this area, the fractures of cleats of main coal seam were comparatively developed; density of endogenetic fracture is 10~30 pieces/ 5 cm. overall, the physical characteristics of coal reservoir are relatively favorable, and permeability of main coal seam is 0.22×10-3 μm2 ~ 8×10-3 μm2. reservoir pressure gradient of the main coal seam is 0.5~0.9 mpa/100 m; coal seam desorbing pressure is 1.7 ~ 4.0 mpa; and gas saturation of coal seam is 55 – 86 % (table 3) (jie, 2010). the gas-bearing feature of the coal seam is relatively favorable, and gas content of the main coal seam is 4 ~ 20.87 m3/t. in particular, coalbed gas content of weibei block is 6 ~ 14 m3/t; the coalbed gas content of linfen block is 9.03 ~ 20.87 m3/t; the coalbed gas content of lvliang block is 4 ~ 14 m3/t (jie, 2010). the above data suggests an excellent potential of coalbed methane resources. some scholars also studied the origin of coalbed methane in the eastern margin of ordos basin. the studying results show that the methane carbon isotope distributes in -70.5 ‰ ~ -36.19 ‰ and evidently from north to south it gradually becomes weighting. mixed gas, including resulting biogas and thermogenic gas, occurs in the peripheral and shallow part, and thermogenic gas is within the middle and profound part (li and zhang, 2013). through the study of the origin of coalbed methane in hancheng region of southeastern ordos basin, ma et al. (2011) discovered that the methane carbon isotope of coal seam was in -42.978 ‰ ~ -32.200 ‰ and ethane carbon isotope belonged to -21.619 ‰ ~ -9.751 ‰. therefore, they considered that the origin type of coalbed methane in this region was predominantly thermal degradation gas. 3 shale gas in coal measures in ordos basin, upper paleozoic carboniferous – permian source rocks are predominantly dark mud shale; organic matter is abundant; thickness is stable; the distributional area is large; the maturity is high. the rocks of origin mainly distribute in benxi formation, taiyuan formation, and shanxi formation. q3geological characteristics of unconventional gas in coal measure of upper paleozoic coal measures in ordos basin, china the total thickness of mud shale is relatively high, with that of dark mud shale in benxi formation averaging at about 29.3 m, that of taiyuan formation at 39.6 m and that of shanxi formation at 63.2 m (table 4). the buried depth of upper paleozoic mud shale is between 2000 and 3500 m (wang et al., 2011). therefore, it provides an appropiate material basis for the formation of shale gas reservoir of coal measures, and its resource total extent is also tremendous. table 4. geological characteristics data of mud shale of upper paleozoic in ordos basin (yan et al., 2013) shale gas is an unconventional kind that is self-generation and selfpreservation. mud shale is not only a hydrocarbon-generating bed but also reservoir. geological characteristics data of the shale of upper paleozoic in ordos basin can be seen in table 5. in the basin, the average organic carbon content of upper paleozoic mud shale reaches 1.72 %; kerogen type is mainly iii; higher maturity of organic matter; ro value distribute in 1.6 ~ 2.82 %, thus it belongs to the mature-postmature stage and begins to generate a large quantity of gas at present. the porosity is averagely 1.12 %; permeability is 0.037×10-3μm2. brittle mineral content in shale is generally 46.51 % ~ 54.04 %; clay content, 43.61 % ~ 47.81 %. on the whole, upper paleozoic mud shale possesses the condition for shale gas reservoir to form (wang et al., 2011; yan et al., 2014). gas-bearing coal rock mud shale and gas-bearing tight sandstone mud shale are the optimal source-reservoir-cap combination type (yan et al., 2014). table 5. geological characteristics of mud shale of upper paleozoic in ordos basin (zhang et al., 2012b; guo et al., 2012; yan et al., 2013; yan et al., 2014) the gas content of the mud shale is an important parameter. several scholars have studied the gas content of upper paleozoic mud shale in ordos basin, and the data were collated in table 6. usually, 1.0 m3/t is taken as the lowest limit of industrial standard for the gas content of mud shale. from table 6, it can be known that gas content of mud shale in shanxi formation distribute in 0.96 ~ 2.71 m3/t and the average value is 1.94 m3/t, therefore, exceeding the minimum threshold of industrial standard. however, gas content of the mud shale of benxi formation of mi 35 well is slightly low, and this preliminarily reflects that the reservoir forming conditions of mud shale of shanxi formation are more favorable than benxi formation. in xiasiwan – yunyan region of eastern basin, gas measure and groove face of the dark mud shale interval of shanxi formation from more than 30 wells show active, and display good gas indicator and produced low-yield industrial gas flow (fig. 2) (gong et al., 2013). table 6. gas content characteristics of partly upper paleozoic mud shale in ordos basin (wang et al., 2011; fu et al., 2013; gong et al., 2013) figure 2. gas logging of shanxi formation of x24 well in eastern ordos basin (gong et al., 2013) several suits of mud shale developed in ordos basin. middle ordovician pingliang formation is difficult to form a large scale of shale gas reservoir; triassic yanchang formation mud shale possesses the reservoir forming condition for shale gas and tight oil. however, exploration potential of the tight oil is broader; carboniferous-permian sea-continental transitional facies shale is relatively favorable for shale gas reservoir. it is also the most certain strata series for exploration of shale gas (wang et al., 2011). 4 tight sandstone gas tight sandstone gas resource of upper paleozoic coal measures in ordos basin is very enormous, and the resource is preliminarily estimated to be more than 5x10-12 m3 (table 1). upper paleozoic in the basin extensively developed fine sandstone, siltite, argillaceous siltite, and other tighter sandstone. porosity and permeability of these sandstones average at 6.58 % and 0.77×10-3 μm2, respectively (deng, 2003). for these reasons, they belong to low porosity and low permeability tight sandstone. in ordos basin, it was developed that the fluvial-delta facies tight sandstone reservoir of shanxi formation and shihezi formation (fig. 3), thus this paper mainly discussed the geological characteristics of tight sandstone in these two horizons. the thickness of the sandstone is large, linxing area may be taken as an example. in this area, the thickness of taiyuan formation reaches 4.0 ~ 40.89 m, additionally the thickness of shanxi formation is larger than of taiyuan formation and the distributional scope is also wider (guo et al., 2012). the tight sandstone of coal measures staggeringly developed with mud shale and coal seam and adjoins gas source rock; therefore, it possesses the potential of tight sandstone gas. currently, exploration in ordos basin has made a significant breakthrough and entered the scale development stage. the lithologic character of shanxi formation is mainly feldspar debris quartz sandstone, followed by a few of feldspathic lithic sandstone and lithic sandstone. cement is predominantly silicic, and calcite cementation; grain-supported structure; type of cementation includes mainly pore and mosaic cementation (fig. 3). box 8 segment is chiefly feldspar debris quartz sandstone; apart from that, it also consists of a few of lithic sandstone and lithic quartz sandstone (zhou et al., 2013). q4 jinxian he, xiaoli zhang, li ma, hongchen wu and muhammad aqeel ashraf figure 3. developed characters of tight sandstone of coal measures in ordos basin (zhou et al., 2013) when the organic matter of coal measures transformed to hydrocarbons, organic acid was constantly produced, therefore coal measures strata are rich in organic acid. organic acid has a greater impact on the diagenesis of tight sandstone reservoir because organic acid may corrode feldspar to form a secondary pore. organic acid plays an active and constructive role in improving the physical properties of tight sandstone reservoir. in the basin, average porosity of the sandstone in taiyuan formation is 8.2 % and the permeability, porosity and permeability of shanxi formation average at 0.58×10-3μm2, 7.2% and 7.68×10-3μm2, respectively. also, the median porosity of sandstone and permeability in shihezi formation are 9.2 % and 10.42×10-3μm2 (table 7). therefore, physical properties of the sandstone of shihezi formation are better than those of shanxi formation and those of taiyuan formation are comparatively poor. meanwhile, coal measures tight sandstone staggeringly deposited and developed with coal seam and dark mud shale riching organic matter (fig. 2) to make a favorable condition for source rocks of tight sandstone. the gas show active in tight sandstone of upper paleozoic in ordos basin and especially in linxiang area(xie et al., 2016;liang et al.,2016b). in the southern basin, tight sandstone of segment 6 of lower shihezi formation underwent fracturing testing to attain daily capacity 8457 m3 gas flow, thus, it exhibits excellent exploration potential of tight sandstone gas. sandstone of shan 2 segment of yan 6 well is 12 m thick (fig. 3). logging showed gas anomaly, log interpretation revealed that gas reservoir was 8 m, the daily capacity including the underlain no. 2 coal seam was 6 000 m3 gas flow (zhou et al., 2013). table 7. physical properties characters of coal measures tight sandstone of upper paleozoic in ordos basin (fu et al., 2013) 5 reservoir formation model unconventional gas in coal measures belongs to “continuous oil and gas reservoir”, which belongs to different types. reservoirs of coalbed methane, shale gas, and tight sandstone gas are coal rocks, mud shale, and sandstone respectively, and all of these different types of reservoirs coexisted in the same coal-bearing series. their source types, such as sedimentary environments, structural, thermal evolutions, were in remarkable consistency. therefore, unconventional gas in coal measures possesses paragenesis characteristics in the same basin (cao et al., 2014). in coal measures, coal seam and dark mud shale are rich in organic matter, which constitutes source rocks of unconventional gas in coal measures. when these organic matters reach the threshold of hydrocarbon formation to generate gaseous hydrocarbon, and then it may supply hydrocarbon for coal seam, mud shale, and tight sandstone to altogether form gas reservoirs of a coal seam, shale, and tight sandstone respectively (fig. 4.). however, gas reservoirs of the coal seam and shale belong to self-generation and self-preservation type that display no straightforward migration process. therefore, they are distinctly different from the formation of the conventional oil-gas reservoir. the hydrocarbon gasses generated from coal seam and mud shale migrate to tight sandstone via a short-distance to gather into reservoir. therefore, the formation of tight sandstone gas reservoir of coal measures is similar to that of the conventional oil-gas reservoir. the slight difference that exists is that reservoir bed of tight sandstone gas reservoir of coal measures is close to hydrocarbon source rock bed, and the migration distance of natural gas is shorter than that of oil and gas of conventional oil-gas reservoir. figure 4. gas reservoir model of upper paleozoic coal measures in ordos basin(wang et al.,2014;cao et al.,2014) 6 conclusions (1) in ordos basin, unconventional gas in coal measures, e.g. coalbed methane, shale gas and tight sandstone gas, developed in upper paleozoic, and the resources amounts are enormous. the geological condition of ordos basin is favorable for three kinds of gasses occur together. therefore, at a later stage, the research on joint exploration and exploitation of the three types of gasses ought to be carried out. (2) unconventional gas of upper paleozoic coal measures mainly distributes in the eastern ordos basin. currently, the geological studying level to the coalbed methane in the eastern ordos basin is relatively high, and several areas have entered the commercial development stage. whereas studying level to the shale gas and tight sandstone gas of coal measures in this region are relatively small. available data indicate that in this region the oil gas shows of upper paleozoic mud shale and tight sandstone are favorable, and resource potential is conspicuous, about which further work is needed (3) unconventional gas of coal measures is accumulated through slightly different processes, of which coalbed methane reservoir and shale gas reservoir of coal measures belong to “self-generation and selfpreservation”, and the gas migration process almost did not happen. whereas the coal measures tight sandstone gas reservoir belongs to “allogenic, and self-preservation” and the natural gas experienced a short distance migration. although reservoir types of coal measures unconventional gas belonging to the same suit of strata are different, the forming factors of the gasses reservoir are closely related and uniform. therefore, it will be more scientific and reasonable that the geological reservoir-forming processes of three different kinds of unconventional gas of coal measures are studied as a whole. acknowledgement this study was supported by the fundamental research funds for the central universities (2015xkzd07) and a project funded by the priority academic program development of jiangsu higher education institutions. references cao, d.y., yao, z. and li, j. (2014). evaluation status and development trend of unconventional gas in coal measure. coal science and technology, 42(1), 89-92. deng, j., wang, q.f., gao, b.f., xu, h. and zhou, y.h. (2006). distribution and tectonic background of various energy resources in ordos basin. earth science, (journal of china university of geosciences), 31(3), 330-336. q5geological characteristics of unconventional gas in coal measure of upper paleozoic coal measures in ordos basin, china deng, l.z. (2003). influential factors on physical property parameters of upper paleozoic reservoirs in ordos basin[j]. journal of chengdu university of technology (sscience & ttechnology eedition), 30(3), 270-272. feng, s.l., ye, j.p., and zhang, s.a. (2002). coalbed methane resources in the ordos basin and its development potential. geological bulletin of china, 21(10), 658-662. fu, j.h., guo, s.b., liu, x.s. and wang, y.g. (2013). shale gas accumulation condition and exploration potential of the upper paleozoic shanxi formation in ordos basin. journal of jilin univcrsity (earth sciencc edition), 43(2), 382-389. gao, w., tian, w.j., qin, w., kong, w.m. and chen, m. (2014). geological optimization of coalbed methane and shale gas co-exploration and concurrent production in guizhou province (in chinese). fault-block oil and gas field, 21(1), 36-38. gong, m.l., ding, w.l., pi, d.d., cai, j.j., zhang, y.q., fu, j.l. and yao, j.l. (2013). forming conditions of shale gas of the shanxi formation of permian in the southeast of orldos basin. journal of northeast petroleum university, 37(3), 1-10. guo, b.g., xu, h., meng, s.z., zhang, w.z., liu, y.n., luo, h.h., li, y. and shen, w.m. (2012). geology condition analysis for unconventional gas co-exploration and concurrent production in linxing area. china coalbed methane coal seam gas china9(4), 3-6. guo, s.b. and zhao, k.y. (2014). gas-bearing influential factors and estimation of shale reservoirs in upper paleozoic, ordos basin. petroleum geology &and experiment, 36(6), 678-683. doi: 10.11781/sysydz201406678 jie, m.x. (2010). prospects in coalbed methane gas exploration and production in the eastern ordos basin. natural gas industry, 30(6), 1-6. li, g.h. and zhang, h. (2013). the origin mechanism of coalbed methane in the eastern edge of ordos basin. science china: earth sciences, 56(10), 1701-1706. doi: 10.1007/s11430-013-4616-3 liang, b., shi,y.s.,sun w.j. and liu, q.(2016a). reservoir forming characteristics of “the three gases” in coal measure and the possibility of commingling in china. journal of china coal society,41(1):167-173. liang, j.s.,zhu,x.s,liu,y.h., wang, c.w. and lv, y.m.(2016b). comparative study on accumulation conditions of the tight gas in the upper paleozoic of the qinshui basin and ordos basin and its exploration potential. journal of china coal society, 41(1):192-201. liu, c.l., zhu, j., che, c.b., yang, h.l. and fan, m.z. (2009). methodologies and results of the latest assessment of coalbed methane resources in china (in chinese). natural gas industry, 29(11), 130-132. ma, x.z., song, y., liu, s.., bai, z.h. and zheng, y.p. (2011). origins and geochemical characteristics of coalbed methane in hancheng, eastern ordos basin. natural gas industry, 31(4), 17-20. qin, y., liang, j.s., shen, j., liu, y.g., and wang, c.w. ( 2014). gas logging shows and gas reservoir types in tight sandstones and shales from southern qinshui basin. journal of china coal society, 39(8), 1559-1565. qin,y.,shen, j. and shen y.l.(2016).joint mining compatibility of superposed gas-bearing systems:a general geological problem for extraction of three natural gases and deep cbm in coal series. journal of china coal society,2016,41(1):14-23. wang, s.j., li, d.h., li, j.z., dong, d.z., zhang, w.z. and ma, j. (2011). exploration potential of shale gas in the ordos basin. natural gas industry, 31(12), 40-46. wang, t., wang, q.w. and fu, x.h. (2014). the significance and the systematic research of the unconventional gas in coal measures. coal geology & exploration, 42(1), 24-27. characteristics of marine-continental transitional organic-rich shale in the ordos basin and its shale gas significance. earth science frontiers, 21(6):1-10. doi: 10.13745/j.esf.2015.06.015 yang, h., liu, x.s., yan, x.x. (2015). the relationship between tectonicsedimentary evolution and tight sandstone gas reservoir since the late paleozoic in ordos basin. earth science frontiers, 22(3), 1-10. doi: 10.13745/j.esf.2015.03.015 zhang, g.s., zhao, w.z., yang, t., guo, b.c. and deng, s.t. ( 2012a). resource evaluation, position and distribution of tight sandstone gas in china.chinese eengineering ssciences, 14(6), 87-93. zhang, l.x., jiang, c.f. and guo, c. (2012b). exploration potential of upper paleozoic shale gas in the eastern ordos basin. journal of xi'an shiyou university ( natural science edition), 27(1), 23-26. zhang, s.l., li, b.f. and chen, x.d. (1996). occurrence and affecting geological factors of coalbed gas of carboniferous-permian in the eastern margin of ordos basin. petroleum exploration and development, 23(4), 11-14. zhao, q.b., sun, b. and li, w.z. (1998). conditions of forming large coal seam gas field and targets for exploration in eastern part of ordos basin. petroleum exploration and development, 25(2), 20-23. zhou, y.t., zhu, g.s. and wang, y. (2013). analysis of natural gas exploration potential of upper paleozoic tight sand of south yanchuan area in ordos basin.progress in exploration geophysicsreservoir evaluation and development, 3(5), 17-23. keywords: benin; scaling exponent; mfdfa; singularity spectrum; temperature; nonlinear dynamics. palabras clave: benín; exponente de escalado; mfdfa; espectro de singularidad; temperatura; dinámica no lineal. how to cite item agbazo, m., koto n'gobi, g., alamou, e., kounouhewa, b., afouda, a., & kounkonnou, n. (2019). multifractal behaviors of daily temperature time series observed over benin synoptic stations (west africa). earth sciences research journal, 23(4), 365-370. doi: https:// doi.org/10.15446/esrj.v23n4.51863 in this study, multifractal detrended fluctuation analysis (mfdfa) is applied to daily temperature time series in benin synoptic stations from 1967 to 2012. the purpose is to examine the degree of multifractality and compare multifractal characteristics of temperature in various climatic areas (subequatorial and sudanian). the analysis reveals that the daily temperature time series shows a multifractal behavior that is sensitive to the station’s geographic position and presents a positive long-term correlation. weather conditions and the geographic position of the synoptic stations affect the shape and the characteristics of temperature spectrum. the multifractality is stronger in subequatorial region than the sudanian. the major source of multifractality in temperature series is found to be the fat-tailed probability density function. however, long-range correlations also partly contribute to the multifractal features. except kandi’s synoptic station, the spectrum is left-skewed. the findings indicate the usefulness of the non-linear analysis in climate research due to the complex interactions among the natural processes. it can help understand the mechanisms governing the dynamics of temperature time series in benin. abstract multifractal behaviors of daily temperature time series observed over benin synoptic stations (west africa) comportamientos multifractales de las series temporales de temperatura observadas en las estaciones sinópticas de benin (africa occidental) issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.51863 en este estudio, el análisis de la fluctuación multifractal descendente es aplicado a diversos tiempos de temperatura cotidiana en estaciones sinópticas en benín de 1967 a 2012. el objetivo es examinar el nivel de multifractalidad y comparar las características multifractales de temperatura en varias zonas climáticas (subecuatorial y sudanés). el análisis muestra que los diversos tiempos de temperatura diaria presentan un comportamiento multifractal, que son sensitivos a la posición geográfica de la estación y que hay una correlación positiva a largo plazo. las condiciones climáticas y la posición geográfica de la estación sinóptica afectan la forma y la característica del espectro de la temperatura. la multifractalidad es más fuerte en la región subecuatorial que en la sudanesa. se encontró que la mayor fuente de multifractalidad en las diversas temperaturas consiste en una función probabilística de colas pesadas. sin embargo, correlaciones de gran autonomía también se corresponden con las características multifractales. excepto la estación sinóptica de kandi, el espectro inclina a la izquierda la asimetría estadística. los hallazgos señalan la utilidad del análisis no lineal en la investigación climática debido a las interacciones complejas entre los procesos naturales. este estudio permite comprender los mecanismos que manejan la dinámica de las series temporales de la temperatura en benin. resumen record manuscript received: 14/07/2016 accepted for publication: 28/06/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 365-370 m. agbazo1*, g. koto n’gobi2, e. alamou3, b. kounouhewa2, a. afouda3, & n. kounkonnou1 1international chair in physics mathematics and applications (cipma-chair unesco), university of abomey-calavi, benin bp: 526 uac benin 2laboratoire de physique du rayonnement, university of abomey-calavi, bp: 526 uac benin 3applied hydrology laboratory (lha), university of abomey-calavi, bp: 526 uac benin * corresponding author: agbazomedard@yahoo.fr m et eo r o lo g y 366 m. agbazo, g. koto n’gobi, e. alamou, b. kounouhewa, a. afouda, & n. kounkonnou introduction according to several studies, linear statistical methods are insufficient for a complete analysis of meteorological parameters time series. researchers have explained these findings by the fact that the atmosphere and the natural processes taking place in it are a complex natural system (kantz and schreiber, 2004; peters and neelin, 2006; vassoler and zebende, 2012; shi, 2014). therefore, meteorological parameters are characterized by a high degree of nonlinearity, non-stationarity and complexity (he, 2015; agbazo et al., 2019a; philippopoulos et al., 2019; kalamaras et al., 2017, 2019; jiang et al., 2016; burgueño et al., 2014 and dong et al., 2016). in this context, multifractal methods are suitable to analyze processes that obey to nonlinearity characteristics. nowadays, the multifractal detrended fluctuation analysis (mfdfa) introduced by kantelhardt et al. (kantelhardt et al., 2002) has been used widely to examine the intrinsic behavior of meteorological parameters in order to explain climate change, to assess climate change impact by improved methods and to do their prediction (kalamaras et al., 2019). for example, recently, mfdfa has been applied to time series of: precipitation by efstathiou and varotsos (2012); agbazo et al. (2019a); temperature by kalamaras et al. (2017); kalamaras et al. (2019); jiang et al. (2016); burgueño et al. (2014); dong et al. (2016); wind speed by kavasseri and nagarajan (2005); feng et al. (2009), and many others. the authors found that mfdfa methods can help reveal some properties, which could not be detected by linear methods (kantelhardt et al., 2002; kalamaras et al., 2017, 2019; philippopoulos et al., 2019; jiang et al., 2016). variations in temperature and precipitations’ records are important indicators of climate change (jiang, 2017). agbazo et al. (2019a) have studied the fractal analysis of the long-term memory in precipitation in benin (west africa). they have shown, (i) the existence of positive long-term memory characteristic in the rainfall of area and found that (ii) fractal theory reveals benin climatic characteristics better. therefore, in order to generalise the differences in the dynamics of meteorological processes and to assess climate impacts, it is important to compare multifractal properties of temperature in various climate conditions. however, long-term memory of temperature has not yet been studied in a fractal framework in benin, one of the west african countries affected by the effects of climate change (ipcc, 2007). thus, the aim of this work is to analyse spatial and time variation of multifractal properties of daily temperature time series observed in benin synoptic stations in order to examine the basic features of their multifractal behavior. materials and methods materials site description the study covers all the synoptic stations of benin (figure 1), the geographical positions of which are presented in table 1 below. according to boko et al. (1988), benin is characterized from the south to the north by two main climatic zones in which synoptic stations are located: cotonou, bohicon, and save are located in the subequatorial region where march is the hottest month (~26°c), while august is the coldest month (~24°c). the daily and annual thermal amplitudes are, respectively, ~10°c and ~5°c. the relative humidity ranges between 70% and 95% because of the proximity to the atlantic ocean. the subequatorial climate has four seasons in a year: a long rainy season (april to july) followed by a short dry season (august to september) and a short rainy season (october to november) followed by long dry season (december to march). however, the stations of parakou, kandi, and natitingou are located in sudanian region in the northern part of the country. the daily mean of air temperatures in natitingou, parakou, and kandi are, respectively, ~25°c, ~27°c, and ~35°c. table 1. geographical coordinates of synoptic stations stations latitude (°n) longitude (°e) altitude (m) cotonou 6.21 2.23 3.9 bohicon 7.10 2.03 166.00 save 7.59 2.26 198.51 parakou 9.21 2.37 392.00 natitingou 10.19 1.23 460.00 kandi 11.08 2.52 289.7 data records data were provided by the agency for aerial navigation safety in africa and madagascar (asecna). daily temperature data from benin synoptic stations: cotonou (co), bohicon (bo), save (sa), parakou (pa), kandi (ka) and natitingou (na) are used for the period from 1967 to 2012. figure 2 shows the mustache boxes of temperature. the figure shows that the smallest and largest values of minimum temperature are observed respectively in save (~ 3°c) and cotonou (~ 39°c). while, the smallest and largest values of maximum temperature are observed respectively in parakou and cotonou. the temperature distribution is asymmetrical. methods description of mfdfa method to investigate multifractal characteristics of non-stationary, non-linear, natural and complex meteorological phenomenon time series (shen et al., 2015, 2017; wan et al., 2016; jiang et al., 2017), the mfdfa method was proposed figure 1. study sites location: (a) benin’s location in west africa, (b) synoptic stations’ location in benin. 367multifractal behaviors of daily temperature time series observed over benin synoptic stations (west africa) if the time series follow the power law, then we can obtain the scaling function: f s sq h q( ) ∝ ( )� (4) where h(q) is the generalized hurst scaling function (kantelhardt et al., 2002; telesca et al., 2004). for monofractal time series, h(q) is independent of q, whereas, for a multifractal time series, h(q) varies with q. to be specific, the scaling exponent h(2) provides information about the average fluctuation of the series. the series can be categorized into one of the following three types depending on the h(2) values. these are: (i) 0 < h(2) < 0.5 for an anti-persistent type long range correlated process (negative long-term memory) where large values (compared to the average) are more likely followed by small values and vice versa, (ii) h(2) = 0.5 for an entirely random uncorrelated distribution, and (iii) 0.5 < h(2) < 1 for a persistent and long range correlated process (positive long-term memory) where large values are more likely to be followed by large values and vice versa. as király and jánosi (2005) suggest, the term “persistent” is used in the sense that an increasing trend in the past implies an increasing trend in the future. thus, it lightly differs from “persistence” in climatology, defined as the continuance of a specific pattern. this dependence is considered to be a characteristic of multifractal process (kantelhardt et al., 2002). in order to avoid a divergence of moments in the fat tails of the fluctuation distribution as mentioned by some authors (wang et al., 2013; ihlen et al., 2012), we restrict the order q within the range -5 ≤ q ≤ 5 as in (wang et al., 2013). multifractal spectrum in order to get more information about the temperature data and to characterize the strength of the multifractality, two functions and few parameters are deduced from h(q) via a legendre transform (feder et al.,1988; kantelhardt et al., 2002; shi et al., 2008; liu et al., 2014a, 2015; jiang et al., 2016). these functions are defined by following equations:    q h q q dh q dq f q h q ( ) = ( ) + ( ) ( ) = − ( )  +       1 (5)  q( ) and f ( ) are respectively singularity strength (or hölder exponent) and multifractal spectrum. these parameters are defined as followed: ∆  = −max min (6) and ∆f f fmin max= −( ) ( )  (7) where max and min are obtained from the relation f ( ) = 0 . f fmax0( ) = (8) if the observed time series is a single-scale fractal series, the function f () is a constant. f () generally, if the time series is a multiscale series, the function has a bell-like shape. the first important property of the spectrum is its width.  is multifractal spectrum width or spectrum width. the parameter  describes the inhomogeneity of the distribution of probability measured on the overall fractal structure, which has been identified as the intermittency and degree of multifractality (liu et al., 2014a). the smaller  is, the more uniformly distributed the fractal region is; while, the bigger  is, the more heterogeneous the probability distribution is. in others words, if a spectrum has a broad width,  is indicative of a strong multifractality (i.e., it has a ‘fine’ structure). if the width becomes smaller, then the time series tends to be a monofractal one. the spectrum width of a pure monofractal time series is equal to zero. f > 0 shows that the number of largest subsets is greater than the minimal one in probability measures, so, multiscale fractal spectrum shows a “left hook” shape; f < 0 shows a contrary condition, and the multiscale fractal spectrum is “right hook” shaped. designing 0 as the singularity strength with maximum spectrum, a small value of 0 means that the underlying process “loses fine-structure”; while a large value of 0 ensures larger complexity. by kantelhardt et al. (2002). it can be briefly described according to shen et al. (2017), as followed: let x t nt | = …{ }1 2, , , be an original time series of n equidistant measurements to which the procedure of the mfdfa method is applied. first, a new series named “profile” is determined as followed: y k x x t k t( ) = −( ) = ∑ 1 (1) where x is the mean value of xt and k = 1, 2, …, n. the profile is then divided into n = int(n/s) equal-sized nonoverlapping windows with a length s. since n is not the integral multiple of s in most cases, there might be a short part at the end of the profile that remains uncovered. to take full account of the series, the same procedure can be repeated starting from the end of the series. hence, we obtain 2ns segments altogether. we then calculate the variance of each window as follows: f s s y y for n i s s i m s 2 1 1 21 1 2     , , , , ( ) = − ( )     = … = −( ) +  ( )∑ i 11 1 2 1 2 s y y for n n i s n n s i m s s = − −( ) +  ( )∑ − ( )    = +   i ( ),.., ss          (2) it should be noted that linear (i.e., polynomial order m=1), quadratic (i.e., m=2), cubic (i.e., m=3), or higher-order polynomials y m ( ) ( )i can be used to fit the local trend. in mfdfam, possible m order trends are eliminated in the profile. the order of the polynomial has been found not to alter the results, with the order varying from 2 to 5 (koscielny-bundea et al., 2006). in this study, a third-order polynomial has been adopted as shown by agbazo et al. (2019a) in benin for the precipitation time series. by averaging over all windows, we obtain the fluctuation as follows: f s n f s for q exp n q s n q q s s ( ) = ( )           ≠ = ∑12 0 1 4 1 2 2 2 1  , / /   = ∑ ( )          =           1 2 2 0 ns f s for qln , (3) figure 2. mustache boxes of temperature data observed at each synoptic station from 1967 to 2012. cotonou (co), bohicon (bo), save (sa), parakou (pa), kandi (ka) and natitingou (na). 368 m. agbazo, g. koto n’gobi, e. alamou, b. kounouhewa, a. afouda, & n. kounkonnou according to shimizu et al. (2006), a measure of the width of the multifractal spectrum can be obtained by fitting a second-order equation to the curve of the spectrum around 0 f a b c    ( ) = −( ) − −( ) +0 2 0 (9) among the three constants a, b, c, the most important is b, which is an asymmetry parameter. coefficient b indicates the asymmetry of the spectrum. according to burgueño et al. (2014) and kalamaras et al. (2019), when b being zero for a symmetric spectrum, the shape of the spectrum is symmetrical. for positive b values, the spectrum is right-skewed and for negative left-skewed (burgueño et al., 2014). a right-skewed spectrum is related to relatively strongly weighted high fractal exponents (with “fine-structure”), while a leftskewed spectrum is indicative of low fractal exponents (a more ‘regular’ time series). c is an additive constant equal to 1. origins of multifractality generally, there are two major types of sources for multifractality in time series: (a) different temporal correlations for small and large fluctuations and (b) a fat-tailed probability distribution for the values of the time series (rak and zieba 2015). the main methods to find the contributions of the two sources of multifractality are the shuffling procedure and the surrogating procedure, respectively (kwapien et al., 2005). indeed, to test the first source of multifractality, we randomly shuffle the series to remove any temporal correlations. thus, if no multifractal feature remains after we conduct the shuffling procedure on the original multifractal series, we can conclude that long-range correlation dominates the multifractality in the original series. the shuffling procedure consists in generating a random permutation of the array elements of time series. in contrast, surrogate data is used to check whether the multifractality comes from fat-tailed probability distribution, because it can eliminate any sort of non-linearities in original series and weaken the nongaussianity of the distributions. in this paper, we use the common method of amplitude adjusted fourier transform (aaft) developed in (theiler et al., 1992; dong et al., 2016; wu et al., 2018) to obtain surrogate data. aaft can be summarized according to dong et al. (2016) as follows: 1) a discrete fourier transform of the original series is conducted, 2) the discrete fourier transform of the data is multiplied by random phases, and 3) an inverse fourier transform is performed to generate a phase-randomized surrogate. to further compare the contributions of multifractality from the two sources, the corresponding multifractal spectrum width is computed for the original series, shuffled series and surrogate series. to better compare the impact of both types of multifractality in time, the following computations are done: (1) the ratio to , which can reflect the contribution of the long-range correlations’ multifractal strength to the whole multifractal strength (shen et al., 2016); (2) the ratio ∆shuf to , which indicates the proportion of the fat-tailed probability distribution multifractality. results and discussion the fluctuation function log f sq ( )( ) versus log s( ) of the original temperature time series is plotted for different statistical moments q= (-5, -2, 2, 5) and shown in figure 3. it is observed that the linear relationship between the mfdfa fluctuation factor fq(s) and the time scale s is obvious, which implies the existence of power-law relationship. in addition, a phenomenon appears in some of the relationships shown in figure 3: the regression lines with different orders of q tend to converge, which indicates the multifractal nature of the original series of temperature records. moreover, the scaling behavior of fq(s) (i.e. slope) for positive and negative q are not the same at each synoptic station. this behavior reveals the existence of different degrees of multifractality at benin synoptic stations. in order to verify if these apparently converging lines are multifractal or monofractal, the plot of the generalized hurst exponent h(q) versus q obtained at each synoptic station for original temperature times series are shown in figure 4. to further compare the contributions of multifractality from the two sources, the corresponding multifractal spectrum f () is computed. in figure 3. log-log plots of the mfdfa fluctuation factor fq(s) versus the time scale s for specific moments q for the temperature time series observed at (a) cotonou, (b) bohicon, (c) save, (d) parakou, (e) natitingou and (f) kandi synoptic stations. figure 4. generalized hurst exponent h(q) as a function of q for the original temperature at all synoptic stations it is observed that multifractal behavior for temperature at each synoptic station is clearly shown in figure 4 because all the curves have obvious negative slopes. to be specific, all the h(2) values are greater than 0.5. one can conclude that the original temperature series is nonstationary signals with long range power-law correlations (shi, 2014). therefore, the fluctuations of the temperature time series are positively correlated in a power-law fashion. 369multifractal behaviors of daily temperature time series observed over benin synoptic stations (west africa) figure 5 the multifractal spectrum is presented as a function of the singularity strength  for the original series, shuffled series and surrogate series. in all cases, the spectrum shape is a humped curve and it is obviously unsymmetrical about its axis. moreover, the spectrum has a left truncation and a long right tail. this is attributed to multifractal structure that is insensitive to the local fluctuations with large magnitudes. the south-north distributions of the main multifractal parameters are presented in figure 6. it is observed in figure 6a that except cotonou and bohicon stations, 0 values increase with the station’s latitude. however, highest and lowest values are respectively observed at cotonou and save stations, which implies that the underlying mechanism governing daily temperatures has larger complexity at cotonou than save. figure 6a reveals that spectrum width  increases with the latitude in subequatorial and sudanian regions. however,  values in the subequatorial region are systematically greater than those of sudanian region. these findings indicate that in sudanian region the fractal range is more uniformly distributed; while in subequatorial region the probability distribution is more heterogeneous. in others words, the multifractality is stronger in subequatorial region. the highest value of  is obtained at parakou station, which implies a stochastic dynamic character and strong fluctuations of the original series at this station as compared to the others (xue et al., 2015). the ratio ∆surr to , and ratio ∆shuf to  are presented in figure 6b. it is shown that no gradient is obvious in their south-north distribution. it is clearly observed that, whatever the station, the contribution of the longrange correlations’ multifractal strength to the whole multifractal strength is systematically greater than the proportion of the fat-tailed probability distribution multifractality. it can be noticed that parakou temperature series have the highest value of the ratio ∆surr to  among all the benin synoptic stations. regarding the spectral asymmetry spatial distribution (figure 6c), it is observed that except kandi, the spectrum asymmetry values are negative; therefore, the spectrum is left-skewed, which is indicative of low fractal exponents (a more ‘regular’ time series). kandi’s spectra are with higher spectrum asymmetry values, therefore, contain a greater number of high fractal components. as shown in figure 6d, f > 0 for the original, shuffled and surrogate temperature times series over all synoptic stations. these findings show that the number of largest subsets is greater than the minimal one in probability measures, so, multiscale fractal spectrum shows a “left hook” shape. however, f values are systematically different for original, shuffled and surrogate temperature times series. therefore, the source of multifractality in temperature depends on long-range temporal correlations and a fat-tailed probability distribution. however, both of them influence the multifractality. moreover, the multifractality in temperature depends on the geographical position and the type of climate of the studied station. the value of f is not systematically reduced after the original series is shuffled and surrogated. therefore, the shuffling and the surrogating procedure do not systematically imply the reduction in the degree of multifractality. conclusions in this paper, the multifractal behavior of daily temperature time series for benin synoptic stations from 1967 to 2012 is investigated by multifractal detrended fluctuation analysis (mfdfa) method. this is achieved through five multifractal parameters (hurst exponent, hölder exponent with maximum spectrum, spectrum asymmetry, spectrum width and singularity spectrum). the main and nontrivial findings obtained are the following: 1. temperature time series exhibit long-range temporal correlation, which cannot be fully described by a single scaling exponent. they are long-term positively correlated, meaning that an increase in temperature is more likely to be followed by another increase. 2. temperature time series have a multifractal nature; however, weather conditions and the geographic position strongly affect the strength of their multifractality. multifractality is stronger in subequatorial region. figure 5. singularity spectrum f () as a function of the singularity strength function  for the temperature of original, shuffled and surrogated series at (a) cotonou, (b) bohicon, (c) save, (d) parakou, (e) natitingou and (f) kandi synoptic stations. figure 6. south-north distribution of the main multifractal parameters obtained for the original, shuffled and surrogate temperature times series. (a) 0 and the spectrum width, ; (b) the ratio ∆surrogated to � ∆original and the ratio ∆shuffled to � ∆original; (c) spectrum asymmetry and (d) f values. 370 m. agbazo, g. koto n’gobi, e. alamou, b. kounouhewa, a. afouda, & n. kounkonnou 3. regarding the spectral asymmetry spatial distribution, it is observed that except in kandi, the spectral asymmetry values are negative, therefore the spectrum is left-skewed. 4. long-range temporal correlations for small and large fluctuations and a fat-tailed probability distribution are two major sources for multifractality of temperature; however, fat-tailed probability distribution contributes more to multifractality. future works will be focused on the investigation of the relationship between the predictability of the meteorological time series and their multifractal nature in different climatic conditions. acknowledgments we, the authors, would like to thank the agency for aerial navigation safety in africa and madagascar (asecna) for providing the data that we used to conduct this study. we also thank moudachirou language institute (mli) for reviewing the english structure in the text. references agbazo, m. n., koto n’gobi, g., alamou, e., kounouhewa, b. & afouda a. (2019). fractal analysis of the long-term memory in precipitation over benin (west africa). hindawi, advances in meteorology, 2019, 1353195. doi: https://doi.org/10.1155/2019/1353195. boko, m. (1988). climats et communautés rurales du bénin. rythmes climatiques et rythmes de développement. thèse d’état, paris iv, 608pp. bunde, a. & havlin, s. (1995). fractals in science. springer-verlag, heidelberg, 1995. burgueño, a., lana, x., serra, c. & martínez, m.d. (2014). daily extreme temperature multifractals in catalonia (ne spain). physics letters a, 378 (11), 874-885. dong, q., wang, y. & li, p. (2016). multifractal behavior of an air pollutant time series and the relevance to the predictability. environmental pollution, 222, 444-457. efstathiou, m. n. & varotsos, c. a. (2012). intrinsic properties of sahel precipitation anomalies and rainfall. theoretical and applied climatology, 109(3-4), 627-633. feder, j. (1988). fractals. plenum press, new york, 90pp. feng, t., fu, z. t., deng, x. & ma, j. y. (2009). a brief description to different multi-fractal behaviors of daily wind speed records over china. physics letter a, 373, 4134-41. ipcc. (2007). climate change: impacts, adaptation and vulnerability, contribution of working group ii to the fourth assessment report of the intergovernmental panel on climate change. cambridge university press, cambridge, uk, 2007. ihlen, e.a. (2012). introduction to multifractal detrended fluctuation analysis in matlab. frontiers in physiology, 3, 141-159. jiang, l., zhao, l. & zhao, z. (2017). on the difference of scaling properties for temperature and precipitation over china. advances in meteorology, 2017, article id 5761275. jiang, l., zhang, j., liu, x. &, li, f. (2016). multifractal scaling comparison of the air temperature and the surface temperature over china. physica a, 462, 783-792. jiang, l., li, n. & zhao, x. (2017). scaling behaviors of precipitation over china. theorical applied climatology, 128, 63-70. kantelhardt, j. w., zschiegner, s. a. & koscielny-bunde, e. (2002). multifractal detrended fluctuation analysis of nonstationary time series, physica a, 316, 87-114. kavasseri, r. g. & nagarajan, r. (2005). a multifractal description of wind speed records. chaos, solitons & fractals, 24, 165-173. kiraly, a. & janosi, i. m. (2005). detrended fluctuation analysis of daily temperature records: geographic dependence over australia. meteorology and atmospheric physics, 88(3-4), 119-128. koscielny-bundea, j., kantelhardt, w., braund, p., bundea, a. & havlin, s. (2006). long term persistence and multifractality of river runoff records: detrended fluctuation studies. journal of hydrology, 322, 120-137. kalamaras, n., philipppopoulos, k., deligiorgi, d., tzanis, c. g. & karvounis, g. (2017). multifractal scaling properties of daily air temperature time series. chaos, solitons & fractals, 98, 38-43. doi: 10.1016/j. chaos.2017.03.003. kalamaras, n., tzanis, c. g., deligiorgi, d., philippopoulos, k. & koutsogiannis, i. (2019). distribution of air temperature multifractal characteristics over greece. atmosphere, 10, 45. doi: 10.3390/ atmos10020045. kantz, h. & schreiber, t. (2004). nonlinear time series analysis. second edition, cambridge university press: cambridge, uk. kwapien, j., oswiecimka, p. & drozdz, s. (2005). components of multifractality in high frequency stock returns. physica a: statistical mechanics and its applications, 350, 466-474. liu, z., xu, j. & shi, k. (2014a). self-organized criticality of climate change. theoretical and applied climatology, 685-691. liu, z., xu, j., chen, z., nie, q. & wei, c. (2015). multifractal and long memory of humidity process in the tarim river basin. stochastic environmental research and risk assessment, 28, 1383-1400. peters, o. & neelin, j. d. (2006). critical phenomena in atmospheric precipitation. nature 2, 393-396. philippopoulos, k., kalamaras, n., tzanis, c. g., deligiorgi, d. & koutsogiannis, i. (2019). multifractal detrended fluctuation analysis of temperature reanalysis data over greece. atmosphere, 10, 336. doi: 10.3390/atmos10060336. rak, r. & zieba, p. (2015). multifractal flexibly detrended fluctuation analysis. acta physica polonica b, 46, 1925. shen, z., shi, j. & lei, y. (2017). comparison of the long-range climate memory in outgoing longwave radiation over the tibetan plateau and the indian monsoon region. advances in meteorology, 2017, 7637351. shen, c. h., li, c. l. & si, y. l. (2015). a detrended cross-correlation analysis of meteorological and api data in nanjing, china. physica a, 419, 417-428. shi, k. (2014). detrended cross-correlation analysis of temperature, rainfall, pm10 and ambient dioxins in hong kong. atmospheric environment, 97, 130-135. shi, k., liu, c. q., ai, n. s. & zhang, x. h. (2008). using three methods to investigate time–scaling properties in air pollution indexes time series. nonlinear analysis: real world applications, 9, 693-707. doi: https://doi.org/10.1016/j.nonrwa.2007.06.003 shimizu, y., thurner, s. & ehrenberger, k. (2002). multifractal spectra as a measure of complexity in human posture. fractals, 10,103-116. teiler, j., galdrikian, b., longtin, a., eubank, s. & farmer, d. j. (1992). using surrogate data to detect nonlinearity in time series. in: m. casdagli and s. eubank (eds). nonlinear model, forecast, pp. 163–188, addisonwesley, redwood city, ca, usa. telesca, l., balasco, m., colangelo, g., lapenna, v. & macchiato, m. (2004). investigating the multifractal properties of geoelectrical signals measured in southern italy. physics and chemistry of the earth, parts a/b/c, 29(4-9), 295-303. vassoler, r. t. & zebende, g. f. (2012). dcca cross-correlation coefficient apply in time series of air temperature and air relative humidity. physica a, 391, 2438-2443. wan, s., liu, q., zou, z. & he, w. (2016). nonlinearity and fractal properties of climate change during the past 500 years in northwestern china. discrete dynamics in nature and society, 2016, 4269431. wu, l., chen, l., ding, y. & zhao, t. (2018). testing for the source of multifractality in water level records. physica a, 508, 824-839. xue, y., pan, w., lu, w.-z. & he, h. d. (2015). multifractal nature of particulate matters (pms) in hong kong urban air. science of the total environment, 532, 744-751. saad bakkali.indd 123 analysis of phosphate deposit “disturbances” using the horizontal-gradient responses of resistivity data (oulad abdoun, morocco) saad bakkali earth sciences department faculty of sciences and techniques, abdelmalek essaadi university, tangier, morocco email: saad.bakkali@menara.ma resumen en la gran cuenca de oulad abdoun en marruecos se inició la explotación de un nuevo depósito de fosfatos para reemplazar al de grand daoui que se encuentra agotado. existen inclusiones estériles de caliche que dificultan la extracción de las rocas fosfatadas y que son difíciles de detectar. la resistividad de los caliches excede un valor de 200 ω.m contra 80 a 150 ω.m para la roca fosfatada. se llevó a cabo un proyecto de prospección eléctrica con un equipo schlumberger sobre una extensión de 50 hectáreas. se obtuvieron modelos del perfil geológico mediante análisis de gradiente horizontal del mapa de los “distorsiones” de fosfatos. se logró localizar modelar la localizacion de las inclusiones de caliche y cubicar las reservas de fosfatos de manera más confiable. palabras clave: prospección geofísica, resistividad, fosfatos, gradiente horizontal, marruecos. abstract in the great oulad abdoun basin exploitation of a new phosphate deposit was begun after the grand daoui horizon was exhausted. inclusions of sterile hardpan—so-called “disturbances” —are hard to detect and interfere with phosphate extraction. their resistivity is above 200 ω.m as against 80 to 150 ω.m for the phosphate-rich mineral. a schlumberger resistivity survey over an area of 50 hectares was carried out. models of the geology were successfully obtained from the analysis of the horizontal-gradient map of “disturbances”. a new field procedure was tested to estimate the depth of disturbances. phosphate reserves were better known. key words: geophysical surveys, resistivity, phosphate, horizontal-gradient, morocco. © 2005 esrj unibiblos. manuscript received august 2005 paper accepted september 2005 earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 123-131 124 saad bakkali introduction morocco is a major producer of phosphate, with an annual output of 19 million tons and reserves in excess of 35 billion cubic meters. this represents more than 75% of world reserves. resistivity surveys have been successfully used in the oulad abdoun phosphate basin in khouribga province (figure 1), which is about 120 km south of casablanca. the present survey was carried out in the sidi chennane deposit which a part of oulad abdoun basin, extending over some 800,000 hectares. the sidi chennane deposit is sedimentary stratiform and contains several distinct phosphate-bearing layers. these layers are found in contact with alternating layers of calcareous and argillaceous hardpan. in this field, extraction was begun after grand daoui deposit was exhausted. however, the new deposit contains many inclusions and lenses of extremely tough hardpan locally known as “dérangements” or disturbances, found throughout the phosphate-bearing sequence. the hardpan pockets are normally detected only at the time of drilling. they interfere with field operations and introduce a severe bias in the estimates of phosphate reserves. direct exploration methods such as well logging or surface geology are not particularly effective. however, the chemical changes which are detectable at the hardpan/phosphate rock interface produce an important resistivity contrast. other factors such as changes in lithofacies and clay content and consistence appear to account for some additional resistivity difference. it was found that normal phosphate-bearing rock has a resistivity of 80 to 150 ω.m while the hardpan typically features resistivity values of between 200 and 1000 ω.m. a pilot resistivity survey was performed over an area of 50 hectares. the objective of this experiment was to try and map and constrain the anomalous regions corresponding to hardpan. a resistivity map was expected to allow the electrical resistivity signals to be imaged in 3d. we used a schlumberger array with a span of 120 m designed to reach a depth of 40 m. the so-called disturbances appear at random so that the apparent resistivity map may be used by the operating personnel as a kind of radar to plan the sequence of field operations. we use traditional analytic analysis by horizontal-gradient to circumscribe the extension in-depth of the effects of the random disturbances, or any other resistivity anomalies that may be present. overview of the area of study the sidi chennane phosphate deposit is within the oulad abdoun basin about 33 km south-east of khouribga (figure 2). its boundaries are : west, meridian 372500 (lambert), south, meridian 22800 (lambert), east, highway rp22, and north, the outcrops of the basement of the phosphate-rock sequence. the climate of the phosphate plateau is essentially arid. rainfall is from november to may and is usually below 400 mm. vegetation is of sparse dwarf palm trees. rural population subsists on cattle ranching and seasonal agriculture in small villages, or douars. ground water is increasingly scarce. scattered wells depend on an aquifer in the turonian limestones at depths of 100 m or more, which is sealed by the senonian marls. figure 1. location of the area of study. 125 analysis of phosphate deposit “disturbances” using the horizontal-gradient responses of resistivity data (oulad abdoun, morocco) figure 2. main phosphate basins in morocco. geology the phosphate mineral was deposited over a long time window from maestrichtian (late cretaceous, about 80 ma), to lutetian (early eocene, 40 ma). however, deposition was irregular. some layers are missing. oulad abdoun basin occupies most of the phosphate plateau which is bounded toward the north by red outcrops of pre-cenomanian sediments forming an extension of the south edge of the central massif. the western boundary is the rhamna range, the beni amir plain is to the south and the upper atlas of beni mellal extends to the east. the geology of the study area is well understood (see figure 3 for stratigraphy). figure 3. general outline of the stratigraphy. t h e g e o l o g i c s e c t i o n rests unconformably on p a l e o z o i c s c h i s t s a n d quartzites. the basement is well located and the s e d i m e n t a r y c o v e r i s fairly thick (kchikach and hiyane, 1991). the u p p e r m o s t f o r m a t i o n s o f t h e m a e s t r i c h t i a n and eocene contain the phosphate-bearing strata which are 30 to 50 m thick. the earlier deposits, i.e. the lower 5 to 28 m, are clayey phosphates of maestrichtian age. the upper 20 to 30 m are less homogeneous. they are layered phosphate marls and sandstones with some limestones of eocene age. below the phosphate-bearing strata one finds up to 70 m of senonian marls and limestone marls; 20 to 60 m of turonian limestones; a cenomanian f o r m a t i o n o f a l t e r n a t e l y gypsum marls and limestone marls; and finally 10 to 60 m of red marls and mudstones of pre-cenomanian age. t h e d i s t u r b a n c e s m a y b e differentiated by size of the pocket or inclusion, type of material, hardness, clay content, or type of contact with the phosphate rock. two main types of disturbances are found. the first type is found throughout the mineral deposit : it appears to be a random mixture of limestones, marls, clays, cherts and low-grade phosphate with large amounts of cherty limestone. the second type is highly disturbed and lacks any dominant facies. it appears as an accumulation of low-grade phosphate limestone blocks with large nodules of chert, marl, some fragments of chert and phosphate rock. the latter type forms inclusions of 10 to more than 150 m and is the most abundant during mining operations (figure 4). 126 saad bakkali figure 4. adverse effects of « disturbances » on mining operations. the study area was selected for its representativity and the resistivity profiles were designed to contain both disturbed and enriched areas (kchikach and hiyane, 1991). the sections were also calibrated by using vertical electrical soundings. high values of apparent resistivity were encountered due to the presence of near-vertical faulting between areas of contrasting resistivity, and fault zones which may contain more or less highly conducting fault gouge. the gouge may contain gravel pockets or alluvial material in a clay matrix (kchikach et al., 2002). such anomalous sections are also classified as disturbances. apparent resistivity values in these profiles locally exceeded 200 ω.m. in order to locate and define the anomalous areas or disturbances, an electric current of intensity i was passed between electrodes a and b, and the voltage drop ∆v was measured between the potential electrodes m and n. the apparent resistivity is found from ρ app = k (∆v/i), where k is the geometric constant of the instrument which depends only on the distance between electrodes (gasmi et al., 2004). thus the ratio between i and ∆v yields the resistivity of the terrain. our schlumberger profiles set required the electrodes to be aligned and equidistant from their central point o so that mn<<ab (bakkali and bouyalaoui, 2004). the lateral inhomogeneities of the ground can be investigated by means of the apparent resistivity obtained from the survey. as the surface extension of the layers is displayed we may infer the presence or absence of any disturbances as well as any facies variations. our resistivity measurements were performed by means of a syscal2 resistivity meter by these pockets are found both in the underlying formation and in the upper members of the phosphate sequence, and as a result there are strong resistivity contrasts between the disturbances and the normal phosphate-bearing rock. these contrasts were confirmed in the test runs (figure 5). geophysical prospection could thus be based on prior evidence from field data. figure 5. a resistivity traverse over three « disturbances ». field procedures resistivity is an excellent parameter and marker for distinguishing between different types and degree of alteration of rocks. resistivity surveys have long been successfully used by geophysicists and engineering geologists and the procedures are well established. 127 analysis of phosphate deposit “disturbances” using the horizontal-gradient responses of resistivity data (oulad abdoun, morocco) brgm instruments using a rectangular array of 20 m x 5 m. in order to reach a mean depth of exploration of 40 m we carried out 51 traverses at a spacing of 20 m. there were 101 stations at 5 m distance for every traverse, which makes 5151 stations altogether in the survey (bakkali and bahi, 2005). methodology interpretation of resistivity anomalies is the process of inferring information on the position and composition of a target mineral body in the underground. in the present case the targets were essentially the inclusions called perturbations. the amplitude of an anomaly may be assumed to be proportional to the volume of a target body and to the resistivity contrast with the mother lode. if the body has the same resistivity as the mother lode no anomaly will be detected. apparent resistivity measurements are obtained from a harmonic potential v which fulfils laplace’s equation ∆v=0 in the surrounding space external to the body. the potential gradient falls off as 1/r2 (blakely, 1995). as a first approximation, the scalar potential in the neighborhood of the perturbations and within the sources complies with the equation ∆v=-2ρi δ(r), where δ(.) is the dirac delta, ρ is the resistivity in the anomalous region and i is the current at a point electrode in an elastic halfspace, i.e. the topographic surface of the study area (dey and morrinson, 1979). the apparent resistivity map which one obtains from such a survey is actually a map of discrete potentials on the free surface, and any major singularity in the apparent resistivities due to the presence of a perturbation will be due to the crossing from a “normal” into a “perturbed” area or vice versa. in other words, the apparent resistivity map may be considered a map of scalar potential differences assumed to be harmonic everywhere except over the perturbed areas. thus one assumes that the potential difference data may be processed by fourier analysis without any terrain correction (cooper, 2004). horizontal-gradient and downward prolongation process figure 6 shows the map of resistivity anomalies corresponding to “disturbed” phosphate zones. figure 7 is the corresponding map of the “disturbed” zones. this procedure enables us to define the sources. the standard procedure was basically fourier filters which enhance some features of the anomaly map at the expense of other features, thus enabling us to connect the mapped surface anomalies to depth (bakkali, 2006). figure 6. a map of apparent resistivity anomalies of the area of study. 128 saad bakkali figure 7. a map of the corresponding “disturbed” zones of the resistivity anomalies map. (figure 8) was calculated starting from a plan obtained by an adjustment by the method of least squares of a grid of 5 cells out of 5 cells centered on the point to determine. we note the good correlation between “disturbed” phosphate zones and the horizontal-gradient intensity) the downward prolongation of the horizontalgradient apparent resistivity map reveals the extension in-depth of the contrasting densities. the operator of prolongation in the spectral domain is e z u v2 2 2+r where is the depth of prolongation and o the frequency defined as u v2 2= +o ] g, where u and v are the frequencies in the eastern and northern directions (gunn, 1975). we have applied this process to the map of the horizontal-gradient apparent resistivity to estimate the extension in-depth of the contrasting of density between the disturbances and the normal phosphate-bearing rock. this singular procedure will enable us to map the extension of anomalic zones in-depth (bakkali, 2006). the data are imaged symmetrically in the space domain in order to minimize edge effects on the transformed map. the second-derivative operation is equivalent to high-pass filtering. it is useful for interpretation as it refines the response of the geophysical instrument, and it produces sharper contours. as derivation is sensitive to noise (cooper and cowan, 2003), this procedure enhances the expression of any structures and discontinuities. the magnitude of the horizontal-gradient is usually estimated by finite difference methods from values measured at gridded points on the apparent resistivity anomaly map. the magnitude is determined using the following equations: , , , x y x x y y x y app app app 2 2 2 2 2 2 2 2 = +t t t l ^ ^ ^ h h h with , x x y ax2 , ,app i j i j1 1 2 2 = -t t t+ p ^ h and , y x y ay2 , ,app i j i j1 1 2 2 = -t t t+ p ^ h where x is the eastern coordinate and y the northern coordinate. ,i jt is the pseudo-apparent resistivity defined at grid point (i,j). grid intervals in the x-direction and y-direction are axp and ayp respectively. maxima horizontal-gradient may be occur immediately over steep or vertical boundaries separating rock masses of contrasting densities. on the horizontal-gradient apparent resistivity map, lines drawn along ridges formed by enclosed high horizontal-gradient magnitudes correspond to these boundaries. the horizontal-gradient represents in this case an indicator of the level of variation of the contrast of density between the disturbances and the normal phosphate-bearing rock. the value of the intensity of the horizontal-gradient anomalies 129 analysis of phosphate deposit “disturbances” using the horizontal-gradient responses of resistivity data (oulad abdoun, morocco) otherwise some spurious high frequency effects may appear around the edges. once the horizontal-gradient resistivity map was filtered in the frequency domain the downwardprolonged corresponding map at a depth 5 to 50 m (figure 9) was restored back to the space domain by inverse fourier transformation. downward prolongation was carried out in steps of 10 meters which enabled us to sample the entire sequence of level of the contrasting of density between the disturbances and the normal phosphate-bearing rock. results and conclusions the resistivity map as obtained by the above procedure in the study area provided a direct image for an interpretation of the resistivity survey. we were able to identify the anomalies which turned out to be strongly correlated with the disturbances. we found that the disturbances as detected from surface measurements were distributed apparently at random as confirmed by the various maps of the downward prolongation of horizontal-gradient apparent resistivity intensity. these maps represent an effective indicator of the intensity level of “disturbance” in depth. the anomalous areas were neatly outlined. as mentioned above, the disturbances were outlined in depth down to 50 m. the contrasting of density between the disturbances and the normal phosphatebearing rock from 5 to 50 m depth attest that the disturbances are contrasted in-depth comparatively on the surface. the high intensities of the horizontal-gradient over anomalous areas are interpreted as reflecting the fact that the pockets of disturbed material tend to become more consolidated at depth. we find that the analytic prolongation procedure of the horizontal-gradient helps to better constrain the location of anomalous areas on the surface. the overall effect is that of scanning the anomalous bodies. we have described an analytical procedure to identify the anomalies of a specific problem in the phosphate mining industry. the results proved satisfying. data processing procedures as horizontal-gradient were found to be consistently useful and the corresponding prolonged maps may be used as auxiliary tools for decision making under field conditions. it was found that the maps of horizontal-gradient at different depths were particular useful to the surveyors for improving and constraining their estimates of phosphate reserves in the deposit. figure 8. horizontal-gradient apparent resistivity intensity map in ω.m-1 of figure 6 (contrast density level between the disturbances and the normal phosphate-bearing rock corresponds to contrast of nuances of gray color. 130 saad bakkali figure 9. downward prolongation of horizontal-gradient apparent resistivity intensity maps . 131 analysis of phosphate deposit “disturbances” using the horizontal-gradient responses of resistivity data (oulad abdoun, morocco) references • bakkali, s. (2006). a resistivity survey of phosphate deposits containing hardpan pockets in oulad abdoun, morocco, geofisica internacional, 45 (1), (under press). • bakkali, s. and l. bahi, (2005). cartographie des “dérangements” de séries phosphatées par mesures de résistivités électriques, journal des sciences pour l’ingénieur, j.s.p.i., 6, (under press). • bakkali, s. and j. bouyalaoui, (2004). prospection géophysique appliquée à l’évaluation des eaux souterraines de mediouna (tanger, maroc), journal des sciences pour l’ingénieur, j.s.p.i., 4, 13-22. • blakely, r.j., (1995). potential theory in gravity and magnetic applications, cambridge university press, pp441 • cooper, g., (2004). the stable downward continuation of potential field data, exploration geophysics, 35, 260265. • cooper, g. and d. cowan., (2003). the application of fractional calculus to potential field data, exploration geophysics, 34, 51-56. • dey a. and h.f morrinson, (1979). resistivity modelling for arbitrarily shaped two-dimensional structures, geophysical prospecting, 27 : 106-136. • gasmi, m., ben dhia., h.., andrieux., p. & amri, f., 2004. contribution de la prospection électrique à l’étude hydrogéologique des aquifères (tunisie méridionale), sécheresse, 15, 2, 201-8. • gunn, p.j., (1975). linear transformation of gravity and magnetics fields, geophysical prospecting, 23, 300312. • kchikach, a. and m. hiyane, (1991). apport de la géophysique à la détermination des dérangements à sidi chennane nord, publications ecole mohammadia d’ingénieurs, rabat, maroc. • kchikach, a., m. jaffal, t. aifa, and l. bahi, (2002). cartographie de corps stériles sous couverture quaternaire par méthode de résistivités électriques dans le gisement phosphaté de sidi chennane (maroc). comptes rendus. geosciences, 334, 379-386. based on actual land use in yancheng city, this paper establishes the ecological security of the land index system from the perspectives of economy, society and ecological environment. using the catastrophe theory and pressure-stateresponse (psr) model, the purpose of the research was to judge the degree of land ecological security in yancheng city from 2002 to 2011 and determine the requirements to ensure the sustainable development of this region. the results showed that over the ten years, the pressure, state, response and synthetic values of the land ecological security were evaluated as safe in 2003 and 2007 and at middle level in 2002 and for 2004 to 2006 and then dropped to insecure for 2008 to 2009 and very insecure for 2011 to 2012. the ecological security level and circumstance of land use in yancheng city has generally declined. during this period, the land use pattern of yancheng city was under tremendous pressure from the conflict between rapid urbanization and economic development, as well as conservation and rehabilitation of the eco-environment. this research shows that conditions for the land ecosystems of yancheng city are not optimistic and, as such, should draw the attention of responsible government departments. future policy options should aim to mitigate these problems through the control of population growth and the improvement to quality of life, protection of wetland and forest land, application of scientific concepts of development, coordination of economic development and land utility, and strengthening the control functions of land-use planning. de acuerdo con el uso actual del suelo en la ciudad de yancheng, en el este de china, este artículo establece la seguridad ecológica del sistema de tierras desde las perspectivas económica, social y ecológica. a partir de la teoría de las catástrofes y el modelo ambiental de presión-estado-respuesta (psr), el propósito de esta investigación fue determinar el grado de seguridad ecológica del suelo en la localidad de yancheng entre 2002 y 2011 y determinar las condiciones para asegurar el desarrollo sustentable de la región. los resultados muestran que sobre estos diez años los valores de presión, estado, respuesta y sintéticos fueron evaluados como seguros en 2003 y 2007, como nivel medio en 2002, entre 2004 y 2006, cayeron a inseguros en 2008 y 2009, y muy inseguros para 2011 y 2012. el nivel de seguridad y las condiciones de suelo en yancheng se han disminuido generalmente. durante este período, el patrón de uso de la tierra en la ciudad de yancheng estuvo bajo una gran presión por el conflicto entre la rápida urbanización y el desarrollo económico y la conservación y la rehabilitación ecoambiental. esta investigación muestra que las condiciones para los ecosistemas terrestres no son óptimas y que por lo tanto es necesario llamar la atención de los departamentos de gobierno responsables. las posibles políticas futuras deben enfocarse en mitigar estos problemas a través del control al crecimiento urbano y el mejoramiento de la calidad de vida, la protección de los humedales y los bosques, la aplicación de conceptos científicos al desarrollo, la coordinación del avance económico y el suelo, y el fortalecimiento de las funciones de control en la planeación del uso de la tierra. earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014): 181 187 abstract resumen key words: land ecological security evaluation, catastrophe theory, yancheng city. palabras clave: evaluación de la seguridad ecológica del suelo; teoría de las catástrofes; yancheng. record manuscript received: accepted for publication: land ecological security assessment for yancheng city based on catastrophe theory wang liang¹,2. bian zhengfu¹, chen hongquan2 ¹school of environment and spatial informatics, china university of mining and technology, xuzhou , china.e-mail:wangsantai123@163.com 2school of urban and resource environment, yancheng teachers university, yancheng , china ecology issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.44642 http://dx.doi.org/10.15446/esrj.v18n2.44642 182 wang liang, bian zhengfu, chen hongquan introduction land is the foundation of global development and ecological survival. land ecosystems play a very important role in social-economic progress at the regional, national and international scales, and land ecological security is related to regional sustainable development. currently, land ecosystems are under great stress with the intensification of agriculture, population growth, urbanization and industrialization, all of which impede human sustainable development. there is significant evidence that the land ecosystems of many regions have become highly stressed and dysfunctional due to continuous and excessive utilization and exploitation of land resources (mark 2006; honson and marvin 2009; wessels et al. 2004; garcı´a et al. 2008; su et al. 2011; salvati and bajocco 2011). aggravation of the ecological environment has caused the ecological security situation to become one of the most pressing problems that society has to face and attempt to solve. in this regard, there is a necessity to develop methods and indicators to accurately diagnose the state of land ecosystems in order to prompt ecological restoration, management and regulation. land ecological security (eco-security) assessment based on an analogy of the threat to survival is such a procedure. most of the currently used methods to evaluate the security of land ecosystems (fuzzy ahp, entropy method, matter element model) have disadvantages in subjectivity and complexity, especially associated with the weight determination procedure (shi et al. 2003; su et al. 2011). as a mathematical model characterized by dialectic and simple construction, catastrophe theory (ct) generally applies to systems that may respond to continuous changes in control variables through a discontinuous change from one equilibrium state to another. ct is naturally chosen to effectively model and analyze the discrete transition mechanism resulting from changes in the internal and external circumstances (xu et al. 2014). by combining catastrophe theory with fuzzy mathematics and only considering the relative importance of the indices, the catastrophe method avoids the subjectivity in weight decisions. it has therefore been used in many fields, including evaluation of rural information levels (zhang et al. 2013), business cycles (khademian 1988), agricultural insect pest-control (piyaratne et al. 2013) and environmental pollution assessment (wang et al. 2011). although the pressure-state-response (psr) model has been developed further to the dpsir, which includes ‘driving force’ and ‘impact’, the decision was made to use the psr model framework in this research, as explained later in the paper in the section on driving force and response. in recent years, yancheng city has faced great change as a result of the jiangsu coastal development to prompt urbanization and industrialization. with a huge population (the population of yancheng city in 2011 was 8.207×106 with an annual growth rate of 5.311%), this region’s land eco-security level has been in continuous decline. the change in the land eco-security state can be considered as a particular catastrophic behavior, a small and gradual change in the steady equilibrium state of a sub-system can rapidly cause the whole system to reach a crash state (su et al.2011). following this idea, this paper proposes a catastrophe model combined with the psr model framework for land ecosecurity assessment, using yancheng city, china, as a case study. the objective is to demonstrate the method and evaluate the degree of land ecological security in yancheng city from 2002 to 2011 as a tool to determine the requirements to ensure sustainable development of this region. study area and data study area yancheng city (32°34’~34°28’ n, 119°27’~120°54’ e), which has the largest land area (16, 972 km2) and the longest coast line in the jiangsu province, is located on the western side of the yellow sea in eastern china. the urbanization rate and gross domestic product (gdp) rate of increase for this region in 2011 was 55.9% and 12.8%, respectively. investment in tertiary industry accounted for 37.82% of gdp, while investment in environmental protection was only 2.09%. the land ecosystems of yancheng city have become highly stressed and dysfunctional due to this continuous, excessive utilization and exploitation of land resources. the trend for both gdp and the total ecological footprint in yancheng city from 2002 to 2011 has steadily increased, putting still greater pressure upon the ecological environment of the region (figure. 1). consequently, there is considerable need for methods and indicators to accurately diagnose the state of the land ecosystem and to prompt ecological restoration, management and regulation. 0 500 1000 1500 2000 2500 3000 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 year 0 500 1000 1500 2000 2500 per gdp($) total ecological footprint(100km2) figure1. changes in per gdp and ecological footprint in yancheng city from 2002 to 2011 data source the data used in this research are from the “statistical yearbook of yancheng city 2003-2012”, “statistical yearbook of jiangsu 2003-2012” and data supplied by the yancheng city prefecture land and resources bureau. some additional data were obtained from field surveys and questionnaires. all the index values are calculated based on the data collected from these sources. materials and methods catastrophe theory catastrophe theory was proposed in an attempt to rationally account for the phenomenon of discontinuous change in behaviors (outputs) resulting from continuous change in parameters (inputs) in a given system. in this section, a brief discussion of the basic assumptions and results of catastrophe theory is presented in a form useful for such applications (woodstock and poston 1974). let f: tk×tn t: a smooth (infinitely differentiable) function representing a dynamical system m. tk: the space of input variables (controls, parameters). tn: the space of output variables (responses, behaviors). the fundamental assumption is that m attempts to locally minimize f. given any such function f, if the point c∝tk is fixed, then a local potential function fc: t n→t is obtained. therefore, fc can be expressed as follows: v=v(x,u) where v: the potential function, x: response variables, and u: control variables. the critical points of the potential function fc form an equilibrium surface (zeeman 1976). the equation of that surface is obtained by calculating the first derivative of fc, f′c（x）= 0, and the singularities are obtained by calculating the second derivative of fc, f′′c（x）= 0. a bifurcation set of the catastrophe system is obtained by eliminating x between f′c（x）= 0 and f′′c（x）= 0. a normalization formula is derived by decomposing the bifurcation set. the values of x and all control variables in the normalization formula range between 0 and 1. these variables are called ‘‘catastrophe progression’’. the catastrophe progression of each control variable can be obtained from the initial membership function using a recursive algorithm subject to the normalization formula. supposing that the response variable is one dimensional, catastrophe models can be classified into four categories according to the dimension of control variables. summary descriptions of these models are given in table 1. 183land ecological security assessment for yancheng city based on catastrophe theory table 1. description of catastrophe models model types control variables function bifurcation set normalization formula fold 1 v(x)=x3+u1x u1=-3x 2 xu1= u cusp 2 v(x)=x4+u1x 2+u2x u1=-6x 2 , u2=8x 3 xu1= 1u ,xu2= 3 u swallowtail 3 v(x)= 5 1 x5+ 3 1 u1x3+ 2 1 u2x2+u3x u1=-6x 2, u2=8x 3, u3=-3x 4 xu1= 1u ,xu2= 3 u xu3= 4 u butterfly 4 v(x)= 6 1 x6+ 4 1 u1x4+ 3 1 u2x3+ 2 1 u3x2+u4x u1=-10x 2,u2=20x 3,u3=-15x 4,u4=4x 5 xu1= 1u ,xu2= 3 2u xu3= 4 u ,xu4= 5 u the psr model the psr model is based on the concept of causality: human activities exert pressures on the environment and change the quality and quantity of natural resources, which then lead to responses in human behavior. three categories of indicators are distinguished. first, eco-environmental pressure indicators describe pressures on the environment by human activities and climate change. second, eco-environmental state indicators describe the status quo of the natural environment and ecosystem function. third, societal response indicators show the degree to which society responds to eco-environmental changes and concerns. this could be the number and type of measures taken, the efforts of implementing measures or the effectiveness of those measures. indices selection and standardization based on previous studies and the framework, as well as the principles explained above, this paper generates a set of assessment indices. initially, a set of 55 indices was developed. subsequently, a three-round delphi process (linstone and turoff 1975) was established, from which 28 indices were selected that favored the consensus needed to validate the analysis. experts with skills in the appropriate fields of study evaluated the set of 28 indices for relevance to the assessment. after performing principal component analysis to reduce data dimensionality, a total of 18 indices were generated (table 2). table 2. land ecological security indices framework of yancheng city target layer project layer first code factor second code index layer properties land ecological security pressure a1 social economy b1 c1 population growth rate（‰） c2 population density（person/km 2) c3 urbanization（%） c4 per capita gdp（yuan/person） + environ ment b2 c5 the usage of fertilizer（kg/hm 2） c6 water resource（m 3/person） + c7 dwelling areas（m 2/ person ） state a2 social economy b3 c8 gdp growth（%） + c9 cultivation index（%） + c10 rural per capita net income（yuan/person） + c11 city per capita net income （yuan/person） + environ ment b4 c12 per farmland（hm 2/ person ） + c13 per public greenland（m 2/ person ） + c14 proportion of irrigated farmland（%） + response a3 method b5 c15 proportion of tertiary industry（%） + c16industrial solid wastes utilized proportion （%） + c17industrial wastewater up to the discharge standards proportion (%) （%） + c18 proportion of environment invest （%） + 184 wang liang, bian zhengfu, chen hongquan note: development of a tertiary industry can be an effective method to alleviate deterioration of land ecosystems, so it is feasible to put c15 in the project layer of response. as different indices have no comparability because they differ completely in dimensions, it is necessary to convert original data to a dimensionless form. the evaluating indicators can be divided into positive term targets and negative term targets. the model takes the following two forms. 1) standardization of positive effect indices: for some indices, the higher their value, the greater the positive effect they will bring to land ecological security, which means this type of index presents much less risk to ecological security. their standardization method is: 2) standardization of negative effect indices: for other indices, the higher their values, the greater the negative effect they will bring to land ecological security, which means this type of index presents much more risk to ecological security. their standardization method is: in the formula, zij is the i standardized value of the j index included in the assessment; xij is the i th actual value; max is the biggest value of this index; and min is the minimum value of this index. application of catastrophe theory the catastrophe assessment index system can be divided into hierarchical sub-systems. if the index at a higher level (response variable) contains two lower level indices (control variable), it can be assumed as a cusp system. the relative importance of these two control variables should be determined (u1, important; u2, less important), and the control variable can then be obtained from the membership function using a recursive algorithm subject to the normalization formula. similarly, when the index at a higher level contains one, three or four lower level indices, it can be, respectively calculated based on fold, swallowtail and butterfly membership functions. the catastrophe assessment model for yancheng city was therefore developed using such an approach (su et al. 2011). the catastrophe model for land eco-security assessment in yancheng city is shown in figure. 2. figure 2 . catastrophe model for land eco-security assessment of yancheng city. score transformation the synthetic values of catastrophe assessment are generally high, and the differences are not obvious (poston and ian 1978). this can be attributed to the fact that catastrophe progression is calculated based on the normalization formula (su et al. 2003). therefore, it is difficult to determine the actual secure level directly using the results obtained by the catastrophe assessment. usually, the synthetic values of multi-attribute assessment are divided into five grades using an equality distribution function (xiong et al. 2007). accordingly, the land eco-security level can be divided into five grades: 0.2 (very insecure), 0.4 (insecure), 0.6 (middle), 0.8 (secure) and 1.0 (very secure). the problem is how to find a way to transform the results obtained by the catastrophe assessment into the ordinary-used synthetic values. the method for score transformation used in this paper is described as follows: suppose the relative membership degree for all indices equals n, then the relative membership degree for higher level indices should also equal n. consequently, the synthetic membership degree can be obtained by applying a suitable catastrophe model. by virtue of this method, the catastrophe progression value for each secure grade was calculated (table 3). table 3. corresponding values between the results of the catastrophe model and ordinary-used values at different security levels security level land eco-security level obtained by catastrophe model ordinary values pressure state response synthetic very secure >0.973 >0.975 >0.985 >0.991 >0.8 secure 0.973-0.948 0.975-0.934 0.985-0.926 0.991-0.977 0.6-0.8 middle secure 0.948-0.897 0.934-0.880 0.926-0.872 0.977-0.957 0.4-0.6 insecure 0.897-0.814 0.880-0.831 0.872-0.839 0.957-0.933 0.2-0.4 very insecure <0.814 <0.831 <0.839 <0.933 <0.2 method demonstration this section is provided to demonstrate this method in terms of a simple example using the 2002 data from the study area (table 4). 185land ecological security assessment for yancheng city based on catastrophe theory table 4. land eco-security value of yancheng city in 2002 index layer primary data standard application of catastrophe theory c1 population growth rate（‰） 1.82 0.343 343.0 0.791 b 1 791.0 0.922 a 1 0.967 c2 population density（person/km 2) 531.01 0.902 3 902.0 c3 urbanization（%） 35 0.602 4 602.0 c4 per capita gdp（yuan/person） 8464 0.212 5 212.0 c5 the usage of fertilizer（kg/hm 2） 399.69 0.538 538.0 0.872 b 2 3 872.0 c6 water resource（m 3/person） 18.69 1 3 1 c7 dwelling areas（m 2/ person） 29.3 0.606 4 606.0 c8 gdp growth（%） 11.4 0.973 973.0 0.817 b 3 817.0 0.872 a 2 c9 cultivation index（%） 162.77 0.454 3 454.0 c10 rural per capita net income（yuan/person） 3867 0.266 4 266.0 c11 city per capita net income （yuan/person） 9015 0.317 5 317.0 c12 per farmland（hm 2/ person） 0.097 0.065 065.0 0.591 b 4 3 591.0 c13 per public greenland（m 2/ person） 6.5 0.893 3 893.0 c14 proportion of irrigated farmland（%） 46.45 0.095 4 095.0 c15 proportion of tertiary industry（%） 32.4 0.215 215.0 0.784 b 5 784.0 0.885 c16industrial solid wastes utilized proportion （%） 92.7 0.445 3 445.0 c17industrial wastewater up to the discharge standards proportion (%) （%） 72.30 0.870 4 870.0 c18 proportion of environment invest （%） 2.13 0.751 5 715.0 results by using the model described above with the statistical data and other data from 2003 to 2012, the land eco-security of yancheng city was calculated (figure. 3). 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 pressure state response synthetic0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9 1 figure3. the land eco-security level of yancheng city between 2002 and 2011. it can be seen that the land eco-security level for ‘pressure’ was evaluated as secure in 2003 and 2007, a middle level in 2002, 2004, 2005 and 2006, insecure in 2008 and 2009, and very insecure in 2010 and 2011. the maximum value of the land eco-security level for ‘pressure’ was 0.951 in 2003, and the minimum value was 0.192 in 2011. in the past decade, rapid urbanization, increasing from 35% in 2002 to 55.9% in 2011, accompanied by population growth (increasing by two hundred thousand between 2002 and 2011), has been a fundamental cause for the current land ecological problems in yancheng city. with a high population density and increasing wealth, demands for public infrastructure (roads, water facilities, and utilities), housing, industrial and commercial growth and additional construction projects start to appear as ‘‘rural sprawl’’ (mann 2009). statistical data provide evidence that urbanization has greatly stimulated the up-scaling and expansion of the road systems in yancheng city. the great opportunities available in the city have also attracted a significant influx of people into the city, which exerts further pressure on the land ecosystems. in addition, overuse of chemical fertilizers, pesticides, and plastic film in croplands has led to land pollution and contributed to the degradation of land ecosystems in yancheng city. similar to pressure, the land security level with regard to the ‘state’ index also generally decreased over the period between 2002 and 2011. the ‘state’ index of land eco-security was considered as very insecure in 2010 and 2011. experience in other parts of the world suggests that intensive urbanization necessitates the transfer of significant areas of land to secondary and tertiary sectors (chan and shimou 1999). in many cases, farmland has to be given up for development, and large areas of high-quality farmland are lost to urban build-up (long et al. 2009). this is exactly the case in yancheng city. during the studied 10 years, per capita farmland and effectively irrigated land significantly declined, posing considerable threat to regional food security. the increased concentration of factories with low treatment efficiency for waste gas, water and other wastes has also contributed to the insecure state. according to the statistics, the load 186 wang liang, bian zhengfu, chen hongquan of industrial wastewater increased from 28,800 ton/km2 in 2002 to 35,400 ton/km2 in 2011. the case of industrial solid wastes was even more serious, with the quantity doubling during the study period. over the decade, there is a great difference of quality change between various land types. from 2002 to 2011, the area of farmland decreased the most in the region by 3,679.77 km2, while forest lands declined by 61.21 km2. all these phenomena indicate that the ecological service function of the land ecosystem in yancheng city has been deteriorating rapidly. if appropriate and timely measures are not adopted, the land ecosystem will continue to degrade. as for ‘response’, the security level was evaluated as secure in the initial 2 years. however, insecure and very insecure levels were identified for the following years. while it is true that yancheng city has invested in a number of environmental protection projects and the proportion of scientific researchers has been increasing, the industrial pollution issues have not been effectively tackled. performance evaluation by local officials is mainly associated with economic growth, and environmental protection is not listed as one of the criteria, unless serious environmental disasters are made public and raise concerns. industry is the major sector for the local labor market and the main financial source for the local government. most officials ignore such industrial activities, and although pollution levy systems have been introduced in the city, they are largely ineffective, as the levies are too low to give polluters an incentive to reduce emissions. similarly, water pollution fees are small relative to the additional costs of pollution control (sinkule and ortolano 1995). these responses have led to the continuing decline of the land eco-security level. over the studied 10 years, the synthetic land eco-security was evaluated as very secure in the initial 2 years, middle and insecure from 2004 to 2009, and very insecure in the last 2 years. these results show that the land ecosystem conditions are not optimistic for yancheng city and should be brought to the attention of policy makers and the public. at present, serious problems, such as rapidly decreasing farmland with a significant reduction in amount of farmland per person, the high demand for construction land, prominent contradiction between land resources and population growth, land utilization increasing yearly, over-exploitation of land resources and declining ecological land, especially forest and natural reserve land, exert huge pressure on the regional environment. land ecosystem conditions are equally pessimistic for shanghai (su et al.2011), liaoning province (han 2009) and hubei province (xu 2012). discussion and conclusions the catastrophe model integrated multiple assessment indices of land ecological security according to the inherent contradictions and relative importance of indices without calculating weights (thom 1975; vadrevu et al 2008; weidlich and huebner 2008; zeeman 1976). the aim of this study on the land eco-security assessment was to develop a means of quantifying the state of land ecosystems over time. the land ecosystems of yancheng city generally remained at or under the middle grade of eco-security across the 10 years between 2002 and 2011. during the 10 year period, the land ecosystems sustained intensified pressure, presented a worsened security state and received reduced positive responses from society. these have led to a significant decline in the synthetic land eco-security. the results presented here, showing the changes over time and the factors that have led to the present state, demonstrate a method that can be used to predict and manage land ecosystems with considerable utility. it is noted that this assessment relied mainly on the available statistical data set, and the evaluated eco-security levels, whether secure or insecure, are thus a relative concept. however, although this approach lacks absolute accuracy, the study clearly demonstrates the changes in land ecosystem conditions over time and should assist officials and citizens in resolving issues regarding regional land eco-security improvements. the land eco-security index described here should therefore be a useful tool for policy makers and land managers to develop measures that strengthen land ecosystem protection and promote ecological reconstruction. further policy options that should be considered to mitigate the problems more thoroughly include control of population growth and in-migration; protection of wetland and forest land; application of scientific concepts of development; coordination of economic development and land utility; and strengthening the control functions of land use planning. acknowledgments the study was supported by the national natural science foundation of china (41071083), and the priority academic program development of jiangsu higher education institutions. references chan rck, shimou y.1999, urbanization and sustainable metropolis and development in china: patterns, problems and prospects. geojournal 49:269–277. garcía m, oyonarte c, villagarcı´a l, contreras s, domingo f, puigdefa´bregas j.2008, monitoring land degradation risk using aster data: the non-evaporative fraction as an indicator of ecosystem function. remote sens environ 112:3720–3736. han tf .2009,assessment and early warning of land use change ecological security in liaoning province. doctoral dissertation, northeastern university. xu bs. 2012,study on landscape pattern evolution of land use and ecological security assessment in jianghan plain. doctoral dissertation, huazhong normal university. honson m, marvin s.2009,urban ecological security: a new urban paradigm? int j urban reg res 33:193–215. khademian h.1988,catastrophe theory: an application to a business cycle model.mathl comput modelling 10(4):257–262. linstone ha, turoff m .1975, the delphi method: techniques and applications.addison-wesley publishing company, reading, ma. long h, liu y, wu x, dong g .2009, spatio-temporal dynamic patterns of farmland and rural settlements in su-xi-chang region: implications for building a new countryside in coastal china. land use policy 26:322–333. mann f .2009, institutional causes of urban and rural sprawl in switzerland. land use policy 26:919–924. mark ls .2006,threat level green: conceding ecology for security in eastern europe and the former soviet union. global environ change16:400–422. piyaratne mkdk, zhao hy, meng qx.2013,aphidsim: a population dynamics model for wheat aphids based on swallowtail catastrophe theory. ecological modelling 253:9–16. poston t, ian s .1978,catastrophe theory and application. lord, pitman. salvati l, bajocco s .2011,land sensitivity to desertification across italy: past, present, and future. appl geogr 31:223–231. shi y, liu y, he j .2003,further study on some questions of catastrophe evaluation method. eng j wuhan univ 36:132–136. sinkule b, ortolano l .1995,implementing environmental policy in china. praeger, westport, london. su s, li d, xiang y .2011,assessing land ecological security in shanghai (china) based on catastrophe theory. stoch environ res risk assess 25:737–746. thom r .1975,structural stability and morphogenesis. w.a. benjamin inc, reading, ma. vadrevu kp, cardina j, hitzhusen f.2008,case study of an integrated framework for quantifying agro-ecosystem health. ecosystems1:283–306. wang wj, liu sl, zhang ss, chen jw .2011,assessment of a model of pollution disaster in near-shore coastal waters based on catastrophe theory. ecological modelling 222:307–312. weidlich w, huebner h .2008,dynamics of political opinion formation including catastrophe theory. j econ behav organ67:1–26. wessels kj, prince sd, frost pe, van zyl d.2004,assessing the effects of human-induced land degradation in the former homelands of northern south africa with a 1 km avhrr ndvi time-series. remote sens environ 91:47–67. 187land ecological security assessment for yancheng city based on catastrophe theory woodstock aer, poston t.1974,a geometrical study of the elementary catastrophes, springer lecture notes in math, vol. 373.springerverlag, new york. xiong y,zeng g,chen g .2007,combining ahp with gis in synthetic valuation of eco-environment quality-a case study of hunan province, china.ecol model 209:97–109. xu y, hu b, wu j, zhang jh.2014,nonlinear analysis of the cooperation of strategic alliances through stochastic catastrophe theory. physica a 400: 100–108. zeeman ec.1976,catastrophe theory. sci am234:65–83. zhang lx, liu x, li dl, fu zt.2013,evaluation of the rural informatization level in four chinese regions: a methodology based on catastrophe theory.mathematical and computer modelling 58:868–876. earth sciences research journal seismology earth sci. res. sj. vol. 16, no. 2 (december, 2012): 31 38 adaptive seismic ground roll attenuation using the double density dual tree discrete wavelet transform (dwt) method ali reza goudarzi1 and mohammad ali riahi2 1 institute of geophysics, tehran university, islamic republic of iran. e-mail: aligoudarzi@ut.ac.ir 2 institute of geophysics, tehran university, islamic republic of iran, phd in exploration geophysics (associate professor of geophysics). phone: +98 (21)61118219. e-mail: mariahi@ut.ac.ir abstract seismologists working on the reflection of seismic data have been dealing with noise and trying to attenuate noise as much as possible without damaging a particular signal. many methods have been used to remove noise types, each being partially effective. none of the transforms have been ideal and, even at the point of transform, noise becomes added to data. the double density dual tree (dddt) discrete wavelet transform (dwt) method was used in this paper to attenuate noise in pre-stack and post-stack seismic data. wavelet domain ground roll analysis (wdga) was used for finding ground roll energy in the wavelet domain; it is a substitute method for thresholding in seismic data processing. the dddtdwt method has four wavelets and two scaling functions. using this method, shot gather data decomposes to sub-scales and each scale contains coefficients related to a particular signal. the procedure involved applying distinct wavelets and scaling functions to data and then eliminating the coefficient of noise from inverse dddtdwt using the wdga technique. the inverse transform provided a section without noise; double density dwt and dual tree dwt methods were combined (i.e. each method having its own characteristics and advantages). resumen los sismólogos que trabajan en la reflexión de los datos sísmicos se han ocupado del ruido y en tratar de atenuarlo tanto como sea posible sin dañar una señal particular. muchos métodos se han utilizado para eliminar ciertos tipos de ruido, cada uno siendo parcialmente efectivo. ninguna de las transformadas han sido ideal e incluso en el punto de transformación, el ruido es añadido a los datos. el árbol de doble densidad dual (dddt) y el método de transformada de ondicula discreta (dwt) fueron los métodos utilizados en este trabajo para atenuar el ruido en los datos sísmicos la sección preapilda y postapilada. el análisis de ground roll en el dominio de la ondícula (wdga) se utilizó para encontrar su energía. este es un método sustituto de umbralización en procesamiento de datos sísmicos. el método dddtdwt tiene cuatro ondiculas y dos funciones escalares. usando este método, datos shot gather fueron descompuestos a subescalas, donde cada escala contiene coeficientes relacionados a una señal particular. el procedimiento consistió en aplicar distintas ondiculas y funciones escalares a los datos y posteriormente eliminar el coeficiente de ruido a partir de la inversión de dddtdwt utilizando la técnica wdga. la transformada inversa proporcionó una sección sin ruido; los métodos de doble densidad dwt y doble árbol dwt se combinaron (es decir, cada método tiene sus propias características y ventajas). palabras claves: sismica, ruido, transformada ondicula, disparo-receptores , sección apilada, wdga. keywords: seismic, noise, wavelet transform, shot gather, stack section, wdga. record manuscript received: 14/10/2011 accepted for publications: 06/05/2012 introduction this study presents a wavelet-transform-based method and wavelet domain ground roll analysis (wdga) technique for seismic ground roll reduction. the double density dual tree (dddt) discrete wavelet transform (dwt) was used with the wdga technique for seismic denoising. exploration seismology has usually dealt with a variety of noise having determined and undetermined sources. this paper was aimed at attenuating ground roll using the dddtdwt method on post-stack (residual noise) and pre-stack data. more than two-thirds of total generated seismic energy is imparted ali reza goudarzi and mohammad ali riahi32 into rayleigh waves in most surface seismic surveys when a compressional wave source is used (richart et al., 1970; choon b, park et al., 1999). receiver arrays are used to attenuate undesired coherent noise in certain wave numbers. shot hole depth and charge size are the important parameters in surface wave generation in land seismic acquisition. since the surface’s physical conditions can vary significantly along a seismic line, measurements taken in the field to eliminate surface waves may be insufficient (saatcilar r and canitez n, 1988). in recent years, halliday d (2011) has used an adaptive interferometry method for ground-roll suppression and henley d (2003) has tried to attenuate coherent noise in the radial trace domain. neelmani et al., (2008) used the curvelet domain to attenuate coherent and random noise in seismic section. more recently, bonar and sacchi (2012) have applied a nonlocal means algorithm to seismic noise attenuation. signal and noise frequency and time domain together have considerable overlap. hence, the fourier transform for noise removal is not always quite efficient because its basic premise is that stationary signal and seismic signals have non-stationary characteristics. the dwt technique can be a viable alternative for the removal of coherent and random noise in seismic data processing. wavelet transform is a simple and reliable tool for signal and image processing and has overcome the limitations of the fourier transform. the conventional dwt uses a wavelet and scaling function; this method was used by deighan et al., (1997) to remove ground roll noise. defects in this type of dwt are aliasing, its shift invariant property and a lack of directivity in higher dimensions (because coefficients related to trace ground rolls traced at the same scale are different and hamper making all coefficients zero, ground rolls therefore not being completely removed). surface waves provide common sources of noise in seismic land data. creating surface waves is closely related to regional geology and the location of sources and receivers (al-husseini et al, 1981; dobrin 1951). deeper sections’ seismic data is similar in bandwidth to that of ground rolls, but ground rolls have higher amplitude. a regular filtering method is 1d frequency filtering; however, the intersection between seismic signal and noise frequency time components means that 1d frequency filtering is not a useful tool for such purpose. another technique that is often used is 2d fourier transform or f-k transform; nevertheless, it disagrees with seismic data’s natural behaviour because of stationary signal assumptions. also, there is no distinct border between signal and noise in 2d fourier transform, consequently it distorts a seismic signal because of windowing and changes its frequency component. the dwt applied by deighan et al., (1997) had various restrictions; different techniques have been proposed to overcome such restrictions, including aliasing which may happen because of dwt’s low redundancy factor. it is also a shift-invariant transform making denoising difficult in practice and it cannot be expanded to m-dimensional applications (i.e. difficult directivity). it has been suggested that a proper wavelet could be applied to avoid aliasing and reduce the frequency band of scales intersection in the wavelet domain. some newer wavelet transform structures have been introduced instead of critically sampled dwt. a common type of dwt is udwt which was introduced by guo et al., (1995); redundancy grows with level and is perfectly shift-invariant in this method. udwt’s redundancy characteristics generate a huge amount of information which cannot be simply applied to seismic data processing because it is time-consuming. udwt can thus be replaced by practical methods. after methods were introduced by kingsbury (2001) and selesnick (2001), a combination of such methods (selesnick et al., 2005) was applied to denoising and image processing. the proposed techniques were based on multi-wavelet frames to avoid aliasing characteristics and achieve better time-frequency representation between scales. kingsbury applied the dwt in separate parallel trees; he used two wavelets that were approximately hilbert transform pairs (a dual tree cwt uses two real dwts: the first dwt gives the real part of the transform while the second dwt gives the imaginary part (selesnick et al., 2005)). selesnick (2001) used figure 1. iterated filter bank for the double-density dual-tree dwt (selesnick, 2004). a similar strategy but one of the wavelets had one sample delay and the wavelets were real. the double density discrete wavelet transform (dddwt) was based on oversampled fir filter banks. this method has much smoother wavelets than orthonormal wavelets (like daubechies’ orthonormal wavelets) from the same support. as ground roll occurs at higher scales, it can be picked up and suppressed clearly in the wavelet domain. dddwt and dtcwt are nearly shift invariant and have redundancy factors of 2, so they might be considered as perfect substitute methods. dddtdwt uses their advantages in signal processing and has a redundancy factor of 4. the wavelet domain ground roll analysis technique (wdga) has been used here to address ground roll coefficients in the wavelet domain, each signal being analysed to scales and each scale has its time-frequency resolution. note that frequency was related to scales (abry, 1997): f a = f c a. δ where a was a scale, δ was the sampling period, f c was the centre frequency of a wavelet in hz and f a was pseudo-frequency corresponding to scale a, in hz. methodology h, i(t) and g, i(t) (for i = 1, 2) were wavelets in the dddtdwt method. wavelets were applied with a sample apart and were nearly the same, so that  h, 1(t) ≈ h, 2(t – 0.5) . other pairs had an approximate hilbert pair of wavelets so that g, 1(t) ≈ h { h, 2(t)}. the results gave wavelets where the zero moments had to be determined, having a high degree of smoothness and the shortest possible length. the dddtdwt method is nearly shift-invariant and has a redundancy factor of 4. double density dwt is nearly invariant with shift, it uses a scaling function and two wavelet functions, it can remove noise, its redundancy factor is 2 and wavelets are smooth and their lengths are minimised. the adaptive seismic ground roll attenuation using the double density dual tree discrete wavelet transform (dwt) method 33 dual tree dwt method is almost invariant with shift, it is a fairly good approximation for continuous wavelet transform, its redundancy factor is 2, time-frequency width is enhanced and wavelet length is minimised. the two methods do have differences, such as both wavelets become linked together in quite different manners, the degree of freedom in dual tree dwt design is less than that of double density dwt and filter bank structure is distinct from each other. dddtdwt has advantages over both these methods. the relationship between wavelets is:  g, 1 (t) ≈ h { h, 1 (t)} (1) g, 2 (t) ≈ h { h, 2 (t)} (2) h, 1(t) ≈ h, 2(t – 0.5) (3) g, 1(t) ≈ g, 2(t – 0.5) (4) figure 1 shows the dddt dwt filter bank. dual tree dwt uses critically-sampled dwt and the structure is such that the filter banks are applied iteratively and in parallel. reconstruction filters in the filter bank are time-reversed analysis filters. the filters applied in this study had the following features which had to satisfy certain conditions: they had to satisfy perfect reconstruction, wavelets had to be approximately hilbert pairs, they had to satisfy zero moment property and they had to have a minimum length. figure 2 shows the wavelets considered for seismic data processing (selesnick et al., 2004). selesnick et al., (2005) encountered four problems and solved them by using the dddtdwt technique used here. oscillations mean that since wavelets are band pass functions then wavelet coefficients tend to oscillate positively and negatively around singularities. shift variance means that a small shift in the signal greatly perturbs wavelet coefficient oscillation pattern around singularities. substantial aliasing arises from wavelet figure 2. the wavelets and scaling functions (selesnick, 2004). the wavelets and scaling functions of double density dual tree dwt 5 0,5 0 -0,5 0 2 4 6 8 10 12 14 1 0,5 0 -0,5 -1 0 2 4 6 8 10 12 14 1 0,5 0 -0,5 -1 0 2 4 6 8 10 12 14 1 0,5 0 -0,5 -1 0 2 4 6 8 10 12 14 1 0,5 0 -0,5 -1 0 2 4 6 8 10 12 14 5 0,5 0 -0,5 0 2 4 6 8 10 12 14 h(t) h, 1(t) h, 2(t) g, 2(t)  g, 1(t) g(t) ali reza goudarzi and mohammad ali riahi34 coefficient samples’ wide spacing or, equivalently, the fact that wavelet coefficients are computed via iterated discrete-time down-sampling operations interspersed with non-ideal low-pass and high-pass filters. a lack of directionality would concern multi-dimension wavelet transforms (but was not the case in this study). unlike dwt based on real-valued oscillating wavelets, the fourier transform is based on complex-valued oscillating sinusoids. the oscillating cosine and sine components (real and imaginary parts, respectively) form a hilbert transform pair (selesnick et al., 2005). the fourier transform does not suffer from these problems. fourier transform magnitude does not oscillate positively and negatively but rather provides a smooth positive envelope in the fourier domain. fourier transform magnitude is perfectly shift-invariant, having a simple linear phase offset encoding the shift. fourier coefficients are not aliased and do not rely on a complicated aliasing cancellation property to reconstruct a signal. the m-d fourier basis sinusoids are highly directional plane waves (selesnick et al., 2005). the idea behind dddtdwt is to use characteristics from fourierbased methods but in short support wavelets and using a filter bank parallel strategy. thresholding the following nonlinear transform was applied to empirical wavelet coefficients to suppress noise: f (x) = x i (|x| > t) where t was a certain threshold. the choice of threshold was a very delicate and important statistical problem. a large threshold leads to atdata preparation shotgather f-k transform of shotgather f-k analusis of the groundroll inverse fourier transform forward dddtdwt of ground roll forward dddtdwt of shotgather inverse dddtdwt output figure 3. wdga technique flowchart. comparing performs of coefficients of ground rol al shot gather in the wavelet domain and set them of zero tenuating the seismic signal inside the ground roll whereas a small threshold provides inappropriate denoising. theoretical considerations yielded the following threshold value: (5)t = √ 2 2 log(n)n where n was input vector length and 2 variance of noise. the only difference between hard and soft thresholding procedures lie in the choice of nonlinear transform on empirical wavelet coefficients. the following nonlinear transform was used for soft thresholding: s(x) = sign(x)(|x| – t)i(|x| > t) (6) where t was a thresholding parameter. soft thresholding, like hard thresholding, has encountered similar problems in seismic denoising, especially for ground roll elimination. thresholding is well behaved for seismic random noise attenuation but causes trouble for data-processors for coherent noise attenuation. there is a trade-off between noise attenuation and signal damage; a practical approach was thus needed to overcome such restriction. the wavelet domain ground roll analysis technique (wdga) was thus used to address ground roll coefficients in the wavelet domain, each signal analysed to scales and each scale having its time-frequency resolution, frequency being related to scales (abry, 1997). it was obvious that ground roll low-frequency signal would be seen in low frequency scales, based on applying a wavelet transform strategy. filter banks of these lowfrequency scales may be in lower scales or upper scales. setting thresholding value depends on user experience; if a user chooses an incorrect thresholding value then thresholding algorithms go through all the scales in a particular wavelet domain. thresholding value means limiting to keep or kill a coefficient and will not ask if it is a part of a particular signal or not. where ground roll is based on the level of decomposition present in a few low-frequency scales, then thresholding procedures are inappropriate for attenuating them. the wdga strategy used in this paper consisted of: 1) transforming seismic data to the f-k domain and analysing it; 2) eliminating the energy relating to a signal because of ground roll linear characteristics in the fourier domain; 3) the inverse transform of the analysed data in the fourier domain providing the seismic section containing ground roll energy; 4) transforming the section containing signal and ground roll energy to the wavelet domain; 5) transforming the analysed section containing just ground roll energy to the wavelet domain by the same level of decomposition as the previous step; 6) comparing two domains from steps 4 and 5 and the coefficients present in step 5 from step 4 set to zero; and 7) step 6 inverse dwt regarding time domain providing a section with no ground roll energy. figure 3 shows a flowchart of the wdga technique. dddtdwt resolution level was controlled by signal length. maximum value was used for better time frequency representation. two types of transform were thus applied to a chain, where each transform’s abilities helped to diagnose noise having a linear pattern. linear analysis was used for ground roll elimination and it was believed that the wdga technique could attenuate each linear complex noise in the wavelet domain, e.g. air waves, multiples and tube waves. the main point concerned the choice of transform (in the chain) to be applied before discrete wavelet transform. no transform in seismic signal processing was ideal and perfect, but a combination of them led to reliable data-processing results. adaptive seismic ground roll attenuation using the double density dual tree discrete wavelet transform (dwt) method 35 application seismic data usually involves the following equation: y(t) = x(t) * w(t) + n(t) (7) where x(t) was the earth’s impulse response. the source wavelet w(t) was convolved with x(t), then noise n(t) was added to the equation (n(t) containing all types of noise). this noise could not be eliminated completely because its time and frequency overlapped the seismic signal. the wavelet used in generating a synthetic seismogram was a ricker wavelet having 30 hz central frequency; the ground rolls were in the 5-15 hz frequency range. figure 4a shows the synthetic section; this shot gather contained four reflectors where ground roll masked them, especially the first low velocity reflector. ground rolls have low frequency and high amplitude, features which can be very useful for attenuating ground roll. the dddtdwt method was thus used here; the data involved 2050 samples and figure 4b gives the sampling interval as 4 ms. selected traces’ frequency spectra can be seen before and after dddtdwt filtering. note in figure 5 that noise amplitude in the 5-15 hz frequency range has been attenuated. the first low velocity reflector appeared successfully and other reflectors’ continuity was enhanced. the dddtdwt method (due to time frequency representation) eliminated all ground roll energy while reflector continuity was enhanced without damaging the signal. note that the dddtdwt method preserved each frequency’s amplitude in the pass band. good shift invariance (and wavelet smoothness) required that cascaded multirate filters’ frequency response side lobes be small (kingsbury (2001). figure 4. a) synthetic seismic section, b) synthetic seismic section after dddtdwt filtering figure 5. a) the 20th signal’s frequency spectrum before (green) and after (red) dddtdwt filtering. sa m bl e n o. 200 400 600 800 1000 1200 1400 1600 1800 2000 100 200 300 400 500 trace no. sa m bl e n o. 200 400 600 800 1000 1200 1400 1600 1800 2000 100 200 300 400 500 trace no. input shotgather filtered shotgather a m p li tu de 1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 20 40 60 80 100 120 frequency (hz) amplitude spectrum of input data amplitude spectrum of output data (a) (b) ali reza goudarzi and mohammad ali riahi36 after applying the method to synthetic data, the method’s performance was investigated on real shot gathers. the real data had 1.92 second record length with 2 ms sampling interval, and the trace interval was 20m. ground roll frequency band in this example was estimated to be 3-12 hz by f-k mapping. dddtdwt was applied to attenuate ground roll (figure 6); this method had an acceptable result and noise became significantly reduced. events in a deeper part of the section had more continuity and signal distortion did not happen with this filtering technique. the signal was still visible and most ground-roll energy was attenuated. figure 7 shows the 20th spectrum’s data trace amplitude and its filtered section. the second type of data used here was a stacked section having 2 ms sampling interval, distance between adjacent traces being 25 m. record length was 1.92 seconds or 960 samples. the wavelet transform method’s efficiency was then tested on shot gather for remnant noise reduction of post stack data. all shot gather noise was attenuated and true velocity analysed to give the best results; the data was then stacked without ground roll attenuation using these velocity values (figure 8a). residual noise in post stack data occurred in some field data examples with complicated behaviour of ground roll or complex geological areas being generated; it might thus be of interest for geophysicists to investigate filtering methods’ behaviour in this respect. figure 8 shows noisy and denoised sections. comparing the sections shown in figure 8, it can be seen that the reflections were quite recovered without distortion, and horizontal continuity was extended and improved by using the dddtdwt method. it is thus worth emphasising that dddtdwt can be used to attenuate ground roll, having an application to both synthetic and real data. the algorithm produced satisfactory results regarding the preservation of wavefigure 6. a) real seismic section, b) real seismic section after dddtdwt filtering figure 7. the 20th signal’s frequency spectrum before (green) and after (red) dddtdwt filtering of real data. input shotgather filtered shotgather sa m bl e n o. 100 200 300 400 500 600 700 800 900 10 20 30 40 50 60 trace no. sa m bl e n o. 100 200 300 400 500 600 700 800 900 10 20 30 40 50 60 trace no. (a) (b) a m p li tu de 1 0,9 0,8 0,7 0,6 0,5 0,4 0,3 0,2 0,1 0 50 100 150 200 120 frequency (hz) amplitude spectrum of input data amplitude spectrum of output data adaptive seismic ground roll attenuation using the double density dual tree discrete wavelet transform (dwt) method 37 form, frequency and bandwidth and improved resolution of seismic data. filtered output had better s/n ratios and reflected signals’ trace-to-trace continuity. conclusions the dddtdwt method was used for seismic ground roll attenuation in this study. dddtdwt is an efficient way to filter seismic data; this method does not suffer from conventional dwt’s limitations. it can be quickly applied to a large volume of data and coherent noise in seismic data was successfully attenuated with this type of discrete wavelet transform. the wdga is an efficient and applicable method for seismic denoising and could be a substitute method for thresholding. wdga gains from two transforms: a regular transform (e.g. f-k) and discrete wavelet transform; the latter overcame fourier restrictions. this denoising method preserved signal information and generated little processing-related noise; filtering partly reduced noise. the advantage of dddtdwt, in conjunction with wdga, was that ground roll was suppressed by insignificant distortion of a particular signal. a reflection seismogram’s vertical resolution was not affected by applying dddtdwt, as happens when using frequency filters. acknowledgments we would like to thank the university of tehran’s institute of geophysics and the earth physics department which supported us in this study. we appreciate the kind advice given by professor kingsbury who helped us in this study. we would like to offer our regards and blessings to all who supported us in any respect during the completion of the study. references abry, p. (1997), ondelettes et turbulence. multirésolutions, algorithmes de décomposition, invariance d’échelles, diderot editeur, paris. al-husseini, m. i. , glover, j. b. , and barley, b. j. , 1981, dispersion patterns of the ground-roll in eastern saudi arabia: geophysics, 43, 121-137. al-yahya, k. m. , 1991, application of the partial karhunen-loeve transform to suppress random noise in seismic sections: geophys. prosp., 39,77-93. alsdorf, d. , 1997. , noise reduction in seismic data using fourier correlation coefficient filtering. , geophysics, v62, no. 5, pp. 1617-1627. baussard, a. , nicolier, f. , and truchetet. f. , 2004, rational multiresolution analysis and fast wavelet transform: application to wavelet shrinkage denoising. signal processing, 84(10):1735–1747. bayram, i. , and selesnick. i. w. , 2009, frequency-domain design of overcomplete rational-dilation wavelet transforms. ieee trans. signal process. , 57(8):2957–2972. beresford-smith, g. , and rango, r. , 1988, dispersive noise removal in t-x space: application to arctic data: geophysics, 53, 346-358. berkner, k. , and wells, jr, r. o. , 1998, a correlation-dependent model for denoising via nonorthogonal wavelet transforms. technical report cml tr98-07, computational mathematics laboratory, rice university, bonar d., sacchi m., 2012, denoising seismic data using the nonlocal means algorithm, geophysics. vol. 77, no. 1 (january-february 2012); p. a5–a8. brown, j. c. , 1991, calculation of a constant q spectral transform. j. acoust. soc. am. , 89(1):425–434. canales, l. l. , 1984, random noise reduction: presented at the 54th annual international meeting, seg. figure 8. a) post stack data: ground roll covered all parts of the section without adopting any particular trend, b) filtered section: dddtdwt attenuated ground roll energy input section filtered section sa m bl e n o. 200 400 600 800 10 20 30 40 50 60 trace no. sa m bl e n o. 200 400 600 800 10 20 30 40 50 60 trace no. (a) (b) 4 2 0 -2 -4 -6 ali reza goudarzi and mohammad ali riahi38 cary, p. w. ,2009. , ground roll attenuation with adaptive eigenimage filtering., seg houston 2009 international exposition and annual meeting chui, c. k. , 1992, an introduction to wavelets: academic press inc. deighan, a. j. , and watts, d. r. , 1997, ground-roll suppression using the wavelet transform: geophysics, 62, 1896–1903. dobrin, m. b., 1951, dispersion in seismic surface waves: geophysics, 16, 63-80. donoho, d. l. 1995, de-noising by soft-thresholding: ieee transactions on information theory, vol. 41, no. 3, 613-627. guo, h,1995,“theory and applications of the shift-invariant, timevarying and undecimated wavelet transforms,”master thesis, rice university”. halliday d.,2011, adaptive interferometry for ground-roll suppression, the leading edge, may 2011, p532-p537. henley d. c., 2003, coherent noise attenuation in the radial trace domain, geophysics, vol. 68, no. 4 (july-august 2003); p. 1408–1416, kingsbury, n. g. , 1998,the dual-tree complex wavelet transform: a new technique for shift invariance and directional filters. in proceedings of the eighth ieee dsp workshop, utah, august 9-12. kingsbury , n. g. 2001, complex wavelets for shift invariant analysis and filtering of signals. applied and computational harmonic analysis, 10(3):234-253. lang, m. , guo, h. , odegard,, j. e. , burrus, c. s. , wells, r. o. , mevis,1996, jr. ,1996,“noise reduction using an undecimated discrete wavelet transform,” ieee signal processingletters, pp. 10-12. nason, g. p. , and siverman, b. w. 1995, the stationary wavelet transform and some statistical applications, in a. antoniadis, ed. , wavelets and statistics: springer-verlag, 281–300. neelmani r., baumstein a. i., and gillard d. g.,2008, coherent and random noise attenuation using the curvelet transform, the leading edge, february 2008, p240-248. porsani, m.j. ,2009,. ground-roll attenuation based on svd filtering,. seg houston 2009 international exposition and annual meeting selesnick, i. w. , 1999, interpolating multiwavelet bases and the sampling theorem. ieee trans. on signal processing, 47(6):1615-1621. selesnick, i. w., baraniuk . r. g., and kingsbury, n., 2005, the dualtree complex wavelet transform a coherent framework for multiscale signal and image processing. ieee signal processing magazine, 22(6):123-151. selesnick, i. w. ,2004 the double-density dual-tree dwt. ieee trans. on signal processing, 52(5):1304-1314. selesnick , i. w., and burrus, c. s,. 1998 maximally at low-pass fir filters with reduced delay. ieee trans. on circuits and systems ii,45(1):53-68. selesnick, i. w. , and sendur, l,. 2000, smooth wavelet tight frames with denoising applications. in proc. ieee int. conf. acoust. , speech, signal processing (icassp). selesnick , i. w. , 2011, wavelet transform with tunable q-factor. ieee transactions on signal processing , 1-18. selesnick , i. w. ,2001, the double density dwt. in a. petrosian and f. g. meyer, editors, wavelets in signal and image analysis: from theory to practice. kluwer. vetterli, m. , 1992, wavelets and filter banks: theory and design: ieee trans. sig. proc. , 40, 2207-2232. this work purposes a general mean velocity and a suspended sediment concentration (ssc) model to express distribution in every point of the cross section of turbulent shear flow by using a probability density function method. in order to solve turbulent flow and avoid multifarious dynamical mechanics, the probability density function method was used to describe the velocity and concentration profiles interacted on directly by fluid particles in turbulent shear flow. the velocity profile model was obtained by solving for the profile integral with the product of the laminar velocity and probability density, through adopting an exponential probability density function to express probability distribution of velocity alteration of a fluid particle in turbulent shear flow. an ssc profile model was also created following a method similar to the above and based on the schmidt diffusion equation. different velocity and ssc profiles were created while changing the parameters of the models. the models were verified by comparing the calculated results with traditional models. it was shown that the probability density function model was superior to log-law in predicting stream-wise velocity profiles in coastal currents; and the probability density function ssc profile model was superior to the rouse equation for predicting average ssc profiles in rivers and estuaries. outlooks for precision investigation are stated at the end of this article. abstract keywords: exponential probability density; mean velocity profile; concentration profile; river flow; coastal current mean velocity and suspended sediment concentration profile model of turbulent shear flow with probability density function velocidad media y modelo de perfil de concentración de sedimentos suspendidos de flujo turbulentos de cizalla turbulento con funciones de densidad de probabilidad. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n3.65172 este trabajo propone un modelo de velocidad media general y un modelo de concentración de sedimentos suspendidos (css) para expresar la distribución en cada punto de la sección de cruce del flujo turbulento de cizalla mediante el uso de funciones de densidad de probabilidad (pdf).. el método de funciones de densidad de probabilidad se usó para describir los perfiles velocidad y concentración que interactuaron directamente con partículas fluidas en el flujo de desprendimiento turbulento para resolver el flujo turbulento y evitar diferentes mecánicas dinámicas. el modelo del perfil de velocidad se obtuvo resolviendo el perfil integral con el producto de la velocidad laminar y la densidad de probabilidad, mediante la adopción de una función de desidad exponencial para expresar la probabilidad de distribución de la velocidad de alteración de la partícula de un fluido en un flujo de desprendimiento turbulento. también se creó un modelo de perfil css siguiendo un método similar al anterior y basado en la ecuación de difusión schmidt. se crearon diferentes perfiles de velocidad y css durante el cambio de parámetros de los modelos. los modelos se verificaron comparando los resultados calculados con los modelos tradicionales. se demostró que el pdf era superior a la ley logarítmica en la predicción de los perfiles de velocidad en corrientes costeras, y que la probabilidad del modelo del perfil de función de densidad ssc fue superior a la ecuación rouse para predecir perfiles ssc promedio en ríos y estuarios. las perspectivas para la investigación de precisión se indican al final de este artículo. resumen palabras clave: densidad de probabilidad exponencial, perfil de velocidad media, perfil de concentración, corriente de río, corriente costera. record manuscript received: 22/05/2017 accepted for publication: 23/08/2017 how to cite item wu, g., zhu, l., & li, f. (2017). mean velocity and suspended sediment concentration profile model of turbulent shear flow with probability density function. earth sciences research journal, 21(3), 129 134. doi: http://dx.doi.org/10.15446/esrj.v21n3.65172 h y d r o g e o l o g y earth sciences research journal earth sci. res. j. vol. 21, no. 3 (september, 2017): 129 134 guanglin wua, b, liangsheng zhua,*, fangcheng lib acollege of civil and transportation engineering, south china university of technology, guangzhou 510641, china bcollege of ocean engineering, guangdong ocean university, zhanjiang 524088, china *corresponding author. tel: +86 02087111030-3682; e-mail: lshzhu@scut.edu.cn (l. s. zhu) 130 guanglin wu, liangsheng zhu, fangcheng li introduction turbulence is much more complex than laminar flow. according to the boussinesq eddy viscosity assumption (boussinesq, 1877), the logarithmic law for velocity profile was solved by using the mixing length method (prandtl, 1925; khan et al., 2017; zawar et al., 2017). some vertical velocity distribution laws, such as the power law and the exponential law, are based on experimental considerations or semi-empirical theories (lin et al., 1953; deissler, 1954; afzal et al., 2007). other laws for turbulent velocity profile or concentration profile are based on an indirect turbulence model (bucci et al., 2008; nucci and fiorucci, 2011). logarithm law is the most popular one used to express the mean velocity in a cross section of turbulent flow for wind flow, river flow, and lake models. the logarithm law fits velocity and concentration distribution in the position far away from the interface in developed turbulent flow (wang and yu, 2007). some researchers have focused on mean velocity profiles of turbulent wall-bounded flows separately (poggi et al., 2002; buschmann and gad-el-hak, 2007; kempe et al., 2007; buschmann et al., 2009). different vertical velocity distribution laws are used to describe streamwise velocity structure in coastal currents (soulsby, 1980; anwar, 1996, 1998; de serio and mossa, 2014), in tidal channels (soulsby and dyer, 1981), and on the continental shelf (soulsby, 1983; de serio and mossa, 2010).vertical distribution of suspended sediment concentration (ssc) in rivers, estuaries, and coastal currents have been studied by many researchers through the years (rouse, 1937; taylor and dyer, 1977; li uet al., 2014).this work derives general formulas to describe the mean velocity and ssc distributions of turbulent shear flow to fit the whole cross section from the bottom boundary to the upper boundary, using a probability density function method. a brief review of classical methods for velocity profile of shear flow consider the laminar shear flow—the steady incompressible developed flow between two interfaces, with one interface moving with velocity u and the other being motionless. the flow can be solved based on 2-dimensional navier-stokes equations and expressed as (1) here h is the distance between two interfaces, μis dynamic viscosity, and d p/ d x is the pressure gradient along the x-direction. the velocity profile of shear flow can be simplified to eq. (2) when d p/ d x = 0 (2) logarithm law (prandtl, 1925), derived by the mixing length method, is the most popular equation used to express the mean velocity in a cross section of turbulent shear flow. the basic logarithm law of mean velocity profile of turbulent shear flow is (3) here, u* is the shear velocity of the fluid and y0 is hydrodynamic roughness length. but eq. (3) cannot fit select velocity profiles not full profiles because of the assumption that the mixing length is proportional to the distance away from the wall boundary. many instances and experiments have proven that the mixing length method can successfully predict velocity distributions far away from the interfaces, but it cannot predict velocity distribution near interfaces such as walls, seabed, gas-liquid surfaces, and so on. as far as we know, turbulent intensity should be dependent on the reynolds number, which is composed of the characteristic velocity, characteristic length, and kinematic viscosity. considering the mixing length method carefully, the core assumption is that the eddy viscosity is in direct proportion to the average velocity gradient and the mixing length is in direct proportion to the distance away from the wall. the method does not directly create the relationship between turbulent parameters and average velocity. exponential probability density model (epdm) for mean velocity profile of turbulent shear flow to solve for turbulent flow without closely associated defferent dynamical mechanics, a probability density function method is adopted to describe the pulsating intensity of velocity of fluid particles in this study. the pulsating intensity of every fluid particle in the profile will impact the whole profile. the effect decays at a distance from its original position and obeys a probability density function. exponential probability density function is often used to describe the decay degree distribution in many physical and chemical problems. in many situations, the exponential probability density function model is an appropriate method to approach the random decay phenomena. the primary expression of the exponential probability density function is (4) here, p(y) is the probability density of a fluid particle impacting any position, λ is the damping index, and y is the distance from the particle position yi to the impacting position y, defined by (5) as shown in fig. 1, the expression of exponential probability density function p1(y), represents the probability density that the velocity of laminar shear flow at position y1, u1, distributes to any position y due to turbulence. because the expression satisfies the boundary condition that p1(y1) = pm1 where y = y1 as shown in fig. 1, λ = pm, such as λ1= pm1, λ2 = pm2, and so on. hence, the expression pi(y) satisfies (6) figure 1. schematic diagram of exponential probability density for pulsating intensity transfer 131diffraction characteristics of small fault ahead of tunnel face in coal roadway the probability density of u, velocity of any position of y, distributes to y0, a certain position at the profile, is defined as p0 as shown in fig. 1, such as p01 and p02. we have (7) average turbulent velocity of a particular position, y0, produced by integrating the product of the local laminar velocity in eq. (7), which is solved with navier-stokes equations, and the exponential probability density function of pulsating intensity at the position, across all fluid particles in the profile, is shown as (8) here α is an adjustment. substituting u(y) with eq. (2) and p0(y) with eq. (7) in eq. (8), we get (9) eq. (9) cannot be analytically and fully solved. to solve eq. (8), we assume that (10) substituting eq. (2) and eq. (10) into eq. (8), the result is (11) integrating eq. (11), the result is (12) setting (13) and substituting y0 with y in eq.(12), the velocity profile of turbulent shear flow with this method (called pdfm) should be (14) assume that (15) eq. (14) can be calculated when the adjustable parameters, m and n, are given. the mean turbulent velocity in eq. (14) can be calculated by giving different m and n, as well as defining the dimensionless turbulent velocity and dimensionless position as (16) the calculated results are shown in fig. 2 and fig. 3. fig. 2 reveals that the mean velocity profile changes when the value of m is changed, but the mean velocity changes are weaker in the region near the wall or other interfaces than in the intermediate region while α=1. the dimensionless velocities are almost the same in the region when the value of y+ is between 0 and 0.25, but they change very evidently near the position where y+ is approximately equal to 0.8. thus, the upper mean velocity will increase along with an increasing value of m. the curve shape fits turbulent velocity profile when the value of m is lower in this case, such as m=0.25, 0.1 and 0.01. figure 2. mean velocity profiles with change of m (n= 0.25) figure 3. mean velocity profiles with change of n (m = 0.3) 132 guanglin wu, liangsheng zhu, fangcheng li de serio and mossa (2010, 2014) investigated vertical profiles of the stream-wise velocity of the coastal currents at the taranto (ta) and bari (ba) sites in italy. the comparison between the measurement data, the log law as shown in eq. (3), and the pdfm as shown in eq. (17) is plotted in fig. 4 at taranto for vertical stream-wise velocity profiles in shallow coastal currents. the calculated results show that the probability density function model (pdfm) more accurately reflects measurement data than the log-law model. ssc profiles model and its verification in river and estuary the rouse equation (rouse, 1937) was the most popular method used to calculate vertical concentration profiles of suspended load in rivers, lakes, estuarine and coastal currents, as well as on the continental shelf. the rouse formula was (18) in which ya is the thickness of the sheet-flow layer, ca is the concentration at ya which equals the bed-load concentration, z*=ω/κu* is the suspension index, and ω is the particle settling velocity. the similar expression of pdfm also can be created by the method below based on the schmidt diffusion equation (schmidt, 1925). (19a) that is (19b) in which ε is the particle diffusion coefficient, set as an assumption of the exponential probability density function (20) then (21) hence, (22) here γ is the suspension index of this model. the results of ssc profiles calculated on pdfm of eq. (22) are shown in fig. 5. the mean velocity profiles are distinct when the values of n are different, as shown in fig. 3, while α=1. the main tendency of the profile curves is shifting from the top left to the bottom right in fig. 3 along with the decreasing value of n. the tendency indicates that the parameter n has a closed relationship with turbulent intensity. verification of pdfm in coastal currents the mean velocity profile model derived in the previous section was applied to fit the investigated profiles to verify the effectiveness of the model. to use eq. (14) conveniently, the profile formula of pdfm should be transformed as (17) here β= αu/u*. figure 4. calculated and measured vertical profiles of the stream-wise velocity of coastal currents at taranto (survey data cited in de serio and mossa (2014)) 133diffraction characteristics of small fault ahead of tunnel face in coal roadway figure 5. calculated results of suspended sediment concentration profile model figure 5 shows that the slope of the profile curves is bigger near the surface than the bottom and the upper slope increases as the lower slope decrease when γ is increasing. the trend means that the particle settling velocity, particle diffusion coefficient and the water depth velocity influence γ to change the ssc distribution in the vertical cross section. the ssc is very small when γ=50, so the bed load is the dominant composition when γ≥50. the curves also show that the ssc at surfaces is greater than zero when γ≤2. on the contrary, the ssc at surfaces tends to zero when γ>2. so, this model provides an approach to avoid the surface limitation of rouse equation. zhang et al. (2007) investigated ssc of the yangtze river in detail and concluded a semi-empirical model of ssc. a comparison of predicting profiles between the pdfm (eq. 22) and the rouse equation [eq. (18)] with the measurements in the yangtze river is shown in fig. 6. as shown in the figure, the pdfm model used for this study more accurately reflects the measurement data than does the rouse equation. figure 6. calculated and measured vertical ssc profile in the yangtze river (survey data cited in zhang et al. (2007)) figure 7. calculated and measured vertical ssc profile in the yangtze estuary (survey data cited in liu, et al. (2014)) liu, yang, zhu, et al. (2014) investigated the average ssc profile in the yangtze estuary in detail. the survey data is cited here to verify the pdfm model, eq. (22). the comparison between the rouse equation, pdfm, and measurements is shown in fig. 7. the pdfm model of this study is also agreed more closely with the measurement data than does the rouse equation in the case of the estuarine area. discussions a model for solving velocity and concentration distribution of turbulent flow by using probabilistic methods is put forward in the article (khan et al., 2017; sultana et al., 2017). general models for describing the mean velocity and ssc profiles of a whole cross section in turbulent shear flow are derived based on the idea. in fact, the velocity profile formula of laminar shear flow is also the mean velocity profile solution of turbulent flow which ignores the turbulent items in reynolds equation, thus the pulsating intensity of fluid particles is apparently related to it. on the other hand, the pulsating intensity is also under the control of certain boundary conditions. the exponential probability density function may describe the pulsating intensity of fluid particles of the whole cross section correctly to study mean velocity and concentration distributions. the calculated results of the derived formulas are as diverse as the changing the damping index of exponential probability density function. an innovative approach may be found to solve complex turbulent flow by expanding on this concept. conclusions the pdfmmodel of this study possesses some advantage in the fields of predicting stream-wise velocity profiles of coastal currents and ssc profiles in river and estuaries as shown in the previous examples.further research is needed to validate the method through continued studies and carried forward to clarify and optimize the parameters of these models to apply for practical problems. for advanced research, the parameters should be studied meticulously, such as its relationship with dynamic parameters, particularly turbulent characteristic parameters. 134 guanglin wu, liangsheng zhu, fangcheng li acknowledgments this work was partially supported by the guangdong province science and technology research project (2010b030600007) and projects of enhancing school with innovation of guangdong ocean university (gdou2016050258 & gdou2017052503). references afzal, n., seena, a., & bushra, a. (2007). power law velocity profile in fully developed turbulent pipe and channel flows. journal of hydraulic engineering, 133(9), 1080-1086. anwar, h. o. (1996). velocity profile in shallow coastal water. journal of hydraulic engineering, 122 (4), 220-223. anwar, h. o. (1998). spread of surface plume in homogeneous and stratified coastal waters. journal of hydraulic engineering, 124 (5), 465-473. boussinesq, j. (1877). essai sur la théorie des eaux courantes. académie des sciences. (essay on the theory of water flow. in french). french academy of sciences 23(1), 666-680. bucci, g., fiorucci, e., russo, s. a., & nucci, c. d. (2008). analysis of turbulent flow speed profiles in pressure pipes using the dissimilar similitude technique applied to an electrolytic tank: implementation and experimental characterization. ieee transactions on instrumentation and measurement, 57(8), 1547-1553. buschmann, m. h. & gad-el-hak, m. (2007). recent developments in scaling of wall-bounded flows. progress in aerospace sciences, 42(5), 419-467. buschmann, m. h., indinger, t., & gad-el-hak, m. (2009). near-wall behavior of turbulent wall-bounded flows. international journal of heat and fluid flow, 30 (5), 993-1006. de serio, f., & mossa, m. (2010). velocity profiles of sea currents. in: proceedings of the 6th international symposium on environmental hydraulics, athens, greece, 23-25 june 2010, crc press, 451-456. de serio, f. & mossa, m. (2014). stream-wise velocity profiles in coastal currents. environmental fluid mechanics, 14(4), 895-918. deissler, r. g. (1954). analysis of turbulent heat transfer, mass transfer and friction in smooth tubes at high prandtl and schmidt. national advisory committee for astronauts, washington d.c., usa. kempe, a., schlamp, s., & rösgen, t. (2007). non-contact boundary layer profiler using low-coherence self-referencing velocimetry. experiments in fluids, 43(2-3), 453-461. khan, m. z., rahman, z. u., khattak, z., & ishfaque, m. (2017). microfacies and diagenetic analysis of chorgali carbonates, chorgali pass, khair-e-murat range: implications for hydrocarbon reservoir characterization. pakistan journal of geology, 1(1): 18-23. lin, c. s., moulton, r. w., & putnam, g. l. (1953). mass transfer between solid wall and fluid streams, mechanism and eddy distribution relationships in turbulent flow. industrial and engineering chemistry, 45(3), 636-640. liu, j. h., yang, s. l., zhu, q., & zhang, j. (2014). controls on suspended sediment concentration profiles in the shallow and turbid yangtze estuary. continental shelf research, 90(1), 96-108. nucci, c. d., & fiorucci, e. (2011). mean velocity profiles of fullydeveloped turbulent flows near smooth walls. comptes rendus mécanique, 339(6), 388-395. poggi, d., porporato, a., & ridolfi, l. (2002). an experimental contribution to near-wall measurements by means of a special laser doppler anemometry technique. experiments in fluids, 32(3), 366-375. prandtl, l. (1925). über die ausgebildete turbulenz. (on fully developed turbulence. in german). z.a.m.m., 5, 136-139. rouse, h. (1937). modern conceptions of the mechanics of fluid turbulence. trans. asce. 102(1), 463-543. schmidt, w. (1925). xxiii. gefügestatistik. zeitschrift für kristallographie. mineralogie und petrographie, 38(1), 392-423. soulsby, r. l. (1980). selecting record length and digitization rate for nearbed turbulence measurements. journal of physical oceanography, 10(2), 208-219. soulsby, r. l. (1983). the bottom boundary layer of shelf seas. physical oceanography of coastal and shelf seas ed., elsevier oceanography series 35, 189-266. soulsby, r. l., & dyer, k. r. (1981). the form of the near-bed velocity profile in a tidally accelerating flow. journal of geophysical research: oceans, 86(c9), 8067-8074. sultana, m. n., akib, s., & ashraf, m. a. (2017). thermal comfort and runoff water quality performance on green roofs in tropical conditions. geology, ecology, and landscapes, 1(1), 47-55. taylor, p. a., & dyer, k. r. (1977). the theoretical models of flow near the bed and their implication for sediment transport. the sea, 6(1), 579-601. wang, y. h., & yu, g. h. (2007). velocity and concentration profiles of particle movement in sheet flows. advances in water resources, 30(5), 1355-1359. zawar, z. u., khan, m., khattak, z., abbas, m. a., & ishfaque, m. (2017). microfacies analysis and reservoir potential of sakesar limestone, nammal gorge (western salt range), upper indus basin, pakistan. pakistan journal of geology, 1(1), 12-17. zhang, r. j., xie, j. h., chen, & w. b. (2007). river dynamics (in chinese). wuhan univ. press, wuhan, hubei, p.r. china. palabras clave: basurero municipal; índice de compresion; propiedades físicas; modelo anfis; modelo ann; estándar estadístico. key words: municipal solid waste, compression ratio, physical properties, anfis model, ann model, statistical criteria the compression ratio of municipal solid waste (msw) is an essential parameter for evaluation of waste settlement. since it is relatively time-consuming to determine compression ratio from oedometer tests and there exist difficulties associated with working on waste materials, it will be useful to develop models based on waste physical properties. therefore, present research attempts to develop proper prediction models using anfis and ann models. the compression ratio was modeled as a function of the physical properties of waste including dry unit weight, water content, and biodegradable organic content. a reliable experimental database of oedometer tests, taken from the literature, was employed to train and test the ann and anfis models. the performance of the developed models was investigated according to different statistical criteria (i.e. correlation coefficient, root mean squared error, and mean absolute error) recommended by researchers. the final models have demonstrated the correlation coefficients higher than 90% and low error values; so, they have capability for acceptable prediction of municipal solid waste compression ratio. furthermore, the values of performance measures obtained for ann and anfis models indicate that the anfis model performs better than ann model. el índice de compresión de residuos sólidos es un parámetro esencial para la evaluación del asentamiento de un basurero municipal. debido al desgaste de tiempo para determinar el índice de compresión a partir de pruebas edométricas y debido a las dificultades asociadas al trabajo con materiales desechados es necesario desarrollar modelos basados en las propiedades físicas de los desechos solidos. además, la presente investigación pretende desarrollar modelos de predicción apropiados a partir de los esquemas anfis y ann. el índice de comprensión se modeló como una función de propiedades físicas de desechos que incluyen el peso seco de una unidad, el contenido de agua y el contenido orgánico biodegradable. de la literatura se tomó una base de datos confiable de pruebas edométricas experimentales que fue empleada para preparar y evaluar los modelos anfis y ann. el desempeño de los modelos desarrollados fue investigado de acuerdo con diferentes criterios estadísticos (por ejemplo, el coeficiente de correlación, el error cuadrático medio y el error medio absoluto) recomendados por investigadores. los modelos finales han demostrado coeficientes de correlación mayores al 90 por ciento y valores bajos de error. esto significa que estos modelos tienen una capacidad de predicción aceptable para el índice de comprensión del basurero municipal. además, los valores de las medidas de desempeño obtenidos para los modelos anfis y ann indican que el modelo anfis tiene una mayor asertividad que el modelo ann. earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014): 165 171 abstract resumen record manuscript received: 10/02/2014 accepted for publication: 29/07/2014 geotechnics compression ratio of municipal solid waste simulation using artificial neural network and adaptive neurofuzzy system maryam mokhtari¹*, ali akbar heshmati r¹, nader shariatmadari¹ ¹ school of civil engineering, iran university of science and technology (iust), narmak, tehran 16846-13114, iran, tel: +98 21 77240565 ¹ school of civil engineering, iran university of science and technology (iust), narmak, tehran 16846-13114, iran, tel: +98 21 77240565, corresponding author: mokhtari_ma@iust.ac.ir ¹ school of civil engineering, iran university of science and technology (iust), narmak, tehran 16846-13114, iran, tel: +98 21 77240565 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.41988 http://dx.doi.org/10.15446/esrj.v18n2.41988 166 ali akbar heshmati r, maryam mok htari, nader shariatmadari introduction settling municipal solid waste (msw) is a significant issue considered in geotechnical design of a landfill. waste settlement influences the stability, deformation, and structural performance of landfills (chen et al., 2010). the settlement of msw is composed of three mechanisms: primary compression, mechanical creep, and bio-compression (sowers, 1973; edgers et al., 1992; wall and zeiss, 1995; el-fadel and khoury, 2000; and bareither et al., 2011). primary compression occurs quickly in response to self-weight and an external load applied to the waste. secondary compression (i.e. mechanical creep) is due to long-term slippages, reorientation of particles, and delayed compression of some waste constituents. bio-compression is related biochemical processes associated with the decomposition of biological solids and gas generation. a commonly used parameter for predicting the primary compression strain in msw is the compression ratio (c’c): (1) where εi is the primary compression in msw, σvo ' is the existing vertical effective stress and δσv ' is the change in the vertical effective stress. the compression ratio represents the slope of the curve of the strain versus the logarithm of the effective pressure. it is conventionally determined through an oedometer test and is related to the compression index (cc), which is commonly applied for soils by the equation c 'c = cc1+e0 , where e0 = initial void ratio. considering some difficulties in calculating the initial void ratio of the waste materials, it is usually preferred to use the compression ratio instead of the compression index. there have been several studies investigating the effect of various parameters on waste compressibility. chen et al. (2010) and landva et al. (2000) examined waste samples with different initial densities. their results demonstrate that an increase in the waste density reduces the c’c. experimental works conducted by sowers (1973), swati and joseph (2008), and chen et al. (2009) on waste settlement revealed an increase in organic compounds and compressible materials could bring about an increase in c’c. karimpour-fard and machado (2009) and chen et al. (2009) tried to examine the compressibility of waste samples that were at different ages. the results demonstrated that an increase in the age; thus, further waste decomposition decreased the waste compressibility and compression ratio. durmusoglu et al. (2006) tested waste materials with a natural moisture content and field capacity moisture. the results indicated that wastes with field capacity moisture had a higher c’c than those with a natural moisture content. other researchers, such as vilar and carvalho (2004) and reddy et al. (2009b), examined waste samples with a similar composition and a dry unit weight. they reported higher c’c values for wastes with higher water contents. earlier experimental works introduced such parameter as dry unit weight, water content, and contribution percentage of biodegradable organic materials (paper, food wastes, and yard wastes) as the most significant factors which affect the compressibility of municipal solid waste. earlier studies have addressed the effect of some parameters on waste compression ratio. it is to be noted that determining compression ratio through an oedometer test is time-consuming. another problem with doing this test on waste materials is the scale effect. owing to the fact that the particles of waste materials are much larger than those of fine-grained soil, it is not possible to precisely determine the waste compressibility by commonly used oedometers such as casagerande oedometer. in many cases, therefore, any precise investigation of the compressibility behavior of waste entails designing and manufacturing oedometer instruments that suit the size of waste materials. also, owing to the specific nature of waste materials, it would be hard to work on them. therefore, it seems necessary to develop models that are based on effective parameters measured through more simple experimental methods. in recent years, soft computing methods such as neural networks and neuro-fuzzy have been used successfully to solve a wide variety of problems in geosciences εi =cc ' σv0 ' +δσv ' σv0 ' and geotechnical engineering. these include the estimation of the prediction of compression ratio of soils (park and lee, 2011; and ozer et al., 2008), strength parameter modeling of different soils (sezer, 2013; khanlari et al., 2012; and heshmati et al., 2009), simulation of waste generation (noori et al., 2009), estimation of probability of the liquefaction (hanna et al., 2007; and venkatesh et al., 2013). in the present study, two soft computing methods, i.e. adaptive neuro fuzzy inference system (anfis) and neural network (ann), have been utilized to propose efficient models for predicting the compression ratio of msw. artificial neural network (ann) a neural network consists of parallel layers of simple computational elements called ‘artificial neurons’. all of the layers are connected to each other by interconnection weights. figure 1(b) shows an artificial neuron model. the artificial neurons basically include three parts: weight, bias and transfer function. in each neuron, coming signals from the previous layer, ix , are multiplied by the related adjustable connection weight, ijw , and then summed in addition to an bias value, θ j , . then the combined input, ji , is passed through a nonlinear transfer function, )( jif , to yield the output of the neuron (yi) (jaksa et al., 2008). output of each neuron creates the input of the neuron in the next layer. this process is summarized in equations (2) and (3), and illustrated in fig 1(b): (2) (3) fig. 1 (a) the feed forward neural network structure used for modeling compression ratio of msw, (b) the artificial neuron model one of the most practical artificial neural networks for solving the problems of engineering and geotechnics is multi-layer perceptron (mlp) network (khanlari et al., 2012). the mlp networks contain an input layer, one or more hidden layers, and one output layer. this network is a feedforward network, the information of which only flows in one direction, from the input to the output. figure 1 depicts a three-layered feed-forward network. mlp utilizes a supervised learning technique called ‘back propagation’ for training the network. back propagation algorithm is one of the most common supervised learning algorithms for training feed forward multilayer networks. this algorithm involves two main phases, namely forward pass and backward pass. in the forward pass, an input training pattern is shown to the network, and its effect propagates in the network layer by layer until the output of the network is obtained. in the forward pass, the error between the network prediction and the target value is calculated and distributed; then, by means of a technique called gradient descent, it is propagated from the output layer to the input layer in order to reduce the error value. this process is iterated, and the weights are updated until the summed square of the errors comes below an acceptable value. to enhance speed and performance of the back propagation algorithm, levenberg-marquardt algorithm can be alternatively used instead of the conventional gradient descent method. i j = wji∑ xi +θ j yi = f (i j) 167compression ratio of municipal solid waste simulation using artificial neural network and adaptive neurofuzzy system adaptive neuro-fuzzy inference system (anfis) neuro-fuzzy systems are the result of the integration of fuzzy logics and artificial neural networks. so they benefit from the advantages of both fuzzy systems and neural networks. in this hybrid method, the fuzzy system provides the relation between input and output variables, while parameters of membership functions of the fuzzy system are optimized by the neural network. one of the well-known neuro-fuzzy models is adaptive neuro-fuzzy inference system (anfis) which is capable of modeling any nonlinear function with a desirable accuracy (jang et al., 1997). anfis uses a feed forward network to optimize the parameters of a fuzzy inference system and perform well on a given task. the most common fuzzy inference system with this compatibility located in an adaptive network is takagi-sugeno fuzzy system. the first order sugeno model was adopted in this study as it is also extensively used in many engineering problems. an example of the first order sugeno fuzzy inference system with two inputs, one output and two membership functions for each of the inputs is illustrated in figure 2(a). two typical if-then rules can be stated for such a model (sugeno and kang, 1998): (4) if x = a2and y = b2 then f2(x,y) = p2x+q2y+k2 (5) where a1, a2, b1, and b2 are the membership functions for inputs x and y, respectively, while k1, p1, q1 and k2, p2, q2 represent the parameters of output functions for the two defined rules. the discussed architecture of the anfis is also depicted in figure 2(b). it can be seen that calculations of the anfis model are implemented in five layers. the detail and formulation on anfis are addressed by jang et al. (1997). anfis utilizes a back propagation learning algorithm or a hybrid algorithm which is an integration of back propagation and least square techniques in order to generate fuzzy rules and tune member function. fig. 2 (a) the first order sugeno fuzzy model, (b) corresponding anfis architecture (padmini et al., 2008) experimental database a valid database was used for modeling. it included the results of 64 oedometer tests on msw as well as information related to the tested waste properties such as dry unit weight (γd ), dry weight water content (ωd ), and percentage of biodegradable organic materials (om ) obtained in previous f x = a1and y = b1then f1(x,y) = p1x+q1y+k1 studies. a considerable part of these data and references cited by bareither et al. (2012) were combined with the data provided by hyun ii et al. (2011), reddy et al. (2011), and karimpour-fard and machado (2012). the employed data are indicated in table 1. the physical properties of wastes (i.e ωd,γd, and om) were used as the input parameters and the cc' as the target in the anfis and the ann models. table 1. the sources of the database reference no. of oedometer test beaven and powrie (1995) 5 chen and lee (1995) 1 gabr and valero (1995) 6 landva et al. (2000) 4 olivier and gourc (2003) 1 vilar and carvaleho (2004) 4 olivier and gourc (2007) 2 stoltz and gourc (2007) 1 chen et al. (2009) 5 reddy et al. (2009a) 3 reddy et al. (2009b) 4 reddy et al. (2009c) 3 reddy et al. (2011) 5 chen et al. (2010) 2 stoltz et al. (2010) 9 breither et al. (2011) 6 hyun ii et al. (2011) 1 karimpourfard and machado (2012) 2 in order to simulate the compression ratio, the database was randomly divided into two subsets, training and testing subsets. the training subset was used to calibrate the models, and the testing subset was used for validation of the developed models based on the training subset. the selection was such that the maximum, minimum, and mean value as well as the standard deviation of the parameters got to be close to each other in the training and the testing of the subsets. eighty percent of the data (51 cases) were assigned for the training subset and twenty percent (13 cases) for the testing subset. to control network overtraining and over-fitting, the training set was divided into two subsets: training and cross validation. the use of about 10% of the training data as a cross validation set has been suggested by researchers (kucuk 2008). as a result, in this study, 10% of the training subset was randomly taken and used for cross validation. the statistical properties of the divided subsets are summarized in table 2. 168 ali akbar heshmati r, maryam mok htari, nader shariatmadari table 2. statistical properties of the training and testing sets parameter statistics training data testing data d (%)ω max 151.00 135.00 mean 71.02 73.79 min 15.60 31.30 std dev 33.72 34.59 γd (kn/ m3) max 12.50 9.14 mean 5.48 5.85 min 2.45 1.86 std dev 1.86 2.18 om (%) max 60.00 60.00 mean 32.91 34.12 min 0.49 2.00 std dev 19.79 20.03 c’c max 0.39 0.34 mean 0.24 0.21 min 0.09 0.14 std dev 0.06 0.07 performance measures various statistical parameters including correlation coefficient (r), root mean squared error (rmse), and mean absolute error (mae) were investigated in order to evaluate the efficiency and accuracy of the proposed models. these parameters are defined as: (6) (7) (8) in equations (6), (7) and (8), n is the number of the data, mi and pi display the measured and predicted values, respectively. also, m and p are the mean values corresponding to the measured and predicted values, respectively. development of ann model as the topology of a network directly influences its computational complexity and generalization capability, it is important to devise appropriate network architecture in neural network modeling. this needs determination of the number of hidden layers and the number of neurons in each layer. theoretical works have revealed that a multilayer feed forward network with one hidden layer is able to approximate any continuous and complex nonlinear function with the desired degree of accuracy (hornik et al., 1989). consequently, one hidden layer was used in this study. the number of neurons in the input and output layers is dependent on the number of model inputs and outputs. the input layer of the ann model, developed in this study, has tree neurons, including one for each input. the output layer has only one neuron representing the measured value of compression ratio. the ann performance is influenced by the number of neurons in hidden layers. however, there are no specific guidelines that can r = (mi −mi=1 n ∑ )(pi −p) (mi −m) 2( (pi −p) 2 i=1 n ∑ i=1 n ∑ rmse = (mi −pi) 2 i=1 n ∑ n mae = 1 n mi −pii=1 n ∑ be used to select the optimum number of neurons in a hidden layer for a given problem. this parameter is often selected through a trial and error method. to obtain the optimum network architecture, different numbers of hidden-layer neurons, from 3 to 8, were employed to train different networks. finally, an appropriate ann network with best performance was selected by considering r and rmse values of the testing data set and the training data set. a back-propagation algorithm was used to train the mlp neural network. specifically, the levenberg–marquardt algorithm was used to improve the speed and efficiency of back propagation algorithm. also, a tan-sigmoid function was used as the activation function in the hidden and output layers. after the training phase, the generalization capability of the trained model was evaluated using the test data set (13 data). the output value of the optimal network, compression ratio, can be calculated by equation (18): (9) where w1 and w2 are the weight matrices, and b1 and b2 are the biases vectors. the values of these parameters are presented in table 3. development of anfis model in the anfis system, each input parameter must be clustered into several class values in order to form fuzzy rules, and each fuzzy rule is made up of two or more membership functions .various methods have been proposed for the clustering of the input parameter and development of the rules, the most common of which are: “subtractive clustering” method (chie, 1994) and “grid partitioning” method (jang and sun, 1995). the former was adopted in this study because it required less computational effort and calculation time. the subtractive clustering algorithm assumes that every data point is a candidate that can become a cluster center. then, a measure of density for each data point is used to determine the potential of a point to become a cluster center (chiu, 1994). the data point with the highest potential is selected as the first cluster center, and the data points near the first cluster center are removed, as determined by the influential radius. the algorithm searches for a new data point having the highest number of neighbors as the next cluster center. then the data points near the new cluster center with fewer neighbors are removed. this is repeated until all the data points are examined and sufficient numbers of cluster centers can be generated. the influential radius is necessary for determining the number of clusters. a smaller radius leads to many smaller clusters in the data space, thereby resulting in more rules, and vice versa. the value of the optimal radius in a trial and error process was determined by changing the radius from 0.1 to 1 with an increment of 0.05. at last, the radius value was chosen to be 0.9 that resulted in approximately two clusters. therefore, each input and output was characterized by two membership functions and, hence, two rules were created. the gaussiantype membership function was assigned to each input variable. after constructing the initial fis structure, the training stage was done. the hybrid learning algorithm was employed as an optimizing method in the training stage. in this step, the number of iterations of the hybrid algorithm for the correction of model parameters and goal error were considered 100 and 0, respectively. the constructed anfis model has 20 parameters (8 linear and 12 nonlinear). the membership function (mf) plots of the dry weight water content, dry unit weight, and biodegradable organic compounds used in the training stage are shown in fig. 3. cc ' = f2 w2[ ] f1 w1[ ] ωd γd om ⎡ ⎣ ⎢ ⎢ ⎢ ⎤ ⎦ ⎥ ⎥ ⎥ + b1[ ] ⎡ ⎣ ⎢ ⎢ ⎢ ⎤ ⎦ ⎥ ⎥ ⎥ + b2[ ] ⎡ ⎣ ⎢ ⎢ ⎢ ⎤ ⎦ ⎥ ⎥ ⎥ 169compression ratio of municipal solid waste simulation using artificial neural network and adaptive neurofuzzy system fig. 3 membership function plots of the compression ratio, (a) dry weight water content, (b) dry unit weight, and (c) biodegradable organic compound results and discussion ann model fig. 1(a) schematically depicts the ann architecture used in this study. a study on different networks with different hidden neurons showed that a network comprising five neurons in a hidden layer approaches the lowest prediction errors and maximum correlation coefficient. a comparison of the prediction made by the ann with the experimental value for each of the training and testing data set is shown in fig 4. as it can be seen, most of the cases in the training and testing subsets are around the middle, indicating a good agreement between the ann model prediction and the experimental values. fig.4. measured versus predicted compression ratio of msw using ann model, (a) training data set, (b) testing data test table 3 shows the weights of the input-hidden and hidden-output layer connections and biases of the proposed ann model after the end of the training procedure. the relative importance of the input parameters can be assessed by partitioning the hidden-output connection weights into components connected with each input variable. table 4 lists the relative importance of the input variables in the ann model. as seen in this table, the dry density and the dry weight water content have the most significant effect on the predicted compression ratio with average relative importance values of 41.88 and 43.47%, respectively. table 3. connection weights and biases bias connection weights hidden neurons b2b1outputom -3.400870.85305-0.16996-0.54242.3920481 -1.01113-0.305351.3268221.993881-0.997252 0.108780.4829270.9026261.1143041.0685023 -0.18422-0.4320.1127751.8204131.3946824 2.493959-0.49832-0.415312.31615 0.445335 0.876667output neuron wd γd table 4. the relative importance of the ann model input variables. om dgdwinput parameter 14.6541.8843.47relative importance (%) according to smith (1986), if a proposed model provides r> 0.8 and the error values (e.g. mae and rmse) are at the minimum, there is a strong correlation between the measured and predicted values. the training and testing results of the ann model are presented in table 5. the results show that the ann model performs well, with r, rmse, and mae at the values of 0.93, 0.0187, and 0.0219 in the training stage and 0.928, 0.0378, and 0.0404 in the testing stage, respectively. anfis model fig. 5 represents the structure of the proposed anfis model in this study. the fuzzy radius of this model is 0.9, and it is formed of two rules. fig. 5 the schematic of anfis architecture based on sugeno fuzzy model developed in this study fig. 6 compares the anfis predictions with the experimental values of the compression ratio of msw for the training and testing data sets. the proposed model shows a very good correlation to the training and testing data. the predictive performance of the anfis model is summarized in table 5. the results demonstrate that the anfis model with a high r and low rmse and mae values is trained well and able to estimate the target values with an acceptable degree of accuracy. fig.6 measured versus predicted compression ratio of msw using anfis model, (a) training data set, (b) testing data test table 5. performance of the proposed models testingtrainingmodel rmsemaerrmsemaer 0.05150.0480.90.06020.02350.904ann 0.03960.03470.9230.02220.01870.923 anfis 170 ali akbar heshmati r, maryam mok htari, nader shariatmadari prediction results obtained from the ann and anfis models are compared in table 5. the high correlation coefficient and the lower error in both models demonstrate the capability of those models in acceptably predicting the compression ratio. as suggested by the results, the error rates of the anfis model are lower than ann model; so, anfis has performed more precisely. table 5 also shows that the results obtained for each of the models during testing are generally consistent with those obtained during training. this suggests that the proposed models are able to generalize within the range of data used for training. it is noteworthy that, although the predictive capability of the proposed models is limited to the range of the data used for their calibration, if a new data set becomes available, these models can be retrained and updated easily without repeating the development procedures from the beginning. conclusions in this research, anfis and ann models were used to predict the compression ratio of municipal solid waste. the models were developed based on a reliable database obtained from the literature. the following emerge as the conclusions of the study: 1 the proposed models developed through ann and anfis modeling techniques can precisely predict the compression ratio of msw. 2 the relative importance study indicates that dry unit weight and dry weight water content have a significant effect on the compression ratio of msw. 3 a major advantage of ann and anfis models is that, once a new set of data becomes available, they can be easily updated and improved without repeating the development procedures from the beginning. references bareither, c.a., benson, c.h. and tuncer, t.b. (2012). compression behavior of municipal waste: immediate compression, journal geotechnical geoenvironmental engineering. 138, 1047-1062. beaven, r.p. and powrie, w. (1995). determination of the hydrogeological and geotechnical properties of refuse using a large scale compression cell, proceedings of the 5th international sardinia landfill conference, s. margherita dipula, cagliari, italy, 745-760. chen, r.h. and lee, y.s. (1995). settlement analysis of a waste landfill, proceedings of the 3th international symposium on environmental geotechnology, san diego, ca, usa, 539-553. chen, y.m., zhan, t.l.t., wei, h.y. and ke, h. (2009).aging and compressibility of municipal solid wastes, waste management journal. 29, 86-95. chen, r.h., chen, k.s. and liu, c.n. (2010). study of the mechanical compression behavior of municipal solid waste by temperaturecontrolled compression tests, journal of environmental earth sciences. 61, 1677-1690. chiu, s.l. (1994). fuzzy model identification based on cluster estimation, journal of intelligent fuzzy systems. 2, 267-278. durmusoglu, e., sanchez, i. and corapcioglu, m. (2006). permeability and compression characteristics of municipal solid waste samples, environmental geology. 50, 773-786. edgers, l., noble, j.j. and williams, e. (1992). a biologic model for longterm settlements in landfills, proceedings of the environmental geotechnology, balkema, rotterdam, the netherlands, 177-183. el-fadel, m. and khoury, r .(2000). modeling settlement in msw landfills: a critical review, environmental science and technology. 30, 361-327. erzin, y.and gul, t.o. (2012). the use of neural networks for the prediction of the settlement of one way footings on cohesionless soils based on standard penetration test, neural computing and applications, doi 10.1007/s00521-012-1302-x. gabr, m.a. and valero, s.n. (1995). geotechnical properties of municipal solid waste, geotechnical testing journal.18, 241-251. hanna, a. m. derin, u. and saygili, g. (2007). neural network model for liquefaction potential in soil deposits using turkey and taiwan earthquake data. soil dynamics and earthquake engineering, 27, 521-540. heshmati, a.a.r., alavi, a.h., keramati, m. and gandomi, a.h. (2009). a radial basis function neural network approach for compressive strength prediction of stabilized soil. geotech spec pub asce, changsha hunan, china, 191, 147–153. hornik, k., stinchcombe, m. and white, h. (1989). multilayer feed forward networks are universal approximators, neural network journal. 2, 359-366. hyun ii, p., borinara, p. and hong, k.d. (2011). geotechnical considerations for end-use of old municipal solid waste landfills, international journal environmental research. 5, 573-584. jaksa, m.b., maier, h.r. and shahin, m.a. (2008). future challenges for artificial neural network modeling in geotechnical engineering, the 12th international conference of international association for computer methods and advances in geomechanics (iacmag), geo, india, 1710–1719. jang, j.s.r., sun, c.t. and mizutani, e. (1997). neuro-fuzzy and soft computing,prentice-hall, new jersey. jang, j. s. r. and sun, c. t. (1995). neuro-fuzzy modeling and control, proceedings of the ieee, 378–406. kalkan, e., akbulut, s.,tortum, a. and celik, s. (2009). prediction of the unconfined compressive strength of compacted granular soils by using inference systems, environmental geology. 58, 1429–1440. karimpour-fard, m. and machado, s.l. (2012). deformation characteristics of msw materials, electronic journal geotechnical engineering. 17(a), 2009-2024. khanlari, g.r., heidari, m., momeni, a.a. and abdilor, y. (2012). prediction of shear strength parameters of soils using artificial neural networks and multivariate regression methods, engineering geology, 131-132, 11-18. kucuk, n. (2008). modeling of gamma-ray energy absorption buildup factors using neural network, ann. nucl. energy. 35, 1787–1790. landva, a.o., valsangkar, a.j. and pelkey, s.g. (2000). lateral earth pressure at rest and compressibility of municipal solid waste, canadian geotechnical journal. 37, 1157-1165. noori, r., abdoli, m.a., farokhnia, a. and abbasi, m. (2009). results uncertainty of solid waste generation forecasting by hybrid of wavelet transform-anfis and wavelet transform-neural network, expert systems with applications. 36, 9991–9999. olivier, f., gourc, j.p., lopez, s., benhamida, s. and van-wyck, d. (2003). mechanical behavior of solid waste in a fully instrumented prototype compression box, proceedings of the 9th international waste management and landfill symposium, cagliari, italy, 1-12. olivier, f. and gourc, j.p. (2007). hydro-mechanical behavior of municipal solid waste subject to leachate recirculation in a large-scale compression reactor cell, waste management journal . 27, 44-58. ozer, m., isik, n.s. and orhan, m. (2008). statistical and neural network assessment of the compression index of clay-bearing soils, bull eng geol environ. 67, 537–545. padmini, d., ilamparuthi, k. and sudheer, k.p. (2008). ultimate bearing capacity prediction of shallow foundations on cohesionless soils using neurofuzzy models, computers and geotechnics. 35, 33-46. park, h.l. and lee, s.r. (2011). evaluation of the compression index of soils using an artificial neural network, computers and geotechnics. 38, 472–481. reddy, k.r., hettiarachchi, h., parakalla, n.s., gangathulasi. and j., bogner, j. e. (2009a). geotechnical properties of fresh municipal solid waste at orchard hills landfill, usa, waste management. 29, 952-959. reddy, k.r., gangathulasi, j., parakalla, n.s., hettiarachchi, h., bogner, j.e. and lagier, t. (2009b). compressibility and shear strength of municipal solid waste under short-term leachate recirculation operations, waste management and research. 27, 578-587. 171compression ratio of municipal solid waste simulation using artificial neural network and adaptive neurofuzzy system reddy, k.r., hettiarachchi, h., gangathulasi, j., bogner, j. and lagier,t. (2009c). geotechnical properties of synthetic municipal solid waste, international journal of geotechnical engineering. 3, 429-438. reddy, k.r., hettiarachchi, h., gangathulasi, j. and bogner, j. e. (2011). geotechnical properties of municipal solid waste at different phases of biodegradation, waste management.31, 2275-2286. sezer, a. (2013). simple models for the estimation of shearing resistance angle of uniform sands, neural computing and applications. 22,111-123. smith, g.n. (1986). probability and statistics in civil engineering: an introduction. collins, london. sowers, g. (1973). settlement of waste disposal fills, proceedings of the 8th international conference on soil mechanics and foundation engineering, balkema, rotterdam, 207-210. sugeno, m. and kang, g.t. (1988). structure identification of fuzzy model, fuzzy sets systems. 28, 15–33. stoltz, g., and gourc, j. p. (2007). influence of compressibility of domestic waste on fluid permeability, proceedings of the 11th international waste management and landfill symposium, cagliari, italy, cisa, environmental sanitary engineering centre, italy, 1-8. stoltz, g., gourc, j.p. and oxarango, l. (2010). characterisation of the physico-mechanical parameters of msw, waste management journal. 30, 1439-1449. swati, m. and joseph, k. (2008). settlement analysis of fresh and partially stabilized municipal solid waste in simulated controlled dumps and bioreactor landfills, waste management journal. 28, 1355-1363. venkatesh, k., kumar, v. and tiwari, r. p. (2013). appraisal of liquefaction potential using neural network and neuro fuzzy approach, applied artificial intelligence: an international journal. 27, 700-720. vilar, o.m. and carvalho, m. f. (2004). mechanical properties of municipal solid waste, testing and evaluation journal. 32, 438-449. wall, d. and zeiss, c. (1995). municipal landfill biodegradation and settlement, environmental engineering journal. 121, 214-224. keywords: artificial intelligence; low permeability reservoir; petroleum logging; lithology identification palabras clave: inteligencia artificial; embalse de baja permeabilidad; explotación de petróleo; identificación de litología how to cite item shang, f., cao, m., & wan, c. (2021). application of artificial intelligence in lithology recognition of petroleum logging in low permeability reservoirs. earth sciences research journal, 25(2), 255-262. doi: https://doi.org/10.15446/ esrj.v25n2.80895 in low permeability reservoirs, the conversion accuracy of the existing petroleum logging lithology identification method to small pore capillary pressure curve is not high, resulting in a low rock mass identification accuracy. therefore, artificial intelligence technology is considered in this study to enhance the accuracy of lithology identification in low permeability reservoirs. firstly, the radar mapping program is used to predict the position of reservoir oil logging, and then the small pore capillary pressure curve is converted by using the conversion method of piecewise power function scale to obtain the pore characteristics of low-permeability reservoir rocks. on this basis, the crossplot method is used to gather the pore characteristic data in well logging and form a plan, and the response parameters of well logging rock mass are obtained to realize the identification and analysis of lithology. the experimental results show that, compared with the existing identification methods, the accuracy of lithology identification in low-permeability reservoir logging is significantly increased after the application of artificial intelligence technology, and the identification process takes less time, which fully proves that the application of artificial intelligence technology is conducive to improving the performance of lithology identification. abstract application of artificial intelligence in lithology recognition of petroleum logging in low permeability reservoirs aplicación de la inteligencia artificial en reconocimiento litológico de extracción de petróleo en yacimientos de baja permeabilidad issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v25n2.80895 en los yacimientos de baja permeabilidad, la precisión de conversión del método de identificación de litología de extracción de petróleo existente a la curva de presión capilar de poros pequeños no es alta, lo que da como resultado una baja precisión de identificación del macizo rocoso. por lo tanto, en este estudio se considera la tecnología de inteligencia artificial para mejorar la precisión de la identificación litológica en yacimientos de baja permeabilidad. en primer lugar, el programa de mapeo de radar se usa para predecir la posición del registro de petróleo del yacimiento, y luego la curva de presión capilar de poros pequeños se convierte utilizando el método de conversión de la escala de función de potencia por partes para obtener las características de los poros de las rocas del yacimiento de baja permeabilidad. sobre esta base, se utiliza el método de parcelas cruzadas para recopilar los datos característicos de los poros en el registro de pozos y formar un plan, y se obtienen los parámetros de respuesta del macizo rocoso del registro de pozos para realizar la identificación y análisis de la litología. los resultados experimentales muestran que, en comparación con los métodos de identificación existentes, la precisión de la identificación litológica en el registro de yacimientos de baja permeabilidad aumenta significativamente después de la aplicación de tecnología de inteligencia artificial, y el proceso de identificación lleva menos tiempo, lo que demuestra plenamente que la aplicación de la tecnología de inteligencia artificial es propicia para mejorar el rendimiento de la identificación litológica. resumen record manuscript received: 05/07/2019 accepted for publication: 02/10/2019 earth sciences research journal earth sci. res. j. vol. 25, no. 2 (june, 2021): 255-262 fuhua shang1, maojun cao1*, caizhi wang2 1school of computer & information technology, northeast petroleum university, daqing, 163318, china 2department of well logging & remote sensing technology, research institute of petroleum exploration and development, beijing, 100083, china *corresponding author: caomj007@126.com g e o l o g y mailto:caomj007@126.com 256 fuhua shang, maojun cao1, caizhi wang introduction with the development and progress of the society, the demand for energy in various fields is constantly increasing. therefore, petroleum geological exploration technology is facing more severe challenges. multi channel, multi means, multi discipline cross fusion has become the development trend of exploration and development technology (deng et al., 2017). the early geological exploration objects in china are mainly sand and mudstone reservoirs, and lithology identification is relatively simple. with the continuous development of geological exploration and development, the requirement of well logging technology is gradually improved, and in order to form a more reasonable well logging interpretation method, lithology identification technology is also improved. lithology identification belongs to the category of logging engineering and is an important research content in reservoir description, formation evaluation and drilling monitoring. the commonly used methods to obtain formation lithologic information include cuttings logging, coring section analysis and comprehensive treatment of logging data (wei et al., 2017). among them, the quality of cuttings logging depends on the operation of technicians, which is greatly influenced by human factors. the process of core slice analysis is time-consuming and laborious, and it is difficult to describe the complete well formation profile. in terms of comprehensive processing of logging data, crossplot method can intuitively reflect the boundary interface and lithological distribution area of different lithologies, and has the advantage of intuitive reading. however, its operation process is complex, and it cannot identify the whole well or interpret lithological information of the well. the mathematical statistical method is suitable for well logging data with good physical characteristics, and has a good application effect when there are few core data and more logging data, but it is very difficult to obtain the empirical formula (xie et al., 2018; jia et al., 2018; jiang et al., 2018). artificial intelligence is a branch of computer science. it is a technology used to simulate, extend and expand human intelligence. it can simulate human consciousness, thinking and information formation process. research in this field includes robotics, language recognition, image recognition, natural language processing and expert systems. it can be said that artificial intelligence is a very challenging science, which requires not only computer knowledge, but also knowledge in fields such as psychology and philosophy (deng et al., 2018). artificial intelligence research, on the whole, one of the main goal is to make the machine to be able to do some human intelligence is needed to complete a complex work, namely through the use of computers to simulate human thinking process and intelligent behavior of the discipline, can say it is almost covered all subjects of natural science and social science, its scope has been far beyond the scope of computer science (guzman & aoyama, 2018). in recent years, artificial intelligence has become one of the main methods of pattern recognition in various fields. for example, bp neural network and fuzzy cluster analysis method. although the bp neural network method has some disadvantages such as local minimization, slow convergence rate and different structure selection, it has the ability of distributed processing, self-learning, self-organization, highly nonlinear mapping and fault tolerance, which greatly makes up for the deficiency of self-recognition performance. therefore, in order to improve the ability of identifying lithology of oil logging in low permeability reservoirs, artificial intelligence technology is introduced in this study. application research on reservoir oil logging location prediction when predicting the location of reservoir oil logging, first of all, do not collect logging data. in this process, various interference factors inevitably exist, which affect the resolution of the signal. linear combination of sensitive indexes with high correlation is an effective method to improve the accuracy of logging interpretation. therefore, on the basis of selecting the logging data of the core well, the logging data is compared with the core data by the multi-index radar chart comparison method. the execution flow of the radar chart drawing program is shown in figure 1. according to the implementation process of the radar chart drawing program as shown in figure 1, sensitive indexes that can reflect the lithologic change characteristics are selected, and then the correlation coefficient of the selected sensitive indexes is obtained, and the indexes with good correlation are combined linearly. although the noise is random in the radar map drawing process, the linear combination process can enhance the formation characteristic information, and highlight the useful information by superposition signal, thus reducing the signal interference (han et al., 2019). the dimensions of different logging indicators are often different. in order to combine the indicators of different dimensions, it is necessary to normalize the indicators and then combine the dimensionless data to form new analysis indicators. firstly, the radar map is used to analyze the logging data, and the log phase identification curve set is established through the standard well. then, the logging location of the reservoir of the unknown well segment is identified by using the curve difference comparison, and the logging data of the core well is obtained after sorting out and correcting the analysis data, and the clustering normalization is completed. after calculation, the following results can be obtained: x x x x x = −∗ ∗ ∗ ∗ min max min (1) among them, x represents the normalized log curve, x* represents the original log data, x*min and x * max represent the maximum and minimum values of the log curve respectively. it is easy to realize the linear normalization method with the powerful matrix operation of matlab, but in the transverse comparative analysis of multiple wells, the difference of normalization results will be caused by the different selection of maximum and minimum values, thus affecting the analysis results. therefore, data expansion or shift processing is carried out in this paper according to the value situation (zhou et al., 2019; jia & deng, 2018). for example, after analyzing logging and core data, it can be seen that a certain reservoir is located in the second section, which is dominated by grayish-white medium-grain lithic sandstone, and its density (den) value is mostly around 2~3 g/cc. at this point, if we can extract the fractional part, magnify the difference between maximum and minimum, we will get a better comparison effect. on this basis, all the above density measurements were summarized and five parameters of ac, den, cnl, gr and rd were selected to depict the logging phase position radar map of the traveling gas layer, gas bearing layer and gas layer, as shown in figure 2. figure 1. radar chart drawing program execution flow diagram 257application of artificial intelligence in lithology recognition of petroleum logging in low permeability reservoirs figure 2. logging locations corresponding to different density values table 1. rock mass characteristic data table serial number of rock mass porosity/ % t2cutoff/ ms nmr bound water saturation/ % maximum throat radius /μm mean throat radius/μm sorting coefficient expulsion pressure/ mpa t2lm/ ms permeability/ 10-3μm-2 12-1 6.71 22.68 64.26 11.92 0.44 0.14 0. 26 1.70 0.07 61-1 6.77 21.99 65.77 11.41 0.60 0.16 0.31 1. 26 0.13 64-1 7.89 11.01 73.22 575 0.55 0.05 -0.31 1.36 0.11 62r 13.50 16.26 43.01 21.25 1.00 0.05 -0.43 0.75 0. 56 18.00 6.90 7.90 42.21 11.06 1.22 0.04 -0.66 0.61 0. 28 51b 10.60 18.00 40.60 26.24 1.15 0.12 -0.18 0.65 0.42 9-35 6.99 2.66 52.40 3.96 1.19 0.02 -0.94 0.63 0.68 70-1 5.62 1.55 57.21 1.93 0.33 0.00 -0.86 2.28 0.08 19-1 9.73 16.79 41.10 28.16 1.17 0.06 -0.31 0.64 0.23 34-4 12.77 5.85 46.97 9.78 2.60 0.21 -0.20 0.29 3.59 52-1 11.37 10.54 83.37 3.01 0.34 0.05 -0.42 2.21 0.15 64-1 8.12 7.41 64.15 6.47 0.89 0.07 -0.46 0. 84 0.41 93-1 16.65 8. 26 42.68 17.19 20.06 0.17 -0.53 0. 20 42.30 115-1 14.92 10.39 52.38 10.46 3.81 0.03 -0.74 0.56 4. 85 9-1 11.57 6. 80 61.11 5.54 1.35 0.07 -0.32 0.17 0.91 20-1 11.76 6.67 46.52 9.32 4.29 0.91 0.41 0.98 17.27 68-1 6.38 2.75 85.97 1.09 0.76 0.02 -0. 94 1.39 157 137.00 10.45 7.77 79.32 2. 98 0.54 0.03 -0.75 1.40 0.79 152-1 8.35 40.30 83.07 7.01 0.54 0.04 -0.48 0.98 0.87 3-1 6.26 27.35 67.35 9.17 0.76 0.30 0.81 2.01 0.63 30-1 5.52 1.68 91.38 0. 87 0.37 0.04 -0.40 0.22 0.52 91-1 10.56 13.81 45.54 15.45 3.35 0.19 -0.20 1. 26 3.38 144-1 8.21 61.52 67.08 35.12 0.60 0.10 -0.02 1.09 0.70 216-1 9.47 52.66 59.96 41.56 0.69 0.17 0.23 1.23 0.97 236-1 8.17 36.92 62.34 27.80 0.61 0.16 0.34 1. 20 0.85 257-1 6.90 46.74 56.18 42.06 0.63 0.21 0.62 0. 20 0.42 according to the different density ranges shown in figure 2, it is obvious that cnl can reflect different density differences to a certain extent, which corresponds to the location of oil logging in low-permeability reservoirs. samples of rock sections are randomly selected within the determined location range to determine the pore characteristics of rocks in low-permeability reservoirs. pore characteristics of low permeability reservoir rock when judging non-uniform properties of low permeability reservoirs, pressure pump experiments should be carried out in low permeability reservoirs according to thin sections obtained by well logging, and rock characteristic data information should be obtained by nuclear magnetic test. the characteristic data are shown in table 1. referring to the data information in table 1, the “three porosity component percentage method” was used to process the above 26 rock samples, to qualitatively classify the pore types of the rock samples, and to draw the corresponding cross plot of pore structure parameters (wang & siau, 2019), as shown in figure 3. by referring to the cross plot of pore structure parameters shown in figure 3, it can be seen that the t2 spectrum measured by nmr consists of three kinds of transverse relaxation time, that is, the t2 spectrum observed is: 1 1 1 1 t t t tb d s2 2 2 2 = + + (2) 258 fuhua shang, maojun cao1, caizhi wang where, t2 denotes transverse relaxation time, t2b denotes volumetric (free) relaxation time of the fluid, t2d denotes diffusive relaxation time, and denotes surface relaxation time. t2s and t2d were characterized by specific surface area, diffusion coefficient, magnetic field gradient, echo interval and other parameters, and then: 1 1 122t t s v d gt b e 2 2 2 '= + + ( ) ρ γ      (3) among them,  2 represents the transverse surface relaxation strength of the rock, s represents the surface area of pores, v represents the volume of pores, v represents the specific surface area of characterized rock samples, d represents the diffusion coefficient, g represents the magnetic field strength, te represents the echo interval,  represents the magnetic rotation ratio, and the remaining parameters remain unchanged. for the uniform magnetic field with a small echo interval, equation (2) can be transformed into: ρ ρ 2 21s v t2      = =' f rs c (4) where, fs represents the geometric shape factor, and for spherical pores, the value of fs is 3; for columnar pores, fs has a value of 2. rc represents the pore radius. generally speaking, the relaxation time of pore fluid is related to the size and shape of pore space (grzonka et al., 2018; gobashy et al., 2020). since there should be a corresponding transverse relaxation time for a particular rc , t2 and rc should have a one-to-one correspondence. however, due to the constant change of fs , the corresponding relationship between t2 and rc also changes. for medium high porosity and permeability reservoirs, reservoir heterogeneity is weak, and the corresponding relationship between t2 and rc has little change. however, for tight sandstone reservoirs with complex pore structure and pore type, the corresponding relationship between t2 and rc changes greatly due to their strong reservoir heterogeneity. according to fluid mechanics, there is a certain relationship between capillary pressure and pore radius: p rc c = 2σ θcos (5) where, pc represents capillary pressure,  represents fluid interfacial tension, and  represents wetting contact angle. for mercury-air system, equation (4) can be rewritten as: p rc c ' .  0 735 (6) the comprehensive calculation formulas (3) and (5) can be obtained as follows: c pc= ×t2 (7) where, c represents the conversion coefficient, and. therefore, if the conversion coefficient can be accurately calculated by using the well log, the capillary pressure curve at different depths of the whole well can be evaluated by using the nmr log, and the variation of various pore structure parameters with depth can also be evaluated (zibret, 2019). generally, the pseudo-capillary pressure curve and the capillary pressure curve of the pore segment cannot be superimposed together, because of the accumulation of thin film bound water with large pores in t2 distribution to small pores (zhang, 2019). based on this idea, a piecewise power function calibration method is proposed to solve the problem of low precision of pressure curve conversion of small pore capillary. in theory, the specific surface area has a linear relationship with the pore size, but for the actual formation, the pore structure is very complex, and the specific surface and pore size often have a nonlinear relationship, which can be expressed as follows: f rc( )= ρ2 1 t2 (8) where, f (rc ) represents the function of rc . according to equation (5) and equation (7), the function of t2 spectrum can be calculated as follows: g pc 1 t2      = (9) according to the above formula, the pore characteristics of lowpermeability reservoir rocks can be divided into large pores and small pores, and the pore characteristics can be described as follows: p a p a cb b cs b =       =               1 2 2 1 1 1 t t 2 2 (10) where, a1 and b1 represent the corresponding parameters of small pores, and a2 and b2 represent the corresponding parameters of large pores. referring to the pore characteristics obtained, crossplot method can be used to obtain logging response characteristics that can represent lithology. obtain logging rock mass response parameters the cross-plot method is used to summarize the pore data obtained above and form a plan. according to the coordinates of the intersection point, the numerical value and aggregation trend of the parameters can be seen. in the process of using the crossplot method to identify lithology, two groups of appropriate logging data are first selected, and coordinate points are drawn in the crossplot coordinate system according to the changes of logging data. the aggregation trend of coordinate points is expressed in the form of intuitive data, and the distribution range and boundary of different lithology can be reflected visually. the crossplot drawn should be compared with the coring lithology data for better accuracy of lithology identification. cross-graphs such as gr-pb and gr-ac can be made according to the core analysis data. according to these cross-graphs, the positions of different measured lithologic points, the lower limit of lithologic logging parameters and the lithologic distribution trend can be determined. referring to logging rock masses of different densities and using the ratio of spectral intensity of elements, structural logging rock characteristic parameters are as follows: figure 3. cross plot of pore structure parameters 259application of artificial intelligence in lithology recognition of petroleum logging in low permeability reservoirs z r sp = ×φ0 5421 1 0458 0 3181 . . . (11) among them, z represents logging rock characteristic parameters,  represents porosity calculated above, r represents average pore throat radius, sp represents sorting coefficient, and two constants represent permeability measured. logging rock characteristic parameters were defined to be around 80-110, and the acoustic wave presented high value, which was generally above 67, with the development of porosity structure. when the characteristic parameters of logging rocks are above 120 and the acoustic values are generally above 70, the fused tuff appears in the logging interior. when the characteristic parameters of logging rocks are distributed around 40-120 and the acoustic values are generally below 65, the characteristic values cannot be determined, and the rock mass is also a fused tuff. according to the calculated logging rock characteristic parameters, the cross map of gr-ρb within the logging location was obtained, as shown in figure 4. as can be seen from the logging characteristic parameters shown in figure 4, different rock assemblages have great differences in layers, and their electrical properties are also different. therefore, before obtaining the response characteristics in the logging, linear transformation should be adopted to normalize the electrical properties in the logging characteristic parameters, and the standardization process is as follows: x x x x n x n x x t n i n i i i i n n n i i n i n n i = − = = − −           = = ∑ ∑ σ σ 1 1 1 1 2 1 ( )  (12) where, where, 111 represents the n the standardized value of xi , x̄i represents the mean value of the electrical variable, i represents the standard deviation of the electrical variable, and n represents the number of logging characteristic parameters. the response values of acoustic time difference (ac), natural gamma (gr) and resistivity (rt) obtained by using the above standardized electrical property values combined with the curve overlap method are comprehensively judged. the difference, acoustic time difference (ac), natural gamma (gr) and resistivity (rt) of porosity d and n extracted by the overlapping method are shown in table 2. the data shown in table 2 are integrated into the calculation formula of logging response characteristic parameters, as shown below: n m t t nf nma ma f f ma ma f = − − = − −        φ φ ρ ρ ρ ρ ∆ ∆ (13) among them, tf represents acoustic time difference of pore fluid, tma represents acoustic time difference of rock skeleton, ma represents pore fluid density, ma represents neutron porosity of rock skeleton, and nf represents neutron porosity of pore fluid. by referring to the calculation results of different characteristic parameters, the response values of corresponding logging lithology were calculated, as shown in table 3. table 3. response values of logging lithology logging lithology ac (us/ft) den (g/cm3) cnl (p.u.) gr (api) cloud mass limestone 45.25~47 2.17~2.79 -1~0 10 gray dolomite 43.5~45.25 2.79~2.81 0~1 10 limestone 48 2.71 -1 10 dolomite 43.50 2.87 1 10 argillaceous limestone 59 2.12 21.50 57.50 anhydrite 49.30~51.80 2.91~3.02 -2 1.50~6 paste dolomite 43.50~46.75 2.87~2.91 -1~0.50 5.80~10 cream quality limestone 49 2.71~2.915 -1.50~1 5.80~10 corresponding to the response values of different characteristics shown in table 3, the properties of different logging rocks can be determined. thus, the application of artificial intelligence in lithology identification of low permeability reservoirs is completed. simulation experiment and result analysis in order to verify the practical application performance of the low permeability reservoir logging lithology identification method designed in this paper by using artificial intelligence, the following simulation experiment is designed for verification.figure 4. gr-ρb cross map of oil logging in low permeability reservoir table 2. parameter set matrix table overlapping layer d − n the mean ac(us/ft) the mean gr(api) the mean rt(·m) 1 -0.179 53.094 49.902 230.441 2 -0.143 53. 512 54.729 235.899 3 -0.135 53. 539 52.784 238.887 4 -0.132 53. 539 47.642 242.040 5 -0.089 52. 870 42.829 253.606 6 -0.011 51.491 39.953 263.617 7 -0.056 50.794 41.874 268.003 8 -0.081 50.525 45.082 285.795 9 -0.105 50.318 47.851 321.116 10 -0.108 50.347 45.627 337.462 11 -0.110 50.453 45.927 339.846 12 -0.109 50.542 43.827 294.277 13 -0.121 50.374 42.100 263.586 14 -0.085 53.412 49.902 213.427 260 fuhua shang, maojun cao1, caizhi wang the experiment to prepare firstly, a basic libs detection system is set up, which mainly includes light source, light splitting system, photoelectric detector, timing control module, collimation, focusing light path, light collection lens group, sample table and so on. imaging of laser-induced plasma is also required during the experiment, so imaging modules should be set up in the libs detection system. the experimental platform structure is shown in figure 5. figure 5. experimental platform structure diagram as shown in figure 5, the main performance parameters of the me5000 medium-step grating spectrometer in the experimental platform are shown in table 4. table 4. main performance parameters of me5000 medium step grating spectrometer parameter data escher spectrometer focal length (mm) 195 aperture f7 wavelength range (nm) 230~950 wavelength accuracy (nm) ±0.05 focal plane size (mm) 13.3×13.3 stray light 1.5×10-4 iccd active pixel 1024×1024 effective pixel size (nm) 13 effective area (mm) 13.3×13.3 peak quantum efficiency up to 18% total insertion delay (ns) ~35 ddg insert delay (ns) ~19 gate delay 0~25s delayed resolution 25ps width of the gate 1ns~25s width resolution 25ps under the control of the performance parameters shown in table 4, 8 logging blocks as shown in figure 6 are prepared. use as shown in figure 6 to mudstone, shale, sandstone, dolomite, asphalt, volcanic ash, conglomerate and paste rock, in the experimental environment, were used in this paper, design of using artificial intelligence low permeability oil reservoir logging lithology identification method and the traditional lithology recognition method based on laser induced breakdown spectroscopy, logging lithology recognition method based on boosting tree algorithm comparing experiment, from lithology recognition accuracy and recognition process takes two angle compared three methods of recognition performance. experimental results and analysis taking mudstone samples as an example, based on the above experimental preparation, the standard reference spectrum of mudstone samples was first obtained by using me5000 medium-step grating spectrometer, as shown in figure 7. figure 7. standard reference spectra of mudstone samples after obtaining the standard spectrum of mudstone samples, under the same experimental conditions, used respectively in this paper, the design of the use of artificial intelligence in the low permeability oil reservoir logging lithological identification method and the traditional lithology recognition method based on laser induced breakdown spectroscopy, logging lithology recognition method based on boosting tree algorithm for mudstone samples of spectral information. in order to avoid the spectral line position drift caused by temperature change, the experimental room temperature should be controlled within the range of 25-30 °c. on this basis, the similarity between the spectral information of mudstone samples obtained by three different identification methods and the standard spectra was compared. when the position and relative strength of spectral lines between the three different identification methods were consistent, the recognition was deemed as a success. the recognition results obtained by the three recognition methods are shown in table 5. according to the experimental results shown in table 5, the three identification methods can be used to identify the same number of different types of logging rock mass information, and the three identification methods figure 6. experiment to prepare rock samples 261application of artificial intelligence in lithology recognition of petroleum logging in low permeability reservoirs show different identification capabilities. after applying the logging lithology identification method based boosting tree algorithm, the number of rock masses identified is too small and the identification accuracy is low. however, the identification accuracy of lithology based on laser-induced breakdown spectroscopy has been improved, but the information of other rock masses has not been fully identified except for the gypsum rock. after the application of this method, except dolomite and shale, the information of other rock mass is correctly identified, and the correct identification rate of this method is the largest among the three methods. based on the above experimental results, it can be seen that the rock mass identification method designed in this paper with artificial intelligence technology can avoid the deficiency of identifying too few rock masses, and the identification accuracy is higher, which is suitable for the identification of lithology of oil logging in low-permeability reservoirs. on this basis, the identification performance of the three methods is verified by taking the identification process time as the index. the results are shown in figure 8. figure 8. time consuming comparison diagram of different methods to identify the process according to the experimental results shown in figure 8, the three identification methods can be used to identify the same number of different types of logging rock mass information, and the time of the three identification methods is greatly different. the well logging lithology identification method based on boosting tree algorithm has a identification process of 8-12min. the lithology identification method based on laser induced breakdown spectroscopy technology is higher and keeps between 12-15min. however, the identification process of the method in this paper takes the least time among the three methods, which is always kept below 7min. based on the above experimental results, it can be seen that the rock mass identification method designed in this paper with artificial intelligence technology can quickly identify the lithologic characteristics of oil logging in low-permeability reservoirs. conclusion automatic identification of lithology is based on the combination of mathematical theory and computer software. this process fully embodies the interdisciplinary application, which can be used as a supplement to the conventional lithology identification method and promote the development of the lithology identification technology with multiple approaches, methods and angles. the appearance of artificial intelligence lithology identification software plays a powerful role in promoting the technology of complex formation lithology classification. in this paper, the application of artificial intelligence technology to low permeability reservoir oil logging lithology identification is proposed. it has been proved by practice that this method can effectively improve the accuracy of lithology identification in low permeability reservoir, and the identification process is time-consuming. acknowledgments this paper is supported by national science and technology major project “well logging interactive fine fusion processing platform” (no. 2017zx05019-005), heilongjiang natural science foundation “quantum intelligent optimization and application in real-time adjustment of logging interpretation model while drilling” (no. lh2019f004). reference deng, l. (2018). artificial intelligence in the rising wave of deep learning: the historical path and future outlook [perspectives]. ieee signal processing magazine, 35(1), 180-177. doi: 10.1109/ msp.2017.2762725 deng, c. x., pan, h. p., & luo, m. (2017). joint inversion of geochemical data and geophysical logs for lithology identification in ccsd main hole. pure and applied geophysics, 174(12), 4407-4420. https://doi. org/10.1007/s00024-017-1650-7 gobashy, m., abdelazeem, m., abdrabou, m., & khalil, m. h. (2020). estimating model parameters from self-potential anomaly of 2d inclined sheet using whale optimization algorithm: applications to mineral exploration and tracing shear zones. natural resources research, 29 (1), 499-519. https://doi.org/10.1007/s11053-019-095260 grzonka, d., jakóbik, a., koodziej, j., & pllana, s. (2018). using a multi-agent system and artificial intelligence for monitoring and improving the cloud performance and security. future generation computer systems, 86(sep.), 1106-1117. doi:10.1016/j.future.2017.05.046 guzman, a., & aoyama, a. (2018). pipeline risk assessment using artificial intelligence: a case from the colombian oil network. process safety progress, 37(1), 110-116. doi:10.1002/prs.11890 table 5. statistical table of identification results of different methods rock mass classification number of test samples /number logging lithology identification method based on boosting tree algorithm method of this paper lithology identification method based on laser induced breakdown spectroscopy correctly identify quantity/number recognition accuracy/% correctly identify quantity/number recognition accuracy/% correctly identify quantity/number recognition accuracy/% mudstone 8 4 50 8 100 6 75 sandstone 8 3 37.5 8 100 5 62.5 shale 8 4 50 7 87.5 6 75 volcanic ash 8 3 37.5 8 100 7 87.5 dolomite 7 3 42.8 6 85.7 5 71.4 cream rock 2 1 50 2 100 2 100 conglomerate 7 5 71.4 7 100 6 85.7 asphalt 4 2 50 4 100 3 75 http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(chengxiang+deng) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(chengxiang+deng) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(chengxiang+deng) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(miao+luo) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(miao+luo) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(mohamed+gobashy) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(mohamed+gobashy) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(mohamed+gobashy) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(maha+abdelazeem) http://scholar.cnki.net/result.aspx?q=%e4%bd%9c%e8%80%85:(mohamed+abdrabou) 262 fuhua shang, maojun cao1, caizhi wang han, f. l., zhang, h. b., guo, q., & rui, j. (2019). lithological identification with probabilistic distribution by the modified compositional kriging. arabian journal of geosciences, 12(18), 1-14. doi:10.1007/s12517019-4775-4. jia, h., & deng, l. h. (2018). water flooding flowing area identification for oil reservoirs based on the method of streamline clustering artificial intelligence. petroleum exploration and development, 45(02), 328335. https://doi.org/10.1016/s1876-3804(18)30036-3 jia, j. w., fu, h. b., & wang, h. d. (2018). lithology identification methods based on laser-induced breakdown spectroscopy technology. chinese journal of quantum electronics, 35(03), 264-270. jiang, k., wang, s. d., & hu, y. j. (2018). lithology identification model by well logging based on boosting tree algorithm. well logging technology, 42(04), 395-400. wang, w. y., & siau, k. (2019). artificial intelligence, machine learning, automation, robotics, future of work and future of humanity: a review and research agenda. journal of database management, 30(1), 61-79. doi:10.4018/jdm.2019010104 wei, j. l., liu, x. n., ding, c., liu, m., jin, m., & li, d. (2017). developing a thermal characteristic index for lithology identification using thermal infrared remote sensing data. advances in space research, 59(1), 7487. https://doi.org/10.1016/j.asr.2016.09.005 xie, y. x., zhu, c. y., zhou, w., li, z., liu, x., & tu, m. (2018). evaluation of machine learning methods for formation lithology identification: a comparison of tuning processes and model performances. journal of petroleum science and engineering, 16(03), 182-193. https://doi. org/10.1016/j.petrol.2017.10.028 zhang, l., liang, y. c., & niyato, d. (2019). 6g visions: mobile ultrabroadband, super internet-of-things, and artificial intelligence. china communications, 16(8), 1-14. doi: 10.23919/jcc.2019.08.001 zhou, y., zhang g. z., gao g., zhao, w., yi, y., & wei, h. (2019). application of kernel principal component analysis in lithologic identification of well logging turbidite. oil geophysical prospecting, 54(03), 667-675, 490. zibret, g. (2019). influences of coal mines, metallurgical plants, urbanization and lithology on the elemental composition of street dust. environmental geochemistry and health, 41(3), 1489-1505. doi: 10.1007/s10653018-0228-3 about:blank about:blank about:blank about:blank about:blank about:blank about:blank about:blank about:blank https://doi.org/10.1016/s1876-3804(18)30036-3 http://www.wanfangdata.com.cn/details/javascript:void(0) http://www.wanfangdata.com.cn/details/javascript:void(0) http://www.wanfangdata.com.cn/details/javascript:void(0) http://www.wanfangdata.com.cn/details/javascript:void(0) http://dx.doi.org/10.4018/jdm.2019010104 https://doi.org/10.1016/j.asr.2016.09.005 https://doi.org/10.1016/j.petrol.2017.10.028 https://doi.org/10.1016/j.petrol.2017.10.028 with the continuous improvement of precision requirements for borehole geophysical exploration, the application of transient electromagnetic method (from now on referred to as tem) in a borehole has become a hot spot. the conventional borehole tem can only determine the longitudinal depth of the geological anomaly, the radial azimuth and depth cannot be resolved. a double-transmitting and sextuple-receiving borehole tem is proposed, through which the radial anomaly is excited by the electromagnetic field generated by the double-emitting loops, and the azimuth and depth of the anomaly will be identified by the difference characteristics of the six receiving loops signals. in this paper, the response equations of the transmitting-receiving mode of double-transmitting and sextuple-receiving borehole tem are deduced, and the response characteristics of the induction segment and the attenuation segment of the receiving loops are obtained based on the response equations under ramp function turnoff condition, providing the basis for theoretical analysis. due to the negative value of the double-transmitting and sextuple-receiving transient electromagnetic response signals, a negative transformation algorithm under the double logarithmic coordinate system is proposed to provide the essential method for the analysis of two kinds of physical simulation experimental data of the radial azimuth and radial depth detection of the anomaly. the results show that the double-transmitting and sextuple-receiving borehole tem has decent resolution ability in detecting the radial azimuth of the anomaly, and the effective resolution is 30°. the geometric difference among induced voltages of different measuring points can be used to evaluate the radial depth of the anomaly qualitatively. it is expected that the double-transmitting and sextuple-receiving borehole tem can provide technical guidance for little borehole geophysical exploration in the fields of oil, natural gas, coal and basic engineering construction. abstract keywords: borehole transient electromagnetic method; double-transmitting and sextuplereceiving; negative transformation algorithm; radial azimuth; radial depth. double-transmitting and sextuple-receiving borehole transient electromagnetic method and experimental study issn 1794-6190 e-issn 2339-3459 earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017): 77 83 bo wang1, shengdong liu1*, shining li3, fubao zhou2 1state key laboratory of deep geomechanics & underground engineering and school of resource and earth science, china university of mining and technology, xuzhou 221116, china 2school of safety engineering, china university of mining and technology, xuzhou 221116, china 3hebei research institute of investigation and design of water conservancy and hydropower, tianjin 300143, china *email of corresponding author: wbsyes@126.com con la necesidad continua de mejorar la precisión en la exploración de perforaciones geofísicas, la aplicación del método electromagnético transitorio (tem, del inglés transient electromagnetic method) se ha convertido en un tema de constantes estudios. la aplicación del tem en perforaciones convencionales solo puede determinar la profundidad longitudinal de las anomalías geológicas, pero no puede resolver el acimut radial y la profundidad. este estudio propone la aplicación del método tem con transmisión doble y recepción séxtuple a través de la cual se altera la anomalía radial con el campo electromagnético generado por la doble emisión de ondas, mientras el acimut y la profundidad de las anomalías se identifican por las características de las señales de las seis ondas de recepción. en este trabajo se dedujo la respuesta de las ecuaciones del modo transmisión-recepción para la transmisión doble y la recepción séxtuple en las perforaciones con el método tem, y se obtuvieron las respuestas características del segmento de inducción y el segmento de atenuación de las ondas de recepción basados en la solución de las ecuaciones en condición de apagado de la gráfica de función, lo que proporciona las bases para el análisis teórico. debido al valor negativo en las señales de respuesta de la transmisión doble y recepción séxtuple del electromagnetismo transitorio, se propone un algoritmo de transformación negativa bajo el sistema de coordenadas logarítmicas dobles para proveer el método esencial del análisis de dos clases de información de simulación física experimental del acimut radial y la detección de la profundidad radial de la anomalía. los resultados muestran que el tem con transmisión doble y recepción séxtuple en perforaciones tiene una buena capacidad de resolución para detectar el acimut radial de la anomalía, con una resolución efectiva de 30 grados. la diferencia geométrica de los voltajes inducidos desde los diferentes puntos de medida se puede utilizar para evaluar cualitativamente la profundidad radial de la anomalía. la expectativa es que el método tem con transmisión doble y recepción séxtuple en perforaciones puede proveer orientación técnica para pequeñas exploraciones de perforación geofísica en los campos de petróleo, gas natural, carbón y construcciones básicas de ingeniería. resumen palabras clave: método electromagnético transitorio; transmisión doble y recepción séxtuple; algoritmo de transformación negativa; acimut radial; profundidad radial. record manuscript received: 01/03/2017 accepted for publication: 01/07/2017 how to cite item: wang, b., liu, s., li, s., zhou, f. (2017). double-transmitting and sextuple-receiving borehole transient electromagnetic method and experimental study. earth sciences research journal, 21(2), 77 83. doi:http://dx.doi.org/10.15446/esrj.v21n2. 63006 estudio experimental del método electromagnético transitorio con transmisión doble y recepción séxtuple http://dx.doi.org/10.15446/esrj.v21n2.63006 p o t e n t ia l m e t h o d s 78 bo wang, shengdong liu, shining li and fubao zhou 1. introduction with the continuous improvement of precision requirements for borehole geophysical exploration, the application of tem in a borehole has become a hot spot (strack, 2014; gan et al., 2016; khoshbakht et al., 2014; karinskiy et al., 2014; gorbatenko et al., 2014). in the study of tem in borehole and well drilling, stolz (2003) examined the borehole electromagnetic method of the gold-bearing tectonic zone, and the borehole electromagnetic method can detect significant anomalies in the shear zone. wilkinson et al., (2005) studied the underground distribution of saline-alkali materials using the aeronautical electromagnetic method, ground electromagnetic methods, and borehole tem, combined with soil, subsurface materials, and groundwater data. school and edwards et al. (2007) studied the resolution of the electromagnetic method and high-resistance targets in submarine borehole dipole-dipole devices. ren et al., (2011) studied the response characteristics of the transient electromagnetic signal in production wells by mean of numerical simulations. wang (2012) supplemented the forward and inverse models of cased well logging and introduced the overall design scheme of the cased well logging system. zhang (2012) studied the design and implementation of high-power electromagnetic pulse source, and detailed description of the proper logging instrument structure and the particular circuit implementation method. guo and song (2012) designed the parameters of the tem receiving loops of the production well by studying the signal response characteristics of the receiving loops. zang et al. (2014) used the open-hole model to study the propagation characteristics of electromagnetic waves in the well and gave a twodimensional spectrum and its dispersion curve describing the propagation of electromagnetic waves in the well. shen et al. (2016) proposed an induction logging method based on transient electromagnetic excitation. borehole tem reduces the size of the excitation source and receiving loops and places them in the borehole for excitation and reception, achieving better coupling between the device and the underground medium and more evident abnormal response. for roslee et al. (2017), the burial depth of the anomaly can be more accurately distinguished by increasing measure points. however, the conventional borehole tem can only determine the longitudinal burial depth of the geological anomaly, the radial azimuth and depth cannot be resolved (liu et al., 2014). to solve this problem, in this paper the response equations of the transmitting-receiving mode of double-transmitting and sextuple-receiving borehole tem will be deduced. a negative transformation algorithm will be proposed, and the borehole tem instrument will be created, and two kinds of physical simulation experiment study of the radial azimuth and radial depth detection of the anomaly will be carried out. 2. principle as shown in figure 1, the probe detects signals with measuring points in the borehole, and it can be seen from the enlarged view that the entire probe consists of two transmitting loops and six receiving loops. the transmitting circuit conveys electromagnetic field with the same value and opposite direction so that the electromagnetic field diffuses to the radial azimuth from the middle of the probe. lindang et al. (2017), as shown in figure 2, the six receiving loops produce a positive primary field response to the magnetic field excited by the transmitting loops. kamsani et al., (2017) the response to the electromagnetic signals generated from the outside anomaly of the borehole is different: for the loops directly facing anomaly, the signal of reply value is positive; for the loops reversely facing anomaly, the induced voltage is negative. figure 2. schematic diagram of borehole tem signal response. when excitation source is ramp function turn-off source (mcneil et al., 1984; fitterman and anderson, 1987), the expression of the ramp function turn-off transient electromagnetic field response can be obtained according to the duhamel integral (zhong and zheng, 1986). induction segment: (1) attenuation segment: where t0 is the turn-off time; r (r) is the transient response excited by the ramp function turn-off source; r = t -s is a replacement function. under the excitation of ramp function turn-off source, the horizontal magnetic component of the single magnetic dipole in the whole space is (yang, 2009): (3) where is the vacuum permeability, , m is the magnetic moment of the single transmitting loop; a is the distance from the center of receiving loop to transmitting loop axis; 2l is the distance between the centers of the two transmitting loops; r0 is the absolute distance from the midpoint of the receiving loop to that of the transmitting loop, . it can be seen from figure 2 that the two transmitting circuits produce an electromagnetic field with the same value and opposite direction. lai et al., (2017) according to the symmetry of the electromagnetic field, the 79double-transmitting and sextuple-receiving borehole transient electromagnetic method and experimental study vertical components cancel each other, and the horizontal components are doubled, so the magnetic field response at the midpoint of the receiving loop of the transmitting-receiving device is: (4) take equation (4) into equations (1) and (2). induction segment: (5) attenuation segment: (6) the induced voltage curve of the receiving loop can be obtained by . the layered medium is a standard medium for studying the propagation of transient fields. the particular case is that there is a small plate in the layered strata with high conductivity, and the response equation of the magnetic field of receiving point at any position in the thin plate: where b = 2/ μs, is the height of the dipole source above the horizontal plate s, x is the horizontal projection distance between the field source and the receiving point, z is the vertical projection distance between the point of receipt and the field source, m is the magnetic moment of the double magnetic dipole source, m=2m. when a receiving loop is directly facing the thin plate, the receiving point receives only the signal of vertical component, and the loop of receipt is coaxial with the dual transmitting source , and x =0 , substitute in equation (7): obtain the time derivative: (10) as shown in figure 2, the transient response of the dual transmitting system shows that the primary response to the six receiving loops is the same, and the secondary response is different. the secondary field response of the loop directly confronting the thin plate is positive, while the secondary field response of the channel reversely facing the thin plate is negative. lai et al. (2017), considering the 2a space between the two faceto-face receiving loops in the device, the secondary field response values are different when the distance between the plates is close, and the value of reply of the reversely facing the plate is: (11) 3. negative transformation algorithm as mentioned above, the response signal value received by some receiving loops is negative when the probe detects a low-resistance anomaly, and the negative value usually cannot be expressed in the double logarithmic coordinate system. lai et al., (2017) it is necessary to use an algorithm to express negative values in the correct geometric form in the double logarithmic coordinate system. the transformed value is obtained through the following transformations of the original data: where n is a symmetric coefficient. taking the induced voltage data of the receiving loop rx4 as an example, the symmetry coefficient n=0. the horizontal axis is chosen as the axis of symmetry before and after transformation. as shown in figure 3, the curve with positive values did not change before, and after the rx4 data transformation, the absolute value of the negative value data is symmetrical to the transformed value along horizontal axis v=1, and the curve attenuation is normal. figure 3. comparison of rx4 measuring data before and after data transformation. 4. physical simulation of abnormal azimuth recognition the transmitting-receiving unit is placed in an open space, and the plate low-resistance anomaly situated on one side of the probe. roslee et al., (2017) the borehole tem instrument is independently created by our research group, and the instrument and the probe structure are shown in figures 4-5 respectively. the physical simulation experiment is shown in figure 6, and the first measuring point is the anomaly directly facing 1# receiving loop. for the following measuring points, the distance between the anomaly and the probe remains the same, and every time the probe is rotated by 5° when each measuring point is set for detection. 80 bo wang, shengdong liu, shining li and fubao zhou figure 5. schematic diagram of probe structure. figure 6. schematic diagram of physical simulation. from the probe structure in figure 5, it is known that the receiving loops are combined to form a regular hexagonal prism structure. since the structure is centrosymmetric and the minimum symmetry angle is 30°, the rotation angle selects: 0°, 5°, 10°, 15°, 30° and 60°, and the measuring data are shown in figure 7. the induced electromotive force of receiving loops rx3, rx4 and rx5 are negative values, and the data is calculated by equation (11) with symmetric coefficient n=0, and transformed and drawn in the double logarithmic coordinate system. as can be seen from the figure 7-a, the secondary field response value of the receiving loop rx1 directly facing anomaly is positive and maximum, and the raw values of the secondary field response of receiving loops rx3, rx4, and rx5 reversely facing the anomaly are negative. kamsani et al., (2017) according to the law of attenuation, the attenuation curves of each receiving loop eventually converge to zero. since the data is transformed, its value should converge to the symmetry axis v=10n, and the curves converge to one in the graph, and the curve configuration of negative values conform to abnormal response characteristics. it can be seen from figure 7 that the response value is the highest when the anomaly directly faces rx1 receiving loop, the response signals of rx2 and rx6 loops are the same and are positive; while the response values of rx3, rx4, and rx5 loops reversely face rx1 loop are negative, and it is shown as the characteristics of the six curves in figure 7-a. when the anomaly is within the normal angle between two adjacent receiving loops, the abnormal response value of the two loops is positive, and the response value of the loop is higher when the anomaly is closer to a certain loop; as shown in figure 7-b and figure 7-c, the anomaly is closer to rx1 loop. when the defect is located on the angular bisector of the two adjacent loops, the responses of the two loops are positive, and the values are the same, as shown in figure 7-e. by comparing figure 7-a and figure 7-f, when the anomaly directly faces a certain loop, the response of the loop is the highest, and the value of the loops on both sides are positive and approximately same, and the remaining three loops have a negative response. from the above phenomenon, the following four laws can be summarized: (1) when there is a lowresistance anomaly, there exist at least two measuring curves with the positive secondary field. (2) the normal direction of the several receiving loops with the positive secondary field of measuring curves must be the azimuth of the anomaly. (3) if there is only one highest curve among the measuring curves with the positive secondary field, then the anomaly is near the normal direction of the receiving loop. (4) if there are two higher curves among the measuring curves with positive secondary field and coincide, the anomaly is near angular bisector of the two receiving loops. therefore, the measuring curves obtained by the borehole transient electromagnetic system can identify the azimuth of the radial anomaly, and the resolution is 30°. it can be seen from figure 7-b, 7-c and 7-d that when the anomaly is at different angles within the range of 30°, the curves are also different, but cannot be directly distinguished from the significant difference among the curves. figure 4. borehole tem instrument. 81double-transmitting and sextuple-receiving borehole transient electromagnetic method and experimental study figure 7. response attenuation curves of the receiving loops at different rotation angles. 5. physical simulation of anomalies with different radial depth from the above theoretical derivation, it is known that the borehole tem has different response signals for different anomalies, and the physical simulation experiments were carried out for lowresistance anomaly with different radial depths of the same azimuth (roslee et al., 2017; lindang et al., 2017; kamsani et al., 2017; lai et al., 2017). as shown in figure 8, the probe moves along a measuring line, the direction of the probe remains the same, and the low-resistance anomalies of different radial depths are placed in the same direction directly facing rx1 receiving loop. there are seven measuring points in the measuring line, and the 1# measuring point is no anomaly detection, and the radial depths of 2#-7# measuring points are from h to 6 h. in the experiment, the anomaly directly faces rx1 receiving loop, so rx1 gets a positive response signal, rx4 gets a negative response signal. the experimental data of the two receiving loops were extracted to obtain the response curves shown in figures 9-10, and the symmetry coefficient n=-1 in negative value transformation algorithm. 82 bo wang, shengdong liu, shining li and fubao zhou figure 8. schematic diagram of physical simulation experiment of different radial depths. figure 9. contrast curve for each measuring point of rx1. figure 10. contrast curve for each measuring point of rx4. it can be seen from figure 9 that when the anomaly directly faces a receiving loop, the secondary field response signal of the anomaly decreases with the increase of the radial depth of the transmitting-receiving device and the low-resistance defect. the secondary field segments of curves are parallel to each other, which is in line with the correct attenuation law. at the same time, the transient time (zhong and zheng, 1986; jiang, 1988) of each curve increases with distance, which accords with the time law of abnormal response. it can be seen from figure 10 that when the anomaly faces the rx4 receiving loop reversely, the anomaly response is negative. the amplitude of the secondary field response signal decreases with the increase of radial depth between the transmitting-receiving device and the abnormality, and the subsequent field segments of curves are parallel to each other, which accords with the correct attenuation law. figures 9-10 show that there is a same law between the receiving loop directly facing the anomaly and reversely facing anomaly. the change of the anomaly depths can be determined by the induced voltage difference of fixed time window. through data analysis, the best resolution window ranges from 138μs to 975μs. the geometric distance between the two curves under the double logarithmic coordinate system can be solved by the following equation: (12) the equation shows the geometric induced voltage difference of the curve distance in the double logarithmic coordinate system, v1 is the transformed value in a fixed time window when receiving loop faces the anomaly directly, and v2 is the transformed value in a fixed time window when receiving loop reversely faces the anomaly. the response values of t=136.8μs, 231.2μs, 390.4μs, 659.2μs and 856μs in the response signals of rx1 and rx4 receiving loops are calculated in equation (8) to obtain the curve graph for the geometric difference of different fixed time windows shown in figure 11, where the length of a standard grid in the vertical axis represents the length of the 10 times of data difference in the double logarithmic coordinate system. it can be seen from figure 11 that the geometric distance between the positive and negative signals decreases as the anomaly depth increases, so the geometric difference between the response signals of different measuring points can be used to evaluate the radial depth of the anomaly qualitatively. two curves of rx1 and rx4 are coincident with zero geometric distance when there is no anomaly. in the fixed time window, the geometric difference is the lower when the time is larger. from the curve comparison, the curves of several different windows are parallel, and the radial depth does not affect the attenuation speed of its secondary field response signal when the same low-resistance anomaly is detected. figure 11. curve graph for the geometric difference of different measuring points. 5. conclusion borehole tem is a kind of fine geophysical method which effectively utilizes the drilling space. in this paper, through theoretical derivation and physical simulation experiment, the response of borehole tem double transmitting loops is analyzed and studied. through the negative transformation algorithm, the negative values obtained by double-transmitting and sextuplereceiving borehole tem can be plotted correctly in the double logarithmic coordinate system, which provides an effective method for the radial direction and depth experimental data analysis. from the analysis of the electromagnetic principle, it can be seen that the borehole tem has efficient resolution ability for the radial azimuth of low-resistance anomaly. by physical simulation experiment, the effective resolve of the double-transmitting and sextuplereceiving borehole tem is 30°. at the same time, the borehole tem can qualitatively analyze the radial depth between the low-resistance anomaly and borehole tem probe by the geometric difference, so the double-transmitting and sextuple-receiving borehole tem is an efficient way to identify the radial azimuth and depth of low-resistance anomaly. 83double-transmitting and sextuple-receiving borehole transient electromagnetic method and experimental study since each measuring point of the double-transmitting and sextuplereceiving borehole tem can obtain different pattern signals of six receiving loops, which has the advantages of less time consuming and a large amount of data. therefore, the next step in research should take advantage of the multi-parameter advantage of double-transmitting and sextuple-receiving borehole tem to achieve high-resolution quantitative interpretation. acknowledgements this research has been performed by the national natural science foundation project (grant no. 41604082, 51323004 and 41474122) and joint funding project of national natural science foundation and shenhua group corporation ltd (grant no.u1261202). a project funded by the priority academic program development of jiangsu higher education institutions. references fitterman d.v., anderson w.l., 1987. effect of transmitter turn-off time on transient soundings. geoexploration, 24(2), 131-146. gan, t., balmain, b., sibgatullin, a., 2016. formation evaluation logoff results comparing new generation mining-style logging tools to conventional oil and gas logging tools for application in coalbed methane (cbm) field development. journal of natural gas science and engineering, 34(8), 1237-1250. gorbatenko, a.a., sukhorukova, k.v., 2016. high-frequency induction logging in deviated and horizontal wells: geosteering and inversion. russian geology and geophysics, 57(7), 1111-1117. guo, b.l., song, x.j., 2012. research on receiving technology of transient electromagnetic method in production wells. journal of northwest university (natural science edition), 6, 890-896. jiang, b.y., 1988. application of transient electromagnetic method in near field magnetic source exploration. geological publishing house, beijing. kamsani s.r., ibrahim n., ishak n.a., 2017. psychological debriefing intervention: from the lens of disaster volunteers, malaysian journal of geoscience, 1(1), 32-33. karinskiy, a.d., daev, d.s., 2016. the effects of tool eccentricity and formation anisotropy on resistivity logs: forward modeling. russian geology and geophysics, 57(10), 1477-1484. khoshbakht, f., rasaie, m.r., shekarifard, a., 2016. investigating induction log response in the presence of natural fractures. journal of petroleum science and engineering, 145, 357-369. lai g.t., razib a.m.m., mazlan n.a., rafek a.g., serasa a.s., simon n., surip n., ern l.k., rusli t., mohamed., 2017. rock slope stability assessment of limestone hills in northern kinta valley, ipoh, perak, malaysia, geological behavior, 1(1), 16-18. lindang h.u., tarmudi z.h., jawan a., 2017. assessing water quality index in river basin: fuzzy inference system approach, malaysian journal of geoscience, 1(1), 27-31. liu, s.d., liu, j., yue, j.h., 2014. development status and key problems of chinese mining geophysical technology. coal science and technology, 39(1), 19-25. mcneil, j.d., 1984. approximate calculations of the transient electromagnetic response from buried conductors in a conductive half-space turn-off duration. geophysis, 49(7), 918-924. ren, z.p., dang, r.r., song, x.j., 2011. study on transient electromagnetic response characteristics of production wells. journal of oil and gas technology, 33(9), 100-104. roslee r., bidin k., musta b., tahir s., tongkul f., norhisham m.n., 2017. gis application for comprehensive spatial soil erosion analysis with musle model in sandakan town area, sabah, malaysia, geological behavior, 1(1), 01-05. school c., edwards r.n., 2007. marine downhole to searfloor dipoledipole electromagnetic methods and the resolution of resistive targets. geophysics, 72, 39-49. shen, j.g., meng, c., pi, y.g., 2016. transient electromagnetic logging theory transient induction logging. progress in geophysics, 31(2), 770-774. stolz e.m., 2003. direct detection of gold bearing structures at st ives, wa-dhem vs dhmmr. exploration geophysics, 34, 131-136. strack, k.m., 2014. future directions of electromagnetic methods for hydrocarbon applications. surveys in geophysics, 35(1), 157-177. wang, y.h., 2012. cased hole electromagnetic logging theory and parameter optimization. xi’an shiyou university, xi’an, china. wilkinson k., chamberlain t., grundy m., 2005. the role of geophysics in understanding salinisation in sw queensland. exploration geophysics, 36, 78-85. yang, h.y., 2009. study on numerical simulation and distribution regularity of transient electromangnetic field with mine-uesd multi small loop. china university of mining and technology, xuzhou, china. zang, d.f., zhu, l.f., zhang, f.m., shen, j.g., shen, y.j., wang, z.l., 2014. theory study for transient electromagnetic logging: electromagnetic wave. well logging technology, 5, 530-534. zhang, h., 2012. design of downhole transient electromagnetic large power source. yangtze university, jingzhou, china. zhong. s.m., zheng, d.z., 1986. transient process analysis. tsinghua university press, beijing. the shock response and dynamic fracture of concrete gravity dams under impact load are the key problems to evaluate the antiknock safety of the dam. this study aims at understanding the effects of impact shock on the elastic response and dynamic fracture of concrete gravity dams. firstly, this paper uses acceleration records of a concrete gravity dam under impact to establish the correct way to determine the concrete gravity dam of the fundamental frequency and present cut sheets multi-degree-of-freedom dynamic modeling. under strong impact loading, the constitutive relation of concrete gravity dam and the highest frequency of the impact are uncertain. so, the main advantage of this method is avoiding the use of elastic modulus in the calculation. the result indicates that the calculation method is a reliable computational method for concrete gravity dams subjected to impact. subsequently, the failure process of dam models was numerically simulated based on abaqus commercial codes. finally, this paper puts forward suggestions for future research based on the results of the analysis. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): m1 -m6 abstract keywords: reservoir, gravity dam, underwater explosion, failure mode, dynamic analysis. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.54133 h id r o l o g y lu lu1,2,3*, xin li2, jing zhou2, genda chen4, dong yun1 1 faculty of management engineering, huaiyin institute of technology, huai’an 223001, china 2the state key laboratory of coastal and offshore engineering, dalian university of technology, dalian, 110623, china 3the state key laboratory of structural analysis for industrial equipment, dalian university of technology, dalian 116023 china 4 the center for infrastructure engineering studies, missouri university of science and technology, rolla, missouri 65401, usa *corresponding author. e-mail: llzhxy@aliyun.com numerical simulation of shock response and dynamic fracture of a concrete dam subjected to impact load record manuscript received: 12/11/2015 accepted for publication: 05/02/2016 how to cite item lu, l., li, x., zhou, j., & chen, g. (2016). numerical simulation of shock response and dynamic fracture of a concrete dam subjected to impact load. earth sciences research journal, 20(1). m1-m6. doi: http://dx.doi.org/10.15446/esrj.v20n1.54133 introduction recently, the novosibirsk hydroelectric plant after a report about a planted explosive, russia has declared a state of emergency. so, this event should cause the attention of researchers and administrative departments. hydraulic dams are critical infrastructure in geotechnical engineering. they are often designed and built to store water for drinking and irrigation in adjacent areas, to add water recreation spaces, to create a water way for the short-distance transport of people and goods across deep canyons in mountainous regions, and to regulate the river during a flood event. a water-filled dam can boost the local economy through various personal and business activities, leading to the establishment of a new community center such as a village or a town, and representing a major capital and long-term investment. on the other hand, the breaching and an accidental damage of a dam can lead to a catastrophic flood event and its chain effects such as engulfing downstream residential areas and washing away agriculture lands. therefore, design and maintenance of dams are not only a serviceability issue but also a life-threatening matter to millions of people. when extreme events such as earthquakes, tsunamis, hurricanes, and tornadoes took place, concrete dams can be subject to extensive shaking and wave impact. the march-11, 2011, japan earthquake event testified the destructive power of the earthquake-induced tsunami. many scholars have studied the high dam subjected to earthquake action. among these are zhou et al. (2000), mir et al. (1995), kong et al. (2012). equally, if not more important, dams are also vulnerable targets for man-made explosion events, particularly with the advent of advanced long-range and precision missile technologies. since the september 11 attacks by terrorists, there has been increasing public concern about the threat of bomb attacks on dam structures (federal emergency management agency, 2003). therefore, protection of dam structures against impact loads is a critical component of homeland security (lu et al., 2013). indeed, as respectively studied by lu et al. (2012, 2014a, 2014b), and zhang et al. (2014), the risk of a tall concrete dam being subjected to underwater explosion shock wave cannot be neglected. m2 lu lu, xin li, jing zhou, genda chen and dong yun 0.075 m 0.1 m 0.56 m 0.075 m 0.1 m 0.56 m 0.75m hammers initial position water bag pressure sensor vertical rope 1#hammer 2#hammer 3#hammer 4#hammer 5#hammer model dam moving hammers fixed point base anchorage hammers initial position water bag pressure sensor vertical rope 1#hammer 2#hammer 3#hammer 4#hammer 5#hammer model dam moving hammers fixed point base anchorage currently, with the development of computational techniques and numerical simulation methods, as well as commercialization of nonlinear dynamic software (e.g. abaqus, ly-dyna), major developments in understanding the structural responses and failure modes of concrete structures under blast load have taken place. many researchers have conducted comprehensive experimental and numerical investigations related to the effects of explosions on building structures(tian et al., 2008; jayasooriya et al., 2011), marine structures (jin, et al., 2011; zhang et al., 2011), underground structures (ma et al., 2011; li et al., (2013), and bridge structures (hao et al, 2010; son et al., 2011). in the modeling of transient loading, it is very critical to describe the propagation velocity of the stress waves correctly. in fact, the value of this velocity depends on the material elastic modulus that is given by the material constitutive relation. from the material point of view, concrete shows an increase in elastic modulus with the strain rate increases, a phenomenon called strain rate effect (georgin, 2003). the relationship between concrete strength and strain rate was extensively investigated by bischoff and perry (1991), georgin and reynouard (2003), grassl(2006) and tai, y.s. (2009). because many problems of the concrete material constitutive parameters and constitutive model have not been a clear understanding, the numerical results obtained by different calculation models are very different. moreover, the modeling of the strain rate effect on concrete is not very easy to tackle. these problems, to a great extent, comprise the uncertainty existing in the macroscopic numerical simulation. one of the efficient methods to study the failure modes and mechanisms of structures is to carry out a large number of model experiments and obtain data from them. considering that the experimental study has its limitations, as well as great difficulties and expensive costs for the underwater explosion test, only a small amount of data can be obtained. for example, lu et al. (2014b) obtained only a small quantity of damage to the relationship between the state and the maximum pressure of the shock wave through model tests. overall, to the best of our knowledge, experimental investigations of concrete gravity dams under underwater shock wave effects have not yet been conducted to date. concrete dams are thoroughly studied in this paper both experimentally and numerically to understand their behavior and failure modes. specifically, this paper is to use acceleration records of the concrete structure under strong impact to establish the correct way to determine the concrete structure of the fundamental frequency and present cut sheets multi-degree-of-freedom dynamic modeling. the dams were numerically modeled to understand further their shorttime failure process in the order of msec based on abaqus. examples of model tests the same test setup and test results as presented by lu (2012), lu (2014a), lu (2014b)were used in this paper which will be briefly depicted as follows. the dimension of the model dam and test layout are shown in figure 1. figure 1. dimension of the model dam and test layout (lu, 2012; lu, 2014a; lu, 2014b). -0.01 0 0.01 0.02 0.03 0.04 0.05 0.2 0.3 0.4 0.5 0.6 rt pr es su re (m pa ) time(sec)-0.01 0 0.01 0.02 0.03 0.04 0.05 0.2 0.3 0.4 0.5 0.6 rt pr es su re (m pa ) time(sec) the mechanical properties of the individually tested samples and their average values are given in table 1. therefore, the actual material density, young’s modulus, compressive strength, and tensile strength of the small-scale model are ρm = 2,900 kg/m3, em = 355 mpa, fcm = 205 kpa, and ftm = 13.2 kpa, respectively. table 1. mechanical properties of concrete material for scale model dams (lu, 2012; lu, 2014a; lu, 2014b). -100 -50 0 50 100 150 200 0 0.2 0.4 0.6 0.8 1 st ra in () time(sec) 610 � -100 -50 0 50 100 150 200 0 0.2 0.4 0.6 0.8 1 st ra in () time(sec) 610 � the main test results of model test are shown in table 2. the pressure, strain recorded and acceleration recorded are shown in figures 2, 3 and 4. table 2. main results of the model tests (lu, 2012; lu, 2014a; lu, 2014b). -12 -7 -2 3 8 13 18 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) time(sec) 22.521 =ow1#acceleration sensor -12 -7 -2 3 8 13 18 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) time(sec) -12 -7 -2 3 8 13 18 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) time(sec) 22.521 =ow1#acceleration sensor figure 2. time history of the impact pressure on the dam surface (lu, 2012; lu, 2014a; lu, 2014b). figure 3. recorded strain (lu, 2012; lu, 2014a; lu, 2014b). m3numerical simulation of shock response and dynamic fracture of a concrete dam subjected to impact load -6 -4 -2 0 2 4 6 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 a cc el er at io n( g) time(sec) 2#acceleration sensor -6 -4 -2 0 2 4 6 0.2 0.25 0.3 0.35 0.4 0.45 0.5 0.55 0.6 a cc el er at io n( g) time(sec) 2#acceleration sensor -3 -2 -1 0 1 2 3 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) 3#acceleration sensor time(sec) -3 -2 -1 0 1 2 3 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) 3#acceleration sensor time(sec) -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) 4#acceleration sensor time(sec) -2 -1.5 -1 -0.5 0 0.5 1 1.5 2 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) 4#acceleration sensor time(sec) -1.5 -1 -0.5 0 0.5 1 1.5 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) 5#acceleration sensor time(sec) -1.5 -1 -0.5 0 0.5 1 1.5 0.2 0.3 0.4 0.5 0.6 a cc el er at io n( g) 5#acceleration sensor time(sec) figure 4. recorded acceleration time histories from the model test (lu, 2012; lu, 2014a; lu, 2014b). as shown in the measured pressure-time curve, the impact load can be simplified to a triangular distribution. the failure mode and damage area are shown in figure 5. figure 5. failure mode and damage area[6-8] note that the accelerometers and pressure sensor were not synchronized during the test. as a result, the peak accelerations seem to occur before the application of the peak pressure. the maximum time delay among the five accelerometers is 23.44 msec. the time delay was likely caused by different periods when five hammers applied impact forces. based on the arrival time of peak accelerations, hammers 4 and 5 were in contact with the model dam later than hammers 1-3. 3. cut sheets dynamic modeling for that, the concrete dam is divided into i unites, the governing equation for the system with i freedom is as follows: [m] {a}+[k]{x}={p} (1) where [m],[k] denote the mass matrix and stiffness matrix; {x}, {a} and {p} are the displacement, acceleration and force vectors, respectively. displacement and acceleration using the vibration mode expand can be got (2) (3) where is {φi} i-th mode vector of the system, ξi is general coordinate corresponding to {φi}. equation 1 after substituting equations 2 and 3 becomes (4) this equation left multiplied by {φj}t becomes (5) use of the orthogonality of the vibration mode on the and , that is (6) by the equations 5 and 6 can be got (7) the governing equation for ξi is (8) (9) (10) (11) (12) let m4 lu lu, xin li, jing zhou, genda chen and dong yun rt p t maxp rt2 p t maxp rt2 p t maxp rt (a) (b) (c) rt p t maxp rt2 p t maxp rt2 p t maxp rt (a) (b) (c) (13) (14) (15) where superscript t denotes transformation of matrix or vector. pi* is the generalized force the corresponding to i-th vibration mode. mi* is the generalized masscorresponding to i-th vibration mode. mii is known quantity, so mi* can be determined. to give expression of , must be assumed for pi*(t) .therefore, three forms of the impact force-time curve were examined, as shown in figure6 (a-c), with the same pmax and impulse pmtr. figure 6. generalized impact force-time curve solutions of for figure 6, a, b, and c, are represented by equations13, 14 and 15, respectively. because |ẍ|t=0 = |ẍ|max for equations13 and 14 that are inconsistent with experimental results, figure 6 (a) and (b) are impossible. to determine the value of must also be determined ω0i and θc from the measured acceleration record. the time of acceleration extremal point(including the maximum and minimum values) tm1, tm2, tm3... tmi and the time of acceleration zero point t01, t02, t03... t0i can be obtained by the measured acceleration records. if tr infinite,by the first type of the equation 15, | |max occurs at the following times: , respectively. by the second type of the equation 15, the following equations can be obtained (16) (17) (18) the specific calculation process is as follows: let tm1 <tr , then the equation 17 substituting ω01 of the solution to equation 18 and t01 to determine tr. if tr > tm1, then the ω01 and the corresponding tr is solution. if tm1 <tr, then by tm1 and t01 from the measured acceleration records, and equations 16 and 17 , the simultaneous solution to get ω01 and tr . the maximum acceleration (shown in table 2) is gradually decreased from top to bottom, which reflects the first mode characteristics of the dam. experiment obtained vibration mode and natural frequency, so do not need too much freedom in the calculation. for that, the model dam is divided into five unites(as shown in fig.7.). y x (x, y) (0, 0) (0, 0.75) (0.56, 0) 1#acceleration sensor(0.06, 0.75) unit: m 2#a(0.126, 0.681) 3#a(0.252, 0.413) 4#a(0.378, 0.244) 5#a(0.500, 0.080) 11m 22m 33m 44m 55m (0.075, 0.65) y x (x, y) (0, 0) (0, 0.75) (0.56, 0) 1#acceleration sensor(0.06, 0.75) unit: m 2#a(0.126, 0.681) 3#a(0.252, 0.413) 4#a(0.378, 0.244) 5#a(0.500, 0.080) 11m 22m 33m 44m 55m (0.075, 0.65) figure 7. discrete model dam and the locations of acceleration sensors mii (shown in fig.7.) in kg, it is reasonable to assume that the measured acceleration distribution is proportional to the first mode, so for the model test, {φ1} t=(1,0.323,0.161,0.111,0.074).from above m*1=5.1928kg, (p*1)max = 1510.016n. for given pressure-time curve shown in figure 6(c) the equation 15 becomes (19) (20) for given case the maximum acceleration of 1# appears in t <tr (where tr is rise time of the pressure pulse, as shown in fig.6.), so is obtained 1 from and the value of ω01 equal to 52.22 by equation 18, we can get calculation results and experimental results are in good agreement. it indicates that the cut sheets dynamic calculation model is a reliable computational model for the concrete dam subjected to impact load. the main advantage of this method is avoiding the use of elastic modulus in the calculation. under impact loading, the constitutive relation of concrete structures and the highest m5numerical simulation of shock response and dynamic fracture of a concrete dam subjected to impact load frequency of the impact are uncertain. the computational work is much smaller than step by step integration. failure process by simulation during impact tests, it is difficult, if not impossible, to observe the failure process of concrete dams over the duration of pressure pulse (~60 msec) except at the end of the tests. on the other hand, the cut sheets dynamic model can calculate the elastic response of the dam under impact, but that cannot simulate the dynamic fracture process, and must assume that the functional form of load. therefore, numerical simulations with a finite element model (fem) were supplemented for further understanding of the concrete dam behavior over time. a fem of the model dam was established with 4-node linear tetrahedron three-dimensional (3d) elements in abaqus commercial codes. due to symmetry in both boundary conditions and loads, only half of the dam was numerically modeled as shown in figure 8. the model dam was fixed at its base. the locations of five accelerometers are also shown in figure 8. the mechanical properties of concrete used for the model are included in table 1. besides, the poisson ratio was taken to be 0.18 in various simulations. figure 8. 3d meshes and locations of five accelerometers of half of the model for convenience, the pressure-time curve in fig. 2 was cleaned up and only the main pulse was included in simulation as shown in red line in figure9. after numerical treading, the peak pressure slightly increases from 48.6 kpa to 49.7 kpa due to the initial negative value of the recorded pressure time history. the horizontal stress at the location of strain gauge was calculated from the fem under the registeredpressure. figure8 compares the calculated stress (black line) with the recorded pressure (red line) and the so-called “predicted stress” (blue line) by multiplying the recorded strain by the modulus of elasticity. it is clearly seen that the three-time histories are in good agreement because the horizontal stress expects to be in equilibrium with the applied pressure at the upstream of the model dam. therefore, the fem is partially validated. figure 9. recorded pressure, calculated stress and measured stress from the measured strain figure 10 shows the distribution of the vertical bending stress at different time instances and the progression of structural failure. since the maximum bending stress at 13.3 msec. exceeds the tensile strength of concrete, 18 kpa, concrete was considered to start cracking at 13.5 msec. in numerical simulations. crack initiated at 0.57 m above the base, which is close to the fracture location observed at 0.5 m high during the testing of the model in table 3. at 20.0 msec, the fracture was extended to the downstream face and, by the end of pressure loading at 50.0 msec, another fracture section was fully developed across the middle of the model dam as well as at the bottom of the dam. the fracture damage at the base is similar to the pattern observed from the photo of the model in table3. the fracture in the middle of model dam seems close to the damage scenario observed during the testing of model underlarger impact loading. figure 10. distribution of vertical stress (bending effect) at various time instances in practical applications, there is no anchorage of dam foundation. it is 150 m tall (h) and 112.5 m wide (l) with a vertical upstream slope and a downstream slope of 1:0.75. let p1 is the tensile strength of concrete material. there p1 = 0.3 mpa p2 isconcrete gravity against sliding stress. mpa when g is the dam gravity. f=1 is the friction coefficient. p3 is concrete gravity against overturning. (a) before damage (b) crack initiated at 13.5 msec. (d) model collapse at 50.0 msec. (c) fracture through the top section at 20.0 msec. m6 lu lu, xin li, jing zhou, genda chen and dong yun to simplify the calculation of the dam as a triangle, centroid dam in 1 / 3h and 2/3 l department, the resultant moment m (t) can be obtained (21) (22) where f is the uniform external force on upstream. hydrostatic pressure pw is expressed as: where ρw is the density of water. if m (t) ≥ 0, when the dam starts to rotate, so the concrete dams may be first toppled off, then fracture at their top of the damand, finally, fracture at their middle height. the performance of concrete under impact load and static load strength and deformation have a certain difference, with the increase of loading rate, the mechanical properties of concrete exhibit different behavior, the so-called rate dependent. at present, the research work is not enough, the research data is relatively small, and the conclusions of the study are not unified. this kind of research needs mesoscopic and macroscopic numerical analysis methods(multis cale model). conclusions the following conclusions can be drawn: under strong impact loading, the constitutive relation of concrete structures and the highest frequency of the impact are uncertain. this paper uses acceleration records of concrete structure under strong impact to establish the correct way to determine the concrete structure of the fundamental frequency and present cut sheets multi-degree-of-freedom dynamic modeling. the main advantage of this method is avoiding the use of elastic modulus in the calculation. calculation results and experimental results are in good agreement. the computational work is much smaller than step by step integration of the finite element. however, the cut sheets dynamic model can not simulate the dynamic fracture process, and must assume that the functional form of load. therefore, numerical simulations with a finite element model (fem) were supplemented for further understanding of the concrete dam behavior over time. the results of the analysis indicate that concrete damage mainly occurred at the dam head and bottom using the material property from table 1.concrete dams may be first toppled off, then fracture at their top of the damand, finally, fracture at their middle height. the performance of concrete under impact load and static load strength and deformation have a certain difference, with the increase of loading rate, the mechanical properties of concrete exhibit different behavior, the so-called rate dependent. at present, the research work is not enough, the research data is relatively small, and the conclusions of the study are not unified. acknowledgements this work was supported by the [the natural science foundation of china] under grant [no. 51379028,71301060], [the research funds of the state key laboratory of structural analysis for industrial equipment of dalian university of technology] under grant [no. gz1409], and the [huai’an applied research and scientific and technological research funds] under grant [no. has2014021-4]. references: bischoff, p.h. and perry, s.h. (1991). compressive behaviour of concrete at high strain rates. materials and structures, v. 24(6), 425–450. federal emergency management agency. (2003). dam safety and security in the united states: a progress report on the national dam safety program in fiscal years 2002 and 2003. u.s. department of homeland security. grassl, p. and jirásek, m. (2006). damage-plastic model for concrete failure. international journal of solids and structures, 43(22): 7166-7196. georgin, j.f. and reynouard, j.m. (2003). modeling of structures subjected to impact: concrete behaviour under high strain rate. cement & concrete composites, 25, 131-143. hao, h, and tang, e.k.c. (2010). numerical simulation of a cable-stayed bridge response to blast loads. part ii: damage prediction and frp strengthening. engineering structures, v. 32, 3193–3205. jayasooriya, r., thambiratnam, d.p., perera, n.j. and kosse, v. (2011). blast and residual capacity analysis of reinforced concrete framed buildings. engineering structures, 33(12), 3483–3495. jin, q. and ding, g. (2011). a finite element analysis of ship sections subjected to underwater explosion. international journal of impact engineering, 38(7), 558–66. kong, x.j., liu, f.h. and liu, j. (2012). shaking table model tests on faceslab dislocation of concrete faced rock-fill dams under earthquakes. chinese journal of geotechnical engineering. 32(2):258-266. li, j.c., li, h.b., ma, g.w. and zhou, y.x. (2013). assessment of underground tunnel stability to adjacent tunnel explosion. tunneling and underground space technology, v. 35, 227–34. lu, l. (2012). study on the damages and protection decision-making of the concrete gravity dam by the action of underwater shock wave. dalian university of technology. lu, l., li, x. and zhou, j. (2013). protection scheme for concrete gravity dam acting by strong underwater shock wave. advanced science letters, 19(1), 238-243. lu, l., li, x. and zhou j. (2014a). risk assessment method and protection goals of high concrete gravity dam subjected to far-field underwater nuclear explosion. advanced materials research, vol. 871, 21-26. lu, l., li, x. and zhou, j. (2014b). study on damage of high concrete dam subjected to underwater shock wave. earthquake engineering and engineering vibration, 13(2), 337-346. lu, y. (2009). modeling of concrete structures subjected to shock and blast loading: an overview and some recent studies. structural engineering & mechanics, 32(2), 235-249. ma, g., zhou, h. and chong, k. (2011). in-structure shock assessment of underground structures with consideration of rigid body motion. journal of engineering mechanics, 137(12), 797–806. mir, r.a. and taylor, c.a. (1995). an experimental investigation into earthquakeinduced failure of medium to low height concrete gravity dams. earthquake engineering and structural dynamics, 24(3), 373-393. son, j. and lee, h.j. (2011). performance of cable-stayed bridge pylons subjected to blast loading. engineering structures, v. 33, 1133-1148. tai, y.s. (2009). flat ended projectile penetrating ultra-high strength concrete plate target. theoretical and applied fracture mechanics, v. 51 (2), 117-128. tian, l, and li z.x. (2008). dynamic response analysis of a building structure subjected to ground shock from a tunnel explosion. international journal of impact engineering, 35(10), 1164–78. zhang, a., zhou, w., wang, s. and feng l. (2011). dynamic response of the non-contact underwater explosions on naval equipment. marine structures, 24(4), 396–411. zhang, s., kong, y. and wang, g. (2014). a comparative study on the dynamic response of concrete gravity dams subjected to underwater and air explosions. journal of performance of constructed facilities, vol. 33(17), 47-54 zhang, s., wang, g., wang, c., pang, b. and du, c. (2014). numerical simulation of failure modes of concrete gravity dams subjected to underwater explosion. engineering failure analysis, 36, 49-64. zhou, j., lin, g., and zhu t. (2000). experimental investigation into seismic failure of high arch dams.journal of structural engineering. 126(8), 926-935. geotisica colombiana pp.71-73 bogota, d.c.n° 7 dieiembre de 2003 issn 0121 2974 true amplitude migration in constant gradient velocity media luis antonio castillo lopez protesor asistente universidad nacional de colombia, curso posgrado de geofisica tacultad de ciencias. departamento de geociencias, grupo de investigaci6n en geofisica e-mail: lacastillol@unal.edu.co resumen ei algoritmo de kirchhoff con trazado de rayo se basa en una funcion de velocidad con stante. esta solucion incorpora informacion de trayectoria de rayos dentro del algoritmo, el cual puede extenderse a un medio de velocidad variable. el modelo de veloci dad incluye la localizacion de reflectores e interval os de velocidades entre dos reflectores adyacentes, generados a traves del analisis de velocidad, pero importante en el procesamiento sismico tradicional. este proceso puede alcanzarse al aplicar un operador de funcion peso de apilado por difraccion, cuyo desarrollo teorico se bas a en la integral de migraci6n tipo kirchhoff. al escoger el propio peso para el peso de apilado de datos, el resultado del proceso de migraci6n es una seccion sismica donde se recupera la amplitud y su valor es proporcional al coeficiente de reflexion, liamado rnigracion con amplitudes verdaderas (recuperacion de amplitudes verdaderas). palabras clave: recuperacion de amplitudes verdaderes, psdm, medid con gradiente de velocidad. abstract the kirchhoff conventional algorithm using ray tracing is based on a constant velocity function. this solution, where is incorporated ray path information into conventional algorithm, can be extended to a variable velocity medium. velocity models include the locations of reflectors and interval of velocities between any two adjacents reflectors that are generally generated through the velocity analyses, process important in traditional seismic processing. this process could be achieved by applying a weighted diffraction stack operator, which theoretical development is based on a kirchhoff type migration integral. by choosing the proper weight for stacking the data, the result of the migration process is a seismic section where the amplitude are recovered and its value are proportional to the reflection coefficient, the so-called true-amplitude migration (true ampiltude recovering). migration process was applied into different models. the objetive is to reproduce an accurate picture of the subsurface model from the data registered in areas where the velocity changes. the migration process was obtained after trying several testes made with different weigths. key words: true amplitude recovery, psdm. gradient velocity media introduction by incorporating ray path information into conventional migration algoritm, the solution can be extended to variable velocity medium. the rays are traced from each depth point in the survey of each source-geophone location in order to determine the difraction traveltime surface difraction (huygens surface). then, the weighted diffraction stack operator could be used to migrate the seismic data. in this paper we present a fast algorithm that is able to recover true amplitude in depth migration process, by considering manuscrrto recibido para evaluacion el15 de junio de 2003. articulo aceptado para publicae ion por el comne editorial el 36 de octubre de 2003. gfofislca cdlombiana. 7, diciembre de 2003 a more particular situation when the seismic velocity model is represented by a linear distribution variable with the depth. this kind of model is important for simulating many situations in the seismic exploration. for instance, variations of the velocity with depth, inversion velocity (negative gradient) and increasing of velocity in a short layer (positive gradient). modelling asymptotic ray theory provide useful extrapolating techniques for modelling, migration and inversion, using ray theory of coordinates tracing equations which determine the second derivatives of the traveltime curve. this is desired in order to determine geometrical spreading, and then true amplitude recovery. in order to know 71 luis antonio castillo l6pez ray irajelory with positive gradient velocity 0.5 e ~1.5 ! 2.51------3l-'::-----'--,j-'---' o 0.5 1.5 2.5 offset (km) 0.5 e c -e 1.5 ! 2.51------........... 3l-----''--' ---'--'----l o 0.5 1.5 2 offset (kml figure i. models for differents gradients in a configuration common shot. gradient positive (left) and gradient negative (rigth), the stability of the diffraction stack migration algorithm, we have applied it to a set of seismic data in a common-shot configuration. diffraction stack operator after data input, two-dimensional diffraction stack operator can be written following the expression given by urban (1999), based on martins et al. (1998) 72 seismogram distance o 0 5 (km) i 1.5 2 i -i------=+=~~~_rt'-i-r-,....,...,~.,....,..-,--l----___+ e 0.5 -i------+-++-i-++-;.+:~~f.l=m"!-'l-+-----____+ f.== o-l------j-lllllllllj...ljlulu--'--'--'--,------t ray trajectories in constant gradient velocity media with commom shot configuration o r-~-----''''-'-"t't'rr-rt'crroct>--rt..--.rrr~~~---' vo=2kml 0.2 i ;s 0.4 c. ojo 0.6 0.8l--~-~~~~-~-~-~-~-~----j o 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 offset (km) figure 2. ray trajectory model(lower) and synthetic seismogram(upper). v(m ,t)= rbfd~w(~,m)~ u (~,t+td) v21t a (1) with the source and receiver pairs (s,g) described by the vector parameter and the traveltime curve defined by all the points of on the earth surface, and a point m in depth. ar~u (~, t +,d ) is the anti-causal half-time derivative (operator that corresponds in the frequency domain to the filter.j -iw) of the seismic trace u (~,t + 'd) registered on the geophone position g( ~), and it represents the principal component of the seismic primary reflected wavefield. the result of the integral (i) is put at the point m into the model. providing that we call depth difraction stack migration, with all lines summed along the stacking line and multiplied by the true amplitude (weight function) w(~,t-'d) and summed, then the result of migration process is proportional to the reflection coefficient recovered. 2.5 diffraction traveltime stack curve being known the macrovelocity model with a distribution of velocity varying linearly with depth the diffraction traveltime stack curve can be built, for the diffraction stack processing v(z)=vo + gz (2) where vo is the velocity near the earth surface, g is the gradient of the velocity function in the direction of the vertical axis z, in depth. this kind of velocity function is the first approximation to be considered when investigating the regional variations of velocity for most sedimentary rocks (lapsen, 1993). geofisiga golombiana, 7, oigiembre de 2003 true amplitude migration in constant gradient velocity media 1.9,----,------,-----,,---,---r----,--,.--...., 1.85r········· ,........... t···········, l ;..............,............."... ~ ~ 1.8 r············ .j ,........... , , ,...........w, ,..,?<. , ~ ~ 1.75 r············,··········· "',....... ; ; ; ····:<'a::···············l ~ ;:: 1.7 r···········+···············f."-:~~==~4:···w! .]. ~ 1.65 r···········t··········· t··········· ; ,........... .i. ; ;.... ~ 0.5 1.5 2.5 distance (km) 3.5 4.5 figure 3. traveltime difraction and traveltime reflection curve for a depth point. inside such model with linear variation of velocity, the ray trajectory is circular and the traveltime is given as solution of the integral (bleistein, 1986) i tan ( a~) i t=-in -in(b,bg) g tan(a~) g (3) being b, = 1 + (g2p + 2g(ji) / 2cco. all the values of this result are known; ii and e, are the circular trajectory and parameter ray, respectively. computational examples the seismic model consists of by a curved reflector bellow an inhomogeneous domo with constant gradient velocity medium, where the near surface velocity is 2.0 kmls (varying with the depth), and its gradient is 1.25/s. the shot position is 0.4 km on the left, with 0.5 km and 1.45 km for the near and far offset, respectively. the source dominant frequency is about 70 hz, and the sample interval is of 2.0 ms. in the figure 2 are represented the tracing rays(lower) and the synthetic seismogram(upper) for a asymmetric common shot configuration. the seismic data is noise free. in the figures 3 and 5 we geofislca colomb lana. 7, diciembre de 2003 have the image and depth migrated seismic data, after appliying of the diffraction stack operator to the respective set of input seismic data. it is important to note that the proposed algorithm provides a good image of the target reflector, and it is able to be used for migrating seismic data in constant gradient velocity media. conclusions in this paper, was develop and tested the diffraction stack migration algorithm, applying it to a set of common-shot seismic data, syntheticaly generated by the ray theory for a constant gradient velocity medium. the algorithm was tested providing us a good image of the target reflector. this result is very important in sense that the proposed algorithm is fast, stable and suitable to be used for true amplitude depth migration. references bleistein, n., 1986, two-and -one-half dimensional in-plane wave propagation. geophysical prospecting, 34, 686-703. japsen, p., 1993, influence of lithology and neogene uplift on seismic velocities in denmark: implications for depth conversion of maps. the aapgb, 77, n. 2, 194-211. schleicher, 1.; tygel, m.; and hubral, p., 1993, 3-d true-amplitude finite-offset migration. geophysics, 58(8): 1820-1830. schneider, jr. w. a.; ranzinger, k. a.; balch, h.; and kuse, c, 1992, a dynamic programming approach to first arrival traveltime computation in media with arbitrarily distributed velocities. geophysics, 57, n. 1,39-50. urban, 1.,1999, two-dimensional true-amplitude migration and introduction to 2.5-d case. master thesis. federal university of para, brazil. (in portuguese). acknowledgments we would like to thank the seismic group of the geophysical institute, charles university, prague, czechoslovakia, for making available the ray tracing software seis88, and the universidad nac10nal de colombia for supporting the author. 73 to analyze the effects of the leakage recharge of the aquifer on the initial dewatering of coalbed methane wells, the mathematical seepage model of water in the coalbed considering the aquifer leakage was established by using the leakage coefficient according to the unsteady seepage theory. the model was solved after laplace transform and the stehfest numerical reverse inversion was used to obtain the solution in right space. then, the log-log type curves of pressure and pressure derivative were created with new combinations of parameters. based on the natural seepage mechanism, the influence of aquifer leakage on curve shape was judged. it is found that the radial flow ends earlier as the leakage coefficient increases. moreover, it was proposed to obtain reservoir permeability, skin factor, and leakage coefficient by using type curve matching. the type curves are useful for quantitatively evaluating the level of leakage, thereby guiding the adjustment of the following production system for cbm wells. este estudio estableció el modelo matemático de filtración de agua en una capa carbonífera al estimar la salida acuífera con el uso del coeficiente de fuga, de acuerdo con la teoría de filtración inestable, para analizar los efectos en la recarga de pérdida de fluidos de un acuífero en el drenado inicial para pozos de gas metano. el modelo se resolvió tras usar la transformación laplace y la inversión numérica stehfest para encontrar la respuesta en el lugar indicado. luego, se creó la representación algorítmica de la presión y la presión derivativa con nuevas combinaciones de parámetros. se evaluó la influencia de la pérdida de fluido del acuífero en la forma de la curva con base al mecanismo físico de filtración. se estableció que el flujo radial finaliza antes de que el coeficiente de pérdida de fluido se incremente. además, se propone el uso de la curva tipo correspondiente para obtener la permeabilidad del reservorio, el factor de daño y el coeficiente de pérdida de fluido. las curvas tipo son útiles para evaluar cuantitativamente el nivel de la pérdida de fluido, y de esta manera guiar el ajuste de un sistema de producción consecuente para pozos de gas metano de carbón. earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017): 1721. abstract resumen keywords: coalbed methane (cbm); formation water; leakage; well test analysis palabras clave: gas metano de carbón (cbm); formación de agua; pérdida de fluido; análisis de pozo. solution and type curves for the seepage model of the water-bearing coalbed with leakage recharge curvas de solución y tipo para el modelo de filtración de capas carboníferas acuíferas con recarga de fugas record manuscript received: 01/03/2017 accepted for publication: 31/03/2017 how to cite item cheng, w., yang, y., sun, h., zhang, c., wen, q., & li, y. (2017). solution and type curves for the seepage model of the waterbearing coalbed with leakage recharge. earth sciences research journal, 21(1), 17-21. http://dx.doi.org/10.15446/esrj.v21n1.63003 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n1.63003 c b m wangang chen 1, yu yang2*, hansen sun 1, chengwei zhang 2, qin wen 2, yanyan li 2 1research center, china united coalbed methane co. ltd., beijing 100011, china 2college of energy, chengdu university of technology, chengdu 610059, china *email of corresponding author: zjstxz@163.com 18 wangang chen, yu yang, hansen sun, chengwei zhang, qin wen and yanyan li 1. introduction the coalbed methane (cbm) exists, in the form of adsorbed gas, in the coal reservoir. to recover the cbm, the coalbed is firstly dewatered and depressurized. only when the coal seam pressure drops to the critical desorption pressure, can the methane be desorbed and diffuse from matrix pores to fracture network. in the wyodak anderson coalbed (onsager and cox, 2000; cox and onsager, 2002) of fort union formation in the powder river basin, u.s., the 15# coalbed in shouyang-yangquan of the qinshui basin, china, and other coal measure strata (meng 2010), water bodies exist in the upper or lower coalbed, in addition to the coalbed cleat system. the silty or muddy aquitard with a certain thickness of the coalbed and the aquifer has a certain permeability. during the coalbed dewatering, the pressure difference between the aquifer and the coalbed may drive the interlayer leakage recharge, and accordingly water production of the gas well may change, thus impressing the coalbed dewatering and depressurizing. the scholars in china and abroad have frequently studied interlayer leakage of the reservoir. in the 1960s and 1970s, some scientist elsewhere investigated the reservoir seepage in the case of leakage (katz and tek, 1962; russell and prats, 1962). in china, jia yonglu established the dynamic model of double-layer reservoir with inter-layer leakage in 1997 (jia 1997). then proposed the double-layer dual-porosity model, double-layer triple-porosity model and triple-layer confined model with inter-layer leakage (zhang et al., 2009; chen et al., 2011; huo et al., 2006). sun hedong, gao chengtai, et al. established the dynamic model of infinite triple-layer oil and gas reservoir with inter-layer leakage. wu yilu analyzed the causes for the leakage in the multi-layer gas reservoir and established the mathematical seepage model for the multi-layer gas reservoir with leakage [12]. for the models established in previous studies, all layers were perforated for exploitation, and horizontal radial seepage occurred in each layer, except for inter-layer leakage. however, when the coalbed is perforated for production, the aquifers in the roof and floor are selectively avoided, so horizontal radial flow only occurs in the coalbed. therefore, the existing models for multi-layer reservoir with leakage are not suitable for the analysis of inter-layer leakage in the cbm wells. according to the unsteady seepage theory, the infiltration model for cbm wells considering the inter-layer leakage can be established to obtain the analytical solution of bottom hole flowing pressure under laplace space. after the solution in true space is obtained by stehfest numerical inversion, new type curves can be established. then, these type curves are used to judge whether any leakage recharge exists in the coalbed during dewatering, so as to determine the formation parameters and leakage parameters and analyze the level of leakage recharge. 2. methods 2.1. physical model as shown in figure 1, for a cbm well in the infinite formation center, only the coalbed is perforated for dewatering and gas production. compared with the water content in the coalbed, the water body of the aquifer in the roof and floor can be regarded as infinite water body with a great volume (su and lin, 2009). aquitard with low permeability exists between coalbed and water body. based on the data statistics of the powder river basin in the united states, the vertical permeability of such aquitard is 0.003-0.03 md, suggesting a certain impermeable capacity (onsager and cox, 2000). according to the research on the leakage process of the underground water by hantush and jacob, et al., the model is based on the following assumptions (hantush and jacob, 1955a; hantush 1955b): 1). each layer is homogeneous and isotropic, neglecting the effect of capillary pressure and gravity; 2). each layer has an infinite boundary; 3). only the coalbed is perforated, and the formation water in the coalbed flows into the wellbore in the form of horizontal radial flow; 4). the formation water in the floor recharges the coalbed through the aquitard in the manner of vertical seepage and then the inter-layer leakage occurs; 5). the compressibility of the rock in the aquitard is not considered; 6). due to the significant volume, the water body in the aquifer can be regarded as the constant pressure recharge boundary for the coalbed water. 2.2. mathematical model and solution in the coalbed with leakage recharge, the basic mathematic equation for water seepage is expressed as (streltsova, 1988): (1) the aquifer is regarded as the infinite water body, so the boundary between the aquifer and the aquitard can be treated as the constant pressure boundary. the flow rate recharged into the coalbed by the aquifer through the aquitard is (2) based on the definition of leakage coefficient , the dimensionless leakage coefficient is defined as: (3) equation (2) is substituted into equation (1). according to the definitions of dimensionless pressure, dimensionless time and dimensionless radius, the dimensionless partial differential equation is obtained after simplification. (4) initial condition: ; outer boundary condition: ; inner boundary condition: 19solution and type curves for the seepage model of the water-bearing coalbed with leakage recharge according to the definition of laplace transform: v equation (4) is transformed to ordinary differential equation: (5) similarly, the expressions of the stablish condition in laplace space are as follows: (6) (7) (8) (9) equation (5) is zero-order bessel equation with imaginary argument, and its laplace spatial solution is (10) due to , where is the euler constant and 1.781ã = , equation (10) is deformed as: (11) 3. results and discussion in the traditional well testing analysis, the reservoir permeability, skin factor, and other formation parameters can be determined, with the measured data in type curve matching. for the coalbed with leakage recharge, the leakage coefficient should also be determined and the level of interlayer leakage recharge be analyzed. next, considering the dewatering test with continuous production and water injection test with constant flow rate, the corresponding type curves are derived. 3.1. type curves of the bottom hole pressure in the dewatering with constant production the drawdown well test is conducted to measure the change of the bottom hole fluid pressure with time in the process of dewatering with constant production. according to the ideas of ramey, gringarten, and bourdet to make type curves (liu, 2003). new type curves applicable to dewatering test analysis with constant production can be derived by establishing a new combination of parameters. the dynamic change of the bottom-hole pressure tested is always susceptible to the wellbore storage effect. even before the wellbore storage effect ends, the leakage effect of the aquifer has superseded the influence of after flow. to avoid the interference of wellbore storage effect, the well should be opened or shut in by using downhole control valve in the test process (yang et al., 2015). the wellbore storage coefficient is minuscle and can be ignored, i.e. 0cd = . let , then . equation (11) can be simplified as: (12) it is proved below that is the laplace transformation result of dp based on . (13) equation (12) can be finally written as: (14) the solution in right space can be calculated by using stehfest numerical inversion. the basic equation is [19]: (15) where (16) (17) in the log-log coordinate system, a group of dimensionless pressure type curves (figure 2) are prepared, with as ordinate against as abscissa and -2sdeb as the curve parameter. each type curve corresponds to a value of -2sdeb . similarly, in the log-log coordinate system, is calculated based on the definition of dimensionless pressure derivative . a group of pressure derivative type curves (figure 3) are drawn, with as ordinate against as abscissa and -2sdeb as the curve parameter. each type curve corresponds to a value of -2sdeb . in figure 3, when the inter-layer leakage does not happen, i.e. 0=-2sdeb , the model only has radial seepage, and the derivative curve of the bottom-hole pressure is a horizontal line, which conforms to the study results of gringarten, bourdet, et al. (yang et al., 2015). according to the definition of db , the effect of aquitard thickness confh and vertical permeability confk ,v on the formation water leakage 20 wangang chen, yu yang, hansen sun, chengwei zhang, qin wen and yanyan li is comprehensively considered. from the underground fluid seepage law, we can see that the smaller the aquitard thickness confh is, or the bigger the vertical permeability confk ,v of the aquitard is, the weaker the barrier action of the aquitard on the vertical flow of the formation water is and the stronger the effect of the formation water leakage on coalbed seepage is. figures 2-3 indicates that the bigger the value of db is (i.e. the bigger the value of -2sdeb is), the earlier the leakage affects the coalbed water seepage and the earlier the coalbed radial flow stage ends. in the actual dewatering and production of the coalbed, it is also proved that when the surrounding rock of the coalbed roof and floor has high permeability and strong leakage effect, desorption of cbm can only occur in the small range near the borehole, which is manifested as the deficiency of gas supply (zhang et al., 2013). according to the definition of dimensionless pressure it is shown that the smaller dp is, the higher the bottom-hole flowing pressure is. in other words, the inter-layer leakage may increase the bottomhole pressure and affect desorption of the methane. according to the characteristics of the type curves, the existence of leakage recharge in the aquifer during testing can be qualitatively judged. figure 2 type curves of dimensionless pressure in case of inter-layer leakage. figure 3 type curves of dimensionless pressure derivative in case of inter-layer leakage. then, the formation parameters can be obtained by using the type curves shown in figures 2-3. 3.1.1. formation parameters matching by using the type curves first, the actual testing data obtained from the well site is processed. on the log-log coordinate with the same size of type curves, the curve is drawn with actual differential pressure as ordinate and actual dewatering time t as abscissa and then matched with the type curves in figure 2. permeability can be obtained through the pressure matching value. (18) the skin coefficient can be obtained by time matching value. (19) the leakage coefficient can be obtained by curve matching value. (20) 3.1.2. formation parameters matching by using the type curves of pressure derivative similarly, on the log-log coordinate with the same size of pressure derivative type curves, the curve is drawn with the product of actually measured differential pressure derivative and time t as ordinate against actual dewatering time t as abscissa, and then matched with the interpretation chart. permeability can be obtained through the pressure derivative matching value. (21) by time matching, the skin coefficient can be obtained and its calculation equation is same with equation (19). by curve value matching, the leakage coefficient can be obtained and its calculation equation is same with equation (20). 3.2. type curves of the bottom-hole pressure in the water injection test with constant production in the process of dewatering test, the bottom-hole pressure may drop below the critical desorption pressure, resulting in the gas and water (two-phase) flow, which may interfere the well test interpretation. in order to prevent such two-phase flow, the coalbed can be tested by using the method of water injection with constant flow rate. in the process of water injection, the added value of the bottom-hole pressure is (22) 21solution and type curves for the seepage model of the water-bearing coalbed with leakage recharge the differential pressure is transformed to , then all the parameters may be determined and the effect of leakage on gas well production can be analyzed by using the type curves in figure.2 and figure.3 directly and matching of equations (18)-(21). 4. conclusions (1) the mathematical model for inter-layer leakage in cbm wells is established. under bottom hole open and shut-in condition, the true spatial solution of the mathematical model is obtained by using laplace transformation and stehfest reverse inversion. (2) new combination of parameters is developed under the precondition of dewatering or water injection with constant production, respectively. on the log-log coordinate, the type curves of and , and are drawn, respectively. so the existence of leakage recharge in the aquifer in the process of testing can be judged. the characteristics of the type curves show that the bigger the leakage efficient is, the earlier the leakage recharge is and the higher the bottom-hole pressure is. (3) by type curve matching, the coalbed permeability, skin factor, and leakage coefficient can be determined. the curve matching contributes to quantitatively evaluating the level of leakage recharge, and judging the effect of the aquifer on the coalbed dewatering and production, so that subsequent dewatering and production can be adjusted timely and properly. explanation of symbols p indicates the pressure, mpa; r indicates the radius, m; indicates the pressure transmitting coefficient, cm2/s; t indicates the production time, h; indicates the fluid viscosity, mpa·s; zv indicates the seepage velocity of aquitard, m/s; k indicates the permeability of coalbed, md; h indicates the coalbed thickness, m; indicates the porosity, decimal; tc indicates the comprehensive compressibility, m3/m3; indicates the permeability of aquitard, md; indicates the aquifer pressure, with value equal to initial formation pressure, mpa; confh indicates the aquitard thickness, m; b indicates the leakage coefficient, md/m; db indicates the dimensionless leakage coefficient, dimensionless; wr indicates the borehole radius, m; dp indicates the dimensionless pressure, dimensionless; dr indicates the dimensionless radius, dimensionless; dt indicates the dimensionless time, dimensionless; dc indicates the dimensionless wellbore storage coefficient, dimensionless; indicates the dimensionless bottomhole flowing pressure, dimensionless; s indicates the skin factor, dimensionless; z indicates the laplace transform variable based on dt , dimensionless; dp indicates the image function of laplace transform about dimensionless bottom-hole pressure, dimensionless; indicates zero-order second class modified bessel function, dimensionless; indicates first-order second class modified bessel function, dimensionless; indicates euler’s constant, ; z' indicates the laplace transform variable based on , dimensionless; i indicates the natural number, dimensionless; n indicates the incremental number; indicates the dimensionless pressure derivative, dimensionless; q indicates daily water production, m3/d; b indicates the water volume factor, m3/m3; indicates the producing pressure drop, mpa; ip indicates the initial formation pressure, mpa; indicates the bottom-hole flowing pressure, mpa; indicates the drawdown pressure derivative. acknowledgements this research was financially supported by china’s national major and special project in the 12th five-year plan “development demonstration project for the coalbed methane vertical wells in the south of shanxi qinshui basin (phase ii)” (grant no. 2011zx05060) and 13th fiveyear plan “combined demonstration project for the coalbed methane, tight gas and shale gas in linxing-shenfu area” (grant no. 2016zx05066). references chen, f. f., jia, y. l., & zhang, f. x. (2011). double-layer triple-porosity seepage model and well test curves with inter-layer leakage. well testing, 20(1), 10-13. cox, d. o., & onsager, p. r. (2002). application of leaky aquifer type curves for coalbed methane characterization. society of petroleum engineers, 1, 1-10. gao, c. t. (1990). interpretation theory for stratified transient well test of leaky multi-layer oil reservoir. acta petrolei sinica, 11(3), 72-81. hantush, m. s., & jacob, c. e. (1955a). non-steady radial flow in an infinite leaky aquifer. eos transactions american geophysical union, 36(1), 95-100. hantush, m. s., & jacob, c. e. (1955b). steady three-dimensional flow to a well in a two-layered aquifer. transactions american geophysical union, 36(2), 286-292. huo, j., jia, y. l., & wang, h. t. (2006). oil and gas reservoir model for multilayer channeling and dynamic monitor down-hole pressure. well testing, 15(2), 1-4. jia, y. l. (1997). dynamic model of down-hole pressure of reservoirs with leakage of two pays. natural gas industry, 17(1), 52-54. katz, m. l., & tek, m. r. (1962). a theoretical study of pressure distribution and fluid flux in bounded stratified porous systems with leakage. spe journal, 68(3), 225-234. liu, n. q. (2003). practical modern well test interpretation method. petroleum industry press, beijing. liu, z. l. (2014). the stagnant air and haze in china: where come and go. biotechnology: an indian journal 10, 14027-14031. meng, z. p. (2010). theories and methods of coalbed methane development geology. science press, beijing. pp. 198-200. onsager, p. r., & cox, d. o. (2000). aquifer controls on coalbed methane development in the powder river basin. wyoming. society of petroleum engineers, 1, 1-8. russell, d. g., & prats, m. (1962). the practical aspects of inter-layer leakage. jpt, 54(6), 589-593. su, x. b., & lin, x. y. (2009). coalbed methane geology. coal industry press, beijing. streltsova, t. d. (1988). well testing in heterogenous formations. john wiley & sons inc., new york. pp. 313-314. stehfest, h. (1970). remark on algorithm 368 numerical inversion of laplace transforms. communications of acm, 13(10), 624-625. sun, h. d., liu, l., & zhou, f. d. (2003). exact solution and type curve for infinite three layers reservoir with leakage. mineralogy and petrology, 23(1), 101-104. wu, y. l. (2004). pressure behavioral characteristics and well testing analysis theoretical research of stratified gas reservoir. southwest petroleum institute, 17-39. yang, y., zhang, f. d., & sun, h. s. (2015). well test analysis. geological publishing house, beijing. zhang, l. h., wang, h. t., & jia, y. l. (2009). seepage model of dualporosity media double-layer reservoir with inter-layer leakage. journal of southwest petroleum university (edition of natural science), 31(5), 178-182. zhang, s. b., sun x. b., guo h. y. (2013). identifying method of leakage recharge in cbm wells. coal geology & exploration, 41(5), 29-32. seismotectonic-geologic hazards zoning map of iran is drawn based on deterministic seismic hazards evaluation using the seismicity records, structural trends, tectonic settings, fault ruptures and neotectonics activities in iran. this map has been prepared to indicate the seismotectonic-geologic hazards of iran. it contains the earthquake hazards parameters such as b value and mmax for the nineteen seismotectonic provinces. furthermore, moho discontinuity depth, seismogenic layer depths and seismic rates for all provinces are investigated. the majority of deformation in iran has been concentrated in the continental crust of the country. zagros and alborz experience deep earthquakes that are an indication of existence of the thick-skinned tectonics. finally, some provinces such as piranshahrborojen, east iran, naienrafsanjan and east alborz, which include main suture zones of iran, have faced high seismic hazards. este estudio se realizó con el fin de preparar un mapa zonal de amenazas seismotectónicas y geológicas de irán con base en la evaluación de amenazas sismícas deterministas a través de los registros sísmicos, las tendencias estructurales, las configuraciones tectónicas, las rupturas de fallas y la actividad neotectónica en irán. este mapa contiene los parámetros de amenaza sismológica como los valores b y mmax para las 19 provincias sismotectónicas del país. además, se investigó la profundidad de la discontinuidad de mohorovičić, las capas sismogénicas profundas y las clasificaciomes sísmicas para todas las provincias. la mayoría de la deformación de irán está concentrada en la corteza continental del país. las provincias de zagros y alborz experimentan terremotos profundos que indican la existencia de una configuración tectónica de capa gruesa. finalmente, algunas provincias como piranshahr-borojen, este de irán, naien-rafsanjan y este de alborz, que incluyen las principales zonas de sutura de irán, han enfrentado grandes riesgos sísmicos. earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 7 13 abstract resumen key words: seismotectonic, hazards, zoning map, iran, seismicity, earthquake palabras clave: sismotectónica, amenazas, mapa zonal, irán, sismicidad, terremoto. seismotectonic-geologic hazards zoning of iran mehran arian department of geology, science and research branch, islamic azad university, tehraniran; mehranarian@yahoo.com seismology record manuscript received: 5/11/2013 accepted for publication: 14/01/2015 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n1.40664 8 mehran arian introduction the deterministic assessment of the seismotectonic-geologic hazards of iran involves specifying the magnitude ranges and dominant mechanism of earthquakes that might bring about destructive effects on the regions and main cities. a seismotectonic province is a homogenous statistical unit that has distinguishable time, location and magnitude ranges for expected seismic events (arian, 2010). each seismotectonic province has a range of seismogenic layer depths and dominant mechanism of earthquakes, as these are related to tectonic settings, neotectonics activities, mechanical stratigraphy and deformational models of crust. therefore, earthquake hazard parameters can be estimated for them. in this study, regional tectonics (qorashi and arian, 2011) was integrated with seismologic data to prepare the seismotectonic-geologic hazards of iran. the main steps in this study have been summarised as follows: -seismic records such as archaeological, historical (berberian and yeats, 1999, 2001) and instrumental seismic data up to the year 2014 which have provided information on the seismic sources have been gathered from different sources. the seismotectonicgeologic hazards map is identified based on the physiographic-tectonic zoning map of iran’s sedimentary basins (arian, 2013). these basins described the locations of different structural units of iran (fig. 1). -seismicity parameters were estimated for each seismotectonic province. materials and methods at the first time, ambraseys and melville (1982) provided iran’s historical earthquakes and, subsequently, earthquake catalogue of iran was compiled by berberian (1994). investigation of seismicity of iran reveals that significant destructive earthquakes are related to the primary active faults of different tectonics settings (engdahl, et al., 2006). iran is one of the seismically active areas of the alpinehimalayan orogenic belt. based on previous work on neotectonics regime in iran, zagros in south iran is the most active zone (arian et al., 2002 and 2003; arian and qorashi 2006; arian and hashemi, 2008; rahimi and arian, 2014; arian and aram, 2014). then, alborz in north iran (arian and pourkermani 2004; feizi et al., 2007; khavari et al., 2009 and 2010; sorbi et al., 2011; poroohan et al., 2009 and 2013; arian et al., 2011a and 2012; arian and bagha, 2012;nouri et al,3013; bagha et al., 2014; moghimi et al., 2015) and central iran (arian, 2010; housini toudeshki et al., 2011;houseni toudeshki and arian, 2011; arian et al., 2011b, c, d; pourkermani and arian, 2001 have been situated in the next orders. the seismotectonic provinces of iran were formerly studied by several researchers such as abdalian (1963), berberian (1976), nowroozi (1976), and tavakoli and ghafory-ashtiany (1999). in this study, based on updated catalogue of significant and damaging iranian earthquakes, a new seismotectonic -geologic hazards zoning map has been proposed. the boundaries of the provinces are established through consideration of geologic materials, seismic sources, tectonic settings, and major faults and physiographic divisions (fig. 2). earthquake hazard parameters seismic hazard analysis requires an assessment of earthquake hazard parameters and the future earthquake potential in a region. earthquake hazard parameters such as, maximum expected magnitude, mmax, seismic rate and b value have been evaluated for each seismotectonic province of iran by kijko and selevoll (1992) method. the results, which are shown in table 1, were estimated for each seismotectonic provinces of iran. result and discussion zagros province the dominant structural trends in zagros province (fig. 2) are nw-se in north-western part and e-w in the south-eastern section. from tectonics figure 1. physiographic-tectonic zoning map of iran’s sedimentary basins iran, modified from arian (2013). numbers in this figure are; 1: zagros-east taurus hinterland, 2: persian gulf mesopotamian foreland basin, 3: makran accretionary prism, 4: bashagard mountains, 5: jazmorianmashkel fore arc basin, 6: shahsavaran -soltan magmatic arc, 7: south lut-south helmand back arc basin, 8: east iran mountain belt, 9: westcentral alborz and lesser caucasus hinterland, 10: great kavir-northen urmieh lake foreland basin, 11: south great kavir fold and thrust belt, 12: south caspian-black sea foreland basin, 13: urmiehdokhtar magmatic arc, 14: naien-kerman retro arc foreland basin, 15: sanandaj-sirjan overthrust belts, 16: east alborz or binalod hinterland, 17: torbat-e am-neyshabour retro arc foreland basin, 18: kopet dagh hinterland, 19: south caspian remnant basin, 20: maiamay -taibad inverted back arc basin, 21: khafkavir plain magmatic arc, 22: lut plain-gonabad back arc basin, 23: tabas hinterland, 24:yazd-khour piggy back basin. figure 2. the seismotectonic-geologic hazards zoning map of iran. 9seismotectonic-geologic hazards zoning of iran point of view, this province contains the overthrust and simple folded belts of zagros which has been formed in the north-eastern part of arabian plate’s passive margin. zagros mountains extend to east taurus mountains in turkey and are called zagros-east taurus hinterland (fig.1). focal mechanisms of numerous earthquakes are reversed and thrust (beberian, 1995) such as naghan and ardal (ms=6.1, 1977) and/ or right slip in related to n-s strike slip faults such as kazeron (se boshehr ms=6.2, 2013) or kareh bas (firozabad-fars ms=6.1, 1994, arian et al, 2006). zagros province experiences low to moderate earthquakes with high frequency, short repeat time and 10-20km focal depth. the intensity of earthquakes is in middle levels in which there are several ductile rocks (thick evaporates) in sedimentary covers such as hormuz, dashtak, and gachsaran formations. sometimes in the northern margin of this province, focal depths exceed to 70km (hatzfeld et al.2003) that is an indication of initial stages of thick-skinned tectonics. the most severe seismic hazards in zagros province, including large cities such as ilam, ahvaz, yasoj, boshehr and shiraz, are landslides in high regions of northern margins, settlements in some lowlands and coastal areas and tsunami in southern margins. piranshahrborojen province the dominant structural trend in piranshahrborojen province (fig. 2) is nw-se. from tectonics point of view, this province contains the north margin of overthrust belts of zagros formed on the northeastern part of arabian plate’s passive margin. focal mechanisms of several earthquakes are a right slip about main recent faults of zagros such as doroud (ms=6.1, 2006). piranshahrborojen province is struck by moderate to high earthquakes with low frequency, long repeat time and 10-15km focal depth. the intensity of earthquakes is at high levels. sometimes, focal depths exceed to 70km which is an indication of initial stages of thick-skinned tectonics. the most severe seismic hazards in piranshahrborojen province, including cities such as kermanshah, khoramabad, borojerd and shahrekod, are landslides in high regions of northern margins, settlements in some lowlands and surface faulting. bandar-e abbas province the dominant structural trend in bandar-e abbas province (fig. 2) is nesw. from tectonics point of view, this province contains the south-eastern part of overthrust and simple folded belts of zagros. focal mechanisms of several earthquakes are reversed and thrust such as khorgo (ms=7, 1977). bandar-e abbas province is hit by moderate to high earthquakes with low frequency, short repeat time and 10-20km focal depth. the intensity of earthquakes is in middle levels in which there are ductile rocks in sedimentary covers. the most severe seismic hazards in bandar-e abbas province, containing large cities such as bandar-e abbas, gheshm, bandar-e lengeh and hajiabad, are rapid uplifting in northern margins, settlements in some lowlands and coastal areas and tsunami in southern margins. makran province the dominant structural trend in makran province (fig. 2) is e-w. from tectonics point of view, this province contains the accretionary prism on trench of oman sea subduction that extends to pakistan (fig. 1). focal mechanisms of a vast number of earthquakes are reversed and thrust such as pasni in pakistan (ms=8.3, 1945). makran province experiences low to moderate earthquakes with low frequency, very long repeat time and down to 20km focal depth. the intensity of earthquakes is in the middle to high levels. the most severe seismic hazards in makran province, which contains large cities such as bandar-e chabahar and nikshahr, are rapid uplifting in northern margins, settlements in some lowlands and coastal area and tsunami in southern margins. jazmorian province the dominant structural trend in jazmorian province (fig. 2) is e-w. from tectonics point of view, this province contains a fore arc basin (jazmorian plain) which extends to mashkel plain in pakistan (fig. 1). focal mechanisms of many earthquakes are reversed and thrustand sometimes normal (in relation to bending of subducting oceanic lithosphere) such as nw saravan with 69 km focal depth (ms=5.2, 2009) and se khash with 82 km focal depth (ms=7.7, 2013). jazmorian province experiences moderate earthquakes with middle frequency, medium repeat time and down to 150km focal depth. the intensity of earthquakes is in middle levels by very deep hypocenters. the most important seismic hazards in the jazmorian province, containing large cities such as saravan and iranshahr, are rapid uplifting and settlements in some lowlands. it should be noted that low seismicity in makran and jazmorian provinces is related to slow subduction rate and steeper dip of benioff zone. current subduction rate of oceanic crust (related to late cretaceous) is 19 millimetres per year (vernant et al. 2004). urmiehsirjan province the dominant structural trend in urmiehsirjan province (fig. 2) is nwse. from tectonics point of view, this province contains two different units: a) urmieh-dokhtar magmatic arc (fig. 1) which is the result of subduction beneath the southern active continental margin of cimmerian plate. b) sanandajsirjan overthrust belts (fig. 1) is formed by metamorphic rocks of the northeastern part of arabian plate (alavi, 1994). focal mechanisms of several earthquakes are reversed and thrust such as changureh (ms=6.4, 2002) the urmieh-sirjan province is struck by low to moderate earthquakes with low frequency, medium repeat time and down to 10km focal depth. the intensity of earthquakes is in high levels in which there are cold igneous rocks. sometimes, focal depths exceed to 70km that it is an indicator for initial stages of thick-skinned tectonics. the most serious seismic hazards in urmiehsirjan province, which contains large cities such as urmieh, sirjan, sanandaj, hamadan, arak, ghom and isfahan, are landslides in high regions of northern margins, settlements in some lowlands and surface faulting. lut province the dominant structural trend in lut province (fig. 2) is nw-se. from tectonics point of view, this province contains a retro arc (back arc) basin (lut plain) which extends to helmand plain in pakistan (fig. 1). focal mechanisms of several earthquakes are reversed and thrust and/ or right slip about n-s strike slip faults such as nayband and nehbandan. lut province experiences low earthquakes with low frequency, medium repeat time and down to 10km focal depth. the intensity of earthquakes is in high levels in which there are hot igneous rocks. the most serious seismic hazards in lut province, which contains lut plain, are settlements, surface faulting and volcanic hazards in southern and western parts of the plain. east iran province the dominant structural trend in east iran province (fig. 2) is n-s. from tectonics point of view, this province contains a collision belt brought about by eastward subduction of east-central iran miniplate (fig. 1) beneath central blocks of afghanistan. this collision belt has been formed by a dextral transpression zone. focal mechanisms of many earthquakes are right slip in relation to n-s strike slip faulting such as zirkuh (berberian et al.,1999) in qa’enat (ms=7.2, 1997) and thrust such as west zabol (ms=6.2, 1993). east iran province is hit by moderate to high earthquakes with middle to high (on fault terminations) frequency, medium repeat time and down to 15km focal depth. the intensity of earthquakes is at high levels. the most serious seismic hazards in east iran province, which contains large cities such as nehbandan, zahedan and khash, are landslides in high regions, settlements in some lowlands, surface faulting and dispersed volcanic hazards. gonabadbirjand province the dominant structural trend in gonabadbirjand province (fig. 2) is nw-se. from tectonics point of view, this province contains the northern part 10 mehran arian of east-central iran miniplate (fig. 1). it has been impressed by splays at the north termination of nehbandan fault system. focal mechanisms of numerous earthquakes are left slip in relation to w-e strike slip faulting such as dasht-e bayaz (ms=7.4, 1968) and right slip in relation to nw-se strike slip with thrust component along longitudinal faults such as south ghaen (ms=5.6, 2012). gonabadbirjand province experiences moderate to high earthquakes with low frequency, long repeat time and down to 10km focal depth. the intensity of earthquakes is at high levels. the most serious seismic hazards in gonabad birjand province, which contains large cities such as gonabad and birjand, are landslides in high regions, settlements in some lowlands, and surface faulting. tabas-minab province the dominant structural trend in tabas-minab province (fig. 2) is n-s. from tectonics point of view, this province contains the tabas, nayband, kuh banan, gowk, bam, sazvaran and minab fault systems. focal mechanisms of several earthquakes are dextral strike-slip faulting with reverse component in southern part such as bam (ms=6.6, 2003) and reverse faulting with dextral strike slip component in northern part such as tabas (ms=7.4, 1978) and western part such as zarand (ms=6.5, 2004). tabas-minab province experiences moderate to high earthquakes with low frequency, long repeat time and down to 15km focal depth. the intensity of earthquakes is at high levels. southward earthquake migration is index characteristic in this province. the most severe seismic hazards in tabas-minab province, containing large cities such as tabas, minab, kerman, bam, jiroft and jask, are landslides in high regions, settlements in some lowlands, surface faulting. kashmaranarakbafgh province the dominant structural trend in kashmaranarakbafgh province (fig. 2) is ne-sw. from tectonics point of view, this province contains the northern and western part of east-central iran miniplate (fig. 1). focal mechanisms of several earthquakes are sinistral strike-slip faulting and thrusting. kashmaranarakbafgh province experiences low to moderate earthquakes with low frequency, medium repeat time and down to 20km focal depth. the intensity of earthquakes is at low levels. the most serious seismic hazards in kashmaranarakbafgh province, which contains large cities such as kashmar, anarak and bafgh, are settlements in some lowlands and surface faulting. naienrafsanjan province the dominant structural trend in naienrafsanjan province (fig. 2) is nw-se. from tectonics point of view, this province contains the extensive part of naien-kerman retro arc foreland basin (fig. 1). focal mechanisms of several earthquakes are thrusting. naienrafsanjan province experiences low to moderate earthquakes with low frequency, medium repeat time and down to 20km focal depth. the intensity of earthquakes is at low levels. the most serious seismic hazards in naienrafsanjan province, which contains large cities such as naien, rafsanjan and yazd, are settlements in some lowlands and surface faulting. great kavirtaibad province the dominant structural trend in great kavirtaibad province (fig. 2) is nw-se in the eastern part and ne-sw in the western part. from tectonics point of view, this province contains the maiamaytaibad marginal back arc basin, southeastern part of great kavirurmieh lake foreland basin and south great kavir folded and thrust belt (fig. 1). the latter case is the result of the collision between the northern part of east-central iran miniplate and southeastern part of north-central iran miniplate.there are many salt diapirs that have been formed by tectonic forces (arian, 2012, asadian and arian 2009, pourkermani and arian, 1997) focal mechanisms of several earthquakes are thrusting and dextral strike-slip faulting in the eastern part and sinistral strike slip faulting and thrusting in western part such as garmsar (ms=5.1, 2007). great kavirtaibad province is hit by low to moderate earthquakes with low frequency, medium repeat time and down to 10km focal depth. the intensity of earthquakes is at low levels. the most serious seismic hazards in great kavir taibad province, which contains large cities such as semnan, sabzehvar and torbat-e heydarieh, are settlements in some lowlands and surface faulting. east alborz province the dominant structural trend in east alborz province (fig. 2) is nw-se in the eastern part and ne-sw in the western part. from tectonics point of view, this province contains the torbat-e jam-neyshabour retro arc foreland basin and east alborz (fig. 1) hinterland formed in a fore arc basin on the north-eastern part of cimmerian plate. an overthrust belt with sw vergence has been formed. focal mechanisms of several earthquakes are thrusting such as neyshabour (ms=7.4, 1405). east alborz province experiences low to moderate earthquakes with high frequency, short repeat time and down to 20km focal depth. the intensity of earthquakes is at high levels. the most serious seismic hazards in east alborz province, which contains large cities such as neyshabour, esfaraien and shahrod, are landslides in high regions, settlements in some lowlands, surface faulting. kopet dagh-middle caspian province the dominant structural trend in kopet dagh-middle caspian province (fig. 2) is nw-se. from tectonics point of view, this province contains the kopet dagh hinterland (fig. 1) or kopet dagh folded and thrust belt that has been formed in the passive margin of eurasia plate. focal mechanisms of several earthquakes are dextral strike slip faulting such as garmkhan-e bojnourd (ms=6.4, 1996) and thrusting. dagh-middle caspian province experiences low to moderate earthquakes with high frequency, short repeat time and down to 20km focal depth. sometimes in the northern margin of south caspian basin, focal depths exceed to 70km which is a depiction of its subduction. the intensity of earthquakes is at low levels. the most serious seismic hazards in dagh-middle caspian province, which contains large cities such as mashhad, ghochan and bojnourd, are landslides in high regions, settlements in some lowlands, and surface faulting. south caspian province the dominant structural trend in south caspian province (fig. 2) is nw-se in the eastern part and n-s in the western part. from tectonics point of view, this province contains the south caspian remnant oceanic basin (berberian, 1983) and northern foreland basin of central alborz hinterland (fig. 1). focal mechanisms of several earthquakes are reversed to thrust (jackson et al., 2002) such as astara (ms=6.2, 1980). south caspian province is struck by low to moderate earthquakes with middle frequency, short repeat time and down to 20km focal depth. the intensity of earthquakes is at low levels. the most serious seismic hazards in south caspian province, which contains large cities such as gorgan, sary and rasht, are settlements in some lowlands and tsunami. central alborz province the dominant structural trend in central alborz province (fig. 2) is nesw in the eastern part and nw-se in the western part. from tectonics point of view, this province contains the middle part of great kavirurmieh lake foreland basin and central alborz hinterland (fig. 1) that has been formed in the northern active margin of the cimmerian plate. focal mechanisms of several earthquakes are sinistral strike-slip faulting such as manjil (ms=7.3, 1990) and reversed with a sinistral strike-slip component such as boien zahra (ms=7.2, 1962). central alborz province experiences moderate to high earthquakes with low frequency, long repeat time and down to 20km focal depth. the intensity of earthquakes is at high levels. the most serious seismic hazards in central alborz province, which contains large cities such as tehran, 11seismotectonic-geologic hazards zoning of iran ghazvin and zanjan, are landslides in high regions, settlements in some lowlands, surface faulting and volcanic hazards around damavand cone. azerbaijan province the dominant structural trend in azerbaijan province (fig. 2) is nwse. from tectonics point of view, this province contains the north-western part of great kavirurmieh lake foreland basin and west alborz – lesser caucasus hinterland (fig. 1). focal mechanisms of several earthquakes are dextral strike slip faulting such as salmas (ms=7.3, 1930) and reversed such as varzaghan (ms=6.2, 2012), that the latter case was an index case of sympathetic fault segments in iran. this segment has become active only 9 minutes after an earthquake (ms=6.4) with the strike-slip dominant mechanism in another segment of the same fault. azerbaijan province experiences moderate to high earthquakes with high frequency, medium repeat time and down to 10km focal depth. the intensity of earthquakes is at high levels. the most serious seismic hazards in azerbaijan province, which contains large cities such as tabriz, and ardebil, are landslides in high regions, settlements in some lowlands, surface faulting and volcanic hazards around sahand and sebalan cones. persian gulf province the dominant structural trend in persian gulf province (fig. 2) is nwse. from tectonics point of view, this province contains the foreland basin of zagros hinterland (fig. 1) which has been formed on the north-eastern passive margin of the arabian plate. it extends to mesopotamian terrain in iraq. focal mechanisms of several earthquakes are reversed or thrust. persian gulf province experiences low earthquakes with low frequency, medium repeat time and 10 to 20km focal depth. the intensity of earthquakes is at middle levels in which there are several ductile rocks in sedimentary covers such as hormuz and dashtak formations. the most serious seismic hazards in persian gulf province, which contains large cities such as khorramshahr, abadan, mahshahr and iranian islands, are settlements in some lowlands and coastal areas and tsunami. table 1. estimated seismic hazard parameters for seismotectonic provinces. 12 mehran arian conclusions seismotectonic-geologic hazards zoning map has been prepared to indicate the seismic hazard in iran. characteristics of this map are as follows: the map is intended as a pattern for the assessment of the geologic hazards in iran. some provinces such as zagros, persian gulf, kopet dagh-middle caspian and south caspian have rarely been surface faulted. zagros, persian gulf, kopet dagh-middle caspian, south caspian makran and bandar-e abbas provinces face tsunami hazards. the principal part of deformation in iran has been concentrated in the continental crust of the country. zagros and alborz experience deep earthquakes which is an indication of the existence of thick-skinned tectonics. jazmorian and southern margin of kopet dagh-middle caspian provinces are struck by deep earthquakes which is an indication of slow subduction of oceanic lithosphere. some provinces such as piranshahrborojen, east iran, naien rafsanjan and east alborz, which include main suture zones of iran, have faced high seismic hazards. acknowledgements this work is funded by the department of geology, islamic azad university, science and research branch, tehran, iran. also, special thanks to vice-president for research in science and research branch, tehran and dr.g.razaghian from islamic azad university, damavand branch. references abdalian, s. (1963). seismo-techtonique de i’ iran, institute of geophysics, tehran university, iran, 105pp. alavi, m. (1994). tectonics of the zagros orogenic belt of iran: new date and interpretations, tectonophysics. 229, 211-238. ambraseys, n.n., melville, c.p. (1982). a history of persian earthquakes, cambridge university press, london, 219pp. arian, m. (2010). applied seismotectonics, farazamin geologic research centre, 304pp. (in persian). arian, m. (2010). earthquake-fault hazard investigations in the kerman quadrangle. journal of sciences. 19, 176-182. arian, m. (2013). physiographic-tectonic zoning of iran’s sedimentary basins, open journal of geology. 3, no.3, 169-177. arian, m., aram, z. (2014). relative tectonic activity classification in the kermanshah area, western iran. solid earth. 5, 1277-1291 arian, m., bagha, n. (2012). active tectonics of tehran area, iran. journal of basic and applied scientific research. 2, 3805-3819. arian, m., hashemi, a. (2008). seismotectonic zoning in the zagros. journal of sciences. 18, 63-76. arian, m. (2012) clustering of diapiric provinces in the central iran basin. carbonates and evaporites. 27, 9-18. asadian, f., arian, m. (2009) identification of diapiric provinces of central iran through geological and geographical analysis, international journal of agriculture environment & biotechnology.2, 3443-3451. arian, m., qorashi, m., pourkermani, m., ahmadnia, a. (2006) the structural significance kareh bas transcurrent fault system in the zagros fold and thrust belt, geosciences. 15, 126-133. arian, m., pourkermani, m. (2004). tectonic elements of south flank in the east-central alborz mountain. journal of sciences. 4, 359-368. arian, m., qorashi, m. (2006). the movement potential evaluation of the major quaternary faults in alborz-central iran border zone, from the east of tehran to the east of semnan. geosciences.15, 184-188. arian, m., ahmadnia, a., qorashi, m., pourkermani, m. (2002). structural analysis of mengharak transcurrent fault system in zagros, iran, geoarabia. 7, 209-210. arian, m., bagha, n., khavari, r., noroozpour, h. (2012). seismic sources and neo-tectonics of tehran area (north iran), indian journal of science and technology. 5, 2379-2383. arian, m., qorashi, m. pourkermani, m., ahmadnia, a. (2003). fractal analysis of mengharak transcurrent fault system in zagros iran, abstracts of fourth international conference on seismology and earthquake engineering, tehran, iran, may, 2223 arian, m., maleki, z., noroozpour, h. (2011a). cenozoic diastrophism and deformational events in the east central alborz. journal of basic and applied scientific research. 1, 2394-2400. arian, m., toudeshki, v.h., noroozpour, h. (2011b). active tectonics of qezel ozan river basin, nw iran. journal of applied environmental and biological sciences. 1, 291-295. arian, m., pourkermani, m., sistanipour, a., noroozpour, h. (2011c). seismicity and fault segmentation of bafq-baghin fault system (central iran). journal of applied environmental and biological sciences. 1, 382-396. arian, m., pourkermani, m., sistanipour, a., noroozpour, h. (2011d) kinematic significance of foldand fault-related fracture systems in the rafsanjan’s northeast highlands (central iran), journal of basic and applied scientific research. 1, 3398-3406. bagha, n., arian, m., ghorashi, m., pourkermani, m., el hamdouni, r., solgi, a. (2014). evaluation of relative tectonic activity in the tehran basin, central alborz, northern iran, geomorphology. 213, 66-87. berberian, m. (1976).seismotectonic map of iran, geological survey of iran,39 berberian, m. (1983). southern caspian: a compressional depression floored by a trapped, modified oceanic crust, canadian journal of earth sciences, 20, 163-183. berberian, m. (1994). natural hazards and the first earthquake catalogue of iran, vol.1: historical hazards in iran prior to 1900, international institute of earthquake engineering and seismology, 603pp. berberian, m. (1995). master blind thrust faults hidden under the zagros folds: active basement tectonics and surface morphotectonics, tectonophysics. 241, 193-224. berberian, m., yeats, r.s. (1999). patterns of historical earthquake rupture in the iranian plateau, bulletin of the seismological society of america.89, no.1, 120-139. berberian, m., yeats, r.s. (2001). contributions of archaeological data to studies of earthquake history in the iranian plateau, journal of structural geology. 25, 563-584. berberian, m., jackson, j. a., qorashi, m., khatib, m. m., priestley, k., talebian, m., ghafuriashtiani, m. (1999). the 1997may 10 zirkuh (qa’enat) earthquakes (mw 7.2): faulting along the sistan suture zone of eastern iran, geophysical journal international. 136, 671-694. engdahl, e. r., jackson, j. a., myers, a. c., bergman, e. a., priestley, k. (2006). relocation and assessment of seismicity in the iran region, geophysical journal international.167, 761–778. feizi, f., arian, a., rahmani, r. (2007). seismotectonic zoning in the eastern part of the central alborz, journal of sciences. 17, 151-164. hatzfeld, d., tatar, m., priestley, k. and ghafory – ashtiany, m. (2003). seismological constraints on the crustal structure beneath the zagros mountain belt (iran), geophysical journal international.155, 403-410. housini toudeshki, v., pourkermani, m., arian, m., khosrotehrani, k.h. (2011). influence of structures on the ghezel ozan river, geosciences. 21, 55-60. housini toudeshki, v., arian, m. (2011) morphotectonic analysis in the ghezel ozan river basin, nw iran. journal of geography and geology. 3, 258-260 jackson, j.a., priestley, k., allen, m., berberian, m. (2002). active tectonics of the south caspian basin, geophysical journal international. 148, 214-242. khavari, r., arian, m., ghorashi, m. (2009). neotectonics of the south central alborz drainage basin, in nw tehran, n iran, journal of applied sciences.9, 4115-4126. khavari, r., ghorashi, m., arian, m., khosrotehrani, k. (2010). geomorphic signatures of active tectonics in the karaj drainage basin in south central alborz, n iran, geosciences. 19, 67-74. kijko, a., selevoll, m.a. (1992). estimation of earthquake hazard 13seismotectonic-geologic hazards zoning of iran parameters from incomplete data files. part ii. incorporation of magnitude heterogeneity, bulletin of the seismological society of america.82, 120-134. moghimi, h., arian, m., sorbi, a. (2015). fault movement potential of marzanabad area, north alborz, iran, open journal of geology. 5, 126-135. nouri, r., jafari, m.r., arian, m., feizi, f. , afzal, p. (2013) correlation between cu mineralization and major faults using multifractal modelling in the tarom area (nw iran), geologica carpathica. 64, 409-416. nowroozi, a. a. (1976). seismotectonic provinces of iran, bulletin of the seismological society of america. 66, no. 4, 1249-1276. poroohan, n., poukermani, m., arian, m. (2009). an assessment on correlations of seismotectonic parameters preceding and following roudbar-manjil earthquake (gilan, north of iran), australian journal of basic and applied sciences. 3, 2643-2652. poroohan, n., pourkermani, m., arian, m. (2013). an assessment of relationship in f-parameter and paleostress fields in heterogeneous lithologies: roudbar area (northwest of iran), australian journal of basic and applied sciences. 7, 933-942. pourkermani, m., arian, m. (1997). salt domes of central iran. journal of humanities. 3, 29-41. pourkermani, m., arian, m. (2001) structural geomorphology of northeastern kurdistan, journal of humanities. 7, 37-48. qorashi, m., arian, m. (2011). tectonics of iran, geologic survey of iran, 336pp. rahimi, n., arian, m. (2014). tectonic geomorphplogy of hamedan-sosangerd region, west iran, advances in environmental biology. 8, 119-124. sorbi, a., arian, m., pourkermani, m. (2011). the application of geomorphic indices to the assessment of relative tectonic activity levels in tehran quadrangle. journal of the earth. 6, 1-9. tavakoli, b., ghaforyashtiany, m.(1999). seismic hazard assessment of iran, annali di geofisica.42, no.6, 1013-1021. vernant, ph. nilforoushan, f., hatzfeld, d., abbasi, m.r., vigny, c., masson, f., nankali, h., martinod, j., ashtiani, a., bayer, r, tavakoli, f., chery, j. (2004). present – day crustal deformation and plate kinematics in the middle east constrained by gps measurements in iran and northern oman, geophysical journal international.157, 381-398. issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n3.103605 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (september, 2022): 255 262 g e o l o g y a detection method of urban underground geological anomalies in the united kingdom based on feature fusion xuemei liu* department of preschool education, hebei women’s vocational college, shijiazhuang, china *corresponding author liuxuemei@mjc-edu.cn abstract engineering geological conditions include the nature of rock and soil, geological structure, landform, hydrogeological conditions, and adverse geological processes. among them, faults, fissures, folds, karst, and lithology changes seriously affect the safety and construction cost of mountain tunnels, hydraulic tunnels, and other projects. for this reason, a new method based on feature fusion is proposed to detect the geological anomalies in london and sheffield. it established a 3d raster data model oriented to attribute information modeling and visualization of urban underground space to obtain geological data. based on this acquired data, authors adopted the feature-level fusion extraction method based on the multi-attribute geological abnormal body to extract, fuse, fill and surface the multi-attribute data of underground space geological data. smooth processing can realize the detection of abnormal geological bodies in underground space. it has been proved that this method can be used in geological data display, feature extraction, feature fusion, and abnormal physical examination. keywords: feature fusion; british city; underground space; geology; abnormal body; detection; método de detección de anomalías geológicas subterráneas en el reino unido con base en la fusión de características resumen las condiciones de ingeniería geológica incluyen la naturaleza de rocas y suelo, estructura geológica, formas de la tierra, condiciones hidrogeológicas y procesos geológicos adversos. entre estas condiciones las fallas, fisuras, plegamientos, kársticos y los cambios de la litología afectan seriamente la seguridad y los costos de producción de túneles en montañas, túneles hidráulicos y otros proyectos. por esta razón, este trabajo propone un nuevo método basado en la fusión de características para detectar las anomalías geológicas en londres y sheffield, en el reino unido. este método parte de un modelo de datos de trama 3d para realizar un modelado de información y visualización del espacio urbano subterráneo con el fin de obtener información geológica. con base en esta información adquirida, los autores adoptaron el método de extraccción de fusión de características con base en los multiatributos geológicos de un cuerpo anormal para extraer, fusionar, llenar y normalizar la información multiatributos del espacio urbano subterráneo. en el proceso de suavizado se pueden detectar los cuerpos anormales en el espacio subterráneo. los resultados prueban que este método puede usarse en la presentación de información geológica, en las características de extracción, características de fusión, y en la examinación física de los cuerpos. palabras clave: fusión de características; reino unido; espacios urbanos subterráneos; geología; cuerpos anormales; detección; record manuscript received: 09/08/2022 accepted for publication: 12/10/2022 how to cite item: liu, x. (2022). a detection method of urban underground geological anomalies in the united kingdom based on feature fusion. earth sciences research journal, 26(3), 255-262. https://doi. org/10.15446/esrj.v26n3.103605 https://doi.org/10.15446/esrj.v26n3.103605 mailto:liuxuemei@mjc-edu.cn https://doi.org/10.15446/esrj.v26n3.103605 https://doi.org/10.15446/esrj.v26n3.103605 256 xuemei liu 1. introduction the geological body refers to the natural objects with spatial forms formed in the geological era. geological bodies are often irregular objects, showing a natural, arbitrary geometry and size. geological bodies can be divided into geological forms, including ore bodies, reef shoals, rock mounds, sand bodies, karst caves, and faults (sun et al., 2018). unlike other parts of the stratum, the geological body usually occupies a space volume and greatly differs from other geological components in the same stratum (montanari et al., 2017). an abnormal geological body is a part of the concept of geological bodies. the geological anomaly body has all the characteristics of the geological body but also has some of its own. in the authors’ geological exploration and geophysical research, the range of abnormal geological bodies is different from other areas in many aspects, such as the mutation of vibration amplitude and the mutation of the waveform shape (wu et al., 2017). geological anomaly bodies play an essential role in many geological structures. the exploration and research of geological anomaly bodies play an active role in the inference of geochemical and geophysical anomalies (li et al., 2019). geochemical and geophysical anomalies are often caused by mineralization, which is the effect of variation of local chemical and physical fields in the area during the formation of the deposit, and often serves as a sign of the existence of the deposit (ping et al., 2017). there are many important geological resources such as oil and natural gas in common geological abnormal bodies such as river courses and sand bodies (zhu et al., 2018). therefore, in geophysical exploration and structural interpretation of geological data, geological anomalies will be specifically detected, identified, and visualized (bai et al., 2018). given the above characteristics and significance, geologists and researchers have done much research and achieved good results (han et al., 2020; zhang et al., 2019). structural interpretation of geological data covers almost all scientific research fields of petroleum and basic geology, including structural stratigraphy, seismic stratigraphy, and other disciplines. three-dimensional visualization technology can directly show the results of structural interpretation and other geological structure research. it can help geologists and relevant researchers better understand the geological characteristics of the surface and below the surface to better carry out subsequent research (liu et al., 2017). the research and visualization of 3d object surface reconstruction are one of the most important technologies of 3d visualization, which is widely used in the fields of geological exploration, environmental science, aerospace, health care, remote sensing, and geophysical exploration. wang xiangsheng proposed to use the geological radar to detect the geology, but the detection of this method is timeconsuming and inefficient; liu mingyang applied the electrical method to the geological survey of water conservancy engineering, but the survey accuracy of this method is controversial, and lack of case verification; han kunying and other scholars designed an automatic inspection system of geological map spatial database to apply in the geological exploration. this paper recognizes that a matching detection method should be introduced for this system to achieve high-precision detection of geological anomalies (xu & chen, 2019). nowadays, 3d visualization technology is a research hotspot of all kinds of geological and seismic software. it is a tool for intuitively expressing and displaying many geological features on and below the surface. geologists and researchers can use three-dimensional visualization software to display the results of seismic research and understand the geological structure more intuitively to pave the way for further research. in this paper, the authors propose a method of detecting the geological anomalies in urban underground space based on feature fusion. firstly, they built a three-dimensional grid data model for attribute information modeling and visualization of urban underground space to obtain the geological features of urban underground space in the uk. then, they used the method of extracting the multi-attribute anomaly data to detect the geological anomalies in urban underground space in the uk (peng et al., 2018). 2. materials and methods 2.1. the 3d raster data model for attribute information modeling and visualization of urban underground space the data of physical and chemical attribute parameters (such as stress, porosity, permeability, water quality, and water temperature) in urban underground space geological bodies differ from the information of spatial geometry structure. they are generally the continuous change (gradual change) in three-dimensional space. it is necessary to model and analyze the attribute data in the geological body and study its space. generally, the three-dimensional volume element model can be used to segment the underground space entity to form the corresponding volume element. then the specific spatial interpolation method and visualization algorithm can be used for analysis. different volume element segmentation methods and data models can be used according to research objectives, such as the three-dimensional grid, octree, and tetrahedron models. this paper uses a three-dimensional grid model to process the physical and chemical attribute parameter data modeling of urban underground space geological bodies (gao et al., 2018). the 3d grid model uses a set of regular-size 3d grids (cubes) to segment the space which will be simulated and establish a real-time association with attribute information. in the three-dimensional grid data model, the threedimensional grid is the basic unit to store the voiceless attribute information inside the geological body, and it is assumed that the function value of the physical and chemical attribute data field is distributed on the eight vertices of the grid, that is, the function value at the vertex (x i,yj,zl)is f. (xi,yj,zl) therefore, the volume data in the storage and grid can be defined as a sampling function in a discrete bounded three-dimensional space, and the sampling is structured by structure; that is, the volume data is regular structured, and the sampling is uniform in three directions of space, which can be expressed as: f x y z x x x x x x x y y y y y y i i i i i i( , , ) , ,..., ( ) , ,..., ( � � � � � � � � � � 1 2 1 1 2 1 �� � � � � � � � � � � � � � � � � � x z z z z z z xi i i ) , ,..., ( ) 1 2 1 (1) with this kind of 3d grid data model, the analysis of spatial attribute data in the geological body can be attributed to the analysis of attribute data among 3d grids, as the visualization technology can also be applied to the 3d visualization of data field represented by the 3d grid data model to analyze the spatial distribution of an attribute value (singh et al., 2017). in this paper, when using the 3d grid model to represent spatial attribute data, each vertex of the 3d grid stores two attribute data (u,v), where u represents the stratum number of the grid vertex, and v represents the physical and chemical attributes at the location of the grid vertex. in this way, the set of vertices with the same stratum number can represent the spatial position and shape of the corresponding stratum. this model can represent not only geological bodies and their attributes with complex morphology and topological structure but also analyze spatial attributes under the constraints of geological body set structure, providing constraints on spatial set structure for direct volume rendering of spatial attributes (gao, 2018). 2.2. feature-level fusion extraction method of geological anomaly based on multi-attribute in order to analyze the related problems of the underground geological abnormal body, it is necessary to introduce a cause of that body. therefore, this paper takes the earthquake as the cause of the abnormal body. based on the underground geological data extracted from the 3d grid data model of urban underground spatial attribute information modeling and visualization constructed in section 2.1, it adopts the method of multi-attribute data fusion to extract the abnormal body, to improve the characteristics of abnormal geological bodies in underground space and integrate them to achieve highprecision abnormal body detection (sheth et al., 2018). because of the complexity of the data, it is often unable to reflect all the features and details of the geological structure from a single geological attribute, so it is essential to interpret the geological structure with different geological attributes. one of the characteristics of geological data is that the amount of data is vast, tens or even hundreds of g. if it is fused directly, it will produce a great deal of computation, which ordinary computer systems cannot process. therefore, this paper proposes a feature-level fusion extraction algorithm based on a multi-attribute geological anomaly body, roughly divided into two steps: 257 a detection method of urban underground geological anomalies in the united kingdom based on feature fusion the feature extraction of multi-attribute data and the anomaly body fusion of multi-attribute data (shrestha et al., 2017). 2.2.1. feature extraction of multi-attribute data of abnormal body based on human-computer interaction researchers are usually more interested in geological structures. when dealing with multiple geological attribute data, they only need to extract their feature points for research. in this paper, the feature extraction method of multigeological attributes is based on human-computer interaction, which requires users to observe and analyze each geological data, select the range of geological attribute values of interest, and extract the scattered points of geological anomalies within the range. this paper introduces the feature extraction of abnormal bodies based on multi-attribute human-computer interaction data. firstly, open the geological texture attribute data and visualize it. secondly, using “karst cave” as an example, human-computer interaction observation data finds the “karst cave” structure. thirdly, contrast the color table to find its attribute range. finally, extract the abnormal-scattered points within the range. figure 1 is the flow diagram of the process: start visual geological data human computer interaction to confirm the attribute value range of the target color comparison table confirmation attribute range outlier scatter within the extraction range end figure 1. the extraction process of geological attribute characteristics 2.2.2. feature fusion of multi-attribute data based on fuzzy logic after the feature extraction of multiple geological attribute data, these features need to be fused to get the scattered point data of the geological anomaly body. the data of the integrated abnormal body contains the characteristics of multiple geological attributes, which reflect the different geological characteristics of the abnormal body according to the user’s emphasis on different attributes. the multi-attribute geological anomaly fusion extraction method in this paper is based on fuzzy logic reasoning. it requires users to set the importance of different geological attribute data according to their objectives to get the ideal effect of anomaly fusion extraction. the fuzzy set was put forward by l. a. zadeh, an american cybernetic expert, in 1965. since the birth of fuzzy set theory, the rapid development of the fuzzy logic system has played a role in pattern recognition, signal processing, health care, and other fields, providing great help for related research. common fuzzy inference systems are divided into three categories: pure fuzzy inference system, mamdani type fuzzy inference system, and takagi sugeno type fuzzy inference system. this paper is based on t-s fuzzy inference. the reasoning process of t-s type fuzzy includes the fuzzy operation, and its output is an exact value. this is determined by the form of t-s fuzzy reasoning itself (cotterill et al., 2017). outlier feature fusion is divided into four steps based on fuzzy logic: initialization, setting parameter list, and reading outlier feature scatter. spatial scatter registration. feature scatter fusion. and binary operation. the flow chart is shown in figure 2. in order to solve the problem of the feature fusion and extraction of abnormal geological bodies, the analytic hierarchy process (ahp) based on fuzzy reasoning is used to fuse the features of multiple geological attribute data. the factor set for the abnormal geological body is “0”. the “1” values on each grid point of each geological attribute data and the value of each grid point obtained after the feature data scatter fusion operation is the final comment set. the focus of this paper is how to extract the geological abnormal body feature scatter point data from each different geological attribute. the factor set is the comment set formed by the scattered points that belong to the target after the fusion. due to the scattered points of geological anomalies have the following characteristics, the value of each grid point is either “0” (the point is the background point) or “1” (the point is the scattered point of geological anomalies), the authors use the fuzzy reasoning algorithm to process the scattered point data of geological anomalies. at this point, adjustments and simplifications to the fusion algorithm were necessary to adapt to the actual seismic interpretation. in short, multi-geological attribute data featurefusion aims to merge multi groups of 3d grid data into a new group and binary these values. the threshold rule must be formulated according to the characteristics of geological data to meet the integration requirements of our subjective interpretation experience combined with the actual geological data (olson et al., 2017). start set the extract parameter list of the exception body fusion read a scatter of anomalous body features with multiple geological attributes three point registration in space the fuzzy logic parameters are weighted and fused end determine if it is greater than the threshold set to the exception body point binary scatter data, abnormal body point is "1", otherwise is "0" y n figure 2. abnormal body feature fusion based on fuzzy logic 2.2.3. abnormal volume data filling and surface smoothing because of the noise interference and the influence of the accuracy of the geological acquisition data, the abnormal geological bodies obtained by the proposed method are often not satisfactory enough. there are many burrs and some “holes” inside the abnormal geological bodies, which are not continuous. then, some treatments are necessary to fill the holes and smooth the surface. based on this purpose, this paper suggests a new algorithm flow for filling and smoothing the scattered point data of abnormal bodies suitable for geological data interpretation, which has had a good effect on the surface smoothing of geological structures. figure 3 is the flow of filling and smoothing the data of three-dimensional geological abnormal bodies. firstly, a binary morphological operation is carried out to fill the small holes in the abnormal body and remove the redundant data points on the 258 xuemei liu boundary of the abnormal body for the scattered point data after the previous step of multi-geological attribute feature fusion. secondly, the gaussian filtering method is used to smooth the scattered point data of the abnormal geological body. finally, the moving cube algorithm is used to find the surface of the isosurface as a three-dimensional geological anomaly body. start scattered data of abnormal body after fusion morphological image technique is used to fill holes in three points of abnormal body gaussian filter smoothing operation the isosurface is used to extract the surface of living geological anomalies end visualization of geological anomalies figure 3. flow chart of abnormal body data filling and surface smoothing in the process of morphological processing, this paper uses the expansion and corrosion algorithm. the corrosion algorithm can be summarized into three steps: 1. taking the cube of size as the structural element and scanning the scattered point data of each spatial geological anomaly successively; 2. the structural elements are fused with the scattered data of the abnormal body in the covered area; 3. if all of them are valid outlier data (represented by “1”), then this point in the result is valid. otherwise, it is not a valid outlier scatter (represented by “0”). this operation reduces the scattered point boundary of the abnormal body by one circle and removes the small and meaningless objects inside the target. the expansion operation algorithm used in this paper includes the following three steps: 1. like the corrosion operation, the cube is used as the structural element to scan the scattered point data of each spatial geological anomaly body in turn 2. make “or” calculation on scattered points of structural elements and geological abnormal body data in the covered area; 3. if there are invalid outliers (represented by “0”), this point in the result is not regarded as an outlier. otherwise, it is an outlier (represented by “1”). the result of this operation is that the small “holes” inside the abnormal geological body are filled to make it more complete. 3. results britain is the first country in the world to develop and utilize underground space on a large scale comprehensively. in 1845, charles pearson, an englishman, first proposed the “underground train” concept. eighteen years later, london built the world’s first subway. with the development of the subway, underground shopping malls, commercial streets, apartments, and parking lots have emerged in succession. at the same time, due to the cold climate in britain, cellar construction is relatively standard. for example, most of the houses in central london have basements. with the increase in the land price, the development of underground space is more common, which causes many practical problems, such as the impact of underground space excavation on the safety of adjacent buildings. for this reason, the method in this paper is used to detect the underground geological anomalies in london and sheffield. 3.1. feature extraction results the esda-gis-based feature extraction method of complex karst surface subsidence area and multi-source information integration-based drilling geological feature parameter estimation and prediction as the comparison method were tested several times to extract and compare the comprehensive characteristics of underground space in london and sheffield, uk. the comparison results are shown in figures 4 and 5. 0.75 0.8 0.85 0.9 0.95 1 1 2 3 4 5 6 7 o ve ra ll test times article method feature extraction method of complex karst ground subsidence area based on esda-gis estimation and prediction of drilling geological parameters based on multi-source information integration figure 4. comprehensive comparison results of extraction of geological characteristics of underground space in london, uk. 0 0.2 0.4 0.6 0.8 1 1.2 1 2 3 4 5 6 7 o ve ra l test times article method feature extraction method of complex karst ground subsidence area based on esda-gis estimation and prediction of drilling geological parameters based on multi-source information integration figure 5. comprehensive comparison results of extraction of geological characteristics of underground space in sheffield, uk. it can be seen from the above figure that there are some differences in the extraction of geological characteristics of underground space between london and sheffield. in this method, the total coefficient of extracting geological characteristics of underground space in london and sheffield, uk, is always the largest, and both are as high as 0.98. the total coefficient of extracting characteristics of complex karst surface subsidence area supported by esdagis and the estimation and prediction of geological parameters of drilling based on the integration of multi-source information are always lower than that 259 a detection method of urban underground geological anomalies in the united kingdom based on feature fusion of this method. when extracting the geological characteristics of underground space in london and sheffield, the scope of extraction is more comprehensive, which provides a reasonable basis for the follow-up study. 3.2. feature fusion results in order, the regional geophysical and geochemical information featurelevel fusion method, the fusion method based on multi-band wavelet transform, and the method in this paper are used to fuse the underground space geological features of london and sheffield in the united kingdom as explained in the 3.1 section. the fusion effect of the three methods is analyzed, and the fusion degree is taken as the comparison index. the comparison results are shown in figures 6 and 7. 97.56 97.25 98.56 97.88 98.88 98.88 98.88 85.32 85.68 85.21 85.45 85.64 85.58 85.69 75.69 75.87 75.35 75.69 75.87 75.35 74.99 0 20 40 60 80 100 120 1 2 3 4 5 6 7 fu si on d eg re e /% test times article method feature level fusion method of regional geophysical and geochemical information fusion method based on multi band wavelet transform figure 6. fusion effects of geological characteristics of underground space in london, uk 98.56 98.88 98.56 98.44 98.88 98.88 98.88 84.57 84.57 84.57 84.57 84.98 84.57 84.57 75.01 75.01 75.05 75.01 75.01 75.05 75.05 0 20 40 60 80 100 120 1 2 3 4 5 6 7 fu si on d eg re e /% test times article method feature level fusion method of regional geophysical and geochemical information fusion method based on multi band wavelet transform figure 7. integration effects of underground geological features in sheffield, uk it can be seen from the above figure that the highest fusion degree of the method in this paper is 98.88% and 98.88% for the geological characteristics of underground space in london and sheffield, uk. the highest fusion degree of the regional geophysical and geochemical information feature-level fusion method for the geological characteristics of underground space in london and sheffield is 85.69% and 84.98%, respectively. 3.3. abnormal body detection results based on the detection accuracy of abnormal bodies and the surface smoothness of the detection results, the detection effect of this method on the underground space geological abnormal bodies in london and sheffield is analyzed. the results are shown in figures 8 and 9. 97.4 97.6 97.8 98 98.2 98.4 98.6 98.8 99 99.2 1 2 3 4 5 6 7 d et ec tio n ac cu ra cy /% number of tests/times london sheffield figure 8. detection accuracies of underground geological anomalies in london and sheffield in this paper 97.4 97.6 97.8 98 98.2 98.4 98.6 98.8 99 99.2 1 2 3 4 5 6 7 d et ec tio n ac cu ra cy /% number of tests/times london sheffield figure 9. in this paper, the surface smoothness of geological anomalies detected in underground space in london and sheffield was studied according to figures 8 and 9, the minimum detection accuracy of the method in this paper is not less than 97% for the geological anomalies in london and sheffield, and the surface smoothness of the method in this paper is higher than 97%. therefore, this paper’s method can detect high-precision geological abnormal bodies in urban underground space, and the surface of the abnormal body is smooth and complete. 260 xuemei liu 3.4. application efficiency test the detection of geological abnormal body in underground space needs to be realized in a short time to realize disaster prevention in natural disasters such as earthquakes. therefore, the application efficiency of the method in the above experiments is tested, and the results are shown in table 1. table 1. application efficiency test results of the method in this paper (unit: ms) test times the data collection feature extraction characteristics of the fusion abnormal body detection 1 105.25 101.51 139.54 128.58 2 105.24 101.23 139.58 128.69 3 105.25 101.22 139.52 128.45 4 105.24 101.51 139.58 128.25 5 105.25 101.23 139.52 128.45 6 105.25 101.22 139.58 128.25 7 105.25 101.22 139.52 128.45 the mean 105.25 101.31 139.55 128.45 it can be seen from the data in the table that the average time consumption of this method in underground space geological data collection, feature extraction, feature fusion, and abnormal physical examination is 105.25 ms, 101.31 ms, 139.55 ms, and 128.45 ms. the overall application efficiency is (105.25 ms+101.31 ms+139.55 ms+128.45 ms) / 4=150.75 ms. it takes less time and can meet the needs of underground geological anomaly detection. 4. discussions britain is the birthplace of the modern administrative licensing legal system. its relatively complete planning and management system for the development and utilization of underground space is of reference significance for china to establish and improve the planning and management system of underground space. based on the above analysis of the relevant legal system and underground engineering planning and management in the united kingdom, the experiences can be mainly summarized as follows: (1) considering that the development and utilization of underground space is a gradual process involving many fields and related departments, the construction of relevant laws and regulations is often scattered in the relevant laws and regulations rather than forming comprehensive legislation at the national level at the beginning. britain is the first country to carry out the practice of underground space development and utilization, but it has not formed a comprehensive law of underground space. on the one hand, this is related to the local differences in the development and utilization of underground space, i.e., the areas with more underground space development are still a few urban areas. on the other hand, the purpose of underground space development is mainly based on various specific requirements such as real estate profits, city scale, and climate reasons. the central or local government will issue various special laws, standards, rules, and other changes according to the situation. it is fast and can guide practice better. in 2011, the british parliament passed the underground development and utilization bill, which reflects the demand for comprehensive legislation with the increase of underground space development and utilization. however, the relevant provisions and practical experience of various national and local laws and regulations in the united kingdom in the past 150 years have undoubtedly laid a solid foundation for formulating comprehensive legislation. (2) considering the different types of underground engineering, it is advisable to classify and carry out professional management. for underground space development projects of different natures, the project scale is often quite different (for example, the difference between the investment and construction scales of a common family basement and large-scale infrastructure has several orders of magnitude), and the technical requirements of the project (such as traffic assessment and construction safety) are the key to approval. therefore, different underground space projects should develop different planning and management procedures. in the uk, underground engineering is divided into general engineering, major infrastructure, and underground transportation facilities, and different review and approval subjects, procedures, and elements are established. local governments or planning departments have further made detailed regulations suitable for the region. it is worth noting that the planning and management cycle of major infrastructure is relatively long, it takes 28 days for the review team to examine the application, and then it takes another three months for the review team to conduct the review and the relevant ministers of state to make a decision, which shows that the review decision of primary infrastructure underground should be cautious, even at the cost of time. this idea of classified management is worth learning (pang et al., 2018). the latest regulations of shanghai on planning and construction of underground space issued in 2014 also made similar provisions, divided into “key areas” and “other areas,” and distinguished in planning formulation, planning content, scheme design, license management, and other aspects. (3) considering the closed nature of underground engineering, limited by geological and hydrological conditions, complex and unpredictable underground conditions, and the interaction between the above-ground and underground development and utilization, the preliminary investigation, research, and evaluation of underground space development and utilization is essential (pang et al., 2020). in the planning and management of underground space, great importance is attached to the investigation and evaluation work in the early stage, mainly reflected in the requirements of application elements. even for general underground engineering, the application materials shall include all kinds of research and evaluation reports on existing buildings, structural stability, geological conditions, surrounding influence, traffic influence, environmental influence, hydrogeological influence, safety, and sudden danger (wang et al., 2019). (4) in consideration of the impact of underground works on surrounding owners and the public, the planning and management process of underground works should pay attention to the publicity of schemes and relevant information, the solicitation of opinions from relevant owners, and the maintenance of public interests (xu et al., 2020). although the ownership of the general underground space development in the uk belongs to the land owner, the owners’ interests around it must be maintained, and unnecessary disputes should be avoided. therefore, in the program design process, the owners’ consent within 9 meters should be obtained in the preliminary preparation stage, and the owners within 50 meters should be ensured in the review and construction stage. for major infrastructure construction, according to the planning and compulsory purchase act 2004 and the planning act 2008, the british public sector or the government can compulsorily purchase private underground space according to the public needs and compensate according to the market value at that time. therefore, the national essential underground large-scale infrastructure facilities are often owned by the state; or their use rights enjoy their respective parts according to the agreement of public-private cooperation to ensure the public interest. however, in this process, the opinions and attitudes of the media, the public, and community organizations can still significantly affect the project’s approval. 5. conclusion in this paper, authors proposed a method of detecting the geological anomalies of urban underground space based on feature-fusion in the united kingdom, tested the geological anomalies of urban underground space in london and sheffield in england, and drew the following conclusions: 1. compared with the same method, the overall coefficient of the method is the largest in london and sheffield, and it is as high as 0.98; 2. the highest fusion degree of underground space geological characteristics of london and sheffield is 98.88% and 98.88%, respectively; 3. the minimum detection accuracy of geological anomalies in london and sheffield is not less than 97%. the surface smoothness of geological anomalies in london and sheffield is higher than 97%; 4. this method’s overall average application time is only 150.75 ms, which can achieve high-precision and high-efficiency detection of geological anomalies in underground space. with the continuous development of modern geoscience, the technicians engaged in geological experiments and tests are also faced with more challenges. the material composition of the objects to be analyzed will be gradually complicated. geologists should attach great importance to the extraction of 261 a detection method of urban underground geological anomalies in the united kingdom based on feature fusion abnormal information and innovate analytical chemistry theory by applying physics, geological chemistry, and other multi-disciplinary knowledge. it also takes this as a breakthrough point, integrating the geological experiment testing resources to innovate more operable geological experiment testing technologies and methods and improve the level of geological experiment testing. acknowledgment hebei social science foundation project: a study on the evaluation system of english language ability of students majoring in preschool education in higher vocational education under the systematic paradigm of working process (project no. hy17jy045). references bai, c. y., he, l. y., li, x. w., & sun, j. y. (2018). simultaneous travel time tomography for updating both velocity and reflector geometry in triangular/ tetrahedral cell model. journal of seismology, 22, 559-574. https://doi. org/10.1007/s10950-017-9722-9 cotterill, c. j., phillips, e., james, l., forsberg, c. f., tjelta, t. i., carter, g., & dove, d. (2017). the evolution of the dogger bank, north sea: a complex history of terrestrial, glacial and marine environmental change. quaternary science reviews, 171, 136-153. https://doi.org/10.1016/j. quascirev.2017.07.006 gao, l. (2018). urban strip road landscape sharp feature extraction integrity optimization simulation. computer simulation, 7, 261-264. gao, r. m., yang, w.j., dai, q. (2018). characteristic polynomials of graphical arrangements corresponding to wheel graphs and join graphs. journal of jilin university (science edition), 56, 859-863. han, x., zhang, d., yan, j., zhao, s., & liu, j. (2020). process development of flue gas desulphurization wastewater treatment in coal-fired power plants towards zero liquid discharge: energetic, economic and environmental analyses. journal of cleaner production, 261, 121144. https://doi.org/10.1016/j.jclepro.2020.121144 li, b., luo, z. x., chen, j. y., cai, h. q., & pang, x. (2019). application research of uav photogrammetry in geological disaster monitoring of hydropower station reservoir area. automation & instrumentation, 5, 209-211+227. liu, x. l., chang, j. c., & liu, r. n. (2017). synthesis method of initial noise samples of oil pump cavitation fault based on characteristic frequency analysis. automation & instrumentation, 12, 62-164. montanari, a., farley, k., claeys, p., de vleeschouwer, d., de winter, n., vansteenberge, s., sinnesael, m., & koeberl, c. (2017). stratigraphic record of the asteroidal veritas breakup in the tortonian monte dei corvi section (ancona, italy). geological society of america bulletin, 129, 9-10. https://doi.org/10.1130/b31476.1 olson, h. n., dilles, j. h., kent, a. j. r., & lang, j. r. (2017). geochemistry of the cretaceous kaskanak batholith and genesis of the pebble porphyry cu-au-mo deposit, southwest alaska. american mineralogist, 102(8), 1597-1621. https://doi.org/10.2138/am-2017-6053 pang, r., xu, b., kong, x., & zou, d. (2018). seismic fragility for high cfrds based on deformation and damage index through incremental dynamic analysis. soil dynamics and earthquake engineering, 104, 432-436. https://doi.org/10.1016/j.soildyn.2017.11.017 pang, r., xu, b., zhou, y., zhang, x., & wang, x. (2020). fragility analysis of high cfrds subjected to mainshock-aftershock sequences based on plastic failure. engineering structures, 206, 110152. https://doi.org/10.1016/j. engstruct.2019.110152 peng, y. w., gu, x. x., cheng, w. b., zhao, x. b., wang, g. n., lu, l. t., han, j. m. (2018). metallogenesis of the late palaeozoic axi-tawuerbieke au-pbzn district in the tulasu basin, western tianshan, china: constraints from geological characteristics and isotope geochemistry. geological journal, 53, 3030-3050. ping, p. z., qiu, m. c., & zhen, j. z. (2017). genesis and implications of the late jurassic hailesitai granites in the northern greater khingan range: evidence from zircon u-pb dating and hf isotope. geological magazine, 154, 963-982. sheth, h., vanderkluysen, l., demonterova, e. i., ivanov, a. v., savatenkov, v. m. (2019). geochemistry and 40ar/39ar geochronology of the nandurbar-dhule mafic dyke swarm: dyke-sill-flow correlations and stratigraphic development across the deccan flood basalt province. geological journal, 54, 157– 176. https://doi.org/10.1002/gj.3167 shrestha, s., larson, k. p., guilmette, c. (2017). the p-t-t evolution of the exhumed himalayan metamorphic core in the likhu khola region, east central nepal. journal of metamorphic geology, 35, 663-693. doi:10.1111/ jmg.12250 singh, a. k., nayak, r., khogenkumar, s., subramanyam, k. s. v., thakur, s. s., bikramaditya singh, r. k., & satyanarayanan, m. (2017). genesis and tectonic implications of cumulate pyroxenites and tectonite peridotites from the nagaland-manipur ophiolites, northeast india: constraints from mineralogical and geochemical characteristics. geological journal, 52, 415-436. doi: 10.1002/gj.2769. sun, x., liu, z., hu, y., zhang, l., & zimmermann, r. (2018). perceptual multi-channel visual feature fusion for scene categorization. information sciences, 429, 37-48. https://doi.org/10.1016/j.ins.2017.10.051 wang, m., zhang, d. q., cheng, y. n., & tan, s. k. (2019). assessing performance of porous pavements and bioretention cells for stormwater management in response to probable climatic changes. journal of environmental management, 243, 157-167. wu, r., calner, m., & lehnert, o. (2017). integrated conodont biostratigraphy and carbon isotope chemostratigraphy in the lower-middle ordovician of southern sweden reveals a complete record of the mdice. geological magazine, 154, 334-353. https://doi.org/10.1017/s0016756816000017 xu, b., pang, r., & zhou, y. (2020). verification of stochastic seismic analysis method and seismic performance evaluation based on multi-indices for high cfrds. engineering geology, 264, 105412. https://doi.org/10.1016/j.enggeo.2019.105412 xu, x., & chen, l. (2019). projection of long-term care costs in china, 2020– 2050: based on the bayesian quantile regression method. sustainability, 11(13), 3530. http://dx.doi.org/10.3390/su11133530 zhang, k., wang, q., chao, l., ye, j., li, z., yu, z., yang, t., & ju, q. (2019). ground observation-based analysis of soil moisture spatiotemporal variability across a humid to semi-humid transitional zone in china. journal of hydrology, 574, 903-914. https://doi.org/10.1016/j.jhydrol.2019.04.087 zhu, z. g., he, b. b., & xue, r. (2018). image feature recognition method for dangerous situation in intelligent substation. chinese journal of power sources, 42, 597-600. https://doi.org/10.1007/s10950-017-9722-9 https://doi.org/10.1007/s10950-017-9722-9 https://doi.org/10.1016/j.quascirev.2017.07.006 https://doi.org/10.1016/j.quascirev.2017.07.006 https://doi.org/10.1016/j.jclepro.2020.121144 https://doi.org/10.1016/j.jclepro.2020.121144 https://doi.org/10.1130/b31476.1 https://doi.org/10.2138/am-2017-6053 https://doi.org/10.1016/j.soildyn.2017.11.017 https://doi.org/10.1016/j.soildyn.2017.11.017 https://doi.org/10.1016/j.engstruct.2019.110152 https://doi.org/10.1016/j.engstruct.2019.110152 https://doi.org/10.1002/gj.3167 https://doi.org/10.1016/j.ins.2017.10.051 https://doi.org/10.1017/s0016756816000017 https://doi.org/10.1016/j.enggeo.2019.105412 https://doi.org/10.1016/j.enggeo.2019.105412 http://dx.doi.org/10.3390/su11133530 https://doi.org/10.1016/j.jhydrol.2019.04.087 https://doi.org/10.1016/j.jhydrol.2019.04.087 ole_link1 ole_link2 maestria en geofisica, una nueva opcion en las geociencias luis alberto briceno guarupe profesor asociado, departamento de geociencias facultad de ciencias universidad nacional de colombia briceno g., l.a. (2000): maestria en geofisica, una nueva opci6n en las geociencias. geofis. colomb. 4:79-80. issn 01212974. santa fe de bogota, d.c. colombia. 1. introduccion ei devenir de las ciencias es mas complejo que lineal. las geociencias en colombia, siguiendo esta hip6tesis, han tenido un desarrollo de ires y venires. la 16gica pareciera ser que se deberian tener primero carreras en las diferentes disciplinas 0 profesiones que las necesidades y planes'de desarrollo de' una comunidad 10 requieran, luego especializaciones, <' maestrias y doctorados. sin embargo, en areas inter 0 multidisciplinarias,. como las geociencias no ha side asi, ni a nivel nacionatni internacional, 10 cual tiene sus pros y contras. ei hecho es que existe la necesidad real de solucionar problemas relacionados con la disciplina de la geofisica, hay la necesidad de profundizar en el tema e investigar sobre una gran variedad de aspectos. la universidad nacional de colombia, en su facultad de ciencias y a traves del departamento de geociencias, con la activa participaci6n del grupo de investigaci6n en geofisica, ha asumido una vez mas el compromiso de impulsar la formaci6n e investigaci6n en el area de las geociencias, reorientando y poniendo en funcionamiento el programa de maestria en geofisica, que habia side aprobado en 1985. en esta ocasion, el acuerdo no. 11, del consejo de sede de bogota, del 20 de mayo de 1998, introdujo modificaciones al plan inicial de estudios, que permitieron reiniciar actividades a partir del primer semestre de 1999, con siete estudiantes, que incluian dos qeoloqos, dos fisicos y tres ingenieros (civil, electrico y catastral y geodesta). para el primer semestre del ano 2000, se recibieron doce estudiantes, que incluian esta vez, siete profesionales en geologia y cinco en ingenierias, fisica y rnaternaticas. como se aprecia, la interdisciplinariedad es tipica del programa, 10 cual permite satisfacer los objetivos de observaci6n, adquisici6n, proceso e interpretaci6n de los procesos naturales que son abarcados por la geofisica. la puesta en marcha de la maestria en geofisica ha contado en esta oportunidad con la activa colaboraci6n directa 0 indirecta de personas e instituciones, ajenas a la universidad, que han entendido 10 positivo del esfuerzo y que estan dispuestos a seguir trabajando para que este posgrado sea cada vez rnejor y sea una opci6n concreta, para quienes quieran formarse a nivel avanzado e investigar en esta area de la geociencias. la maestria en geofisica, junto con el posgrado en meteorologia y el recientemente aprobado posgrado en geologia, fortifican el departamento de geociencias, incentivan a seguir trabajando por otros programas de pre y posgrado y son un paso positive en el desarrollo de las geociencias en colombia. 2. el programa maestria en geofisica objetivos: • obtener un mejor entendimiento de la dinarnica, estructura y composici6n de la tierra y aplicar ese conocimiento en la exploraci6n del subsuelo y en la prevenci6n de catastrofes de origen geol6gico. • formar a un nivel avanzado aquellos profesionales con interes y capacidades de profundizar en la geofisica, mediante la utilizacion de rnetodos cientificos y tecnol6gicos de frontera, a traves de programas ligados a las realidades y necesidades de nuestra sociedad. • proveer al pais y en particular a las instituciones que 10 requieran, de personal id6neo en el area de geofisica, capaces de planear y desarrollar programas investigativos que se integren a las politicas de desarrollo econ6mico, tecnol6gico y social, y que a la vez sean interlocutores en programas cientificos internacionales. • incentivar y fortalecer los grupos de investiqacion en las diferentes lineas de la geofisica, para colaborar con el desarrollo integral de la region. g • colaborar con la capacitaci6n cienlifica de profesionales que puedan contribuir en la formaci6n de futuras generaciones de geofisicos a nivel de pre o posgrado. titulo otorgado: magister en geofisica requisitos de admisi6n: la admisi6n al programa sera anual, comenzando actividades en febrero, con inscripciones en septiembreoctubre del ano anterior. se evaluara: • prueba de conocimientos • hoja de vida • entrevista • comprensi6n ingles 30 puntos 30 puntos 30 puntos 10 puntos puntaje minima de admisi6n 70/100 plan de estudios: • dos asignaturas basicas, que tienen por objeto impartir conocimientos avanzados, fundamentales para cualquier linea de la geofisica, con el objeto de ampliar y profundizar la fundamentaci6n cientifica. • cuatro asignaturas electivas, cuya finalidad es lograr mayor profundizaci6n y diversidad de conocimientos te6ricos y aplicados de temas especializados, referentes a las iineas geofisicas. • tres seminarios de investigaci6n que permiten desarrollar actividades de temas necesarios para el cumplimiento de actividades investigativas. • una tesis de caracter formativo, que permita el planteamiento claro y la soluci6n de un problema de investigaci6n. dependiendo de la formaci6n basica de los participantes, asi como de sus intereses, el trabajo de investigaci6n y tesis se puede orientar en dos iineas: prospecci6n y sismotect6nica. las asignaturas ofrecidas son: basicas: geofisica avanzada; series de tiempo y teoria de inversi6n. electivas: campos potenciales; sismica de reflexi6n; sismologia y ondas sismicas; geologia estructural avanzada; geologia de hidrocarburos avanzada; tect6nica y volcanologia; evaluaci6n geofisica de yacimientos;' geofisica de pozos; sistemas de informaci6n geoqraflca en geociencias. complementarias: geologia especial; maternaticas especiales. duraci6n: cuatro semestres. costos: derechos acadernicos derechos de matricula servicio medico 125 puntos 15 puntos 10 puntos informaci6n complementaria: geoweb@ciencias.ciencias.unal.edu.co geofisic@ciencias.ciencias.unal.edu.co road construction has increased significantly worldwide in the last decades to meet the demands of the increasing human population and this has led to serious soil erosion problems, the bulk of which is unaccounted for, especially in the developing world. for comprehensive land management decisions and monitoring strategies, a review of work that has been done to assess soil erosion due to roads is critical. this article, therefore, reviews the causes of road‒related soil erosion, assessment methods and available control measures. specifically, this work provides an overview of (i) the linkages between roads and soil erosion; (ii) measurement and prediction of road‒related erosion; and (iii) erosion control and rehabilitation techniques. literature shows that road construction results in hill-slope profile modification; removal of vegetation cover; as well as the formation of steep slopes that are prone to severe erosion. furthermore, there is a variety of erosion control measures for controlling road‒related erosion although no study has demonstrated the method that is cost efficient and operational across different landscapes. we are of the view that this study provides guidance in future research on road‒related soil erosion across the developing world where sophisticated monitoring techniques are limited due to resource scarcity for assessing large areas. la construcción de carreteras se ha incrementado ampliamente en todo el mundo durante las últimas décadas para cumplir con las demandas de la creciente población humana, lo que ha llevado a serios problemas de erosión de suelos, muchos de los cuales no se previeron, especialmente, en los países en desarrollo. sobre las decisiones y supervisión de estrategias de un manejo completo del terreno se realizó una revisión al crítico trabajo que se ha hecho para medir la erosión en suelos causados por las carreteras. por esta razón, este artículo revisa las causas de la erosión relacionada con la construcción de rutas y evalúa los métodos y medidas de control disponibles. específicamente, este trabajo ofrece una revisión de (a) las relaciones entre las carreteras y la erosión de los suelos; (b) la medida y la predicción de la erosión vinculada a las carreteras, y (c) las técnicas de control de erosión y rehabilitación. la literatura muestra que la construcción de carreteras produce modificaciones en el perfil inclinación, remueve la vegetación superficial y aumenta la inclinación en pendientes propensas a erosión severa. además, existen varias medidas para controlar la erosión causada por la construcción de carreteras, a pesar de que ningún estudio ha demostrado el método que sea más eficiente y operacional para diferentes paisajes. este estudio guía futuras investigaciones en la erosión causada por la construcción de caminos en los países en desarrollo donde las técnicas de supervisión sofísticas para la evaluación de grandes áreas son limitadas debido a la escasez de recursos. earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 73 80 abstract resumen key words: roadcut and fill embankments, road drainage structures, runoff, soil loss, measurement and prediction, revegetation. palabras clave: diseño de carreteras, llenado de diques, estructuras de desagüe de carreteras, escorrentía, pérdida de suelos, medida y predicción, revegetación. a review of road‒related soil erosion: an assessment of causes, evaluation techniques and available control measures khoboso elizabeth seutloali, heinrich reinhard beckedahl record manuscript received: 04/06/2014 accepted for publication: 30/05/2015 geotechnics issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n1.43841 74 khoboso elizabeth seutloali, heinrich reinhard beckedahl introduction road construction has increased significantly worldwide in the last decades for the provision of effective human mobility and transportation of commodities (bochet et al., 2010). this development has resulted in permanent alteration of the geomorphic and hydrological settings of the landscape leading to increased soil erosion (ramos-scharron and macdonald, 2007). for instance, road construction can result in the modification of natural hill-slope profiles, the construction of roadcut and fill embankments and impervious roadbeds that concentrate runoff (jordan and martinezzavala, 2008). roads concentrate runoff, critical for enhancing increased hill-slope soil loss and sediment yield which later impairs the quality of surrounding open waterbodies (forsyth et al., 2006; lane and sheridan, 2002; ramos-scharron and macdonald, 2007; sheridan and noske, 2007). for instance, lane and sheridan (2002) in their study observed a water quality deterioration as shown by increased turbidity and total dissolved solids downstream of a road stream crossing. the major sediment source at the road stream crossing was the result of erosion on the road verge and the road fill slopes. environmental challenges caused by the accelerated soil erosion due to roads have economic ramifications related to soil rehabilitation and water treatment. it is therefore, a necessity to provide an overview of literature on road-related soil erosion for a better understanding of the causes and methods of assessment that have been considered so as to (1) guide future development; and (2) provide the necessary guidance and informed recommendations on possible efficient and cheap monitoring approaches and erosion control efforts especially in resource-scarce environments. this review therefore seeks to provide an overview of: (i) the effects of armoured roads on soil erosion by water, (ii) related structural designs that facilitate soil erosion processes, and (iii) available approaches for assessing roadrelated soil erosion and the available erosion control techniques. so far, to the best of our knowledge, no studies have been done to assess soil erosion related to paved roads. for instance, previous studies on road‒ related erosion have been dominated by the work on forest roads (i.e. unpaved roads) which include those by burroughs and king (1989) who addressed the potential for reduction of onsite sediment production by different treatments on various components of the forest road prism. croke and hairsine (2006) reviewed the interaction of forest road and track network with both sediment and runoff delivery in managed forests. the study by macdonald and coe (2008) discussed the underlying processes of forest roads sediment production from surface erosion and land sliding. although baird et al. (2012) also reviewed forest road erosion, their focus was on the processes of erosion and sediment delivery from these roads, whereas the other studies either considered land sliding or the process of runoff from the forest road network only. the limitation of the reviews mentioned above is that none addressed the post-construction case of armoured roads except focusing on erosion from unpaved forest roads. furthermore, none of the studies conceptualized assessment of road-related erosion, as well as its control. road-related soil erosion road construction creates numerous roadcut and fill embankments, as well as ditch relief or culvert sites (figure 1) that contribute to runoff and high sediment production that cause extreme land degradation (ramos-scharron and macdonald, 2007). roadcut and fill embankments have bare and steep gradients that cause the generation of runoff and sediment yield (bochet and garcía fayos, 2004). lack of vegetation cover also intensifies soil detachment by raindrops and proliferates susceptibility to erosion as a result of reduced cohesion and shear strength of the soil (jankauskas et al., 2008). similarly, steep gradients increase erosion on these slopes due to reduced water infiltration and increased runoff accumulation (arnaez et al., 2004; cerdà, 2007). numerous studies have documented soil erosion on roadcut and fill embankments (arnaez et al., 2004; jordan and martinez-zavala, 2008; megahan et al., 2001; xu et al., 2009). for example, a study by arnaez et al. (2004) recorded a significant generation of runoff and sediment from roadcut embankments and fill slopes in the iberian range, spain. roadcut embankment soil loss rates exceeded those from the fill slopes by 16 times, and this was attributed to the steep gradients, presence of embedded gravels and low vegetation cover. similarly, jordan and martinez-zavala (2008) recorded a total soil loss of 106 g m-2 and 17 g m-2 from roadcut and side-cast fills respectively in southern spain. the highest erosion rate was observed on the roadcuts due to steep slopes, low vegetation cover and the presence of loose colluvium. moreover, megahan et al. (2001) evaluated the effects of slope gradient, slope length, slope aspect, rainfall erosivity and ground cover density on erosion on the roadcuts in idaho, usa. the multiple regression analysis showed that slope gradient was the most significant of all site variables in affecting roadcut erosion. xu et al. (2009) on the other hand investigated the effects of rainfall and slope length on runoff and soil loss on the qinghai-tibet highway side-slopes in china and found that rainfall intensity correlated with sediment concentration and soil loss, while soil loss decreased with increasing slope length. in summary these studies highlight that slope properties (viz. slope gradient and length, vegetation cover and soil properties, particularly soil texture) of the roadside embankments are critical in determining the degree of soil erosion along these areas. roads initiate soil erosion through drainage structures diverting water from their impervious surfaces as well as from roadcut embankments. road surfaces (including unpaved roads) increase runoff generation (ziegler and giambelluca, 1997). furthermore, the road surfaces transect the hillslope hydrology, creating the need for draining the roadcut embankment and road surface through culverts at regular intervals (as indicated by point 1, in figure 1), with the consequential change from diffuse surface flow downslope to concentrated flow. extensive surface erosion may occur where this concentrated flow is discharged down-slope at discharge points (point 2 and 3 in figure 1). geomorphic impacts of concentrated runoff from road drainage have been documented by numerous studies (jungerius et al., 2002; kakembo, 2000; montgomery, 1994; beckedahl and de villiers, 2000). montgomery (1994) conducted a field survey of road drainage concentration in the western united states and observed that the discharge of road surface concentrated runoff and of intercepted subsurface flow result in initiation and enlargement of a gully and slope instability below the drainage outfall. gully initiation was related to ground slope and contributing area thresholds. kakembo (2000) reported a case of ephemeral stream incision triggered by runoff concentration through a series of railway culverts on a steep hillslope at kwezana, eastern cape, south africa and concluded that concentrated runoff coupled with the steep slope of the drainage discharge area and the rainstorms of high magnitude influenced gully initiation. although not a case study of roads, the scenario is similar in this instance too, the slope hydrology is disrupted and concentration of runoff initiated gullies and triggered hillslope instability. jungerius et al. (2002) reported gully formations where concentrated surface water was diverted to the verges alongside the road in west pokot, kenya. the study found that gully formation is influenced by the steep slopes, lack of vegetation cover, torrential rainfall and the fine grained soils of the alluvial fans. beckedahl and de villiers (2000) investigated the causal relationship between road drainage and pipe erosion in the eastern cape province, south africa. their findings showed that soil pipes and gullies developed where road drainage resulted in a high concentration of surface water on sensitive or dispersive soils. these studies have shown that erosion initiation at road drainage discharge sites is influenced by the contributing area, slope steepness, rainfall intensity and soil properties. the studies by kakembo (2000) and montgomery (1994) however, did not include the estimation of the quantity of soil loss in their agenda. investigations of the impact of concentrated road runoff on soil erosion, to be complete and comprehensive, should also consider an estimation of the amount of soil loss rather than simply focussing only on the contributing factors. these estimations are necessary as they could provide a clear and detailed evidence of the effects of concentrated road runoff discharge on the actual soil loss. after analysis of the possible impacts of road construction on soil erosion, it is important to highlight the methods that can be used to investigate road‒ related erosion. this knowledge will help for accurate assessment of erosion levels and soil loss along the road networks. 75a review of road‒related soil erosion: an assessment of causes, evaluation techniques and available control measures figure 1. a typical cut and fill road cross section and features. the numbers one (1) to four (4) refer to potential impacts, and these are discussed in the text. adapted from (fu et al., 2010). methods of assessing road‒related soil erosion road‒related soil erosion field measurement techniques available field methods of measuring road-related erosion have been principally based on rainfall simulation and volumetric surveys of erosion features. the choice of a particular technique primarily depends on the part of the road component that will be monitored (table 1). for instance, rainfall simulation method has been widely used to explore runoff and soil loss processes related to roadcut and fill slopes, as well as unpaved road surfaces in many parts of the world (table 1). rainfall simulators create controlled rainfall events (jordan and martinez-zavala, 2008), and their design depends on the type of experiments to be carried out (clarke and walsh, 2007). control of rainfall allows determination of the relationship between soil loss and rainfall parameters (lascelles et al., 2000) as well as generation of runoff and soil loss under repeatable conditions (hamed et al., 2002). moreover, in semi-arid regions, with high rainfall variability and recurrent droughts, rainfall simulation could be useful (cerdà, 2007). however, rainfall simulation is uncertain for extrapolating results to larger scale (arnaez et al., 2004) and also underestimates soil loss as compared to natural rainfall as it supplies a constant rainfall intensity (boix-fayos et al., 2006) and short duration rainfall (jin et al., 2008). nevertheless, simulation results remain useful for comparative purposes (foster et al., 2000; jordan and martinez-zavala, 2008) and for planning, despite challenges of underestimating soil loss and limitation to small scale applications. on the other hand, the volumetric survey of erosion features for assessing road-related soil erosion involves the use of measured dimensions (viz. lengths, widths and depths) of the erosion features either directly in the field or from the use of photographic images to estimate soil loss (okoba and sterk, 2006). these dimensions are then utilized to calculate the volume of the erosion features excavated, which is equivalent to the volume of soil lost (hagmann, 1996). although actual soil loss is underestimated since inter-rill erosion is excluded when measuring pipe, gully, and rill erosion, the approach produces the best approximation of erosion (bewket and sterk, 2003). a number of studies have been carried out using the volumetric survey of erosion features to estimate soil erosion on roadcut and fill embankments and most of these have focused on measurement of erosion related to concentrated runoff from road culverts (table 1). other studies used an erosion index for rill and gully erosion to determine its severity on motorway slopes, such as that of bochet and garcía‐fayos (2004), in valence, spain. the erosion index is based on the percentage cover of erosion on the sampling area. however, unlike other studies based on quantitative estimation of erosion, this semi-quantitative approach did not reveal the effect of aspect on erosion intensity and this was attributed to the fact that this method might not be precise enough to detect such differences. although field methods provide the necessary understanding of erosion processes, the obtained results are, however, difficult to generalize due to the complex interaction of erosion processes and field conditions (ande et al., 2009). prediction of road-related erosion could, therefore, help consider the complex interactions that affect erosion rate. modeling of road-related soil erosion soil erosion models vary from simplified procedures, such as the universal soil loss equation (usle) to more complex methods requiring a series of input parameters, such as water erosion prediction project (wepp) (oliveira et al., 2012). usle, and its modifications, the revised universal soil loss equation (rusle) computes the average annual soil loss caused by rill and inter-rill erosion by multiplying the natural factors (rainfall erosivity-r, erodibility-k, slope length and steepness-ls) and anthropogenic factors (cover and management-c, and conservation practices-p) (angima et al., 2003; oliveira et al., 2012). literature has shown that usle/rusle approaches give better estimates of erosion on an overall basis. for instance, oliveira et al. (2012) stated that the usle/rusle provides a right approach for soil loss prediction since it is applicable in terms of required input data, and the obtained soil loss estimates are reliable. however, the use of this model is based upon erosion rates from landscapes larger than road plots hence application for roads is at a smaller scale than for which it was intended (riedel, 2003). in contrast to the usle/rusle, the wepp model was developed to provide a spatial and temporal distribution of soil loss (baird et al., 2012; clinton and vose, 2003). this model utilizes climate, infiltration, water balance, soil chemistry, plant growth and residue decomposition, tillage and consolidation to predict soil erosion deposition and sediment delivery (baird et al., 2012; clinton and vose, 2003). wepp model is applied to roads by including multiple road features such as road surface, cut-slope, ditch, fill slope and lower hillslope (fu et al., 2010; cheng et al., 2013; elliot et al., 1995; forsyth et al., 2006). the road features are modeled separately by defining them as different overland flow elements with unique soil and vegetation parameters assigned (fu et al., 2010). although some models exist for predicting road-related erosion, these are primarily used to predict erosion from the road surfaces (forsyth et al., 2006; sheridan et al., 2006) and few studies have focused on modelling erosion on roadside slopes and erosion due to road drainage ditches/culverts (elliot and tysdal, 1999; megahan et al., 2001) (see table 2). erosion models, however, suffer from a range of problems (barrett et al., 1998). firstly, the model development was often based on data derived from the united states or european conditions and the application of these models to different climatic and management conditions in other regions has not yet been fully established. secondly, the models were created for field plot scale, and application for large scales is still questionable. thirdly, the model predictions are not entirely accurate as a result of incomplete knowledge of the entire set of aspects and interaction processes arising from a limited set of variables. for instance, the disturbance associated with construction frequently exposes the subsoil (or new soil may be brought in from elsewhere) hence the erodibility values along the road will differ to those of the region (barrett et al., 1998). therefore, for road applications, these models still require further testing, and modifications to include additional factors specially designed for road erosion (fu et al., 2010). measurement of soil erosion using the volumetric survey of erosion features, therefore, could provide a reasonable estimation of erosion (sidle et al., 2004) and does not involve expensive instrumentation, long lead times and/or sophisticated modeling (bewket and sterk, 2003). 76 khoboso elizabeth seutloali, heinrich reinhard beckedahl table 1: overview of the techniques of field measurement of road-related erosion used to date table 2: mathematical models used for predicting road-related erosion 77a review of road‒related soil erosion: an assessment of causes, evaluation techniques and available control measures methods used to control road-related soil erosion soil erosion control measures, i.e. non-engineering and bioengineering (e.g., vegetation, soil erosion control blankets, silt fences and geotextiles) and engineering techniques (e.g., diversion drains and lattice) are formulated to reduce accelerated soil erosion rates on roadside slopes (rickson, 2006; xu et al., 2006). this is because roadside slopes have been demonstrated as major contributors towards road-related soil erosion, accounting for 70 to 90% of the total soil loss from the disturbed roadway area (grace iii, 2000). most of the erosion control measures are specifically designed to minimise the contact of rainfall with the soil as well as reduce runoff velocity (de oña et al., 2009). while these soil erosion control methods are effective in minimising road-related soil erosion, however, some of these methods are failing to meet their intended objectives while others are even expensive to use especially in resourcescarce environments. amongst all these control methods, vegetation cover is probably the most widely used measure for controlling erosion on roadside slopes (xu et al., 2006). this is because vegetation cover intercepts rainfall and increases water infiltration (claridge and mirza, 1981; faucette et al., 2006), stabilizes the soil with roots that hold soil particles together (bochet and garcíafayos, 2004; collison and anderson, 1996), and moderates and dissipates the energy exerted by water (lal, 2001; ande et al., 2009). grace iii (2000) and xu et al. (2006) emphasised the importance of vegetation cover in reducing soil erosion and their findings are also supported by the inserts (figure 2) that indicate the importance of vegetation cover on roadside slopes. grace iii (2000) observed a reduction in sediment yield by over 30% on vegetated roadcut and fill slopes compared to the bare roadside slopes and concluded that vegetation has the greatest potential to mitigate soil erosion through stabilizing the roadside slopes. similarly, xu et al. (2006) found that vegetation provided a longterm soil erosion control on roadside slopes and concluded that soil erosion is significantly reduced when vegetation cover is well established. the effectiveness of vegetation cover to control erosion, however, starts when the vegetation is established (rickson, 2006) and mature (vishnudas et al., 2006). for instance, vetiver grass (vetiveria zizanioides l. nash) application significantly controls soil erosion and stabilizes the slopes, although it may take at least one year to become fully effective (sanguankaeo et al., guangzhou, china). this implies that a site may be susceptible to erosion during the period when there is no vegetation or immature stage, also making the establishment of vegetation difficult, since there is no immediate and adequate protection (vishnudas et al., 2006). additionally, the absence of initial binding material in the slope soils may result in poor vegetation growth (bhattacharyya et al., 2008). for these reasons, soil erosion control blankets and geotextiles are short-term vegetation cover replacement that have been used to offer immediate soil protection (smets et al., 2009). erosion control blankets reduce runoff and soil erosion by improving soil quality (bhattarai et al., 2011) and enhancing vegetation (faucette et al., 2006) that would offer a permanent erosion control. likewise, geotextiles control rain splash and runoff (bhattacharyya et al., 2010) and promote a micro-climate for subsequent vegetation growth (sutherland and ziegler, 2006). geotextiles are applied on bare slopes after spreading seed mixture for long-term erosion protection (sutherland and ziegler, 2007). erosion control geotextiles are made from natural or synthetic material (smets et al., 2009) with synthetic geotextiles dominating the commercial market (jankauskas et al., 2008). synthetic geotextiles such as silt fences are used for highway and other construction projects to provide a temporary sediment control (barrett et al., 1998). silt fences reduce runoff velocity and filters sediments thereby enhancing sedimentation (barrett et al., 1998). silt fences are preferred because they are cheap and easy to install (robichaud et al., 2001; wachal et al., 2009). the limitations of synthetic geotextiles, however, are that they are non-degradable and may cause soil pollution, and their production may cause air and water pollution (bhattacharyya et al., 2010). according to jankauskas et al. (2008), however, natural geotextiles constructed from organic materials are more efficient in controlling soil erosion since they adhere to the surface’s microtopography and can follow slope contours and stay in close contact with the soil (bhattacharyya et al., 2010). additionally, natural geotextiles are easily available in many parts of the world, less costly to produce, apply and are environmentally friendly as they are made of biodegradable material (bhattacharyya et al., 2008). some previous studies have evaluated the effectiveness of erosion control blankets and geotextiles in reducing erosion on roadside slopes and found that they reduce soil loss as a result of improvement in vegetation growth (bakr et al., 2012; de oña and osorio, 2006; jankauskas et al., 2008; pengcheng et al., 2008). bakr et al. (2012) examined the influence of compost/mulch on stormwater runoff rates on highway embankments in louisiana. they found out that compost/mulch was effective for soil erosion control since it increased crop cover and reduced soil loss. others such as pengcheng et al. (2008) evaluated the application of sewage sludge compost on highway embankments in china and observed an improvement of soil quality parameters, increased growth of ryegrass and a reduction in volume of runoff and soil loss. similarly, osorio and de ona (2006) observed that compost application on road embankments in southern spain increases vegetation cover and reduces soil loss. additionally, it was found that soil loss decreased with the addition of greater quantities of compost. jankauskas et al. (2008) investigated the use of palm-leaf geotextiles to control erosion on roadside slopes in lithuania. they found that soil erosion from bare fallow soil was reduced by 91.15 – 94.8%, and this was attributed to the multiple benefits such as soil conservation and improved soil moisture that encouraged better plant growth. on the other hand, engineering soil erosion control techniques (e.g. diversion drains and lattice structures) like non-engineering methods, also reduce erosion on roadside slopes by diverting runoff away from the surface of the roadside slope (claridge and mirza, 1981) and intercepting runoff (xu et al., 2006), respectively. these techniques, however, do not provide a protective layer on the surface of the roadside slope; hence soil detachment from direct rainfall impact could still occur. the combination of engineering and vegetation measures could, therefore, provide an effective method for reducing runoff and direct rainfall impact thereby reducing soil loss on roadside slopes (xu et al., 2006). on the basis of the above discussion, the most efficient and economic soil erosion control strategy is re-vegetation. this is because vegetation cover provides a cheap long-term erosion control (benik et al., 2003), requires less maintenance than complex engineering structures (montoro et al., 2000) and improves the landscape aesthetic value (albaladejo montoro et al., 2000). hence, soil erosion control through the establishment of a dense vegetation cover is a priority for restoration of roadside slopes (garcía-palacios et al., 2010). for instance, figure 2a illustrates roadside slopes that have successfully stabilized due to the use of vegetation cover as a control mechanism. on the other hand, it can be observed in figure 2b that areas without vegetation cover are prone to erosion. while the use of soil erosion control techniques has been widely recognised and investigated, these investigations have, in most cases, focused on the non-engineering and bio-engineering techniques, and less attention has been given to engineering measures although they could provide an efficient erosion control on roadside slopes (xu et al., 2006). therefore, there is a need to test the effectiveness of 78 khoboso elizabeth seutloali, heinrich reinhard beckedahl engineering measures for erosion control on roadside slopes. figure 2. (a) successful application of vegetation cover to control erosion on a roadside slope and (b) signs of erosion on a roadside slope due to the absence of vegetation cover. conclusion roads and road construction result in soil erosion due to the impacts of rainfall affecting geomorphic and hydrologic processes. research has shown that the creation of roadcut and fill embankments with steep slopes and little vegetation cover, as well as the concentration of runoff from the road surface and intercepted subsurface flows influence the hydrologic and geomorphic processes. roadcut embankments, however, are the major sources of erosion than other parts of the road with slope gradient being the most important factor influencing soil erosion. a variety of techniques are used to investigate roadrelated erosion, ranging from field measurements to soil erosion prediction models. these methods could assist in understanding the nature and severity of road-related erosion and can help guide future development and erosion control efforts. however, besides the strengths of erosion measurement methods, soil erosion prediction models, although appropriate for predicting soil loss for the field plot scale, have challenges when applied to small land parcels. therefore, there is a need for further testing and modification of soil erosion prediction models for road application. it has been shown in the literature that soil erosion control techniques have the potential to reduce runoff and soil loss. numerous studies that have investigated the effectiveness of soil erosion control techniques utilised on roadside embankments showed that the most effective methods are those that promote revegetation and reduce both velocity and quantity of runoff. since the extent of road networks is ever-increasing, lessons learned from this research may be applied in the future construction of road systems. as such, research still needs to be done (i) to fully understand the underlying determinants of soil erosion related to road design and construction to limit the effect of embankments; (ii) to quantify road-related soil loss; (iii) to evaluate the effectiveness of erosion control methods on both roadcut and fill embankments; and (iv) to identify new approaches such as remote sensing technologies, to try to improve soil erosion mapping along roads for future monitoring and management strategies. this review therefore provides the necessary insight and inspiration to geomorphologists, road engineers and environmentalists to move towards identifying the most suitable, cheap and readily available techniques for assessing and controlling soil erosion, necessary for reliable and informed approaches for monitoring and managing road-related soil erosion across the world, especially in under resourced-countries. acknowledgements authors would like to thank the university of kwazulu-natal for funding this research. references albaladejo montoro j., alvarez rogel j., querejeta j., diaz e., castillo v. (2000). three hydro-seeding revegetation techniques for soil erosion control on anthropic steep slopes, land degradation & development. 11, 4, 315-325. ande o., alaga y., oluwatosin g. (2009). soil erosion prediction using mmf model on highly dissected hilly terrain of ekiti environs in southwestern nigeria, international journal of physical sciences. 4, 2, 53-57. angima s. d., scott d. e., o’neill m. k., ong c. k., weesies g. a. (2003). soil erosion prediction using rusle for central kenyan highland conditions, agriculture, ecosystems and environment. 97, 295 308. arnaez j., larrea v., ortigosa l. (2004). surface runoff and soil erosion on unpaved forest roads from rainfall simulation tests in northeastern spain, catena. 57, 1, 1-14. baird e. j., floyd w., van meerveld i., anderson a. e. (2012). road surface erosion, part 1: summary of effects, processes, and assessment procedures, watershed management bulletin. 15, 1, 1-9. bakr n., weindorf d. c., zhu y., arceneaux a. e., selim h. (2012). evaluation of compost/mulch as highway embankment erosion control in louisiana at the plot-scale, journal of hydrology. 468, 257-267. barrett m. e., malina j. f., charbeneau r. j. (1998). an evaluation of geotextiles for temporary sediment control, water environment research. 70, 3, 283-290. beckedahl h. r., de villiers a. b. (2000). accelerating erosion by piping in the eastern cape province, south africa, south african geographical journal. 82, 157-162. benik s., wilson b., biesboer d., hansen b., stenlund d. (2003). evaluation of erosion control products using natural rainfall events, journal of soil and water conservation. 58, 2, 98-105. bewket w., sterk g. (2003). assessment of soil erosion in cultivated fields using a survey methodology for rills in the chemoga watershed, ethiopia, agriculture, ecosystems & environment. 97, 1, 81-93. bhattacharyya r., davies k., fullen m. a., booth c. a. (2008). effects of palm-mat geotextiles on the conservation of loamy sand soils in east shropshire, uk, advances in geoecology. 3, 9, 527-538. bhattacharyya r., smets t., fullen m. a., poesen j., booth c. a. (2010). effectiveness of geotextiles in reducing runoff and soil loss: a synthesis, catena. 81, 3, 184-195. 79a review of road‒related soil erosion: an assessment of causes, evaluation techniques and available control measures bhattarai r., kalita p. k., yatsu s., howard h. r., svendsen n. g. (2011). evaluation of compost blankets for erosion control from disturbed lands, journal of environmental management. 92, 3, 803-812. bochet e., garcía-fayos p. (2004). factors controlling vegetation establishment and water erosion on motorway slopes in valencia, spain, restoration ecology. 12, 2, 166-174. bochet e., garcia fayos p., tormo j. (2010). how can we control erosion of roadslopes in semiarid mediterranean areas soil improvement and native plant establishment, land degradation & development. 21, 110 121. boix-fayos c., martínez-mena m., arnau-rosalén e., calvo-cases a., castillo v., albaladejo j. (2006). measuring soil erosion by field plots: understanding the sources of variation, earth-science reviews. 78, 3, 267-285. burroughs e. r., king j. g. (1989). reduction of soil erosion on forest roads. cerdà a. (2007). soil water erosion on road embankments in eastern spain, science of the total environment. 378, 1, 151-155. cheng b., lv y., zhan y., su d., cao s. (2013). constructing china’s roads as works of art: a case study of “esthetic greenway” construction in the shennongjia region of china, land degradation & development. 1-7. claridge f. b., mirza a. m. (1981). erosion control along transportation routes in northern climates, arctic. 34, 2, 14z7-157. clarke m. a., walsh r. p. (2007). a portable rainfall simulator for field assessment of splash and slopewash in remote locations, earth surface processes and landforms. 32, 13, 2052-2069. clinton b. d., vose j. m. (2003). differences in surface water quality draining four road surface types in the southern appalachians, southern journal of applied forestry. 27, 2, 100-106. collison a., anderson m. (1996). using a combined slope hydrology/ stability model to identify suitable conditions for landslide prevention by vegetation in the humid tropics, earth surface processes and landforms. 21, 8, 737-747. croke j. c., hairsine p. b. (2006). sediment delivery in managed forests: a review, environmental reviews. 14, 1, 59 87. de oña j., osorio f. (2006). application of sludge from urban wastewater treatment plants in road’s embankments, journal of hazardous materials. 131, 1, 37-45. de oña j., osorio f., garcia p. a. (2009). assessing the effects of using compost–sludge mixtures to reduce erosion in road embankments, journal of hazardous materials. 164, 2, 1257-1265. elliot w., foltz r., luce c. (1995). validation of water erosion prediction project (wepp) model for low-volume forest roads. sixth international conference on low-volume roads, washington d.c, united states, june, 178-186. elliot w. j., tysdal l. m. (1999). understanding and reducing erosion from insloping roads, journal of foresftry. 97, 8, 30-34. faucette l. b., risse l. m., jordan c. f., cabrera m. l., coleman d. c., west l. t. (2006). vegetation and soil quality effects from hydroseed and compost blankets used for erosion control in construction activities, journal of soil and water conservation. 61, 6, 355-362. forsyth a., bubb k., cox m. (2006). runoff, sediment loss and water quality from forest roads in a southeast queensland coastal plain pinus plantation, forest ecology and management. 221, 1, 194-206. foster i., fullen m., brandsma r., chapman a. (2000). drip-screen rainfall simulators for hydro-and pedo-geomorphological research: the coventry experience, earth surface processes and landforms. 25, 7, 691-707. fu b., newham l. t., ramos-scharron c. (2010). a review of surface erosion and sediment delivery models for unsealed roads, environmental modelling & software. 25, 1, 1-14. garcía-palacios p., soliveres s., maestre f. t., escudero a., castillomonroy a. p., valladares f. (2010). dominant plant species modulate responses to hydroseeding, irrigation and fertilization during the restoration of semiarid motorway slopes, ecological engineering. 36, 10, 1290-1298. grace iii j. (2000). forest road sideslopes and soil conservation techniques, journal of soil and water conservation. 55, 1, 96-101. grace iii j. m. (2005). application of wepp to a southern appalachian forest road. american society of agricultural engineers, 05-2016. hagmann j. (1996). mechanical soil conservation with contour ridges: cure for, or cause of, rill erosion?, land degradation & development. 7, 2, 145-160. hamed y., albergel j., pépin y., asseline j., nasri s., zante p., berndtsson r., el-niazy m. and balah m. (2002). comparison between rainfall simulator erosion and observed reservoir sedimentation in an erosion-sensitive semiarid catchment, catena. 50, 1, 1-16. jankauskas b., jankauskienė g., fullen m. a., booth c. a. (2008). utilizing palm-leaf geotextiles to control soil erosion on roadside slopes in lithuania, agricultural sciences. 15, 3, 22-28. jin k., cornelis w. m., gabriels d., schiettecatte w., de neve s., lu j., buysse t., wu h., cai d., jin j. (2008). soil management effects on runoff and soil loss from field rainfall simulation, catena. 75, 2, 191-199. jordan a., martinez-zavala l. (2008). soil loss and runoff rates on unpaved forest roads in southern spain after simulated rainfall, forest ecology and management. 255, 3, 913-919. jungerius p., matundura j., van de ancker j. (2002). road construction and gully erosion in west pokot, kenya, earth surface processes and landforms. 27, 11, 1237-1247. kakembo v. (2000). artificial drainage induced erosion: the case of railway culverts on the kwenza ridge, near alice, eastern cape, south african geographical journal. 82, 3, 149 153. lal r. (2001). soil degradation by erosion, land degradation & development. 12, 519 539. lane p. n. j., sheridan g. j. (2002). impact of an unsealed forest road stream crossing: water quality and sediment sources, hydrological processes. 16, 13, 2599 2612. lascelles b., favis-mortlock d. t., parsons a. j., guerra a. j. (2000). spatial and temporal variation in two rainfall simulators: implications for spatially explicit rainfall simulation experiments, earth surface processes and landforms. 25, 7, 709-721. macdonald l. h., coe d. b. r. (2008). road sediment production and delivery: processes and management. proceedings of the first world landslide forum, tokyo, japan, 2008, 381–384. megahan w. f., wilson m., monsen s. b. (2001). sediment production from granitic cutslopes on forest roads in idaho, usa, earth surface processes and landforms. 26, 2, 153-163. montgomery d. r. (1994). road surface drainage, channel initiation, and slope instability, water resources research. 30, 6, 1925 1932. montoro a. j., rogel a. j., querejeta j., diaz e., castillo v. (2000). three hydroseeding revegetation techniques for soil erosion control on anthropic steep slopes, land degradation & development. 11, 4, 315-325. moodley m., hill t. r., beckedahl h. r. (2011). influence of unpaved forestry access roads on surface runoff, sediment loss and soil water. water research commission, k5/1807/4. okoba b. o., sterk g. (2006). quantification of visual soil erosion indicators in gikuuri catchment in the central highlands of kenya, geoderma. 134, 1, 34-47. oliveira p. t. s., wendland e., nearing m. a. (2012). rainfall erosivity in brazil: a review, catena. 100, 139-147. osorio f., de ona j. (2006). using compost from urban solid waste to prevent erosion in road embankments, journal of environmental science and health 41, 2311 2327. pengcheng g., xinbao t., yanan t., yingxu c. (2008). application of sewage sludge compost on highway embankments, waste management. 28, 9, 1630-1636. ramos-scharron c. e., macdonald l. h. (2007). runoff and suspended sediment yields from unpaved road segment, st john, us virgin islands, hydrological processes. 21, 1, 35 50. rickson r. j. (2006). controlling sediment at source: an evaluation of erosion control geotextiles, earth surface processes and landforms. 31, 5, 550 560. 80 khoboso elizabeth seutloali, heinrich reinhard beckedahl riedel m. s. (2003). collaborative research and watershed management for optimization of forest road best management practices. proceedings of the 2003 international conference on ecology and transportation, raleigh, us 2003, 148-158. robichaud p., mccool d., pannkuk c., brown r., mutch p. (2001). trap efficiency of silt fences used in hillslope erosion studies. proceedings of the international symposium, soil erosion research for the 21st century, honolulu, united states of america, january 541-543. sanguankaeo s., chaisintarakul s., veerapunth e. the application of the vetiver system in erosion control and stabilization for highways construction and maintenance in thailand. in proceedings of the third international conference on vetiver and exhibition 1-6. proceedings of the third international conference on vetiver and exhibition 2003. selkirk j., riley s. (1996). erodibility of road batters under simulated rainfall, hydrological sciences journal. 41, 3, 363-376. sheridan g. j., noske p. j. (2007). catchment-scale contribution of forest roads to stream exports of sediment, phosphorus and nitrogen, hydrological processes. 21, 23, 3107 3122. sheridan g. j., noske p. j., lane p. n., sherwin c. b. (2008). using rainfall simulation and site measurements to predict annual interrill erodibility and phosphorus generation rates from unsealed forest roads: validation against in-situ erosion measurements, catena. 73, 1, 49-62. sheridan g. j., noske p. j., whipp r. k., wijesinghe n. (2006). the effect of truck traffic and road water content on sediment delivery from unpaved forest roads, hydrological processes. 20, 8, 1683-1699. sidle r. c., furuichi t., kono y. (2011). unprecedented rates of landslide and surface erosion along a newly constructed road in yunnan, china, natural hazards. 57, 2, 313-326. sidle r. c., sasaki s., otsuki m., noguchi s., rahim nik a. (2004). sediment pathways in a tropical forest: effects of logging roads and skid trails, hydrological processes. 18, 4, 703-720. smets t., poesen j., langhans c., knapen a., fullen m. a. (2009). concentrated flow erosion rates reduced through biological geotextiles, earth surface processes and landforms. 34, 4, 493-502. sutherland r. a., ziegler a. d. (2006). hillslope runoff and erosion as affected by rolled erosion control systems: a field study, hydrological processes. 20, 13, 2839-2855. sutherland r. s., ziegler a. d. (2007). effectiveness of coir-based rolled erosion control systems in reducing sediment transport from hillslopes, applied geography. 27, 150 -164. vishnudas s., savenije h., van der zaag p., anil k., balan k. (2006). the protective and attractive covering of a vegetated embankment using coir geotextiles, hydrology and earth system sciences discussions. 10, 4, 565-574. wachal d. j., banks k. e., hudak p. f., harmel r. d. (2009). modeling erosion and sediment control practices with rusle 2.0: a management approach for natural gas well sites in denton county, tx, usa, environmental geology. 56, 8, 1615-1627. xu l., liu w., kong y., zhang k., yu b., chen j. (2009). runoff and water erosion on road sideslopes: effects of rainfall characteristics and slope length, transportation research. 14, 7, 497 501. xu x., zhang k., kong y., chen j., yu b. (2006). effectiveness of erosion control measures along the qinghai–tibet highway, tibetan plateau, china, transportation research. 11, 4, 302-309. ziegler a. d., giambelluca t. w. (1997). importance of rural roads as source areas for runoff in mountainous areas of northern thailand, journal of hydrology. 196, 1-4, 204-229. swelling behavior of clays is of great importance for numerous engineering applications due to the prevalence of expansive clays around the world. expansive clays are present in mexico city, united states, australia, africa and south america, among others. in some cases, these types of clays could present swell pressure values greater than 400 kpa. in this technical note, correlation equations are developed to estimate the swell pressure of clays using laboratory tests performed on swelling clays in the city of barranquilla, colombia. correlations are based on atterberg limits and water content among other soil properties. equations with statistically significant coefficients were selected and compared with equations found in the literature. developed correlations evidenced that swell pressure decays quickly as water content increases. it was found that for the studied soils, atterberg limits did not present statistical significance in the estimation of swell pressure. abstract keywords: expansive clays, swell pressure, atterberg limits, water content. empirical correlations for the swelling pressure of expansive clays in the city of barranquilla, colombia correlaciones empíricas para la determinación de la presión de expansión de arcillas expansivas en barranquilla, colombia issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n1.60226 earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017): 45 49 víctor cantillo, vicente mercado, césar pájaro* universidad del norte, colombia *capajaro@uninorte.edu.co el comportamiento expansivo de las arcillas presenta vital importancia alrededor del mundo, ya que este tipo de arcillas se encuentras en diversas y extensas zonas del planeta. arcillas con este tipo de comportamiento se encuentran presentes en diversas regiones del mundo como la ciudad de méxico, los estados unidos, australia, algunas zonas de áfrica y sudamérica, entre otras. este tipo de arcillas pueden presentar en algunos casos presiones de expansión superiores a los 400 kpa. en la presente nota técnica se desarrollan correlaciones para estimar la presión de expansión de arcillas con base en ensayos de laboratorio practicados a arcillas con comportamiento expansivo presentes en la ciudad de barranquilla, colombia. las correlaciones desarrolladas se basan en los límites de atterberg y contenido de humedad, entre otras propiedades de los suelos. se seleccionaron aquellas ecuaciones que presentaban coeficientes estadísticamente significativos y se realizaron comparaciones con correlaciones presentes en la literatura. las correlaciones mostraron que la presión de expansión decae rápidamente a medida que el contenido de humedad aumenta. además, se observó que para los suelos estudiados los límites de atterberg no son estadísticamente significativos para la estimación de la presión de expansión. resumen palabras clave: arcillas expansivas, presión de expansión, límites de atterberg, contenido de humedad. record manuscript received: 20/09/2016 accepted for publication: 10/02/2017 how to cite item cantillo, v., mercado v., & pajaro, c. (2017). empirical correlations for the swelling pressure of expansive clays in the city of barranquilla, colombia. earth sciences research journal, 21(1). 45-49. doi:http://dx.doi.org/10.15446/esrj.v21n1.60226 g e o t h e c n ic s 46 víctor cantillo, vicente mercado and césar pájaro 1. introduction the soils of barranquilla, a city of colombia, in south america, are mainly composed of layers of sand and gravel (moreno, 2013). however, the western and southwestern part of the city (figure 1) present extensive clay layers with swelling behavior (geotecnología ltda., 2006). in this part of barranquilla, the overall type of construction is essentially residential housing. for this reason, the demographic density in this sector of the city is high. since the mid-1970s, the western and southwestern zone of barranquilla have presented large landslides and slopes instability problems (figure 2) as a consequence of both anthropic and natural factors, and these phenomena have affected around two thousand lowincome families (guardo, 2000). figure 1. location of the study area. (ingeominas, 2011) some of the geotechnical problems in barranquilla´s western and southwestern zone are due to swell behavior of underlying clays (geotecnología ltda, 2006). swelling of clays affects the structural stability of the overlying infrastructure; structural problems are often accentuated due to the topographical characteristics of the terrain, which is constituted by inclined slopes, leading to the collapse of structures, as illustrated in figure 2. studies developed to assess and understand the swelling behavior of these clays have been carried out by geotecnología ltda. (2006), and by edgar rodríguez (2014). the main goal of the work undertaken by geotecnología ltda. (2006) was to increase and improve the geological information regarding the western and southwestern part of the city. on the other hand, rodríguez (2014) aimed to study the geological structure of clay materials in the zone. the results of rodríguez (2014), geotecnología ltda. (2006), and other researchers revealed that most of the geotechnical problems of the area, are related to the water seepage through the clay layers. geotecnología ltda. (2006) explains these phenomena in detail in their work. they state that the upper soil layers in this zone are constituted by sands and gravels with high permeability, followed by barranquilla’s marine clays. during rainy seasons, the granular soils act as huge water retainers in the interface between these soils and the clays beneath. this accumulated water causes clays to swell and at the same time, it causes the loss of slope stability (geotecnología ltda., 2006). a typical soil profile for the study area is presented in figure 3. another important cause of landslides is that earth fills’ inclination in some cases is greater than the clay’s effective friction angle; for instance, some slopes have inclinations between 16° and 25° (guardo, 2000), while the internal effective angle of friction for these materials ranges between 5° and 12° (rodríguez, 2014). figure 2. consequences of landslides at western and south western of barranquilla (angulo, 2014). figure 3. typical soil profile of the study area. (after geotecnología ltda., 2006) clays in barranquilla´s western and southwestern zone present prominent swelling behavior which is mainly produced by the montmorillonite clay mineral content (feferbaum and beltran, 1974). soils with similar characteristics can be found in different zones around the world, for example in the united states, south-america, africa, australia, china and india (murthy, 2011). swell phenomena in clayey soils have been studied by komornik and david (1969), hensen and smith (2002), day (1991, 1992), among others. several authors presented a variety of correlation equations to determine swell pressure of expansive clays, and examples of these equations are present in erguler and ulusay (2003), sridharan and gurtug (2004) and phanikumar (2006). these investigations define the swell pressure value as a function of water content, atterberg limits, dry unit weight and other parameters proposed by the authors, with laboratory tests designed by them. erguler and ulusay (2003), for instance, suggest the “maximum water content in 24 hours” and the “maximum water content in 72 hours” as predictive parameters, while phanikumar (2006) introduces a parameter called free swell index. there is still a great necessity for civil and geotechnical engineering practice in the city to deepen the understanding of the mechanical properties of the soils in this zone of the city. in an effort to extend the knowledge about the mechanical behavior of clays in this part of barranquilla, this technical note presents correlation equations to estimate swell pressure as a function of water content and atterberg limits. correlations found in this study could be useful in the study of expansive clays around with similar characteristics in different zones around the world. predictive equations presented in this technical note are based on the water contents and atterberg limits of clays in barranquilla’s western and southwestern zone as explanatory variables. these properties were selected 47empirical correlations for the swelling pressure of expansive clays in the city of barranquilla, colombia because they are widely used to develop swell pressure and swell index correlations in the literature, and at the same time, these measurements are perhaps the most prevalent in most geotechnical reports. the associated laboratory tests were performed following the procedures outlined by the technical specifications of invias (colombian national institute of roads): inv e-125-07 (corresponding to astm d 4318-00) to determine liquid limit, inv e-126-07 (corresponding to astm d 4318-00) to determine plastic limit, and inv e-122-07 (corresponding to astm d 2216-98) to determine water content. 2. measurement of swell pressure there are three standardized methodologies which determine swell pressure using laboratory testing (sridharan, 2009): swell – load method (curve 1 on figure 4), swell under load method (curve 2 on figure 4), and constant volume method (curve 3 on figure 4). it should be mentioned that, as illustrated in figure 4, each laboratory methodology to determine swell pressure yields different swell pressure values (thompson et al., 2006; kayabali and demir, 2011). figure 4. typical stress-strain curve in laboratory methodologies to determine swell pressure. (after sridharan et al., 2009) in this study, the constant volume method was used to determine swell pressure values implemented to develop the correlations equations. the constant volume method, usually called strain controlled method, can be conducted by taking a specimen in a consolidation ring and applying a pressure equal to the result of the addition of the effective overburden pressure and the approximate anticipated surcharge caused by the foundation. water is then added to the specimen to start swelling. as the specimen starts to swell, a small pressure increment is applied to prevent swelling. when the total stress causes specimen consolidation the swell pressure is attained. more details on the implemented constant volume method can be found in das (2015). 3. correlation equations. correlation equations presented in this section, are based on test results of barranquilla´s western and southwestern zone clays. samples were obtained mainly from the area of campo alegre, because this zone is perhaps the most affected in the city as a consequence of landslides and swelling behavior of clays. a total of 38 samples were used in this study. for each sample, water content, liquid limit, plastic limit, and void ratio were measured. table 1 summarizes the maximum and minimum values for each soil property within the collected samples. the complete water content, liquid limit, plastic limit, plasticity index and void ratio information for each of the 38 samples is summarized in appendix 1. table 1. measurement variables ranges. logarithmic models ( y = a + b ln x ), and exponential models (ln y = a + bx) were used to model swell pressure (dependent variable) as a function of the selected properties. only two models presented correlation coefficients and determination coefficients higher than 0.5; furthermore, only water content coefficients and models using water content as unique dependent variable were statistically significant. correlation equations, determination coefficients ( r2 ) and correlation coefficients (r) of the developed models are presented in table 2. table 2. correlation equation. as can be seen in table 2, statistically, the best fit and correlation equation to predict swell pressure is equation 2. table 3 summarizes goodness of fit parameters for the correlation equations. figure 5(a) shows swell pressure-water content pairs measured for each specimen and swell pressure predictions using equations 1 and 2 from table 2. among the different functional forms used to develop correlation equations between swell pressure and other soils properties, only a third equation (equation 3) which incorporated plasticity index (pi) and water content as explanatory variables, presented high determination coefficient and correlation coefficient ( r2=0.86 and r=0.93 ): ln (( sp( kpa )) = 7.77-0.12w (%) + 0.0054 pi (%) (3); however, the plasticity index was not statistically significant using a statistical significance of 0.05, having a p-value of 0.28 using a t-student distribution (table 3). figure 5(b) shows the correlation from equation 3 along the samples data; it results evident that the plasticity index does not have a major impact on swell pressure for the studied soil samples. figure 5. developed correlation models. 48 víctor cantillo, vicente mercado and césar pájaro table 3. goodness of fit parameters. 4. discussion a broader revision of swell pressure correlation equations can be found in al-rawas and goosen (2006). table 4 presents some relevant correlation equations and corresponding correlation coefficients reported by the authors. table 4. correlation equations found in the literature. w is water content; pl is plastic limit; ll is liquid limit; il is liquidity index; γdry is dry unit weight. table 4, shows that for most of the presented correlation equations, correlation coefficient values are relatively high, which means that the equations could adequately predict the trend of the swell pressure measurements. although correlation coefficients are high, it should be emphasized that the equations presented above be only valid in the measurement range of properties used to develop the equations. another important fact is that the equations are only valid for the soils for which they were developed because swell properties vary among different clays. however, all the equations present similar trends among the various properties and swell pressure values. similarly to the trend presented in this work, some of the proposed correlations yield a decrease in the logarithm of the swell pressure as a function of the water content. erguler and ulusay (2003), for instance, present the variation of swell pressure as a function of water content for soils samples collected at three different sites. in their work, swell pressure decays quickly as water content increases, reaching values of nearly 0 for water contents higher than 30%. such behavior is similar to that shown in figure 5 of this study. 5. conclusions this research presents correlation equations developed to estimate swell pressure of clays using laboratory tests performed on swelling clays in the city of barranquilla, colombia. swell pressure values were estimated using the constant volume method. estimated pressures reached values of more than 600 kpa for low levels of water content. swell pressure values decayed quickly as water content increased, reaching values of nearly 0 for water contents higher than 40%. for the tested soils, it was found that swell pressure could be best modeled as an exponential function of the water content. while previous works by other authors use several additional soil properties (e.g., liquid limit, plastic limit, plasticity index or unit weight), it was found that for the studied soils adequate swell pressure estimates were obtained based solely on water content. moreover, unlike previous works, it was found that atterberg limits did not have statistical significance in the estimation of swell pressure for the studied samples. appendix 1. database of material parameters obtained experimentally. 49empirical correlations for the swelling pressure of expansive clays in the city of barranquilla, colombia references. al-rawas, a. a., & goosen, m. f. (2006). expansive soils: recent advances in characterization and treatment. taylor and francis. angulo, e. (2014). defensor exige medidas urgentes para la ladera suroccidental. diario el heraldo. [online] available at: http:// www. elheraldo.co/local/defensor-exige-medidas-urgentes-paralaladera-suroccidental-139845 [accessed 15 mar. 2017]. das, b. (2015). principles of foundation engineering. cengage learning. day, r. w. (1991). expansion of compacted gravelly clay. journal of geotechnical engineering, 117(6), 968-972. day, r. w. (1992). swell versus saturation for compacted clay. journal of geotechnical engineering, 118(8), 1272-1278. erguler, z. a., & ulusay, r. (2003). a simple test and predictive models for assessing swell potential of ankara (turkey) clay. engineering geology, 67(3), 331-352. erzin, y., & erol, o. (2004). correlations for quick prediction of swell pressures. electronic journal of geotechnical engineering, 9, 1. feferbaum, s. & beltrán, l. (1974). arcillas expansivas en colombia. segundo simposio colombiano de geotecnia e ingeniería geológica. sociedad colombiana de geotecnia. geotecnología ltda. (2006). propuesta general definitiva del plan de manejo y obras del plan maestro para la estabilización geotécnica de las laderas del barrio campoalegre en barranquilla. barranquilla: jaime suárez diaz. guardo, j. (2011). la problemática geotécnica del suroccidente de barranquilla. revista científica ingeniería y desarrollo, (8), 119-129. hensen, e. j., & smith, b. (2002). why clays swell. the journal of physical chemistry b, 106(49), 12664-12667. ingeominas (2011). zonificación de amenaza por movimientos en masa de las laderas occidentales de barranquilla, departamento del atlántico. bogotá, grupo de edición y diagramación de información de ingeominas. kayabali, k., & demir, s. (2011). measurement of swelling pressure: direct method versus indirect methods. canadian geotechnical journal, 48(3), 354-364. komornik, a., & david, d. (1969). prediction of swelling pressure of clays. journal of the soil mechanics and foundations division, 95. moreno, n. (2013). zonificación y caracterización geotécnica de los suelos de barranquilla. fase 1. murthy, d. s. (2011). advanced foundation engineering. cbs publishers & distributors, new delhi, bangalore, india. nayak, n. v. (1985). foundation design manual. dhanpat rai and sons pub. phanikumar, b. r. (2006). prediction of swelling characteristics with free swell index. expansive soils: recent advances in characterization and treatment. taylor and francis group, 173-184. rodríguez, e. a. (2014). evaluación del comportamiento geomecánico de arcillas en el sector de campoalegre–ciudad de barranquilla (doctoral dissertation, universidad nacional de colombia). schneider, g. l., & poor, a. r. (1974). the prediction of soil heave and swell pressures developed by an expansive clay. university of texas. sridharan, a. (2009). critical evaluation of determining swelling pressure by swell-load method and constant volume method. sridharan, a., & gurtug, y. (2004). swelling behavior of compacted finegrained soils. engineering geology, 72(1), 9-18. thompson, r. w., perko, h. a., & rethamel, w. d. (2006). comparison of constant volume swell pressure and odometer load-back pressure. in unsaturated soils 2006 (pp. 1787-1798). asce. using neuronal networks to determine site response from coda waves: application in armenia, colombia earth sciences research journal earth sci. res. j. vol 8 nº 1 (dec. 2004): 52 55 manuscript received july 2004 paper accepted octubre 2004 52 ossodas, a portable digital system for seismological signal acquisition pedro riascos1, andres racines2, jhon caicedo 3, jorge mejia4 & hans meyer 5. e-mail address: pepar@osso.org.co observatorio sismológico del sur occidente – osso/uv universidad del valle, facultad de ingeniería corporación osso §www:http://www.osso.org.co e-mail:corposso@osso.org.co tel : 57(2)6821078, 57(2)6827662 fax: 57(2)6821078, 57(2)3313418 cali colombia sudamérica abstract this paper presents the concept and design of a digital system to record seismological signals. the hardware and software designed has been interfaced to the pc 104 bus architecture in order to achieve a modular system to acquire, process, record and transmit seismic data by telemetry. the modularity of this system offers independence technology from the manufacturers of this kind of equipments both in hardware and software, also because the system was developed under the linux operating system gnu, which allows easy update of hardware; the software is an open platform developed in c, which allows the user to configure the system to operate online and updating according to needs. in brief, we propose in this paper a system design option for seismological instrumentation, which presents significant advantages also for other regions and countries with high seismicity and needs of earthquake monitoring, but have serious limitations in acquiring and maintaining commercial systems imported from developed countries. key words: seismology, instruments, ossodas, pc104, linux gnu, spread spectrum resumen el artículo presenta el diseño de un sistema digital para registrar señales sísmicas. el hardware y software diseñado son adicionados a la arquitectura de bus pc/104 con el propósito de tener un sistema modular para adquirir, procesar, almacenar y transmitir datos sísmicos por telemetría. la modularidad de este sistema ofrece independencia tecnológica con los fabricantes de estos equipos tanto en hardware como en software, ya que el sistema fue desarrollado bajo una plataforma linux gnu que permite al usuario configurarlo o modificarlo de acuerdo con sus necesidades en conclusión, proponemos una alternativa de sistema para la instrumentación sismológica, el cual es ventajoso para aquellas regiones con alto grado de sismisidad, necesidad de equipos para instrumentación sismológica, severas limitaciones para adquirir y mantener sistemas importados de países desarrollados. palabras clave: sismología, ossodas, instrumentación, pc104, linux gnu, espectro disperso © 2004 esrj unibiblos. ossodas, a portable digital system for seismological signal acquisition 53 introduction presently there is an increasing number of portable field system for seismic data acquisition available on the market, with high dynamic range, high data recording capacity and with ethernet and radio connectivity; these gave many seismological observatories the opportunity to improve the observation of seismicity, including observatories in colombia, but with the disadvantage of high costs for maintenance, repair and updating, as well as dependency of a single provider, (i.e., the manufacturers abroad). on the other hand, commercially available seismological instrumentation typically is expensive, rather specialized, predominantly built with proprietary hard and soft components, and very difficult to keep in a good condition. besides overcoming these disadvantages, it is also our aim to further our capacities in scientific instruments design and construction. taking this into account, it is necessary to develop our own seismology instrumentation equipment, with state-of-the-art design but at very low manufacturing, operation and maintenance costs, and transparent to the user, so that they could easily be adapted to specific needs, but also looking for the possibility of upgrading it with the lowest possible expenses. system description the digital acquisition system has been provided with six analog input channels available for instrumentation, which can be connected to two sensors (i.e., a tri-axial seismometer or 3components accelerometer). the signals are conditioned with a programmable gain amplifier and low pass filtered to 250 hz in order to avoid aliasing effects. a 16-bit a/d converter is used to sample the amplified and filtered signals. the digitized data is processed with a digital filter developed for a wide range of frequencies, programmable by the user [1]. the processed data are stored locally and can be accessed online by radio through a serial or ethernet interface. in applications where online acquisition is not required (i.e., the equipment is a standalone station), the data is recorded on a hard disk and downloaded periodically. the user can access remotely, initialize and check the operation of the system, load and observe the acquired data and update the software and recording parameters used for acquisition. the system uses a 40 gb hard disk, which allows one month of continuous recording at 1000 samples/second using 3 channels, or eight months of continuous recording at 125 samples/second. in applications where data transmission (radio, internet, etc.) is used, the hard disk will not store until its maximum capacity because there is a stored option to erase data once have been transferred a high volume of data by telemetry into a data collected station. hardware description the acquisition system includes a signal conditioning stage, a power card to supply ±12 and 5 vdc to the signal conditioning stage, an internet card for communication with an external network, an 8 serial port card to connect the gps system and a spread-spectrum two-way radio, a 16bit a/d converter, and an interconnection card to access the external devices. the signal conditioning stage was designed with a programmable gain amplifier, to amplify the signal supplied by the sensor with a gain range from 1 to 8000 (i.e., 80 db), and input impedance of 10 e 10 ohm || 6 pf. to avoid aliasing effects during the digitisations; a 6th order butter worth low-pass analog filter with a cut-off frequency of 250 hz was added tthis stage. the 16 bits a/d converter has 8 differential or 16 single-ended inputs, with a maximum sampling rate of 100 ksps, 96 db of resolution, and programmable gains of 1, 2, 4 and 8. the maximum input voltage of the converter is ± 10 vdc for a linear operation. figure 1 shows the block diagram of the system as well as the signal conditioning stage. pedro riascos et al. 54 figure 1: block diagram of the ossodas acquisition system the operating system the acquisition system was developed under the linux operating system gnu, which is distributed according to the open source model where the programmers are free to adapt the source code for custom purposes. the actual configuration uses a cpu with the following features [2]: • host platforms supported: redhad linux 7.3 • processor: 2x x 86 • clock speed: 100 mhz • memory: 32mb sdram sodimm module • power supply: atx, 250watt • dev104+, bus extender board (pc/104+ to 2-slot isa and 2-slot pci backplane) • mz104+, pc/104 board with dual ethernet additionally, the programs to acquire, record, and transmit the signals are written in c under linux and they are included in the following modules: • automatic data acquisition • gps configuration • data processing and storing • network configuration • time synchronization features and applications the system designed has the following features: • real-time acquisition of seismic signals • remote monitoring • gps time synchronization • two 3-channel analog inputs • 16-bit digital output • rs232/ethernet connectivity • telemetry access (spread spectrum) • removable hard disk data storage • 176 db full dynamic range • 80 db user selectable analog amplifier • 96 db digital output • programmable sample rate up to 1000 sps • weather sealed housing (for hostile environments) • rapid deployment by unskilled personnel the ossodas system can be used in applications such as: • measurement of environmental noise • seismic signal monitoring • seismic profiling • bridge and building monitoring • passive seismic monitoring (oil fields) • quarry blast vibration control • strong motion seismology technical specification at full operation, the system requires 1.8 a, which could be supplied with a 12 v, 100ah battery and two solar panels of 80w. the system is enclosed in a hermetically sealed pelican pfv case to operate under outdoor conditions. all connectors for accessories of the system are circular plastic connectors amp. the sensor used with the system is the electronic lennartz le3d-light which has a natural frequency of 1 hz and bandwidth of 80 hz [3], although the user can use sensors from different suppliers. figure 2 shows the ossodas layout and table 1 shows the ossodas physical characteristics. ossodas, a portable digital system for seismological signal acquisition table 1:ossodas physical char conclusions and futu a low cost digital system acquisition has been developed acquiring, process, record seismological data by telem propose a system design optio instrumentation, which pr advantages for other regions high seismicity and needs of ear as well as serious limitations maintaining imported commerci the system reliability was chec that is, all of the measured valu the experimental results by val stages that were verified conditioning, the acquisition c and the digital signal processing the future work will be focus digitally configurable analog user to select the gain directly and design of analog and digital figure 2: ossodas layout 55 acteristics re work for seismic signal . the system allows and transmitting etry. in brief, we n for seismological esents significant and countries with thquake monitoring, in acquiring, and al systems. ked stage by stage, es correspond with idated systems. the are: the signal ard, time accuracy, . ed on the design of gains, allowing the from the software, filters using programmable circuits [4]. this technology allows configuring the system in a wide range of frequencies, from 1 to 1000 hz. also, the future work will aim at decreasing the power required by the system in order to operate at low current. this is possible due to the availability of solid-state storage (i.e., disk on chip and compact flash card), which allows sharing the operating systems and acquisition process with the hard disks during the operation of the system. references diamond system corporation, 2002. “rugged embedded solutions for harsh environments and mobile applications”. s. w smith, (1999) “the scientist and engineer guide to digital signal processing”. california technical publishing san diego california, second edition. “le-xd geophone family”, document number 9900003, lennartz electronic, email: info@lennartzelectronic.de “programmable analog circuits”, ispac handbook, lattice semiconductor corporation, january 2000. e. freaking, (1997). “digital signal processing in communication systems”, chapman & hall, new york, ny. f. scherbaum. (1994) “basic concepts in digital signal processing for seismologists”, springerverlag. earth sciences research journal geochemistry earth sci. res. sj. vol. 16, no. 2 (december, 2012): 57 67 the geochemistry of claystone-shale deposits from the maastritchian patti formation, southern bida basin, nigeria olugbenga a okunlola and olubunmi idowu geology department, university of ibadan, ibadan, nigeria. e-mail: gbengaokunlola@yahoo.co.uk abstract an inorganic geochemical study of a claystone and shale sequence from the patti formation around ahoko in the southern bida basin, nigeria, was carried out to determine the basin’s depositional conditions, provenance and tectonics. representative samples underwent mineralogical and geochemical analysis involving major, trace and rare earth element analysis. mineralogical studies using x-ray diffraction analysis revealed prominent kaolinite, dickite and illite peaks; accessory minerals included quartz and microcline. major element abundance showed that the shale samples had sio 2 (61.26%), al2o3 (16.88%) and fe2o3 (3.75%) constituting more than 79% of bulk chemical composition whilst claystone samples contained sio2 (67.79%), al2o3 (17.81%) and fe2o3 (1.67%). higher sio2, ba, sr, v, and nb concentrations were observed in claystone samples rather than shale whereas the shale samples were observed to be more zr-, niand zn-enriched than the claystone ones. the shale and claystone samples showed slightly light rare earth enrichment and slightly flat heavy rare earth depleted patterns having a negative eu and tm anomaly. high tio 2 and rb/k2o values also indicated that the shale and claystone samples were matured. geochemical parameters such as u, u/th, ni/co and cu/zn ratios indicated that these shales were deposited in oxic conditions; the al 2o3/tio2 ratio suggested that intermediate igneous rocks were probable source rocks for the shales, while mafic rocks were suggested as being source rocks for the claystone. however, the la/sc, th/sc, th/ co ratios and shales and claystone plots revealed that they came within the range given for felsic rocks as provenance, thereby suggesting mixed provenace for the sediments. a passive-margin tectonic setting was adduced for the sedimentary sequences. resumen se realizó un estudio de geoquímica inorgánica de una secuencia de arcillolitas y limolitas de la formación patti en ahoko, cuenca sur de bida, nigeria, con el fin de determinar las condiciones de sedimentación de la cuenca, la procedencia de los sedimentos y la tectónica de la zona. a las muestras representativas se les realizaron análisis mineralógicos y geoquímicos: estudio de trasas y análisis de tierras raras. los análisis de difracción de rayos x revelaron importantes niveles de caolinita, dickita e ilita y, minerales accesorios incluido cuarzo y microclina. en las muestras de limolita se encontró sio 2 (61,26%), al2o3 (16,88%) y fe2o3 (3,75%) como constituyentes de más del 79% de la composición química, mientras que las muestras de arcillolita contienen sio2 (67,79%), al2o3 (17.81%) y fe2o3 (1,67%). las concentraciones más altas de sio2, ba, sr, v, nb se observaron en las muestras de lutita mientras que en las muestras de arcillolita se observó una mayor concentración de zr, niy enriquecimiento de znque en las lutitas. las muestras de lutitas y arcillolitas presentan un ligero enriquecimiento de tierras raras y algo de tierras raras pesadas teniendo patrones de anomalía negativa de eu y tm. los altos valores de tio 2 y rb/k2o también indican que las muestras de limolita y arcillolita fueron maduradas. los parámetros geoquímicos como la relación de u, u/th, ni/co y cu/zn indican que las limolitas y arcillolitas fueron depositadas en condiciones aeróbicas, la relación al2o3/tio2 sugiere que las rocas ígneas intermedias fueron probablemente la roca fuente de las limolitas, mientras que se sugiere que las rocas máficas son las rocas fuente de las arcillolitas. sin embargo, la relación palabras claves: geoquímica, inorgánica, esquisto,arcillas, procedencia, tectónica. keywords: geochemical, inorganic, shale, claystone, provenance, tectonic. record manuscript received: 29/09/2010 accepted for publications: 06/05/2012 olugbenga a okunlola and olubunmi idowu58 introduction shale represents the most abundant type of sediment in sedimentary basins worldwide (pettijohn, 1975) and are considered to represent average crustal provenance composition much better than any other detrital sedimentary rocks (mcculloch and wasserburg, 1978). some researchers have suggested that sedimentary rocks’ major element geochemistry is more useful in discriminating a tectonic setting (bhatia, 1983; roser and korsch, 1986); however, major trace elements, such as la, y, sc, cr, th, zr, hf and nb, particularly in combination withtio 2, are best suited for provenance and tectonic setting determination studies because of their relatively low mobility during sedimentary deposition (mclennan et al., 1983). the relative distribution of immobile elements differing in concentration in felsic and basic rocks such as la and th (felsic rock-enriched) and sc, cr and co (enriched in basic rocks relative to felsic rocks) has been used to infer felsic and basic sources’ relative contribution in shales from different tectonic environments (wronkiewicz and condie, 1990). several authors have used geochemical parameters to understand ancient sediments’ paleo-oxygenation conditions (calvert and pedersen, 1993; jones and manning, 1994; nath et al., 1997; cullers, 2002; armstrong-altrin et al., 2003; dobrzinski et al., 2004). this study was therefore aimed at identifying patti formation claystone-shale sequence source rock characteristics, tectonic setting and paleo-environmental conditions in the bida basin around ahoko in central nigeria using their major, trace and rare earth elements’ geochemistry. geology and stratigraphy the bida basin is a northwest-southeast-trending embayment, lying perpendicular to the main axis of the benue trough (figure 1). the south atlantic–tethys seaway was routed through this basin during the campanian to maastrichtian age. the basin is usually regarded as a north-western extension of the anambra basin (whiteman, 1982; ladipo et al., 1994; akande and ojo, 2002; akande et al., 2005), both of them being major depocenters during this period. adeleye (1975a,b) viewed the bida basin as a gently down-warped trough, whose genesis may be closely connected with south-eastern nigeria and the benue valley’s santonian orogenic movements, while whiteman (1982) has suggested that the basin was formed from simple cratonic sag. agyingi (1991) has proposed a post-santonian origin for the bida basin; he reported tension build-up in the benue trough during africa’s separation from south-america (aptian-albian). the adjacent area probably sagged along a northwest-southeast orientated fracture during this period to ease tensional build-up, resulting in the nupe depression forming a side basin. he stated that the sediments in the basin were generally undisturbed as in the case of the post-santonian sediment of the adjacent sokoto and figure 1. map of the study area and sampling locations. longitude n008.21 e06.5 e06.51 e06.52 e06.53 e06.54 e06.55 e06.5 e06.51 e06.52 e06.53 e06.54 e06.55 n008.2 n008.19 n008.18 n008.17 n008.16 n008.15 n008.21 n008.2 n008.19 n008.18 n008.17 n008.16 n008.15 longitude l a t it u d e l a t it u d e la/sc, th/sc, th/co y limolitas indica que la zona de procedencia está dada en el rango de rocas félsicas, lo que sugiere una fuente mixta para los sedimentos. un ambiente tectónico de margen pasivo se adujo para las secuencias sedimentarias. the geochemistry of claystone-shale deposits from the maastritchian patti formation, southern bida basin, nigeria 59 figure 2. stratigraphic successions in the mid-niger basin (taken from akande et al.). figure 3. sections of the patti formation showing dark grey-black shales and claystone inter-bedded with siltstone and sandstone. anambra basins. a maximum sedimentary pile of up to 3,300m has been recorded from aeromagnetic interpretation, suggesting a post-santonian origin (agyingi, 1991). the bida basin’s stratigraphy varies, depending on the area being considered; several authors have discussed this basin’s stratigraphy in two sectors: the northern bida basin (adeleye, 1971) and the southern bida basin (agyingi, 1991; akande and ojo, 2002). the northern bida basin is successively made up of bida sandstone, sakpe ironstone, enagi siltstone and batati ironstone (figure 2) whereas the southern bida basin lithostratigraphy successively consists of lokoja sandstone and the patti and agbaja ironstone formation. this study has concentrated on the southern bida basin’s patti formation. methodology the twenty-five samples used for this study were taken from the exposed shale section in road cuttings at ahoko, along the koton-karifi/ abaji highway. the location being studied lies within the patti formation in the southern bida basin. representative shale and claystone samples were collected from different vertical sections of the outcrops (figure 3). efforts were made to avoid weathered horizons. representative whole rock shale samples were studied by x-ray diffraction (phillips – pw 1011 diffractometer). the diffractograms were recorded using a 1o 2/min/cm scanning rate with ni-filtered fe–k alpha radiation. x -ray diffraction curves were interpreted by comparing notable intensity peaks with those for standard minerals established by carrol (1971). quantitative determination was made by a real method; the shale samples’ clay fraction was not separated for speciation evaluation in this study because, initially, it was necessary to establish the shale sequence’s holistic compositional features to have an unobliterated pattern of its evolution within the basin. geochemical characterisation involved analysing thirteen shale and two claystone samples at the acme analytical laboratories ltd for major, minor and trace element geochemistry using an icp mass spectrometer (perkinelmer, elan 6000) and inductively coupled plasma mass spectrograph on powdered, pressed pellets prepared from 3-5g samples. it was digested by weighing 0.2g aliquot in a graphite crucible mixed with 1.5g libo 2/lib4o7 flux. the crucibles were placed in an oven and heated at 980ºc for 30 minutes. the cooled bead was dissolved in 5% hno 3 (acs grade nitric acid diluted in demineralised water). calibration standards and reagent blanks were added to sample sequences. the basic package consisting of thirty-four elements (ba, co, cs, ga, hf, nb, rb, sn, sr, ta, th, u, v, y, zr, la, ce pr, nd, sm, eu, gd, lu, etc) was determined for the shale and claystone samples. a second 0.5g split sample was digested in aqua regia and analysed by icp-ms to determine au, ag, as, bi, cd, cu, hg, mo, ni, pb, sb, se and zn. an icp emission spectrograph (spectro ciros vision or varian 735) was used for determining major oxides and some trace elements (i.e. sio 2, al2o3, fe2o3, cao, mgo, tio2, p2o5, cr2o5, ba, nb, ni, sr, sc, y and zr). loss on ignition (loi) was determined for both packages by measuring the weight loss after heating a 1g split sample at 95ºc for 90 minutes. results and discussion mineralogy the x-ray diffractograms (xrd) for the whole rock bida shale samples from location a revealed minerals such as kaolinite, quartz, mordenite, microcline, zircon and gamma alumina (figure 4a); kaolinite was the dominant clay mineral. the xrd results for shale in location b also revealed the predominance of kaolinite, quartz, microcline, pyrite and zircon present in subordinate amounts (figure 4b) and whole at location c. kaolinite, illite, quartz, microcline, dickite, mordenite, gamma alumina and iron rich minerals, such as pyrite and hematite, constituted the mineral makeup. the percentage of kaolinite in the shale in location c was 46.91%. typical diffractograms of the claystone revealed kaolinite as the dominant mineral (66%), while illite, dickite, quartz, microcline and zircon were present in subordinate amounts (figure 4d). age northrn bida basin southern bida basin m a st r ic h t ia n batati formation agbaja ironsstone formation enagi formation patti formationsakpe formation c a m pa n ia n jima member claystone (member) bida formation lokoja formation sandstone (member) doko member basal conglomerate (member) l o w e r pa l e o z o ic location c olugbenga a okunlola and olubunmi idowu60 figure 4. (a) x-ray diffractogram for the shale samples from location a. (b) x-ray diffractogram for the shale samples from location b. (c) x-ray diffractogram for the shale samples from location c. (d) x-ray diffractogram for the claystone samples. k clay mineral (probably kaolinite) qz quartz mo mordenite m microcline ga gamma alumina z zircon k clay mineral (probably kaolinite) m microcline qz quartz il clay mineral (probably illite) mo mordenite d dickite h hermatite p pyrite ga gamma alumina k clay mineral (probably kaolinite) qz quartz m microcline p pyrite ga gamma alumina z zircon loi loss on ignition *a4 & *a5 claystone samples tfe2o3 as total iron k clay mineral (probably kaolinite) qz quartz il clay mineral (probably illite) m microcline ga gamma alumina a1 a2 a3 *a4 *a5 b1 b2 b3 b4 b5 c1 c2 c3 c4 c5 sio 2 56.78 57.84 56.91 67.97 67.6 64.08 65.22 64.42 65.2 65.43 59 60.55 59.95 60.6 60.41 al 2 o 3 18.11 18.61 18.74 17.95 17.67 15.29 14.42 15.42 13.94 14.27 18.21 17.85 18.32 17.76 18.53 tfe 2 o 3 5.09 4.4 4.5 1.35 1.98 2.9 3.69 2.95 4.12 3.99 3.76 3.16 3.64 3.3 3.18 mgo 0.18 0.18 0.18 0.13 0.13 0.18 0.15 0.18 0.17 0.18 0.13 0.13 0.13 0.12 0.13 cao 0.06 0.06 0.04 0.08 0.07 0.04 0.05 0.05 0.04 0.05 0.04 0.05 0.05 0.04 0.04 na 2 o 0.05 0.05 0.05 0.07 0.07 0.06 0.07 0.07 0.07 0.07 0.06 0.06 0.06 0.06 0.06 k 2 o 1.13 1.14 1.12 1.64 1.61 1.49 1.5 1.53 1.53 1.5 1.39 1.43 1.44 1.43 1.45 tio 2 1.67 1.76 1.73 2.41 2.43 1.8 1.56 1.84 1.58 1.57 1.8 1.81 1.81 1.82 1.83 p 2 o 5 0.06 0.07 0.07 0.16 0.15 0.07 0.07 0.1 0.07 0.06 0.08 0.08 0.08 0.08 0.09 mno 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.02 0.01 cr 2 o 3 0.013 0.014 0.013 0.015 0.014 0.011 0.011 0.013 0.01 0.01 0.014 0.014 0.014 0.014 0.014 loi 16.5 15.6 16.3 7.8 7.9 13.7 12.8 13 12.8 12.4 15.2 14.6 14.2 14.5 14 total 99.66 99.74 99.67 99.57 99.64 99.64 99.56 99.59 99.56 99.52 99.70 99.75 99.71 99.74 99.74 table 1. major element concentration of the bida shale and claystone samples. (a) (b) (c) (d) the geochemistry of claystone-shale deposits from the maastritchian patti formation, southern bida basin, nigeria 61 geochemistry major element composition table 1 gives the major oxide analysis of thirteen shale and two claystone samples; table 2 summarises the average major element oxide (wt %) data for the samples being studied. these were compared to average shales worldwide (pettijohn, 1957), nasc (gromet et al., 1984), turekan & wedephol (1961) and shales from other parts of nigeria (table 3). it was apparent that most shale samples in table 1 varied slightly regarding sio 2 (56.78-65.43%), had moderately high al2o3 (13.94 -18.74%) and a small variation in fe2o3 values (2.90%-5.09%); they were, however, low in tio2, cao, na2o and k2o. such low k2o content indicated the low amount of illite or k-feldspar present (akpokodje et al., 1991). the claystone was relatively rich in al2o3, tio2 and sio2 compared to the shale, even though the shale had higher fe2o3 content than the claystone. the table 3. comparing average chemical composition of the bida shale studied here to published average shales table 2. average chemical composition of bida shale compared to shale from other sedimentary basins in nigeria oxide present study asu river group (amajor,1987) ezeaku shale (amajor,1987) auchi shale ifon shale (ajayi et al, 1989) bida shale bida claystone sio 2 61.26 67.79 69.94 44.91 51.68 63.3 tio 2 1.74 2.42 0.52 0.65 1.95 1.02 al 2 o 3 16.88 17.81 10 15.71 18.76 18.47 fe 2 o 3 3.75 1.67 4.04 6.24 4.67 1.26 mno 0.02 0.02 0.04 0.06 0.06 0.01 mgo 0.16 0.13 0.87 2.58 4.39 0.82 cao 0.05 0.08 3.38 15.42 1.9 0.09 na 2 o 0.06 0.07 0.4 0.42 0.93 0.42 k 2 o 1.39 1.63 1.15 2.36 1.16 2.36 p 2 o 5 0.08 0.16 0.17 0.46 0.25 0.46 loi 14.2 7.8 9.21 11.1 14.05 11.6 total 99.59 99.58 99.69 99.91 99.87 99.81 oxide average bida shale average shale (pettijohn, 1957) turekan & wedephol (1961) paas nasc (gromet et al., 1984) sio 2 61.26 58.1 58.5 62.40 64.82 tio 2 1.74 0.6 0.77 0.99 0.8 al 2 o 3 16.88 15.4 15 18.78 17.05 fe 2 o 3 3.75 6.9 4.72 7.18 5.7 mno 0.02 trace _ _ mgo 0.16 2.4 2.5 2.19 2.83 cao 0.05 3.1 3.1 1.29 3.51 na 2 o 0.06 1.3 1.3 1.19 1.13 k 2 o 1.39 3.2 3.1 3.68 3.97 p 2 o 5 0.08 0.2 0.16 0.16 0.15 sio 2 /al 2 o 3 3.63 k 2 o/na 2 o 23.16 k 2 o/al 2 o 3 0.08 al 2 o 3 /tio 2 9.70 cu/zn 0.12 ni/co 0.58 olugbenga a okunlola and olubunmi idowu62 claystone (17.67-17.95%) had less variation in al2o3 content than the shale (13.94-18.74%). most samples had low p2o5 content; p2o5 depletion could have been due to the lower amount of accessory phases, such as apatite and monazite, compared to paas (ramasamy et al., 2007). the mgo, mno and na 2o content values were very low in both samples, collectively only accounting for less than 2%. the shale and claystone mgo and cao content was low; the samples thus did not indicate associated carbonates or dolomitisation. the shale and claystone alumina to silica ratio was low (0.27). the 14.28% average loss on ignition (loi) for shale was high, showing great shale potential for carbonaceous compounds. the 3.63 silica-alumina ratio showed that that shale and claystone samples were highly siliceous. the average sio 2 (61.26%) and al2o3 (16.88%) chemical composition in shale constituted about 78% of the samples’ total chemical composition. the other chemical impurities in shale samples from the patti formation were fe 2o3 (3.75%) and tio2 (1.74%). the patti formation shale and claystone samples had higher tio2 values than post-archean australian average shale (paas; taylor and mclennan, 1985). sediments’ k2o/al2o3 ratio can be used as an indicator of ancient sediments’ original composition. the k2o/al2o3 ratios for clay minerals and feldspars are different (0.0 to 0.3, 0.3 to 0.9, respectively) according to cox ( cox et al., 1995). the average k2o/al2o3 ratio for shale varies from 0.06 to 0.1 and is 0.09 for claystone. the k2o/al2o3 ratios were closer to the lower clay mineral range limit in both the shale and claystone samples. comparing the chemical composition of the shale and claystone samples from bida with that of shales from other parts of nigeria revealed that the shale and claystone samples were relatively rich in sio 2, al2o3, tio2 and fe2o3 comparable with black lignite shale from ifon and shale from the auchi area (akpokodje et al.,). the present study’s cao and mno values also had values comparable with ifon shale. the samples were rich in sio2, al2o3 and tio2 compared to ezeaku shales; however, the ezeaku shale had higher fe2o3 and cao compared to bida shale and claystone. the asu river group had higher sio2, fe2o3 and cao values but lower tio2 and fe2o3 values compared to the bida shale and claystone samples from the a1 a2 a3 *a4 *a5 b1 b2 b3 b4 b5 c1 c2 c3 c4 c5 ba 334 343 340 630 563 462 460 465 468 430 356 373 364 359 371 sr 54.2 49.6 51.1 77 73.3 59 59.9 58.4 56.7 53.2 60 66.8 64.8 65.3 73.1 ni 31.7 37.6 24.9 <0.1 0.3 14.5 18.4 47.8 14 18.2 14.5 7.5 10.4 6.3 6 co 49.4 56.5 35.7 4.1 4.2 27.3 28.7 71 24.7 27.8 35.2 19.9 23.8 18.3 16.1 cu 14.1 12 14.2 6 5.1 12.8 10.3 12.6 10.5 10.9 17 17.1 19.8 18.4 18.2 zn 214 177 180 3 3 117 307 159 110 117 34 24 40 16 18 v 126 137 134 111 116 109 76 105 72 69 117 113 123 118 115 y 53.4 52.4 54.9 71.7 78.1 78.3 83.5 76.2 84.2 85.3 77.5 68.4 69.4 66.1 69.4 zr 922.4 829.2 838.4 1358.3 1432.4 1270.5 1735.9 1252.5 1731.4 1702.9 927.5 1001 928.8 977.9 916.6 mo 1.2 1.1 1 0.08 0.2 0.3 1.2 0.6 0.9 1.2 0.4 0.3 0.3 0.4 0.4 nb 48.2 50.3 49.1 70.8 71.3 54.5 49.2 57.6 49 48 55.2 56.3 55 54.8 54.8 pb 28.8 28.8 28.6 14.6 11.6 21.7 22.4 22.5 21.1 22.1 18.3 16.6 21.2 18.1 19.4 rb 36.8 34.2 35.6 47.4 47.3 53 49.4 52.1 50.2 47 48.8 51 47 46.8 48.5 th 24.7 28.3 26.6 32 29.9 28.5 28.9 29.3 33.1 29.9 30.8 29.1 28.9 29.5 32.2 u 10.5 10.6 10.6 13 13 13 11.1 13.4 11.5 11.1 14.5 16.8 15.3 15.8 15.7 table 4. trace element distribution of shales and claystone samples from ahoko. *a4 & a5 claystone samples a1-a3: shale from location a a4-a5: claystone from locations b1-b5: shale from location b c1-c4: shale from location c figure 5. chondrite normalised rare earth element plot for shale and claystone samples from the patti formation. chondrite normalisation values were taken from wakita et al., (1971) patti formation. the shale’s mgo and cao contents were generally low by contrast with values reported by emofurieta et al., (1994) for ewekoro and gombe shales. when compared to shales from other parts of the world (table 4), the bida basin shale had similar averages to those reported by pettijohn (1957), turekian and wedephol (1961) and nasc (gromet et al., 1984) in terms of sio 2, al2o3 and fe2o3; the tio2 composition of the bida shale was perhaps slightly higher than that of the other shales. the bida shale samples were, on average, relatively rich in sio 2, al2o3, tio2 the geochemistry of claystone-shale deposits from the maastritchian patti formation, southern bida basin, nigeria 63 and p2o5 comparable with paas and nasc (gromet et al.,1984). however, the bida shale had depleted mgo, cao, na2o, k2o composition compared to world shale averages. provenance, tectonic setting and depositional environment the geochemical signatures of clastic sediments have been used to ascertain provenance characteristics (taylor and mclennan, 1985; condie et al., 1992; cullers, 1995; armstrong-altrin el al., 2004). most clastic rocks’ al 2o3/tio2 ratios are essentially used to infer source rock composition because al2o3/tio2 ratios increase from 3 to 8 for mafic igneous rocks, from 8 to 21 for intermediate rocks and from 21 to 70 for felsic igneous rocks (hayashi et al., 1997). the al2o3/tio2 ranged from 8.38 to 10.84 in the ahoko shales (patti formation) and 7.27 to 7.44 in the claystone samples; hence, this study’s al2o3/tio2 ratio suggested intermediate rocks as being probable source rocks for the shale samples while mafic rocks were suggested as being source rocks for the claystone samples. cr and ni abundance in siliciclastic sediments was considered a useful provenance tool. cr and ni concentration was low in the shale and claystone samples; a low cr concentration indicates felsic provenance, according to wrafter and graham (1989), and high cr and ni content is mainly found in ultramafic rock-derived sediment (armstrong-altrin et al., 2004). the cr and ni content in these shales was low, therefore suggesting felsic provenance. ratios such as la/sc, th/sc and th/co are significantly different in felsic and basic rocks and may lead to constraints on average provenance composition (wronkiewicz and condie, 1990; cox et al., 1995; cullers, 1995). the th/co, th/sc and la/sc ratios for shale and claystone samples from this study were compared to those of felsic and basic rock-derived sediment (fine fraction) upper continental crust (ucc) and paas values (table 8). these comparisons also indicated that such ratios came within the range of felsic source rocks. the th/co compared to la/sc plot (figure 5; cullers, 2002) also suggested that the patti formation shale and claystone samples were both derived from felsic source rocks. table 7 shows the shale and claystone samples’ rare earth element (ree) concentration; a slight variation in ∑ree content was observed between the shale (361.18 to 422.75) and claystone samples (365.83 to 371.04). bulk ree normally reside in the fine fraction (silt or clay) and it has also been inferred that trivalent ree is readily accommodated in most clay minerals enriched with alumina and ferric iron (cullers et al., 1987, 1988). the chondrite-normalised ree plot (figure 5) showed that shale and claystone samples from the patti formation were lreeenriched and had an almost flat hree pattern, with negative eu and tm anomalies. table 5. average trace element chemical composition compared to bida shale and shale from other sedimentary basins. *paas post archean austrialian average shale *nasc – north-american shale composite n.a. not analysed present study levinson (1974) vine & tourtelot (1970) turekan & wedephol (1961) *paas *nasc (gromet et al., 1984)bida shale claystone average ba 394.23 596.5 300 700 580 650 636 sr 59.39 75.15 200 300 300 200 142 ni 19.37 0.19 50 70 68 55 58 co 33.42 4.15 10 20 _ 23 n.a cu 14.45 5.55 70 50 45 50 n.a zn 116.39 3 300 100 95 85 n.a v 108.77 113.5 150 130 130 150 130 y 70.69 74.9 30 25 _ n.a zr 1156.54 1395.35 70 160 160 210 200 mo 0.72 0.14 10 3 _ n.a nb 52.46 71.05 20 20 n.a 1.90 n.a pb 22.28 13.1 20 n.a n.a 20 n.a rb 46.19 47.35 140 n.a n.a 160 n.a th 29.22 30.95 12 n.a n.a 14.60 n.a u 13.07 13 4 n.a n.a 3.10 n.a cu/zn 0.12 1.85 (cu +mo)/zn 0.13 1.90 ni/co 0.58 0.05 rb/k 2 o 33.23 29.05 u/th 0.45 0.42 olugbenga a okunlola and olubunmi idowu64 present study paas codo shale (mclennan 1980) average shale (levinson,1974)bida shale claystone la 77.40 76.35 38.2 29.7 121 ce 170.42 160.45 79.6 63.4 50 pr 18.71 17.62 8.83 _ _ nd 67.66 57.75 33.9 27.9 24 sm 12.46 10.78 5.55 _ _ eu 2.25 1.78 1.08 _ _ gd 11.01 8.91 4.66 _ _ tb 1.98 1.74 0.744 _ _ dy 11.79 11.13 4.68 _ _ ho 2.48 2.52 0.991 _ _ er 7.69 8.32 2.85 _ _ tm 1.14 1.24 0.405 _ _ lu 1.18 1.31 0.433 _ _ table 6. average rare earth elements of bida shale compared to world averages. table 7. rare earth element distribution of the shale and claystone samples from the location studied here. *paas post archean australia shale a1 a2 a3 *a4 *a5 b1 b2 b3 b 4 b 5 c1 c2 c3 c4 c5 la 71.10 74.30 75.3 77.7 75 77.3 79.1 73.5 76.6 77.1 78.7 78.3 80.2 80.6 84.1 ce 160.7 172.5 168.7 163 157.9 162.5 171.7 162.3 166.7 167 170.8 172.7 176.1 177.6 186.2 pr 16.58 18 17.58 17.73 17.51 18.37 19.17 18.57 18.86 18.99 19.17 19.07 19.34 19.11 20.36 nd 59.6 68.1 65.9 57.8 57.7 75.6 67.1 66.6 64.7 65.7 65.1 68.3 70.7 68.9 73.3 sm 11.65 12.22 12.4 10.67 10.89 13.73 12.24 12.79 12.1 12.35 12.51 12.72 12.4 12.06 12.86 eu 2.15 2.35 2.42 1.81 1.75 2.24 1.98 2.29 1.95 2.04 2.42 2.31 2.35 2.3 2.41 gd 10.53 10.19 10.37 8.59 9.22 12.02 10.9 11.09 11.45 11.71 11.32 10.92 10.99 11.02 11.52 tb 1.78 1.82 1.83 1.73 1.74 2.09 2.1 2.04 2.12 2.17 2.01 1.94 1.93 1.89 1.96 dy 10.22 10.06 10.99 10.67 11.59 12.49 13.06 12.19 12.87 13.49 12.34 11.69 11.21 11.29 11.42 ho 2.22 2.07 2.24 2.4 2.63 2.76 2.87 2.58 2.84 2.83 2.57 2.37 2.32 2.23 2.36 er 6.34 6.5 6.54 8.11 8.53 7.87 9.42 8.12 9.26 9.24 8.18 7.39 6.97 6.87 7.2 tm 0.94 0.89 0.92 1.2 1.27 1.24 1.35 1.23 1.4 1.35 1.16 1.11 1.07 1.06 1.11 yb 6.41 6.19 6.41 8.37 8.74 8.85 9.11 8.17 9.18 9.1 7.67 7.42 6.93 6.92 6.84 lu 0.96 0.96 0.97 1.26 1.36 1.33 1.44 1.26 1.46 1.41 1.17 1.11 1.09 1.11 1.11 the sedimentary rocks’ ree pattern and eu anomaly also help in providing important clues regarding source rock characteristics (taylor and mclennan, 1985). higher lree/hree ratios and negative eu anomalies are generally found in felsic rocks, whereas mafic rocks have lower lree/ hree ratios and no or small eu anomalies (cullers, 1995). higher lree/ hree ratios and negative eu anomalies further confirmed the felsic source rock characteristics of the shale and claystone samples studied here. bhatia and crook (1986) identified la, th, zr, nb, y, sc, co, and ti as being the most useful elements in discriminating different tectonic settings. distinctive field for four environments (oceanic island arc, continental island arc, active continental margin and passive margin) are recognised on la-th-sc and th-sc-zr/10 trivariate plots. the tectonic setting identified in the present study was passive-margin type. figure 7 shows the shale and claystone units plot within the passive margin using a discriminant function diagram (bhatia 1983). the roser and korsch (1986) plot of log (k 2o/na2o) cf sio2 discrimination diagram and th-sc-zr/10 greywackes discrimination diagram (after bhatia and crook, 1986) indicated a passive-margin tectonic setting (figures 8 and 9). the passive-margin comthe geochemistry of claystone-shale deposits from the maastritchian patti formation, southern bida basin, nigeria 65 figure 6. th/co cf la/sc for shale and claystone samples from the patti formation (fields taken from cullers, 2002). shale clay 10 1 th /c o la/sc 0,1 0,01 0,01 basic rocks silic rocks 0,1 1 10 table 8. range of claystone and shale element ratios in this study compared to ratios for similar fractions derived from felsic rocks, mafic rocks, upper continental crust and post-archean australia shale prised atlantic-type rifted continental margins developed along the edges of the continent, remnant, ocean basins adjacent to collision orogens, and inactive or extinct convergent margins. intra-cratonic and rift-bounded grabens (e.g. the benue trough) were formed on a thick continental crust which are included in the passive-margin type tectonic setting. the bida basin is an intra-cratonic sedimentary basin. passive-margin type sandstone is generally enriched in sio 2 and depleted in na2o, cao and tio2, suggesting their highly recycled and mature nature (bhatia, 1983). major element analysis of the studied shale and claystone samples confirmed this, as the samples were also enriched in sio 2 but depleted in na2o, cao and tio2, respectively. paleo-oxygenation condition cu/zn and (cu+mo)/zn ratios have been put forward by hallberg (1976) as redox parameters. according to hallberg (1976), high cu/zn ratios indicate reducing depositional conditions, while low cu/zn ratios suggest oxidising conditions; he showed that high values should indicate more reducing conditions in the depositional basins than lower values, which may indicate more oxidising conditions. in this present study, the shale and claystone samples had ~0.03-1.15 and ~1.7-2.0 cu/zn ratios, respectively (table 5), indicating more oxidising conditions. wedephol (1968) noted that average shales reflect shallow marine sediments accumulating in oxidising conditions. the u/th ratio may be used as a redox indicator with u/th ratio being higher in organic-rich mudstones (jones and manning, 1994). a u/th ratio below 1.25 suggests oxic deposition conditions whereas values above 1.25 suggest suboxic and anoxic conditions (nath et al., 1997). in this study, the shale and claystone samples had low u/th ratios (table 5: ~0.37-0.58, ~0.41-0.44, respectively), indicating that both shale and claystone samples were deposited in an oxic environment. dypvik (1984) and dill (1986) have used the ni/co ratio as a redox indicator. jones and manning (1994) have suggested that ni/co ratios below 5 indicate oxic environments, whereas ratios above 5 indicate suboxic and anoxic environments. thus, the low ni/co ratio for shale (~0.340.70) and claystone (~0.01-0.07) samples indicate that patti formation shale and claystone was deposited in well-oxidising conditions. conclusions field studies have indicated that patti formation shales are found at the base of an outcrop section at ahoko along the koton-karfi and abaji axis, occurring alongside siltstones, claystones and sandstone inter-bedded with bioturbated ironstone. the mineralogical composition of the whole rock shale samples based on x-ray diffraction analysis identified clay mineral constituents as probably being kaolinite and illite, although, subsequent detailed clay fraction determination is needed to further confirm this. other non-clay minerals included microcline, quartz, zircon, pyrite and hematite. the 1 this study, 2 cullers (1994, 2000), cullers and podkovyrov (2000), cullers et al., (1988), 3 taylor and mclennan (1985). element ratio 1shale 1claystone range of sediments2 upper continental crust3 post-archean average shale3felsic rocks mafic rocks th/sc 1.37-2.36 1.57-1.60 0.84-20.5 0.05-0.22 0.79 0.9 th/co 0.5-2.0 7.11-7.81 0.67-19.4 0.04-1 0.63 0.63 la/sc 3.95-5.65 3.89-3.95 2.5-16.3 0.43-0.86 2.21 2.4 figure 7. a discriminant function diagram for sandstone (taken from bhatia, 1983) showing fields for shale samples from passive continental margins, oceanic island arcs, continental island arc and active continental margins. the discriminant function was as follows: discriminant function 1 = -0.447 sio2 + 0.972 tio2 + 0.008 al2o3 -0.267 fe2o3 + 0.208 feo -3.082 mno + 0.140 mgo + 0.195 cao + 0.719 na2o – 0.032 k2o + 7.510 p2o5+0.303 discriminant function 2 = -0.421 sio2 + 1.988 tio2 0.526 al2o3 -0.551 fe2o3 -1.610 feo + 2.720 mno + 0.881 mgo – 0.907 cao 0.177 na2o – 1.840 k2o + 7.244 p2o5+ 43.57 -6 -4 -2 0 2 4 -6 6 5 4 3 2 1 0 -1 -2 -3 -4 -5 -6 discriminant function 1 d is cr im in an t fu n ct io n 2 passive margin samples oceanic island-arc continental island-arc active continental margin olugbenga a okunlola and olubunmi idowu66 proportions of minerals identified in the shale and claystone samples varied. the patti formation’s claystone and shale samples showed considerable variation regarding their major, trace and rare earth elements. major element abundance showed that shale samples had sio 2 (61.26%), al2o3 (16.88%) and fe2o3 (3.75%), constituting more than 79% of the bulk chemical composition while claystone samples contained sio2 (67.79%), al2o3 (17.81%) and fe2o3 (1.67%). there was close correlation between the major element and rare earth element composition of shale and claystone samples in the bida basin. higher sio 2, ba, sr, v and nb concentrations were observed in claystone than shale, whereas shale samples were more zr-, niand zn-enriched than the claystone samples. zr was extremely high in both samples. the discrimination diagrams (bhatia, 1983) for distinguishing the tectonic setting mostly indicated a passive-margin type. the provenance indicated a non-homogenous source rock. analysis and interpretation using geochemical parameters, such as la/sc, th/sc and th/co ratios and la/sc cf th/co, suggested that the shale and claystone samples might have been derived from felsic source rock rather than intermediate or mafic source rocks, such provenance being indicated from interpretation using the al 2o3/ tio2 ratio. geochemical parameters like u, u/th, ni/co and cu/zn ratios strongly implied that these shales were deposited in an oxic environment. references adeleye, d.r. 1971. stratigraphy and sedimentation of the upper cretaceous strata around bida, nigeria. unpublished ph.d. thesis, department of geology, university of ibadan, 297 pp. adeleye, d.r. 1975a. a fauna from the ironstones of the middle niger valley, nigeria. journal of minning geology, vol. 8. pp. 45-48. adeleye, d.r. 1975b. nigeria cretaceous stratigraphy and paleography. american association of petroleum geologists, vol. 59, pp. 23022313. adeleye, d.r. 1976. the geology of the middle niger basin. in: c.a. kogbe (ed) the geology of nigeria, univ. ife., pp. 283 – 287. agyingi, c.m., 1991. geology of upper cretaceous rocks in the eastern bida, nigeria. unpublished ph.d. thesis. department of geology, university of ibadan, 501pp. akande, s.o. and ojo o.j., 2002. organic petrology and rock-eval studies on source rock facies of champion-maastrichtian patti formation, southern bida basin, nigeria. nigerian association of petroleum explorationists bulletin, no. 16, pp. 21-39. akande, s.o., ojo o.j. and ladipo, k.o., 2005. upper cretaceous sequences in the southern bida basin, nigeria. a field guide book (nigerian association of petroleum explorationists), 60pp. akpokodje, e.g., etu-efeotor, j.o and olorunfemi, b.n. 1991. the composition and physical properties of some ceramic and pottery clays of south eastern nigeria. journal. mining and geology. 27,pp 9-15. amajor ,l.c, 1987. major and trace element geochemistry of albian and turonian shales from the southern benue trough nigeria, jour .afric earth sci 6 pp633-641. armstrong-altrin, j.s., lee, y.i., verma, s.p., ramasamy, s., 2004. geochemistry of sandstones from the upper miocene kudankulam formation, southern india: implications for provenance, weathering, and tectonic setting: journal of sedimentary research, 74(2), 285297. bhatia, m.r., 1983, plate tectonics and geochemical composition of sandstones: journal of geology, 91, 611-627. bhatia, m.r., crook, k.a.w., 1986. trace element characteristics of greywackes and tectonic setting discrimination of sedimentary basins. contrib. mineral. petrol. 92, 181–193. calvert, s.e., pedersen, t.f., 1993, geochemistry of recent oxic and anoxic marine sediments, implications for the geological records: marine geology, 113(1-2), 67-88. condie, k.c., wronkiewicz, d.j., 1990. a new look at the archean-proterozoic boundary: sediments and the tectonic setting constraint, in naqvi, s.m. (ed.), precambrian continental crust and its economic resources: elsevier, amsterdam, 61-84. condie, k.c., boryta, m.d., liu, j., quian, x., 1992. the origin of khondalites: geochemical evidence from the archean to early pro100 10 1 0,1 60 70 80 90 100 l og (k 2o /n a 2 o ) sio 2 (%) island arc active continental margin pasive margin b figure 8. roser and korsch (1986) log (k2o/na2o) cf sio2 discrimination diagram for sandstone-mudstone suites showing the fields for a passive continental margin, an active continental margin and an island arc. the plotting coordinates for the field boundaries have been extracted from roser and korsch (1986). 90/10 80/20 80/20 80/20 70/30 70/30 70/30 60/40 60/40 60/40 50/50 50/50 50/50 50/50 40/60 40/60 30/70 30/70 30/70 30/70 20/80 20/80 10/90 10/90 10/90 90/10 90/10 thth/sc/zr/10 distribution ahoko aoutcrop samples sc zr/10 figure 9. th-sc-zr/10 greywackes discrimination diagram (taken from bhatia and crook, 1986).the fields are: a oceanic island–arc, b continental island–arc, c -active continental margin, d passive margin b a c d location a location b location c the geochemistry of claystone-shale deposits from the maastritchian patti formation, southern bida basin, nigeria 67 terozoic granulitic belt in the north china craton: precambrian research, 59(3-4), 207-223. cullers, r.l., 1995. the controls on the major and trace element evolution of shales, siltstones and sandstones of ordovician to tertiary age in the wet mountain region, colorado, u.s.a: chemical geology, 123(1-4), 107-131. cox, r., lowe, d.r., cullers, r.l., 1995. the influence of sediment recycling and basement composition on evolution of mudrock chemistry in the southwestern united states: geochimica et cosmochimica acta, 59(14), 2919-2940. dill, h., 1986. metallogenesis of early paleozoic graptolite shales from the graefenthal horst (northern bavaria-federal republic of germany): economic geology, 81, 889-903. dobrzinski, n., bahlburg, h., strauss, h., zhang, q.r., 2004. geochemical climate proxies applied to the neoproterozoic glacial succession on the yangtze platform, south china, in jenkins, g. et al. (eds.), the extreme proterozoic: geology, geochemistry and climate: american geophysical union monograph series, 146, 13-32. dypvik, h., 1984. geochemical compositions and depositional conditions of upper jurassic and lower cretaceous yorkshire clays, england:geological magazine, 121(5), 489-504. emofurieta, w.o.,ogundimu, t.o and imeokpara, e.c.1994. mineralogical geochemical and economic appraisal of some clay and shale deposits in southwestern nigeria. nig. jour. min.geol.vol. 30, p151159. hallberg, r.o., 1976, a geochemical method for investigation of palaeoredox conditions in sediments : ambio, special report, 4,139-147. hayashi, k., fujisawa, h., holland, h., ohmoto, h., 1997, geochemistry of ~1.9 ga sedimentary rocks from northeastern labrador, canada: geochimica et cosmochimica acta, 61(19), 4115-4137. gromet, l.p., dymek,r.f., haskin,l.a.,andkorotev,r.l,1984. the north american shale composite. its compilation, major and trace element characteristics. geochim.cosmochim.acta 48,2469-2482 jones, b., manning, d.c., 1994. comparison of geochemical indices used for the interpretation of paleo-redox conditions in ancient mudstones: chemical geology, 111(1-4), 111-129. ladipo, k.o., akande, s.o. and mucke, a., 1994. genesis of ironstones from middle niger sedimentary basin, evidence form sedimentological ore microscopic and geochemical studies. journal of mining and geology, vol. 30, pp. 161-168. levinson, a.a. 1974. introduction to exploration geochemistry, applied publishing, illnois. mclennan,s .m ,and taylor,s.r, 1980. geochemical standards for sedimentary rocks: trace elements data for usgs standards scoi,mag-i and sgr-i.chem.geol.29,333-343. mcculloch, m.t., wasserburg, g.j., 1978, sm-nd and rb-sr chronology of continental crust formation: science, 200, 1003-1011. nath, b.n., bau, m., ramalingeswara rao, b., rao, ch.m., 1997, trace and rare earth elemental variation in arabian sea sediments through a transect across the oxygen minimum zone: geochimica et cosmochimica acta, 61(12), 2375-2388. pettijohn, f.j. 1957. sedimentary rocks, 2nd edn. harper & row, new york. roser, b.p., korsch, r.j., 1986. determination of tectonic setting of sandstonemudstone suites using sio2 content and k2o/na2o ratio. j. geol. 94, 635–650. roser, b.p., korsch, r.j., 1988. provenance signature of sandstone-mudstone suite determined using discriminant function analysis of major element data. chem. geol. 67, 119–139. taylor, s.r., mclennan, s., 1985.the continental crust: its composition and evolution: blackwell, oxford, 312 p. turekian, k.k., and wedepohl, k.h., 1961. distribution of the elements in some major. units of the earth’s crust: geological society of america bulletin, v, 72, p. 175-191. vine, j.d. and tourtelot, e.b.1970. geochemistry of black shale deposits a summary report. econ. geol. vol.65, p253-272. whiteman, a. 1982. nigeria: its petroleum geology, resources and potentials. vol. 1 & 2. graham and trotman ltd. london. uk. wrafter, j.p., graham, j.r., 1989. ophiolitic detritus in the ordovician sediments of south mayo ireland: journal of the geological society, london, 146, 213-215. wronkiewicz, d.j., condie, k.c., 1990. geochemistry and mineralogy of sediments from the ventersdorp and transvaal supergroups, south africa: cratonic evolution during the early proterozoic: geochimica et cosmochimica acta, 54(2), 343-354. palabras clave: guangdong coast; maritime cyclones; statistical characteristics; temporalspatial characteristics; path analysis. how to cite item chen, c., wang, l., chen, r., xing, f., & chen, j. (2018). temporal-spatial characteristics and path analysis of maritime cyclones in guangdong coastal areas in the south china sea. earth sciences research journal, 22(4), 319-325. doi: https://doi.org/10.15446/esrj.v22n4.77361 this study was conducted to analyze the maritime cyclone characteristics in guangdong coast in the years of 1949 to 2016, including inter-annual variation, the intensity of tropical cyclones, generating location and time, and path direction. also, the temporal-spatial characteristics were also studied. results show there were 183 tropical cyclones landed in guangdong coast in the past 68 years, with an average of 2.7 each year, which more than 60 percent were a typhoon. most of the tropical cyclones were generated in the northwest pacific, spanning from april to december. the path directions were mainly north, northwest, and west. the strengths of the tropical cyclones were reduced from central guangdong coast to the east and the west sides, and the section of zhanjiang city to shenzhen city was the most vulnerable to tropical cyclones. tropical cyclones that generated in the south china sea tend to attack the west of the guangdong coast, while the ones that produced in the northwest pacific tend to attack the east of the guangdong coast. in the study area, the tropical cyclones frequently occurred from july to september and became strongest in september. there are a most common landing section and path direction for each month. finally, based on the statistical data and research results, the tropical cyclone paths in guangdong coast were preliminarily analyzed. abstract keywords: guangdong coast; maritime cyclones; statistical characteristics; temporal-spatial characteristics; path analysis. temporal-spatial characteristics and path analysis of maritime cyclones in guangdong coastal areas in the south china sea características temporales y espaciales y análisis de recorrido de ciclones marítimos en las areas costeras de guangdong, en el mar de la china meridional issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n4.77361 este estudio se realizó para analizar las características de los ciclones marítimos en la costa de guangdong entre los años 1949 y 2016, además de su variación interanual, la intensidad de los ciclones tropicales, el lugar y el momento de creación, y su recorrido. las características temporales y espaciales también fueron estudiadas. los resultados muestran que 183 ciclones tropicales tocaron tierra en la costa de guangdong en estos 68 años, con un promedio de 2.7 por año, de los cuales el 60 por ciento se consideran tifones. muchos de los ciclones tropicales se generaron en el noroeste del pacífico, y van de abril a diciembre. las direcciones de recorrido son principalmente norte, noroeste, y oeste. la fuerza de los ciclones tropicales se redujeron desde la costa central de guangdong hacia el este y el oeste, mientras que el sector entre las ciudades de zhanjiang y shenzhen es el más vulnerable a los ciclones. los ciclones tropicales generados en el mar del sur de china tienden a tocar tierra en el oeste de la costa de guangdong, mientras que aquellos que se producen en el noroeste del pacífico suelen recorrer hacia la costa este. en el área de estudio, los ciclones tropicales frecuentemente ocurren de julio a septiembre y son más fuertes en septiembre. se evidenciaron recorridos y direcciones de los ciclones propias a cada mes. finalmente, de acuerdo con la información estadística y los resultados de investigación, el recorrido de los ciclones tropicales en la costa de guangdong fueron analizados preliminarmente. resumen record manuscript received: : 24/07/2018 accepted for publication: 01/12/2018 earth sciences research journal earth sci. res. j. vol. 22, no. 4 (december, 2018): 319-325 cheng chen1*, lei wang2, ruozhou chen2, fangliang xing2, jun chen2 1college of civil engineering, fuzhou university, fuzhou, 350116, china 2the pearl river hydraulic research institute, guangzhou, 510611, china *corresponding author: chencheng_1117@163.com h y d r o m e t e o r o l o g y 320 cheng chen, lei wang, ruozhou chen, fangliang xing, jun chen 1. introduction a tropical cyclone is an intense natural phenomenon, and the destructiveness of tropical cyclones over the past years is increasing (emanuel, 2005). it is generated in the ocean, which may cause substantial wind waves and storm surge. therefore, devastating damages will be induced in the coastal areas due to a tropical cyclone (yuan et al., 2010). china is one of the most vulnerable countries that suffered from the wind waves and storm surge disasters (jia et al., 2010), and according to the official records, there were thirteen major maritime disasters possibly caused by the tropical cyclones (liu et al., 2017). consequently, more and more researchers are focusing on this topic (li et al., 2004; yuan et al., 2011; zhang & cui, 2013). guangdong province is located on the west bank of the pacific ocean, and it is on the brink of the south china sea. it is frequently subjected to the invasion of tropical cyclones that generated in the northwest pacific or the south china sea (wen, 2002; tang et al., 2003). the cyclone weather conditions and huge waves generated by tropical cyclones will affect the fishing vessels (rezaee et al., 2016). also, huge waves are a severe threat to the structural safety of coastal, offshore, and marine projects, as well as the daily life of coastal residents (xu et al., 2008; zhao et al., 2010). three rings of the coastal planning record cyclones when they attack the coast (tucker et al., 2018). also, coastal planning such as mangrove creeks provides the service of protection (williams et al., 2007). it is essential to study the characteristics of the tropical cyclones that generated in the ocean to mitigate the disasters in coastal areas (zheng et al., 2014). also, observations of the maritime cyclones help forecasts of the maritime disasters (liou and elsberry, 1987). in this study, the landing tropical cyclones characteristics in guangdong coast in the years of 1949 to 2016 were analyzed, including inter-annual variation, the intensity of tropical cyclones, generating source and time, and path direction. based on the statistical data and research results, the tropical cyclone paths in guangdong coast were preliminarily analyzed, which may provide some useful information for wind wave simulations and disaster prevention in coastal engineering. 2. material and methods 2.1. tropical cyclone classification on the sea in the south china sea, most of the tropical cyclones that affect the guangdong coastal areas are generated in the south china sea and northwestern pacific. the intensity of the tropical cyclone is based on the maximum wind speed near the center of the tropical cyclone. generally, the intensity of the tropical cyclone is divided as follows: 1) super strong typhoon, the maximum average wind speed near the bottom centre is greater than 51.0 m/s (16 stages and above); 2) strong typhoon, the maximum average wind speed near the bottom center is 41.5-50.9 m/s (14-15 stage); 3) typhoon, the maximum average wind speed near the bottom centre is 32.7 41.4 m/s (12-13 stage); 4) strong tropical storm, the maximum average wind speed near the bottom center is 24.5-32.6 m/s (10-11 stage); 5) tropical storm, the maximum average wind speed near the bottom center is 17.2-24.4 m/s (8-9 stage); 6) tropical depression, the maximum average wind speed near the bottom center is 10.8-17.1 m/s (6-7 stage). the typhoon, strong typhoon and super strong typhoon are classified as typhoons in this study because the cyclones stronger than typhoon will demonstrate destructive power to coastal engineering. 2.2. data collection in this study, the data of the cyclones were collected from the fujian water conservancy information network in china. the cyclone data include all the previous tropical cyclone numbers, chinese and english names, hourly path (in coordinates), wind stage, maximum wind speed, central pressure, shift speed and direction of the wind center, radius of 7 and 10 stages wind, etc. the cyclones that weaker than 7 stage were not analyzed in this study because they hardly cause huge damages. the cyclones are chosen from the years of 1949 to 2016, with the period of 68 years. the statistical space area is a square area (17-29 °n, 108-120 °e), including guangdong province and its coastal area. in data selection, 4 data points (i.e. 0:00, 6:00, 12:00, 18:00) are selected at a daily interval. as the wind power is weaker than stage 7, it was assumed the cyclone is disappeared, and the data collection is stopped. the coastal line of guangdong is divided into six sections, to distinguish the landing area of the tropical cyclone in guangdong province, starting from the xuwen county in the west, to the raoping county in the east. figure 1 shows the classification of the sections: a xuwen county; b zhanjiang city; c yangjiang city; d shenzhen city; e shanwei city; f shantou city; g raoping county. 3. results 3.1. inter-annual variation figure 2 presents the number of tropical cyclones landing in guangdong province every year (from 1949 to 2016). results show that the frequency distribution of tropical cyclones which landed in guangdong province every year is not uniform, and the tendency is irregular. during the 68 years, 183 tropical cyclones landed in guangdong province, with an average of 2.7 per year. to be more specific, the largest number of landing time is in the year 1999, which exceeded seven times. the second largest amount of the landing time is the years of 1961, 1993, and 2008, which exceeded six times. the smallest number of the landing time is the years of 1950, 1956, 1959, and 2000, which is 0. that means no tropical cyclone was observed in the study area in these years. for most of the years, the landing time of the tropical cyclones is 2 4, which also accords with the landing time in recent eight years. 3.2. tropical cyclone intensity in the coastal area in this study, the strength of the tropical cyclone that lands in guangdong coast is divided into four types: 1) tropical depression, 2) tropical storm, 3) strong tropical storm, and 4) typhoon. figure 3 shows the intensity distribution of tropical cyclones. results show that the landing times increases as increasing cyclone intensity. to be more specific, typhoon occurs for the most of the landing times, which exceeded to 114 times (accounting for more than 60% of the total), and the averaged landing time in each year is 1.7. for the strong tropical cyclone, it is the second most of the landing times, which exceeded to 40 times (accounting for about 20% in total), and the averaged landing time in each year is 0.6. however, the tropical depression that landing guangdong coast occurred for only seven times during the 68 years (only accounting for 4% of the total), which means it lands guangdong coast once every ten years. according to the statistical results, the strongest typhoon during these 68 years is no. 16 "sally" in 1946. the highest stage of typhoon "sally" was 17 stage, and the maximum wind speed near center during the moving exceeded to 87 m/s. the second strongest cyclones during these 68 years are no. 27 in 1954 and no. 14 in 1968. the intensities of both of them also exceed to the 17 stage, and their maximum wind speeds near center during the moving exceeded to 77 m/s. 3.3. generation location and time the generation sea areas of the tropical cyclones that land in guangdong province are the south china sea and the northwest pacific. one hundred eighty-three tropical cyclones fell in guangdong province during these 68 years, in which 134 cyclones were generated in the northwest pacific, and only 49 cyclones were generated in the south china sea. the first and the last cyclones that landed in guangdong coastal area each year were selected and analyzed in this study. for the first cyclones every year during the 68 years, the earliest time is april 19, which occurred in no. 1 cyclone in 2008, while the latest time is october 6, which is the no. 13 321temporal-spatial characteristics and path analysis of maritime cyclones in guangdong coastal areas in the south china sea figure 1. study area and section classification. figure 2. times of tropical cyclones that landing guangdong each yea figure 3. intensity distributions of tropical cyclones 322 cheng chen, lei wang, ruozhou chen, fangliang xing, jun chen cyclone in 1975. in general, the first cyclones that landed in guangdong coastal area every year occurred in june or july. for the last cyclones every year during the 68 years, the earliest time is june 6, which occurred in no. 5 cyclone in 1955, while the latest time is december 2, which is the no. 27 cyclone in 1974. in general, the last cyclones that landed in guangdong coastal area every year occurred in september, october or november. 3.4. moving directions when the tropical cyclone is moving into the guangdong coastal area, it presents a general moving direction. table 1 listed the moving direction distributions of the tropical cyclones. from the table, it can be seen that in the past 68 years there were mainly three moving directions when the cyclones entered the study area, north, northwest and west, and the northwest direction was the most common direction, which took up half of the total. it is relatively rare for the moving directions of southwest and south because their total number was only four. the four of them are as follows: (1) no. 18 tropical cyclone in 1952, moving from fujian province to meizhou city via guangzhou city, guangdong province. after that, it was heading south and finally entered the south china sea. (2) no. 14 tropical cyclone in 1966, moving from fujian province to shantou city. after that, it was heading south and finally entered the south china sea. (3) no. 3 tropical cyclone in 1999, landing between shantou city and shanwei city, heading south along the coast of guangdong province, turning to the southwest near yangjiang city, and then entering to guangxi province. (4) no. 7 tropical cyclone in 2009, landing near yangjiang city, heading southwest to zhanjiang city, and then moving towards southwest to hainan province. the four cyclones above were unique. also, there was only one cyclone that ran towards the south for 68 years, which only account for 0.5 %. it should be noted that there were no cyclones that moved towards east or southeast during the 68 years. table 1. moving directions of tropical cyclones in guangdong coastal areas moving direction frequency of occurrence percentage (%) north 38 20.8 northeast 8 4.4 south 1 0.5 southwest 3 1.6 west 49 26.8 northwest 84 45.9 3.5. spatial characteristics of landing sections the coastal line of guangdong province that was attacked by the tropical cyclones is divided into six sections (shown as figure 1). table 2 lists the spatial characteristics of landing locations. results show that the tropical cyclones most frequently attacked the section b-c (from zhanjiang city to yangjiang city), followed by c-d (from yangjiang city to shenzhen city), and these two sections are subjected to 43% cyclones in total. therefore, the coastal line from zhanjiang city to shenzhen city, which is the most developed area in guangdong province, is most vulnerable to the tropical cyclones. to be more specific, during 68 years 30 tropical cyclones were landing from xuwen county to zhanjiang city, with an annual average of 0.44. forty-three tropical cyclones fell from zhanjiang city to yangjiang city, with a yearly average of 0.63. forty-one tropical cyclones landed from yangjiang city to shenzhen city, with an annual average of 0.60. thirty-four tropical cyclones fell from shenzhen city to shanwei city, with a yearly average of 0.50. there were 27 tropical cyclones landing from shanwei city to shantou city, with an annual average of 0.40. and there were eight tropical cyclones landing from shantou city to raoping county, with a yearly average of 0.12. for the first landing month in each year, the most common month is may; the earliest month is april, and the latest month is june. for the last landing month in each year, the most common months are october and november; the earliest month is september, and the latest month is december. the period of the first landing month to the last landing month decreases from the center of the guangdong coastal line to the east and west sides (decreasing from 9 months to 5 months). for the moving direction, most of them are west and northwest. notably, the moving direction is north for the sections c-d and f-g. for the intensity of the tropical cyclones, most of them are typhoons. specifically, the strongest intensity of the tropical cyclones occurred between the yangjiang city and shenzhen city (the occurrence frequency of the typhoon is 25 with the annual average of 0.37). the occurrence frequency of typhoon decreases from the coastal center line to east and west sides (dropping from 25 to 5). the generation locations of the typhoon landing in guangdong sea areas are also studied. the tropical cyclones affecting the guangdong coastal area mainly generated from northwest pacific. the cyclones that produced from the south china sea tend to land in the west of the guangdong coastal line, i.e., xuwen county to shenzhen city (section a–d), while the cyclones that generated from northwest pacific tend to land in the east of the guangdong coastal line, i.e., shenzhen city to raoping county (section d–g). it can be concluded that the landing location of the cyclones is likely influenced by its generation location. also, according to the data of the cyclones from 1949 to 2016, the first cyclones every year is likely to attack yangjiang city to shenzheng city (section c–d). however, the last cyclones every year is expected to attack zhanjiang city to yangjiang city (section b – c). table 2. spatial distribution of landing sections for tropical cyclones landing section times first landing month latest landing month direction intensity generation location td ts sts ty south china sea northwest pacific a b 30 june october west / northwest 1 4 5 20 9 21 b c 43 june november west / northwest 3 6 11 23 17 26 c d 41 april december north / northwest 2 5 9 25 10 31 d e 34 may november west / northwest 1 4 5 24 7 27 e f 27 may october west / northwest 0 2 8 17 4 23 f g 8 may september north 0 2 1 5 2 6 note: location a, b, c, d, e, f, g are shown in figure 1; td tropical depression; ts tropical storm; sts strong tropical storm; ty typhoon 323temporal-spatial characteristics and path analysis of maritime cyclones in guangdong coastal areas in the south china sea 3.6. month distribution of cyclones in the coastal area although the occurrence time of tropical cyclones attacking guangdong province is not evenly distributed among each year, it demonstrates certain regularity in the distribution of the months. table 3 lists the monthly distribution of tropical cyclone landings guangdong coastal area. results show that the generation time of the tropical cyclones is from april to december. most cyclones are generated in july, and there are more than 70% of cyclones generated in three months (july, august, and september), which indicates the concentricity of the tropical cyclones in the generation time. to be more specific, during 68 years, there were 49 tropical cyclones striking guangdong coastal area in july, with an annual average of 0.72; there were 40 tropical cyclones attacking guangdong coastal area in august, with a yearly average of 0.59; and there were 42 tropical cyclones attacking guangdong coastal area in august, with an annual average of 0.62. variations of the landing locations are also observed for each generation month. specifically, the most frequent landing location for july is in between shanwei city to shantou city (in section e-f). the most frequent landing locations for june and august are in between zhanjiang city to yangjiang city (in part b-c). and the most frequent landing locations for the rest of the months are in between yangjiang city to shenzhen city (in section c-d). the results above suggest that the landing locations may be influenced by the generation month. also, the main moving directions for each month are north, northwest, and west. for the intensity, the strongest tropical cyclones tend to occur in september (29 typhoons, with an annual average of 0.43), followed by july and august. for the generation location, the tropical cyclones that generated in april and may are likely produced in the south china sea, while the tropical cyclones that generated in other months are probably caused in the northwest pacific. it was found that in the earlier time each year, guangdong coastal areas are vulnerable to the tropical cyclones generated in the south china sea; while in the latter time each year, guangdong coastal areas are vulnerable to the tropical cyclones generated in the northwest pacific. 4. discussion 4.1. analysis procedure of moving characteristics this study collected the data of the tropical cyclones in guangdong coastal areas during 68 years (from 1949 to 2016) and selected all on the tropical cyclone path of them in 6 hours interval (in latitude and longitude). therefore, a total of 2040 path points were analyzed. also, to analyze the tropical cyclones that occurred in summer (the peak period), 511 records, 467 records, and 502 records for july, august and september were also analyzed, respectively. the moving path characteristics on the sea were analyzed according to the statistical results. the analysis procedure is as follows: (1) the most frequent landing location was marked as the constraint condition for a certain month, and the middle point of the section in guangdong coast was taken as the landing location of the specific path. (2) the trend line for a particular month was fitted from the path points of the tropical cyclone that in the study area. (3) the moving direction for the path was used as the check condition of the path characteristics analysis. 4.2. equations for moving path the center coordinates of the tropical cyclones are used in the trend line fitting. the latitudes are used as independent variables, and the longitudes are used as dependent variables. the least squares approximation method is used to propose the functions for the tropical cyclone paths in this study area, so the moving path characteristics are described. the tropical cyclone paths in the peak period months (july, august, and september) are shown in figure 4(a), figure 4(b), and figure 4(c), respectively. the equations are described as follows: the traditional path for july is: 0 3843 87 879 5000 0 115 1202. . ,x x y x− + + = ( )   (1) the traditional path for august is: 0 2561 50 0 110 8 120. , .x y x+ − = ( )   (2) the traditional path for september is: 1 1294 150 0 110 8 120. , .x y x+ − = ( )   (3) in which, is the longitude (in °e), y is the latitude (in °n). based on the frequency analysis of the path of all tropical cyclones, a dense area of path points can be obtained, which falls between the upper boundary and the lower boundary, as shown in figure 4 (d). the upper boundary is: 0 55 85 0 112 120. ,x y x+ − = ( )   (4) the lower boundary is: 0 27 50 0 109 2 120. , .x y x+ − = ( )   (5) in which, x is the longitude (in °e), y is the latitude (in °n). 5. conclusions this study analyzes the tropical cyclones that affect the guangdong coastal areas in the south china sea. the tropical cyclone data from 1949 table 3. temporal distribution of landing month for tropical cyclones month times moving path intensity generation location most common landing location main direction td ts sts ty south china sea northwest pacific april 1 c d west 0 0 0 1 1 0 may 6 c d west 0 0 1 5 5 1 june 25 b c northwest 2 6 4 13 10 15 july 49 e f northwest 4 4 14 27 8 41 august 40 b c northwest 0 6 12 22 14 26 september 42 c d west 1 6 6 29 9 33 october 15 c d north 0 1 1 13 2 13 november 4 c d northwest 0 0 1 3 0 4 december 1 c d north 0 0 0 1 0 1 note: location a, b, c, d, e, f, g are shown in figure 1; td tropical depression; ts tropical storm; sts strong tropical storm; ty typhoon 324 cheng chen, lei wang, ruozhou chen, fangliang xing, jun chen to 2016 were collected. key characteristics of the tropical cyclones on the sea were investigated, such as the inter-annual variation, the intensity of tropical cyclones, the generating source and time, and the path directions. the results are summarized as follows: 1) there were 183 tropical cyclones landed in guangdong province in the past 68 years, with an average of 2.7 each year, which more than 60 percent was a typhoon. 2) most of the tropical cyclones were generated in the northwest pacific, spanning from april to december. the path directions were mainly north, northwest, and west. 3) the strengths of the tropical cyclones were reduced from central guangdong coast to both sides, and the section of zhanjiang to shenzhen was the most vulnerable to tropical cyclones. 4) tropical cyclones that generated in the south china sea and the pacific northwest tend to land in the west and east of the guangdong coastline, respectively. 5) cyclones frequently occurred from july to september and became strongest in september. 6) there are a most common landing section and path direction for each month. finally, the temporal-spatial characteristics were investigated, and the functions for describing moving path characteristics are proposed based on the data. acknowledgments the authors gratefully acknowledge the support of the national natural science foundation of china (grant no. 51809047) and the youth teacher education scientific research project in the education department of fujian province, china (grant no. jt180049). references emanuel, k. (2005). increasing destructiveness of tropical cyclones over the past 30 years. nature, 436(7051), 686. jia, x., lu, c. t., lu, j., & zhou, z. p. (2010). statistical characteristics of typhoons affecting china coast and numerical simulation of typhoon waves. journal of waterway and harbor, 31(5), 433-436. li, y., chen, l. s., & zhang, s. j. (2004). statistical characteristics of tropical cyclone making landfalls on china. journal of tropical meteorology, 20(1), 14-23. liou, c. s., & elsberry, r. l. (1987). heat budgets of analyses and forecasts of an explosively deepening maritime cyclone. monthly weather review, 115 (9), 1809-1824. liu, y. c., chen, h. f., liu, x., & chang, y. p. (2017). insight into tropical cyclone behaviour through examining maritime disasters over the past 1000 years based on the dynastic histories of china–a dedication to ocean researcher v. quaternary international, 440, 72-81. figure 4. typical paths for tropical cyclones in the guangdong coastal areas: (a) for july; (b) for august; (c) for september; (d) upper and lower boundaries for overall months 325temporal-spatial characteristics and path analysis of maritime cyclones in guangdong coastal areas in the south china sea rezaee, s., pelot, r., & finnis, j. (2016). the effect of extratropical cyclone weather conditions on fishing vessel incidents’ severity level in atlantic canada. safety science, 85, 33-40. tang, x. c., liu, h. p., pan, a. d., liang, b. q., li, y. t., & wang, t. m. (2003). analysis of disastrous features of landing typhoon in coastal regions of guangdong province in recent 50 years. scientia geographica sinica, 23(2), 182-187. tucker, c. s., trepanier, j. c., harley, g. l., & delong, k. l. (2018). recording tropical cyclone activity from 1909 to 2014 along the northern gulf of mexico using maritime slash pine trees (pinus elliottii var. elliottii engelm.). journal of coastal research, 34(2), 328-340. wen, y. (2002). the characteristics of typhoon disaster and disaster countermeasures in guangdong province. journal of catastrophology, 17(3), 54-59. williams, m. j., coles, r., & primavera, j. h. (2007). a lesson from cyclone larry: an untold story of the success of good coastal planning. estuarine, coastal and shelf science, 71(3-4), 364-367. xu, f. m., huang, y. f., & song, z. y. (2008). numerical simulation of typhoondriven-waves from the east china sea to yangtze estuary. chinese journal of hydrodynamics, 23(6), 604-611. yuan, f. c., chen, m. n., & chen, d. w. (2010). changes of tropical cyclones in china and fujian in recent 60 years. marine forecasts, 27(5), 3439. yuan, j. j., ding, z. y., & wang, l. (2011). a statistical study and composite analysis on the characteristics of the extratropical transition of landfall typhoons during 1949-2007. journal of tropical meteorology, 27(4), 529-541. zhang, w. l., & cui, x. p. (2013). review of the studies on tropical cyclone genesis. journal of tropical meteorology, 29(2), 337-346. zhao, h. j., song, z. y., xu, f. m., & cheng, c. (2010). numerical simulation of typhoon waves in the south china sea – a case study of typhoon chanchu. the ocean engineering, 28(3), 128-134. zheng, c. w., zhou, l., song, s., & pan, j. (2014). simulation of the wave field caused by no. 1307 typhoon “soulik.” journal of xiamen university (natural science), 53(2), 257-262. letter from the editor it's time to present our second volume of 2017. the result of a great effort to put into your hands some works of the best quality. with this selection of articles, an invitation to the colombian geology congress. as every two years, in a city strategically picked up to show the best of the colombian and latinamerican geology, the sociedad colombiana de geologia entity (colombian geology society) presents the xvi congress of geology and the third explorers symposium. the venue is santa marta, a touristic city in the caribbean. register for technical courses and field visits are still open. the event is scheduled august 28 until september 1, 2017. recent months, it has been confirmed in colombia a gas reservoir with at least 520 meters of netgross at the kronos well, which suggests a substantial increase in the gas reserves of the country. sure, a subject to be analyzed during the congress. the discovery, which left the first results in 2015, is proved in a depth of 4,795 meters, including a 1,835-meter water table. the state-managed company ecopetrol holds a 50% share in this block, according to news releases by them, as the 50% remaining is of anadarko, an operator company. these figures are on the same way that the energy reserves around the world, where new techniques and exploration have reached peaks to cover the current demand for oil and related products for decades. this growth of reserves in the world goes hand in hand with the investment. ecopetrol, for example, announced the increase of its exploration budget this year to us$ 650 million with the final purpose of intensifying the search for new hydrocarbon reserves. but mining in the earth would be not the unique perspective to analyze. goldman sachs wrote a recent note explaining that “space mining could be more realistic than perceived,” according to an interesting article published by the washington post on 28 april 2017. the report analyzes the possibilities and bets the mining out of the earth is a decade away. for the next year, the nasa plans to vacuum materials from a 2,000-foot-wide asteroid called bennu; sure it would be the cornerstone. thanks for your attention to this number of the earth sciences research journal and the little reflections proposed in this letter. with the best regards, carlos alberto vargas jimenez editor in chief l e t t e r f r o m t h e e d it o r earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017) v16n1c.indb earth sciences research journal geotechnical analysis earth sci res sj vol 16, no 1 (june, 2012): 65 74 multicriteria decision-making analysis based methodology for predicting carbonate rocks’ uniaxial compressive strength hakan ersoy and derya kanik karadeniz technical university, geological engineering department, 61080, trabzon, turkey, phone: + 90 462 377 35 06, fax: +90 462 325 7405 e-mail: blavetirraa@hotmail com; kanik@ktu edu tr abstract uniaxial compressive strength (ucs) deals with materials’ to ability to withstand axially-directed pushing forces and especially considered to be rock materials’ most important mechanical properties however, the ucs test is an expensive, very time-consuming test to perform in the laboratory and requires high-quality core samples having regular geometry empirical equations were thus proposed for predicting ucs as a function of rocks’ index properties analytical hierarchy process and multiple regression analysis based methodology were used (as opposed to traditional linear regression methods) on data-sets obtained from carbonate rocks in ne turkey limestone samples ranging from devonian to late cretaceous ages were chosen; travertine-onyx samples were selected from morphological environments considering their surface environmental conditions test results from experiments carried out on about 250 carbonate rock samples were used in deriving the model while the hierarchy model focused on determining the most important index properties affecting on ucs, regression analysis established meaningful relationships between ucs and index properties; 0 85 and 0 83 positive coefficient correlations between the variables were determined by regression analysis the methodology provided an appropriate alternative to quantitative estimation of ucs and avoided the need for tedious and time consuming laboratory testing resumen la  resistencia a la compresión uniaxial  (rcu) trata con la capacidad de los materiales para soportar fuerzas empujantes dirigidas axialmente y, especialmente, es considerada ser uno de las más importantes propiedades mecánicas de los materiales rocosos sin embargo, una prueba de rcu es costosa, lleva mucho tiempo para hacerlo en el laboratorio y  requiere muestras de núcleos de alta calidad que tienen una geometría regular por lo tanto, ecuaciones empíricas fueron propuestas para la predicción de rcu como una función de las propiedades índice de las rocas las metodologías de proceso analítico jerárquico (paj) y análisis de regresión múltiple fueron utilizados (en vez de los métodos tradicionales de regresión lineal) en conjuntos de datos obtenidos de las rocas carbonatadas en el noreste de turquía muestras de rocas calizas que van desde el devónico hasta finales del cretácico fueron escogidas;  muestras de travertino y  ónix fueron seleccionadas de ambientes morfológicos teniendo en cuenta sus  condiciones ambientales de superficie los resultados de los experimentos llevados a cabo en alrededor de 250 muestras de rocas carbonatadas fueron utilizados para derivar un modelo mientras que el modelo de jerarquía se centró en determinar las propiedades índice más importantes afectados por la rcu, el análisis de regresión establece relaciones significativas entre la rcu y las propiedades del índice; coeficientes de correlación positivas de 0,85 y 0,83 fueron determinadas por análisis de regresión entre las variables la metodología proporciona una alternativa adecuada para la estimación cuantitativa de la rcu y evita la necesidad de realizar pruebas del laboratorio las cuales son tediosas y dispendiosas palabras claves: resistencia a la compresión, ecuaciones empíricas, rocas carbonatadas, jerarquía analítica keywords: compressive strength, empirical equations, carbonate rock, analytical hierarchy record manuscript received: 06/02/2012 accepted for publications: 01/06/2012 hakan ersoy and derya kanik66 introduction intact rocks’ uniaxial compressive strength (ucs) is the main parameter used for almost all rock mechanical studies in most civil, geological and mining projects (bieniawski 1974; cargill and shakoor 1990); however, regular geometry, high-quality core samples are necessary for determining ucs standard cores cannot always be extracted from weak, highly-fractured, thinly-bedded, foliated rocks this test is thus expensive, time-consuming and requires well-prepared rock samples resent trends in estimating ucs from simple laboratory index tests has been improved to overcome such difficulties, and simple prediction models have become attractive for engineering geologists using non-destructive and easily-applied techniques such as rocks’ ultrasonic wave velocity (uwv) and index properties many attempts have been made to predict intact rocks’ ucs (kahraman 2001; katz et al., 2000; koncagul and santi 1999; chau and wong 1996) some researchers have found that sound velocity is closely related to rock properties (gaviglio 1989; chang et al., 2006; yalçınalp et al., 2008; babacan et al., 2009; moradian and behnia, 2009) whilst other have correlated ucs with index properties such as porosity, density and uwv (ramana and venkatanarayana 1973; yasar and erdogan 2004; kanik 2010) simple statistical methods-based multiple regression techniques have been used to establish predictive models; new techniques such as artificial neural networks and fuzzy inference systems have been used for developing predictive models for estimating the required parameters during recent years (grima and babuska 1999; kayabasi et al., 2002, gokceoglu and zorlu 2004; sonmez et al., 2003; karakus and tutmez 2006; dehfigure 1. map showing the location of the area being studied multicriteria decision-making analysis based methodology for predicting carbonate rocks’ uniaxial compressive strength 67 ghan et al., 2010, yagiz et al., 2011) different evaluation methods have been developed to establish predictive models; such methods include linear vector approach, matrix method, fuzzy set theory, checklist methods, parametric ranking methods and multi criteria decision analysis based methodologies analytic hierarchy process (ahp), was developed as a type of multi-criteria analysis, to standardise multi-criteria decision-making (saaty 1980) however, ahp uses a quantitative comparison method based on pair-wise comparisons of decision-making criteria, rather than utility and weighting functions many engineering geological projects have adopted such as approach (cook et al., 1984; siddiqui et al., 1996; ersoy and bulut 2009) the paper presents multi-criteria decision-making and multiple regression analysis-based methodology for predicting ucs regarding carbonate rocks’ index properties the model is based on non-destructive and relatively easy to apply laboratory tests research methodology sampling location and geology the morphology of mountainous region being studied in north-eastern turkey (figure 1) consist of rough, irregular land having steep slopes and peaks reflecting the eastern black sea region’s geology and tectonic features the main morphological units (mainly faults and folds) have been shaped by structural elements in the region trending ne-sw the region is drained by the harşit and çoruh rivers, forming the most significant fluvial system deep incision forms like v-shaped valleys are characterised by deep gorges and steep slopes in these drainage systems turkey is located in the alpine-himalayan orogenic belt having the world’s richest natural stone formations; it is the country having 5 billion tons of reserves and almost 35% of the world’s natural stone reserves the eastern black sea region has rich potential in terms of the variety of carbonatebearing rocks the region has around 450 million tons natural stone reserves almost 70% of production today consists of travertine and limestone, the remaining 30% being volcanic rocks the rock samples for this study were collected from 5000 km2 in the southern zone of the eastern pontides (ne turkey) the eastern pontide belt is a major metallogenetic province in the eastern black sea coastal region and forms a 500 km long and 100 km wide mountain chain along the black sea coast the eastern pontides may be subdivided into northern and eastern zones on structural and lithological differences (özsayar et al., 1981; okay and sahinturk 1997) the northern zone is dominated by late cretaceous and middle eocene volcanic and volcaniclastic rocks, whereas pre-late cretaceous sedimentary rocks are widely exposed in the southern zone (arslan et al., 1997; eyuboglu 2006; sen 2007) (figure 2) the limestone samples were selected from pre-late cretaceous platform carbonates which are widely exposed in the area and liassic-aged formations; travertine and onyx samples were selected from the gümüşhane, bayburt and giresun area three rock types from ten rock formations were sampled and tested for this study table 1 gives the rock type, age, location and description of the samples sample preparation and laboratory tests several carbonate rock samples were collected in this study to determine physical and strength properties seven oriented block samples for each group of carbonates were collected from the field for laboratory testing; each block was represented by at least ten core specimens 225 nxsize core specimens having a 2 5:1 length to diameter ratio were prepared from the block samples rock samples were carefully inspected before testing, to obtain the most representative stone samples for performing laboratory tests figure 2. main tectonic units and zones in the eastern pontides naf: north anatolian fault; neaf: northeast anatolian fault (eyüboğlu 2006) hakan ersoy and derya kanik68 uwv, unit weight, water absorption and content are the rock materials’ most important index properties and they are often related to porosity porosity is the ratio of the non-solid volume to the total volume of material and it also describes how densely the material is packed in the study, firstly unit weight, water content, apparent porosity and water absorption by weight were determined with respect to the description criteria of isrm (1981) a pundit plus ultra-sonic pulse (usp) instrument giving more precise rock sample measurements and two 54 khz transducers having piezoelectric properties were used in this study to calculate ultrasonic longitudinal wave velocity the transducers were located parallel to the surface of the sample, transit time was measured and ultrasonic wave velocities were calculated from transit time ucs was determined for the study according to isrm (1981) description criteria core surface flatness was supplied for the ucs test to obtain an even load distribution; specimens were loaded axially up to failure establishing decision-making methods regression analysis includes any techniques for modelling and analysing several variables when the focus is on the relationship between a dependent variable and one or more independent variables (freedman 2005) multiple regression is aimed at learning more about the relationship between several independent or predictable variables and a dependent or criterion (davis 1986) regression analysis was used in this with analytic hierarchy process (ahp), a widely accepted decision making method constraints are compared to each other in ahp, to designate each variable’s relative importance in accomplishing an overall goal numerical values were assigned to each pair of constraints using established guidelines and a constrained matrix is built the sum of each column within the matrix was then normalised and weighting was calculated mathematic formulation (simple additive weighting) was defined following an equation for calculating final grading values in multiple criteria problems (saaty 1980); where, v(y) was area’s suitability index, w was a criterion’s weighting or importance factor of a, y was a criterion’s degree or compliance level, i was the criterion number and q was the number of criteria pairwise comparisons were used to determine each criterion’s relative importance; ahp is based on such approach decision makers can quantify their opinions about the criteria magnitude by using a verbal scale when comparing pairs of criteria the pairwise comparison matrix (pcm) constructed by decision-makers in the previous had to have the following attributes; w ij = w ji -1 the next step was to calculate the criteria’s relative importance weighting implied by previous comparisons saaty (1980) proposed estimating pcm’s right principal eigenvector which can be approximated using the geometric mean for each row of the pcm (by multiplying the elements in each row and then taking the nth root, where n is the number of criteria) this mode is known as multiplicative ahp (saaty and millet 2000) and was used in the present work the calculated geometric means were then normalised and relative importance weighting extracted (2) table 1. location and general properties of the carbonate rocks rock type rock description age location 1a. travertine (28 samples) porous and soft white-pale cream in colour quaternary bahçecik (gümüshane) 1b. travertine (25 samples) porous and soft pale cream in colour quaternary kalecik (gümüshane) 1c. travertine (25 samples) porous and soft white-pale cream in colour quaternary çamoluk (giresun) 1d. travertine (10 samples) compact and soft white-pale cream in colour quaternary çamoluk (giresun) 1e. travertine (13 samples) compact and soft white-pale cream in colour quaternary gölova (giresun) 2a. onyx (23 samples) compact and soft white-pale grey and cream in colour contained cream colored band quaternary yaylapınar (bayburt) 2b. onyx (15 samples) compact and soft light green, pale cream in colour contained green colored band quaternary yaylapınar (bayburt) 3a. limestone (20 samples) hard and compact light grey in colour contained some micro fossils l. cretaceous dogger-malm kelkit (gümüshane) 3b. limestone (24 samples) fractured and slightly metamorphosed reddish in colour contained a number of fossils liassic esenyurt (gümüshane) 3c. limestone (20 samples) hard and compact light grey and white in colour cretaceous jurassic kemah (erzincan) 3d. limestone (20 samples) hard and slightly metamorphosed light purple and white in colour contained light brown colored band cretaceous devonian tercan (erzincan) v(y)= w j y ij q ∑ j-1 (1) multicriteria decision-making analysis based methodology for predicting carbonate rocks’ uniaxial compressive strength 69 table 2. water absorption, water content and apparent porosity values of the rock groups studied table 3. unit weight, p wave velocity and ucs values for the rock groups being studied sample location and type water content (%) water absorption by weight (%) apparent porosity (%) max min sd mean max min sd mean max min sd mean kalecik travertine (1a: 30 samples) 0.20 0.08 0.04 0.13 3.33 0.88 0.73 1.39 7.37 2.16 1.36 3.29 bahçecik travertine (1b: 15 samples) 0.77 0.12 0.15 0.34 1.20 0.48 0.24 0.77 2.88 1.18 0.57 1.90 çamoluk travertine (1c: 25 samples) 2.48 0.49 0.43 1.40 2.90 0.98 0.48 1.88 6.65 2.37 1.01 4.28 çamoluk (2) (1d: 10 samples) 0.62 0.45 0.06 0.59 1.30 0.91 0.12 1.07 3.03 2.44 0.28 2.61 gölova travertine (1e: 13 samples) 0.31 0.18 0.04 0.23 0.62 0.37 0.063 0.47 1.55 0.94 0.15 1.19 bayburt onyx (2a and b: 21) 0.79 0.01 0.18 0.07 0.72 0.02 0.02 0.10 1.95 0.06 0.32 0.26 kelkit limestone (3a: 12 samples) 0.06 0.03 0.01 0.05 0.18 0.13 0.01 0.15 0.48 0.34 0.04 0.40 esenyurt limestone (3b: 15 samples) 0.46 0.04 0.09 0.14 0.31 0.06 0.07 0.19 0.08 0.16 0.19 0.51 kemah limestone (3c: 21 samples) 0.14 0.01 0.03 0.03 0.07 0.02 0.01 0.04 0.18 0.05 0.03 0.11 tercan limestone (3d: 25 samples) 0.09 0.01 0.01 0.06 0.12 0.04 0.02 0.09 0.31 0.12 0.05 0.24 sample location and type unit weight (kn/m3) velocity (m/s) p wave ucs (mpa) max min sd mean max min sd mean max min sd mean kalecik travertine (1a: 30 samples) 24.7 22.2 0.8 24.0 4541 3813 220 4050 65 34 8 53 bahçecik travertine (1b: 15 samples) 25.1 24.1 0.3 24.6 5072 4091 297 4598 44 25 5 37 çamoluk travertine (1c: 25 samples) 24.6 24.1 0.2 24.4 3693 3506 61 3590 44 30 5 38 çamoluk (2) travertine (1d: 10 samples) 25.4 24.9 0.2 25.2 4727 3913 268 4323 43 39 1.2 41 gölova travertine (1e: 13 samples) 29.2 22.2 0.9 25.0 5072 3813 374 4430 66 20 10 40 bayburt onyx (2a and b: 21 samples) 27.4 26.5 0.3 27.1 6800 4121 805 5240 70 20 13 45 kelkit limestone (3a: 12 samples) 27.1 26.8 0.1 27.0 6321 5988 98 6200 154 30 38 80 esenyurt limestone (3b: 15 samples) 27.4 26.6 0.2 27.0 6103 5562 169 5760 117 40 24 70 kemah limestone (3c: 21 samples) 27.1 26.8 0.1 27.0 6353 6127 59 6250 162 25 39 90 tercan limestone (3d: 25 samples) 27.3 25.3 0.6 26.8 6057 4327 371 5414 101 79 9 85 hakan ersoy and derya kanik70 integrating site selection criteria was based on multi-criteria assessment methods (eastmen, 1995): = ∑ (sk*wk) where, sk was the land’s suitability for landfill for objective k (priority groups), (fi) k was factor i (discriminating features) for objective k, (wi)k was the weighting for factor i (score given experimental studies) for objective k, (rj) k was constraint j for objective k (value 0 or 1), s was multi-objective suitability and wk was the weighting for objective k results the rocks’ index and strength properties ucs is an expensive and very time-consuming test the core surfaces have to be ground down to make them parallel at a specified tolerance and a high load capacity loading frame is usually required thus, in the study, ten groups of carbonate rock samples were collected in the southern zone of the eastern pontides (ne turkey) to establish some relationships between ucs and index properties in the study the index and strength properties of the rocks were determined in accordance with isrm (1981) the results of the laboratory studies are given table 2 and table 3 when taking into consideration the mean values of the water content, water absorption by weight, apparent porosity, p wave velocity and ucs value data, the highest values were observed for erzincan limestone and the lowest values were observed for gümüşhane travertines however, the values of the ucs were obtained between 20 and 66 mpa for travertines, 20 and 70 mpa for onyx and 25 and 162 for different limestones depending on the mineral shapes, weathering features and occurrence of the microcracks, the ucs values varied in a wide range as found in the study empirical relationship between index and strength properties rock materials’ uniaxial compressive strength greatly depends on their index properties such as porosity, water content, water absorption and mineral composition p wave velocity and porosity are the most preferred properties for predicting ucs amongst such properties (sachpazis 1990; tugrul and zarif 1999; palchik 1999) regression analysis is usually preferred for establishing statistical relationships between different variables relationship intensity of between variables is defined by regression values and correlation coefficients; explanatory variables must thus be weighted regarding dependent variables before mulp w av e ve lo ci ty ( m /s ) 6500 r2=0 70 6000 5500 5000 4500 4000 3500 0 50 100 ucs (mpa) a p p ea ra n ce p or os it y (% ) 5 r2=0 45 4 3 2 1 0 0 50 100 ucs (mpa) w at er a bs ro rb ti on b y w ei gh t (% ) 2 0 r2=0 45 1 5 1 0 0 5 0 0 50 100 ucs (mpa) u n it w ei gh t (g r/ cm ) 2 8 r2=0 40 2 7 2 6 2 5 2 4 0 50 100 ucs (mpa) w at er c on te n t (% ) 1 5 r2=0 32 1 0 0 5 0 0 50 100 ucs (mpa) figure 3. the empirical relationship between ucs and index parameters for different carbonate rock groups table 4. pairwise comparison matrix and relative importance weighting for index properties a b c d e eigenvector weighting a 1 2 3 3 4 3.23 0.45 b 1/2 1 2 3 3 2.34 0.33 c 1/3 1/2 1 2 3 0.73 0.10 d 1/3 1/3 1/2 1 2 0.60 0.08 e 1/4 1/3 1/3 1/2 1 0.30 0.04 a: ultrasonic wave velocity, b: apparent porosity, c: water absorption by weight, d: water content, e: unit weight 1: equal importance, 2: weak or slight (two parameters contributed equally to the objective) 3: moderate importance, 4: moderate plus experience and/or analysis slightly favoured one parameter over another s k = ( ) k ∑ fi*wi i s ∏ rj j (3) (4) multicriteria decision-making analysis based methodology for predicting carbonate rocks’ uniaxial compressive strength 71 figure 4. the empirical relationship between ucs and vp/n for different travertines and onyx (a: bayburt onyx, b: bahçecik and kalecik, c: çamoluk and d: gölova travertines) figure 5. the empirical relationship between ucs and vp/n for the different limestones (a: esenyurt, b: kelkit, c: tercan and d: kemah) u c s (m p a) 80 70 60 50 40 30 20 10 0 25000 50000 75000 100000 vp/n u c s (m p a) 160 140 120 100 80 60 40 20 14000 15000 16000 17000 18000 vp/n u c s (m p a) 60 55 50 45 40 35 30 25 2600 3000 3400 3800 4000 vp/n u c s (m p a) 70 60 50 40 30 20 10 500 1500 2500 3500 4500 5500 vp/n u c s (m p a) 160 120 60 40 30000 50000 70000 90000 110000 130000 vp/n u c s (m p a) 85 75 65 55 45 10000 20000 30000 40000 50000 vp/n u c s (m p a) 110 90 70 50 30 6000 10000 14000 18000 22000 vp/n u c s (m p a) 50 40 30 20 10 500 750 1000 1250 1500 1750 vp/n ucs = 14 in (vp/n)-96 r2=0 83 n: 20 ucs = 0 0184 (vp/n)-22 r2=0 80 n: 12 ucs = 0 0049 (vp/n)+12 r2=0 84 n: 14 ucs = 0 001 (vp/n)+30 r2=0 71 n: 25 ucs = 0 001 (vp/n)+26 r2=0 81 n: 21 ucs = 0 0032 (vp/n)-427 r2=0 86 n: 12 ucs = 0 022 (vp/n)+0 r2=0 84 n: 33 ucs = 19 5 in (vp/n)-107 r2=0 87 n: 35 (a) (a) (c) (c) (b) (b) (d) (d) hakan ersoy and derya kanik72 rock description formulas r2 onyx (21 samples) ucs= 14 ln (vp/n) 96 83 travertine (76 samples) ucs= 15 ln (vp/n) 73 86 limestone (60 samples) ucs= 0.0009 (vp/n) + 38 83 table 5. the formulas used for estimating ucs in this study tiple regression analysis to establish an empirical model ahp used the data obtained from linear regression analysis results concerning ten rock groups (table 1 and figure 3) to decide which material properties should be selected as explanatory variables from rock materials’ index properties this represents a tool which reduces time and the cost involved in many engineering models table 4 lists the priority vectors for all criteria; relative importance weightings are included in the final column of this table ahp parameters are also shown in the table, indicating that the judgments (and therefore final relative importance weightings) seemed to be reasonable the method needed a scale of numbers indicating how many times more important or dominant one element was over another figure 6. the empirical relationship between ucs and vp/n for all travertine (a) and limestone (b) samples figure 7. line-scatter plot diagrams for measured compared to calculated ucs values and histograms showing frequency versus difference in measured and calculated ucs (a and b: travertines, c and d: limestones) u c s (m p a) 60 50 40 30 20 10 0 1000 2000 3000 4000 5000 vp/n u c s (m p a) 140 120 100 80 60 40 20 0 0 20000 40000 60000 80000 100000 vp/n ucs = 15 in (vp/n)-73 r2=0 8671 n: 276 ucs = 0 0009 (vp/n)+38 r2=0 83 n: 60 (a) (a) (b) (c) (d) (b) c al cu la te d u c s (m p a) 70 60 50 40 30 20 10 10 20 30 40 50 60 70 measured ucs (mpa) fr eq u en ci e (n u m be r) 25 20 15 10 5 0 -9 -6 -3 0 3 6 9 measured ucs calculated ucs (mpa) fr eq u en ci e (n u m be r) 16 14 12 10 8 6 4 2 0 -16 -12 -8 -4 0 4 8 12 16 measured ucs calculated ucs (mpa) c al cu la te d u c s (m p a) 140 120 100 80 60 40 20 20 40 60 80 100 1200 140 measured ucs (mpa) multicriteria decision-making analysis based methodology for predicting carbonate rocks’ uniaxial compressive strength 73 element regarding the criterion or property to which they were being compared the ahp analysis showed that the most important criteria affecting on ucs for the carbonate rocks being studied were uwv (45% weighting) and apparent porosity (33 % weighting) the other index properties’ weighting did not surpase %10 as regression analysis provides a means of summarising the relationship between variables, multiple regression analysis-based methodology was used to establish some numerical relationships between rock materials’ ultrasonic wave velocity, apparent porosity and ucs furthermore it is known that vp decreases with increased porosity vp and n-1 were thus used in the equations for predicting ucs in this study index properties were considered to be explanatory variables, and the ucs a depending variable figure 4 and 5 show multiple regression analysis results, involving some positive correlations between uwv rate regarding apparent porosity and ucs: such relationships were characterised by 0 71 and 0 87 regression coefficients however, figure 6 indicates meaningful empirical relationships for all travertines and limestones and the relationships are represented by the formulas given in table 5 a goodness of fit test establishes whether an observed frequency distribution differs from a theoretical distribution; however, a model’s suitability is tested using the difference between observed and expected values normal distribution is a continuous probability distribution which is often used as a first approach for describing real-valued random variables tending to cluster around a single mean value a normal distribution would thus be expected in the histograms showing difference between observed and expected values line-scatter plot diagrams of observed-expected values (figure 7a and c) and histograms of difference between observed and expected values (figure 7b and d) were prepared and normal distributions were observed histograms having a normal distribution showed that these equations did fairly well in estimating ucs using vp/n the results showed that the model being tested was suitable for the study conclusions this study involved three types of carbonate rocks being collected from ten rock formations in the north-eastern turkey laboratory studies were conducted in line with isrm (1980) to establish some relationships between p wave velocity, apparent porosity and ucs and ahp-based multiple regression analyses methodology was used for statistically analyzing the suggested methods and results ahp analysis indicated that uwv (45% weighting) and apparent porosity (33% weighting) were the most important index properties affecting on ucs regarding the carbonatebearing rocks being studied multiple regression analysis correlations indicated by the 0 81-0 87 regression coefficient were determined for uwv rate regarding apparent porosity and ucs the equations obtained by such analysis results were practical, simple and accurate enough to apply and may be recommended for use in practice acknowledgments this study was partly supported by the karadeniz technical university scientific research fund (project no: 2008 112 005 10) the authors would like to thank bülent yalçınalp and ali babacan from karadeniz technical university in trabzon (turkey) for field and laboratory work the authors would also like to thank nazmiye yazıcı from güvencem marble limited company for helping during all stages of the study references arslan, m , tuysuz, n , korkmaz, s , kurt, h (1997) geochemistry and petrogenesis of the eastern pontide volcanic rocks, northeast turkey chemie der erde, 57: 157–187 babacan, a e , ersoy, h , gelisli, k (2009) determination of physical and mechanic properties of rocks with direct and indirect methods: a case study on the beige limestones in the eastern pontides, proceedings of the 21st international mining congress and exhibition of turkey, antalya, 123-130 bieniawski, z t (1974) estimating the strength of rock materials journal of the south african institute mining metallurgy, 74: 312-320 cargill, j s , shakoor, a (1990) evaluation of empirical methods for measuring the uniaxial compressive strength international journal of rock mechanic and mining science, 27: 495-503 chang, c , mark, d , zoback, a , abbas, k b (2006) empirical relations between rock strength and physical properties in sedimentary rocks journal of petroleum science and engineering, 51: 223-237 chau, k t , wong, r h c (1996) uniaxial compressive strength and point load strength international journal of rock mechanic and mining science, 33: 183–188 cook, t , falchi, p , mariano, r (1984) an urban allocation model combining time series and analytic hierarchical methods management science, 30 (2): 198-208 davis, j c (1986) statistics and data analysis in geology john wiley and sons, new york dehghan, s , sattari, g h , chelgani, c , aliabadi, m (2010) prediction of uniaxial compressive strength and modulus of elasticity for travertine samples using regression and artificial neural networks mining science and technology, 20: 41-46 ersoy, h , bulut, f (2009) spatial and multi-criteria decision analysisbased methodology for landfill site selection in growing urban regions waste management and research, 27(5):489-500 eyuboglu, y (2006) description and geotectonic important of the alaskan-type ma c–ultrama c rocks in the eastern pontide magmatic arc (ne turkey), phd thesis, karadeniz technical university, trabzon, turkey (unpublihed) freedman, d (2005) statistical models theory and practice london: cambridge university press gaviglio, p (1989) longitudinal waves propagation in a limestone: the relationship between velocity and density rock mechanic and rock engineering, 22: 299-306 gokceoglu, c , zorlu, k (2004) a fuzzy model to predict the uniaxial compressive strength and the modulus of elasticity of a problematic rock engineering applications of artificial intelligent, 17(1): 61–72 grima, a m , babuska, r (1999) fuzzy model for the prediction of unconfined compressive strength of rock samples international journal of rock mechanic and mining science, 22: 339-349 isrm (international society for rock mechanics) suggested methods: rock characterization, testing and monitoring e t brown (ed ), pergamon press, london; 1981 kahraman, s (2001) evaluation of simple methods for assessing the uniaxial compressive strength of rock international journal of rock mechanic and mining science, 38: 981–994 kanik, d (2010) predicting of uniaxial compressive strength of carbonate rocks using simple test methods karadeniz technical university natural sciences institute master thesis, turkey (unpublished) karakus, m , tutmez, b (2006) fuzzy and multiple regressions modelling for evaluation of intact rock strength based on point load, schmidt hammer and sonic velocity rock mechanic and rock engineering, 39(1): 45–57 katz, o , reches, z , roegiers, j c (2000) evaluation of mechanical rock properties using a schmidt hammer international journal of rock mechanic and mining science, 37(4): 723-728 kayabasi, a , gokceoglu, c , ercanoglu, m (2002) estimating the deformation modulus of rock masses: a comparative study international journal of rock mechanic and mining science, 40: 55-63 hakan ersoy and derya kanik74 koncagul, c e , santi, p m (1999) predicting the unconfined compressive strength of the breathitt shale using slake durability, shore hardness and rock structural properties international journal of rock mechanic and mining science, 36: 139-153 moradian, z a , behnia, m (2009) predicting the uniaxial compressive strength and static young’s modulus of intact sedimentary rocks using the ultrasonic test asce international journal of geomechanics, 9(1): 14-19 okay, a i̇, sahinturk, ö (1997) geology of the eastern pontides in: robinson ag (ed), regional and petroleum geology of the black sea and surrounding region, american association of petroleum geologist (aapg) memoir, 68: 291-311 özsayar, t , pelin, s , gedikoglu, a (1981) cretaceous in the eastern pontides, karadeniz technical university journal of earth sciences, 1: 65-74 (in turkish with english abstract) palchik, v (1999) influence of porosity and elastic modulus on uniaxial compressive strength in soft brittle porous sandstones rock mechanics and rock engineering, 32: 303-309 ramana, y v , venkatanarayana , b (1973) laboratory studies on kolar rocks international journal of rock mechanic and mining science, 10: 465-489 saaty, t l (1980) the analytic hierarchy process mcgraw-hill, new york saaty, t , millet, i (2000) on the relativity of relative measures accommodating both rank preservation and rank reversals in the ahp european journal of operational research, 121: 205-212 sachpazis, c i (1990) correlating schmidt hardness with compressive strength and young’s modulus of carbonate rocks bulletin international association of engineering geology, 42: 75-83 siddiqui, m z , everett, j w , vieux, b e (1996) landfill siting using geographic information systems: a demonstration journal of environmental engineering, 122: 515-523 sonmez, h , gokceoglu, c , ulusay, r (2003) an application of fuzzy sets to the geological strength index (gsi) system used in rock engineering engineering applications of artificial intelligent, 16(3): 251-269 sen, c (2007) jurassic volcanism in the eastern pontides turkish journal of earth sciences, 16: 523-539 yalçınalp, b , ersoy, h , fırat ersoy, a (2008) geological and geotechnical properties of bahcecik (gümüshane, ne turkey) travertines. journal of earth science, 32: 24-36 (in turkish with english abstract) tugrul, a , zarif, i h (1999) correlation of mineralogical and textural characteristics with engineering properties of selected granitic rocks from turkey engineering geology, 51: 303-317 yagiz, s , sezer, e a , gokceoglu, c (2011) artificial neural networks and nonlinear regression techniques to assess the influence of slake durability cycles on the prediction of uniaxial compressive strength and modulus of elasticity for carbonate rocks, international journal for numerical and analytical methods in geomechanics, doi: 10 1002/nag 1066 yasar, e , erdogan, y (2004) estimation of rock physicomechanical properties using hardness methods engineering geology, 71: 281–288 given the difficulty of accurate online detection for massive data collecting real-timely in a strong noise environment during the complex geological mineral grade analysis process, an order self-learning arhmm (autoregressive hidden markov model) algorithm is proposed to carry out online outlier detection in the geological mineral grade analysis process. the algorithm utilizes ar model to fit the time series obtained from “online x ray fluorescent mineral analyzer” and makes use of hmm as a basic detection tool, which can avoid the deficiency of presetting the threshold in traditional detection methods. the structure of traditional bdt (brockwell-dahlhaus-trindade) algorithm is improved to be a double iterative structure in which iterative calculation from both time and order is applied respectively to update parameters of arhmm online. with the purpose of reducing the influence of outlier on parameter update of arhmm, the strategies of detection-before-update and detection-based-update are adopted, which also improve the robustness of the algorithm. subsequent simulation by model data and practical application verify the accuracy, robustness, and property of online detection of the algorithm. according to the result, it is obvious that new algorithm proposed in this paper is more suitable for outlier detection of mineral grade analysis data in geology and mineral processing. abstract keywords: arhmm; bdt; kicvc; outlier detection; online detection. online outlier detection for time-varying time series on improved arhmm in geological mineral grade analysis process detección en tiempo real de valores atípicos sobre series de tiempo variable en arhmm mejorado durante el proceso de análisis de grado mineralógico issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n3.65215 jianjun zhaoa,b, junwu zhoubb, weixing suc*, fang liuc a school of information science & engineering, northeastern university, shenyang 110004, china; b beijing general research institute of mining & metallurgy, beijing 100160, china c school of computer science & software engineering, tianjin polytechnic university, tianjin 300387, china *email of corresponding author: 15900201597@163.com existe gran dificultad para la detección en tiempo real para series de datos masivos con altos niveles de ruido de valores atípicos. se propone un algoritmo de autoaprendizaje arhmm (modelo autoregresivo oculto de markov) para llevar a cabo la detección de dichos valores atípicos en el proceso de análisis del grado mineral geológico. el algoritmo usa un modelo ar para ajustar la serie de tiempo obtenida del “analizador de fluorescencia de rayos x” y hace uso del hmm como una herramienta básica de detección, la cual puede evitar la deficiencia de predeterminar el umbral en métodos tradicionales de detección. para actualizar los parámetros del arhmm en tiempo real, la estructura del algoritmo bdt (brockwell-dahlhaus-trindade) tradicional se mejora para ser una doble estructura iterativa en la que se aplica el cálculo iterativo en tiempo y en orden respectivamente. con el propósito de reducir la influencia de valores atípicos (o extremos) en la actualización del parámetro de arhmm, se adoptan las estrategias de detección-antes-que-actualización y la detección-basada-en-actualización, lo que también aumenta la robustez del algoritmo. la subsiguiente simulación por modelos de datos y aplicación práctica comprueba la precisión, fortaleza y capacidad de la detección en línea del algoritmo. de acuerdo con el resultado, es evidente que el nuevo algoritmo propuesto en este artículo es más apropiado para la detección de datos de valores atipicos para el análisis del grado mineral en geología y el procesamiento mineral. resumen palabras clave: modelo autoregresivo oculto de markov; detección en tiempo real; brockwelldahlhaus-trindade. record manuscript received: 21/02/2017 accepted for publication: 28/07/2017 how to cite item zhao, j., zhou, j., su, w., & liu, f. (2017). online outlier detection for time-varying time series on improved arhmm in geological mineral grade analysis process. earth sciences research journal, 21(3), 135-139. doi: http://dx.doi.org/10.15446/esrj.v21n3.65215 g e o st a t is t ic s earth sciences research journal earth sci. res. j. vol. 21, no. 3 (september, 2017): 135 139 136 jianjun zhao, junwu zhou, weixing su, fang liu introduction mineral composition analysis is a key factor in determining whether or not to carry out mining. over the years, many scholars have proposed some new ideas and methods for accurate mineral grade assessment, many of which are based on chemical or physical test equipment to obtain the data for ingredient grade analysis (kameshwara, rao, & narayana, 2014; de'nan, naaim, & leong, 2017). therefore, the accuracy of the data used for ore composition analysis is critical to the ore grade analysis. at present, automated testing equipment is used in ore grade analysis, such as “box-a type on-stream x-ray fluorescence analyzer”, which uses spectral obtain by irradiating x-rays to the pulp to get the results of ore grade. it is worth noting that box-a type on-stream x-ray fluorescence analyzer by default is that the spectral data obtained is correct. but whether it is chemical or physical testing equipment are inevitably produced abnormal data. those outliers directly affect the analysis results of the mineral products analysiser (clarke, & levis, 1998; rivoirard, demange, & freulon, 2013). therefore, the detection and elimination of these abnormal data is the premise and key to the above ore grade analysis work. a new algorithm is proposed here to especially do outlier detection for ore inspection data which obtain from chemical or physical testing equipment. the algorithm utilizes ar model to fit the time series and makes use of hmm as a basic detection tool, which can avoid the deficiency of presetting the threshold in traditional detection methods. to update parameters of arhmm online, the structure of traditional bdt (brockwell-dahlhaus-trindade) algorithm is improved here, and a double iterative structure in which iterative calculation from both time and order is applied respectively. with the purpose of reducing the influence of outlier on parameter update of arhmm, the strategies of detection-before-update and detection-based-update are adopted, which also improve the robustness of the algorithm. subsequent simulation by model data and practical application verify the accuracy, robustness, and property of online detection of the algorithm. in this paper our innovations are shown as follow: 1. unlike other outlier detect method (such as the traditional ar model detection method), the outliers detect method proposed in this paper does not need to set the detection threshold. 2. considering the problem that the model order of chemical or physical testing equipment’s hard to be determined, the new detected method which is based on residual error has the function of model order self-learning. 3. in a view to avoid the influence of outliers on the test results, this paper proposes a detection-before-update and detection-based-update strategies. the predecessors’ achievements on outlier detection many good ideas and methods are put forward for the research of outliers detection problem, such as that barnett and lewis proposed an outliers detected method based on statistics in their word named ‘outlier in statistical data (barnett & lewis, 1994). outlier detection method based on distance is proposed by knorr and raymond (knorr & ng, 1999; edwin & raymond, 1998), an new detected method based on density is suggested by ramaswamy et al. (2000). but for ore inspection data, the detection methods based on distance, density or variance is a lack of feasibility since an online real-time detection method be needed for the ore test data. with the research of anomaly data detection technology, many new ideas and techniques are introduced, such as clustering analysis (almeida & barbosa, 2007) and neural network method (bullen, cornford, & nnbney, 2003; prakobphol, & zhan, 2008). but clustering analysis method is also not suitable for online outlier detection for extensive test data, and neural network method requires a lot of data to model learning. in 1995, ragaran and argrawal put forward the concept of "sequence anomaly" (han & micheline, 2001) and proposed the detection method based on deviation (takeuchi, & yamanishi, 2006). because this method needs to know the order of the model, it can not be directly used for the outlier detection of mineral grade analysis data. structure of double iteration in bdt to make the bdt algorithm can be calculated online, the improved bdt algorithm with double repetition structure is proposed in this paper. traditional bdt algorithm the traditional bdt algorithm is improved by levinson-durbin algorithm which is proposed by brockwell et al. (2002). for traditional bdt algorithm, using all the data to the iterative calculation of model order, in a view to obtain the order of the forward and backward ar model. xt, t =1,2,∙∙∙ is the test data waiting for detection, where xt is m -dimension vectors. so forward ar model can be express as equation 1. (1) in which, ɛk, (t) is forward residual under k order model, which obeys gauss distribution with zero means. ak (i) is the coefficient of forwarding ar model under k order model. so backward ar model can be written as equation 2. (2) with minimizing all data forward and backward residual as the target, so the generalized objective function written as equation 3 (trindade, 2003). (3) n is the number of the data. ω1 and ω2 are weighted coefficient matrix for forwarding and backward ar model, and in bdt algorithm, the values of ω1 and ω2 is 1. ἓk (t) and are estimated residual values for forward and backward ar model individually. (4) (5) in which, are m -dimension matrix. the traditional bdt algorithm can be written in full as followed: (6) (7) (8) (9) (10) (11) (12) (13) in equation 6 to equation 13, ûk and v̂k are estimated variance for forwarding and backward noise. the initial condition for traditional bdt algorithm are: (14) 137online outlier detection for time-varying time series on improved arhmm in geological mineral grade analysis process (15) (16) the subscript ø express that when the initial iteration, the model order set is empty. • double iteration bdt algorithm the objective function of improved bdt algorithm also is equation 3. the dynamic performance of the algorithm is enhanced by the forgetting factor . the improved bdt algorithm has double loop structure which model order is inner loop and time is the outer loop. we set which is part of equation 6. so: (17) in equation 7, k a set maximum value for model order. considering the time-varying characteristics of the model parameters, the forgetting factor is added to the outer loop(time loop) updates. (18) (19) in equation 18, ri t is the mean of the covariance matrix for ἓi-l (t) and ἢi-l (t-i) in time . similarly, equation 13 can be rewritten as: (20) so the calculation process of double iteration algorithm is illustrated in figure 1. conditional probability. the other is expressed as observation probability matrix calculated by ar model. (21) in equation 21, n(•) is gauss function, and ʃ̂k is estimated the variance of gauss distribution. arhmm outlier detection algorithm also composed of two steps: one step-preliminary detection from equation 1, we can see that there is a deviation between estimated process data by ar model and real process data. (22) if the deviation ɛk (t) is only noise, it obeys gauss distribution. so the preliminary criteria for outlier detection are to determine the probability that the deviation follows gauss distribution. (23) in equation 23, st = 1 indicates that the real data detected is normal, st = 0 means it is the outlier. so the detection criteria can be expressed that: (24) in equation 23, the subscript p is the optimal model order calculated by kicvc criteria whose expression is: (25) in equation 25, is the mean of residual ɛk (t) under various model order (bilmes, 2006). (26) (27) two step-final detection in final detection, the result of preliminary detection is the observed value of hmm. so the final detection result obtained by viterbi algorithm (abd-krim, 2006): (28) for improved arhmm algorithm, when the data at t time is detected, the data before t time already is detected. so the traditional viterbi algorithm is request into: (29) (30) parameters updating by outlier the parameters of order self-learning arhmm algorithm need update online, and the parameters are estimated residual mean , state transition matrix a , and in improved bdt algorithm. specific update algorithm is as follows: figure 1. flow chart of double iteration algorithm implementation of order self-learning arhmm detection algorithm the traditional arhmm structure is composed of two parts (wang, & chiang, 2008): one is markov chain, which is expressed as initial state probability π and state transition matrix in which st is the state at time t, n is the total state for hmm, and is a 138 jianjun zhao, junwu zhou, weixing su, fang liu (a) : if is normal data, then maintained by equations 30, 31, otherwise, not updated at time t. (b) : since there are two states in arhmm algorithm. so the updated algorithm is: (31) in equation 31, n (ay) indicates the times of the situation that st-1 = i, st = j, appears (lou, 195). (c) , : is normal data, then , calculated by equation(14), otherwise, using data means to replace xt. (d) is normal data, then calculated by equation 19; otherwise, it calculated by equation (32). (32) (e) is normal data, then calculated by equation (18); otherwise, it calculated by equation 33. (33) results and discussion model-based validations to verify the accuracy of the algorithm to detect the order of the model, the three order model which reacting ore detection interference process are used to generate the data. the data is shown in figure 2-(a), and the order detection results are shown in figure 2-(b). in which: (35) as can be seen from (34), there are time vary parameters on the denominator. to verify the robustness of the algorithm, we add 10% of the white noise and eight anomalies. the data is shown in figure 3-(a), and the order detection results and outliers detected results are shown in figure 3-(b) and figure 3-(c). figure 2. data and results of order detection as can be seen from the order detection results, the proposed algorithm can accurately detect the model order through the short-term adjustment process. with a view to verify that the proposed algorithm not only can identify the optimal model order but also can detect the abnormal data accurately. we modified the open-loop model mentioned in the alex alexandridis paper to get the second set of data (alexandridis, sarimveis, & bafas, 2003). rabiner 1989). the modified model is as follows: (34) figure 3. data and detection results for model data fig.3-(a) is the data waiting for detecting. fig.3-(b) is the order estimated result, and fig.3-(c) is the outlier detection result. as can be seen from the result, for the third-order nonlinear time-varying system, the model-order self-learning algorithm proposed in this paper can find its optimal model order and accurately detect all the anomaly data. application to further verify the practicality of the improved arhmm outlier identified method, it is applied to the on-stream x-ray fluorescence analyzer, which, first of all, in the last century 70's by the finotec outotec company successfully developed and implemented to mineral processing practice. so far, finland outotec company still has more than 80% market share. in china, the beijing institute of mining and metallurgy following its analytical principles, developed in 2014 with the same function grade analyzer boxa type on-stream x-ray fluorescence analyzer (hekimoglu, eernoglu, & kalina, 2009). according to foreign reports, the measurement accuracy of the analyzer increased by 1%, will effectively improve the metal recovery rate of 0.2 or more, and now from the hardware to improve measurement accuracy has been very difficult, or input-output serious disproportionate, so more scholars turn to the analysis of the modeling technology to improve research. based on this situation, the accuracy of the data for the model of learning is essential our comparative tests are as follows: the first set of data is obtained as follows: first of all, we use the improved arhmm outlier detection algorithm proposed in this paper to pre-process the spectral signal which as the input of the analyzer, and get the ore grade results as the first set of data. 139online outlier detection for time-varying time series on improved arhmm in geological mineral grade analysis process the second set of data is obtained as follows: we use the traditional ar outlier detection algorithm proposed in paper by northey, mohr, & mudd (2014) to pre-process the spectral signal which as the input of the analyzer, and get the ore grade results as the second set of data. finally, the two sets of data are compared with the results of the ore grade laboratory test, the error of the two groups of test results as shown in table 1. table 1. the result of comparison between with outlier detection process and without outlier detection process it can be seen from the table that there is higher accuracy when using the improved arhmm outlier detection algorithm proposed in this paper to do the pre-processing of the spectrum compared to tradition ar outlier detected method, which since the improved arhmm outlier detected algorithm have more robustness and more suitable for non-linear systems. conclusions taking into account the lack of arhmm algorithm for ore grade analysis process, an order self-learning arhmm algorithm is proposed in this paper, whose innovation points are summarized as: first, unlike other outlier detection method (such as the traditional ar model detection method), the outliers detection method proposed in this paper does not need to set the detection threshold. second, considering the problem that the model order of control system‘s hard to be determined, the new detected method which is based on residual error has the function of model order self-learning. and third, to avoid the influence of outliers on the test results, this paper proposes a detectionbefore-update and detection-based-update strategies. so under above improving, arhmm algorithm can more accurately use to analysis the data in the geological mineral grade analysis process. in other words, the application field of arhmm algorithm has been expanded. subsequent simulation by model data and practical application verify the accuracy, robustness, and property of online detection of the algorithm. according to the result, it is evident that new algorithm proposed in this paper is more suitable for outlier detection in the geological mineral grade analysis process. acknowledgments this research is partially supported by national natural science foundation of china under grant 51607122, 61602343. references abd-krim, s. (2006). vector autoregressive model-order selection from finite samples using kullback’s symmetric divergence. ieee transactions on circuits and systems i: regular papers, 53(10), 2327. alexandridis, a., sarimveis, h., & bafas, g. (2003). a new algorithm for online structure and parameter adaptation of rbf networks. neural networks, 16(7), 1003-1017. almeida, j. a. s., & barbosa l. m. s. (2007). a new method with outlier detection and automatic clustering. chemometrics and intelligent laboratory systems, 87(2), 208. barnet, v., & lewis, t. (1994). outlier in statistical data. john wiley & sons, chichester. bilmes, a. j. (2006). what hmms can do. ieice transactions on information and systems, e89-d(3), 1. brockwell, p. j., dahlhaus, r., & trindade, a. a. (2002). modified burg algorithms for multivariate subset autoregression. technical report 2002-015, department of statistics, university of florida. bullen, r. j., cornford, d., & nnbney, i. t. (2003). outlier detection in scatterometer data: neural network approaches. neural networks, 16(3-4), 419. clarke, b. r., & levis, t. (1998). an outlier problem in the determination of ore grade. journal of applied statistics, 25(6), 751-662. de'nan, f., naaim, n., & leong, l. c. (2017). behaviour of flush end-plate connection for perforated section. engineering heritage journal, 1, 11-20. edwin, m. k. & raymond, t. n. (1998). algorithms for mining distancebased outliers. proceedings of the twenty-fourth international conference on very large data bases, c. 392. han, j. w., & micheline, k. (2001). data mining concepts and techniques. machinery industry press, china. hekimoglu, s., eernoglu, r. c., & kalina, j. (2009). outlier detection by means of robust regression estimators for use in engineering science. journal of zhejiang university-science a, 10(6), 909. kameshwara, r., rao, c. r., & narayana, a. c. (2014). assessing grade domain of iron ore deposit using geostatistical modelling: a case study. journal of the geological society of india. 83(5), 549-554. knorr, e. m. & ng, r. t. (1999). finding intentional knowledge of distance-based outliers. proceedings of the twenty-fifth international conference on very large data bases, c. 211. lou, h. l. (1995). implementing the viterbi algorithm. ieee signal processing magazine, 1053-5888, 42. northey, s., mohr, s., & mudd, g. m. (2014). modeling future copper ore grade decline based on a detailed assessment of copper resources and mining. resources conservation and recycling, 83, 190-201. prakobphol, k., & zhan j. t. (2008). a novel outlier detection scheme for network intrusion detection systems. proceedings of the second international conference on information security and assurance, c. 555. rabiner, r. l. (1989). a tutorial on hidden markov models and selected applications in speech recognition. proceedings of the ieee. 77(2), 257. ramaswamy, s., rastogi, r., & kyuseok, s. (2000). efficient algorithms for mining outliers from large data sets. proceedings of the acm sigmod international conference on management of data dallas, c. 427. rivoirard, j., demange, c., & freulon, x. (2013). a top-cut model for deposits with heavy-tailed grade distribution. mathematical geosciences, 45(8), 967-982. takeuchi, j., & yamanishi, k. (2006). a unifying framework for detecting outliers and change points from time series. ieee transactions on knowledge and data engineering, 18(4), 482. trindade, a. a. (2003). implementing modified burg algorithms in multivariate subset auto-regressions modeling. department of statistics, university of florida. wang, j. s. & chiang, j. c. (2008). a cluster validity measure with outlier detection for support vector clustering. ieee transactions on systems, man, and cybernetics, part b (cybernetics), 38(1), 78. issn 1794-6190 e-issn 2339-3459 earth sciences research journal earth sci. res. j. vol. 21, no. 4 (december, 2017): 183-188. http://dx.doi.org/10.15446/esrj.v21n4.66103 e n g in e e r in g g e o l o g y the present study focuses on some experimental laboratory tests using a newly-constructed modified direct shear test apparatus. the single-stage and multistage direct shear tests were performed to determine the shear rate and test scheme of the unsaturated shear test. shear strength parameters of unsaturated soil in different conditions are obtained, and the criteria indicate good agreement with standard theories of unsaturated soil. the nonlinear matric suction failure envelope is determined. some shear strength equations are also fitted through the experimental results. abstract keywords: unsaturated soil; direct shear test; matric suction direct shear test of unsaturated soil peng he1, 2, liuying sun2, zhen wang3, * 1 school of civil engineering and architectur, xi’an university of technology, xi’an 710048, china. 2 college of civil engineering, north china university of water resources and electric power, zhengzhou 450011, china. 3 department of geotechnical engineering, tongji university, shanghai 200092, china. el presente estudio se basa en varias pruebas de laboratorio montadas en una máquina de experimentos recientemente construida y modificada para este objetivo. se realizaron las pruebas monofase y multifase de cizallamiento para determinar el punto de corte y el modelo de cizallamiento no saturado. de esta manera se obtuvieron los parámetros de cizallamiento de suelos no saturados bajo diferentes condiciones y los resultados coincidieron con las teorías evaluadas en esta materia. también se determinó la curva de corte para la succión matricial no lineal. algunas ecuaciones de fuerza de cizallamiento también encajan en los resultados experimentales. resumen palabras clave: suelos no saturados; cizallamiento directo; succión matricial. record manuscript received: : 05/07/2017 accepted for publication: 01/11/2017 how to cite item he, p., sun, l., & wang, z. (2017). direct shear test of unsaturated soil. earth sciences research journal, 21(4), 183-188. doi: http://dx.doi.org/10.15446/esrj.v21n4.66103 ensayo de cizallamiento directo en suelos no saturados 184 peng he, liuying sun and zhen wang introduction expansive soil mostly spreads in arid, and semi-arid areas, and it has been found in more than 40 countries (fredlund, 1993), including 20 regions of china. in addition to the unsaturated state most expansive soil always been in practical application, the study on mechanical properties of expansion soil in the unsaturated state is of great significance (lee, 2010; abbasi et al., 2017). unsaturated soil test features time consuming and expensive in general, mainly due to the low permeability of the unsaturated soil and high air-entry ceramic disc (even though it is permeable but airproof). test samples in triaxial tests are relatively bigger (39.1 mm diameter, 80 mm thickness; 50 mm diameter, 100 mm thickness and 70 mm diameter, 140 mm thicknesses). thus, determining the matric suction requires much more time; comparatively time associated with shear strength testing of unsaturated soils is significantly reduced from the smaller size of test samples, generally 61.8 mm diameter, 20 mm thickness (bishop et al., 1960; bishop et al., 1963; ali et al., 2017; anjum et al., 2017). method experimental procedures the newly-constructed modified direct shear test apparatus the conventional direct shear apparatus has been modified, and now it can control the pore air pressure ua and pore water pressure uw. the direct shear box is placed in a sealed pressure chamber; its internal structure is presented in figure 1. figure 1. schematic diagram of the modified direct shear test apparatus. as shown in figure 1, the soil specimen is placed in a high air-entry ceramic disc with a porous-permeable stone on top. the pore air pressure ua can be controlled by adjusting the air pressure in the pressure chamber, and the air pressure can be directly added to the porous-permeable stone (yasin et al., 2017; shazad et al., 2017). the pore water pressure uw can be controlled by the high air-entry ceramic disc which is placed at the bottom of the soil specimen to provide continuity of the water phase in the soil specimen and under the ceramic disc (basarian and tahir, 2017). nevertheless, with the test time passed by, the air in soil pore may diffuse into the bottom of the high air-entry ceramic disc with the water flow and generated bubbles. the flushing device is placed at the bottom of the test apparatus to exhaust the air under ceramic disc which can speed up the testing process efficiently. the overall test system is controlled by a computer recording the shear strength, horizontal displacement, vertical displacement and water phase inlet and outlet. test soil samples the study is conducted using undisturbed expansive soils from yuzhou, middle of henan, china, and the clay mineral mainly consists of calcium montmorillonite (harith and adnan, 2017). before testing, soil sample of 25% moisture is dispensed and placed in the moist chamber for 24 hours to permit uniform distribution of water, and then expansive soil specimens in different dry density conditions are created by being extruded from lifting jack and using a cutting ring. the diameter and height of the example are 61.8 mm and 20 mm, respectively. the sample is placed in the moist chamber for spare after being pumped and saturated. laboratory test shear rate determination shear rate determination is a crucial problem that directly affects the time and results associated with the shear strength testing of unsaturated soils. in the consolidated undrained shear test, it takes some time for pore water pressure of the soil transferred to the specimen bottom through the specimen or filter paper (rahman et al., 2017; simon et al., 2017). if the shear rate is too high, pore water pressure will not dissipate completely, values of pore water pressure measured at the bottom of the sample will perform lagging, and useful stress parameters cannot be obtained, which apparently makes numerical value measured low. thus, the shear rate should be determined appropriately according to the soil permeability coefficient (fredlund et al., 1997; ismail et al., 2017). several single-stage direct shear tests were carried out at four shear rates on modified apparatus as experimental tests by maintaining a constant the net normal stress, considering the effect of shear rate on test results. the test detail and results are presented in table 1 and figure 2. table 1. tentative experiment scheme for determining test shear rate. figure 2. results of shear displacement and shear stress in different shear rates. 185direct shear test of unsaturated soil as the test results showed, the ultra pore water pressure can dissipate, peak shear stress can remain stable when the shear rate is controlled by 0.012 mm/ min. thus, 0.012 mm/min is proposed as the standard shear rate in the entire test. single-stage shear test of unsaturated soil the loading steps of the matric suction for the single-stage tests were 0 kpa, 50 kpa, 100 kpa, 200 kpa; under different matrix suctions, the vertical stress measures were 100 kpa, 200 kpa, 300 kpa. table 2 summarizes the single-stage direct shear test scheme of expansive unsaturated soil. table 2. the single-stage direct shear test scheme of expansive unsaturated soil. twenty samples of same dry density were prepared once a time and placed in the moist chamber after being pumped and saturated to eliminate the effects on test results from sample preparation. each time the sample was transferred to direct shear apparatus and applied predetermined air pressure until the suction arrived at equilibrium (generally last 12 hours), then the vertical load was applied (maintaining for 12 hours) with keeping the air pressure and vertical stress constant and then be sheared. the shear rate was determined to 0.012 mm/min. terminated shear displacement was set to 6 mm and 9 10 hours. it was needed to guarantee that the peak strength was not surpassed. results and discussion (a) single-stage direct shear test results of unsaturated soil in dry density of 1.4 g/mm3 the effects of shear displacement and shear stress in different matric suction and vertical pressure are presented in figure 3. table 3. single-stage direct shear stress parameters of expansive unsaturated soil in the dry density of 1.4 g/mm3. it can be summarized from figure 3 that shear strength of soil sample increased with the rise in vertical pressure and matric suction. under different matric suctions, the peak shear stress was chosen as failure shear stress if the displacement didn’t arrive 4 mm; if exceeded, it should be 4 mm. shear stress under different matric suctions, and vertical pressure concluded from figure 3 are presented in table 3. figure 3. the results of shear displacement and shear stress (ρd = 1.4 g/cm3) (b) single-level direct shear test data of unsaturated soil in dry density of 1.5g/mm3 figure 4 presents relationships between shear displacement and shear stress in different matric suctions and net vertical pressure. table 4. single-stage direct shear test stress parameters of expansive unsaturated soil in a dry density of 1.5 g/mm3 it can be summarized from figure 4 that shear stress of soil sample increased with the rising of vertical stress and matric suction. table 4 shows the single-stage direct shear test strength parameters under different matric suctions. from table 4 and figure 4 it can be summarized that shear stress increase with the rising of net vertical pressure and matric suction. cohesion value c increased sharply with the rising of net vertical pressure, which can mean that the increase of vertical force has a significant effect on cohesion enhancements; while, the internal friction angle under vertical pressure changed little. 186 peng he, liuying sun and zhen wang figure 4. results of shear displacement and shear stress (ρd = 1.5 g/cm3) (c) single-stage direct shear test results of unsaturated soil in dry density of 1.6 g/mm3 figure 5 presents the results of shear displacement and shear stress under different matric suctions and vertical pressure. figure 5. results of shear displacement and shear stress (ρd = 1.6 g/cm3). it can be summarized from figure 5 that soil shear stress increased with the rising of vertical pressure and matric suction. under different matric suctions, the peak shear stress was chosen as failure shear stress if the displacement didn’t reach 4mm; if exceeded, it should be 4mm. shear stress under different matric suctions, and vertical pressure concluded from figure 5 are presented in table 5. table 5. single-stage direct shear test stress parameters of expansive unsaturated soil in a dry density of 1.6 g/mm3. (d) single-stage direct shear test results of unsaturated soil in dry density of 1.7 g/mm3 figure 6 presents results of shear displacement and shear stress in a dry density of 1.7 g/mm3 under different matrix suctions and vertical pressure. figure 6 allows determining that shear stress of soil samples continuously increased with the rising of matric suction and vertical pressure. shear strength parameters under different matric suction are presented in table 6. figure 6. results of shear displacement and shear stress (ρd=1.7g/cm3). table 6. single-stage direct shear test stress parameters of expansive unsaturated soil in a dry density of 1.7 g/mm3. results analysis shear strength parameters of four dry density specimens under different matric suction and vertical pressure are listed in table 7 187direct shear test of unsaturated soil table 7. direct shear test stress parameters of unsaturated soil. (a) effects on shear strength parameters from matric suction 0 50 100 150 200 0 5 10 15 20 25 30 35 40 c o h e s i o n £ k̈ p a £ © mat r i c s uct i on£ k̈pa£ © model polynomial adj. r-square 0.97506 value standard error -intercept 4.35391 2.16438 -b1 0.36442 0.05495 -b2 -0.00101 2.54822e-4 dr y dens i t y 1. 4g/ cm3 0 50 100 150 200 10 15 20 25 30 35 40 45 dr y dens i t y 1. 6g/ cm3 c o h e s i o n ( k p a ) mat r i c s uct i on( kpa) model polynomial adj. r-square 0.9946 value standard error -intercept 10.70927 0.97392 -b1 0.28565 0.02473 -b2 -6.18545e-4 1.14664e-4 0 50 100 150 200 5 10 15 20 25 30 35 40 45 dr y dens i t y 1. 5g/ cm3 c o h e s i o n ( k p a ) mat r i c s uct i on( kpa) model polynomial adj. r-square 0.99296 value standard error -intercept 5.63591 1.21969 -b1 0.35034 0.03097 -b2 -8.72818e-4 1.43599e-4 0 50 100 150 200 10 15 20 25 30 35 40 45 dr y dens i t y 1. 7g/ cm3 c o h e s i o n ( k p a ) mat r i c s uct i on( kpa) model polynomial adj. r-squar 0.99225 value standard error -intercept 13.66291 1.12924 -b1 0.25601 0.02867 -b2 -4.93818e1.3295e-4 figure 7. results of cohesion and matric suction in different dry density conditions. the experimental results are represented by a equation including cohesion, internal friction angle and matric suction where s＝ matric suction, c＝ cohesion and ø＝ internal friction angle, a1, a2, b1, b2, b3＝ fitting parameters. fitting curves in different dry density conditions are presented in figure 7 and 8; parameters are listed in table 8. 2 321 ** sasaac ++= )**exp( 2321 sbsbb ++=f figure 8. results of internal friction angle and matric suction in different dry density conditions. table 8. fitting parameters. (b). effects of cohesion and internal friction angle from dry density values of fitting parameters in table 9 are known parameters. consider the effect on fitting parameters from dry density, fitting curves and parameters are presented in figure 9. table 9. parameters fitted by polynomials. (1) (3) (4) (5) (6) (7) (8) (9) (2) fitting formulas (3) ~ (7) are presented below; fitting coefficients are listed in table 9. 188 peng he, liuying sun and zhen wang figure 9. relationship between dry density and fitting parameters (a): a1; (b): a2; (c): a3; (d): b1; (e): b2; (f): b3. matric suction conditions can be calculated substituting (3) (7) to (8) and (9), cohesion and internal friction angle in any dry density. (10) (11) conclusions methods for unsaturated shear testing are more complicated, more time consuming and more expensive when compared to conventional test methods for saturated soils. an extensive investigation was carried out to study the validity of rapid procedures in obtaining shear strength parameters of saturated and unsaturated soils using the single-stage direct shear test. in this paper, the tests used a newly-constructed modified direct shear test apparatus that uses smaller samples and allows independent control of pore air pressure ua and pore water pressure uw. the shear strength was established using single-stage loading over a range of net normal stresses and matric suction values. the study was carried out using soil samples from the middle route of southto-north water transfer project in yuzhou section. the main conclusions from the study are presented as below: (1) the tests to determine appropriate shear displacement rate was successfully applied before unsaturated direct shear tests. the results of experimental tests showed that 0.012 mm/min is the most suitable shear rate for the samples of yuzhou. (2) it was shown that efficiency of the direct shear test of unsaturated soil is higher compared with the conventional analysis. because of this time-saving test method, it cost less time to get shear strength parameters of unsaturated soil under different conditions. (3) numerous parameters can be controlled in modified tests, from which we got shear strength parameters of unsaturated soil under different conditions; some classical theories about unsaturated soil are also verified. also, the nonlinear failure envelope of matric suction was obtained. (4) the test results confirmed the nonlinear interrelationship among the cohesion, internal friction and dry density, which can be used in some practical engineering about unsaturated soil as a kind of empirical formula. (5) it was shown that the shear rate must be controlled within 0.012 mm/min to get reliable shear strengths parameters for unsaturated soil samples of yuzhou. thus, the test of determining appropriate shear displacement rate before the shear test is necessary. thus, the recommendations are given excessively, it is essential to shear the soil sample beyond the peak shear stress to the strain softening region, and make sure that the samples are wholly consolidated at every new effective normal stress-stage. acknowledgements this research was financially supported by the science and technology project of henan provincial transportation department (grant no. 2012p51) and key project of science and technology of henan provincial education department (grant no. 14b580001). references abbasi, n., khan, m.s., mughal, m.s., & yasin, m. (2017). sedimentary facies analysis of nagri formation, kashmir basin, subhimalayas, pakistan. pakistan journal of geology, 1(2): 03–06. ali, s.s., ijaz, n., aman, n., nasir, a., anjum, l. & randhawa, i.a. (2017). clinical waste management practices in district faisalabad. earth sciences pakistan, 1(2):01–03. anjum, t., khan, h.i.u.h., tariq, w., farooq, u. & shauket, i. (2017). production of soil-cement bricks using sludge as a partial substitute. earth science malaysia, 1(2): 10-12. basarian, m.s., & tahir, s.h. (2017). groundwater prospecting using geoelectrical method at kg gana, kota marudu, sabah. earth science malaysia, 1(2): 07-09. bishop, a. w., & blight, g. e. (1963). some aspect of effective stress in saturated and partly saturated soils. geotechnique, 13,177-197. bishop, a. w., alpan, i., blight, g. e. & donald, i. b. (1960). factors controlling the shear strength of partly saturated cohesive soil. research conference on shear strength of cohesive soils asce, 503-532. chen, j. & fredlund, d. g. (2003). advance in research on shear strength of unsaturated soils. rock and soil mechanics 24, 655-660. fredlund, d. g. (1993). soil mechanics for unsaturated soil. new york, wiley, 1-10. fredlund, d. g., rahardjo, h., & chen, z. (1997). the unsaturated soil mechanics. beijing: china architecture press, 52-55. gan, j. k. m., fredlund, d. g. & rahardjo, h. (1988). determination of the shear strength parameters of an unsaturated soil. canadian geotechnical journal, 25, 500-510. harith, n.s.h. & adnan, a. (2017). estimation of peak ground acceleration of ranau based on recent earthquake databases. malaysian journal geosciences, 1(2): 06-09. ismail, m.n., rahman, a. & tahir, s.h. (2017). wave-dominated shoreline deposits in the late miocene sedimentary sequence in the miri formation north sarawak, malaysia. geological behavior, 1(2):14–19. lee, h. (2010). considering arbitrary drying wetting cycle changes in the history of unsaturated soil constitutive theory. ph.d. diss., institute of rock and soil mechanics, chinese academy of sciences. rahman, n.a., tarmudi, z., rossdy, m. & muhiddin, f.a. (2017). flood mitigation measures using intuitionistic fuzzy dematel method. malaysian journal geosciences, 1(2):01-05. shahzad, a., munir, m.u.h., yasin, m., umar, m., rameez, s., samad, r., altaf, s. & sarfraz, y. (2017). biostratigraphy of early eocene margala hill limestone in the muzaffarabad area (kashmir basin, azad jammu and kashmir). pakistan journal of geology, 1(2): 16–20. simon, n., roslee, r. & lai, g.t. (2017). temporal landslide susceptibility assessment using landslide density technique. geological behavior, 1(2):10–13. yasin, h., usman, m., rashid, h., nasir, a., & randhawa, i.a. (2017). alternative approaches for solid waste management: a case study in faisalabad pakistan. earth sciences pakistan, 1(2):04-06. the aim of this article is the identification and characterization of complex factors acting simultaneously on forming large sized reservoirs in the deep south caspian basin, as well as the development of a model based on a combination of synchronous natural processes of the formation of hydrocarbon deposits. the article analyses reasons for high layer pressures in the hydrocarbon fields of the south caspian depression (scd), impacting factors. the tectonic factor and density-barometric model can become the main criterion for selection of priority projects for exploration and site selection for drilling wells. obtained results allow conducting effective studies of the exploration and development of oil and gas condensate fields in the scd area. abstract keywords: folding genesis; hydrocarbon deposits; neo-tectonic activity; horizontal compressive strain; vertical forces. formation of oil and gas reservoirs in the great depths of the south caspian depression issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n4.64008 earth sciences research journal earth sci. res. j. vol. 21, no. 4 (december, 2017): 169-174 gahraman nariman ogly gahramanov state oil company of azerbaijan republic gahraman@inbox.ru el objetivo de este artículo es la identificación y caracterización de los complejos factores que actúan en la formación de depósitos de gran tamaño en lo profundo de la cuenca al sur del mar caspio, al igual que el desarrollo de un modelo de formación de depósitos de hidrocarburos basado en la combinación de procesos naturales sincrónicos. este artículo analiza las causas de la presión de las capas altas en los campos de hidrocarburos ubicados al sur de la depresión cáspica y sus factores de impacto. además, se propone el componente tectónico y el modelo de densidad barométrico como los criterios más importantes para la selección de proyectos prioritarios de exploración y la escogencia de lugares para perforación de pozos. los resultados obtenidos permiten realizar estudios de exploración y desarrollo en los campos de petróleo y gas condensado en el sur de la depresión cáspica. resumen palabras clave: formación de plegamientos, depósito de hidrocarburos; actividad neotectónica; esfuerzo horizontal de compresión; fuerzas verticales. record manuscript received: 13/04/2017 accepted for publication: 25/08/2017 how to cite item gahraman gahramanov, o.n.(2017). formation of oil and gas reservoirs in deep water areas of the south caspian depression. earth sciences research journal, 21(4). 169-174 doi:http://dx.doi.org/10.15446/esrj.v21n4.64008 formación de depósitos de petróleo y gas en las profundidades al sur de la depresión cáspica g e o l o g y o f o il a n d g a s 170 gahraman nariman ogly gahramanov introduction the scd is the most deep-sunk sedimentary basin in the world and occupies the entire southern deep-water part of the caspian sea and its adjoining kur and western-turkmen depressions in the west and east, respectively. the scd is the most deep-sunk sedimentary basin in the world and occupies the entire southern deep-water part of the caspian sea and its adjoining kur and western-turkmen depressions in the west and east, respectively. it is a unique geological structure in the earth’s crust in terms of sedimentation conditions, thickness and rate of deposition, history of geotectonic development, oil and gas formation, oil and gas accumulation and depth of their deposition, reserves, volumes, and geological structure of reservoirs. the scd is a relic of the large caucasian south caspian back-arc marginal sea, which emerged in the middle jurassic period and differs from deep-water basins in inland and marginal seas of the world: the scd has the thinnest (6-8 km) consolidated oceanic crust and the thickest sedimentary cover (25-30 km). the thickness of pliocene-quaternary sediments is 10-12 km. at that, the thickness of the main oil and gas -pay red rocks horizons of the pliocene – is 7-8 km. the scd was formed in the arabian-eurasian collision zone. it is surrounded by the mountain ranges of the greater caucasus, kopet dag, talysh, and elbors (mamedov, 2008; aliyev, 2005; eybatov, 2010). effective exploitation of hydrocarbon (hc) resources in the scd requires a development of a new scientific bases and methodological solutions of relevant tasks. the creation of theoretical models of these scd deposit formations is an unprecedented and complicated task. at present, hydrodynamic modeling methods for hydrocarbon deposits, computer and programming technologies are being developed rapidly, which allows designing detailed numeric hydrodynamic models with account for the most factors that affect the development of deposits. in particular, this approach is described in detail in by j. hunt (2013). this allows using in the output data for mathematical modeling the entire source material that is accumulated at oil fields by the documentation of the state of development by wells and the deposit in general. however, one should not forget that a numerical geological model is the basis for the detailed hydrodynamic model (mstislavskaya et al., 2005; zhang, 2012; carvalho, 2013) shaymardanov et al. (2003) and abasov et al. (2007) established that the most productive structures (deposits) at great depths are formed in favorable conditions of earliergenerated and preserved hydrocarbon gases, i.e. where gas did not escape through cap rocks or they were not destroyed by mud volcanoes or other geological processes. preservation of hc at great depths is also facilitated by “sealed” deposits and lithological traps (bochever et al., 1972). under these conditions, hc preservation is determined by a weakened draining-out of formed gas during oil destruction due to significant deceleration of its diffusion. theory and practice showed that the probability of finding “sealed” deposits and high-capacity lithological traps is low (kartsev et al, 1992; gurbanov et al., 2014). the creation of comprehensive geological three-dimensional filtration models, based on the combination of their preconditioning natural simultaneous processes (norman, 2001), is a promising area of theoretical and applied geology that allows taking a big step towards the solution of this problem. intensive hc generation due to the existence of a powerful deep source is a geochemical prerequisite of hc deposit formation (abasov, 2008). another question of interest: can folded structures of the sedimentary basin, which appeared in the scd, or poor folds be further developed by means of internal force impact in the hc deposit itself? at that, the creation of a theoretical model of formation of such deposits is a complicated task. the difficulties are determined by a wide range of geological conditions and the lack of relevant information (popkov et al, 2014). limited data on these underground processes in hc deposits, located in the scd (kazmin et al., 2011), allow designing only a conceptual model of hc reservoir formation (baak, 2010), based on the natural peculiarities of oil and gas deposits at great depths. the conditions of the tectonic development of the scd and its subsidence primarily in the neogene-quaternary period determined the accumulation of a thick flysch formation and the accumulation of large oil and gas deposits thereon (gurbanov et al., 2014). interdisciplinary modelling shows that subsidence and sedimentation in the scd resulted from thermal subsidence following jurassic rifting and further enhanced subsidence due to sedimentation during in the neogene-quaternary (alizadeh et al., 2017). based on advanced analytical methods, hydrocarbons of the lower pliocene productive series (the main reservoir) are of epigenetic origin (feyzullaev et al., 2015). they have been formed due to subvertical impulsive migration from the oligocene-miocene sediments. two main phases of oil migration produced commercial accumulation in the productive series. the following questions arise: 1. which factors preconditioned the accumulation of such thick hydrocarbon-bearing sedimentary layers and which forces played the main role in the formation of large oil and gas deposits? 2. why are deposits with satisfactory reservoir properties (rp) encountered at such great depths in the scd area? this should not happen at such depths. the purpose of this study is to determine and characterize the complex of simultaneously acting factors of the formation of large reservoirs at great depths of the scd, and the development of a model, based on the combination of their preconditioning natural simultaneous processes. method kuliyev et al. (1998) shows that there are several variants of the mechanism of formation of folds by means of “internal” factors of the sedimentary basin. in the present paper, these results were taken as the theoretical aspects of the proposed concept. under certain conditions, abnormally high reservoir pressures (ahrp) are formed and, consequently, sediments are compacted. this process mostly occurs in the central, the most submerged parts of the sedimentary basin, mostly comprised of clay rocks. rock compaction primarily depends on the effective pressure (pef) (kuliyev et al., 1998): pef = prc cppor (1) where (prc) is the pressure of the overlying sedimentary rocks, (ppor) is the pressure of fluids that saturate rocks (pore pressures); c is the load coefficient that varies from 0 to 1. the studies conducted in this area, showed that rock porosity can be determined by the following formula (2) where , is the rock porosity under pef pressure and without compacting pressure, respectively, β is the irreversible compaction coefficient. taking into account the fact that porosity is related to density by formula (3) where is the matrix density, is the proportionality coefficient, formula (3), taking into account (1), can be written as (4) 171formation of oil and gas reservoirs in the great depths of the south caspian depression figure 1. left: location of the petroleum and anticlinal structures of the south caspian megadepression. 1 anticlinal structures; fields: 2 oil, 3 gas, 4 oil and gas, 5 oil and condensate. from formulas (1) and (4) it follows that under other various conditions, marginal areas of basins will be characterized by the greatest level of effective pressure and rock density (kuliev et al., 1998). in the central parts, where pore pressures are close to geostatic ones, effective pressure will be significantly lower and can even theoretically be close to zero (mamedov, 2008; kuliyev et al., 1998). consequently, such discrepancies in the levels of compacting load directly affect rock density, in particular: rock density in the horizontal section reduces from the periphery to the center of the basin. results the location of oil and gas and anticlinal structures of the south caspian megadepression, which gives an idea of the location of major hydrocarbon deposits in the study region is presented in the fig. 1. due to specific geological, geochemical, pressure, and temperature conditions of hc accumulation in traps and its preservation in deposits at great depths, excessive reservoir pressure is a crucial factor. it is determined by the size of the reservoir, the presence of ahrp beyond its boundaries, and the difference in fluid density. the formation of larger and thicker structures is investigated as a process of accumulation, selforganization, and self-preservation of natural oil and gas reservoirs at great depths. this allows for a new approach to the problem of folding and formation of large fields on a local scale in the scd itself. the formation of anticlinal traps is known to be accompanied by an almost simultaneous development of the sedimentary basin itself. the theoretical substantiation and adequate modeling of values, the direction and nature of the effect of pef, prc, and phyd (hydrostatical pressure) factors, and the possibility of gas-dynamic and geodynamic pressure effect are the determinants of the scale and form that implement structural changes. in the anticlinal structures of a gas condensate field, especially with low hc density, in this phase, pressure, which corresponds to the far-wing parts of the structure can be transferred to its crest without significant changes. this zone is characterized by excessive reservoir pressure. the considered mechanism of formation and reformation of gas condensate fields up to large sizes due to excessive pressure in the elevated structure parts of the scd provides a more reasonable explanation of their vertical and horizontal zonation. in the authors’ opinion, the preservation of large hc accumulations in the scd deposits can be governed, primarily, by two reasons. the first one is the type and size of traps, with anticlinal structure and deposit thickness being of great importance. the second one is the intensive generation of gas due to catagenetic destruction of organic matter (om) under the elevated pressure and temperature conditions and influx of gaseous fluids into traps. it is worth noting that the scale of oil catagenetic transformation is affected by both its genetic peculiarities, i.e. the initial hc content of oil, and the pressure and temperature conditions, required for the change of the hc phase structure. therefore, liquid hcs, which have the greatest resistance to thermal action, are located at great depths. at present, the initial level of reservoir pressure decreased significantly in the gas condensate fields of many regions. due to the development of gas condensate fields, the reduction of reservoir pressure could affect the changes of hc phase structure in these fields to the liquid phase, since under favorable temperature conditions, the level of reservoir pressures is important during this process. the decisive factor of great depths is the capacity and reservoir properties (rp) of rocks. the temperature conditions in the scd are favorable for the preservation of hc. the temperature reaches 120°c at 6 km depth (guliyev et al., 2011). therefore, the main factor that deteriorates the rp of rocks with depth is pressure, i.e. increased rock pressure (yusifzade, 1985, 2013). in the authors’ opinion, the preservation of large hc accumulations in the south caspian depression deposits can be determined by two reasons. hc-containing reservoirs, in the thick elevated anticlinal structures, have a relatively increased reservoir pressure due to the growth of excessive pressure. the growth of excessive pressure in the elevated part of an anticlinal structure is caused by the difference in the density of deposit water and gas. it increases with temperature growth. the reservoir pressure in the contour, where the relative depth increases, in thick structures in gas fields, is almost transferred to the crest of the deposit. the deposition depth in the crest of the anticlinal structure decreases significantly compared to the wings. reducing the effective pressure level facilitates the deceleration of rock compaction, sometimes even decompaction. this helps preserve or even slightly reduce the primary capacity and reservoir properties of rocks. in reducing the negative impact of rock pressure on the horizon and preserving rp, high reservoir pressure in the crest of a depressed structure during simultaneous hc accumulation facilitates the loss of stability of the considered environment, the growth of the anticlinal structure itself and, subsequently, the continuation of hc accumulation in this trap, which ultimately forms large reservoirs. the most important criterion for the estimation of oil and gas basins’ prospect is the volume of sediment accumulation and the frequency of discovery of large and giant oil and gas fields (aliyev, 2007). the scd volume of sediment accumulation is estimated to be ca. 3.25 million km3. the thickness of the sediment complex is 25 km. the complex consists of a thick stratum of sediments of primarily terrigenous and carbonate content from the jurassic to the quaternary period. most oil and gas (gas condensate) deposits, discovered in the pay horizon of azerbaijan are limited to the “absheron facies”, formed by deltaic deposits of the paleo-volga (aliyev, 2007). the sediment accumulation of compacted sand and clay, alongside a mass of diatoms, high temperatures in the depression zone (absheronbalkhan depression) and the counter count of high-temperature fluids, created ideal conditions for oil and gas accumulation (abasov et al., 2007). since the 1970s, a number of promising oil and gas condensate fields have been discovered in the absheron archipelago of the azerbaijan sector of the caspian sea. these include such large fields as azeri-chyrag-guneshlil, shah deniz, large structures of kyapaz, umid, safar-mashal, nakhchivan, and others. the main deposits of hydrocarbons (over 70%) in abovementioned fields are concentrated in the “pereriv” suite and the x layer of the pay horizon (yusifzade, 1985). the geological structure of the area is formed mostly by pliocene-quaternary sediments with a thickness of 3700-4000 m. on the regional scale a certain growth in reservoir pressures is observed, where the anomaly coefficient reaches 1.05-1.07 on average. on the local scale, i.e. within hc deposits, the level of reservoir pressures can exceed the hydrostatic pressure significantly. at that, the anomaly coefficient can go up to 1.7 and more. 172 gahraman nariman ogly gahramanov of pef, prc, and ppor strains (figure 2). in the anticlinal structures of gas condensate fields, especially in the conditions of low hc density, the pressure that corresponds to the far wings of the structure also affects its crest without significant changes. thus, such zones are characterized by excessive reservoir pressure. due to excessive pressure in the elevated parts of the structure in the scd, distribution of vertical forces in the deposit area provides for a more substantiated explanation of the mechanism of formation and reformation of gas condensate fields up to large sizes (kuliyev, 1998). estimations of excessive pressures for various thicknesses and depths of gas condensate deposits were carried out in order to determine the regularities of the changes in excessive pressure and the gradients of reservoir pressure with depth (table 1). figure 2. schematic cartoon that explains the distribution mechanism of pressures (prc, ppor, pef) in the mentioned fields that created excessive pressure in the crest of the anticlinal structure, which causes structure growth (a1, a2, a3) with increase of deposition depth. a) density-baric model; b) discovered and assumed structure; c) diagram of the hydrostatic, geostatic, and reservoir pressures. 1 – current position of the structure; 2 – structure at the early stage of development; 3 – mud volcano; 4 – assumed structure; 5 gas-dynamic system; 6 sea bottom; 7 smooth bank; 8 sea level; 9 water and gravity system; 10 decompaction zone; 11, 14 pef – 38 mpa; 12, 13 ppor – 70 mpa; 15 compaction zone; 16 0,1 mpa/m; 17 prv (excessive pressure) in well no. 1; 18 0,24 mpa/m; 19 well no. 1; 20 well no. 2. table 1. information on the distribution of primary geological reserves of gas with depth at the western edge. the following question arises can folded structures of the sedimentary basin in the scd, or poor folds be further developed by means of internal force impact in the hc deposit itself notwithstanding the external force impacts, in particular, tectonic forces? the reason for the emergence of intrabasin linear structures (there are seven such lines in the absheron archipelago and four in the baku archipelago) is the horizontal compressive strain, which causes folding in the sedimentary cover in the scd and subducting in the consolidated crust. another question arises: can folded structures of a sedimentary complex be formed by internal force impact of the basin itself? during sedimentation, horizontal gradients of effective pressure and density are formed in the basin, which are directed from the center to the periphery. i.e. the basin, essentially, generates internal compressive horizontal forces, independent on the external forces effect. the magnitude of such strain can be sufficient for the development of folds. the probability of folding in this region is high. the period of origination and youth of the sediment basin are inherent in all basins. therefore, the described mechanism is typical of all basins, irrespective of their geological age. both lithological and geotectonic reconstruction should consider the possibility of folding due to the internal forces of the sediment cross-section (kuliyev et al., 1998; babayev et al., 2014). a similar process occurred in the pay horizon (ph) of the scd, since in the deep-lying horizons, there is a high probability of favorable conditions for the formation of hc in cases of deposition at depths of less than 2.54.5 km. the constructed profile for the determination of the influence of effective pressure shows changes in the sizes of the structure with the increase of the deposition depth of the pay horizon and the distribution figure 3 presents charts of changes in the anomaly of reservoir pressures with depth, depending on deposit thickness, which shows how important the thickness of gas condensate deposits for the formation of large fields and the conditions of preservation of deposits (reservoirs) at great depths. in the crest and far wings of the structure, pef values differ significantly and increase with the growth of the structure and, consequently, the preservation of reservoir properties occurs primarily in the elevated parts of the structure. the local maximum pef in the elevated part of the structure creates the difference in anomalies both horizontally and vertically. the ∆pef gradient, which affects the long period even at small values, affects the formation (growth) of the anticlinal structure. thus, other thing being equal, the greatest levels of rock pressure and density will be characteristic of the marginal and aquifer zones of the deposit. in the central parts, the effective pressure will be significantly lower. in the crest of the structure, due to a relatively small deposition depth, rock pressure will be the lowest, while internal pressure will be the highest. it is also expedient to examine the mechanism of emergence of horizontal compressive strains in a quick subsidence regime (figure 4). in a submerging basin, other mechanisms of formation of horizontal compressive strains are possible. assume that within the abovementioned geological situation, i.e. due to substance differentiation, the values of forming horizontal compressive strains are insignificant. 173formation of oil and gas reservoirs in the great depths of the south caspian depression figure 3. gas condensate deposits. changes in excessive pressure (prv/pex) with depth, depending on the deposit thickness of 1, 2, 3, 4 – 500, 1000, 1500, 2000 m, respectively. figure 4. the schematic cartoon of structural elevation under the effect of forces. it is considered geostatic, hydrostatic, effective, and excessive pressure, with account for sedimentary basin pressures. sedimentary basin pressure is only explained by influence of sedimentary cover. more preferential mechanism is the combination of loading the edges of the basin and large horizontal compressive forces due to the syn compressional bending associated with subduction of the scd (abdulaev et al., 2015). after a certain geological time, the magnitude of such horizontal strains can be sufficient for the process of elastic instability of the competent bed (kuliyev et al., 1998). next, it is expedient to investigate the mechanism of formation of scd structures during the emergence of vertical forces. the studied structures differ from other folds in the studied region in relatively large sizes and mostly isometric dislocation. these structural features confirm the dominance of normal vertical forces in their formation. a very important peculiarity of the folding process is the possibility of the growth of folds together with sedimentation. this process should result in the natural change of thicknesses and facies of individual stratums. their thickness should mostly be reduced in the crests of anticlines and increased in the most depressed parts. the facies in the crests of the anticlinal syngenetic horizon should include several coarse sediment fractions, while in the depressions they should be thinner. the sediments will not “die out” with depth, while the dying out of deformations can occur in the upper intervals. in certain cases, epigenetic deposits can occur at great depths. for example, due to intensive subsidence of ph deposits in the scd, the structures of the absheron peninsula were subjected to strong effect of folding. consequently, the crests of the lower and upper structures of the ph often do not coincide, sometimes shifting over large distances. discussion in general, the formation of large reservoirs at great depths of the scd is determined by the effect of a set of simultaneous factors, the most important of which, in the authors’ opinion, are: 1. the presence of excessive pressure determined by the hc deposit thickness, which height changes depending on the deposit thickness and depth, and hc density. for example, at a depth of 7 km and a deposit thickness of 2000 m, excessive pressure is 14-15 mpa. 2. the presence of anomalously high reservoir pressures in ph, which increase significantly with hypsometric depth. for example, in the study region of the scd, at a depth of 7 km, ahrp level can reach relatively high levels, since in marine fields at a depth of 6-6.5 km, the prv/phyd value varies from 1.3 to 1.7. 3. under intensive subsidence of the scd due to the effect of avalanche sedimentation and flow, the thickness of the ph and its depth created appropriate pressure and temperature conditions at great depths of the scd, where hc migrated from the area with high reservoir pressures to the area with low reservoir pressures. in the conditions of intensive submersion of the scd because of the effect of abovementioned forces, depositional structures are formed and grew, which created large reservoirs. hc migrated from the area of higher initial gradients of reservoir pressure (depressed zones) to areas of low gradients (the most elevated zone). traps, filled to the hinges ended up an enclosed on all sides by high potential zones, where the process of hc migration and accumulation continued, which required a growth of the gas-containing capacity of traps at great depths under high reservoir pressures. at that, high potential zones grew under the effect of abovementioned forces, creating a range of high strains across the entire ph section above the structure. consequently, the mechanism of formation of large reservoirs in the ph at great depths of the scd is determined by a simultaneous effect of a set of factors and the presence of conditions, related to the peculiarities of the geological structure of the depths. the peculiarities are lithofacial change in the sediments of the ph and its intermittent structure along the section, which are characterized by the presence of reservoirs or interconnected reservoirs that form a single hydrodynamic system, on the one hand, and a powerful cap of clay strata with continuous migration of hc under pressure gradients from submerged zones and their accumulation and self-preservation in traps, on the other hand. from this, it follows that at such great depths, small deposits cannot exist. favorable conditions for the natural preservation of reservoir properties are created only in large and thick deposits. the suggested density-structure-fluid theory of formation of large fields in the scd can be used for both explored fields and for explorations in this region, with a view to modelling above mentioned fields. the suggested theory can increase the probability of discovering oil and gas fields during explorations of regions, located in similar geological conditions. at that, the main factor that resists the negative effect of rock pressure on the rp of rocks at great depths is the reservoir pressure. thus, the main factor that resists the negative impact of rock pressure on rp of rocks at great depths is the reservoir pressure, which is directly dependent on its size. under these conditions, hc-containing reservoirs in the thick elevated anticlinal structures have a relatively increased reservoir pressure due to the growth of excessive pressure. the growth of excessive pressure in the elevated part of an anticlinal structure is caused by the difference in the density of deposit water and gas. the excessive pressure increases with temperature growth and the reservoir pressure in the contour, where the relative depth increases, in thick structures in gas fields, is almost transferred to the crest of the deposit. in addition, the deposition depth in the crest of the anticlinal 174 gahraman nariman ogly gahramanov structure decreases significantly compared to the wings. (other sources exist, which facilitate the reduction of rock pressure towards the crest). reducing the effective pressure level facilitates the deceleration of rock compaction, sometimes even decompaction. this, in turn, helps to preserve or even slightly reduce the primary capacity and reservoir properties of rocks (perhaps, form fractures). thus, continuous accumulation of hc in traps, growth of anticlinal structure sizes, especially the increase in the thickness of the structure, deceleration of rock compaction, especially in the elevated part of the structure are interconnected. these facilitate the preservation of large hc clusters at great depths of the scd, which is an important scientific substantiation of hydrocarbon extraction processes and ensures enrichment of scientific knowledge with the possibility of their direct use for the solution of practical problems. conclusions the result of the study is the compilation of a list and characteristics of the set of simultaneously acting factors invoked in forming reservoirs at great depths in the scd. they are the presence of excessive pressure; the presence of abnormally high reservoir pressure in pay horizons, the thickness and occurrence depth of which created respective pressure and temperature conditions at great depths of the scd, as well as developing a model based on the combination of their preconditioning natural simultaneous processes. this paper suggests the possibility of development of autonomous folding due to the formation of internal pressures in the hc field (deposit) itself which the author called the structure-fluid theory of large-amplitude structure formation in the scd. the density-baric model reflects one of the most important physical, geological, and industrial characteristics of reservoirs. consequently, the tectonic factor and its density-baric model can be the main criterion of the primary objects for exploration, the optimal choice of primary programs for drilling, and for the placement of exploratory and industrial wells. this model allows distinguishing potentially dangerous areas for drilling wells at the beginning of the field exploitation. to conclude, it should be noted that both studied types of folding with differences in their formation mechanism, could have correlations that are more complex. in the authors’ opinion, folds of the singular large structures can be relatively simple. noted regularities in the distribution of facies and thicknesses of deposits that form the studied large folds, are convincing proof of the leading role of vertical forces in their formation. references abasov, l. g., kondryushkin, m.t., aliyarov, yu.m., krutykh & r.yu. (2007). study and prediction of parameters of complex natural oil and gas reservoirs of the south caspian depression. baku, azerbaijan: azerbaijan oil&gas. abasov, m. t., kondrushkin, yu.m., aliyarov, r.yu. & krutykh, l.g. (2007). steep methodical approach to the study and forecasting of parameters of natural oil and gas reservoirs of the south caspian basin. oil and gas geology, 2(1), 27-33. abdullaev, n.a., kadirov, f. & guliev, i.s. (2015). subsidence history and basin-fill evolution in the south caspian basin from geophysical mapping, flexural backstripping, forward lithospheric modelling and gravity modelling. in: brunet, m.-f., mccann, t. & sobel t (eds) (2015). geological evolution of central asian basins and the western tien shan range, 427 geological society, london, special publications 175-196. doi: https://doi.org/10.1144/sp427.5. aliyev, a.i (2005). all about oil. baku, azerbaijan: azerbaijan oil&gas.. aliyev, a. i. (2007). south caspian depression: its location among the large and giant oil and gas basins, conditions of formation and preservation of oil and gas deposits, hydrocarbon resource estimation. azarbayan neft tyasyarrtsvaty, 6(1), 15–23. baak, c. g. c. van (2010). glacio-marine transgressions of the early and middle pleistocene caspian basin, azerbaijan. msc thesis, paleomagnetic laboratory ‘fort hoofddijk’, utrecht university, faculty of geosciences. babayev, m.s., sultanov, l.a., ganbarov, sh. & aliyev, t.a. (2014). on the results of petrophysical studies productive strata deposits baku archipelago areas of oil and gas. proceedings of the higher technical educational institutions of azerbaijan, 2(1), 7-12. bochever, a. k. & oradovskaya, f.m. (1972). hydrogeological substantiation of the protection of underground water and water intakes from pollution. moscow, russia: nedra carvalho, i. d. s., mendes, j.c. & costa, t. (2013). the role of fracturing and mineralogical alteration of basement gneiss in the oil exhsudation in the sousa basin (lower cretaceous). northeastern brazil. journal of south american earth sciences, 47(1), 47– 54. eybatov, t.m. (2010). caspian seal (pusa caspica gmel.) endemic of caspian. news of the azerbaijan national academy of sciences, geosciences, 4(1), 151-169. feyzullayev, a.a., tagiyev, m.f. & lerche, i. (2015). on the origin of hydrocarbons in the main lower pliocene reservoirs of the south caspian basin, azerbaijan. energy exploration & exploitation, 33 (1): 1–14. guliyev, e. aliyeva, d. huseynov, a. feyzullayev & p. mamedov (2011). hydrocarbon potential of ultra deep deposits in the south caspian basin. search and discovery article #10312. gurbanov, v.s., sultanov, l.a. & abbasov, g.g. (2014). the lithological and petrographic and reservoir properties of meso-cenozoic deposits of the pre-caspian-guba petroleum district. geophysical news in azerbaijan, 3-4(1), 10-13. hunt, j. (2013). oil and gas geochemistry and geology. moscow, russia: book on demand. kartsev, v. p. vagin, a.a. & shugrin, s.b. (1992). oil and gas hydrogeology. moscow, russia: nedra. kazmin, v.g. & verzhbitskii, e.v. (2011). age and origin of the south caspian basin. oceanology, 51(1), 131–140. kuliev, g.g., djevanshir, r.d., kasimova, s.m. & kulieva, g.z. (1998). folding caused by inner stresses effect in sedimentary basins. proceedings of azerbaijan academy of sciences. the sciences of earth, 2, 97–102. mamedov, p.z. (2008). the reasons for the rapid subsidence of the earth's crust in the south caspian basin. azerbaijan oll industry, 1(1), 8-20. mstislavskaya, v. & filippov, l. (2005). oil and gas geology and exploration. moscow, russia: tsentrlitneftegaz. norman, j. h. (2001). nontechnical guide petroleum geology, exploration, and production. tulsa, oklahoma: pennwell. popkov, v. i., zatsepina, s. v. & mironova, e. a. (2014). reactive-diffusive geo-dynamics of porous media, analytics of navier-stokes equation. iv int. conf. euro-asian community of scientists. modern concepts of scientific studies, moscow, russia, september, 58-61. shaymardanov, v. g., voronin, m.m., orentlikherman, d.v., mikhaylin & e.i. (2003). on the conduction of works in accordance with the arwal (acoustic rehabilitation of wells and layers) technology at extracting wells of ‘pokachevneftegaz’ territorial industrial enterprise. intensification of oil and gas extraction: papers of the international technological symposium, august, 301–308. yusifzade, kh. b. (1985). geological model of formation of deep structures in the scd (problems and estimations). azerbaijan oil ind., 9(1), 23–45. yusifzade, kh. b. (2013). application of modern technologies in the field of exploration and production of oil and gas fields in azerbaijan. azərbaycan neft təsərrüfat, 7-8(1), 3-13. zhang, l. (2012). north ustyurt basin structure evolution and basic petroleum geological characteristics. inner mongolia petrochemical, 3(1), 133-134. earth sciences research journal geotechnical earth sci. res. sj. vol. 16, no. 2 (december, 2012): 13 19 introduction due to their low permeability, clays are the main material used as a liner in solid waste disposal landfills. they become exposed there to various chemical, biological and physical events and the clay is affected by the resulting leachate (yılmaz et al., 2008); researchers are thus interested in surfactants and polymers for modifying clays’ engineering properties. surfactants are surface-active agents which alter fluid interface properties. surfactant modified clay (smc) or surfactant–clay complexes have been considered appropriate landfill liners (lo, 2001; ashmawy et al., 2002; gates et al., 2004; matott et al., 2006) and also potential sorbents for wastewater and contaminated soils (zhu and zhang, 1997; mulligan et al., 1999a, b; mulligan et al., 2001; al-asheh et al., 2003; li et al., 2003; wibulswas, 2004; ghiaci et al., 2004; yang et al., 2005). consistency limits (atterberg limits) have been repeatedly shown to be useful indicators of clay behaviour (jefferson and rogers, 1998; dolinar et al., 2007). reconstituted clay sample liquid and plastic limits can be correlated with various engineering properties, such as specific surface area, cation exchange capacity, permeability, shrinking and swelling behaviour, shear strength and soil compressibility (sharma and lewis, 1994; abdullah et al., 1999; yukselen and kaya, 2006; dolinar et al., 2007). consequently, soil consistency limits are determined, then some other geotechnical properties whose determination may take a long time can be easily estimated with acceptable accuracy. evaluating consistency surfactant modified clays’ consistency limits and contact angles suat akbulut, z. nese kurt and seracettin arasan* ataturk university, civil engineering department, 25240 erzurum, turkey. * corresponding author. phone: + 90 4422314618; fax: + 90 4422360957. e-mail: arasan@atauni.edu.tr (s. arasan). abstract this study was aimed at preparing a surfactant modified clay (smc) and researching the effect of surfactants on clays’ contact angles and consistency limits; clay was thus modified by surfactants for modifying their engineering properties. seven surfactants (trimethylglycine, hydroxyethylcellulose, octyl phenol ethoxylate, linear alkylbenzene sulfonic acid, sodium lauryl ether sulfate, cetyl trimethyl ammonium chloride and quaternised ethoxylated fatty amine) were used as surfactants in this study. the experimental results indicated that smc consistency limits (liquid and plastic limits) changed significantly compared to those of natural clay. plasticity index and liquid limit (pi-ll) values representing soil class approached the a-line when zwitterion, nonionic, and anionic surfactant percentage increased. however, cationic smc became transformed from ch (high plasticity clay) to mh (high plasticity silt) class soils, according to the unified soil classification system (uscs). clay modified with cationic and anionic surfactants gave higher and lower contact angles than natural clay, respectively. resumen este estudio tiene como objetivo la preparación de un surfactante de arcilla modificada (smc) y la investigación de sus efectos en los ángulos de contacto y los límites de consistencia de las arcillas es decir, sus propiedades de ingeniería. siete surfactantes (trimetilglicina, hidroxietilcelulosa, octil fenol etoxilato, ácido lineal alquilbenceno sulfónico, sulfato sódicol auril éter, cloruro de cetil trimetil amonio cuaternario y amina grasa etoxilada) fueron utilizados como en este estudio. los resultados experimentales indicaron que los límites de consistencia de smc (límites líquido y plástico) cambiaron significativamente en comparación con los de arcilla natural. el índice de plasticidad y límite líquido (pi-ll) que representan el tipo de suelo, se acercaron a la línea a cuando aumentó el porcentaje de zwitterion, nonionic y el surfactante aniónico. sin embargo, catiónicos smc se transformaron a partir de ch (arcilla de alta plasticidad) a los suelos de la clase mh (limo alta plasticidad), según el sistema de clasificación de suelos unificado (uscs). la arcilla modificada con surfactantes catiónicos y aniónicos dio mayores y menores ángulos de contacto que la arcilla natural, respectivamente. palabras claves: limite de cosistencia, angulo de contacto, surfactante, arcilla. keywords: consistency limit, contact angle, surfactant, clay. record manuscript received: 14/10/2011 accepted for publications: 06/05/2012 suat akbulut, z. nese kurt and seracettin arasan14 chemical composition clay sio 2 % 53.28 al 2 o 3 % 20.67 fe 2 o 3 % 6.13 cao % 1.71 mgo % 2.82 k 2 o % 0.82 na 2 o % 0.02 ti 2 o % 0.63 loi % 13.9 table 1. the chemical composition of clay. table 2. clay’s index properties table 3. some properties of the surfactants used in the tests index properties clay clay content <0.002 mm (%) 74 finer content <0.075 mm (%) 99 specific gravity gs 2.72 liquid limit wl (%) 88.4 plastic limit wp (%) 38.0 plasticity index ip (%) 50.4 cation exchange capacity (meq./100 g dry soil) 38.59 contact angle o 35 2005; dharaiya and jana, 2005). rogers et al., (2005) examined the contact angles of some compatibilisers for polymer-silicate layer nanocomposites. dharaiya and jana (2005) and cipriano et al., (2005) have also investigated smc contact angles. cipriano et al., (2005) observed that imidazolium bf4 surfactant becomes adsorbed onto a mica surface, resulting in a hydrophobic surface. similarly, jouany and chassin (1987) indicated that a cationic surfactant will become adsorbed with its positively-charged head group next to the negatively-charged clay surface, thereby forcing the hydrophobic surfactant tail to adsorb and become exposed to the solution. thus, while native montmorillonite surface is hydrophilic, adsorption of a small amount of surfactant on the surface can render it hydrophobic. many studies can be found dealing with surfactant (anionic, cationic and nonionic) modified clays (fu and qutubuddin, 2000; gungor et al., 2001; he et al., 2005; isci et al., 2008; guegan et al., 2009; gurses et al., 2009). most have been focused on investigating smcs’ electrokinetic properties, such as zeta potential, cation exchange capacity, electrical conductivity, etc; however , no comprehensive study has been found comparing smc contact angles and consistency limits. this paper was thus aimed at investigating the effect of seven surfactants (zwitterion, nonionic, anionic, and cationic surfactants) on smectite clay’s consistency limits and contact angles. the surfactants used in suspensions were 5% (250 ppm), 10% (500 ppm), and 15% (750 ppm) dry natural clay by weight. materials and methods clays the clayey soil sample used came from a clay-pit in the oltu-narman deposits in erzurum, turkey, classed as high plasticity clay (ch) according to the unified soil classification system (uscs). these deposits are concentrated in two different stratigraphic horizons, namely late oligocene and early miocene sequences. clay-rich fine-grained sedimentary units were deposited in shallow marine and lagoon-type mixed environments. their clay minerals originated from the alteration of eocene calc-alkaline islandarc volcanics, mainly from pyroclastics (trachite and andesite flow) which form the basement for the oltu depression (kalkan and bayraktutan, 2008). x-ray diffraction (xrd) and x-ray fluorescence (xrf) methods (table 1) have been used to identify the major minerals and chemical compounds present in these clays. the soil’s chemical compositions and xrd patterns are given in table 1 and fig. 1, respectively. table 2 gives some of the clay’s index properties. smectite (56%) appeared to be the predominant clay mineral according to x-ray diffraction (figure 1), whereas kaolinite (34%) and illite (3%) appeared in lower proportions. silt and sand-size particles were composed of quartz (4%) and feldspar (3%). surfactant abbreviation formula surfactant type trimethylglycine tmg (ch 3 ) 3 n+ch 2 co 2 zwitterion hydroxyethylcellulose hec (c 6 h 10 o 5 ) n nonionic octyl phenol ethoxylate triton x-100 c 14 h 22 o(c 2 h 4 o) n nonionic linear alkylbenzene sulfonate acid labsa ch 3 (ch 2 ) 11 c 6 h 4 so 3 h anionic sodium lauryl ether sulfate sles ch 3 (ch 2 ) 10 ch 2 (och 2 ch 2 ) n oso 3 na anionic cetyl trimethyl ammonium chloride ctac (c 16 h 33 )n(ch 3 ) 3 cl cationic quaternised ethoxylated fatty amine qefa (fatty amine) r (och 2 ch 2 ) n nh cationic limits provides some very basic mechanical data about a particular soil and a first insight into a particular clay’s chemical reactivity. the liquid limit and plasticity index are mainly influenced by clay minerals’ ability to interact with liquids (schmitz et al., 2004) and hydraulic conductivity tends to decrease when the liquid limit and plasticity index become increased (alawaji, 1999; met et al., 2005; arasan and yetimoglu, 2008). contact angle measurements are often used to indicate clay wettability and interfacial tension (rogers et al., 2005) and provide an insight into surfactant behaviour. the pertinent literature contains only a limited number of studies on smc contact angles (jouany and chassin, 1987; janczuk et al., 1989; lopez-duran et al., 2003; cipriano et al., surfactant modified clays’ consistency limits and contact angles 15 surfactants and modified clay preparation seven surfactants were used in this research; table 3 gives the surfactants’ abbreviation, formula and type. smcs were prepared following the procedure described by xi et al., (2007) and liu et al., (2008). briefly, 40 g of clay was first dispersed in 8 l of deionised water, then stirred with a magnetic stirrer at about 1,000 rpm for about 2 h. previously-prepared surfactant solution was slowly added to the clay suspension at 30oc. all modified products were dried at room temperature. the surfactants were used at 5% (250 ppm), 10% (500 ppm), and 15% (750 ppm) of clay by weight. consistency limits consistency limit tests (i.e. liquid limit and plastic limit) were carried out following the procedure outlined in bs 1377, part 2, 1990. the cone penetrometer (fall cone) method was used to determine liquid limit. specimens were prepared for the liquid limit tests by mixing an air-dried smc mass (passing through a 425-μm sieve). plastic limit tests were performed on material prepared for the liquid limit test; both liquid and plastic limit tests were conducted at room temperature. contact angle measurements a contact angle is defined as being the angle made by the liquid/solid interface intersection with the liquid/air interface; alternately, it can be described as being the angle between a solid sample’s surface and the tangent of a droplet’s ovate shape at the edge of the droplet. a high contact angle indicates low solid surface energy or chemical affinity; this is also referred to as a low degree of wetting. a low contact angle indicates high solid surface energy or chemical affinity and a high, or sometimes complete, degree of wetting (anonymous, 2009). figure 2 illustrates a drop of the reference liquid (water for fig. 2a and air for fig. 2b) resting on a solid surface in the presence of another fluid (air for fig. 2a and water for 2b). the interface between the two fluids meets the solid surface at contact angle  (mitchell and soga, 2005). the simplest way of measuring a contact angle is with a goniometer which allows the user to measure the contact angle visually. the droplet is deposited by a syringe pointing down vertically onto the sample surface and a high-resolution camera captures the image which can then be analysed either by eye (with a protractor) or using image analysis software. the contact angles were measured with a goniometer (cam 101, ksv instruments, finland) in this study, using clay and modified clay pellets. figure 1. a typical xrd analysis result for clay figure 2. wetting two fluids (water and air) on a solid surface (taken from mitchell and soga, 2005) figure 3. wtmg effect on clay’s consistency limits (taken from akbulut et al., 2010). figure 4. hec effect on clay’s consistency limits. water (reference fluid) water (reference fluid)   solid surface solid surface air air (a) (b) c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (tmg/clay), % c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (hec/clay), % plastic limit liquid limit plasticity index plastic limit liquid limit plasticity index suat akbulut, z. nese kurt and seracettin arasan16 results and discussion consistency limits figures 3, 4, 5, 6, 7, 8 and 9 give the variation of modified clays’ consistency limits by tmg, hec, triton x-100 labsa, sles, ctac, and qefa, respectively. for clays modified with zwitterion, nonionic and anionic surfactants (tmg, hec, triton x-100, labsa and sles), it was seen that the liquid limit and plasticity index increased when surfactant percentage was increased. the plastic limit also decreased when increasing the percentage of surfactant. on the other hand, the liquid limit and plasticity index for clays modified with ctac and qefa decreased drastically when surfactant percentage was increased. however, plastic limit values’ variation was insignificant for ctac and qefa (figures 8-9). the consistency limit test results were marked on the cassagrande plasticity chart to determine the new soil classification according to the uscs. figure 10 shows the changes after modification with anionic surfactants and figure 11 shows the changes after modification with cationic surfactants. it can be clearly seen that the points representing soil classification approached the a-line when anionic surfactant percentage was increased. however, cationic surfactants changed the clay class when the figure 5. triton x-100 effect on clay’s consistency limits. figure 7. sles effect on clay’s consistency limits (akbulut et al., 2010). figure 9. qefa effect on clay’s consistency limits. figure 8. ctac effect on clay’s consistency limits. figure 6. labsa effect on clay’s consistency limits (taken from akbulut et al., 2010) c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (triton x-100/clay), % c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (sles/clay), % c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (qefa/clay), % c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (ctac/clay), % c on si st en cy l im it s, % 140 120 100 80 60 40 20 0 0 5 10 15 20 surfactant precent (labsa/clay), % plastic limit liquid limit plasticity index plastic limit liquid limit plasticity index plastic limit liquid limit plasticity index plastic limit liquid limit plasticity index plastic limit liquid limit plasticity index surfactant modified clays’ consistency limits and contact angles 17 percentage was increased, resulting in cationic smcs becoming mh (high plasticity silt) (figure 11). consistency limit test results would suggest that clay water affinity became significantly increased by labsa and hec; however, tmg and sles did not significantly change water affinity. ctac and qefa also decreased water affinity. it should be pointed out that there has been no general consensus regarding the effect of surfactants and chemicals on clays’ consistency limits. most researchers have reported that chemicals have decreased the liquid limit of clays (gleason et al., 1997; shackelford et al., 2000; schmitz et al., 2004; lee et al., 2005; jo et al., 2005). however, little research has indicated that chemicals have increased cl or kaolinite clay liquid limit (rao and mathew, 1995; sivapullaiah and manju, 2005; park et al., 2006; arasan and yetimoglu, 2008). different patterns have most likely arisen from a difference in clay mineralogy and surfactant/chemical type (i.e. zwitterion, anionic, nonionic and/or cationic). in line with previous studies, this research has shown that anionic smcs were the most hydrophilic and cationic smcs were the most hydrophobic (figures 12, 13). nevertheless, it could be said that the net electrical forces between clay mineral layers were affected by surfactant percentage and type; anionic and cationic surfactants would result in an increase and a decrease in net repulsive forces, respectively (increased and decreased repulsive forces cause dispersion and flocculation of clay particles, respectively). figure 12. zwitterion, nonionic and anionic surfactants modified clays’ contact angle results. figure 13. cationic surfactants modified clays’ contact angle results. figure 14. the contact angles for some modified clays. a) contact angle image of natural clay (35o); b) contact angle image of tmg (10%) modified clay (14o); c) contact angle image of qefa (15%) modified clay (54o) figure 10. zwitterion, nonionic, and anionic surfactants modified clay results on the plasticity chart. figure 11. the cationic surfactants modified clays results on the plasticity chart. c on ta n ct a n gl e, ° 50 40 30 20 10 0 0 5 10 15 20 surfactant precent (surfactant/clay), % c on ta n ct a n gl e, ° 90 70 50 30 10 0 5 10 15 20 surfactant precent (surfactant/clay), % tmg labsa sles hec triton x-100 qefa ctac p la st ic it y in de x, % 80 60 40 20 0 0 20 40 60 80 100 120 liquid limit, % p la st ic it y in de x, % 80 60 40 20 0 0 20 40 60 80 100 120 liquid limit, % cl cl ml ml mh mh ch ch a-line a-line tmg labsa sles hec triton x-100 ctac qefa (a) (b) (c) suat akbulut, z. nese kurt and seracettin arasan18 contact angle measurements figure 12 gives the effect of zwitterion, nonionic and anionic surfactants on modified clays’ contact angles and figure 13 shows the cationic surfactants effect. figure 14 presents some images of the smallest contact angle obtained from natural and smcs. figure 12 shows that zwitterion, nonionic and anionic surfactants significantly decreased modified clays’ contact angles; however, cationic surfactants increased the contact angles (figure 13). tmg and qefa were the most effective surfactants when contact angle results were taken into consideration; similar to consistency limit results, contact angle measurements indicated that clay water affinity was increased by zwitterion, nonionic and anionic surfactants and also became decreased by cationic surfactants. cipriano et al., (2005) indicated that cationic surfactants increased modified clays’ contact angles and produced a hydrophobic surface. conclusions the following conclusions were thus drawn: • consistency limits were significantly changed compared to those for natural clay. the points representing soil class came further towards the a-line when zwitterion, nonionic and anionic surfactant percentage increased. cationic surfactants changed the clay classification from ch to high plasticity silt (mh) when the percentage of surfactant added to the clay was increased; • clays modified with zwitterion, nonionic and anionic surfactants gave the lowest contact angles compared to those for natural clay; however, the clays modified with cationic surfactants gave the highest contact angles; and • it could also be said that clay water affinity was increased by zwitterion (tmg), nonionic (hec, triton x-100) and anionic surfactants (labsa, sles), also that cationic surfactants (ctac and qefa) decreased the water affinity used in this research. hence, zwitterion, nonionic and anionic smcs may be used as hydrophilic materials in waste water remediation and the cationic smcs may also be used as hydrophobic materials (liner) in waste disposal landfills and dams. it should be pointed out that further studies on smcs’ engineering properties (e.g. xrd, xrf, dta and tg for mineralogy and cation exchange capacity, zeta potential for electro-kinetic properties) are needed to make more reasonable judgments. such studies should aim at explaining smc behaviour more reasonably. acknowledgements we gratefully acknowledge tubitak’s financial support (grant 107y295). references abdullah, w.s., alshibli, k.a., al-zou’bi, m.s., 1999. influence of pore water chemistry on the swelling behavior of compacted clays. applied clay science, 15: 447-462. akbulut, s., arasan, s., acikyildiz, m., 2010. water affinity of surfactant modified clay. proceedings of the first makassar international conference on civil engineering (micce2010), isbn 978-60295227-0-9. alawaji, h.a., 1999. swell and compressibility characteristics of sandbentonite mixtures inundated with liquids. applied clay science 15,411-430. al-asheh, s., banat, f., abu-aitah, l., 2003. adsorption of phenol using different types of activated bentonites, sep. purif. technol. 33: 1–10. anonymous, 2009. http://en.wikipedia.org/wiki/contact_angle arasan, s., yetimoglu, t. 2008. effect of salt solutions on the consistency limits of two clays. turkish journal of engineering and environmental sciences, 32: 107-115. ashmawy, a.k., el-hajji, d., sotelo, n., muhammad, n., 2002. hydraulıc performance of untreated and polymer-treated bentonite in inorganic landfill leachates. clays and clay minerals, 50:546-552. cipriano, b.h., raghavan, s.r. mcguiggan, p.m. 2005. surface tension and contact angle measurements of a hexadecylimidazolium surfactant adsorbed on a clay surface. colloids and surfaces a: physicochem. eng. aspects 262: 8–13. dharaiya, d., jana, s.c., 2005. thermal decomposition of alkyl ammonium ions and its effects on surface polarity of organically treated nanoclay polymer, 46(23):10139-10147 dolinar, b., misic, m., trauner, l., 2007. correlation between surface area and atterberg limits of fine-grained soils. clays and clay minerals, 55: 519-523. fu, x., qutubuddin, s., 2000. synthesis of polystyrene–clay nanocomposites. materials letters, 42:12–15. gates, w.p., nefiodovas, a., peter, p., 2004. permeability of an organo-modified bentonite to ethanol-water solutions. clays and clay minerals, 52(2):192-203. ghiaci, m., abbaspur, a., kia, r., seyedeyn-azad, f., 2004. equilibrium isotherm studies for the sorption of benzene, toluene, and phenol onto organozeolites and assynthesized mcm-41, sep. purif. technol. 40: 217–229. gleason, m.h., daniel, d.e., eykholt, g.r., 1997. calcium and sodium bentonite for hydraulic containment applications. j. geotech. geoenv. eng. 123(5), 438-445. guegan, r., gautier, m., beny, j.m., muller, f., 2009. adsorptıon of a c10e3 non-ionic surfactant on a ca-smectite. clays and clay minerals, 57: 50 -509. gungor, n., alemdar, a., atici, o., ece, i.o., 2001. the effect of sds surfactant on the flow and zeta potential of bentonite suspensions. materials letters, 51: 250–254 gurses, a., karaca, s., acikyildiz, m., ejder (korucu), m., 2009. thermodynamics and mechanism of cetyltrimethylammonium adsorbtion onto clayey soil from aqueous solutions. chemical engineering journal, 147: 194-201. he, h., ding, z., zhu, j., yuan, p., xi, y., yang, d., frost, r.l., 2005. thermal characterization of surfactant-modified montmorillonites clays and clay minerals, 53: 287 -293. isci, s., gunister, e., alemdar, a., ece, o.i., gungor, n., 2008. the influence of dtabr surfactant on the electrokinetic and rheological properties of soda-activated bentonite dispersions. materials letters, 62:81–84. janczuk, b., chibowski, e., hajnos, m., bialopiotrowicz, t., stawinski, j., 1989. influence of exchangeable cations on the surface free energy of kaolinite as determined from contact angles. clays and clay minerals, 37: 269-272. jefferson, i., rogers, c.d.f., 1998. liquid limit and the temperature sensitivity of clays. engineering geology, 4(9), 95-109. jo, h.y., benson, c.h., shackelford, c.d., lee, j.m., edil, t.b., 2005. long-term hydraulic conductivity of a geosynthetic clay liner permeated with inorganic salt solutions. j. geotech. geoenv. eng. 131(4),405-417. jouany, c., chassin, p., 1987. determination of the surface energy of clay-organic complexes by contact angle measurements. colloids surface, 27 (4): 289–303. kalkan e, bayraktutan ms (2008) geotechnical evaluation of turkish clay deposits: a case study in northern turkey. environmental geology, 55:937–950. surfactant modified clays’ consistency limits and contact angles 19 lee, j.m., shackelford, c.d., benson, c.h., jo, h.y., edil, t.b., 2005. correlating index properties and hydraulic conductivity of geosynthetic clay liners. j. geotech. geoenv. eng., 131(11), 1319-1329. li, z., willms, c.a., kniola, k., 2003. removal of anionic contaminants using surfactant-modified palygorskite and sepiolite. clays and clay minerals, 51: 445-451. liu, r., frost, r.l., martens, w.n., yuan, y., 2008. synthesis, characterization of mono, di and tri alkyl surfactant intercalated wyoming montmorillonite for the removal of phenol from aqueous systems. journal of colloid and interface science, 32(2): 287-294. lo, i.m.c., 2001. organoclaywith soil–bentonite admixture aswaste contaminant barriers, j. environ. eng. 127: 154–161. lópez-durán, j.d.g., khaldoun, a., kerkeb, m.l., ramos-tejada, m.m., gonzález-caballero, f., 2003. wettability of montmorillonite clays in humic acid solutions. clays and clay minerals, 51: 65-74. matott, l.s., bartelt-hunt, s.l., rabideau, a.j., fowler, k.r., 2006. application of heuristic optimization techniques and algorithm tuning to multilayered sorptive barrier design, environ. sci. technol. 40: 6354–6360. met, i., akgun, h., turkmenoglu, a.g., 2005. environmental geological and geotechnical investigations related to the potential use of ankara clay as a compacted landfill liner material, turkey. environmental geology 47, 225-236. mitchell, j.k., soga, k., 2005. fundamentals of soil behavior, 3rd edition, john wiley & sons, inc., hoboken, new jersey, p 577. mulligan, c.n., yong, r.n. and gibbs, b.f., 1999a. on the use of biosurfactants fort he removal of heavy metals from oil-contaminated soil. environ. prog. 18: 31–35. mulligan, c.n., yong, r.n. and gibbs, b.f., 1999b. removal of heavy metals from contaminated soil and sediments using the biosurfactant surfactin. j. soil contam. 8: 231–254. mulligan, c.n., yong, r.n. and gibbs, b.f., 2001. surfactant-enhanced remediation of contaminated soil: a review. engineering geology. 60: 371-380. park, j., vipulanandan, c., kim, j.w., oh, m.h., 2006. effects of surfactants and electrolyte solutions on the properties of soil. eviron geol. 49: 977-989. rao, s. n. and mathew, p. k., 1995. effects of exchangeable cations on hydraulic conductivity of a marine clay. clays and clay minerals, 43(4), 433-437. rogers k., takacs e., thompson m.r., 2005. contact angle measurement of select compatibilizers for polymer-silicate layer nanocomposites. polymer testing, 24 (4): 423-427. schmitz, r.m., schroeder, c., charlier, r.,2004. chemo-mechanical interactions in clay: a correlation between clay mineralogy and atterberg limits. applied clay science, 26,351-358. shackelford, c.d., benson, c.h., katsumi, t., edil, t.b., lin, l., 2000. evaluating the hydraulic conductivity of gcls permeated with non-standart liquids. geotextiles and geomembranes, 18,133161. sharma, h.d., lewis, s.p., 1994. waste containment systems, waste stabilization, and landfills: design and evaluation. john wiley & sons inc., canada, pp.588. sivapullaiah, p.v., manju, “kaolinite-alkali interaction and effects on basic properties”, geotechnical and geological engineering, 23,601-614, 2005. wibulswas, r., 2004. batch and fixed bed sorption of methylene blue on precursor and qacs modified montmorillonite, sep. purif. technol. 39: 3–12. xi, y., zhou, q., frost, r.l., he, h., 2007. thermal stability of octadecyltrimethylammonium bromide modified montmorillonite organoclay. journal of colloid and interface science, 311(2):347-353. yang, j.w., lee, y.j., park, j.y., kim, s.j., lee, j.y., 2005. application of apg and calfax 16l-35 on surfactant-enhanced electrokinetic removal of phenanthrene from kaolinite. engineering geology, 77: 243–251. yılmaz, g., yetimoglu, t., arasan, s., 2008.  hydraulic conductivity of compacted clay liners permeated with inorganic salt solutions. waste management & research, 26: 464-473. yukselen, y., kaya, a., 2006. prediction of cation exchange capacity from soil index properties. clay minerals, 41: 827-837 zhu, l., li, y., zhang, j., 1997. sorption of organo-bentonites to some organic pollutants in water. environ. sci. technol. 31: 1407–1410. claudiabravo.indd 132 earth sciences research journal earth sci. res. j. vol. 9, no. 2 (dec. 2005): 132-138 (manuscript received august 2005, paper accepted november 2005) wave equation datuming to correct topography effect on foothill seismic data claudia lily bravo niño1 and luis alfredo montes vides2 1 sandox co. bogotá. colombia. e mail: sandox@col.net.co 2 universidad nacional de colombia. departamento de geociencias. e mail: lamontesv@unal.edu.co resumen en el procesamiento convencional se aplican las correcciones estáticas para atenuar el efecto de la topografía, basado en la suposición de una velocidad en la capa superficial mucho menor que la del substrato, lo cual hace que los rayos viajen en una trayectoria cercana a la vertical. sin embargo, cuando el contraste de velocidades entre esas capas no es grande, las trayectorias de los rayos que emergen no son verticales, por lo que las correcciones estáticas posicionan erradamente los reflectores. una mejor alternativa para corregir este efecto es continuar el campo de onda hasta una superficie plana (datum), porque no asume una trayectoria vertical, solucionando la ecuación de onda acústica para extrapolar fuentes y receptores. para conocer el desempeño del algoritmo de kirchhoff se usó para continuar el campo de onda de datos sintéticos. primero se continuaron fuentes y receptores hasta la base de la capa meteorizada y luego hacia arriba hasta un datum plano. comparando los resultados se observa que la solución por continuación del campo de onda provee un realce notable de los reflectores, que exhiben una mejor continuidad y un incremento en el contenido de frecuencia. palabras clave: campo de onda, topografía rugosa, superficie de referencia, trayectoria del rayo, continuación del campo de onda, extrapolación, integral de kirchhoff. abstract the current seismic processing applies static corrections to overcome the effects associated to rough topography, based in the assumption that velocity in near surface is lower than in the substratum, which force going up rays travel near to vertical. however, when the velocity contrast between these layers is not large enough, the trajectory of the up going rays deviate from vertical raveling the reflectors erroneously. a better alternative to correct this is to continue the wave field to a datum, because it does not assume vertical ray trajectory and solves the acoustic wave equation to extrapolate sources and receivers. the kirchhoff approach was tested in synthetic shots continuing their wave field to a datum and finally it was applied instead of static corrections in real data acquired in foothill zones. first shot and receiver gathers were downward continued to the base of weathering layer and later upward continued to a final flat datum. comparing the obtained results we observed that continuation approach provides a noticeable enhancement of reflectors in seismic records, displaying a better continuity of the reflectors and an increment in frequency content. 133 wake equation datuming to correct topography effec on foothill seismic data key words: wave field, rugged topography, datum, ray path, wave equation datuming, extrapolation, kirchhoff integral. © 2005 esrj unibiblos introduction in exploration foothill zones, rugged surface cause damage in seismic data quality. statics corrections have been the conventional technique proposed to overcome this problem. in another approach, berryhill presented the concept of wave field continuation, named wave equation datuming (wed) based in an extrapolation scheme using the kirchhoff integral, applying it to poststack (1979) and prestack data (1984). other solutions emerged based on finite difference (claerbout, 1985) and phase-shift (gazdag, 1978). reshef (1991) gave a datuming principle that permits direct prestack migration of data recorded on a not flat topographic surface. yilmaz and lucas (1986) replaced a layer of arbitrary shape and known velocity with another one with different velocity in order to eliminate bending trajectories in the area with severe velocity contrast. bevc (1997) upward continued the data from the surface to a flat datum above the highest topography using a constant velocity simulating a flooded topography effect without requiring a prior knowledge of the replacement velocity. hindriks and verschuur (2001) presented a redatuming approach based on shot gather amplitude using the cfp (common focus point) technology. benxi and others (2004) presented a novel wave equation datuming method which just uses a single shot gather to downward extrapolate the receivers and vertically time shift the source from the surface to the datum, using the near surface velocity model and depth-velocity curve at the source location respectively. recently effects of topography in foothill data have been studied (malaver et al., 2005; montes et al 2003) and new approaches have been proposed to increase quality in seismic data recorded in these regions (montes et al. 2005). wed approach was applied to a seismic line acquired in region with rough topography and processed using similar seismic steps applied in a flow with static corrections, the obtained results showed an improvement in image quality and a increment in frequency content. static corrections its objective is determine the arrival times of the reflections that could be observed if all measures were done on a flat datum without weathering layer. so, time corrections are applied to data recorded in a, b and c positions of figure 1 to convert them in traces that should be observed if the data were registered in a d , b d and c d points on flat datum. figure 1. static corrections: time correction from the surface to the bottom weathering layer and to the datum figure 2. observe the rays deviating from the vertical trajectory as assumed by static corrections. static corrections estimation needs the replacement velocity, especially if there is a significant change in elevation or in the base of the lvl, a wrong value can produce false structures in the final section. the 134 claudia lily bravo niño and luis alfredo montes vides replacement velocity normally is determined from well data, seismic refraction or tomography. the conventional static corrections apply vertical time shift assuming the waves propagate almost vertically due to the high velocity contrast between the near surface and the substrate as seen in figure 2. wave equation datuming the wed can be done through extrapolation methods as kirchhoff integral which is a solution of the acoustic wave equation, the integral equation corresponds to: , , cos p x z t da r2 1 j 0 2= r i ^ h # , , , ,p x z c r c r t p x z c r i ij i ij0 0 0 02 2 + -x xb bl l; e p j is the scalar wave field propagating with velocity c /r cij =x and is the time used to travel from pi (x0, y 0 , z 0 ) on the input datum to the point p j (x, y, z) on the output datum. the objective is to calculate p j for all times at point (x, y, z) on an infinite flat surface a 0 as seen in figure 3. the wave equation datuming is a two-step process: first it is applied to all shot gathers and later to the receiver gathers, extrapolating shot and receiver gathers separately. observed in figure 5a recorded on a flat surface and in figure 5b affected by topography. this dataset was created using a complete processing sequence in seismic unix (the colorado school of mines), because promax does not include wave field continuation through kirchhoff integral. figure 3. using data recorded in an input datum the wave field value in an output datum is estimated. methodology and aplication a synthetic dataset was generated in a two layers model (figure 4), 180 raw records with 25 m shot interval and 8 ms sample rate were obtained. the shots were upward continued to a rough surface through wed, simulating seismic data recorded in foothill zones, one of these synthetic shots is figure 4. two layers model used to generate synthetic data. figure 5. (a) synthetic shot registered on a flat surface, (b) synthetic shot affected by topography. then, the shots were continued to a plane datum. the shot 400 generate on the flat surface is observed in figure 6a, this shot affected by topography is shown in figure 5b and the shot after wed applied to eliminate the effects introduced by the topography is observed in figure 6b. comparing figures 6a and 6b, it is noticeable the kirchhoff integral performance to unravel the reflector. in the second step the wed approach was applied to a 2d seismic line of the colombian foothill area, with 140 m of topographic variations as can be observed in figure 7, 12 km length, 240 channels, 344 shots, 15 m group interval and 15 m shot interval. 135 wake equation datuming to correct topography effec on foothill seismic data (a) (b the stacked section furnished by the use of static corrections is showed in figure 11, there some structures and reflectors are observed, giving an approximation of the geologic model in the zone. at the top-left part of the figure 11 a package of reflectors is observed which conforms the crest of an anticline, being visible until 800 ms. but below the anticline is hard to visualize other geologic features, except between 800 and 1500 ms where there are weak evidence of reflectors. at the central part of the image an anticline structure is seen and in the right part the image is unclear. the stacked section of the figure 12 contains the stacked section obtained replacing wed instead of static corrections in the processing flow, where the reflectors of the left part are enhanced exhibiting a nice continuity also in deeper zone. the right part of the image displays more information content than figure 11, being notorious the anticline structure in the upper section that descends to create a syncline structure. besides, some diffraction points are observed in this central zone indicating the existence of a fault described in previously done surficial geological reports. the presence in figure 11 of unidentified seismic signals crossing the section dipping in both left and right directions are strongly attenuated in figure 12, allowing to depict the geological structure of zone below the foothill zone. conclusions in the conventional processing sequence, vertical static time shift is used to carry out sources and receivers to a flat datum, based in the assumption than the near surface velocity is smaller than the substrate, which is correct in some areas, where its use can correct the nmo curve distorted by the topography. however when the contrast is not great enough the ray path from substrate is not vertical. so, the conventional processing may conduce to wrong location of reflectors. the comparison between the shot gathers and the images obtained by the two considered approaches indicates that the application of wed to attenuate the effect introduced in seismic records by rough topography provides images with higher information content that those images furnished when static corrections are applied in this kind of zones. the nmo curves corrected by wed applications ensures that others methods based in this assumption figure 6. (a) synthetic shot registered in flat surface, (b) shot after corrected by wed. figure 7. topography of the acquired seismic survey the line was processed using conventional sequence in promax including filters, gain, amplitude recovery, static refraction, nmo, residual static to obtain finally a stacked section. the line was processed using a second processing sequence, but instead of use static corrections wed was applied using seismic unix (cwp-su). the difference between the images provided by the two processing sequences is associated to this different step. discussions the raw shot 78 with geometry assigned is showed in figure 8, the same shot after applied static corrections is observed in figure 9 whereas the shot after applied wed is displayed in figure 10. comparing the images in figures 9 and 10 it is noticeable the difference among them. in the figure 9 it is hard to appreciate any defined reflector, whereas in figure 10 several reflectors observed clearly around 1000, 1600, 2000 and 3000 ms, in addition other events with limited continuity are observed. similar results were obtained in others shot gathers, confirming as expected, that the wed enhance the seismic information contained in a record acquired in zones with rough topography. 136 claudia lily bravo niño and luis alfredo montes vides can be correctly applied or others analysis like avo provides information consistent with their own hypothesis. the stacked sections obtained with wed are structurally consistent diminishing the interpretation risk. otherwise although the shot gathers provided by wed should be processed using a different flow sequence to obtain a clearer image, in order to compare the methods performance they were submitted to the same sequence. the wave field continuation method is a solution of the wave equation which allows to sources and receivers at whatever datum, without assumptions about the velocity contrast in shallow layers model. this method has into account the ray path when the velocity contrast between the substrate and the near surface is small, validating the snell law. references • benxi ke, bo zhao, chaoying liu and yunfeng fang. 2004 wave-equation datuming based on a single shot gather. seg expanded abstracts ann. met.23, 2156. • berryhill, j. r., 1979, wave equation datuming. geophysics 44, 1329-1344. • berryhill, j. r., 1984, wave equation datuming before stack. geophysics 49, 2064-2066. • bevc d.,1997, flooding the topography: wave equation datuming of land data with rugged acquisition topography. geophysics 62, 1558-1569. • claerbout, j. f., 1985, imaging the earth’s interior: blackwell scientific publications. • gazdag, j., 1978, wave equation migration with the phase shift method: geophysics 43, pag. 1342-1351. • hindriks c. o. h. and verschuur d. j. 2001, an imaging approach to complex near surface effects eage 2001 p-163 63rd conference & technical exhibition —amsterdam, the netherlands. • malaver l. and montes l. 2005. elastic approach to remove surficial waves from shot gathers. earth sciences researh journal. vol. 9. no. 1, 65-69. • montes l ,vargas c and pérez g. 2005 modeling and removal of back-scattered noise from rough topography in land seismic data, ct&f vol. 3, núm. 1, 131-138. • montes l., quintana r., cespedes s., espíndola n., salinas t. and pérez g. 2003. attenuation of the seismic dispersion associated to foothills topography: application to real data. ct&f vol. 2, núm. 4, 7-12. • reshef, m., 1991, depth migration from irregular surfaces with depth extrapolation methods (short note): geophysics 56, pag. 119-122. • yilmaz, o. and lucas d. 1986. prestack layer replacement. geophysics 51, 1355. figure 8. shot 78 with geometry assigned. 137 wake equation datuming to correct topography effec on foothill seismic data figure 9. shot 78 after applying static corrections. figure 10. shot 78 after applied wed. 138 claudia lily bravo niño and luis alfredo montes vides figure 11. stacked section obtained with static corrections. figure 12. stacked sections provided by the use of wed. lake water temperature is one of the key parameters in determining the ecological conditions within a lake, as it influences both chemical and biological processes. therefore, accurate prediction of water temperature is crucially important for lake management. in this paper, the performance of soft computing techniques including gene expression programming (gep), which is a variant of genetic programming (gp), adaptive neuro fuzzy inference system (anfis) and artificial neural networks (anns) to predict hourly water temperature at a buoy station in the yuan-yang lake (yyl) in north-central taiwan at various measured depths was evaluated. to evaluate the performance of the soft computing techniques, three different statistical indicators were used, including the root mean squared error (rmse), the mean absolute error (mae), and the coefficient of correlation (r). results showed that the gep had the best performances among other studied methods in the prediction of hourly water temperature at 0, 2 and 3 meter depths below water surface, but there was a different trend in the 1 meter depth below water surface. in this depth, the ann had better accuracy than the gep and anfis. despite the error (rmse value) is smaller in ann than gep, there is an upper bound in scatter plot of ann that imposes a constant value, which is not suitable for predictive purposes. as a conclusion, results from the current study demonstrated that gep provided moderately reasonable trends for the prediction of hourly water temperature in different depths. la temperatura del agua es uno de los parámetros básicos para determinar las condiciones ecológicas de un lago, ya que está influenciada por procesos químicos y biológicos. además, la exactitud en la predicción de la temperatura del agua es esencial para el manejo del lago. en este artículo se evalúa el desempeño de técnicas de soft computing como la programación de expresiones de genes (peg), que es una variante de la programación genética (pg), el sistema neuro-fuzzy de inferencia adaptativa (anfis, en inglés) y las redes neuronales artificiales (rna) para predecir la temperatura del agua en diferentes niveles de una estación flotante del lago yuan-yang (yyl), en el centro-norte de taiwán. se utilizaron tres indicadores estadísticos, el error cuadrático medio (ecm), el error absoluto medio (mae, en inglés) y el coeficiente de correlación (r) para evaluar el desempeño de las técnicas de computación. los resultados muestran que la peg es más exacta en la predicción de la temperatura del agua entre 0,2 y 3 metros de profundidad. sin embargo, se evidencia una tendencia diferente a partir del metro de profundidad. a esta distancia de la superficie, las rna son más exactas que la peg y el anfis. los resultados de este estudio probaron claramente la usabilidad del peg y las rna en la predicción de la temperatura del agua a diferentes profundidades. abstract resumen keywords: soft computing techniques, statistical indicators, subalpine lake, water temperature. palabras clave: técnicas soft computing, indicadores estadísticos, lago subalpino, temperatura del agua. water temperature prediction in a subtropical subalpine lake using soft computing techniques record manuscript received: 29/04/2014 accepted for publication: 26/02/2016 how to cite item earth sciences research journal, 20(2), b-b. doi:http://dx.doi.org/10.15446/esrj.v20n2.43199 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.43199 saeed samadianfard1,*, honeyeh kazemi1, ozgur kisi2, wen-cheng liu3 1department of water engineering, university of tabriz, tabriz, iran (*corresponding author, e-mail: s.samadian@tabrizu.ac.ir, tel: +989141101845, fax: +984113356007) 2department of civil engineering, canik basari university, samsun, turkey 3department of civil and disaster prevention engineering, national united university, miao-li, taiwan earth sciences research journal earth sci. res. j. vol. 20, no. 2 ( june, 2016 ) : d1-d11. h y d r o l o g y d2 saeed samadianfard, honeyeh kazemi, ozgur kisi and wen-cheng liu 1. introduction water temperature is a fundamental physical property with a direct impact on all organisms inhabiting the aquatic environment (webb et al., 2008). because predicting the water temperature is important for maintaining water quality and for ecosystem management, several authors have investigated methods for simulating water temperature of lakes (lawrence et al., 2002; lee et al., 2009; schwab et al., 2009). hondzo and stefan (1996) simulated daily water temperature and dissolved oxygen profiles in minnesota lakes by deterministic process-based water quality models with daily meteorological conditions as input. derived empirical formulas for lake water quality and stratification indicators from the simulation results gave good predictions of temperature and dissolved oxygen characteristics estimated from measurements in seven minnesota lakes. fang and stefan (1996) substantially modified a water temperature and ice cover model for freshwater lakes and combined it with a summer model to simulate water temperature structures and ice thickness in two small lakes in the north central us. the best value of volume averaged water temperature for the two studied small lakes was higher than was previously found for a larger lake. benyahya et al. (2007) provided an overview of the existing statistical water temperature models. they categorized them in two major groups: deterministic and statistical/stochastic models. they stated that the deterministic models require numerous input data and they are appropriate for analyzing different impact scenarios, but the main advantage of the statistical models is their relative simplicity and relative minimal data requirement. sharma et al. (2008) developed models to predict annual maximum near-surface lake water temperatures for lakes across canada using four statistical approaches: multiple regression, regression tree, artificial neural networks and bayesian multiple regression. although artificial neural networks were marginally better for three of the four data sets, multiple regression was considered to provide the best solution based on the combination of model performance and computational complexity. trumpickas et al. (2009) tried to construct empirical relationships between surface water temperatures and local air temperatures that could be used to estimate future water temperatures using future air temperatures generated by global climate models. zhao et al. (2011) calculated the surface water temperature, deep water temperature and mean annual epilimnetic temperature and compared the obtained values with empirical data in lake taihu by using the dynamic water temperature model. the simulated values were consistent with empirical data. results showed that this model could be used for the temperature simulation in the studied lake. thiery at al. (2014) evaluated a set of one dimensional lake models for lake kivu, east africa. in this study, meteorological observations from two automatic weather stations were used to drive the models, whereas a unique dataset, containing over 150 temperature profiles recorded since 2002, was used to assess the model’s performance. simulations were performed over the freshwater layer only (60 m) and over the average lake depth (240 m). the good agreement between the deep simulations and the observed meromictic stratification also showed that a subset of models was able to account for the salinityand geothermal induced effects upon deep-water stratification. also, some attempts have been made to relate lake water temperature to meteorological parameters such as air temperature. in this context, piccolroaz et al. (2013) developed air2water, a simple physically based model to relate the temperature of the lake superior (usa–canada), considering a 27 years record of measurements, to air temperature only. the results proved that their model was suitable to be applied over long timescales (from monthly to interannual), and could be easily used to predict the response of a lake to climate change, since projected air temperatures were usually available by large-scale global circulation models. toffolon et al. (2014) reconstructed temperature of the surface layer of temperate lakes by means of a simplified model on the basis of air temperature alone. the comparison between calculated and observed data showed a remarkable agreement for all 14 lakes investigated (mara, sparkling, superior, michigan, huron, erie, ontario, biel, zurich, constance, garda, neusiedl, balaton, and baikal, in west-to-east order), which presented a wide range of morphological and hydrological characteristics. given the significant impact of lakes on surface atmosphere interactions, the need for an accurate simulations of lake temperatures at different depths arises. for this purpose, soft computing techniques such as gene expression programming (gep), adaptive neuro fuzzy inference system (anfis) and artificial neural networks (ann) can play undeniable roles in mentioned simulations. gene expression programming (gep) has been applied to a wide range of problems in artificial intelligence, artificial life, engineering and science. gep can be successively applied to areas where (i) the interrelationships among the relevant variables are poorly understood (or where it is suspected that the current understanding may well be wrong), (ii) finding the size and shape of the ultimate solution is difficult, (iii) conventional mathematical analysis does not, or cannot, provide analytical solutions, (iv) an approximate solution is acceptable (or is the only result that is ever likely to be obtained), (v) small improvements in performance are routinely measured (or easily measurable) and highly prized, (vi) there is a large amount of data in computer readable form, that requires examination, classification, and integration, e.g., molecular biology for protein and dna sequences, astronomical data, satellite observation data, financial data, marketing transaction data, or data on the world wide web (banzhaf et al. 1998). during the last decade, genetic programming has been used as a viable alternative approach to physical models. aytek and kisi (2008) applied gp to suspended sediment transport, and found it to perform better than conventional rating curve and multi-linear regression techniques. shiri and kisi (2011) compared gep and anfis methods for predicting groundwater table depth fluctuations and found gep to be better than anfis in this regard. samadianfard (2012) examined the potential of the gep technique in estimating flow friction factor in comparison with the most currently available explicit alternatives to the colebrook–white equation. results revealed that by using gep, the friction factor could be identified precisely. samadianfard et al. (2012) studied the capabilities of the gp in simulating the wetting patterns of drip irrigation. results showed that the gp method had good agreement with the results of hydrus 2d software considering the full set of operators in estimation of radius and depth of wetting patterns. also results obtained from field experimental in a sandy loam soil showed reasonable agreement with the gp results. results of the study demonstrated the usefulness of the gp method for estimating wetting patterns of drip irrigation. anfis is a neuro-fuzzy system, which uses a feed-forward network to search for fuzzy decision rules that perform well on a given task. using a given input/output data set, anfis creates a fuzzy inference system whose membership function parameters are adjusted using a back-propagation algorithm alone or a combination of a back-propagation algorithm with a least mean square method. this allows the fuzzy systems to learn from the data being modeled. kisi (2006) investigated the ability of anfis technique to improve the accuracy of daily evaporation estimation. based on his results, the anfis computing technique could be used successfully in modeling evaporation process from the available climatic data. shiri et al. (2011) compared anfis to ann to estimate daily pan evaporation values from climatic data and found anfis to be better than ann. the artificial neural network (ann) approach provides a viable solution to the environmental problems because it is based on training not on analytical models or statistical assumptions. ann models can be trained to predict results from examples and once trained; they can perform predictions at very high speed (mellit et al., 2006). ann is an intelligent data-driven modeling tool that is able to capture and represent complex and non-linear input/output relationships. anns are massively parallel, distributed processing systems that can continuously improve their performance via dynamic learning. moghaddamnia et al. (2009) explored evaporation estimation methods based on ann and anfis techniques. they found that ann and anfis techniques had much better performances than the empirical formulas. zaier et al. (2010) developed ann ensemble models to improve the results of single artificial neural network (single ann) for the estimation of the ice thickness in a number of selected canadian lakes during the early winter ice growth period. ann ensemble models for the estimation of ice thickness proved to be more accurate than single ann models. kisi et al. (2012) applied three artificial intelligence approaches, d3water temperature prediction in a subtropical subalpine lake using soft computing techniques namely anns, anfis and gep to forecast daily lake-level variations of lake iznik in turkey. the results obtained by the gep approach indicated that it performed better than anfis and anns in predicting lake-level variations. liu and chen (2012) compared the performance of the ann technique with a physically based three-dimensional circulation model for prediction of water temperature at a buoy station in the yuan-yang lake in north-central taiwan at various measured depths. the simulated results revealed that the accuracy of the three-dimensional circulation model was better than the ann model. fallah-mehdipour et al. (2013) investigated the capability of anfis and gp as two artificial intelligence tools to predict and simulate groundwater levels in three observation wells in the karaj plain, iran. results indicated that gp yielded more appropriate results than anfis when different combinations of input data were employed in both prediction and simulation processes. the main objective of this paper is to investigate the accuracy of soft computing techniques such as gene expression programming, adaptive neuro fuzzy inference system and artificial neural network methods for the prediction of hourly water temperature in a lake at different measured depths. some statistical parameters for error estimation are used herein as comparing criteria for the evaluation of the performance of the studied models. 2. material and methods 2.1. study site and data collection yuan-yang lake (yyl, 24°34’60.00’n, 121°24’0.00’e, area= 3.7×104 m2; maximum depth= 4.5 m) in north-central taiwan is a subtropical, humid lake in the clilan national forest preserve (figure 1). figure 2 shows the bathymetry of the yyl (contours in m). also, yyl is a subalpine lake situated in a natural reserve area that has been virtually undisturbed by human activity for a long time (chen and wu, 1999). it is surrounded by a cloud belt forest of taiwan yellow cypress, an important timber tree and a relic species. the lake is located 1670 m above the mean sea level which is at the subalpine region. the geography of the drainage basin allows large quantities of terrestrial runoff from the surrounding mountains to enter the yyl (wu et al., 2001). the mean annual air temperature is approximately 13°c (monthly averages range from -5 to 15°c), and annual precipitation can exceed 4000 mm. wind speed over the lake, which was measured 1 m above the lake by an anemometer, (between 0 to 4.220 m.s-1) is relatively weak. the dominant wind directions are from the east and the south west because of the v-shaped valley facing east to west. concerns about the water quality in yyl, have been rapidly increasing recently due to the natural and anthropogenic pollution. in order to understand the underlying physical and chemical processes as well as their associated spatial distribution in yyl, liu et al. (2011) analyzed fourteen physico-chemical water quality parameters recorded at the eight sampling stations by using multivariate statistical techniques and a geostatistical method. their results showed that four principal components i.e., nitrogen nutrients, meteorological factor, turbidity and nitrate factors, account for 65.52% of the total variance among the water quality parameters. the spatial distribution of principal components further confirmed that nitrogen sources constitute an important pollutant contribution in the yyl. since april 2004, the deepest point of yyl (approximately 4.5 m) was instrumented with a buoy that measures environmental parameters every 10 min (see figure 2) ; these data are accessible at the global ecological lakes observatory network (gleon) website. all data from the instrumented buoy and associated meteorological variables were downloaded from the gleon publicly accessible database. the yyl buoy measured surface dissolved oxygen, wind speed, wind direction, water temperature profiles, and air temperature. a weather station approximately 1 km from the lake also measured rainfall, humidity, and soil temperature. high-resolution water temperature profiles collected from the buoy were used for the training and validation of the studied models. also meteorological variables given in table 1 were used to develop the soft computing techniques. figure 1. location of yuan-yang lake in taiwan figure 2. the bathymetry of the yyl (contours in m) table 1. meteorological variables used to develop the models table 2 represents the hourly statistical parameters of the applied variables. in this table, the terms xmean, xmin, xmax, sx, cv and csx denote the mean, minimum, maximum, standard deviation, coefficient of variation and skewness coefficient, respectively. as can be seen clearly, rainfall has the maximum skewness. solar radiation and relative humidity also indicate a skewed distribution. air pressure, soil temperature and air temperature show normal distributions because they have significantly low skewness values. table 3 shows the correlations between the meteorological and hourly water temperature parameters. as can be seen from the table, the soil temperature has the highest correlations with hourly water temperatures in all depths. air temperature also has higher correlations than the other variables. time variation graphs of the meteorological variables used as inputs to the gep, anfis and ann models are illustrated in figure 3. d4 saeed samadianfard, honeyeh kazemi, ozgur kisi and wen-cheng liu table 2. hourly statistical parameters of the observed data. note: the terms xmean, xmin, xmax, sx, cv and csx denote the mean, minimum, maximum, standard deviation, coefficient of variation and skewness, respectively. table 3. correlations between meteorological and hourly water.temperature parameters. 0 200 400 600 800 1000 1200 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date s ol ar ra di at io n (w m -2 ) 610 615 620 625 630 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date a ir pr es su re (h pa ) 70 80 90 100 110 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date r el at iv e hu m id ity (% ) 24 25 26 27 28 29 30 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date s oi l t em pe ra tu re (c ) 10 15 20 25 30 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date a ir te m pe ra tu re (c ) 0 1 2 3 4 5 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date w in ds pe ed (m .s -1 ) 0 10 20 30 40 50 5/1 5/6 5/11 5/16 5/21 5/26 5/31 6/5 6/10 6/15 6/20 6/25 6/30 date r ai nf al l ( m m ) figure 3. time variation graphs of the meteorological conditions. d5water temperature prediction in a subtropical subalpine lake using soft computing techniques 2.2 general overview of genetic programming in this section, a brief overview of the gp and gep is given. detailed explanations of gp and gep are provided by koza (1992) and ferreira (2006), respectively. gp was first proposed by koza (1992). it is a generalization of genetic algorithms (gas) (goldberg, 1989). the fundamental difference between ga, gp, and gep is due to the nature of the individuals. in the ga, the individuals are linear strings of fixed length (chromosomes). in the gp, the individuals are nonlinear entities of different sizes and shapes (parse trees), and in gep the individuals are encoded as linear strings of fixed length (the genome or chromosomes), which are afterwards expressed as nonlinear entities of different sizes and shapes (ferreira, 2001 a,b). gp is a search technique that allows the solution of problems by automatically generating algorithms and expressions. these expressions are coded or represented as a tree structure with its terminals (leaves) and nodes (functions). gp applies gas to a “population” of programs, typically encoded as tree-structures. trial programs are evaluated against a “fitness function” then the best solutions are selected for modification and re-evaluation. this modificationevaluation cycle is repeated until a “correct” program is produced. there are five major preliminary steps for solving a problem by using gep. these are the determination of (i) the set of terminals, (ii) the set of functions, (iii) the fitness measure, (iv) the values of the numerical parameters and qualitative variables for controlling the run, and (v) the criterion for designating a result and terminating a run (koza, 1992). a gep flowchart is presented in figure 4. figure 4. gep flowchart there are five major steps in preparing to use gep of which the first is to choose the fitness function. the fitness of an individual program i for fitness case j is evaluated by ferreira (2006) using: where p is the precision and e(i,j) is the error of an individual program i for fitness case j. for the absolute error, this is expressed by: (2) where p(i,j) is the value predicted by the individual program i for fitness case j (out of n fitness cases) and tj is the target value for fitness case j. again for the absolute error, the fitness fi of an individual program i is expressed by: (3) where r is the selection range. the second major step consists of choosing the set of terminals t and the set of functions f to create the chromosomes. for this study, the function set consists of 7 functions including four basic arithmetic operators, i.e., (+, -, ×, /) and some basic mathematical functions, i.e., (√, ln(x), exp) selected among all the available functions in gep. the function selection was based on simplicity and its relevance to the nature of the problem thus ensuring a simple and efficient final gep model. the third major step is to choose the chromosomal architecture, i.e., the length of the head and the number of genes. values of the length of the head, h = 10, and six genes per chromosome were employed based on the discussion in ferreira (2001b). the fourth major step is to choose the linking function. in this study, the subprograms were linked by addition on the basis of recommendations made by ferreira (2001a) and findings of other studies (e.g. guven and aytek, 2009). finally, the fifth major step is to choose the set of genetic operators that cause variation along with their rates. a combination of all genetic operators, i.e., mutation, transposition and recombination, was used for this purpose. the parameters of the training of the gep are given in table 4. table 4. parameters of the gep model 2.3. general overview of adaptive neuro-fuzzy inference system (anfis) anfis (jang, 1993), using a given input/output data set, constructs a fuzzy inference system (fis) whose membership function parameters are tuned (adjusted) using either a back propagation algorithm alone or in combination with a least squares type of method. this adjustment allows fuzzy systems to learn from the data set. figure 5 shows the anfis structure for a fuzzy inference system with two inputs. d6 saeed samadianfard, honeyeh kazemi, ozgur kisi and wen-cheng liu figure 5. anfis structure firstly, the fis type should be selected. in this regard, the sugeno method was selected for the present study (ozger, 2009; ullah and choudhury, 2013). to train a fis, a training data set that contains the desired input/output data of the system to be modeled must be loaded. before training, an initial fis model structure must be specified: (1) hybrid algorithm was selected as the optimization method. the optimization methods train the membership function parameters to emulate the training data. (2) the number of training epochs and the training error tolerance were selected. these values are entered to set the stopping criteria for training. the training process stops whenever the maximum epoch number is reached or the training error goal is achieved. to validate the trained fis, after loading the test data, the number of membership functions (mfs) and the type of input and output membership functions were selected. there are only two choices for the output membership function: constant and linear. this limitation of output membership function choices is because anfis only operates on sugeno-type systems (jang, 1993). the builtin membership function composed of difference between two sigmoidal membership functions (dsigmf) and the constant membership function were selected as the input and output membership functions, respectively. the number of mfs assigned to each input was three. it should be noted that there is no basic rule for selecting the optimization method and determining the optimal number and type of mfs and they are usually considered by trial and error. nevertheless, it should be taken into consideration that large numbers of mfs will increase the calculation time and efforts (keskin et al., 2004). the values assigned to each parameter for the anfis model are given in table 5. table 5. parameters for the anfis model 2.4. general overview of artificial neural networks (anns) the qnet neural network development system which is a complete solution for back propagation neural network modeling (qnet, 2000) was used for the present study. back propagation type neural networks process information in interconnecting processing elements termed nodes. these nodes are organized into groups termed layers. there are three distinct types of layers in a back propagation neural network: the input layer, the hidden layer(s) and the output layer. a network consists of one input layer, one or more hidden layers and one output layer. connections exist between the nodes of adjacent layers to relay the output signals from one layer to the next. information enters a network through the nodes of the input layer. the input layer nodes are unique in that their sole purpose is to distribute the input information to the next processing layer (i.e., the first hidden layer). figure 6 shows the structure of the anns with one hidden layer including three nodes. figure 6. the structure of the artificial neural networks to design the neural network, the modeler must specify the layer and node quantities, transfer functions and network connections. including the input layer, one hidden layer and the output layer, the specified number of the layers for the network was three. the number of nodes in the input layer will be equal to the number of input data values in the model. according to table 1, it was specified seven for the prediction of hourly water temperature for the present study. the number of output nodes for the network must correspond to the number of outputs in the network. it was one because we only had one output. choosing the number of hidden layers and the number of hidden nodes in each layer is not so trivial. the construction of the hidden processing structure of the network is arbitrary. generally, it is best to start with simple network designs that use relatively few hidden layers and processing nodes. however, in practice, it is usually better to employ multiple hidden layers for solving complex problems. a single hidden layer including three single nodes was specified herein to avoid an unnecessary large and complex model. the sigmoid function is qnet’s default transfer function and it is the most widely used function for back propagation neural networks. another network design consideration concerns how to control the network’s connections. while the connection editor gives the modeler almost unlimited flexibility in designing a network, the fact is that the vast majority of designs work best fully connected. qnet’s connection editor is best suited for highly advanced models that require groups of input data to be processed through separate network pathways. the default fully connected configuration was used for the present study. the values assigned to each parameter for the ann model are given in table 6. as it is explained above, qnet has a high and intelligent ability to simulate complex networks easily while preparing a code for these networks could be excruciating. for practical problems, using an easy method, which is usable for different cases, is more acceptable than sophisticated methods. in summary, qnet is professional user friendly software and it has been used for simulating different complex problems (kuo et al. 2004, yang et al. 2009) and that is why it was used in this research. table 6. parameters for the ann model the performance of three soft computing techniques, namely gep, anfis and ann to predict the hourly water temperature at yyl at various measured depths was compared. hourly water temperature data from may 1 to june 11, 2008 were taken as the training data set, while the measured d7water temperature prediction in a subtropical subalpine lake using soft computing techniques data from june 12 to 30, 2008 served as the validation data set. furthermore, time-series meteorological conditions served as inputs for the studied models for data from may 1 to june 30, 2008 (figure 3). 2.5. evaluation parameters several parameters can be considered for the evaluation of the model-predicted values of hourly water temperature in the yyl lake. in this study, root mean squared error (rmse), mean absolute error (mae) and correlation coefficient (r) were used as the evaluation criteria and they can be computed as follows (liu and chen, 2012): where tp(i) and to(i) represent the model-predicted and the observed water temperature, respectively, and n is the number of observations. 3. results and discussion a comprehensive comparison was made to compare the accuracy of gep, anfis and ann methods for the prediction of hourly water temperature in different depths in the yyl. rmse, mae and r values of each model for the prediction of hourly water temperature at different depths are shown in table 7. it is clear from the table that in the case of surface water temperature, the gep model has the lowest rmse (1.49 oc), mae (1.21 oc) and the highest r (0.73) values. the ann model seems to be the second best from the rmse and mae viewpoints. also, in the case of 1 meter depth, the ann model shows better accuracy than the gep and anfis models. anfis model is ranked as the second best. despite the error (rmse value) is smaller in ann than gep, there is an upper bound in scatter plot of ann (see figure 7.e) that imposes a constant value, which is not suitable for predictive purposes. furthermore, at the 2 meter depth below water surface, the gep model has the lowest rmse (0.35 oc) and mae (0.24 oc) values. also, the ann model has a better accuracy than the anfis. finally, at the 3 meter depth below water surface, the superiority of the gep model over the other models is clearly seen from table 7. the ann model seems to be the second best from the rmse, mae and r viewpoints. figure 7 shows the observed (x-axis) and predicted (y-axis) hourly water temperature values in different depths and in validation period in the form of scatter plots. gep model seems to be better than the other soft computing techniques. in the case of surface water temperature, the estimates of gep model seem to be closer to the exact line than those of the ann and anfis models. the anfis model’s estimates are also less scattered than the ann. in the case of 1 meter depth, although the ann model has the least scattered estimates, but there is an upper bound in scatter plot of ann that imposes a constant value, which will make additional problems in future predictions. so, gep model having less scattered estimations seems to be the best. table 7. performance assessment of different models for predicting hourly water temperature at different depths at the 2 meter depth below water surface, the estimates of the gep model closer to the exact line than those of the other models. the ann and anfis models underestimate all the high values (>13.5 oc). the ann model has a better accuracy than the anfis because anfis model significantly underestimates high values (>13 oc). it is clear from the figure, at the 3 meter depth below water surface, the estimates of the gep model are closer to the ideal line than those of the other models. the ann model’s estimates seem to be less scattered than the anfis model. it should be noted that actual physical processes of the lakes such as vertical mixing, which controls the vertical distribution of temperature, have undeniable roles in changing water temperatures at different depths. vertical mixing, in small, seasonally stratified lakes such as yyl, might result in partial or complete mixing of the water column, depending on the thermal conditions of the lake and the strength of meteorological forces driving the mixing process. due to the fact that the purpose of the current research is statistical investigation of the effects of metrological parameters in predicting hourly water temperatures of yyl, the roles of physical phenomena including vertical mixing and phase lag between the sub-daily variations of the variables, affecting the heat flux, have been ignored. so, ignoring the physical factors might be one of the sources, which increased the error parameters of the predictions. figure 8 shows the observed and predicted values as time series plot in the validation period. from the figure, it can be said that the gep model is generally more successful than the ann and anfis model especially for the high water temperature values. one of the advantages of gep in comparison to the other theories is its ability in producing analytical formula for determination of output parameters. table 8 summarizes the gep mathematical equations for the prediction of hourly water temperature in different depths. from the table, the ap seems to be not effective variable on the water temperature in the case of 2and 3-meter depth below water surface. it can be said that the ap has a more effect on wt0 than the wt1. decreasing ap effect by increasing depth is clearly seen from the table. d8 saeed samadianfard, honeyeh kazemi, ozgur kisi and wen-cheng liu 10 15 20 25 10 15 20 25 measured wt0 (°c) g e p p re di ct ed w t0 (° c ) r=0.733 mae=1.206 °c rmse=1.492 °c a 10 15 20 25 10 15 20 25 measured wt0 (°c) a n n p re di ct ed w t0 (° c ) r=0.636 mae=1.315 °c rmse=1.637 °c b 10 15 20 25 10 15 20 25 measured wt0 (°c) a n fi s p re di ct ed w t0 (° c ) r=0.630 mae=1.341 °c rmse=1.770 °c c 12 14 16 18 12 14 16 18 measured wt1 (°c) g e p p re di ct ed w t1 (° c ) r=0.644 mae=0.568 °c rmse=0.675 °c d 12 14 16 18 12 14 16 18 measured wt1 (°c) a n n p re di ct ed w t1 (° c ) r=0.796 mae=0.351 °c rmse=0.441 °c e 12 14 16 18 12 14 16 18 measured wt1 (°c) a n fi s p re di ct ed w t1 (° c ) r=0.414 mae=0.508 °c rmse=0.639 °c f 12 14 16 12 14 16 measured wt2 (°c) g e p p re di ct ed w t2 (° c ) r=0.821 mae=0.236 °c rmse=0.345 °c g 12 14 16 12 14 16 measured wt2 (°c) a n n p re di ct ed w t2 (° c ) r=0.842 mae=0.307 °c rmse=0.395 °c h 12 14 16 12 14 16 measured wt2 (°c) a n fi s p re di ct ed w t2 (° c ) r=0.586 mae=0.665 °c rmse=0.807 °c i 10 12 14 10 12 14 measured wt3 (°c) g e p p re di ct ed w t3 (° c ) r=0.782 mae=0.245 °c rmse=0.319 °c j 10 12 14 10 12 14 measured wt3 (°c) a n n p re di ct ed w t3 (° c ) r=0.746 mae=0.442 °c rmse=0.552 °c k 10 12 14 10 12 14 measured wt3 (°c) a n fi s p re di ct ed w t3 (° c ) r=0.521 mae=0.576 °c rmse=0.684 °c l figure 7. scatter plots of observed (x-axis) and predicted values (y-axis) of hourly water temperature in different depths. d9water temperature prediction in a subtropical subalpine lake using soft computing techniques 14 16 18 20 22 24 26 0 100 200 300 400 500 time (hr) w t 0 (° c ) observed gep ann anfisa 12 13 14 15 16 17 18 0 100 200 300 400 500 time (hr) w t 1 (° c ) observed gep ann anfisb 12 13 14 15 0 100 200 300 400 500 time (hr) w t 2 (° c ) observed gep ann anfisc 11 12 13 0 100 200 300 400 500 time (hr) w t 3 (° c ) observed gep ann anfisd figure 8. time series plots of observed and predicted values of hourly water temperature in different depths. table 8. mathematical expressions of gep model. d10 saeed samadianfard, honeyeh kazemi, ozgur kisi and wen-cheng liu as a conclusion, output results showed that gep provided reasonable and moderately accurate trends for the prediction of hourly water temperature in different depths. 4. conclusion in the present study, the performance of some soft computing techniques, namely gene expression programming, adaptive neuro fuzzy inference system and artificial neural network to predict hourly water temperatures at different layers of the yuan-yang lake in north-central taiwan has been compared. a time-series set of data for the hourly water temperature at different measured depths from may 1 to june 11, 2008 was taken as training dataset, while the data measured from june 12 to june 30, 2008 were served as a validation dataset for the studied models. the relative performances of these models were comprehensively evaluated using various statistical indices including rmse, mae and r coefficients. results showed that the gep had the best performances in predicting hourly water temperatures at the surface and both 2 and 3 meter depths below water surface, whereas, a different trend was seen for the 1 meter depth below water surface. in this depth, in spite of smaller error in ann than gep, there is an upper bound in scatter plot of ann that imposes a constant value, which is not appropriate for predictive purposes. conclusively, results obtained from this study showed that gep can provide reasonable trends for the prediction of hourly water temperature in different depths especially in shallow waters. acknowledgments sincere thanks are given to the two anonymous reviewers for their invaluable and constructive comments and suggestions for improving the paper quality. references aytac, a. and kisi, o., (2008). a genetic programming approach to suspended sediment modelling. journal of hydrology, 351, 288-298. banzhaf, w., nordin, p., keller, r.e. and francone, f.d., (1998). genetic programming. morgan kaufmann, san francisco, ca. benyahya, l., caissie, d., st-hilaire, a., ouarda, t.b.m.j. and bobee, b., (2007). a review of statistical water temperature models. canadian water resource journal, 32(3), 179-192. chen, s.h. and wu, j.t., (1999). paleolimnological environment indicated by the diatom and pollen assemblages in an alpine lake of taiwan. journal of paleolimnology, 22, 149-158. fallah-mehdipour, e, borzorg haddad, o. and marino, m.a., (2013). prediction and simulation of monthly groundwater levels by genetic programming. journal of hydro-environment research, 7(4), 253-260. fang, x. and stefan, h.g., (1996). long-term lake water temperature and ice cover simulations/measurements. cold regions science and technology, 24, 289-304. ferreira, c., (2001a). gene expression programming in problem solving. 6th online world conf. on soft computing in industrial applications (invited tutorial). ferreira, c., (2001b). gene expression programming, a new adaptive algorithm for solving problems. complex systems, 13 (2), 87. ferreira, c., (2006). gene expression programming: mathematical modeling by an artificial intelligence. springer, berlin 2006, p. 478. gleon, global lake ecological observatory network. http://gleon.org/. goldberg, d.e., (1989). genetic algorithms in search, optimization, and machine learning. addison-wesley, reading, ma. guven, a. and aytek, a., (2009). new approach for stageedischarge relationship: geneexpression programming. journal of hydrologic engineering, 14, 812-820. hondzo, m. and stefan, h.g., (1996). long-term lake water quality predictors. water research, 30(12), 2835-2852. jang, r., (1993). anfis: adaptive network-based fuzzy inference system. ieee transactions on systems, man, and cybernetics, 23, 665-685. keskin, m.e., terzi, o. and taylan, d., (2004). fuzzy logic model approaches to daily pan evaporation estimation in western turkey. hydrological sciences journal, 49 (6), 1001–1010. kisi, o., (2006). daily pan evaporation modeling using a neuro-fuzzy computing technique. journal of hydrology, 329, 636-646. kisi, o., shiri, j. and nikoofar, b., (2012). forecasting daily lake levels using artificial intelligence approaches. computers & geosciences, 41, 169-180. koza, j.r., (1992). genetic programming, on the programming of computers by means of natural selection. mit press, cambridge, ma, isbn 0-262-11170-5. kuo, y.m., liu, c.w. and lin, k.h., (2004). evaluation of the ability of an artificial neural network model to assess the variation of groundwater quality in an area of blackfoot disease in taiwan. water research, 38 (1), 148-158. lawrence, s.p., hogeboom, k. and le core, h.l., (2002). a threedimensional general circulation model of the surface layer of lake baikal. hydrobiologia, 487 (1), 95-110. lee, h.s., yamashita, t. and haggag, m., (2009). modelling hydrodynamic in yachiyo lake using a non-hydrostatic general circulation model with spatially and temporally varying meteorological conditions. hydrological processes, 23 (14), 1973-1987. liu, w.c. and chen, w.b., (2012). prediction of water temperature in a subtropical subalpine lake using an artificial neural network and threedimensional circulation models. computers & geosciences, 45, 13-25. liu, w.c., yu, h.l. and chung, c.e., (2011). assessment of water quality in a subtropical alpine lake using multivariate statistical techniques and geostatistical mapping: a case study. international journal of environmental research and public health, 8, 1126-1140. mellit, a., benghanen, m. and kalogirou, s.a., (2006). an adaptive wavelet network model for forecasting daily total solar radiation. applied energy, 83, 705-722. moghaddamnia, a., ghafari gousheh, m., piri, j., amin, s. and han, d., (2009). evaporation estimation using artificial neural networks and adaptive neuro-fuzzy inference system techniques. advances in water resources, 32, 88-97. ozger, m., (2009). comparison of fuzzy inference systems for streamflow prediction. hydrological sciences journal, 54(2), 261-273. piccolroaz, s., toffolon, m. and majone, b., (2013). a simple lumped model to convert air temperature into surface water temperature in lakes. hydrology and earth system sciences, 17, 3323-3338. qnet, (2000). qnet 2000 neural network modelling for windows 95/98/ nt, qnettoll user’s guide and datapro user’s guide. vesta services, incorporated, usa. samadianfard, s., (2012). gene expression programming analysis of implicit colebrook-white equation in turbulent flow friction factor calculation. journal of petroleum science and engineering, 92-93, 48-55. samadianfard, s., sadraddini, a.a., nazemi, a.h., provenzano, g. and kisi, o., (2012). estimating soil wetting patterns for drip irrigation using genetic programming. spanish journal of agricultural research, 10(4), 1155-1166. schwab, d.j., beletsky, d., depinto, j. and dolan, d.m., (2009). a hydrodynamic approach to modeling phosphorus distribution in lake erie. journal of great lakes research, 35 (1), 50-60. sharma, s., walker, s.c. and jackson, d.a., (2008). empirical modelling of lake water-temperature relationships: a comparison of approaches. freshwater biology, 53, 897-911. shiri, j. and kisi, o., (2011). comparison of genetic programming with neuro-fuzzy systems for predicting short-term water table depth fluctuations. computers & geosciences, 37 (10), 1692-1701. shiri, j., dierickx, w., pour-ali baba, a., nemati, s. and ghorbani, m.a., (2011). estimating daily pan evaporation from climatic data of the state of illinois, usa using adaptive neuro-fuzzy inference system and artificial neural network. hydrological research, 42 (6), 491-502. d11water temperature prediction in a subtropical subalpine lake using soft computing techniques thiery, w., stepanenko, v.m., fang, x., johnk, k.d., li, z., martynov, a., perroud m., subin, z.m., darchambeau, f., mironov, d. and lipzig, n.p.m.v, (2014). lakemip kivu: evaluating the representation of a large, deep tropical lake by a set of onedimensional lake models. tellus a, 66, 21390, http://dx.doi. org/10.3402/tellusa.v66.21390 toffolon, m., piccolroaz, s., majone, b., soja, a.m., peeters, f., schmid, m. and wuest, a., (2014). prediction of surface temperature in lakes with different morphology using air temperature. limnology and oceanography, 59 (6), 2185-2202. trumpickas, j., shuter, b.j. and minns, c.k., (2009). forecasting impacts of climate change on great lakes surface water temperatures. journal of great lakes research, 35, 454-463. ullah, n. and choudhury, m., (2013). flood flow modeling in a river system using adaptive neuro-fuzzy inference system. environmental management and sustainable development, 2(2), 54-68. webb, b.w., hannah, d.m., moore, r.d., brown, l.e. and nobilis, f., (2008). recent advances in stream and river temperature research. hydrological processes, 22, 902-918. wu, j.t., chang, s.c., wang, y.s., wang, y.f. and hsu, m.k., (2001). characteristics of the acidic environment of the yuan yang lake (taiwan). botanical bulletin of academia sinica, 42, 17-22. yang, c.t., marsooli, r. and aalami, m.t., (2009). evaluation of total load sediment transport formulas using ann. international journal of sediment research, 24(3), 274-286. zaier, i., shu, c., ouarda, t.b.m.j, seidou, o. and chebana, f., (2010). estimation of ice thickness on lakes using artificial neural network ensembles. journal of hydrology, 383, 330-340. zhao, l., yu, z. and acharya, k., (2011). modeling water temperature of lake taihu. american geophysical union, fall meeting. research on the effect of climate variability/climate change on runoff is limited in humid tropical regions. climate change has implications beyond the water resources sector, such as effects on agriculture and fisheries. hence, such studies are becoming increasingly important. this study uses both historical field data and future climate forecasts from general circulation model hadley centre coupled model, version 3 gcm hadcm3. these data are further downscaled using third generation of the hadley centre’s regional climate model (hadrm3), with providing regional climates for impacts studies (precis) software under the three quantifying uncertainties in model projections (qumps) with a horizontal resolution of 25 km x 25 km by the centre for climate change and adaptation research, anna university chennai, for the state of tamil nadu. these downscaled data are used to study runoff changes due to climate change for the kosasthaliyar sub-basin in south india. a trend analysis of the hydro-meteorological data for the sub-basin indicates that future rainfall is expected to decrease by approximately 10%, while the mean temperatures will increase by the year 2100. the runoff changes from 2011 to 2040 are not perceptible when compared to the historical period of 1971 to 2000. this study attempts to provide information on climate variability and its impacts on runoff in the kosasthaliyar sub-basin. la investigación del efecto variabilidad climática/cambio climático en el modelo pluviosidad/escorrentía es limitada en regiones tropicales húmedas, donde el cambio climático tiene implicaciones tanto en los recursos acuíferos, como en la agricultura y la pesca. por lo tanto este tipo de estudios han incrementado su importancia. este estudio utiliza tanto los datos adquiridos en este campo como las predicciones climáticas del modelo general de circulación de la célula de hadley, en su versión 3gcm hadcm3. estos datos fueron reducidos luego al utilizar la tercera generación del modelo climático regional del centro hadley (hadrm3), con el programa de estipulación de climas regionales para estudios de impacto (precis, en inglés), bajo los tres modelos de cuantificación de la incertidumbre en proyecciones (qumps, en inglés). el centro para el cambio climático y la investigación de adaptación de la universidad anna, de chennai, en el estado tamil nadu, utiliza una escala horizontal de 25 kilómetros por 25 kilómetros. esta reducción de datos se utiliza para estudiar los cambios de escorrentía por el cambio climático en la subcuenca de kosasthaliyar, al sur de la india. un análisis de tendencia de los datos hidrometeorológicos en la subcuenca indica que en el futuro la pluviosidad caerá en un 10 %, mientras que la temperatura media se incrementará para el año 2100. los cambios de escorrentía para el período 2011-2040 no difieren de los del período 1971-2000. este estudio es uno de los primeros acercamientos para proveer información sobre la variabilidad climática y sus impactos en la escorrentía de la subcuenca de khosastaliyar. earth sciences research journal eart sci. res. j. vol. 18, no. 1 (june, 2014): 45 49 abstract resumen key words: gcm, hadcm3, hadrm3, qump, runoff palabras clave: gcm; hadcm3; hadcm3; qump; escorrentía. climate variability and its impacts on runoff in the kosasthaliyar sub-basin, india balambal usha1 and b v mudgal2 1. research scholar, centre for water resources, anna university chennai. email: u_seshadri@rediffmail.com 2. professor, centre for water resources, anna university chennai. introduction evidence is mounting that climate change is occurring (alavian et al (2009)). every region of the world is expected to experience climate change. the impact of climate change on water resources may be positive or negative, depending on the geographical region (rao et al (2012)). each river basin faces a unique set of climate-related water challenges, depending on the hydrological regime. the extent that each water system will be affected by climate change will depend on the area’s degree of vulnerability, including its internal capacity to adapt to such change. internal capacity refers to the ability or resilience of the system to cope with this externality. in general, warming increases the variability in precipitation, i.e., the number of rainy days may decrease, while the intensity of rainfall may increase (mukerji 2009). this trend is likely to create excessive runoff within a short period (mall et al, 2006). additionally, as the atmosphere warms, the evaporation from soil moisture increases (chattopadhyay and hulme 1997; mall et al 2006). runoff is decreasing in some regions and increasing in other regions. whether runoff increases or decreases in a region is basin-specific within the context of climate change. meteorological data are fundamental inputs for hydrological investigations. a critical issue is that “due to changing climate past data and hydrological experience may no longer be a guide for decision making”, i.e., the stationarity of climate, which is a key assumption in planning for water resources, is no meteorology record manuscript received: 19/09/2013 accepted for publication: 31/05/2014 46 balambal usha and b v mudgal longer valid [ebnflo environmental aqua resource inc.(2010)]. hence, all impact assessments of climate change will need to incorporate a wide range of climate scenarios to study the potential of future water resources potential. the study was conducted by considering the above factors. the main aim is to assess the impacts of climate variability on the reservoir inflows of the major reservoirs located within the sub-basin of kosasthaliyar. a number of authors have used weap (water evaluation and planning system) to study stream flow impacts under climate change. a notable study was conducted by harma (2010) on the okanagan basin in british columbia. in this study, land use and climate, which are changing simultaneously, were considered for modeling stream flows using weap. the results demonstrate that all future climate conditions will critically reduce streamflows relative to the societal and ecological demands in at least a few months of “normal” and “dry” years. in all cases, the combination of demand, reservoir operations and climate variability will result in less than optimal conditions for instream flow needs. in another study by yates et al (2009), the sacramento basin in california was chosen to study the effects of climate change on reservoir inflows. the sacramento basin was divided into more than 100 sub-basins. a 37-year monthly climate time series from 1962 to 1998 was applied to a distributed hydrological model in weap. the results showed that the weap model was able to simulate the inflows well for this period. the problem the ipcc (2007) ar4 states that india could suffer from water stress – an annual water availability of less than 1000 cubic meters per capita – by 2025, and groundwater availability could decrease as much as 37 percent by mid-century. chennai basin is situated in a humid tropical climate in south india. water availability in this region depends on rainfall, particularly its spatial and temporal distribution. the basin receives approximately 1300 mm of rainfall annually. approximately 90 percent of this annual rainfall is received from two monsoons spread over a period of six months from june to december. the remaining rainfall occurs during a six-month period encompassing the summer months. most of the reservoirs in the basin are sources of domestic water for the city (indicated in the chennai basin micro level report). hence, the inflow into these reservoirs is very important because it dictates the supply of water for the basin. groundwater, which is a major water source, has been subjected to large withdrawals. therefore, surface water resources must be judiciously conserved and managed. quantifying hydrological responses to climate change is extremely important for proper water resources management (ipcc 2007). global climate models (gcms) are the best tools to study climatic variations at the global scale. however, gcms do not capture local details, which are necessary for impact assessments at the regional level. hence, there is a need to link the output from gcms by further downscaling so that they can be used as input for hydrological modeling at basin level . so far, such studies have apparently not been reported at the regional level for chennai basin. study area and methodology 3.1 study area chennai basin is situated in south india. the kosasthaliyar subbasin is the part of chennai basin located between 12°40’n-13°40’n and 79°10’e-80°25’e at the northeast corner of tamil nadu. the total area of the basin is 7282 km², 5542 km² of which is in tamil nadu; the remaining area is in andhra pradesh. the kosasthaliyar sub-basin covers an area of 2013.58 km². the poondi reservoir is a single purpose reservoir that was constructed in 1944 as a source of drinking water for the city of chennai. cholavaram and red hills are the other two reservoirs of the kosasthaliyar sub-basin. the map of the study area is given in figure 1. figure 1 study area (source: water resources department, tamil nadu) 3.2 trend analysis the non-parametric mann-kendall (mk) test (mann, (1945); kendall, (1975)) has been commonly used to assess the significance of trends in hydrometeorological series (chiew and mcmohan 1993 and others). first, the presence of a monotonic increase or decrease has been tested using the slope of the linear trend estimated from regression analysis. second, the significance of the trend is estimated using the mann –kendall method. 3.3 weap model the water evaluation and planning system (weap) is a physically based, semi-distributed river basin model developed by the stockholm environment institute. weap was developed to assess water resource allocation problems, and it is also used to study the impacts of climate change on hydrology. it is a complex integrated river basin model that operates with a flexible time step, i.e., hourly, daily, etc. the weap rainfall runoff model requires data on soil, land use, climate, etc., for the assessment of water resource availability at desired locations of a drainage basin. 3.4 research stages the methodology flowchart is given in figure 2 below. figure 2 methodology flowchart trend and variability analysis (1971-2000) projection (2011-2098) climate change impact assessment on water resources of chennai basin hydrometeorolical parameters of rainfall and temperature runoff changes due to changes in hydrometeorological parameters inferences based on results from above 47climate variability and its impacts on runoff in the kosasthaliyar sub-basin, india to conduct this research, rainfall and temperature field data of the past 30 years were analyzed; subsequently, the virgin flows for the same 30-year historical period (1971-2000) were used to develop a hydrological model. the rainfall and temperature data are used to drive the model to study past runoff impacts. along with this analysis, rcm data of rainfall and temperature for the same period are used to simulate the runoff. then, the results from the two simulations are compared. furthermore, rcm data are used for predicting future flows for the period of 2011-2040. for past impact analysis, field data of monthly rainfall and virgin flows were obtained from the institute for water studies, taramani, chennai. data for temperature were obtained from the imd (india meteorological department). to study the future changes in the basin, 120 years of daily climatic data were obtained by dynamically downscaling gcm hadcm3 output using precis software. the a1b emission scenario under the three qump (quantifying uncertainties in model projections) was considered. the spatial resolution of hadrm3 (regional climate model) was 25 km x 25 km. a land use map from satellite images was obtained from the institute of remote sensing (irs) , anna university, chennai. the land use map of the catchment of poondi reservoir is given in figure 3. figure 3 land use map of the catchment of poondi reservoir irs 1997. data for the weap input parameters, such as soil water capacity, hydraulic conductivity, crop coefficients, net irrigated area, deep water capacity and deep conductivity, were obtained from the institute for water studies, chennai. the leaf area index (lai) was calculated by using the fao method. the leaf area index is obtained from crop coefficients of both the mid-season and harvesting season. the initial value of the crop coefficient is 0.1. to assess the effects of climate change on future flows, weap was used to first calibrate and validate the model for the kosasthaliyar reservoirs with virgin flow data. a monthly time step was selected, as climate change is a longterm phenomenon. hence, monthly flows aggregated to annual flows were considered more representative of stream flows than shorter time steps. upon consultation with a few field experts, it was found that the catchment of the red hills is a highly urbanized sub-watershed. the virgin flows into both the red hills reservoir and the cholavaram tank are negligible. the runoff from kosasthaliyar is stored in poondi reservoir. therefore, only the virgin flows of poondi were considered, and the watershed was considered a single reservoir system. the stream flows were modeled accordingly. results and discussion 4.1 trend analysis – past (1971-2000) the main aim of this paper is to study the climate variability impacts on runoff. therefore, a detailed description of the work conducted on trend analysis is not presented. instead, a snapshot of the results of the trend analysis is shown in figure 4. an initial analysis indicates increases for both rainfall and temperature during 1971-2000 (control climate). the mk test at a significance level of 0.85 showed that these trends are statistically significant. further rainfall and runoff trends for the sub-basin of kosasthaliyar showed an increase. 4.1 trend analysis – future (2011-2100) the graphs show that rainfall will decrease in the coming years, whereas temperatures will increase. the maximum mean temperature (during may) is likely to increase from 34.5 degrees celsius in 1970-2000 to 37.5 degrees celsius by 2098. 4a 4b figure 4(a-b) chennai basin climatology – past and future monthly rainfall and temperature trends 4.3 weap model the upstream catchment of the poondi reservoir was considered for the analysis. the same period of analysis was taken into consideration, where1971-2000 is the control period and 2011 – 2040 is the future climate. the runoff in relation to the rainfall for the kosasthaliyar basin was plotted. the runoff increases in accordance with the rainfall for this sub-basin. to test the statistical significance of this trend, the mann-kendall test was performed 48 balambal usha and b v mudgal using xlstat. the trend is statistically significant at the significance level of alpha equal to 0.3 with a p-value of 0.288. to assess the effects of climate change on future flows, weap was used to first calibrate and validate the model for the poondi reservoir with virgin flow data. for rainfall data that used field data for the first simulation run, the mean of five relevant rain gauge stations was considered for the analysis. for the second simulation run using rcm data, a monthly bias correction factor was applied to the rcm rainfall data for the average rainfall of the nearby grids because there were seasonal biases compared to the field data. literature on rcm studies indicates that rcm data should never be used in hydrological studies without bias correction (maraun et al 2010). the details of the calibration and validation periods are given in table 1. table 1 model testing period for the calibration and validation reservoir calibration validation poondi 1971-1998 1998-2000 the model performance was assessed using nse and other indices, as shown in table 2.the nash sutcliffe efficiency (nse) produced very high values for both the calibration and validation of the field data. for the rcmderived runoff, the nse values were within an acceptable range. per the performance rating recommended by moriasi et al (2007) for the monthly time step, this model is very good for chennai basin. table 2 indices for testing the model efficiency nse calibration validation field data 0.92 0.85 rcm bias corrected 0.26 0.4 4.4 future impacts on runoff of the kosasthaliyar sub-basin the weap model was used to predict inflows under climate change. to build a future scenario, a single land use change was considered for each decade i.e., 2011-2020, 2021-2030 and 2031-2040. for poondi, the change in land use was an increase in the wasteland, which leads to a reduction in water bodies, as shown in figure 5. it might be noted that there were no major perceptible changes in the land use extrapolated for the future period. this process, coupled with climate change under the a1b emission scenario, was used to model future inflows. these results are graphically represented in figure 6. 5a figure 5 (a-c) future land use changes/land cover for the poondi reservoir 6a 6b 6c figure 6(a-c) climate change effects on runoff – baseline and future periods the graphs show that overall, there will be no significant change in the runoff in the coming years. 49climate variability and its impacts on runoff in the kosasthaliyar sub-basin, india conclusions in this study, runoff changes due to climate change were investigated for the kosasthaliyar sub-basin, south india. both past and future impacts on runoff were mapped using field data and climate model data, i.e., data from rcm precis using a horizontal resolution of 25 km x 25 km. the climatic data obtained were superimposed on a hydrological model using weap software. from a trend analysis of the climatic data and runoff data, the following are observed: approximately a 3°c rise in temperature is expected by the end of 2100. this is consistent with other studies on chennai basin using the same a1b scenario (nic report). additionally, the rainfall projections for the same scenario in the regionindicate a 10 percent reduction for the basin. this result is also consistent with the cited literature. runoff is mainly influenced by rainfall in the area. the weap model is able to simulate the hydrology of chennai basin under changing climatic conditions. this study attempts to estimate runoff changes due to climate change in the study area using the very high spatial resolution rcm hadrm3, the a1b scenario, and qump projections. notably, a larger set of climate simulations is necessary to improve the robustness of the conclusion that there is no significant change in the future runoff when compared to the baseline period for this sub-basin. this conclusion is based on the rcm data, which was derived from one ensemble run. this limitation was mainly due to the run-time of the simulations (approximately six months). using simulations from multiple runs could provide a better estimate of the range of variability in the future stream flows. references alavian, v., qaddumi, h. m., dickson, e., diez, s. m., danilenko, a. v., hirji, r. f., and blankespoor, b. (2009). water and climate change: understanding the risks and making climate-smart investment decisions. world bank, washington dc http://water. worldbank. org/ publications/water-and-climate-change-understanding-risks-andmaking-climate-smart-investment-decisions. chattopadhyay, n., and hulme, m. (1997). evaporation and potential evapotranspiration in india under conditions of recent and future climate change. agricultural and forest meteorology, 87(1), 55-73. chennai basin micro level report (2006), institute of water studies, taramani, chennai. chiew, f. h. s., and mcmahon, t. a. (1993). detection of trend or change in annual flow of australian rivers. international journal of climatology, 13(6), 643-653. ebnflo environmental aqua resource inc . (2010). guide for assessment of hydrologic effects of climate change in ontario,www.env.uwaterloo.ca harma, k. j. (2010). changing with the flow: an analysis of water supply and demand in a subwatershed of the okanagan basin, british columbia. ipcc (intergovernmental panel on climate change), (2007). the physical science basis. contribution of wg 1 to the fourth assessment report of the intergovernmental panel on climate change, cambridge university press, cambridge. available at: http://ipcc-wg1/wg1-report.html. kendall m. g. (1975). rank correlation methods, 4 th ed., charles griffin, london mall, r. k., gupta, a., singh, r., singh, r. s., & rathore, l. s. (2006). water resources and climate change: an indian perspective. current science, 90(12), 1610-1626. mann, h. b.(1945). non-parametric test against trend, econometrica, 13, 245-259. maraun, d., wetterhall, f., ireson, a. m., chandler, r. e., kendon, e. j., widmann, m and thiele-eich, i. (2010). precipitation downscaling under climate change: recent developments to bridge the gap between dynamical models and the end user. reviews of geophysics, 48(3). moriasi, d. n., arnold, j. g., van liew, m. w., bingner, r. l., harmel, r. d., & veith, t. l. (2007). model evaluation guidelines for systematic quantification of accuracy in watershed simulations. trans. asabe, 50(3), 885-900. mukerji, r. (2009). vulnerability and adaptation experiences from rajasthan and andhra pradesh, water resources management sdc v & a programme, india, pp 19 nic report. observed climate and climate change projections. available from: http://www.environment.tn.nic.in/doc/pdf/chapter%204.pdf rao p.g.s.l.h.v., peter k.v., gopakumar c. s. (2012). impact of climate change on agriculture in india, proceedings of the international meet on impact of climate change on water resources, pp 41 yates, d., purkey, d., sieber, j., huber-lee, a., galbraith, h., west, j and rayej, m. (2009). climate driven water resources model of the sacramento basin, california. journal of water resources planning and management, 135(5), 303-313. earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014): 99 105 abstract resumen key words: acceleration data, shear wave velocity, antakya, soil conditions, 1d velocity model. palabras clave: información de aceleración; ondas de corte; antakya; condiciones de suelo, modelos de velocidad 1d. record manuscript received: 30/01/2014 accepted for publication: 23/09/2014 seismology this manuscript presents a site response analysis and an estimation of s-wave velocity that are dependent on acceleration data. first, existing data, such as density, seismic wave velocity, and soil cross-sections, are obtained from previous seismic microzonation studies and used to prepare input data for a suite of matlab routines, which are referred to as sua software. acceleration data are obtained from four free-field strong-motion stations of the seramar project, which was conducted between 2006 and 2009 in conjunction with a turkish-german joint research project, and inputted into the software as basic data. the results include a 1d velocity cross-section versus depth and an amplification model of the site. three different depth levels can be determined for the ranges of 0-5 m, 5-15 m and 15-25 m. the seismic velocities vary between 380 and 470 m s-1 for the first 5 m; 320 and 480 m s-1 for 5-15 m; and 470 and 750 m s-1 for 15-25 m. these results are comparable with the amplification values from the microtremor data from previous studies. the 1d velocity models are appropriate for the soil conditions. este trabajo presenta el análisis a una respuesta de sitio y una estimación de la velocidad de la onda de corte que son dependientes de la información de aceleración. los datos adicionales como la densidad, la velocidad de onda sísmica y los cortes transversales de suelo, se obtuvieron de estudios previos de microzonificación sísmica y se utilizaron para preparar el registro de datos en una plataforma de rutinas matlab, que se refieren al software sua. los datos de información de la aceleración se tomaron de cuatro estaciones de monitoreo de movimientos fuertes a campo abierto del proyecto seramar, que se realizó entre 2006 y 2009 en una investigación conjunta turco-alemana, y se ingresaron en el programa como la información básica. los resultados incluyen una sección cruzada de velocidad 1d versus profundidad y el modelo amplificado del sitio. se pudieron determinar tres niveles diferentes a partir de los rangos de 0-5 m, 5-15 m y 15-25 m. las velocidades sísmicas pueden variar entre 380 y 470 m s-1 para los primeros 5 metros; 320 y 480 m s-1 para el rango 5-15 m, y 450 y 750 m s-1 para el rango 15-25 m. estos resultados son comparables con los valores de amplificación del perfil microtemor de estudios previos. los modelos de velocidad 1d son apropiados para las condiciones del suelo. estimating shear wave velocity using acceleration data in antakya (turkey) aydın büyüksaraç1, semir över2, m. cemal geneş3, murat bikçe4, selçuk kaçın4, özcan bektaş5 1. bitlis eren university, civil engineering dept., tr-13000 bitlis-turkey 2. mustafa kemal university, geophysical engineering dept. tr-31200 iskederun/antakyaturkey 3. zirve university, civil engineering dept. tr-27260 gaziantepturkey 4. mustafa kemal university, civil engineering dept. tr-31200 iskederun/antakyaturkey 5. cumhuriyet university, geophysical engineering dept. tr-58140 sivasturkey correspondence author: aydın büyüksaraç. e-mail: absarac@comu.edu.tr issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.41810 http://dx.doi.org/10.15446/esrj.v18n2.41810 100 aydın büyüksaraç, semir över, m. cemal geneş, murat bikçe, selçuk kaçın, özcan bektaş introduction site characteristics of urban areas are primarily estimated by analyses of recorded ground motions. the estimation of site characterization is very important for disaster mitigation planning. however, the types of recorded ground motions differ. because largeor moderate-sized earthquakes do not occur on a frequent basis, they cannot be considered as a source of ground motion. in this case, the site characteristics may be estimated from microtremors or small-sized frequent earthquakes (shrikhande and basu, 2004), which are divided into zones according to their similarities after estimation of the site characteristics. this process is referred to as microzonation. microzonation establishes several parameters that are associated with disasters such as earthquakes and landslides. however, microzonation provides a detailed description of earthquake hazards for the selected locality in a seismic microzone. this process is known as seismic microzonation. data from seismic microzonation includes the intensity of shaking at the bedrock level, the site characteristics, the predominant period and the spectral amplification of surface motion with respect to bedrock motion (shrikhande and basu, 2004). the distribution of damage caused by earthquake ground shaking commonly reflects regional areal differences in local soil conditions. earthquake damage is significantly greater on thick deposits of unconsolidated sediments compared with nearby bedrock locations. as a result, the site conditions should be considered as part of the assessment of ground motion hazards. studies indicate that shear-wave velocity is a critical factor in determining the intensity of ground shaking and is a useful parameter for characterizing local soil conditions in ground motion studies (over et al., 2011). the behavior of soil layers at large strains associated with strong earthquakes is normally nonlinear; therefore, it should be considered in seismic microzonation studies. the majority of researchers currently prefer to use microtremor measurements to estimate seismic microzonation maps. however, to estimate site characteristics from microtremors or weak motion data, all techniques are based on linear system theory (shrikhande and basu, 2004). microtremors have been used to estimate the resonant period of a site since 1957; the results were used to specify the earthquake design forces in various zones in japan depending on the predominant period (kanai, 1957). the characteristics of microtremors were compared with strong earthquake accelerograms recorded in el centro, california by udwadia and trifunac (1973). they concluded that the characteristics of microtremors significantly differ from the characteristics of strong earthquake accelerograms. many researchers have investigated ground motion behavior to establish a correlation between average shear wave velocity and the site amplification factors. surface wave recordings are very useful for seismic response analyses of the sites. this approach is very important to understand the behavior of soils because the soil conditions are related to the mechanical characteristics of surface soils, such as the slope of the soils, and the depth of the ground water table. the estimation of these conditions, which are related to the soils, increases the cost of the site investigations. the soil behavior during earthquakes or microtremors reflects its conditions. seismic hazard analysis provides a framework for which uncertainty in the size, location and likelihood of future earthquakes can be incorporated to model earthquake hazard. in this paper, a simple approach was applied to estimate the level of uncertainty in the modelled amplification factors due to variations in alluvial thickness and velocity structure. matlab routines, which are referred to as sua, are provided to implement an equivalent linear site-response analysis with the option of an assessment of uncertainty (robinson et al., 2006). the presence of soils, geological sediments and weathered rocks can amplify the level of ground shaking experienced during an earthquake. consequently, the effect of the type of soil on earthquake ground shaking is an important component for a seismic hazard analysis. this paper provides a detailed and comprehensive description of an equivalent linear site-response analysis, which is a technique for modelling the amplification of seismic waves due to propagation through alluvial soil in antakya. the description includes a theoretical solution for the wave equation, derivation of a transfer function that relates bedrock acceleration to surface acceleration, calculation of a response spectral acceleration and computation of an amplification factor. 1. overview of the geology, seismicity and soil conditions of antakya in the eastern mediterranean, plate motions occur between the arabia/ anatolia, africa/arabia and anatolia/africa boundaries along the left lateral amanos fault, the dead sea fault and the cyprus arc, and the extension of the latter on land [named the cyprus-antakya transform fault by over et al. (2004a)]. active faults that are associated with the previously mentioned structural zones join to form a triple junction at the quaternary amik basin near antakya (over et al., 2004a). the seismicity and filling of the quaternary age in antakya and its surrounding area have been controlled by these active major faults. the length of the nne-trending segment is approximately 145 km and is known as the amanos fault (lyberis et al., 1992) or karasu fault (westaway, 1994), which has been considered to be a southern continuation of the east anatolian fault zone (eafz) (over et al., 2002; over et al., 2004a). it is located along the western flanks of the karasu valley and the amik basin, which has a width of approximately 30 km, and is filled with plio-quaternary sediments to create a thickness greater than 1000 m (perincek and eren, 1990). some isotopic dating studies suggest a quaternary to recent age for the basaltic rocks along the karasu valley (rojay et al., 2001). plio-quaternary sediments and quaternary volcanics unconformably overlie miocene beds (fig. 1). several authors (ergin et al., 1967; ambraseys, 1970; mckenzie, 1972, 1978; soysal et al., 1981; rotstein and kafka, 1982) have described the general trends of seismicity in the eastern mediterranean, anatolia and levant countries. during this century, this region has been characterized by a low seismicity (jackson and mckenzie, 1988). however, it is known to have been seismically active based on historical catalogues (poirier and taher, 1980; ambraseys and barazangi, 1989). two large historical earthquakes (denoted by stars in fig. 2) destroyed the town of antakya on august 13, 1822 (ms=7.4) and april 13, 1872 (ms=7.2) (ergin et al., 1967), which caused more than 20,000 deaths in the region (kalafat and bagci, 2001). these historical events caused surface ruptures with lengths of 20 km and 30 km along directions of n30° and n20°e, respectively (ambraseys and jackson, 1998). figure 1: geological map of antakya. inset map shows the location and main tectonic units. nafz: north anatolian fault zone, eafz: east anatolian fault zone (over et al., 2011). the most important earthquake during the instrumental period occurred on january 22, 1997; it had a body-wave magnitude (mb) of 5.5 and aftershocks with mb of 5.2 and 5.3 and maximum intensity of io=vi-vii, which caused damage in certain buildings (erdik et al., 1997). 101estimating shear wave velocity using acceleration data in antakya (turkey) figure 2: seismicity and seismotectonic map of the hatay region and surroundings, and some focal mechanisms of shallow earthquakes. the red points represent the earthquakes in 1997. eafz: east anatolian fault zone, dsfz: dead sea fault zone, kofz: karataş osmaniye fault zone, rf: reyhanli fault, and af: amanos fault. data on the near-surface geological conditions are important for understanding the site amplifications that were experimentally measured using the earthquakes. two deep borehole logs provide a vertical profile of the surface deposits up to 60 m (w1) and 100 m (w2), respectively (table 1). antakya municipality housing is situated along the two sides of the asi river, which originates in syria and runs to the mediterranean sea through antakya city. the sediments within the antakya region are primarily composed of quaternary alluvial fill, which consists of clay, silt, sand and gravel. the soil description in the borehole logs show that the surface soil consists of quaternary materials composed of clay, gravel detritic formations of conglomerates and alluvial sands. the average ground water level in the area is 3 m according to the deep boreholes. table 1: deep boreholes in the study area (w1=60 m and w2=100 m) (over et al., 2008) w1 w2 depth (m) formation depth (m) formation 0 1 agricultural soil table 0 7 gravel with clay 1 7 gravel 7 11 clay stone 7 – 26 clay 11 18 clay stone with gravel 18 22 gravel 22 26 conglomerate 26 29 conglomerate with clay 26 36 gravel 29 33 conglomerate 33 53 conglomerate with clay 36 76 clay 53 60 clay stone 76 – 80 76 80 gravel 80 90 clay 90 100 gravel with clay 2. site effects from h/v ratios and vs30 analysis the fundamental frequencies of the near-surface deposits at a site are a valuable tool for developing land-use plans for urban regions that consider earthquake disaster mitigation. here, the resonant periods and site amplifications are determined by dividing the smoothed fourier amplitude spectrum of the horizontal component by the smoothed spectrum of the vertical component that was recorded at the same site, which eliminates the need for a reference site. the spectral ratio technique for estimating the transfer function for the site response, involves the ratio of the fourier amplitude spectrum of the motion recorded at the site to the fourier amplitude spectrum of the same motion recorded at a nearby reference site, has been proposed to estimate the resonant periods of the site (edac, 2004). over et al. (2011) analyzed the microtremor data for antakya. the h/v spectra were obtained for all observation sites, and the predominant periods of all sites were identified. the microtremor measurements show that the predominant period for the sites in antakya range from approximately 0.2 s to 0.8 s (fig. 3). figure 3: locations of sections from combination of microtremor and remi measurement positions. p represents microtremor observations and ë represents remi observations. 3. data 3.1. strong-motion stations in close collaboration with mustafa kemal university (mku) and other local partners, the earthquake damage analysis center (edac) at bauhaus-universität weimar initiated a turkish-german joint research project on seismic risk assessment and mitigation in the antakya-maras-region (seramar) (edac 2004). in the framework of the seramar project, three reinforced concrete structures were equipped with a building monitoring system manufactured by syscom instruments of switzerland. in this building monitoring system, three or four sensors were installed on the structure and one sensor was installed on the free field (schwarz et al., 2007). the sensors were mr2002+ triaxial accelerometers with a linear 0 to 150 hz frequency response range and a ±2 g measuring range. the free-field station was installed at a distance that was equivalent to the minimum height of one instrumented or neighboring building (schwarz et al., 2007). data obtained from the fourth station, which were instrumented by the general directory of disaster affairs of turkey, were employed in this study (fig. 4). the instruments consisted of kinemetrics etna with three channels and an internal triaxial episensor force balance accelerometer. 102 aydın büyüksaraç, semir över, m. cemal geneş, murat bikçe, selçuk kaçın, özcan bektaş 3.2. acceleration data in the considered time-window between 2006 and 2009, approximately 30 earthquakes were recorded and identified at these stations (abrahamczyk et al., 2008; schwarz et al., 2009). all earthquake parameters were derived from the website of the kandilli observatory and the earthquake research institute (koeri) bulletins, as shown in table 2. figure 4: locations of accelerometer stations: stations i, ii and iii are free-field stations of the building monitoring systems, whereas station iv is an independent free-field station. table 2: list of the measured earthquakes in the project area. no. date location h (km) md ml recorder lat. long. sta-ii sta-i sta-iii sta-iv 1 19/09/2006 35.95 35.77 22 3.8 ● ● ● 2 09/10/2006 35.82 35.60 39 4.3 ● ● ● ● 3 07/01/2007 37.13 36.09 20 3.9 ● 4 25/01/2007 36.12 35.90 17 3.0 ● ● 5 01/02/2007 35.84 35.90 12 3.6 ● ● ● 6 03/02/2007 36.95 35.42 27 3.6 ● ● 7 11/03/2007 36.56 35.88 6 3.8 ● 8 14/03/2007 36.47 35.86 4 3.2 ● 9 12/05/2007 36.39 35.63 8 3.8 ● ● 10 15/05/2007 35.97 35.83 5 3.1 ● 11 21.01.2008 37.00 36.12 7.0 2.9 ● ● 12 07.03.2008 36.78 36.42 8.7 3.2 ● ● 13 10/03/2008 36.61 36.47 6.8 2.8 ● ● 14 17/04/2008 36.28 36.13 7.3 3.3 ● ● ● 15 03/05/2008 36.73 36.62 7.1 3.9 ● 16 12/06/2008 36.25 36.06 7.4 3.0 ● ● 17 19/06/2008 36.63 36.05 6.8 3.7 ● 18 01/11/2008 36.06 35.86 24.2 3.3 ● 20 18/11/2008 36.22 36.98 7.0 2.9 ● ● 21 30/11/2008 36.24 36.11 7.0 2.7 ● ● 22 04/12/2008 36.18 36.02 7.2 2.9 ● 23 10/12/2008 36.04 36.25 8.4 3.2 ● 24 19/12/2008 36.21 36.29 7.0 3.0 ● 25 19/12/2008 35.93 36.36 7.0 2.8 ● 26 17/01/2009 37.16 36.31 7.5 4.6 ● ● 27 22/01/2009 36.11 35.88 24.6 3.4 ● ● 28 22/01/2009 36.09 35.83 22.8 3.1 ● ● 29 05/02/2009 36.20 36.09 7.4 3.1 ● ● 30 09/02/2009 36.09 36.05 2.1 3.3 ● the magnitudes of the earthquakes (md) ranged from 2.7 4.3. a total of 30 acceleration data were obtained from the time histories of this area. several earthquakes with magnitudes (md) that ranged from 3.0 4.3 were measured. examples of recorded acceleration time histories are given in figure 5 (schwarz et al., 2007). the response spectrum is the most suitable tool for expressing the excitation response relation in earthquake engineering and seismic design. although it is an indirect measure of ground intensity, it directly expresses the maximum response, which is a major concern in structural design. if sets of response spectra for ground motions recorded at different locations during previous earthquakes are generated, a large variation would be observed in both the response spectral values and the shape of the spectra will vary among the sets. figure 5: recorded building response due to a magnitude ml = 3.1 earthquake (march 3, 2008); sensor: free-field of the building (a) recorded acceleration time histories. these variations are dependent on many factors, such as the energy release mechanism in the vicinity of the focus or the hypocenter and along the fault interfaces, the epicentral distance and the focal depth, the geology and its variations, the energy transmission paths, the richter magnitude and the local soil conditions at the recording station. thus, the response spectral values s (spv, spa and sd) (eq. 1) for earthquake ground motion should be considered to be a function expressed in the following form (clough and penzien, 1993; tehranizadeh and hamedi, 2002) s = s(sm, ed, fd, gc, m, sc, ξ ,t) (1) s=s(sc,ξ) (2) where the independent variables include sm: source mechanism, ed: epicentral distance, fd: focal depth, gc: geological conditions, m: richter magnitude, sc: soil conditions, x: damping ratio and t: period. the effects of sm and gc on both the spectral values and the shapes of the response spectra are not well understood; therefore, these effects cannot be quantified when defining response spectra for design purposes. the effects of ed, fd and m are usually considered when specifying the intensity levels of the design response spectra; however, they are frequently disregarded during the specification of the shape of these spectra due to a lack of knowledge regarding their influences. the effect of sc on both the intensities and the shapes of the response spectra are considered in the definition of the design response spectra (tehranizadeh and hamedi, 2002). in many cases, the ratio of the spectral acceleration to the peak ground acceleration (acceleration amplification) is plotted as a function of either frequency or period (mohraz and elghadamsi, 1989). studies have shown that the response spectra from accelerograms recorded on similar soil conditions reflect similarities in shape and amplification. for this reason, the response spectra from records with common characteristics are averaged and smoothed before they are used in design (mohraz and elghadamsi, 1989). 103estimating shear wave velocity using acceleration data in antakya (turkey) 4. antakya case study robinson et al. (2006) collectively described the presence of soils, geological sediments and weathered rock as regolith and developed matlab routines, which are referred to as sua, to model the amplification of seismic waves due to propagation through regolith from sydney, australia. antakya exhibits high risk due to earthquakes. the approach of robinson et al. (2006) was employed with available geotechnical information to investigate the potential for ground motion amplification in antakya. antakya is located within the amik basin, which comprises a sequence of plio-quaternary sediments with a minimum thickness of 1000 m from a variety of geological environments, including sand, gravel, clay and silt. five site classes were identified in antakya using microtremor data and seismic velocities obtained by over et al. (2011). the surface acceleration time history and the response spectral acceleration for the recorded earthquakes in the stations (table 2) were obtained using sua software (robinson et al., 2006). the methodology enabled a broad-scale site-response analysis despite limited geotechnical data. the analysis provided an indication of the level of uncertainty associated with the two geotechnical parameters layer thickness and seismic velocity, for which the amplification factors are the most sensitive. the sua software packages are designed for individual site-response analysis and do not define the site response for a larger region. accelerometers were installed in two different soil types to record multiple earthquake data in the framework of the seramer project. we prepared input data for sua software for each of the four locations using vs data on the densities of shallow wells drilled by private firms. one-dimensional velocity versus depth was calculated for all four locations by controlling the acceleration data for each station (figures 6a, b, c, d). the cross-sections were simulated from the site classes using the approach suggested by robinson et al. (2006) to produce amplification factors for acceleration time history combinations. figure 6: 1-d velocity models generated from acceleration data using input data site class 4 for station i and site class 2 for stations ii, iii and iv. (a) station i, (b) station ii, (c) station iii, and (d) station iv. gray colored drilling logs show the soil condition at the station. lf: landfill, css: clay with silt and sand, ssc: silt with sand and clay, sss: sand with silt and sand, sgcs: sand with gravel, clay and silt, gcs: gravel with clay and sand, sc: sand with clay, and cs: clay with sand. 104 aydın büyüksaraç, semir över, m. cemal geneş, murat bikçe, selçuk kaçın, özcan bektaş amplification factors are assumed to be log-normally distributed. note that the mean in log-space is equivalent to the median in normal space. the median amplification factors are shown within the 16th and 84th percentiles in figures 7a, b, c, d. figure 7: median amplification factor (solid line) and 16th and 84th percentiles (dashed lines) (a) station i, (b) station ii, (c) station iii, and (d) station iv. 5. discussion and conclusions this paper describes the site response and velocity changes using acceleration data with inversion for four acceleration stations in antakya. the study also uses the microzonation results of over et al. (2011). 1 the median amplification factor indicated the following maximum levels of amplification: 1.5 at 0.4 s periods for station i, 1.5 at 0.5 s periods for station ii, 2 at approximately 0.25 s periods for station iii, and 2 at 0.4 s periods for station iv, which approximately corresponds to the natural period calculated by the microtremors. 2 the methodology enabled a broad-scale site-response analysis to be conducted despite limited geotechnical data. o analysis provided an indication of the level of uncertainty associated with the two geotechnical parameters soil thickness and seismic velocity, for which the amplification factors are the most sensitive. the seismic velocities are comparable with the drilling results. 3 the soil in the study area presented different velocity structures. three different depth levels were determined: 0-5 m, 5-15 m and 15-25 m. the seismic velocities varied between 380 and 470 m s-1 for the first 5 m; 320 and 480 m s-1 for 5-15 m; and 470 and 750 m s-1 for 15-25 m. however, stations 1 and 3 exhibit a similar seismic velocity structure. station 2 exhibits different velocity characteristics due to improvements in seismic velocity values. station 4 presents a similar soil structure along the deep profile. 4 our analysis determined the correlations, calculated periods and amplifications from microtremor and acceleration data. the periods were approximately identical (fig. 8a). the shape and amplitude of the h/v spectral ratio of the microtremor for each site were similar to the shape and amplitude of the h/v spectral ratio for the seismic motion. the amplifications showed differences due to different sources of the data (fig. 8b). this difference can be explained by the effects of the local topography. the study indicated that the alluvial soils in antakya can significantly amplify the ground shaking associated with an earthquake. the incorporation of uncertainty revealed that a range of possible amplification factors can be expected. the method assumed that the velocity in each soil model unit can be represented by a linear gradient. figure 8: comparison of acceleration and microtremor data (a) period changes and (b) amplification changes. 105estimating shear wave velocity using acceleration data in antakya (turkey) acknowledgements the authors are grateful to dr.-ing. jochen schwarz and dipl.-ing. lars abrahamczyk from the earthquake damage analysis center (edac) of bauhaus university in weimar, germany for the permission to use the recorded data provided by the collaborate seramar project . the instruments donated by syscom inc., switzerland and the financial support from the scientific and technological research council of turkey (tubitak) under the project grant no. 106m420 are also appreciated. this paper has been reviewed by dr. hakan cinar of karadeniz technical university, turkey, which has significantly improved its quality. references abrahamczyk, l., schwarz, j., lang, d.h., leipold, m., golbs, ch., genes, m.c., bikce, m., kacin, s., and gülkan, p. (2008). building monitoring for seismic risk assessment (i): instrumentation of rc frame structures as a part of the seramar project. 14th world conference on earthquake engineering (wcee), beijing 2008, china. ambraseys, n.n. (1970). some characteristic features of the anatolian fault zone, tectonophysics. 143-165. ambraseys, n. n. and barazangi, m. (1989). the 1759 earthquake in the bekaa valley: implications for earthquake hazard assessment in the eastern mediterranean region, journal of geophysical research, 94 (b4), 4007-4013. ambraseys, n.n., jackson, j.a. (1998). faulting associated with historical and recent earthquakes in the eastern mediterranean region, geophysical journal international. 133, 390-406. clough, r. and penzien, j. (1993). dynamics of structures. new york: mcgraw hill edac (2004). turkish-german-swiss joint project on seismic risk assessment and mitigation in the antakya-maraş regon on the basis of microzonation, vulnerability and preparedness studies (seramar)project description. earthquake damage analysis center, bauhausuniversitat weimar, october 2004. erdik, m., aydınoglu, n., kalafat, d. and pınar, a. (1997). january 22, 1997 (mw=5.8) antakya, turkey. earthquake international association of seismology and physics of the earth’s interior (iaspei), thessaloniki, greece, august 1997, 17-28. ergin, k., guclu, u., uz, z. (1967). a catalogue of earthquakes of turkey and surrounding area (11 ad to 1964 ad). maden fakultesi, arz fizigi enstitusu yayinlari, istanbul. 24, 1-28. jackson, j. a. and mckenzie, d. (1988). the relationship between plate motions and seismic moment tensors, and the rates of active deformation in the mediterranean and middle east, geophys. j. r. astr. soc., 93, 45-73. kalafat, d. and bagci, g. (2001). adana ve doğu anadolu fay zonunun depremsellik özellikleri. tmmob, jeofizik mühendisleri toplantısıadana, 36-43. kanai, k. (1957). the requisite conditions for predominant vibration of ground. bulletin of earthquake research institute, university of tokyo, 35, 457. lyberis, n., yurur, t., chorowicz, j., kasapoglu, e. and gundogdu, n. (1992). the east anatolian fault: an oblique collisional belt. tectonophysics, 204, 1–15. mckenzie, d. (1978). some remarks on the development of sedimentary basins, earth and planet sci. lett., (40), 25-32. mckenzie, d. (1972). active tectonics of the mediterranean region, geophysical journal of the royal astronomical society, 30(2), 109–185. mohraz, b., elghadamsi, f.e. (1989). the seismic design handbook. new york: van nostrand reinhold. over, s., büyüksaraç, a., bektaş, o. and filazi, a. (2011). assessment of potential seismic hazard and site effect in antakya (hatay province), se turkey. environmental earth sciences, 62, 313-326. over, s., büyüksaraç, a., bektaş, o., arisoy, m. o. and filazi, a. (2008). hatay i̇li merkezinin (antakya) deprem duyarlık ve mikrobölgelendirme projesi, mustafa kemal university (internal project, in turkish). over, s., ozden, s. and yilmaz, h. (2004a). late cenozoic stress evolution along the karasu valey, se turkey. tectonophysics, 380, 43-68. over, s., unlugenc, u.c. and bellier, o. (2002). quaternary stress regime change in the hatay region (se turkey). geophysical journal international, 148, 1-14. perincek, d. and eren, a.g. (1990). doğrultu atımlı doğu anadolu ve olu deniz fay zonları etki alanında gelişen amik havzasının kökeni. türkiye 8. petrol kongresi bildiri kitabı, 180–192. poirier, j.p. and taher, m.a. (1980). historical seismicity in the near and middle east, north africa, and spain from arabic documents (viith-xviiith century), bulletin of the seismological society of america, 70(6), 2185-2201. robinson, d., dhu, t. and schneider, j. (2006). sua: a computer program to compute regolith site-response and estimate uncertainty for probabilistic seismic hazard analyses, computers & geosciences, 32, 109–123. rojay, b., heimann, a. and toprak, v. (2001). neotectonic and volcanic characteristics of the karasu fault zone (anatolia, turkey): the transition zone between the dead sea transform and the east anatolian fault zone. geodinamica acta, 14, 197–212. rotstein, y. and kafka, a. l. (1982). seismotectonics of the southern boundary of anatolia, eastern mediterranean region: subduction, collision, and arc jumping, journal of geophysical research, 87(b9), 7694–7706. schwarz, j., abrahamczyk, l., langhammer, t., leipold, m., genes, m.c., bikce, m. and kacin, s. (2009). building typology for risk assessment: case study antakya (hatay). earthquake and tsunami, istanbul, turkey. schwarz, j., lang, d.h., abrahamczyk, l., bikce, m., genes, m.c. and kacin, s. (2007). seismische instrumentierung mehrgeschossiger stahlbetonbauwerke-ein beitrag zum seramar project. der österreichischen gesellschaft für erdbebeningenieurwesen und baudynamik (d-a-ch tagung 2007), 28 september 2007, wien. shrikhande, m. and basu, s. (2004). strong motion v/s weak motion: implications for seismic microzonation. journal of earthquake engineering, 8, 159-173. soysal, h., sipahioglu, s., kolcak, d. and altinok, y. (1981). turkiye tarihsel deprem katalogu. tubitak project (no: tbag-341), ankara. tehranizadeh, m. and hamedi, f. (2002). engineering structures, 24, 933-943. udwadia, f.e. and trifunac, m.d. (1973). comparison of earthquake and microtremor ground motion in el centro, california. bulletin of the seismological society of america, 63, 1227-1253. westaway, r. (1994). present-day kinematics of the middle east and eastern mediterranean. journal of geophysical research, 99 (b6), 12071-12090. application of artificial neural network for predicting shaft and tip resistances of concrete piles ehsan momeni¹, ramli nazir¹, danial jahed armaghani², harnedi maizir³ ¹department of geotechnics and transportation, faculty of civil engineering, universiti teknologi malaysia. ²young researchers and elite club, qaemshahr branch, islamic azad university, qaemshahr, iran. ³jurusan teknik sipil, sekolah tinggi teknologi pekanbaru, indonesia. correspondent author: mehsan23@live.utm.my abstract resumen key words: axial bearing capacity, artificial neural network, high strain dynamic testing, pile shaft resistance, pile tip resistance. palabras clave: capacidad axial de soporte; red neuronal artificial; dinámicas de alto esfuerzo de pilotes; resistencia de punta. record manuscript received: 11/07/2013 accepted for publication: 26/10/2014 geotechnics axial bearing capacity (abc) of piles is usually determined by static load test (slt). however, conducting slt is costly and time-consuming. high strain dynamic pile testing (hsdpt) which is provided by pile driving analyzer (pda) is a more recent approach for predicting the abc of piles. in comparison to slt, pda test is quick and economical. implementing feed forward back-propagation artificial neural network (ann) for solving geotechnical problems has recently gained attention mainly due to its ability in finding complex nonlinear relationships among different parameters. in this study, an ann-based predictive model for estimating abc of piles and its distribution is proposed. for network construction purpose, 36 pda tests were performed on various concrete piles in different project sites. the pda results, pile geometrical characteristics as well as soil investigation data were used for training the ann models. findings indicate the feasibility of ann in predicting ultimate, shaft and tip bearing resistances of piles. the coefficients of determination, r², equal to 0.941, 0.936, and 0.951 for testing data reveal that the shaft, tip and ultimate bearing capacities of piles predicted by ann-based model are in close agreement with those of hsdpt. by using sensitivity analysis, it was found that the length and area of the piles are dominant factors in the proposed predictive model. la capacidad axial de soporte (abc, en inglés) de un pilote de construcción se determina usualmente a través de una prueba de carga estática (slt, inglés). sin embargo, estas pruebas son costosas y demandan tiempo. la evaluación de las dinámicas de alto esfuerzo de pilotes (hsdpt, inglés), que la provee el programa de análisis de excavación (pda, inglés), es una forma de aproximación más reciente para preveer la capacidad axial de soporte. en comparación con la prueba de cargas estática, la evaluación pda es rápida y económica. la implementación de redes neuronales arficiales (ann, en inglés) que permita resolver problemas geotécnicos ha ganado atención recientemente debido a su posibilidad de hallar relaciones no lineales entre los diferentes parámetros. en este estudio se propone un modelo predictivo ann para estimar la capacidad axial de soporte de pilotes y su distribución. para fines de una red de construcción se realizaron 36 pruebas pda en pilotes de diferentes proyectos. los resultados de los análisis de excavación, las características geométricas de los pilotes, al igual que los datos de investigación del suelo se utilizaron para probar los modelos ann. los resultados indican la viabilidad del modelo ann en predecir la resistencia de los pilotes. los coeficientes de correlación, r², que alcanzaron 0.941, 09.36 y 0.951 para la evaluación de los datos, revelan que la capacidad del pilotaje en el último rodamiento, en el cojinete del eje y en la punta que se predijeron con el modelo ann concuerda con las establecidas a través del hsdpt. a través del análisis de respuesta se determinó que la longitud y el área de los pilotes son factores dominantes en el modelo predictivo propuesto. earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 85 93 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n1.38712 86 ehsan momeni, ramli nazir, danial jahed armaghani, harnedi maizir 1. introduction pile foundations are used extensively to transfer structural loads deep enough into the ground. proper estimation of axial bearing capacity is of prime importance in designing geotechnical structures. the ultimate amount of the load, which can be carried by the pile shaft, determines the type of pile as piles are classified according to their load-transfer mechanism (friction piles and end-bearing piles). hence, in addition to determining ultimate bearing capacity of piles, obtaining the pile shaft capacity is also of advantage (nazir et al, 2013). there are numerous methods for assessing pile bearing capacity and its distribution. although many attempts have been made to develop analytical or empirical methods for pile bearing capacity estimation (e.g. meyerhof, 1976; vesic, 1977; coyle and castello, 1981), most of these methods rely on empiricism and they are site specific (randolph, 2003). the most direct way for determining the axial bearing capacity of piles is static load test (slt). the test is standardized by american standards test methods (astm d114307). however, conducting slt is time consuming, expensive and difficult (likins and rausche, 2004). high strain dynamic testing (hsdt) of piles is a current approach for predicting the abc of piles and its distribution. hsdt is based on one dimensional wave propagation theory and is performed by using a pile driving analyzer (pda). in essence, in hsdt, a pile is hit by a hammer while pda monitors and records the necessary data for implementing wave equation analysis. consequently, through an iterative procedure using capwap software, the pile bearing capacity and its distribution can be predicted. the hsdt procedure is standardized by astm (astm d4945-08). utilization of artificial neural network (ann) in civil engineering has recently drawn considerable attention. it is generally attributed to the ann power in finding complex relationship between different parameters when the contact nature between them is unknown (garret, 1994). although many researchers have attempted to show the superiority of ann in predicting the bearing capacity problems, most of them focused on the prediction of ultimate bearing capacity of piles rather than its separate shaft and tip resistances (goh, 1995; goh, 1996). the main objective of this study is to propose an ann-based predictive model of bearing capacity using real pda and site investigation data. the predictive model is built for predicting shaft, tip and ultimate resistances (qs, qp and qu) of piles. nevertheless, it is worth mentioning that this study uses the capwap predicted pile bearing capacity rather than the determined bearing capacity of piles through slt. there are several published works concerning utilization of artificial intelligence techniques for predicting abc of piles (chan et al. 1995; chow et al. 1995; abu-kiefa, 1998; shahin, 2001; lok and che, 2004; das and basudhar, 2006; ardalan et al. 2009; shahin, 2008; shahin, 2010; adarsh et al. 2012; alkroosh and nikraz, 2012). among researchers who have addressed anns for predicting bearing capacity of pile foundations, goh (1995; 1996) developed an ann-based predictive model to estimate the ultimate load capacity of driven piles in sandy soils. his findings suggest that compared to conventional methods of pile bearing capacity estimation, ann-based predictive model works better. in another study, lee and lee (1995) employed ann for estimation of pile bearing capacity. their study focused on small scale laboratory tests where the horizontal and vertical chamber pressure, the number of hammer blows, pile penetration depth ratio, and mean normal stress of the soil were set as inputs of the network model while the ultimate bearing capacity was selected as the model output. according to their conclusion, ann can provide good prediction performance in bearing capacity problems. teh et al. (1997) also addressed the workability of neural network for predicting the pile bearing capacity. abu-kiefa (1998) implemented ann to predict the abc of driven piles in cohesionless soils. for network construction purpose, he compiled the data of 59 recorded cases of good-quality pile load tests. in his study, friction angles of the soil, the effective overburden pressure around the tip of the pile, the length of the pile and its equivalent cross-sectional area were considered as input layers of the ann model. his conclusion showed the feasibility of ann for predicting shaft and tip resistances of piles. however, among more recent studies, pal and deswal (2008) studied the ann application in predicting the total capacity of concrete spun pipe piles. they used stress-wave data for building their annbased predictive model. based on their conclusion, in comparison to support vector machines, the prediction performance provided by ann was more reliable. shahin and juksa (2009) proposed an ann-based predictive model of bearing capacity for drilled shafts. their model dataset comprised cone penetration test results and drilled shaft load tests on 94 recorded cases. jianbin et al. (2010) developed an ann model for predicting the ultimate abc of pipe piles. the influential parameters that they have considered for network construction included the effective length and diameter of pile, unit weight, cohesion and internal friction angle of soil as well as the standard penetration test (spt) results. benali and nechnech (2011) suggested an ann-based predictive model of pile bearing capacity in cohesionless soils. for training the ann, they had collected the mechanical properties of purely coherent soil and geometrical characteristics of 80 axially loaded piles. the correlation coefficient, r, equals to 0.92 reveals the reliability of their ann-based predictive model. 2. methods 2.1 high strain dynamic testing of piles it was mentioned earlier that high strain dynamic testing (hsdt) is an innovative method for estimating the abc of piles. in dynamic testing of pile (pda test), it is hypothesized that a pile with uniform cross section is a slender element surrounded by material with much lower stiffness. hence, any mechanical impact through a hammer leads to downward propagation of wave (timoshenko and goodier, 1951; salgado, 2008). therefore, the principle of one dimension wave propagation theory can be implemented for piles. details of the solution to the partial differential equations of wave propagation theory for predicting axial bearing capacity of piles can be found elsewhere (salgado, 2008). nevertheless, the finite difference-based model introduced by smith (1960) was considered as the bench mark for dynamic testing of piles. to estimate the axial bearing capacity, smith developed a discrete solution for wave propagation in piles. in smith’s model, the pile was simulated using a number of masses attached to each other by elastic springs, while the soil was modelled by a number of springs, and linear viscous dampers to represent its behavior. however, there were some deficiencies in smith`s model due to lack of knowledge on hammer energy and cushion characteristics. smith`s model later on was enhanced by a group of researchers at case western reserve university (goble et al.1970; rausche et al. 1972; rausche et al. 1985). in their developed method also known as case method, it was not necessary to model the hammer and driving systems. instead, they reported the use of force and acceleration records in a simplified model for predicting the static bearing capacity. as stated by fellenius, (1999) the full power of the wave equation analysis was first realized when it was coupled with dynamic monitoring of piles. the latter can be determined by installing a pair of accelerometer and strain transducer on top of the pile (see figure 1). subsequently, the data recorded by aforementioned instruments are transmitted by means of a cable to pda. the pda will then transform the recorded data into force and velocity. in the next step, the bearing capacity of the pile is predicted using capwap program. capwap combines the measured force and velocity with the wave equation analysis to determine soil resistance and its distribution along the pile. the capwap approach is based on an iterative curve-fitting technique in which pile response, estimated through wave equation analysis of a model pile, is matched to the measured response of the actual pile for a single hammer blow (fhwa, 2006). 87application of artificial neural network for predicting shaft and tip resistances of concrete piles figure 1. example of installed accelerometer (right) and strain transducer (left) on the pile. it should be mentioned that in performing pda tests, to obtain more reliable abc, some criteria such as hammer weight and impact condition must be considered. for full mobilization of soil strength along pile shaft, susilo (2006) suggested minimum hammer weight equals to 1% of the required ultimate pile capacity whereas for piles with larger expected end bearing contributions, the recommended percentage is increased to at least 2% of the ultimate pile capacity. 2.2 artificial neural network artificial neural network is a flexible non-linear function approximation tool that estimates a relationship between given input and output parameters. simpson (1990) reported that a specific ann can be defined using three important components: transfer function, network architecture and learning law. more details on ann structure are addressed elsewhere (e.g. hechtnielsen, 1990; maren, et al. 1990; zurada, 1992; fausett, 1994; ripley 1996). however, study by haykin (1999) recommends that the most wellknown type of feed-forward anns is multi-layer perceptron (mlp). in feedforward anns, the neurons are usually grouped into layers. using neuron connections, signals move from input layer through the hidden layer(s) to output layer. in essence, anns are composed of a set of parallel layers and several interconnected nodes or neurons. there is also a transfer or activation function along each node which transmits signals to either other nodes or output of the network. the activation function in each node is applied to the net input of that node. the net input of the node is obtained by summation of connection weights as well as a threshold value known as bias. among different algorithms for training anns, back-propagation (bp) algorithm is recognized as the most common training algorithm (dreyfus, 2005). basically, bp algorithm consists of two passes; a forward pass and a backward pass. in the former, using transfer function, the outputs are calculated and the errors at the actual output unit are determined (demuth et al. 2007). if the obtained error (mean squared difference between the actual and predicted outputs) is more than adequate, then the error is propagated back through the network and updates the individual weights. this procedure is called backward pass. forward and backward passes are repeated several times until the error is converged to a level specified by a cost function such as mean square error (mse) or root mean square error (simpson 1990, kosko 1994, singh et al. 2004). 3. dataset using the procedure suggested by astm (d4945-08), 36 pda tests were conducted at various project sites in indonesia. the tested piles were reinforced and pre-stressed concrete piles with different diameters and lengths. most of the tests were conducted in cohesionless soils. an example of pda test performed in one of the project sites is shown in figure 2. to represent soil characteristics, the results of spt were collected. the average spt (n) values along the pile shaft and tip were calculated. it is worth mentioning that for obtaining the average spt (n) value around the pile tip, the meyerhof ’s recommendation (1976) was considered. the average spt (n) value for 10d above and 4d below the pile tip was obtained where d represents pile diameter. table 1 lists the pda test results including ultimate bearing capacity of piles, pile set, pile shaft and tip resistances. in addition to pda results, pile geometrical characteristic and the average spt (n) values around the pile shaft and tip are also tabulated in table 1. figure 2. pda test 4. model development the prediction performance of anns is closely related to the architecture of the selected network. therefore, defining the optimum network architecture is crucial in designing ann models. hornik et al. (1989) mentioned that a network with one hidden layer can approximate any continuous function in ann. lawrence (1994) reported that in designing ann architectures, increasing the number of the hidden layers should be the last options; instead focus should be on adding the number of hidden nodes. nevertheless, for network construction, the optimum number of hidden nodes should be determined. several researchers suggested that numbers of hidden nodes are related to the number of input and output parameters. in table 2 some of the formulas suggested by a number of researchers for obtaining the optimum number of hidden nodes are presented. however, the network performance is often evaluated based on the root mean squared error (rmse) as well as regression values. that is to say, using the trial-and-error method, several network architectures are trained with same input and output data and the network that performs best is selected as the optimum network. 88 ehsan momeni, ramli nazir, danial jahed armaghani, harnedi maizir table 1. dataset used for ann-based predictive model. no. pile name pile length (m) pile set (mm) pile area (cm2) nshaft ntip qu (kn) qs (kn) qp (kn) 1 kn 189 25 16 1159.25 9 30 2988 1665 1323 2 kn 204 24.7 23 1159.25 9 30 3940 1445 2495 3 kn 215 19 19 1159.25 3 10 3200 1966 1234 4 p 105 bp6 22.7 3 11309.73 7 37 6753 3649 3103 5 fe p3 bp5 23 4 11309.73 4 21 8501 6035 2466 6 fe p6 bp6 23 3 11309.73 5 21 5322 3529 1794 7 p 112 bp 12 21 3 11309.73 5 15 6089 4293 1796 8 p 108 bp 11 23 5 11309.73 3 12 8500 5903 2597 9 p 109 bp 16 23 4 11309.73 3 12 10930 6412 4518 10 p 105 b2 23 7 11309.73 4 13 11601 7211 4390 11 p 105 b6 22.7 4 11309.73 4 13 8755 5535 3220 12 p 4 a 30 7 1159.25 20 27 1347 1197 150 13 p 18 f 30 7 1159.25 20 27 950 800 150 14 p 33 c 30 6 1159.25 20 27 1176 980 196 15 p 01 34.9 8 876.5 6 10 829 789 40 16 p 02 _ 1 34.9 8 876.5 6 10 605 591 14 17 p 04 34.8 13 552.92 6 10 772 744 28 18 p 05 34.9 11 552.92 6 10 781 774 7 19 p 1 as6bc.187 12.3 16 625 13 11 864 764 100 20 p 02 as21ed.349 10.6 8 625 15 13 689 637 52 21 p 03 as19ed.355 12.8 13 625 14 12 1520 1386 134 22 p01 10.3 9 735.13 18 26 500 485 15 23 p02 8.8 11 735.13 19 27 452 438 14 24 p03 10 6 735.13 18 25 584 556 29 25 p04 101 10 7 735.13 18 25 603 576 27 26 p06-112 10.3 6 735.13 18 26 811 768 44 27 p07-122 10 9 735.13 18 25 770 733 36 28 p08-46 10 6 735.13 18 25 748 700 48 29 p1-p330_3 sp 350 5.3 18 678.58 13 15 788 696 92 30 p2-p169 sp 350 5.3 15 678.58 13 15 794 700 94 31 p3-p144 sp 350 5.2 15 678.58 13 15 824 721 103 32 p1-e7 squar 25 19.6 15 625 11 22 1604 1437 167 33 p2-h 5_4 squar 25 19.7 12 625 11 22 1419 1269 150 34 p3-h 12 squar 25 19.1 12 625 9 19 1920 1720 200 35 b2 _1 5 16 625 8 11 764 670 94 36 d4_3 7.4 17 625 13 19 650 509 141 89application of artificial neural network for predicting shaft and tip resistances of concrete piles to determine the optimal network architecture of the ann model which is designed for predicting qs, qp and qu of piles respectively, using a matlab code created stochastically between -1 and 1, nine networks made of different hidden nodes in the range of 2 to 10 (based on the recommendations presented in table 2) were trained and tested. it should be mentioned that each model was iterated 5 times. for training the networks, levenberg-marquardt (lm) algorithm was used. several studies reported that lm converges while other gradient descendent training algorithms diverge (e.g. hagan and menhaj, 1995). details of this algorithm reported elsewhere (martin et al. 1995). it is also worth mentioning that in developing ann models, the sigmoid function was used as transfer function. assessments of the networks performance were made based on the obtained coefficient of determination, r² as well as rmse. the former indicates the reliability and strength of the correlation between actual and predicted outputs. table 3 lists the obtained r² and rmse for training and testing datasets. as shown in this table, the fourth model which comprises 5 hidden nodes in one hidden layer performs best. the obtained r2 and rmse values for the selected model are 0.998, 0.942, 0.012 and 0.091 for training and testing datasets respectively. the architecture of the selected model is shown in figure 3. it is worth noting that in designing the ann models, 80 percent ( 29 out of 36) of the datasets were assigned randomly for training purpose, and the last 20 percent was used for testing the performance of the model. table 3. ann model performances. model no. nodes in hidden layers network result iteration 1 iteration 2 iteration 3 iteration 4 iteration 5 train test train test train test train test train test r2 rmse r2 rmse r2 rmse r2 rmse r2 rmse r2 rmse r2 rmse r2 rmse r2 rmse r2 rmse 1 2 0.974 0.078 0.761 0.136 0.978 0.064 0.910 0.515 0.902 0.096 0.674 0.392 0.988 0.089 0.879 0.153 0.961 0.081 0.853 0.250 2 3 0.980 0.065 0.674 0.288 0.978 0.166 0.783 0.226 0.990 0.065 0.832 0.152 0.982 0.104 0.842 0.149 0.976 0.133 0.891 0.155 3 4 0.994 0.071 0.783 0.178 0.994 0.043 0.906 0.135 0.990 0.058 0.784 0.149 0.992 0.037 0.637 0.217 0.968 0.036 0.861 0.194 4 5 0.998 0.016 0.896 0.155 0.998 0.040 0.904 0.107 0.998 0.012 0.942 0.091 0.998 0.019 0.774 0.177 0.998 0.095 0.758 0.157 5 6 0.998 0.017 0.763 0.156 0.998 0.032 0.776 0.252 0.998 0.021 0.794 0.237 0.998 0.024 0.605 0.277 0.998 0.033 0.777 0.117 6 7 0.998 0.022 0.721 0.167 0.998 0.028 0.740 0.187 0.998 0.043 0.629 0.209 0.998 0.031 0.921 0.125 0.998 0.067 0.790 0.168 7 8 0.998 0.049 0.850 0.150 0.998 0.116 0.885 0.139 0.998 0.044 0.589 0.231 0.998 0.033 0.784 0.189 0.998 0.020 0.785 0.211 8 9 0.998 0.051 0.851 0.185 0.998 0.042 0.571 0.262 0.998 0.039 0.752 0.201 0.998 0.098 0.870 0.262 0.998 0.072 0.767 0.291 9 10 0.998 0.038 0.706 0.321 0.998 0.055 0.794 0.202 0.998 0.088 0.887 0.282 0.998 0.075 0.715 0.371 0.998 0.061 0.727 0.299 figure 3. architecture of the selected ann-based predictive model. 5. result and discussion the reliability of the ann-based predictive model of bearing capacity can be seen in figures 4 to 6. these figures show the predicted qs, qp and qu of piles versus their measured values. figure 4 shows a comparison between predicted and measured qs for training and testing data. the obtained r² values equal to 0.999 and 0.941 suggest the reliability of the model in predicting qs. similarly figure 5 suggests that the predicted qp is in good agreement with the measured qp. as displayed in figure 5-b, the coefficient of determination equals to 0.936 for testing data recommends the feasibility of the ann-based predictive model of bearing capacity. table 2. the proposed number of neurons for hidden layer ni: number of input neuron, n0: number of output neuron. 90 ehsan momeni, ramli nazir, danial jahed armaghani, harnedi maizir figure 4. shaft bearing capacity (qs) of piles predicted by ann model versus their measured values. figure 5. end-bearing capacity (qp) of piles predicted by ann model versus their measured values. in figure 6, a comparison is made between the measured and predicted qu of piles for both training and testing data. coefficient of determination equals to 0.951 for testing data suggests that the ann-based predictive model is good enough in capturing the ultimate bearing capacity of piles. figure 6. ultimate bearing capacity (qu) of piles predicted by ann model versus their measured values 91application of artificial neural network for predicting shaft and tip resistances of concrete piles to have a better understanding of the prediction performance of ann, in figures 7 to 9 the predicted qs, qp, and qu are checked against their measured values for testing dataset. in fact, the idea behind using testing dataset is to verify the generalization capability of proposed neural network model. close agreement between measured and predicted values suggests that by using pile geometrical characteristics, pile set and the results of spt insitu test, the ann-based predictive model can be implemented for estimating qs, qp, and qu of piles. figure 7. ann-model performance in predicting pile shaft resistance (testing data). figure 8. ann-model performance in predicting pile tip resistance (testing data). figure 9. ann-model performance in predicting ultimate bearing capacity of piles (testing data). the overall prediction performance (for all dataset) of the annbased model is summarized in table 4. in this table, apart from r², rmse and value account factor (vaf) were also used to control the capacity performance of the model. for determining rmse and vaf, equations 1 and 2 were used respectively. rmse = vaf = (1) (2) in the aforementioned equations, y and y′ denote the obtained and estimated values, respectively and n is the total number of data. it is worth mentioning that the model is excellent if the rmse is zero and vaf is 100. table 4. performance indices of the predictive model output r2 rmse vaf (%) qp 0.982 0.081 98.199 qs 0.984 0.075 98.361 qu 0.988 0.064 98.770 overall, the general trend of the results shows that ann as a method which does not require prior assumptions and can provide a relatively reliable solution for assessing the pile bearing capacity and its distribution. although direct determination of pile bearing capacity through slt is still recommended due to the amount of uncertainties in other semi empirical methods (momeni et al, 2013), the use of proposed ann-based predictive model is of advantage as it can reduce the required number of pda tests in each project. 6. sensitivity analysis sensitivity analysis was performed to recognize the importance of each input variable on the axial bearing capacity of piles. for this reason, the strength of the relations between the output parameters and the input parameters was evaluated using cosine amplitude method (cam). this use of this sensitivity analysis is reported in several studies (yang and zhang, 1997; jong and lee, 2004). to utilize cam, all data pairs were expressed in common u-space. the data pairs used to construct a data array u is defined as: u= the elements ui in the array u is a vector of lengths m that is: (3) (4) therefore, each data pair can be considered as a point in m-dimensional space, where each point requires m-coordinates for a full description. the strength of relation between data pairs, ui and uj, is represented by the following equation: (5) the strength of the relation (rij value) indicates the influence of different input parameters on one of the output parameters. the larger the value of rij becomes, the higher is the effect on the output. for example, if the output has no relation with the input, then the rij value is zero, while the value of rij closer to 1 expresses the further influence of the input parameter. nevertheless, the obtained strength of relations of the problem in hand is shown in figure 10. this figure suggests that the most influential parameters on qs, qp and qu are pile area and pile length. 92 ehsan momeni, ramli nazir, danial jahed armaghani, harnedi maizir figure 10. strengths of relation (rij) between qs and qp and input parameter 8. summary and conclusion to develop an ann-based predictive model for estimating the axial bearing capacity of piles, 36 pda tests were performed on different concrete piles with various diameters and lengths. the tests were mostly performed in cohesionless soils. for network construction purpose, the pda results, pile length and cross sectional area, and the average spt (n) values along the pile shaft and tip were used as inputs while the pile bearing capacity and its distribution were set as outputs. through a trial-and-error procedure, it was found that a network with five hidden nodes in one hidden layer yields the best performance. the coefficients of determination equal to 0.941, 0.936, and 0.951 for testing data revealed the reliability of the proposed ann model in predicting the shaft, tip and ultimate bearing capacities of piles, respectively. additionally, through a sensitivity analysis, it was found that the pile length and cross sectional area are the most influential parameters in predicting the bearing capacity of piles. acknowledgement the authors would like to thank the research management centre of universiti teknologi malaysia (utm) and ministry of science, technology and innovation (mosti) for providing financial support through research vote: 4s077 for bringing the idea into fruition. references abu-kiefa, m. (1998). general regression neural networks for driven piles in cohesionless soils, journal of geotechnical and geoenvironmental engineering. 124 (12), 1177–1185. adarsh, s., dhanya, r., krishna, g., merlin, r. and tina. j. (2012). prediction of ultimate bearing capacity of cohesionless soils using soft computing techniques, artificial intelligence. doi:10.5402/2012/628496 alkroosh, i. and nikraz, h. (2012). predicting axial capacity of driven piles in cohesive soils using intelligent computing, engineering applications of artificial intelligence. 25(3), 618-627. american society for testing and materials (2010). standard test methods for deep foundations under static axial compressive load. d 1143-07, annual book of astm standards. astm, philadelphia, pa, 4:07. american society for testing and materials (2010). standard test method for high-strain dynamic testing of piles, d 4945-08, annual book of astm standards. astm, philadelphia, pa, 4, 08. ardalan, h., eslami, a. and nariman-zadeh, n. (2009). piles shaft capacity from cpt and cptu data by polynomial neural networks and genetic algorithms, computers and geotechnics. 36, 616–625. benali, a. and nechnech, a. (2011). prediction of the pile capacity in purely coherent soils using the approach of the artificial neural networks, international seminar, innovation and valorization in civil engineering and construction materials, morocco – rabat / november 23-25. chan, w. t., chow, y. k. and liu, l. f. (1995). neural network: an alternative to pile driving formulas, computers and geotechnics. 17(2), 135-156. chow, y. k., chan, w. t., liu, l. f., and lee, s. l. (1995). prediction of pile capacity from stress-wave measurements: a neural network approach, international journal for numerical and analytical methods in geomechanics. 19(2), 107-126. coyle, h. m. and castello, r. r. (1981). new design correlations for piles in sand. journal of the geotechnical engineering division, asce. 107, 965–986. das, s. k. and basudhar, p. k. (2006). undrained lateral capacity of piles in clay using artificial neural network, computers and geotechnics. 33 (8), 454–459. demuth, h., beal, m. and hagan, m. (2007). neural network toolbox user’s guide, natick, ma: the mathworks. dreyfus, g. (2005). neural networks: methodology and application, germany, springer berlin heidelberg. fausett, l. v. (1994). fundamentals neural networks: architecture, algorithms, and applications, prentice-hall, inc., englewood cliffs, new jersey. fellenius, b. h. (1999). using the pile driving analyzer, pile driving contractors association, pdca, annual meeting, san diego, february 19-20, 1999, 4 p. fhwa. (2006). design and construction of driven foundations-lesson learned on the central artery/tunnel project, report no. fhwahrt-05-159, washington, dc. garret, j. h. (1994). where and why artificial neural networks are applicable in civil engineering, journal computer and civil engineering. 8(2), 129–130. goble, g. g., rausche, f. and moses, f. (1970). dynamic studies on the bearing capacity of piles phase iii, final report to the ohio department of highways, case western reserve univ, cleveland, ohio. goh, a. t. c. (1995). back-propagation neural networks for modeling complex systems, artificial intelligence in engineering. 9, 143-151. goh, a. t. c. (1996). pile driving records reanalyzed using neural networks, journal of geotechnical engineering. 122(6), 492–495. haykin, s. (1999). neural networks, 2nd ed, englewood cliffs, nj: prentice-hall. hagan, m.t. and menhaj, m.b. (1994). training feed forward networks with the marquardt algorithm. ieee transactions on neural networks 5 (6), 861–867. hecht-nielsen, r. (1990). neuro computing, addison-wesely publishing company. hornik, k. stinchcombe, m. white, h. (1989). multilayer feedforward networks are universal approximators, neural networks 2, 359–366. jianbin, z., jiewen, t. and yongqiang, s. (2010). an ann model for predicting level ultimate bearing capacity of phc pipe pile, earth and space. pp, 3168-3176. jong, y. h. and lee, c. i. (2004). influence of geological conditions on the powder factor for tunnel blasting, international journal of rock mechanics and mining sciences. 41, 533–538. kaastra, i. boyd, m. (1996). designing a neural network for forecasting financial and economic time series. neurocomputing 10, 215-36. kanellopoulas i, wilkinson g. g. (1997) strategies and best practice for neural network image classification. international journal of remote sensing 18, 711-25. kosko, b. (1994). neural networks and fuzzy systems: a dynamical systems approach to machine intelligence, prentice hall, new delhi. lawrence, j. (1994). introduction to neural networks: design, theory, and applications, 6th ed. nevada city, ca: california scientific software. lee, i. m. and lee, j. h. (1996). prediction of pile bearing capacity using artificial neural networks, computers and geotechnics. 18(3), 189-200. likins, g. and rausche, f. (2004). correlation of capwap with static load test. proceedings of the seventh international conference on the application of stress wave theory to piles 2004, the institute of engineers malaysia. lok, t. m. h. and che, w. f. (2004). axial capacity prediction for driven piles using ann: model comparison, geotrans 2004, geotechnical engineering for transportation projects (gsp no. 126), m. k. yegian and e. kavazanjian, eds. 93application of artificial neural network for predicting shaft and tip resistances of concrete piles maren, a., harston, c. and pap, r. (1990). handbook of neural computing applications, academic press, inc., san diego, california. martin, t. h. howard, b. d. mark, b. (1995). neural network design. pws publishing company, boston, ma. masters, t. (1994). practical neural network recipes in c++. boston ma: academic press. meyerhof, g. g. (1976). bearing capacity and settlement of pile foundation, journal of the geotechnical engineering division asce. 102, 197–228. momeni, e., maizir, h., gofar, n., nazir, r. (2013). comparative study on prediction of axial bearing capacity of driven piles in granular materials. jurnal teknologi, 61(3) 15-20. nazir, r., momeni, e., gofar, n., maizir, h. (2013). numerical modelling of skin resistance distribution with depth in driven pile, electronic journal of geotechnical engineering (ejge), 18 (l) 2477-2488. pal, m. and deswal, s. (2008). modeling pile capacity using support vector machines and generalized regression neural network, journal of geotechnical and geoenvironmental engineering. 134(7), 1021-1024. paola j. d. (1994). neural network classification of multispectral imagery. master thesis, the university of arizona, usa. randolph, m. f. (2003). science and empiricism in pile foundation design, 43rd rankine lecture, geotechnique. 54(1). rausche, f., goble, g. g. and likins, g. e. (1985). dynamic determination of pile capacity, journal of geotechnical engineering. 111(3), 367–383. rausche, f., moses, f. and goblen, g. g. (1972). soil resistance predictions from pile dynamics, journal of the soil mechanics and foundation division asce. september 1972. ripley bd (1993) statistical aspects of neural networks. in: barndoffneilsen oe, jensen jl, kendall ws, editors. networks and chaos-statistical and probabilistic aspects. london: chapman & hall, pp 40-123. ripley, b. d. (1996). pattern recognition and neural networks, cambridge university press. salgado, r. (2008). the engineering of foundations, mcgraw hill book, usa. shahin, m. a. (2010). intelligent computing for modelling axial capacity of pile foundations, canadian geotechnical journal. 47(2), 230-243. shahin, m. a. (2008). modelling axial capacity of pile foundations by intelligent computing, proceedings of the bga international conference on foundations, dundee, scotland, his bre press, isbn 978-1-84806-044-9, 283-294. shahin, m. a., jaksa, m. b. and maier, h. r. (2001). artificial neural network applications in geotechnical engineering, australian geomechanics journal. 36(1), 49-62. shahin, m. a., jaksa, m. b., and maier, h. r. (2009). recent advances and future challenges for artificial neural systems in geotechnical engineering applications, advances in artificial neural systems. doi:10.1155/2009/308239. simpson, p. k. (1990). artificial neural system: foundation, paradigms, applications and implementations, new york, pergamon. singh, t. n., kanchan, r., saigal, k. and verma, a. k. (2004). prediction of p-wave velocity and anisotropic properties of rock using artificial neural networks technique, journal of scientific and industrial research. 63, 32-38. smith e. a. l. (1960). pile-driving analysis by the wave equation, journal of soil mechanic and foundation division asce. div, 86 (em 4):35-61. susilo, s. (2006). distribusi gesekan tanah pada pondasi tiang bor dalam. pertemuan llmiah tahunan-x hatti, jakarta. 107–119 teh, c. i., wong, k. s., goh, a. t. c. and jaritngam, s. (1997). prediction of pile capacity using neural networks, journal of computing in civil engineering. 11(2), 129-138. timoshenko, s. and goodier, j. m. (1951). theory of elasticity, (2nd ed), mcgraw-hill book co.p.438. usa. vesic, a. s. (1977). design of pile foundation, national cooperative highway research program synthesis of practice no.42, transportation research board, washington, dc. wang, c. (1994). a theory of generalization in learning machines with neural application. phd thesis, the university of pennsylvania, usa. yang, y. and zang, o. (1997). a hierarchical analysis for rock engineering using artificial neural networks, rock mechanics and rock engineering. 30, 207–222. zurada, j. m. (1992). introduction to artificial neural systems, west publishing company, st. paul. more than 40% area of the xixian new area is a loess deposit region, and most of the loess landform is tableland and terrace where the thickness of loess is considerable large. therefore, loess collapsibility will be the most important geotechnical problem in future foundation investigation and construction. to explore loess collapsibility in the xixian new area, we conducted the k0 compression test, based on the collapsibility mechanism, which has different combinations of pressure (0~1.2 mpa) and water content (4%~sat). based on the σ-ε curve under different water content, we calculated the generalized collapse settlement and collapsibility coefficient of every water content under every pressure by subtracting the relevant curve from the saturated curve and analyzed the cross action of pressure and water on loess compressibility. the results show that the average collapsibility level of the northern xixian new area is self-weight collapsible level ii, with a lower limit of 14 m. the compressibility of loess is proportional to pressure and water content. under low water content, the collapsibility coefficient δs increased while the pressure increased, but under medium and high water content, δs will reach a peak with increasing pressure and after that, δs will decrease until its value is close to constant. when under the same pressure, δs decreases when water content increases. if set the additional strain 1.5% as collapse start criterion, then the first collapse pressure pi will linear proportional to water content. the initial collapse water content wi will increase sharply when pressure raises under low pressure, but wi will reach a constant value of 26% when pressure is more than 200 kpa. this consequence will be meaningful for future geotechnical investigation and design in the xixian new area. abstract keywords: xixian new area; loess collapsibility; wetting; pressure; initial collapse water content influence of pressure and water content on loess collapsibility of the xixian new area in shaanxi province, china issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n4.66106 earth sciences research journal earth sci. res. j. vol. 21, no. 4 (december, 2017): 197-202. jiading wang*, yan ma, qianyi guo, di chu state key laboratory of continental dynamics, northwest university, xi’an 710069, china *corresponding author of email: wangjd@nwu.edu.c para estimar la colapsabilidad en xixian se realizó el test de compresión (k0), de acuerdo con el mecanismo de colapsabilidad, con diferentes combinaciones de presión (0~1.2 mpa) y contenido de agua (4%~sat). basado en la curva σ-ε con diferentes contenidos de agua se calcularon los coeficientes de asentamiento y colapsabilidad a partir de substracción de la curva en condiciones de saturación y analizando los efectos de presión y contenido de agua. los resultados muestran que el promedio de colapsabilidad por su propio peso para el área de xixian es de nivel ii, con un límite bajo de hasta 14 m. la compresibilidad de loess es proporcional a la presión y al contenido de agua. con un bajo contenido de agua, el coeficiente de colapsabilidad (δs) se incrementó mientras la presión se aumentó; pero, con contenidos medios y altos de agua, el coeficiente de colapsabilidad alcanzó su pico con el incremento de la presión, y luego bajó para prácticamente estabilizarse. bajo presión constante, la colapsabilidad descendió al aumentar el contenido de agua. si se asume como criterio inicial de colapso (pi) una deformación adicional de 1.5%, pi es linealmente proporcional al contenido de agua. el contenido de agua inicial de colapso (wi) se incrementa drásticamente cuando la presión se eleva a baja presión, sin embargo, el contenido de agua alcanza un valor constante de 26 % cuando la presión es de más de 200 kpa. estos resultados son significativos para la investigación geotécnica y el diseño estructural en xixian. resumen palabras clave: distrito especial de xixian (china); colapsabilidad de loess; humectación; presión; contenido de agua incial de colapso. record manuscript received: 05/07/2017 accepted for publication: 16/11/2017 how to cite item: wang, j., ma, y., guo, q., & chu, d. (2017). influence of pressure and water content on loess collapsibility of the xixian new area in shaanxi province, china. earth sciences research journal, 21(4). 197-202. doi: http://dx.doi.org/10.15446/esrj.v21n4.66106 influencia de la presión y el contenido de agua sobre el índice de colapsabilidad de loess, en el distrito especial xixian, provincia de shaonxi, china h y d r o l o g y 198 jiading wang, yan ma, qianyi guo, di chu 1. introduction china’s loess plateau is a vast territory with a large population. however, since the late quaternary, the climate and ecological environment in this area have been deteriorating, resulting in frequent loess disasters, which seriously restrict economic development and urban construction (zhang, 1980; abbasi et al., 2017). loess is a special soil that has a loose structure, high porosity and collapse and vibration sensitivity (derbyshire, 1995). therefore, the study of collapsible loess plays an important role in urban construction in the loess area. because of its special overhead pore structure and high sensitivity of its microstructure to interactions between water and force, aeolian loess has significant collapsibility. previous researchers (rabinovich & urinov, 1974; kozubal & steshenko, 2015) have studied the collapsibility of loess. gao (1980) and lei (1987) came up with a classification for loess microstructure that achieved good results, which has been used up to today. many scholars (zhenghan & zudian, 1986; yang, 1988; miao, 1999; wang & gu, 2013, zhang, 2000; bakar et al., 2017) conducted an in-depth study of the mechanism of collapsible deformation of loess. chinese studies (zhang, 1995; liu, 1997; zhang, 2000; ma, 2014; usman et al., 2017) developed research on engineering properties of loess in china. in engineering practice, a standardized evaluation and treatment system has been established by collapsibility and lower limit depth (people’s republic of china national standard for building construction in collapsible loess area gb 50025-2004). however, from the above common features, the engineering properties of loess in china have regional characteristics (bata et al., 2017). differences in spatial distribution of loess deposits, provenance differences, and a series of physical and chemical reactions that differ from region to region after deposition affect the microstructure characteristics of loess and change its macroscopic physical and mechanical properties. more than 40% area of the xixian new area is a loess deposit region. most of the loess landforms are tableland and terrace where loess thickness is very large. therefore, loess collapsibility is the most important geotechnical problem in future foundation investigation and construction. as a new state-level development area, the xixian new area is an important part of the western development and is the integration of xi’an and xianyang. it is also an important link in the strategy of building an international metropolis. the loess plateau area in the northern part of the xixian new area will allow the development of logistics and cultural tourism industries around the existing xianyang airport (airport metro) and historical and cultural scenic spots (qin and han dynasties). the scale and span of construction in this area will expand rapidly with an influx of investment over a short period. the number and scale of original construction located in the loess plateau area in the northern part of the xixian new area is small and large portions to be developed are still farmland; therefore, we have accumulated less basic engineering-related data in this area. meanwhile, there is little research on the engineering properties of loess in this area. also, the engineering properties of loess vary in space. therefore, studying the collapsibility of loess in the xixian new area will have significant application value and scientific significance (gao et al., 2017). the traditional collapsibility of loess is the nature of the additional settlement caused by the saturation of saturated loess when the deformation of loess under certain pressure is stable. to deepen the understanding of the mechanism of loess collapsibility, some researchers (zhang, 1990, 1992; sun & liu, 2000, francisca, 2007; yasin et al., 2017) have studied the collapsible deformation of loess during the processes of humidification and dehumidification. the definition of collapsibility is extended to the nature of additional settlement of loess by humidification under a certain pressure to form a stable deformation (yew & rahim, 2017). the collapsible deformation of loess is considered a kind of plastic deformation, and the amount of collapsible deformation is related to the state of humidification (hussin et al., 2017). it is independent of the infiltration path, which indicates that loess has a memory of moistening deformation under certain pressures. however, the amount of humidification deformation and its variation law in loess are both different from the soil, so it is impossible to establish a relational expression for calculating deformation (ismail et al., 2017; sarkar et al., 2017). based on this, we carried out a pre-investigation of loess in the xixian new area, and the k0 test for the typical horizon under different water content and pressure. the purpose of the study is to provide reference and scientific guidance for future engineering investigation and design in the loess distribution area in the northern part of the xixian new area (tariq et al., 2017; tunggolou & payus, 2017). 2. test schemes 2.1 physical parameters of soil foundation the study area is located between the cities of xian and xianyang. the location of this area in shaanxi province is shown in figure 1. five exploratory wells as the cross-shaped of the five endpoints were placed on the large thickness loess plateau in the north of xixian new area (as fig. 1), and the depth of exploratory wells is 30 m. the sampling (fig. 2) and testing of the t1, t2, t3, t4 and t5 exploration wells were completed by the state key laboratory of continental dynamics in the northwest university of china (fig. 3). t1, t2, t3 and t5 exploration wells were located in the loess plateau, and t4 was located in the second terrace of the weihe river. by comparing the five groups of experimental data, we found that the t4 site was a site of first-grade self-weight collapsibility. also, the four sites in the loess plateau are sites of second-grade selfweight collapsibility. moreover, the closer a sampling point is to the center of the loess plateau, the greater the amount of collapsibility and self-weight collapsibility and the deeper the depth of collapse. although the amount of collapsibility and self-weight collapsibility in each sampling site are different, the homogeneity of soil samples at a depth of 5 m is higher, which is shown in table 1. therefore, selected the malan loess located 5 m deep at the t3 site near the xianyang airport as the representative layer. we further studied the influence of pressure and water content on the collapsibility of loess in the xixian new area. figure 1. the sketch of location, landform, and sample diagrammatic in xixian new area 199influence of pressure and water content on loess collapsibility of the xixian new area in shaanxi province, china figure 2. sampling in site exploration wells figure 3. automatic air pressure consolidation instrument (gzq-1) at t3, two 6-m-deep exploratory wells were re-excavated, and 40 soil samples were taken from the borehole wall at 5 m. in the field, the soil samples were sealed with plastic wrap to keep moisture content unchanged and then wrapped in bubble film as a buffer. basic physical and mechanical parameters tests were carried out immediately after samples were transported to the laboratory. these tests measured the natural density, dry density, plastic limit, liquid limit, plastic index, liquid index, void ratio, water content, compression coefficient, compression modulus, the coefficient of collapsibility and coefficient of self-weight collapsibility of soil. the collapsibility test used the single line method. table 1. physical and mechanical properties of ma lan loess at a depth of 5 m in each sampling well 2.2 high-pressure k0 compression test scheme for loess with different water content to explore the effects of interaction between water content and pressure on the compressibility of undisturbed loess in the xixian new area, we designed a k0 compression test scheme with ten kinds of water content and nine levels of pressure (see table 2). the preparation methods of soil samples with different water content are as follows. first, we cut ring-knife samples from massive soil samples at the site. each sample was 2 cm high and had a surface area of 50 cm2. we tested soil moisture content in the upper and lower parts of the ring-knife sample and took the average value. based on the moisture content of each sample, we calculated the amount of water that needed to be increased according to the target moisture content. to ensure uniform moisture content, we dropped water into soil samples that needed wetting, with 6 g/d as an upper limit. soil samples that are necessary to be dehumidified were uniformly dried to a lower water content (about 3.5%). then we calculated the current water content according to the sample’s measured mass. finally, we prepared soil samples using the same humidification method. we prepared saturated soil samples by saturating the soil sample with the natural water content after the first stage pressure compression. to prevent a humidity difference between permeable stone, filter paper and soil samples from affecting humidity test results, we put the permeable stone and filter paper into the soil, which was mixed with water to the same water content as in table 2 after being dried to reach a humidity balance. the test instrument was an automatic air pressure consolidation apparatus (gzq-1 type) produced by nanjing soil instrument factory (fig. 3), and the test methods refer to ‘the standard of soil test methods’ (highway research institute of ministry of communications gb/t 50123 1999 geotechnical test method standard). table 2. k0 compression test scheme 3. test results and analysis 3.1 effect of pressure and water content on compressive deformation properties of loess in the xixian new area the relation curves of pressure and strain at different levels of water content (fig. 4) and the curves of water content and strain under different levels of loess pressure in the xixian new area (fig. 5) were all obtained using the k0 test with 10 kinds of water content and 9 levels of pressure. the bivariate trend surface of compressibility coefficient, water content and pressure are plotted based on the two sets of data 200 jiading wang, yan ma, qianyi guo, di chu figure 4. p-ε curves of loess under different water content figure 5. w-ε curves of loess under pressure at all levels figure 7. p-ε curves of loess under different water content figure 8. δs-w curves of loess under different pressures figure 6. bivariate trend surface of compressive strain and water content, pressure of loess as figure 4 suggests, compressive deformation increases with the increase of pressure at all levels of water content, but the increasing regularity with increasing water content is different. under the low water content conditions (4%, 7%, and 10%) the pε curves of loess in the xixian new area are linear. as the water content increased to 13%, the p ε curve began to show an inflection point, which indicated the axial strain increased significantly with the increase of pressure, and the shape of the curve changed from linear to s-sharped. the position of the pressure value corresponding to the inflection point decreased with increased water content. this phenomenon reflects a law that the mechanical stability of loess microstructure decreases as water content increases. as figure 5 suggests, compressive deformation increases with the increase of water content at all pressure levels, but the increasing regularity with increasing pressure is different. under low-pressure conditions (50, 100 and 150 kpa), when moisture content increases from 4% to 40%, strain increases linearly. with the increase of pressure, the wε curve, similar to the p-ε curve, also began to show an inflection point. the water content corresponding to the inflection point decreases as pressure increases, and the rise of water content means a decrease in suction. this phenomenon reflects a law that the water stability of loess microstructure decreases with increasing pressure. as figure 6 suggests, pressure and water content affect the compression deformation of loess. in general, there is a positive correlation between water content, pressure and compression deformation. these two factors and the compression coefficient form a three-dimensional trend surface similar to a ‘loess hill’ in which the lower part is gentle, the middle is steep, and the top is smooth again. the gentle lower part indicates that humidification under low-pressure conditions and compression under low water content conditions will not cause a large compressive deformation. the steep middle shows that humidification under middle-pressure conditions and compression under middle water content conditions will cause compressive deformation to increase significantly. the recovery of the smooth slope at the top indicates that the growth of this compressive strain was restored to a gentle trend under high pressure and high-water content. 3.2 effect of pressure and water content on compressive deformation properties of loess in xian new area the data in figure 4 are equivalent to carry out nine sets of double wire test of soil samples with different initial water content. the difference of strain was obtained by subtracting the compression curves of unsaturated soil samples under nine levels of water content from the full compression that is loess collapsibility. the relation curves of loess collapsibility coefficient and pressure at different levels of water content are plotted in figure 7, which shows relation curves between collapsibility coefficient and water content of loess under different pressure levels 201influence of pressure and water content on loess collapsibility of the xixian new area in shaanxi province, china as figure 7 suggests, under low water content conditions (4%, 7%, 10%, and 13%), as pressure increases, the loess collapsibility coefficient in the xixian new area increases slowly at first, then decreases gradually after reaching a particular pressure, and finally, the growth rate gradually slows down. the δs-p curve is a sigmoid. under medium-high water content conditions (16%, 19%, 22%, 25% and 28%), as pressure increases, the collapsibility coefficient increases to a peak value and then decreases, eventually becoming constant. the reason for this phenomenon is that the strength of the cement in the skeleton and particles of loess is larger under low water content conditions, and the effective stress caused by matrix suction on the particles is larger, which leads to an increase in friction between particles. therefore, the strength and stiffness of soil microstructure are lower under low water content conditions, and an increase of pressure cannot do a large amount of damage to loess pore structure. finally, the compression deformation is small before soaking. if the deformation of saturated soil under pressure at all levels is the total macroscopic deformation of the deformable microstructure under that pressure, the amount of deformation before saturation will be smaller, the amount of collapsibility after saturation will be greater, and the corresponding coefficient of collapsibility will also be greater. for soil samples under high water content conditions, as water content increases, the infiltration of water will weaken the contribution of the cement and matrix suction to the stable state of the particles. the collapsibility coefficient increases as pressure increases before the pressure has not reached the microstructural strength at this saturation state. however, when the pressure exceeds the micro structural strength at this saturation state, the large amount of deformation of the compression stage is caused by a large amount of destroyed overhead pore structure. once the structural strength is exceeded, deformation increases as pressure increases and the proportion of the total deformation is larger. finally, the collapsibility coefficient decreases as pressure increases. as figure 8 suggests, the collapsibility coefficient decreases with the increase of water content at all pressure levels. when loess is under the combined action of certain pressure and soaking saturation, the adjustable upper limit of pore structure is the total strain of saturated soil under this pressure. the whole strain is composed of compression deformation before saturation and collapsible deformation after saturation. the reasons for the decrease of the collapsibility coefficient with the increase of water content are as mentioned earlier. under constant pressure, microstructure strength decreases as soil moisture content increases, and the deformation is greater in the pre-compression stage. after that, the proportion of the deformation in the collapse stage (i.e. the collapsibility coefficient) will be reduced accordingly. the initial collapse pressure is a crucial parameter in the investigation and design of a collapsible loess foundation and takes an additional settlement caused by saturation of more than 1.5% as the criterion of collapsibility. based on the data in figure 7, we used an interpolation method to calculate the initial pressure, pi, of the collapsible loess at all levels of water content and then draw the curve of pi with water content (fig. 9). meanwhile, based on the data in figure 8, we used an interpolation method to calculate the initial water content, wi, of the collapsible loess at all levels of pressure and then draw the curve of wi with pressure (fig. 10). as figure 9 suggests, the initial pressure of loess collapsibility in the xixian new area increases linearly as water content increases. this phenomenon is due to the compressibility of soil rising as water content increases. therefore, with the increase of water content, a part of the adjustable overhead pore structure can be compressed before it reaches saturation, and to produce more than 1.5% of the collapsible strain after soaking, a greater axial pressure is required. as figure 10 suggests, under low-pressure conditions (100 and 150 kpa), the initial moisture content of loess in the xixian new area increased sharply with the increase of pressure. this shows that the microstructure of loess is more sensitive to humidification under low pressure. but after the pressure reached 200 kpa, and up to 1.2 mpa, the moisture content of collapsibility was stable at about 26%. the results show that loess is not sensitive to the lower level of humidity under high pressure. this is because porous loess structure appears to be micro arch, and the normal stress between particles increases as pressure increases before reaching the structural strength under the initial water content. accordingly, friction resistance is also stronger, and the suction drop effect caused by low-level humidification is offset. the moisture content of 26% is the critical value of moistening deformation of loess in the xixian new area under pressures of more than 200 kpa. when the moisture content does not exceed the critical value, it will not produce significant collapsibility. 4 conclusion based on five representative sites in the loess plateau and the high order of loess deposition area reconnaissance, digging and indoor tests, we evaluated collapsibility grade of loess in the northern part of the xixian new area. at the same time, we carried out a special experimental study on the collapsibility of typical loess stratum. conclusions are as follows: 1. we found that the t4 site was a site of first-grade self-weight collapsibility. in addition, the four sites in the loess plateau are sites of second-grade self-weight collapsibility. meanwhile the closer to the loess plateau hinterland, the larger the collapsibility of the site, which makes for a lower depth of collapsibility. 2. pressure and water content affect the compression deformation of loess. in general, there is a positive correlation between water content and pressure. these two factors and the compression coefficient form a surface with an initial gentle slope, a steep slope in the middle, and return to gentle slope surface at the top. the surface has a shape similar the ‘loess hill’ three-dimensional trend surface. figure 9. relationship between the initial pressure of collapsible loess and moisture water figure 10. relationship between initial water content of collapsible loess and pressure 202 jiading wang, yan ma, qianyi guo, di chu 3. under the same water content, when wi was lower (4%, 7%, 10% and 13%), the coefficient of collapsibility of loess in the xixian new area increased slowly as pressure increased, and then the coefficient of collapsibility increased sharply. finally, the growth rate slowed down gradually. at the same time, the δs -p curve was s shaped. for middle and high water content (16%, 19%, 22%, 25%, and 28%), the collapsibility coefficient increased to a peak value as pressure increased. then it began to decrease and tended to increase with increasing pressure. under the same pressure, the collapsibility coefficient decreases gradually as moisture content increased. 4. if the additional settlement is greater than 1.5% for collapsibility criterion, the initial pressure of loess collapsibility increases linearly with the increase of water content in the xixian new area. however, the initial moisture content of collapsibility increases sharply with the rise of pressure at low pressure (100 and 150 kpa). additionally, when the pressure is between 200 and 1.2 mpa, the moisture content of collapsibility is stable at about 26%. acknowledgement the paper is mainly funded by the state key program of national natural science of china (41630639) and national natural science foundation of china (41372269) and the state key laboratory of continental dynamics foundation (2014). herein, i express my deepest gratitude to them and those who have given me further help during the completion of this paper. references abbasi, n., khan, m.s., mughal, m.s., & yasin, m. (2017). sedimentary facies analysis of nagri formation, kashmir basin, subhimalayas, pakistan. pakistan journal of geology, 1(2), 03–06. bakar, b., tahir, s.h., & asis, j. (2017). deep marine benthic foraminiferal from temburong formation in labuan island. earth science malaysia, 1(2), 17-22 bata, t., samaila, n.k., maigari, a.s., & ikyoive, m.b.s.y. (2017). common occurences of authentic pyrite crystals in cretaceous oil sands as consequence of biodegradation processes. geological behavior, 1(2), 26–30. derbyshire, e., meng, x., & wang, j. (1995), collapsible loess on the loess plateau of china. in: genesis and properties of collapsible soils. volume 468 of the series nato asi series, pp 267-293 francisca, f. m. (2007). evaluating the constrained modulus and collapsibility of loess from standard penetration test. international journal of geomechanics, 7(4), 307-310. gao, g. (1980). microstructure classification and collapsibility of loess. chinese science, 12, 1203-1208, 1237-1240. (in chinese) gao, w., kanna, m.r.r., suresh, e., & farahani, m.r. (2017). calculating of degree-based topological indices of nanostructures. geology, ecology, and landscapes, 1(3), 173-183. highway research institute of ministry of communications (1999). gb/t 50123 1999 geotechnical test method standard. beijing: china architecture and building press. (in chinese) hussin, h., fauzi, n., jamaluddin, t.a., & arifin, m.h. (2017). rock mass quality effected by lineament using rock mass rating (rmr) – case study from former quarry site. earth science malaysia, 1(2), 13-16. ismail, i., husain, m.l., & zakaria, r. (2017). attenuation of waves from boat wakes in mixed mangrove forest of rhizophora and bruguiera species in matang, perak. malaysian journal geosciences, 1(2), 32-35. kozubal, j., & steshenko, d. (2015). the complex compaction method of an unstable loess substrate. arabian journal of geosciences, 8(8), 6189–6198 lei, x. (1987). the pore type chinese loess and collapsible. china science (series b chemical biology medical science, agronomy ), 12: 1309-1318. (in chinese) liu, z. (1997). loess mechanics and engineering. shaanxi science and technology press. (in chinese) ma, y., & wang, j. (2014). immersion tests on characteristics of deformation of self-weight collapsible loess under overburden pressure. chinese journal of geotechnical engineering, 36(3), 537-546 mustafaev, a. a. (1987). problems of the mechanics of collapsible soils and methods of solving them. soil mechanics and foundation engineering, 24(6), 213–218 people’s republic of china national standard for building construction in collapsible loess area gb 50025-2004. (2004). beijing: china architecture and building press. (in chinese) rabinovich, i. g., & urinov, m. i. (1974). characteristics of the development of collapse in loess soils with time. soil mechanics and foundation engineering, 11(5), 331–333 sarkar, m.i., islam, m.n., jahan, a., islam, a., & biswas, j.c. (2017). rice straw as a source of potassium for wetland rice cultivation. geology, ecology, and landscapes, 1(3), 184-189. sun, j., & liu, j. (2000). unsaturated collapsibility, residual collapsibility and multiple collapsibility of loess. journal of geotechnical engineering, 3, 365-367. (in chinese) tariq, w., hussain, s.q., nasir, a., tayyab, n., gillani, s.h., & rafiq, a. (2017). experimental study on strength and durability of cement and concrete by partial replacement of fine aggegate with fly ash. earth sciences pakistan, 1(2), 07-11. tunggolou, j., & payus, c. (2017). moringa oleifera as coagulant used in water purification process f or consumption. malaysian journal geosciences, 1(2), 29-31. usman, m., yasin, h., nasir, a., & mehmood, w. (2017). a case study of groundwater contamination due to open dumping of municipal solid waste in faisalabad, pakistan. earth sciences pakistan, 1(2), 12-13. wang, j. & gu, t. (2013). a mechanism for self-load collapsibilitysaturated loess liquefaction caused by earth micro-tremors. applied mechanics and materials, 405-408,541-547 yang, y. (1988). research of loess collapse mechanism. chinese science (b series chemical biology agronomy medicine science), 7, 756-766. (in chinese) yasin, m., khan, m.s., & khan, m.r. (2017). the modal analysis of rocks in the dwelling of poonch and sudhunhoti, azad jammu and kashmir, pakistan. pakistan journal of geology, 1(2), 07–15. yew, l.k., & rahim, i.a. (2017). prediction of rock mass properties, tunnel stability and support pressure by geological strength index (gsi) in crocker formation: a case study. geological behavior, 1(2), 31-33. zhang, l. (2000). discussion on the laboratory testing method to determine the collapsibility of loess. site investigation science & technology. zhang, s. (2000). terminology and basic concepts of collapsible loess. geotechnical engineering technology, 1, 42-46. (in chinese) zhang, s., & zhang, w. (1992). the collapsibility of loess in the process of reducing humidity and moistening. journal of geotechnical engineering, 1, 57-61. (in chinese) zhang, s., & zheng, j. (1990). wetting deformation characteristics of collapsible loess (q3). chinese journal of geotechnical engineering, 4, 21-31. (in chinese) zhang, w., & zhang, s. (1995). development of research on engineering properties of loess in china. chinese journal of geotechnical engineering, 6, 80-88. (in chinese) zhang, z. (1980). loess in china. geojournal, 4(6), 525–540 zhenghan, c., & zudian, l.(1986). mechanism of collapsible deformation of loess. chinese journal of geotechnical engineering, 2, 1-12. (in chinese) nota editorial hei comenzar las casas es tenerlas media acabadas" miguel de cervantes saavedra en el trasegar de las actividades cotidianas, con frecuencia olvidamos los esfuerzos que hemos llevado a cabo en la realizacion de las mismas, solo cuando volvemos la vista al camino andado y confrontamos los desafios que se ten ian y las metas conseguidas, percibimos la incertidumbre ante la empresa y las satisfacciones de los logros. henos un afio despues, nuevamente aqui, con la publicacion del septimo numero de la revista geofisica colombiana, en la cual, con gran teson y mediante el esfuerzo encomiable de investigadores latinoamericanos de las ciencias de la tierra, presentamos a la comunidad cientifica los resultados de investigaciones en los campos de la geofisica. coincide esta ocasion con los cinco afios de reapertura del programa de maestria en geofisica en la universidad nacional de colombia, en su sede bogota, labor que se ha visto fructificada con una nueva promocion de egresados de este programa de estudios avanzados, que con certeza va a redundar en el devenir de un mayor desarrollo de esta rama del saber en beneficio de la nacion colombiana. otras dos efernerides nos circundan; luctuosa la primera, el proximo 25 de enero se cumplen cinco afios del terremoto de armenia, even to que puso a cuestionar a la ciencia sobre el conocimiento de la corteza bajo los andes colombianos; de regocijo la otra, los diez afios de creacion de la red sisrnologica nacional, que ha suministrado un gran volumen de informacion valiosa en el entendimiento de la geodinamica colombiana. recibimos con beneplacito la propuesta de creacion de una carrera de estudios de pregrado en geofisica por parte de! departamento de geociencias de la universidad nacional de colombia, la cual sera bien recibida por parte de la industria y de los circulos intelectuales del pais, por su naturaleza innovadora ante los retos que asume frente a la sociedad y la comunidad cientifica. invitamos a nuevos iniciados en esta ciencia, para que se acerquen a la luz del saber y el conocimiento cienrifico, proyectando su interes al tenor de la geofisica, ciencia que, por su naturaleza aplicada, reconforta al entre gar sus frutos, y por su fundamentacion teorica henche de curiosidad la mente de inquietos jovenes, finalmente, extendemos la invitacion a toda la comunidad cientifica a participar en el primer congreso latinoamericano de sismologia que va a realizarse en la ciudad de armenia, tanto con la presentacion de sus alcances en investigaciones como con su grata presencia en esta acogedora ciudad; para ella incluimos en este numero la informacion pertinente de este acontecimiento. por el avance de la ciencia y en beneficio de todos, los esperamos. luis alfredo montes vides editor keywords: physic-chemical parameters, liners, residual soil, isotherms, batch test, sorption. palabras clave: parámetros fisicoquímicos, revestimientos, suelo residual, isotermas, prueba de lotes, sorción. how to cite item marques, j. p., ferreira da silva, e. a., patinha, c., kasemodel, m. c., rodrigues, v. g. s. (2019). adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil). earth sciences research journal, 23(4), 385-395. doi: https:// doi.org/10.15446/esrj.v23n4.77869 lateritic soils have been widely used in the construction of landfill base liners for solid waste disposal. on the other hand, there is little practice in the use of cambisols (less clayey) for this finality. the use of local soil for the construction of liners in order to waterproof and retain contaminants represents an alternative to control this type of contamination. thus, the objective of this study was to determine the physical and chemical properties of a cambisol collected in the city of eldorado (ribeira valley) and to evaluate the ability of this soil to adsorb lead (pb). the contamination of soil and water by pb is a worldwide problem, and in the ribeira valley the contamination by pb is common in specific areas mainly due to inadequate mining waste disposal. the efficiency of pb adsorption by the cambisol was indirectly proportional to the concentrations of pb in solution. when the initial concentration of 20 mg/l was used, the adsorption percentage of pb was 92.5% and when the initial concentration was 100 mg/l, the adsorption rate was 81.7%. in addition to the retention of pb, this soil presented the following characteristics which are favorable for using soils in the construction of liners: a silt loam (a high plasticity clay, ch), with liquid limit (ll) of 57%, plasticity index (pi) of 33%, oxidizing medium and the predominance of negative charges on the surface of the colloidal particles. thus, it is concluded that this tropical soil (cambisol) exhibits characteristics that allow it to be used as a liner in the disposal of residues containing pb. abstract adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil) adsorción de plomo (pb) en suelos tropicales meteorizados (región del valle de ribeira brasil) issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.77869 los suelos lateríticos se han utilizado ampliamente en la construcción de revestimientos de basureros para residuos sólidos. por otro lado, hay poca práctica en el uso de cambisols (menos arcillosos) para esta finalidad. el uso de suelo local para la construcción de revestimientos para impermeabilizar y retener contaminantes representa una alternativa para controlar este tipo de contaminación. por lo tanto, el objetivo de este estudio fue determinar las propiedades físicas y químicas de un suelo tropical fuertemente degradado recolectado en la ciudad de eldorado (valle de ribeira), y evaluar la capacidad de este suelo para adsorber plomo (pb). la contaminación del suelo y el agua por pb es un problema mundial, y, en el valle de ribeira, la contaminación por pb es común en áreas específicas, debido principalmente a la eliminación inadecuada de los desechos mineros. la eficiencia de la adsorción de pb por el cambisol fue indirectamente proporcional a las concentraciones de pb en la solución. cuando la concentración inicial de 20 mg/l, el porcentaje de adsorción de pb fue del 92,5% y cuando la concentración inicial fue de 100 mg/l, la tasa de adsorción fue del 81,7%. además de la retención de pb, este suelo presentó las siguientes características que son favorables para el uso de suelos en la construcción de revestimientos: es un suelo franco limoso (arcilla de alta plasticidad, ch), con límite de líquido (ll) del 57%, índice de plasticidad (pi) del 33%, medio oxidante y el predominio de las cargas negativas en la superficie de las partículas coloidales. por lo tanto, se concluye que el suelo residual tropical (cambisol) presenta características que le permiten ser utilizado como revestimiento en la eliminación de residuos que contienen pb. resumen record manuscript received: 14/02/2019 accepted for publication: 19/07/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 385-395 jéssica p. marques1, eduardo a. ferreira da silva2, carla patinha2, mariana c. kasemodel1, valéria g. s. rodrigues1* 1department of geotechnical engineering, são carlos school of engineering, university of são paulo, 400 trabalhador sãocarlense ave, são carlos, brazil 2department of geosciences, geobiotec – geobiosciences, geotechnologies, and geoengineering research center, campus de santiago aveiro university, aveiro, portugal * corresponding author: valguima@gmail.com so il s 386 jéssica p. marques, eduardo a. ferreira da silva, carla patinha, mariana c. kasemodel, valéria g. s. rodrigues introduction lead (pb) is a nonessential potentially toxic element (pte) that naturally occurs at low concentrations in the environment (adriano 1986; alloway 1995). however, the excess of pb is common in the soil due to anthropogenic activities. these activities are related to the use of this metal in industrial applications, mainly in the automotive industry, the improper disposal of mining and smelting wastes; air-borne sources; use of pesticides; biosolids and manure (alloway 1995; skerfving and bergdahl 2007; wuana and okieimen 2011). pb is also commonly found in urban landfill leachates as a result of the improper disposal of dangerous wastes. the contamination of the study region of this paper is associated with inadequate disposal of solid mining waste. the contamination by pb is a global concern due to its high toxicity and bioaccumulative properties. it can cause severe biological effects, increasing anemia, cancer and mortality rates (lustberg and silbergeld, 2002; papanikolaou et al., 2005; skerfving and bergdahl 2007; jin et al. 2008; hegazy et al. 2010). the joint fao/who expert committee on food additives (jecfa) had established a provisional tolerable weekly intake (ptwi) of 25 µg/kg by for all population groups. however, in 2010 the committee concluded that this value could no longer be considered health-protective, and the ptwi of pb was withdrawn (who 2011). there is no known safe blood level for pb, concentrations as low as 5 µg/dl may be associated with decreased intelligence in children, behavioral difficulties and, learning problems (who 2018). a recurrent problem related to pb contamination is the inadequate disposal of solid mining waste. pb is usually related to galena (lead ore pbs) in mining areas, and during the extraction and beneficiation of ores, residues such as slags and tailings are generated. their inadequate disposal can result in soil, water, and air contamination by pb. the characterization of the soil is extremely important to assess the contamination (level and extension). it is equally important to have research efforts aimed at characterizing local soils to be used as liners in the waste disposal area. this may reduce costs involving the transportation of the wastes and/or materials for the construction of landfills. in soils, the mobility and the fate of metals are regulated via their partitioning between soil and soil solution (freeze and cherry 1979; lee et al. 1998). metal mobility in the soil also depends on soil conditions and the chemical form of the metal (alloway 1995; rieuwerts 2007). several studies have addressed metal behavior in different soils, such as zuquette et al. (2008); korf (2011); braz et al. (2013); odukoya et al. (2013); diagboya et al. (2015), zia et al. (2018); gabarrón et al. (2018); colzato et al. (2018). other researchers specifically studied the use of tropical soils as liners for waste disposal (boscov et al. 1999; afolayan and nwaiwu 2005; knop et al. 2008; boscov et al. 2011; morandini and leite 2015; agbenyeku et al. 2016; miguel et al. 2017; onyelowe et al. 2019; ehrlich et al. 2019). pb in the soil is cumulative and traditional treatments for metal contamination are very expensive (usda 2000). treatments can be done in situ or ex-situ, these include high-temperature treatments, washing processes and the use of solidifying agents (usda 2000). for example, pump-and-treat is one of the most traditional techniques for remediation. but the high rate of energy required and the need for effluent disposal may increase the costs (usepa, 2014). thus, preventing metal contamination by using adequate liners is important to avoid the need for expensive remediation techniques. lateritic soils have been widely used in the construction of liners in landfill base (boscov et al. 2011; amadi and eberemu 2012; gabas et al. 2016). however, there is little practice in the use of cambisols as liners. there is increasing interest in these soils because of their availability in brazil, or in other countries. according to boscov (2011), it is important to expand the knowledge on the contaminant mobility behavior in tropical soils to encompass all the main types of soil of the country. researchers have addressed this issue, such as soares (2004), morandini and leite (2015), almeida (2015), colzato et al. (2018). cambisols are very common, and the use of local soils in geoenvironmental applications may reduce costs, which is essential, especially in developing countries (lavagnolo 2018; shubbar et al. 2019). in this context, it is fundamental to perform a detailed study of the soil, covering its physical and chemical characteristics as well as its adsorption capacity. the ability of soil to retain metals is directly proportional to its buffering capacity, and the retention mechanisms differ according to the ph values (freeze and cherry 1979; bourg and loch 1995). depending on environmental conditions (ph, redox potential eh and temperature) and soil composition, metals can be retained in the form of oxides, hydroxides, carbonates, or confined in exchangeable cations and organic matter (om) (tessier et al. 1979). the om content and the mineralogy of the soil can influence contaminant retention due to high cation exchange capacity (cec) (langmuir 1997). soil texture is also an important factor, as it can control the migration of contaminants through the soil profile. in soils with more clay content, soil texture favors the retention of metals. the retention capacity of pte by adsorption is directly related to the surface area. small particles in the clay fraction have higher adsorption capacity (langmuir 1997) due to their large surface area. adsorption is the accumulation of a particular element or substance at an interface. the phase that adsorbs is referred to as the adsorbent and the phase that accumulates at its surface is the adsorbate. the nature of the force involved in the adsorption process may vary, they may be of physical nature, such as van der waals and electrostatic, or chemical, such as reactions with electron sharing (weber 1972). from the point of view of soil and groundwater contamination, the adsorption is the soil capacity to attenuate contaminant mobility and decrease the possibilities of these elements entering the food chain (freeze and cherry 1979; sposito 1984; sparks 1995). according to alleoni et al. (2005), pte soil adsorption depends mainly on: (a) clay content, (b) om, (c) iron (fe), aluminum (al) and manganese (mn) oxides content, (d) ph, (e) specific surface (ss) and (f) soil mineralogy. braz et al. (2013) studied pb, cadmium (cd), cobalt (co), copper (cu), mercury (hg) and zinc (zn) adsorption by soils from the eastern amazon, and the results showed correlation between the adsorption capacity and ph, organic carbon content and fe and al oxides content. diagboya et al. (2015) also showed a significant role of soil organic matter and oxides content in adsorption of pb, cd, and cu by four tropical soils from nigeria. the electrostatic charge of soil particles is linked to the adsorption rate (zuquette et al. 2008). in tropical soils, these charges are highly variable (positive and negative) due to the mineralogical assembly and ph. for instance, gupta and bhattacharyya (2008) showed that the increase of ph led to increasing adsorption capacity of pb by natural kaolinite. according to mohamed et al. (2018), the ph effect in adsorption by soils is related to the changes in particle net proton charge. the electric charge of soil particles influences the adsorption of cations and anions, and the number of adsorbed ions depends on the charge density, related to the surface of the colloid, depending mainly on the type and content of mineral and organic constituents of soils (alleoni et al. 2016). for a given ph, the net charge resulting from all sources of the surface of a colloid can be null, corresponding to the point of zero charge (pzc). the point of zero salt effect (pzse) is the ph value when the difference between the number of moles of h+ and ohadsorbed is invariable with the salt concentration. if the charge densities of dissociated ions and adsorbed ions remain unchanged with the variation of the salt concentration, the pzc, and pzse values are equal. however, this condition rarely occurs in soils (sposito 1984; sparks 1995). the adsorption curve obtained in the batch equilibrium tests is useful to obtain preliminary information about the soil attenuation capacity (daniel 1993). the adsorption isotherm is the ratio between the mass of solute adsorbed and the concentration at equilibrium. the isotherm describes the relation of the solute concentration between two separate phases in equilibrium at a constant temperature (sparks 1995; fagundes and zuquette 2009). thus, the sorption isotherm expresses the relation between the amount of vapor or solute adsorbed as a function of vapor or solute equilibrium concentration, not considering the mechanisms. the isotherm method consists of stirring soil subsamples with solutions containing increasing amounts of the metal ion. the amount of ion adsorbed by the soil is the difference between the initial ion concentration and the remaining ion concentration in solution subsequent to the experiment (linhares et al. 2009). although the adsorption isotherm does not consider chemical mechanisms involved in adsorption, it allows describing the adsorption characteristics by the correlation between the parameters and soil properties (vinhal-freitas et al., 2010). the study area is located in the ribeira valley region, which presents three major geological domains: pre-cambrian metamorphic rocks (with oriented structures: schists, migmatite, and gneiss) and cataclastic (generated by shear stresses); magmatic rocks (represented by intrusive granite, basic and alkaline bodies) and cenozoic sedimentary cover (cetec, 1999). 387adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil) the soil ph in kcl solution (1.0 mol/l) was obtained with a digimed dh21 ph meter with a glass electrode. the relation between soil mass and volume of solution used was 1:2.5 (20 g of soil and 50 ml of kcl solution). the solution was stirred for 30 minutes and then left resting for 1 hour before the measurement (embrapa 2011). the ∆ph is obtained from the numerical difference between the ph kcl and the ph h20. the ph is an important parameter in the adsorption process. according to fageria et al. (1994), the soil acidity is described quantitatively as slightly acid (ph between 6.0 and 7.0), moderately acidic (ph between 5.5 to 6.0), strongly acidic (ph between 5.0 and 5.5), very acid (ph 4.5 to 5.0) and extremely acidic (ph less than 4.5). to determine the eh value, a mixture of 1:2.5 (20 g of soil and 50 ml of distilled water) was prepared. the solution was stirred for 30 minutes and then left resting for 1 hour. the eh was read using a platinum ring electrode connected to a ph meter micronal b374, reference electrode ag/clag. to determine the ec value, a mixture of 1:2 (10 g of soil and 20 ml of water) was prepared and manually shaken for 30 seconds. the mixture was left resting for 1 hour before measurements were made using hanna instruments hi 9033 conductivity meter. pzse pzse is a ph value where the salt concentration of the solution in a soil suspension has no charge effect on the surface of soil particles. according to silva et al. (1996) and coringa and weber (2008), the comparison between soil ph and pzse allows to define whether the net surface charge of the particles is positive (ph < pzse), negative (ph > pzse) or zero (ph = pzse). soil pzse depends on the content and degree of decomposition of organic matter and mineral composition of the clay fraction. to obtain the pzse, a potentiometric acid-base titration was performed, varying to ionic forces. hcl was used as acid, naoh as base and kcl as support electrolyte. in the first step, a series of solutions were prepared with increasing concentrations (0.002 m, 0.004 m, 0.008 m, and 0.016 m) of acid and base diluted in 0.1 m kcl, 0.01 m and 0.001 m solutions. next, suspensions were prepared with 4.0 g of soil and 20 ml of each solution. the suspensions were maintained at room temperature for 24 hours with occasional manual shaking and ph values measured using a digimed dh 21 ph meter. equilibrium ph values (on the ordinate axis) versus the amount of adsorbed h+ (for the acid solutions) or oh(for the basic solutions) were plotted. the pzc was obtained based on the equation proposed by keng and uehara (1973): pcz = 2 ph kcl – ph h2o. the surface potential (⌡o ) was obtained based on the equation proposed by raij and peech (1972): ⌡o = 59.1 (pzse – ph h2o). cec and ss the methylene blue method described by arab et al. (2015) was used in this study. this test is based on the linear relationship between the methylene blue index and fundamental soil properties such as cec and ss. initially 2.0 g of soil sample was dispersed in 10 ml of water, under constant agitation with a magnetic stirrer. the suspension was titrated with a methylene blue solution (1.5 g/l). a volume of 1 ml of methylene blue was titrated in the soil suspension; the suspension was left mixing for 3 minutes. then a droplet of the mixture was taken with a glass rod and placed on filter paper (whatman nº 42). the test is considered negative if there is no presence of a light blue halo surrounding the droplet, and then the procedure is repeated adding an extra 1 ml of the methylene blue solution to the mixture. when a clear light blue halo is formed around the blue drop, the test is considered positive. according to chen et al. (1974 apud pejon 1992)1, the cec is obtained by eq. 1 where v is the volume of methylene blue (ml) used in the titration, c is the concentration of the methylene blue solution (normal) and m is the dry soil mass (g). 1 chen, t. j., santos, p. s., ferreira, h. c., calil, s. f., zandonadi, a. r., & campos, l. v. (1974). determinação da capacidade de troca de cátions e da área específica de algumas argilas e caulins cerâmicos brasileiros pelo azul de metileno e sua correlação com algumas propriedades tecnológicas. cerâmica, 79, 305-326. the ribeira valley (ribeira de iguape water basin and cananéia-iguape estuarine complex) is a well-known region due to its historic pb exploitation. the improper disposal of pb slags in the soils and the upper ribeira de iguape river led to the contamination of many municipalities downstream, including the cananéia-iguape estuarine complex (bosso and enzweiler, 2002; guimarães and sígolo 2008; rodrigues et al., 2012; mahiques et al., 2013; piedade et al., 2014; kasemodel et al., 2016; tramonte et al., 2016). there is a necessity of proper landfill implantation for the mining waste. besides, landfills for urban wastes from the cities itariri and pedro de toledo (both cities are located near eldorado) were considered inadequate in the biennial report provided by the são paulo state environmental company (cetesb, 2017). this problem entails the need to undertake studies that address the ability of local soils to adsorb pb. the main objective of this research was (a) to determine the physical and chemical properties of a tropical soil (cambisol) collected in the city of eldorado (ribeira valley), and (b) to evaluate the adsorption capacity of pb by this soil. materials and methods approximately 10 kg of soil was collected in the city of eldorado (geographic coordinates: 24º31’12” s and 48º06’29” w), ribeira valley region, brazil. the sample studied is a representative one, composed of the collection at 3 different points in the area. the disturbed samples, which were collected from b horizon, were air-dried at room temperature until constant weight, disintegrated with agate mortar and homogenized. tests were performed in duplicates to determine the physical and chemical parameters of soil and the adsorption rates (batch equilibrium test) with pb solution. particle size distribution special attention was given to the soil fine fraction given that the purpose of the study was evaluating pb adsorption. a micromeritics sedigraph 5100 equipment carried out the particle size analysis, for particles up to 100 μm. the analysis was performed based on the iso 13317-3/2001 method (iso 2001). soil (3.5 g) was mixed with distilled water to humidify with sodium hexametaphosphate shmp (1%, 50 ml) as a dispersant. the suspension was placed in sedigraph and the sieved fractions were dried in an oven at 50 °c and weighed. the particle size and the granulometric fractions are estimated based on the stokes law and the measurement was carried out by x-ray absorption (lima and luz, 2001). the determination was also carried out adding only distilled water to the sample. atterberg limits the determination of the ll was performed according to abnt mb30 (1984a) using the casagrande apparatus. the determination of the plastic limit (pl) was performed according to abnt mb31 (1984b). the plasticity index (pi), an important parameter used to evaluate the soil in the retention of contaminants, is the numerical difference between the ll and the lp (pi = ll – pl). the activity index (ai) was calculated from the ratio between the pi and the clay content. this relation indicates the activity of clay in the soil. if the value of ai is less than 0.75, the soil is considered inactive; if the ai value is integrally comprised between 0.75 and 1.25, the soil is normally active and; if the ai value exceeds 1.25, the soil clay is considered active (skempton 1953). physico-chemical parameters of the soil the physical-chemical parameters determined were: (i) ph in water, in cacl2 and in kcl solution; (ii) eh; (iii) electrical conductivity (ec); (iv) point of zero salt effect (pzse). ph, eh and ec the ph was determined both in water and cacl2 solutions (0.01 mol/l) based on the iso 103090/2005 (iso 2005) method. the relation between soil mass and volume of solution used was 1:5 (5 g of soil and 25 ml of water or cacl2 solution). the suspensions were stirred for 5 minutes and then left resting for 24 hours. the measurement was conducted with a glass electrode and ph meter hanna hi 9025. 388 jéssica p. marques, eduardo a. ferreira da silva, carla patinha, mariana c. kasemodel, valéria g. s. rodrigues cec vx cx m = 100 (1) the ss values were empirically estimated using cec results obtained by the pejon (1992) method. according to beaulieu (1979 apud pejon, 1992)2 the specific surface is obtained using eq. 2, where a is the area of the methylene blue molecule (130 å2), n is the avogadro number (6.02 x 1023), m is the mass of dry methylene blue and mw is the molecular weight of methylene blue in anhydrous form. eq. 4 is obtained by substituting the values. ss ax nx m mw x = 1000 (2) ss x x x x x x m ml = ( ) ( ) = −( . . . . 130 10 6 02 10 1 5 319 9 1000 3 67 20 23 2 (3) ss x v m =     3 67. (4) a second method in accordance with iso standard 13536/1995 (iso 1995), and adapted from bascomb (1964), was used to estimate the cec values. the method consists of saturating the soil with bacl2 solution buffered to ph 8.1. exchangeable cations in soils replaced by ba were analyzed by atomic absorption. in order to precipitate barium adsorbed in the soil and the free cations in solution, mgso4 solution was added to the soil sample. magnesium occupies the exchange sites on the soil and the excess of mg (not absorbed) is determined by titration with edta and eriochrome black t indicator. the cec is calculated according to the following formula: cec msoil n m b a=     −     + −( )          1 250 10 25. . . (5) where cec is the cation exchange capacity in cmolc/kg; msoil is the weight in grams of soil used in the test; n is the volume in ml of edta used in the titration; m is the volume in ml of edta used in the blank titration; and b a is the amount in grams of water in the soil sample. the ss values of particles are related to the soil’s ability to retain and/ or release soil chemical elements. higher ss implies greater possibility of interaction with soil elements in solution, which can facilitate the retention process and attenuation of contaminants. om om content was obtained using the method described by eusterhues et al. (2005). this method is conducted using a mixture 5 g of soil sample and 50 ml of hydrogen peroxide that is heated to 90°c. after the peroxide attack, the soil was oven dried at 60°c. the determination of the om content is based on the mass difference of the sample before and after the procedure. pb concentrations icp-ms, analytical technique with high sensitivity was used to estimate pb concentrations in soil samples. the analysis was performed after acid decompositions (hno3, hclo4, and hf) by icp -ms at acme laboratory in canada. mineralogical analysis x-ray diffraction (xrd) xrd is a non-destructive technique used to identify the crystalline phases, which has been widely used in the identification of clays (tan et al. 1986; albers et al. 2002). for the total sample analysis, approximately 1 2 beaulieu, j. identification géptechnique de matériaux argileux naturels par la mesure de leur surface au moyeu du bleu de méthylène. thèse de doctorat de 3° cycle, univ. de paris-sud, orsay, 133 p. g of soil was separated and gently broken down with the agate mortar. the sample was placed in an aluminum sample holder of the x-ray diffractometer, slightly compressing the powder to avoid a preferred crystal orientation. the diffractograms were obtained using the diffractometer philips x’pert consisting of a mppc generator, pw3050 goniometer pw 3040/60 with a microprocessor in the printer registration. the radiation used was copper with a wavelength of 1.5405 å. the diffraction angle 2θ ranged from 4° to 65° at a rate of 0.02°/s. for the fraction analysis diameter less than 0.063 mm the clay thin-sections were prepared using a solution with about 1 g of soil sample, 30 ml of distilled water and two drops of sodium hexametaphosphate (1%) were added as dispersant. the mixture was stirred for about 40 seconds with ultrasonic stirrer. 1 ml of the solution was pipetted onto three glass slides; the droplet was left to dry at room temperature. one of the slides was subjected to treatment with ethylene glycol, another was heated to 500°c, and the third received no treatment. the three slides with different treatments were analyzed in the same diffractogram. differential thermal analysis (dta) and thermogravimetric analysis (tga) this analysis was performed heating the samples to 1000°c at a rate of 12.5°c per minute. in the dta a sample of the material to be studied was heated side by side with a sample of a thermally inert material (alumina) and the temperature difference between the two samples was recorded as the samples were progressively heated (mackenzie 1957). the identification of minerals is by shape, position and intensity of the peaks of the curve produced in atd. the obtained curve was compared with standard curves of various minerals (santos 1989). in atg, the sample weight was recorded as a function of temperature, while it was heated at a controlled rate. changes in weight were assigned to rupture or formation of chemical or physical connections corresponding to the release of volatile substances or formation of compounds (tan et al. 1986). the interpretation of the curves was made based on mackenzie (1957) and santos (1989). batch equilibrium tests the study of adsorption is necessary to understand soil’s ability to attenuate, degrade or immobilize contaminants of organic and/or inorganic origin. in the batch test method, soil and solutions containing the ion of interest are brought into contact, under agitation and during a predetermined period. at the end of the experiment, it is admissible that the system reached the equilibrium of the reaction (soares and casagrande, 2000). in order to determine the sorption characteristics of the soil, batch tests were performed using solutions of five different pbcl2 concentrations: 20, 40, 60, 80, and 100 mg/l. this range was chosen based on pb concentrations in leachate of mining tailings and slag from adrianópolis (ribeira valley). these concentrations reached 77.10 mg/l (raimondi 2014; kasemodel 2017). for this test, we followed the methodology proposed by roy et al. (1992) and zuquette et al. (2008). the soil/solution ratio used was 1:5 (20 g of soil and 100 ml of solution). samples were shaken in a nt 155 novatecnica (between 25 to 30 rpm) benchtop orbital shaker. after 48 hours, the samples were centrifuged and filtered through whatman filter paper nº 42. during the test, room temperature was controlled, and physicochemical parameters were measured. the ph, eh and ec were measured before the test (solutions without soil sample), after soil and solution contact, and in the final stage, after equilibrium. for the data interpretation the authors used the linear, freundlich and langmuir isotherms. the linear isotherm corresponds to the simpler model of adsorption, and estimates the solute adsorption of infinitely increasing concentrations (fagundes and zuquette 2009). the linear isotherm model is represented by eq. 6: s k xcd e= �� (6) where: s is the mass of contaminant adsorbed per unit of soil mass (µg/g), ce is the final concentration of the equilibrium solution (mg/l), kd is the distribution coefficient of linear isotherm (cm3/g). the langmuir method, represented by the eq. 7, uses two different linearization processes to obtain the parameters of this isotherm. the linearization shown in eq. 8 and eq. 9 are called langmuir linearization i and 389adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil) langmuir linearization ii, respectively. in these equations: sm is the maximum amount of solute adsorbed by the soil (µg/g) and b is a constant related to the adsorption binding energy (cm3/g). s s x b x c b x c m e e = + ( ) 1 (7) c s s xb c s e m e m =     +     1 �� � (8) 1 1 1 s s xbx c sm e m =     +    �� �� � (9) the langmuir equation is often preferred since its parameters allow estimating the maximum adsorption capacity and adsorbate affinity of the adsorbate (linhares et al. 2009). the freundlich isotherm is represented by eq. 10 and linearization was used to obtain the parameters by eq. 11. in these equations, s is the mass of contaminant adsorbed per unit of soil mass (µg/g), ce is the final concentration of the equilibrium solution (mg/l), n (dimensionless) and kf (cm 3/g) are coefficients of the freundlich equation. s k xcf e n= �� � (10) log s n xlog c log ke f� � �� � � � �( ) = ( ) + ( ) (11) results and discussion physico-chemical and mineralogical properties table 1 presents the results of particle size distribution. as expected the experiment with and without sodium hexametaphosphate (shmp) presented different results. for example, the clay fraction obtained in the soil when applying shmp was 26.12% (table 1), while the percentage of clay was lower (20.19%) without smhp. silt fraction was 52.47% with shmp, and 63.09% without shmp. the results of the test with dispersant show higher clay content and lower silt content. this is explained by flocculation of small clay particles when there was no dispersant. table 1. particle size distribution of the studied soil in tests with and without dispersant. diameter distribution with shmp distribution without shmp µm (%) (%) 0 – 2 (clay) 26.12 20.19 2 – 60 (silt) 52.47 63.09 60 100 0.79 1.18 > 100 20.62 15.54 according to rowe et al. (1995), a clayey compacted soil can be used as a liner in a waste disposal area, if it presents 15 to 20% of clay fraction in its composition. according to the results (table 1), this soil is suitable to be used in this application. qian et al. (2002), indicates that a soil must to present at least 50% fines (silt and clay fractions) to be applied in a liner, a requirement that the analyzed residual soil also satisfies. the importance of fine particles in soils, used in the construction of liners, is to ensure lower hydraulic conductivity. furthermore, particle size and sorption capacity are also related. according to langmuir (1997), the sorptive abilities of minerals are proportional to their surface areas and surface-site densities. the bigger the proportion of fine particles, the larger the surface area exposed to reaction with the solution. table 2 summarizes the results of the physical, physico-chemical and mineralogical characteristics of the soil. table 2. physical, physico-chemical and mineralogical characteristics of the soil. parameters result pl (%) 24 ll (%) 57 pi (%) 33 ph h2o 4.5 ph cacl2 4.0 ph kcl 4.2 ∆ph -0.3 pzse 3.4 pzc 3.9 surface potential (⌡o) ⌡o = 59.1 (pzse – ph h20) (mv) -65.01 ec (µs/cm) 41.55 cec (cmolc/kg) method: methylene blue 3.62 cec (cmolc/kg) method: saturation with ammonium acetate 3.20 ss (m2/g) 28.29 om (g/kg) 13.40 mineralogy k, qz, gth, i, m k: kaolinite; qz: quartz; gth: goethite, i: illite; m: muscovite soil ll was 57% and pl was 24%, therefore, the pi is 33%. this pi value is not within the ideal range for soil liners. daniel (1993) recommends a minimum pi of 7% for soil liners to achieve the hydraulic conductivity required. however, the author also recommends special attention for soils with high pi. according to daniel (1993) and rowe et al. (1995), soils with pi > 30%, such as the studied soil, may form hard clods that render difficult the construction of the liner, and may be susceptible to shrinkage and cracking if dry. ai was 0.61 when using clay content without shmp, and 0.79 when considering clay content with shmp. thus, the classification of the ai is dependent on the methodology used to obtain clay content, as when determining clay content without the use of the dispersant, ai is considered inactive; and when determining clay content with the use of shmp, ai is considered normal. this is a positive aspect for liners because high activity clay soils are susceptible to shrinking and swelling, causing problems in the liner structure (rowe et al. 1995). the ph measured in water (ph 4.5) was higher than in cacl2 and kcl solutions (4.0 and 4.2, respectively). in general and according to the results soil is very acidic. this may adversely affect the contaminant attenuation. some metals, including pb, remain soluble in acidic environments. in addition, ph influences the magnitude of the soil surface charge, which is also important for metal adsorption (yong 2001). the investigated soil pzse (3.4) is lower than ph value in water (4.5), and δph was negative (-0.3), indicating that the net surface charge of the particles is negative. the surface potential (⌡o) obtained in this study was also negative (-63.24 mv). similar values were obtained by silva et al. (1996) in different soils collected from several regions of brazil, and by braz et al. (2013), in soils from eastern amazon. the electric charge of soil particles influences the adsorption of cations and anions, and the quantity of ions adsorbed depends on the charge density, which depends on the colloid surface (alleoni et al. 2016). according to sposito (1984), solid surfaces of soil particles and/or reactive minerals develop electric charge in two ways: (a) isomorphic substitutions between ions of different valences in minerals, or (b) from reactions of surface 390 jéssica p. marques, eduardo a. ferreira da silva, carla patinha, mariana c. kasemodel, valéria g. s. rodrigues functional groups with ions present in the solution. these charges can be positive, null, or negative. according to langmuir (1997), a sorbent phase endowed with a net negative surface charge exhibits a cec, while a phase containing a net positive surface charge exhibits an anion exchange capacity. cec is a very important property that affects the retention of organic and inorganic species (sposito 1984). the cec of the total soil sample obtained using the methylene blue method was 3.62 cmolc/kg, while cec obtained by nh4ac saturation method was 3.20 cmolc/kg. it should be noted that these cec values are related to the total soil sample. if we consider that all that capacity is solely from the clay fraction (26% of the total soil), we can estimate the actual clay cec dividing the total cec by 0.26. this calculation results in the value of 13.8 cmolc/kg for the methylene blue method, and 17.9 cmolc/kg for the nh4ac saturation method. in this context, we can estimate that the sample may have kaolinite, and a certain amount of illite as confirmed in the mineralogical analysis (table 2). according to langmuir (1997), cec values related to kaolinite range from 3 to 15 cmolc/kg, which is strongly ph dependent, while for illite it ranges from 10 to 40 cmolc/kg, which is slightly ph dependent. according to rieuwerts (2007) and coringa and weber (2008), well drained soils of humid tropical regions, such as those studied in this work, are intensely weathered and dominated by minerals with variable load, as previously proven, which is responsible for low cec values, which could vary with ph. the ss result of the residual soil was 28.29 m2/g. this result confirms the presence of kaolinite in the analyzed soil. kaolinite ss ranges from 10 to 38 m2/g (langmuir 1997). according to sposito (1984), kaolinite ss ranges from 7 to 30 m2/g. the sorptive abilities of minerals are proportional to their surface areas and surface-site densities, which can range widely, reflecting differences in how they are measured (langmuir, 1997). the om content of the residual soil was 13.40 g/kg, which is low. according to rieuwerts (2007) tropical soils are relatively deficient in om as a result of its rapid decomposition. but even these low concentrations of om may be important for the soil cec, mainly for soils dominated by 1:1 clays. humic substances in om can be highly reactive. om can provide a higher cec (150 to 300 cmolc/kg, according to sparks (1995) in comparison with some minerals. according to alleoni et al. (2016), the presence of hematite, goethite and gibbsite explains the higher pzc values, and higher levels of om and kaolinite are responsible for lower pzc values. the pzc value of the analyzed soil was 3.9, reflecting the presence of kaolinite. the mineralogical analysis indicated the presence of non-expansive clay minerals in the soil, such as kaolinite and illite (table 2). according to silva et al. (1996), the presence of kaolinite content in the soil explains the pzse value. this mineral tends to contribute in lowering the average soil pzse, due to its low isoelectric point (between 4.0 and 5.0). according to langmuir (1997), kaolinite is common in humid-climate, well-leached acid soils, and stream sediments where there is good drainage. kaolinite results from weathering of the k-feldspar (microcline and orthoclase) and muscovite in rocks such as granite. kaolinite is a low activity and nonexpansive clay mineral (mohamed et al. 2018). illite clays may result from the weathering of micas and feldspars. in addition to the non-expansive clay minerals, the analyzed soil is composed of goethite, quartz and muscovite. according to rieuwerts (2007), the intense weathering and leaching observed in the tropics results in desilication, i.e. “the leaching of si from silicates and the enrichment of primary quartz, secondary al and fe oxides, and two-layer clay minerals (mainly kaolinite) relative to all other soil constituents”, as observed in the analyzed soil. according to araujo et al. (2002), kaolinite and fe and al oxides are important components in the pte retention. many authors studying the characteristics of tropical humid weathered soils reported the establishment of relations between the electrochemical, chemical and mineralogical attributes, which facilitates understanding the soil ion adsorption phenomena (coringa and weber 2008). in the batch test, it is essential to control the physicochemical parameters before and after the test. thus, electrical conductivity (ec), redox potential (eh) and ph measurements were measured before and after the batch test (fig. 1). the ph values of the soil in contact with contaminant solutions were within the range of 4.0 and 4.5. results show little ph variation after the contact of pbcl2 solution with the residual soil. in this acidic environment pb solubility is favored. the eh of the samples ranged from 320 to 400 mv at the initial stage, from 310 to 440 mv at the contact, and from 370 to 405 mv at the final stage. these positive values show an oxidizing environment, which favors adsorption. pb solubility is lower in oxidizing than in reducing environments. this is probably explained by the dissolution of al, fe and mn oxides in reducing environments. these oxides play an important role in metal adsorption (chuan et al. 1995). significant variation of eh during the test was observed for the samples with the solution of 20 mg/l and 60 mg/l pbcl2. after the contact between the 20 mg/l solution and the soil, the eh increased 92 mv after 48 h. when using 60 mg/l pbcl2 solution, eh decreased 49 mv after 48 h of contact with the residual soil. other tested concentrations had minor eh variations (fig. 1). figure 1. values of ph, redox potential (eh) and electrical conductivity (ec) of the solution used in the batch test, in the initial stage (t = 0 s) and final stage (t = 48 h) of the test. 391adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil) ec decreased after 48 hours of soil solution contact for all tested concentrations, indicating a reduction in the concentration of salt solution due to the adsorption. a considerable ec decrease (approximately 40% of the initial ec value) occurs using 60 mg/l of pbcl2 solution, while smaller variations were observed using 80 and 100 mg/l of pbcl2 solution. batch tests icp/ms results show low concentrations of cd (0.08 mg/kg), pb (16.5 mg/kg) and zn (19.4 mg/kg) in the studied residual soil. these concentrations are similar to results available in the literature (wedepohl 1995) for soils collected in uncontaminated areas (0.10 mg/kg cd; 17 mg/kg pb, 52 mg/kg zn). the results are also comparable to the background concentrations of pte found by nogueira et al. (2018) for natural soils from são paulo state, brazil (0.1 mg/ kg cd; 10.1 mg/kg pb, 21.8 mg/kg zn). table 3 and fig. 2 shows the percentage of pbcl2 adsorbed by the soil. the percentage of adsorbed pbcl2 decreased when the concentration entry was higher, which was also evidenced by ec measurements (fig. 1). table 3. data obtained during batch test: pb initial concentration (c0 ); pb concentration at the end of the experiment (ce ); difference between the initial and equilibrium concentration (δc ); mass of pb adsorbed by mass of soil (s). co (mg/l) ce (mg/l) ∆c (mg/l) s (µg/g) pb adsorbed (%) 20 1.50 18.50 92.17 92.5 40 3.82 36.18 181.35 90.5 60 7.74 52.26 260.78 87.1 80 12.20 67.80 339.34 84.8 100 18.26 81.74 407.88 81.7 as shown in table 3, the adsorption capacity s of the soil increases with increasing solution concentration (c0 ). for the solution of 20 mg/l, soil adsorbed 92.17 μg/g. for the highest concentration (100 mg/l), s was 407.88 μg/g. it should be noted that the increase of the capacity s does not occur at the same rate as the increase in concentration, i. e. the isotherm is not perfectly linear. such behavior is typical of dissolved species in soils. at low pb concentrations, the affinity of the metal with the solid surface of the soil is higher. at increasing concentrations, as the adsorption sites are occupied, this affinity decreases. consequently, the highest percentages of pb adsorption were obtained when the lowest initial concentrations were used (langmuir 1997). based on table 3, using the initial concentration 20 mg/l, the percentage of lead adsorbed by soil was 92.5%. for the solution with 100 mg/l of pb, the adsorption percentage is comparatively smaller (81.7%). the figure 3 presents the langmuir i and ii, freundlich and linear isotherms. table 4 shows the data obtained when using langmuir i and ii, freundlich and linear isotherms. based on the correlation coefficients obtained from linear, langmuir and freundlich equations, all models adequately describe the experimental figure 2. adsorbed concentration of the pb by tropical soil. figure 3. linearization graphs of the linear, freundlich and langmuir (i and ii) adsorption isotherms. 392 jéssica p. marques, eduardo a. ferreira da silva, carla patinha, mariana c. kasemodel, valéria g. s. rodrigues pb adsorption results. the langmuir ii (r2 = 0.9982) and freundlich (r2 = 0.9927) models had the best adjustments (table 4). the linear equation had the lowest adjustment compared to freundlich and langmuir i and ii. similar results were obtained by linhares et al. (2009) studying the pb adsorption in highly weathered tropical soils collected in the region of minas gerais (brazil). for these authors the adsorption was best represented by langmuir (0.98 < r2 < 0.99) and freundlich (0.94 < r2 < 0.99) equations. freundlich equations had also provided better adjustments for pb adsorption by 4 soils from goiás state (0.96 < r2 < 0.99) (oliveira et al., 2010) and by a transported soil from ribeira valley (r2 = 0,98) studied by araujo (2015). paranavithana et al. (2016), observed that the langmiur isotherm presented the best fit for pb2+ and cd2+ in coconut shell biochar (fine granules), in a local soil from sri lanka, and in a biochar-mixed soil (1:1 mixture of biochar and soil). the parameter b (langmuir), which is related with the adsorption energy, was 0.1149 cm3/g for langmuir i and 0.1407 cm3/g for langmuir ii (table 4) models. according to linhares et al. (2009) the pb is an easy adsorption cation. according to pierangeli et al. (2001), this element is retained initially by negative charges of the surfaces of soil colloids, later forming more stable bonds. in the latter case, according to these authors, pb becomes part of the surface of colloids and available to plants in limited quantities. the pb interactions with soil colloids are predominantly more specific and less dependent on surface charges. the specific adsorption involves the exchange of metals with ligands present on the surface of the colloids, which forms covalent bonds and is assumed as the cause for which some soils adsorb metals above their predicted capacity based on cec (linhares et al. 2009). in soils with variable loads, it is difficult to distinguish the contribution of each of its components in the adsorption of metallic cations. even when regarding pure minerals, the adsorption capacity of a certain oxide or clay mineral depends on its crystallinity degree, which is extremely variable with the age and environmental conditions of its formation (pierangeli et al. 2001). the maximum adsorption capacity of the soil is 588.2353 µg/g. this value is lower than the capacity of transported soils from the same region. rodriguez (2013) and araujo (2015) studied pb adsorption by 3 transported soils from ribeira valley, using batch equilibrium tests (with soil-solution ratio 1/5). the authors reported maximum adsorption capacities varying between 893.00 and 1111.11 mg/l. even so, the soil studied in this research proved to be efficient to adsorb pb in solutions with concentrations close to those found in leachate from mining waste. kasemodel (2017) studied soil and slag samples from a waste disposal area in adrianópolis. the pb concentrations found in the leach test extracts ranged from 5.16 to 77.1 mg/l. raimondi (2014) studied mining tailings also from adrianopolis and the leach extracts showed pb concentrations varying from 13.5 to 77.1 mg/l. for this range of concentrations in batch tests, the soil presented adsorption efficiency higher than 84.8% (for the c0 = 80 mg/l). conclusions the tropical soil (cambisol) presented some favorable physical and chemical properties, and others less suitable, to be used in the construction of liner in the waste disposal area. the soil is fine and its granulometry is compatible with that recommended in the literature for use in liners of compacted clay. the clay fraction is composed of 1:1 clay minerals such as low activity, nonexpansive kaolinite. the oxidizing medium and the predominance of negative charges on the surface of the colloidal particles (evidenced by delta δph = 0.3 and pzse = 3.4) favor the retention of metallic cations. on the other hand, some characteristics of this tropical soil are less favorable, when observed separately, for the retention of contaminants. this is the case of low cec, which may result from the presence of kaolinite. low om content and acidic ph also discourage contaminant retention. in addition, the pi of 33% is in the limit of the recommended range, thus, the material can be used as a liner. as for adsorption, it was noted that the highest percentages of pb adsorption were obtained at the lowest initial concentrations. thus, the adsorption percentage of pb when using initial concentration of 20 mg/l was 92.5% and, when the initial concentration was 40 mg/l adsorption rate was 90.5%. for higher pb concentration (100 mg/l) the adsorption percentage is comparatively smaller (81.7%). based on the correlation the langmuir ii (r2 = 0.9982) and freundlich (r2 = 0.9927) models had the best adjustments. the results indicate that the maximum adsorption smax (langmuir) was 588.23 µg/g for langmuir i and 526.31 µg/g for langmuir ii models. thus, it is concluded that the tropical residual soil presents characteristics that allows it to be used as liner in the disposal of residues containing pb. but we recommend the study of possible additives to the soil to improve the plasticity and ion exchange characteristics. acknowledgements the são paulo research foundation (fapesp) for granting a scholarship to the first author (process number 2015/02529-4) and for provided financial support (process number 2014/07180-7) and the national council for scientific and technological development (cnpq) (for productivity in research fellowship process number 305096/2015-0) are gratefully acknowledged. references associação brasileira de normas técnicas – abnt (brazilian association of technical standards) (1984a). brazilian standard nbr 6459. determination of liquid limit. rio de janeiro, brazil. associação brasileira de normas técnicas – abnt (brazilian association of technical standards) (1984b). brazilian standard nbr7180. determination of plasticity limit. rio de janeiro, brazil. adriano, d. c. (1986). trace elements in the terrestrial environments. springer-verlang, new york, 533 pp. afolayan, j. o. & nwaiwu, c. m. o. (2005). reliability-based assessment of compacted lateritic soil liners. computers and geotechnics, 32(7), 505-519. https://doi.org/10.1016/j.compgeo.2005.08.001. agbenyeku, e. e., muzenda, e. & msibi, m. i. (2016). chemical alterations in three clayey soils from percolation and interaction with acid mine drainage (amd). south african journal of chemical engineering, 21, 28-36. https://dx.doi.org/10.1016/j.sajce.2016.04.003. albers, a. p. f.; melchiades, f. g.; machado, r.; baldo, j. b. & boschi, a. o. (2002). a simple method for the characterization of clay minerals by x-ray diffraction. cerâmica, 48(305), 34-37. https://dx.doi. org/10.1590/s0366-69132002000100008. alleoni, l. r. f., iglesias, c. s. m., mello, s. c., de camargo, o. a., casagrande, j. c. & lavorenti, n. a. (2005). atributos do solo relacionados à adsorção de cádmio e cobre em solos tropicais. acta scientiarum agronomy, 27(4), 729-737. https://dx.doi.org/10.4025/ actasciagron.v27i4.1348. alleoni, l. r. f., mello, j, w. v. & rocha, w. s. d. (2016). eletroquímica, adsorção e troca iônica no solo. in: melo, v. f. & alleoni, l. r. f. química e mineralogia do solo parte ii – aplicações. sociedade brasileira de ciência do solo, viçosa, brasil, 1380pp. table 4. adsorption isotherms parameters: distribution coefficient of linear isotherm (kd); langmuir maximum adsorption capacity (smax); langmuir adsorption constant related to binding energy (b); freundlich equilibrium constant (kf); freundlich exponential coefficient (n); correlation coefficient of the isotherms (r 2). linear langmuir i langmuir ii freundlich r2 kd r 2 smax b r 2 smax b r 2 kf n cm3/g µg/g cm3/g µg/g cm3/g cm3/g 0.9576 18.216 0.9895 588.2353 0.1149 0.9982 526.3158 0.1407 0.9927 76.3308 0.5925 393adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil) alloway, b. j. (1995). heavy metals in soils. second edition, blackie academic & professional, netherlands, 368 pp. almeida, a. f. (2015). evaluation of clay materials from corumbataí formation to use in compressed liners (ccl). ms dissertation, são carlos school of engineering, university of são paulo, são carlos, brazil. amadi, a. a. & eberemu, a. o. (2012). delineation of compaction criteria for acceptable hydraulic conductivity of lateritic soil-bentonite mixtures designed as landfill liners. environmental earth sciences, 67, 9991006. https://dx.doi.org/10.1007/s12665-012-1544-z. arab, p. b.; araújo, t. p. & pejon, o. j. (2015). identification of clay minerals in mixtures subjected to differential thermal and thermogravimetry analyses and methylene blue adsorption tests. apllied clay science, 114, 133-140. http://dx.doi.org/10.1016/j.clay.2015.05.020. araújo, h. a. s. (2015). estudo da adsorção de pb, zn e cd por material inconsolidado transportado da região do vale do ribeira (sp). graduate. são carlos school of engineering, university of são paulo, são carlos, brazil. araújo, w. a. sobrinho, n. m. b. a.; mazur, n. & gomes, p. c. (2002). relação entre adsorção de metais pesados e atributos químicos e físicos de classes de solo no brasil. revista brasileira de ciência do solo, 26(1), 17-27. http://dx.doi.org/10.1590/s0100-06832002000100003. bascomb, c. l. (1964). rapid method for the determination of cation-exchange capacity of calcareous and non-calcareous soils. journal of the science food and agriculture, 15(12), 821-823. https://doi.org/10.1002/ jsfa.2740151201. boscov, m. e. g., oliveira, e., ghilard, m. p. & silva, m. m. (1999). difusão de metais através de uma argila laterítica compactada, iv congresso brasileiro de geotecnia ambiental. regeo, são josé dos campos, brazil, 323-330. boscov, m. e. g., hachich, w. c., mahler, c. f. & oliveira, e. (2011). properties of a lateritic soil for pollutant containment. journal of environmental protection, 2, 923-931. https://doi.org/10.4236/jep.2011.27105. boscov, m. e. g. (2011). estudo de materiais locais para utilização nas obras geoambientais. revista fundações & obras geotécnicas, são paulo, 1(5), 36-41. bosso, s. t. & enzweiller, j. (2002). evaluation of heavy metal removal from aqueous solution onto scolecite. water research, 36, 4795-4800. https://doi.org/10.1016/s0043-1354(02)00208-7. bourg, a. c. m. & loch, j. p. g. (1995). mobilization of heavy metals as affected by ph and redox conditions. in: salomons, w. & stigliani, w. m. (eds). biogeodynamics of pollutants in soils and sediments. springer-verlag, new york. 87-102. braz, a. m. s., fernandes, a. r., ferreira, j. r. & alleoni, l. r. (2013). distribution coefficients of potentially toxic elements in soils from the eastern amazon. environmental science and pollution research, 20(10), 7231-42. https://doi.org/10.1007/s11356-013-1723-9. centro tecnológico da fundação paulista de tecnologia e educação – cetec. (1999). situação dos recursos hídricos do ribeira de iguape e litoral sul, são paulo, brazil. companhia ambiental do estado de são paulo – cetesb. (2017). inventário estadual de resíduos sólidos urbanos, série relatórios, são paulo, brazil. https://solo.cetesb.sp.gov.br/solo/publicacoes-e-relatorios/ (last accessed july 2019). chalermyanont, t.; arrykul, s. & charoenthaisong, n. (2008). potential use of lateritic and marine clay soils as landfill liners to retain heavy metals. waste management, 29(1), 117-127. https://doi.org/10.1016/j. wasman.2008.03.010. chuan, m. c., shu, g. y & liu, j. c. (1995). solubility of heavy metals in a contaminated soil: effects of redox potential and ph. water, air, and soil pollution, 90, 543-556. https://dx.doi.org/10.1007/bf00282668. colzato, m., alleoni, l. r. f. & kamogawa, m. y. (2018). cadmium sorption and extractability in tropical soils with variable charge. environmental monitoring and assessment, 190, 345-355. https://doi.org/10.1007/ s10661-018-6666-7. coringa, e. a. o. & weber, o. l. s. (2008). ponto de efeito salino nulo de latossolos da microbacia chico nunes, mato grosso. revista brasileira de ciência do solo, 32(1), 441-448. https://doi.org/ 10.1590/s010006832008000100042. daniel, d. e. (1993). geothecnical practice for waste disposal. chapman & hall: londres, 683 pp. diagboya, p. n., olu-owolabi, b. i., & adebowale, k. o. (2015). effects of time, soil organic matter, and iron oxides on the relative retention and redistribution of lead, cadmium, and copper on soils. environmental science and pollution research, 22(13), 10331-10339. https://doi.org/ 10.1007/s11356-015-4241-0. ehrlich, m., almeida, m. s. s., & curcio, d. (2019). hydro-mechanical behavior of a lateritic fiber-soil composite as a waste containment liner. geotextiles and geomembranes, 47(1), 42-47. https://doi. org/10.1016/j.geotexmem.2018.09.005. empresa brasileira de pesquisa agropecuária – embrapa (brazilian agricultural research corporation) (2011). manual de métodos de análise de solo. second edition, rio de janeiro, brazil. eusterhues, k., rumpel, c., & kögel-knabner, i. (2005). stabilization of soil organic matter isolated via oxidative degradation. organic geochemistry, 36, 1567-1575. https://doi.org/10.1016/j. orggeochem.2005.06.010. fageria, n. k., barbosa filho, m. p., & zimmermann, f. j. p. (1994). characterization of physical and chemical properties of lowland soils of some states of brazil. pesquisa agropecuária brasileira, 29, 267274. fagundes, j. r. t. & zuquette, l. v. (2009). capacidade de sorção de materiais inconsolidados residuais da formação botucatu, região de são carlos (sp), brasil. revista brasileira de geociências, 39(3), 494-506. freeze, a. r. & cherry, j. a. (1979). groundwater. prentice hall inc., englewood cliffs, 604 pp. gabarrón, m., faz, a., martínez-martínez, s. & acosta, j. a. (2018). change in metals and arsenic distribution in soil and their bioavailability beside old tailing ponds. journal of environmental management, 212, 292300. https://doi.org/10.1016/j.jenvman.2018.02.010. gabas, s. g., sarkis, j. e. s., & boscov, m. e. g. (2016). heavy metal diffusion and retention in a mineral barrier of compacted lateritic soil. revista brasileira de geologia de engenharia e ambiental, 4, 9-22. guimarães, v. & sígolo, j. b. (2008). detection of contaminants in a bioindicator species (corbicula flumínea) – ribeira de iguape river, são paulo state. química nova, 31(7). 1696-1698. http://dx.doi. org/10.1590/s0100-40422008000700018. gupta, s. s. & bhattacharyya, k. g. (2008). immobilization of pb(ii), cd(ii) and ni(ii) ions on kaolinite and montmorillonite surfaces from aqueous medium. journal of environmental management, 87, 46-58. https://doi.org/10.1016/j.jenvman.2007.01.048. hegazy, a. a., zaher, m. m., abd el-hafez, m. a., morsy, a. a., & saleh, r. a. (2010) relation between anemia and blood levels of lead, copper, zinc and iron among children. bmc research notes, 3(133), 1-9. https:// doi.org/10.1186/1756-0500-3-133. international organization for standardization – iso. (1995). iso 13536 soil quality determination of the potential cation exchange capacity and exchangeable cations using barium chloride solution buffered at ph = 8,1. international organization for standardization – iso. (2001). iso 13317-3 determination of particle size distribution by gravitational liquid sedimentation methods – part 3: x-ray gravitational technique. international organization for standardization – iso. (2005). iso 10390 soil quality determination of ph. jin, c., li, y., li, y. l., zou, y., zhang, g. l., normura, m., & zhu, g. y. (2008). blood lead: its effects on trace element levels and iron structure in hemoglobin. nuclear instruments and methods in physics research section b: beam interactions with materials and atoms, 266, 36073613. https://doi.org/10.1016/j.nimb.2008.05.087. 394 jéssica p. marques, eduardo a. ferreira da silva, carla patinha, mariana c. kasemodel, valéria g. s. rodrigues kasemodel, m. c., lima, j. z., sakamoto, i. k., varesche, m. b. a., trofino, j. & rodrigues, v. g. s. (2016). soil contamination assessment for pb, zn and cd in a slag disposal area using integration of geochemical and microbiological data. environmental monitoring and assessment, 188, 698-732. http://dx.doi.org/10.1007/s10661-016-5708-2. kasemodel, m. c. (2017). avaliação integrada da contaminação por metais potencialmente tóxicos em área de disposição de resíduo de mineração de chumbo – adrianópolis (pr). phd thesis, são carlos school of engineering, university of são paulo, são carlos, brazil. keng, j. c. & uehara, g. (1973). chemistry, mineralogy, and taxonomy of oxisols and ultisols. proceedings of soil crop science society of florida, us, 33, 119-926. knop, a., vangulck, j., heineck, k., & consoli, n. c. (2008). compacted artificially cemented-soil – acid leachate contaminant interactions: breakthrough curves and transport parameters. journal of hazardous materials, 155, 269-276. http://dx.doi.org/10.1016/j. jhazmat.2007.11.056. korf, e. p. (2011). comportamento hidráulico e reativo de uma mistura solocimento para aplicação em barreiras de contenção de resíduos ácidos contendo chumbo e cádmio. m.sc. dissertation, federal university of rio grande do sul, porto alegre, brazil. langmuir, d. (1997). aqueous environmental geochemistry. prentice hall, new jersey, usa, 600pp. lavagnolo, m. c. (2018). landfilling in developing countries. in: cossu, r. & stegmann, r. (eds.) solid waste landfilling: concepts, processes, technologies. elsevier, 773-796. lee, s. z., chang, l., yang, h. h., chen, c. m., & liu, m. c. (1998). adsorption characteristics of lead onto soils. journal of hazardous materials, 63, 37-49. https://doi.org/10.1016/s0304-3894(98)00203-9. lima, r. m. f. & luz, j. a. m. (2001). análise granulométrica por técnicas que se baseiam na sedimentação gravitacional: lei de stokes. rem – international engineering journal – revista escola de minas, 54(2), 155-159. http://dx.doi.org/10.1590/s0370-44672001000200014. linhares, a. l., egreja filho, f. b., oliveira, c. v., & bellis, v. m. (2009). adsorção de cádmio e chumbo em solos tropicais altamente intemperizados. pesquisa agropecuária brasileira, 44(3), 291-299. http://dx.doi.org/10.1590/s0100-204x2009000300011. lustberg, m. & silbergeld, e. (2002). blood lead levels and mortality. archives of internal medicine, 162(21), 2443-2449. https://doi.org/10.1001/ archinte.162.21.2443. mackenzie, r. c. (1957). the differential thermal investigation of clays. mineralogical society, london, uk, 456 pp. mahiques, m. m., figueira, r. c. l., slaroli, a. b., alves, d. p. v. & gonçalves, c. (2013). 150 years of antropogenic metal input in a biosphere reserve: the case study of the cananéia-iguape coastal system, southeastern brazil. environmental earth sciences, 68, 1073-1087. http://dx.doi.org/10.1007/s12665-012-1809-6. miguel, m. g., barreto, r. p., & pereira, s. y. (2017). study of a tropical soil in order to use it to retain aluminium, iron, manganese and fluoride from acid mine drainage. journal of environmental management, 204, 563-570. mohamed, a. m. o., paleologos, e. k., rodrigues, v. g. s. & singh, d. n. (2018). fundamentals of geoenvironmental engineering: understanding soil, water, and pollutant interaction and transport. butterworth-heineman, elsevier. 708p. morandini, t. l. c., & leite, a. l. (2015). characterization and hydraulic conductivity of tropical soils and bentonite mixtures for ccl purposes. engineering geology, 196, 251-267. http://dx.doi.org/10.1016/j. enggeo.2015.07.011. nogueira, t. a. r., abreu-junior, c. h., alleoni, l. r. f., he, z., soares, m. r., vieira, c. s., lessa, l. g. f., & capra, g. f. (2018). background concentrations and quality reference values for some potentially toxic elements in soils of são paulo state, brazil. journal of environmental management, 221, 10-19. https://doi.org/10.1016/j. jenvman.2018.05.048. odukoya, a. m., oresanya, o., & abimbola, a. f. (2013). biogeochemical and engineering characteristics of soils and groundwater around a dumpsite. earth sciences research journal, 17(1), 53-60. oliveira, l. f. c., lemke-de-castro, m. l., rodrigues, c., & borges, j. d. (2010). isotermas de sorção de metais pesados em solos do cerrado de goiás. revista brasileira de engenharia agrícola e ambiental, 14(7), 776-782. https://doi.org/10.1590/s1415-43662010000700014. onyelowe, k., van, d. b., igboayaka, c., orji, f. & ugwuanyi, h. (2019). rheology of mechanical properties of soft soil and stabilization protocols in the developing countries – nigeria. materials science for energy technologies, 2(1), 8-14. papanikolaou, n. c., hatzidaki, e. g., belivanis, s., tzanakakis, g. n. & tsatsakis, a. m. (2005). lead toxicity update. a brief review. medical science monitor, 11(10), 329-336. https://www.ncbi.nlm.nih.gov/ pubmed/16192916. paranavithana, g. n., kawamoto, k., inoue, y., saito, t., vithanage, m., kalpage, c. s., & herath, g. b. b. (2016). adsorption of cd2+ and pb2+ onto coconut shell biochar and biochar-mixed soil. environmental earth sciences, 75, 484-496. doi: 10.1007/s12665-015-5167-z. pejon, o. j. (1992). mapeamento geotécnico da folha piracicaba-sp (escala 1:100.000): estudo de aspectos metodológicos, de caracterização e de apresentação dos atributos. ph.d. thesis. são carlos school of engineering, university of são paulo, são carlos, brazil. piedade, t. c., melo, v. f., souza, l. c. p., & dieckow, j. (2014). threedimensional data interpolation for environmental purpose: lead in contaminated soil in southern brazil. environmental monitoring and assessment, 186, 5625-5638. http://dx.doi.org/10.1007/s10661-0143808-4. pierangeli, m. a. p., guilherme, l. r. g., curi, n., silva, m. l. n., oliveira, l. r. e., & lima, j. m. (2001). efeito do ph na adsorção-dessorção de chumbo em latossolos brasileiros. revista brasileira de ciência do solo, 25(2), 269-277. http://dx.doi.org/10.1590/s010006832001000200003. qian, x., koerner, r. m., & gray, d. h. (2002). geotechnical aspects of landfill design and construction. prentice hall, upper saddle river, 717 pp. raij, b. v. & peech, m. (1972). electrochemical properties of some oxisols and alfisols of the tropics. soil science society of america, proceedings, 36, 587-598. raimondi, i. m. (2014). estudo e caracterização geológica e geotécnica de rejeitos de mineração – adrianópolis (pr). ms dissertation, são carlos school of engineering, university of são paulo, são carlos, brazil. rieuwerts, j. s. (2007). the mobility and bioavailability of trace metals in tropical soils: a review. chemical speciation and bioavailability, 19(2), 75-85. https://doi.org/10.3184/095422907x211918. rodrigues, v. g. s., fujikawa, a., abessa, d. m. s., hortellani, m. a., sarkis, j. e., & sígolo, j. b. (2012). using the freshwater bivalve anodontites tenebricosus (lea, 1834) as a biomonitor of metals in the ribeira de iguape river. química nova, 35, 454-456. http://dx.doi.org/10.1590/ s0100-40422012000300003. rodriguez, j. r. (2013). estudo da viabilidade geológica, geotécnica e geoquímica de solos argilosos como barreiras selantes – vale do ribeira (sp). ms dissertation, são carlos school of engineering, university of são paulo, são carlos, brazil. rowe, r. k., quigley, r. m., & booker, j. r. (1995). clayey barrier systems for waste disposal facilities. taylor & francis ltd, london, uk, 404 pp. roy, w. r., krapac, i. g., chou, s. f. j., & griffin, r. a. (1992). batch type procedures for estimating soil adsorption of chemicals. technical resource document. epa/530-sw-87-006-f, cincinnati, usa. santos, p. s. (1989). ciência e tecnologia de argilas. edgar blücher. são paulo, brazil, 408pp. shubbar, a. a., sadique, m., kot, p., & atherton, w. (2019). future of clay-based construction materials – a review. construction and building materials, 210, 172-187. https://doi.org/10.1016/j. conbuildmat.2019.03.206. 395adsorption of lead (pb) in weathered tropical soil (ribeira valley region brazil) silva, m. l., curi, n., marques, j. j. g. s. m., guilherme, l. r. g., & lima, j. m. (1996). ponto de efeito salino nulo e suas relações com propriedades mineralógicas e químicas de latossolos brasileiros. pesquisa agropecuária brasileira, 31(9), 663-671. https://ainfo. cnptia.embrapa.br/digital/bitstream/ai-sede/19333/1/pab96_10_ set.pdf. skempton a. w. (1953). the colloidal activity of clays, proceedings of the third international conference on soil mechanics and foundation engineering. zurich, switzerland, icosomef, 57-61. skerfving, s. & bergdahl, i. a. (2007). lead. in: nordberg, g. f., fowler, b. a., nordberg, m., & friberg, l. t. (eds.) handbook on the toxicology of metals, third edition, academic press. 599-643. https://doi. org/10.1016/b978-012369413-3/50086-0. soares, m. r. & casagrande, j. c. (2000). adsorção e modelos. in: ribeiro, m. r., nascimento, c. w. a., filho, m. r. r., cantalice, j. r. b. tópicos em ciência do solo. volume 1. sociedade brasileira de ciência do solo, viçosa, brazil. soares, m. r. (2004). distribution coefficient (kd) of heavy metals in soils of the state of são paulo. phd thesis. luiz de queiroz college of agriculture, univeristy of são paulo, piracicaba, brazil. sparks, d. l. (1995). environmental soil chemistry. academic press, san diego, 267 pp. sposito, g. (1984). the surface chemistry of soils. university press, oxford, 234 pp. tan, k. h., hajek, b. f., & barshad, i. (1986). thermal analysis techniques. in: klute, a. (ed.) methods of soil analysis – physical and mineralogical methods. second edition. american society of agronomy. 151-183. tessier, a., campbell, p. g. c., & bisson, m. (1979). sequential extraction procedure for the speciation of particulate trace metals. analytical chemestry, 51, 844-851. tramonte, k. m., figueira, r. c. l., ferreira, p. a. l., ribeiro, a. p., batista, m. f., & mahiques, m. m. (2016). environmental availability of potentially toxic elements in estuarine sediments of the cananéiaiguape coastal system, southeastern brazil. marine pollution bulletin, 103, 260-269. https://doi.org/10.1016/j.marpolbul.2015.12.011. united states department of agriculture usda. (2000). heavy metal soil contamination. soil quality – urban technical note, nº 3. auburn, alabama, usa. 7p. united states environmental protection agency usepa. (2014). reference guide to treatment technologies for mining-inlfuenced water. epa 542-r-14-001. vinhal-freitas, i. c., maldonado, a. c. d., alvarenga, c. b., camargo, r., & wendling, b. (2010). adsorção e dessorção de metais no solo e coeficientes de isotermas de freundlich e langmuir. revista agropecuária técnica, 31(2), 153-163. https://doi.org/10.25066/ agrotec.v31i2.4516. weber, w. j. (1972). physicochemical processes for water quality control. john wiley & sons. new york. 639p. wedepohl, k. h. (1995). the composition of the continental crust. geochimica et cosmochimica acta, 59, 1217-1232. https://doi.org/10.1016/00167037(95)00038-2. world health organization who. (2011). evaluation of certain food additives and contaminants. who technical report series 960. 227p. available at: https:// apps.who.int/iris/bitstream/handle/10665/44515/who_trs_960_eng. pdf;jsessionid=00319ea0aa2ef11b68af829791626867?sequence=1. world health organization – who. (2018). lead poisoning and health. available at <https://www.who.int/en/news-room/fact-sheets/detail/ lead-poisoning-and-health>. wuana, r. a., & okieimen, f. e. (2011). heavy metals in contaminated soils: a review of sources, chemistry, risks and best available strategies for remediation. international scholarly research notices ecology. 20 p. http://dx.doi.org/10.5402/2011/402647. yong, r. n., mohamed, a. m. o., & warkentin, b. p. (1992). principles of contaminant transport in soils. elsevier, amsterdam, netherlands, 327 pp. yong, r. n. (2001). geoenvironmental engineering: contaminated soils, pollutants fate, and mitigation. boca raton: crc press, 320 pp. zia, a., berg, l. v. d., ahmad, m. n., riaz, m., zia, d., & ashmore, m. (2018). controls on accumulation and soil solution partitioning of heavy metals across upland sites in united kingdom (uk). journal of environmental management, 222, 260-267. https://doi.org/10.1016/j. jenvman.2018.05.076. zuquette, l. v., silva jr. e. m., & garcia, a. (2008). aspectos de sorção para os materiais inconsolidados da região de são carlos (sp). rem – international engineering journal – revista escola de minas, 1(2), 219-230. http://dx.doi.org/10.1590/s0370-44672008000200017 letter from the editor on september 2017, two large earthquakes struck méxico. numerous casualties and at least 40 buildings collapsed in mexico city. the earthquakes are explained by the tectonic setting of central america, the cocos plate subducts underneath the north american plate at about 7-8 cm a year, making mexico a seismically active region. but the two earthquakes with hypocenter depths between 50 and 60 km – did not occur in the contact between the two tectonic plates as is usually expected, but rather within the cocos plate as it bends downward within the mantle. both events showed a normal faulting mechanism, and although they were widely felt, their major impact was in densely populated mexico city, where wave amplification is expected due to the geological features of the city's soil structure. we know that we cannot predict earthquakes, and although this is the holy-grail in seismology, it seems like we are not close to accurately predict them. nevertheless, early warning systems have been developed in various places, including mexico, and they were successful in issuing an alarm, although with a few seconds before the strong shaking. these developments can save lives and continue to be relevant and are likely to be implemented in other regions, including the pacific coast of the us. another aspect that can save lives is people’s awareness. the m7.1 earthquake occurred in the afternoon of the anniversary of the 1985 michoacan earthquake, just a couple of hours after an earthquake drill in mexico city, so people had just recently been remained of what to do. this most likely saved lives. we should ask ourselves, are we prepared? is our city prepared? germán prieto editor in chief carlos a. vargas former editor l e t t e r f r o m t h e e d it o r earth sciences research journal earth sci. res. j. vol. 21, no. 3 (september, 2017) new models using scale analysis and perturbation methods were derivated starting from the shallow water equations based on barotropic fluids. in the paper, to discuss the irregular topography with different magnitudes, especially considering the condition of the vast terrain, some modified quasi-geostrophic barotropic models were obtained. the unsteady terrain is more suitable to describe the motion of the fluid state of the earth because of the change of global climate and environment, so the modified models are more rational potential vorticity equations. if we do not consider the influence of topography and other factors, the models degenerate to the general quasi-geostrophic barotropic equations in the previous studies. este trabajo deduce nuevos modelos con el uso de los métodos de análisis a escala y de perturbación a partir de las ecuaciones de aguas poco profundas con base en fluidos barotrópicos. en este artículo se obtuvieron algunos modelos semigeostróficos barotrópicos para aplicar en zonas de topografía inestable con diferentes magnitudes y considerar especialmente la condición del extenso terreno. la topografía inestable es más propicia para describir el movimiento del estado fluido de la tierra debido al cambio del clima y ambiente, por lo tanto los modelos modificados son ecuaciones de vorticidad potenciales más razonables. si no se considera la influencia de la topografía y otros factores, los modelos se reducirían a las ecuaciones generales semigeostróficas barotrópicas de estudios anteriores. abstract resumen keywords: quasi-geostrophic; potential vorticity; scale analysis; topography. palabras clave: semigeostrofía; vorticidad potencial; análisis a escala; topografía. the modified quasi-geostrophic barotropic models based on unsteady topography record manuscript received: 01/03/2017 accepted for publication: 31/03 /2017 how to cite item zhao, b., sun, w., & zhan, t. (2017). the modified quasi-geostrophic barotropic models based on unsteady topography. earth sciences research journal, 21(1), 23-28. doi:http://dx.doi.org/10.15446/esrj.v21n1.63007 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.63007 modelos semigeostróficos barotrópicos modificados con base en topografía inestable g e o m o r p h o l o g y earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017): 23 28 baojun zhao 1, wenjin sun 1,*, tianming zhan 2 1 college of oceanography, hohai university, nanjing 210098, china 2 school of technology, nanjing audit university, nanjing 211815, china *email of corresponding author: sunwenjin950925@126.com 24 baojun zhao, wenjin sun and tianming zhan 1. introduction in recent decades, many scholars have conducted extensive research on large-scale atmospheric and oceanic dynamics. among them, the topographic effect has great influences on the dynamic mechanism of potential vorticity equations (pedlosky, 1974; collings, 1980; pedlosky, 1980; treguier, 1989). in the atmosphere, the topographic height plays an importance role in atmospheric cyclone and anticyclone changes. luo (1990) pointed out the topographic effect was also an important factor of forming atmospheric blocking, and the global atmospheric circulation would even get affected by the topographic effect. lu (1987) discussed the effects of topographic height and shaped on rossby wave activation and the influences of topographic south-north and east-west slopes on waveform and energy propagation. chen (1998) and jiang (2000) derived the quasi-geostrophic potential vorticity equation with large-scale topography, friction, and heating under the barotropic model, and the large-scale effects of qinghai-tibet plateau on atmosphere were discussed. in addition, the oceanic topography is very complex, such as the north west shelf of australia (holloway, 1997), portugal shelf sea area (sherwin, 2002), etc. the relationship between topography and ocean circulation was pointed out in the literature (roslee et al., 2017b; kamsani, 2017; la, 1990; marshall, 1995; alvarez, 1994; sou, 1996). cessi (1986) discussed the important role of topography in ocean circulation. holloway (1992) introduced the interaction of eddies with seafloor topography and argued that ocean circulations would be significant interaction between turbulent vortices and topography rather than gravity wave drag. then, general expressions for the eddy-topographic force, eddy viscosity, and stochastic backscatter, as well as a residual jacobian term, are derived for barotropic flow over mean topography by frederiksen (1999). all the above researches, actually the topography height also changes with time in the earth fluid. changes in topography can lead to tsunamis, floods and natural disasters(abdullah, 2017; elfithri, 2017; rahim, 2017). yang (2011,2012) and song (2012, 2013) discussed topography changes over time discussed topography changes over time t, the influence of the nonlinear long wave amplitude and waveform. da (2013) discussed the shallow water equation forms when underlying surface slowly changes with time and obtained the vorticity equation with an underlying surface. this model considered the actual circumstances that the topography changes with time-space (erfen et al., 2017; roslee, 2017a) . in this paper, we discuss different magnitude topography under spatial-temporal variable and obtain some modified models, which have important effects on the future discussion about the waveform changes of nonlinear long waves. this paper is organized as follows: in section 2, starting from the rotating shallow water equation set, we give the bottom topography which is not smooth boundary and simplify the equation set with unsteady topography. section 3 is given scale analysis and perturbation methods; then we obtain new equation set by topographic conditions with different magnitudes. later we derive the equations with different orders and get some modified models in section 4. finally, make the relevant conclusions in section 5. 2. basic equations assuming the static equilibrium condition, the fluid can be regarded as barotropic, incompressible, frictionless state. the upper boundary height is ( , , )h x y t , and , free surface pressure intensity is constant. the basic equation set can be written as (pedlosky, 1987) (1a) (1b) (1c) (1d) especially, the bottom topography is about spatial-temporal variable ( , , )b bh h x y t= , the coriolis parameter is (2) first of all, we integrate eq. (1c) (3) and get (4) so (5) (6) eqs. (5) and (6) indicate that the horizontal pressure gradient under the barotropic model can be expressed by the gradient of the free surface gravitational potential (taharin & roslee, 2017). we assume that the preliminary horizontal velocity is independent of the z , eqs. (1a) and (1b) can be rewritten as (7a) (7b) then integrate continuity eq. (1d) from ( , , )bz h x y t= to the free surface ( , , )z h x y t= , we can get (7c) assuming that the free surface height h is constant when the fluid is static (pedlosky, 1987). (8) the barotropic equation set with unsteady topography can be written as (9a) 25the modified quasi-geostrophic barotropic models based on unsteady topography (9b) (9c) if (10) the equation set can be written as (11a) (11b) (11c) 3. scale analysis, perturbation methods making dimensionless analysis on the eqs. (11a), (11b) and (11c), for the large-scale motions (12) where l is an undetermined parameter. substituting eq. (12) into eqs. (11a), (11b) and (11c), then (13a) (13b) (13c) eqs. (13a) and (13b) divided by 0f u , (13c) divided by 2 0c u ghu l l = , we get (14a) (14b) (14c) where (15) let (16) meantime so (17) then we get (18a) (18b) (18c) where (19) 4. derive the barotropic models making classified discussion on the magnitude of λ. 4.1. λ ~ 1 magnitude the scale of the ϕb is consistent with the small amplitude function ϕ, most of the topography parameters following with this situation in the real world. 0ro order approximation (20) or a dimensional form (21) (0) (1) 1 1 1u u rou= + + (0) (1) 1 1 1v v rov= + + (0) (1) 1 1 1 ˆ̂̂ rof f f= + + 26 baojun zhao, wenjin sun and tianming zhan 1ro order approximation (22a) (22b) (22c) after calculation, the dimensionless vorticity equation is derived (23a) (23b) eliminating from the eqs. (23a), (23b), we get (24) namely (25) where (26) λr is the rossby radius of deformation, eq. (25) is a modified model. 4.2. λ ~ 10 magnitude large-scale atmospheric motion of large topography is suitable for such conditions (l ~ 106 m, u ~ 10m/s, f 0 ~ 10-4 s -1 ). for example, a case study of tibetan plateau topography, the height is 3 4 ×103m approximately which is suit for the situation. 0ro order approximation (27) or a dimensional form (28) in order to get the results, we should consider the magnitude of , according to the eq. (18c), 20m can be considered as a parameter. 1ro order approximation (29a) (29b) (29c) or dimensional form (30a) (30b) (30c) eqs. (30a), (30b) and (30c) can be transformed into the vorticity equations (31a) (31b) eqs. (31a), (31b) are quasi-geostrophic barotropic models, so (32a) (32b) where (33) 27the modified quasi-geostrophic barotropic models based on unsteady topography assuming , we eliminate from eqs. (32a) and (32b), then (34) finally (35) eq. (35) is a modified model. 4.3. λ ≤ 10 -1 magnitude according to (18c), it is observed that the magnitude of topography expression form λ is too small, which approximately ignores the influences of topographic effect. now we can observe that the eqs. (25), (35) are two quasigeostrophic barotropic models under variable topographic conditions with time t . if the topographic effect is independent of time, equations degrade into the following forms. under the stable topography, eq. (35) degenerates into (36) where (37) the model (25) under the stable topography can be written as (pedlosky,1987) (38) where (39) otherwise, if we don’t consider the topographic effect and the divergence of the fluid, eqs. (36) and (38) degenerate into (liu, 1991) (40) eq. (40) is a classics dynamics model used by the large-scale atmospheric and oceanic motions. 5. conclusion (a). under the unsteady topography, some new modified models (25), (35) are derived. these models meet the general rule that the topography changes with time in reality. when the topography has nothing to do with the time, eq. (35) degenerates into a dynamics model (36), eq. (25) degenerates into eq. (38) which is a quasi-geostrophic barotropic model under the spatial topography. (b). the modified models under the topographic effect with different magnitudes are presented, we can see the pattern under the condition of large terrain, which is the improvement of the model. after the above-modified models are given, we will also derive the mathematical model for rossby wave in the further study, and the further exploration of the large-scale factual influences of topography on atmosphere and ocean are required. acknowledgements this work was supported by the national nature science foundation of china under grant (no. 61502206), key projects of water conservancy science and technology of jiangsu province, china (no. 2010500312), and the research innovation program for college graduates of jiangsu province (no .cxzz12_0224). references abdullah, m. p. (2017). stakeholders’ response and perspectives on flood disaster of pahang river basin. malaysian journal of geosciences, 1(1) 43-49. alvarez, a. (1994). effect of topographic stress on circulation in the western mediterranean. journal of geophysical research atmospheres, 99(c8), 16053-16064. cessi, p., & pedlosky, j. (1986). on the role of topography in the ocean circulation. journal of marine research, 44 (3), 445-471. chen, j., & liu, s. k. (1998). the solitary waves of the barotropic quasi-geostrophic model with large-scale orography. advances in atmospheric sciences, 15 (03), 404-41. collings, i. l., & grimshaw, r. (1980). the effect of topography on the stability of a barotropic coastal current. dynamics of atmospheres and oceans, 5(2), 83-106. da, c. j. (2013). dynamic modification of the shallow water equation on the slowly changing underlying surface condition. acta physica sinica, 62(3), 039202. elfithri, r. (2017). pahang flood disaster: the potential flood drivers. malaysian journal of geosciences, 1(1) 34-37. erfen, h. f. w. s. (2017.) geochemical characterization of sediments around nukakatan valley, tambunan, sabah. geological behavior, 1(1)13-15. frederiksen, j. s. (1999). subgrid-scale parameterizations of eddytopographic force, eddy viscosity, and stochastic backscatter for flow over topography. journal of the atmospheric sciences, 56(11), 1481-1494. holloway, g. (1992). representing topographic stress for large-scale ocean models. journal of physical oceanography, 22(9), 1033-1046. holloway, p. e., et al., 1997. a nonlinear model of internal tide transformation on the australian north west shelf. journal of physical oceanography, 27(6), 871-896. jiang, x. (2000). a barotropic quasi-geostrophic model with largescale topography, friction and heating. journal of tropical meteorology, 6(01), 65-74. la, v., & paul, e. (1990). the western mediterranean circulation experiment (wmce): introduction. journal of geophysical research: oceans, 95(c2), 1511-1514. liu, s. k., & liu, s. d. (1991). atmospheric dynamics. beijing, bj: peking university press. lu, k. l. (1987). the effects of orography on the solitary rossby waves in a barotropic atmosphere. acta meteorologica sinica, 45, 267-273. luo, d. h. (1990). topographically forced rossby wave instability and the development of blocking in the atmosphere. advances in atmospheric sciences, 7(4), 433-440. marshall, d. (1995). influence of topography on the large-scale ocean circulation. journal of physical oceanography, 25(7), 1622-1635. 28 baojun zhao, wenjin sun and tianming zhan pedlosky, j. (1974). longshore currents, upwelling and bottom topography. journal of physical oceanography, 4(2), 214-226. pedlosky, j. (1980). the destabilization of shear flow by topography. journal of physical oceanography, 10(11), 1877-1880. pedlosky, j. (1987). geophysical fluid dynamics second edition. springer, new york, 57-246. rahim, i. a. (2017). kampung mesilou landslide: the controlling factors. geological behavior, 1(1) 19-21. roslee, r. (2017). gis application for comprehensive spatial soil erosion analysis with musle model in sandakan town area, sabah, malaysia. geological behavior, 1(1) 01-05. roslee, r. (2017). flood potential analysis (fpan) using geo-spatial data in penampang area, sabah. malaysian journal of geosciences, 1(1) 01-06. sherwin t. j., vlasenko v. i., & stashchuk, n. (2002). along-slope generation as an explanation for some unusually large internal tides. deep sea research part i: oceanographic research papers, 49(10), 1787-1799. song, j., et al., 2012. beta effect and slowly changing topography rossby waves in a shear flow. acta physica sinica, 61(21), 210510, http:// dx.doi.org/10.7498/aps.61.210510. song, j. (2014). nonlinear rossby waves excited slowly changing underlying surface and dissipation. acta physica sinica, 63(6), 060401, http://dx.doi.org/10.7498/aps.63.060401. sou, t. (1996). effects of topographic stress on caribbean sea circulation. journal of geophysical research: oceans, 101(c7), 16449-16453. taharin, m. r., & roslee, r. (2017). comparison of cohesion (c’), and angle of internal friction (ф’) distribution in highland area of kundasang by using ordinary kriging and simple kriging. geological behavior, 1(1) 16-18. treguier, a. m. (1989). topographically generated steady currents in barotropic turbulence. geophysical and astrophysical fluid dynamics, 47, 43-68. yang, h. w. (2011). forced solitary rossby waves under the influence of slowly varying topography with time. chinese physics b, 20(11), 120203. yang, h. w. (2012). generation of solitary rossby waves by unstable topography. communications in theoretical physics, 57(3), 473-476. quality control can effectively improve the quality of surface meteorological observations. to ensure the stability and effectiveness of a quality control model under different terrain and climate conditions, it is necessary to structure a quality control model with strong generalization ability. algorithms such as the random forest algorithm provide such generalization ability. however, machine learning algorithms are slower than traditional mathematical models. therefore, a random forest quality control algorithm based on the principal component analysis (pca-rf) is proposed in this paper. fifteen target stations under different climatic and geomorphological conditions were selected and tested using observations collected four times daily at neighboring stations from 2005-2014. the results show that using pca to analyze the elemental composition and select elements with high correlation factors, as well as applying the random forest algorithm, can effectively reduce the run time and keep the accuracy of the model. the training sample dependence, model prediction accuracy and error detection rate of the pca-rf model are superior to those of the spatial regression method. therefore, the pca-rf method is a better quality control model for the spatial quality control of multiple elements of surface air temperature observations. abstract keywords: surface air temperature; quality control; random forest; principal component analysis; spatial quality control method for surface temperature observations based on multiple elements issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n2.65185 m e t e o r o l o g y earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017): xiaoling ye1,2, xing yang1*, xiong xiong1, shuai yang1, yang chen1 1school of information and control, nanjing university of information science and technology, nanjing 210044, china 2collaborative innovation center on forecast and evaluation of meteorological disasters, nanjing university of information science and technology, nanjing 210044, china *email of corresponding author: harryyangx@gmail.com el control de calidad puede mejorar efectivamente la calidad de las observaciones meteorológicas. para asegurar la estabilidad y efectividad de un modelo de control de calidad bajo condiciones diferentes de terreno y climáticas es necesario estructurar un esquema con una fuerte habilidad de generalización. algoritmos como el método de bosques aleatorios (del inglés random forest) cumplen con estas condiciones. sin embargo, los algoritmos de maquinas de aprendizaje son más lentos que los modelos matemáticos tradicionales. en este artículo se propone un algoritmo de control de calidad tipo bosques aleatorios basado en el análisis de componentes principales (pca-rf). se seleccionaron 15 estaciones objetivo bajo diferentes condiciones climáticas y geomorfológicas y se evaluaron con observaciones realizadas cuatro veces por día en estaciones vecinas desde 2005 hasta 2014. los resultados muestran que usando pca para analizar la composición elemental y seleccionar elementos con factores de correlación alta, al igual que la aplicación del algoritmo random forest, se puede reducir efectivamente el tiempo de ejecución y mantener la exactitud del modelo. la dependencia de la muestra de prueba, la exactitud del modelo de predicción y la tasa de detección de error del modelo pca-rf son superiores a aquellos del método de regresión espacial. por lo tanto, el método pca-rf es un mejor modelo para el control de calidad de elementos múltiples en las observaciones superficiales de aire y temperatura. resumen palabras clave: método de control de calidad espacial para observaciones de temperatura superficial basado en múltiples elementos record manuscript received: 23/05/2017 accepted for publication: 30/06/2017 how to cite item: ye, x., yang, x., xiong, x., yang, s., & chen, y. (2017). spatial quality control method for surface temperature observations based on multiple elements. earth sciences research journal, 21(2), 95 99. doi:http://dx.doi.org/10.15446/esrj.v21n2.65185 método de control de calidad espacial para observaciones de temperatura superficial basado en múltiples elementos 102 xiaoling ye, xing yang, xiong xiong, shuai yang and yang chen102 1. introduction meteorological observations are important for identifying and understanding variations and changes in regional and global climate (feng et al., 2004). these observations are also essential to a wide range of meteorological applications, such as climate monitoring, weather forecasting and the evaluation of numerical weather prediction (nwp) models (he et al., 2016; ingleby and lorenc, 1993). therefore, it is of particular importance to improve the accuracy of meteorological observations, especially surface observations (barnes, 1964; gandin, 1988). on the one hand, surface meteorological observations have a longterm history in china. on the contrary, surface meteorological observations can more accurately represent the atmospheric characteristics of the nearsurface compared to other meteorological observations (xu et al., 2013; hu et al., 2002; mi et al., 2014). however, surface observation stations are easily affected by the gross error and random error during the data acquisition process. thus, the primary task and purpose of quality control (qc) are to identify the gross error and large random error associated with large numbers of observations (shi-wei et al., 2009). in surface meteorological observations, air temperature is one of the most important elements because it has a considerable impact on human activity (cheng et al., 2016). hence, a variety of qc methods has been recently proposed for surface air temperature observations. the qc methods include the extreme range check (reek et al., 1992; kubecka, 2001; allen et al., 1998; forsythe et al., 1995), internal consistency check (baker, 1992), time consistency check (shafer et al., 2000; lanzante, 1996), and spatial consistency check methodologies. qc research has focused on spatial consistency check, which is used for multi-station qc. the spatial qc method predicts the value at the target station based on the values at neighboring stations, and it can evaluate the reliability of observations. the inverse distance weighting (wade, 1987) method (idw) uses the proportions of the inverse distances between multiple neighboring stations and the target stations as weights, and it requires the terrain features and distribution of neighboring stations. the bayesian qc method (lorenc and hammon, 1988) uses the probability of observations as a gross error. in 2007, hubbard et al. (2007) proposed the spatial regression test method (srt). srt calculates weights based on the standard error between the target station and neighboring stations, and it is less influenced by terrain and the distribution of neighboring stations compared to the idw method. based on the srt method, xu et al. (2014) proposed the probabilistic spatial-temporal method (srt-ps). this method considers the uncertainty in temperature observations and eliminates the associated effect on temperature fluctuations. the most widely used traditional methods discussed above are the idw method and srt method, and the srt method is considered superior to the idw method (hubbard and you, 2005). thus, the srt method is for comparison with the random forest quality control algorithm based on the principal component analysis (pca-rf) method proposed in this paper. the pca-rf method is used to reconstruct the training samples by extracting high correlation factors from meteorological elements using pca. therefore, the efficiency and accuracy of rf model are improved. the specific objectives of this paper are as follows: (1) to compare the evaluation indexes of the rf and srt methods using training samples over different periods (ridzuan et al., 2017). (2) to set different correlation factor thresholds and compare the evaluation indexes and run times of the pca-rf qc model based on these different thresholds. and (3) to compare the detection rates of the srt and pca-rf methods and evaluate the predictive performances of the qc models. 2. data china is a large country with an area of approximately 9.6 million km2. about 98% of the land area stretches between the latitudes of 20° n and 50° n. thus; the country extends from the subtropical zones in the south to the temperate zones (including warm temperate and cool temperate) in the north (yang et al., 2013; jiang et al., 2016). with recent global climate warming, the national annual average temperature increased by approximately 1.1 degrees celsius, with high temperatures in north and low temperatures in the south. additionally, the magnitude of the temperature increase in winter was larger than that in summer (asis et al., 2017) moreover, other climate factors exhibited fluctuations in different regions (zheng et al., 2010, cheng et al., 2009). therefore, it is more meaningful in spatial qc to consider different climates and geomorphological regionalization. in this paper, data were collected from january 1, 2005, to december 31, 2014, four times daily (02:00, 08:00, 14:00 and 20:00 beijing time (cst)) in different climate regions in china. the data include six types of meteorological elements: ground surface temperature (gst), pressure (pre), relative humidity (rh), temperature (tem), wind direction (wd) and wind speed (ws). these data were obtained from the china national meteorological center and had undergone basic qc; thus, the accuracy of the data is acceptable. missing data, changes in station locations and changes in precipitation measurement instruments during the observation period can influence the homogeneity of datasets (irannezhad et al., 2016). small percentages of missing data at some stations can influence an analysis, but pca can reduce the effect of missing data to a certain degree. zheng et al. (2010) suggested dividing china into 12 temperature zones, including a humid region, a semi-humid region, a semi-arid region and an arid region. li et al. (2013) discussed the specific steps and methods associated with dividing china into six major categories. in this paper, 15 stations with different climates and topographies are selected; however, stations 59948, 59985 and 59287 (in the mid-tropic, equatorial tropic and south subtropical areas) are not selected because they do not have sufficient numbers of neighboring stations (rahim et al., 2017) the target stations are shown in figure 1, and the selection of target stations accounted for the distribution of climate zones and geomorphologic regionalization. neighboring stations are selected based on the distance between the target station and neighboring stations, where the radius of the neighboring stations is not more than 100 km. the number of neighboring stations is 5-10, as shown in table 1. however, stations 55472 and 54823 have a limited number of neighboring stations or significant gaps in data. table 1. the number of neighboring stations for 15 target stations. 103spatial quality control method for surface temperature observations based on multiple elements x x xe s q= figure 1. the distribution of target stations based on geomorphological regionalization in china. to test the feasibility of the pca-rf method, artificial error is added randomly to the observations at the target station. a comparative analysis is performed using observations with artificial random error and the predicted values of the prf method. hubbard et al. (2005) presented a method of adding false error to data. a random number q with a mean of zero and a range of ± 3,5 was selected and added to the observations (hubbard et al., 2007): (1) where e and s denote the error magnitude and standard deviation, respectively, of temperature observations at the target station. additionally, x is the location of mistake value insertion. 3. method 3.1. random forest algorithm random forest (rf) is an ensemble learning method for classification and regression that constructs some randomized decision trees during the training phase and make predictions by averaging the results (scornet et al., 2015). in ensemble learning, each decision tree is a weak classifier, and some weak classifiers are combined into a strong classifier. this approach is much easier than looking for a strong classifier directly. rf generates many regression trees to improve the prediction performance of the model, and each split in the tree is determined using a randomized subset of variables/factors at each node (rahmati et al., 2016). the outcome is the average of the results of all the trees (cutler et al., 2007). 3.2. principal component analysis algorithm principal component analysis (pca) is a method for analyzing a positive semi-definite hermitian matrix. the goal of pca is to identify the most urgent basis by using a new base to filter out the noise and reveal hidden structures (shlens, 2014). notably, pca is used to analyze the covariance matrix of sampled random vectors (lloyd et al., 2014). high-dimensional data are mapped into a new coordinate system and transformed to low-dimensional data that can partly represent the original data. in this manner, pca can significantly reduce the run time of algorithms, especially in big data processing. 3.3. pca-rf algorithm combined with the above concepts, the pca-rf algorithm is proposed in this paper. the concrete steps in the algorithm are divided as follows (asis et al., 2017) step 1: the pca method is used to analyze the multiple element observations in space. assuming that n represents the number of samples, m represents the number of features, and x n,m represents the feature set, the average value of the ith feature iy is calculated using (2). (2) the standard error si of the ith feature is calculated using (3). (3) then, through standardization, (4) is obtained. (4) finally, the correlation matrix of feature sample rm,m is obtained, as shown in (5). (5) an optimal feature sample is established by extracting the matrix with high correlation factors. step 2: the rf algorithm is used to train and test the optimal feature samples, and the pca-rf model produces a predictive value esty . step 3: the predicted values esty and observed values obsy are tested for a given threshold, as shown in (6): (6) where f is the qc parameter and ∂ is the standard error between observed values and predicted values at the target station. if the conditions of (6) cannot be met, the data are considered incorrect (or associated with error). if the conditions of (6) are met, the data are considered correct values (ridzuan et al., 2017). 3.3. evaluation of model performance to analyze different forecasting models with different conditions, the root means square error (rmse) and mean absolute error (mae) were selected as evaluation metrics. rmse is the square root of the ratio of the square of the predicted value and the real value, and mae can accurately reflect the size of the prediction error. their formulas are as follows. (7) (8) 4. results and discussion the estimates using pca-rf were evaluated against the measurements at 15 target stations, and selected observations are shown in figure 1 and table 1 (you et al., 2017). in the analysis of sample dependence, observations over different periods were selected to predict observations of daily mean temperature, and the training samples were the six meteorological elements observed four times daily in the 1, 3, 5, 7 and 9 years before 2014. the testing samples were the six meteorological factors founded four times daily in 2014. in the analysis of the correlation factor, the same observations were used as the training samples and 104 xiaoling ye, xing yang, xiong xiong, shuai yang and yang chen testing samples at different correlation factor thresholds (>0, >0.5, >0.8, >0.9, >0.95). in the analysis of the detection rates of the pca-rf and srt methods, the evaluation indexes (mae and rmse) and two types of error (type i and type ii) were used to compare the performances of pca-rf and srt, and observation selection was the same as that utilized in the correlation factor analysis. observations over different periods were selected to analyze the sample dependence based on rf and srt, and the results are shown in figure 2. in general, as the number of training samples increased, the error in the testing samples decreased. as shown in figure 2, the error trends at 15 target stations exhibited this pattern. regardless of the observation period (one year, three years, five years, seven years or nine years), the mae and rmse values of the rf method are smaller than those of the srt method, and the fluctuations in training samples are smaller. hence, the rf method performed better than srt method based on the sample dependence. although the error associated with the training samples for seven years of observations is less than that associated with the training samples for nine years of observations at target station 57494, the training samples based on nine years of observations exhibited less error in general. therefore, the training sample based on nine years of observations was used in the following analysis. figure 2. evaluation indexes of the rf and srt methods over different observation periods. moreover, figure 2 shows that stations 54823, 56951, 57245 and 58606 have relatively small evaluation index values because the neighboring stations around these target stations are relatively dense (you et al., 2017). additionally, the main reason for the performance difference between the srt method and rf method is the sparse distribution of neighboring stations or low spatial correlation between values. the rf method has relatively less error at the target stations compared to the large error of the srt method, such as at stations 51477, 56684 and 57245. this result suggests that the rf method is superior to the srt method. to investigate the performance of the pca-rf method under different thresholds of the correlation factor, the evaluation index values and run time of the model obtained for seven cases (no, >0, >0.5, >0.8, 0.9, 0.95, alone) are shown in figure 3, where “no” means that the rf model does not perform pca and “alone” means that modeling is only based on temperature observations at neighboring stations. additionally, the unit of “time” (run time of the model) is 100 seconds. figure 3 shows that the evaluation index (mae and rmse) values increased slightly as the correlation factor threshold increased. moreover, the run time of the model decreased as the correlation factor threshold increased. the results show that the choice of correlation factor determines the run time of model and affects the size of the type error. a correlation factor threshold of 0.5 was chosen as the optimal value to construct the qc model. compared to the other correlation factor thresholds, the correlation factor threshold of 0.5 produced the shortest run time of the model, and the model error is relatively small. on the one hand, although the model error is reduced without pca, the run time of the model increases considerably, as pca effectively reduces the run time of the model. on the other hand, although the run time of the quality control model with only temperature elements is short, the error is larger than the error when six elements are considered in the spatial qc model. in summary, the accuracy and run time of the qc model can be improved by choosing a correlation factor threshold of 0.5. thus, the rf model performed better than the srt approach based on the evaluation indexes and error detection rate analysis. 105spatial quality control method for surface temperature observations based on multiple elements 105 figure 3. evaluation indexes and run time of the model for the 15 target stations under different thresholds of the correlation factor. the error detection rate is the ultimate criteria for a qc model. the error detection rate is related to the size of the qc parameter, and the choice of qc parameter is related to the two types of error. in type i error, the real value is treated as an incorrect value. in type ii error, an incorrect value is treated as the true value. in the realm of qc, the intersection of the two types of error is chosen as the optimal qc parameter. the optimal qc parameters for the pca-rf and srt methods obtained based on the intersection of the two types of error are listed in table 2, in which the 15 target stations are numbered 1 to 15 according to the order of their previous names. the corresponding error detection rate is shown in figure 4. table 2 and figure 4 illustrate that the error detection rate of the pca-rf method is superior to the error detection rate of the srt method. thus, the pca-rf method is more stable and efficient based on the target stations with different terrain and climate characteristics. 106 xiaoling ye, xing yang, xiong xiong, shuai yang and yang chen table 2. the optimal quality control parameters of the pca-rf and srt methods. figure 4. the detection rates of the pca-rf and srt methods at 15 target stations. 5. conclusions this paper studies a spatial qc model of six meteorological elements and temperature observations using the pca-rf model, and the results are compared to those of the srt method. the performance of qc has a considerable effect on the selection of the correlation factor threshold and the density of neighboring stations around the target station. the simulation results show that the pca-rf method is less dependent on training samples than is the srt method, and it performed better at target stations with sparse neighboring stations or low spatial correlation. the run time of the model can be reduced effectively by choosing the appropriate correlation factor at the expense of a small increase in the error. the pca-rf method is more stable and has a higher error detection rate compared to the stability and detection rate of the srt method. in the next study, time series will be combined with the pca-rf method, and the error detection rate of the pca-rf qc model may be improved by eliminating the influence of noisy time series. the physical connection between multiple elements at neighboring stations must be taken into account to improve the performance of the model through the establishment of a new sample set based on physical connections. acknowledgements this research was financially supported by the national natural science foundation of china (grant no. 41675156), the postgraduate research and practice innovation program of jiangsu province (grant no. sjzz16_0155), the national public welfare industry (meteorological) scientific research (no. gyhy20110604) and the six talent peaks project in jiangsu province (wlw-021). references allen, r.g., et al. “crop evapotranspiration-guidelines for computing crop water requirements-fao irrigation and drainage paper 56.” fao, rome 300 (1998): d05109. asis j., tahir s.h., rahim a.r., konjing z., kob r.c., tjia h.d., “smaller benthic foraminifera analysis of kudat formation, kudat, sabah: preliminary interpretation.” geological behavior 1 (2017): 27-29 baker, n.l., “quality control for the navy operational atmospheric database.” weather and forecasting 7 (1992): 250-261. barnes, s. l., “a technique for maximizing details in numerical weather map analysis.” journal of applied meteorology 3 (1964): 396-409. cheng a.r., lee t.h., ku h.i., chen y.w., “quality control program for real-time hourly temperature observation in taiwan.” journal of atmospheric and oceanic technology 33 (2016): 953-976. cheng w., zhou c., chai h., zhao s., li b., “quantitative extraction and analysis of basic morphological types of land geomorphology in china.” journal of geo-information science (in chinese) 6 (2009): 725-736. cutler d.r., edwards t.c., beard k.h., cutler a., hess k.t., gibson j., lawler j.j., “random forests for classification in ecology.” ecology 88 (2007): 2783-2792. feng s., hu q., qian w., “quality control of daily meteorological data in china, 1951–2000: a new dataset.” international journal of climatology 24 (2004): 853-870. forsythe, w.c., rykiel, e.j., stahl, r.s., wy, h.i., schoolfield, r.m. “amodel comparision for daylength as a function of latitude and day of year.” ecological moelling 80 (1995): 87-95. gandin, l.s., “complex quality control of meteorological observations.” monthly weather review 116 (1988) 1137-1156. he q., wang z., he j., “bias correction for retrieval of atmospheric parameters from the microwave humidity and temperature sounder onboard the fengyun-3c satellite.” atmosphere 7 (2016): 156. hu, q., feng, s., schaefer, g., “quality control for usda nrcs sm-st network soil temperatures: a method and a dataset*.” journal of applied meteorology 41 (2002): 607-619. hubbard k., goddard s., sorensen w., wells n., osugi t., “performance of quality assurance procedures for an applied climate information system.” journal of atmospheric and oceanic technology 22 (2005): 105-112. hubbard k.g., guttman n.b., you j., chen z., “an improved qc process for temperature in the daily cooperative weather observations.” journal of atmospheric and oceanic technology 24 (2007): 206-213. hubbard k.g., you j., “sensitivity analysis of quality assurance using the spatial regression approach-a case study of the maximum/ minimum air temperature.” journal of atmospheric and oceanic technology 22 (2005): 1520-1530. ingleby n.b. and lorenc a.c., “bayesian quality control using multivariate normal distributions.” quarterly journal of the royal meteorological society 119 (1993): 1195-1225. irannezhad m., marttila h., chen d., kløve b., “century-long variability and trends in daily precipitation characteristics at three finnish stations.” advance in climate change research (2016). jiang g.j., lv h.d., liu l., “process safety evaluation model based on lec and grey theory.” journal of mechanical engineering research and developments, 39(2016): 24-29. ridzuan a.a., zahar u.a.u., noor n.a.m., “association of evacuation dimensions towards risk perception of the malaysian students who studied at jakarta, medan, and acheh in indonesia.” malaysian journal of geoscience, 1 (2017): 07-12. kubecka p., “a possible world record maximum natural ground surface temperature.” weather 56 (2001): 218-221. lanzante, j. r., “resistant, robust and non-parametric techniques for the analysis of climate data: theory and examples, including applications to historical radiosonde station data.” international journal of climatology 16 (1996): 1197-1226. li b., pan b., cheng w., han j., qi d., zhu c., “research on geomorphological regionalization of china.” acta geographica sinica (in chinese) 68 (2013): 291-306. lloyd s., mohseni m., rebentrost p., “quantum principal component analysis.” nature physics 10 (2014): 631-633. lorenc a., hammon o., “objective quality control of observations using bayesian methods. theory, and a practical implementation.” quarterly journal of the royal meteorological society 114 (1988): 515-543. rahim i.a., usli m.n.r., “slope stability study around kampung kuala 107spatial quality control method for surface temperature observations based on multiple elements abai, kota belud, sabah, malaysia.” malaysian journal of geoscience 1 (2017): 38-42. rahmati o., pourghasemi h.r., melesse a.m., “application of gis-based data driven random forest and maximum entropy models for groundwater potential mapping: a case study at mehran region, iran.” catena 137 (2016): 360-372. mi c., et al., “a novel experimental teaching approach for electrical engineering based on semi-physical simulation.” world transactions on engineering and technology education, 12 (2014): 779-783. reek t., doty s.r., owen t.w., “a deterministic approach to the validation of historical daily temperature and precipitation data from the cooperative network.” bulletin of the american meteorological society 73 (1992): 753-762. scornet e., biau g., vert j.-p., “consistency of random forests.” the annals of statistics 43 (2015): 1716-1741. shafer, m.a., fiebrich, c.a., arndt, d.s., fredrickson, s.e., hughes, t.w., “quality assurance procedures in the oklahoma mesonetwork.” journal of atmospheric and oceanic technology 17 (2000): 474-494. shi-wei t., qi-qin z., zhi-fang x., min h. “quality control schemes and its application to automatic surface weather observation system.” plateau meteorology (in chinese) 5 (2009): 029. shlens j., “a tutorial on principal component analysis.” arxiv preprint arxiv (2014): 14041100. wade c.g., “a quality control program for surface mesometeorological data.” journal of atmospheric and oceanic technology 4 (1987): 435-453. xu c.d., wang j.f., hu m.g., li q.x., “estimation of uncertainty in temperature observations made at meteorological stations using a probabilistic spatiotemporal approach*.” journal of applied meteorology and climatology 53 (2014): 1538-1546. xu, z., wang, y., fan, g., “a two-stage quality control method for 2-m temperature observations using biweight means and a progressive eof analysis.” monthly weather review 141 (2013) 798-808. yang l., yan h., xu y., lam j.c. “residential thermal environment in cold climates at high altitudes and building energy use implications.” energy and buildings 62 (2013): 139-145. you l.k., rahim i.a., “application of gsi system for slope stability studies on selected slopes of the crocker formation in kota kinabalu area, sabah.” geological behavior 1 (2017): 10-12. zheng j., yin y., li b., “a new scheme for climate regionalization in china.” acta geographica sinica (in chinese) 65 (2010): 3-12 letter from the editor great challenges come for the world. climatic change, clean energy, petroleum, politic points of view using science for their own benefit. during the last months, the catastrophes have been in the spotlight, not only for those who work on the sciences side to relieve the pain caused or avoid the consequences, but also those states' managers trying to reach political objectives as the earth stills moving, the water continues running through streets and housings; when the forest fires are not completely extinguished, some people is looking for economic or ideology tributes. it's part of our history. from the character rhett butler in gone with the wind, the inspiring american novel, who made a vast fortune with the war, toward the contractors around the world drying public budgets as they explain the "risk management" concept. but science and politic are not going hand-in-hand. recently, the andean region in south america has been the scenery of record-breaking floods and landslides; peru was hit by heavy rains causing damages in the capital city lima and the northern part of the country. in colombia, the same heavy rains caused by el niño climate cycle destroyed two neighborhoods in mocoa, southern city capital of a department, state equivalent. three hundreds of people died. at the same time, bolivia is struggling with the drought consequences of climatic change. country by country in this subcontinent the landscape is similar. a political earthquake was the recent marches celebrated around the globe claiming for respect and support for the science. "we are marching today to remind people everywhere, our lawmakers especially, of the significance of science for our health and prosperity," said in washington bill nye, a science educator with the fame of a rock star. people knows the solution for a better future is in the hands of our politicians; their decisions will allow the development or the constraint of the science. "science, not silence," "oceans are rising, and so are we," "fund science = fun future," "policy without science is a biohazard," "science is not a partisan issue, it is the foundation of our society," "science is not an opinion," were some of the messages exhibited by demonstrators in their marches. humbly, from the earth sciences research journal, we want to adhere the same message: “it’s not magic, it’s science!” we present to you this new volume, is not so so much, but we hope every little bit helps. carlos alberto vargas jiménez. editor in chief, earth sciences research journal l e t t e r f r o m t h e e d it o r earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017) geoffslca colombiana nr 1, octubre de 1992 una solucion al problema inverso en geomagnetismo ramiro l. diaz profesor asociado depto de geociencias-facultad de ciencias-universidad nacional de colombia 53 diaz, r.l.: una solución al problema inverso en geomagnetismo . g eofís. colo mb. 1:53-56 , 1992. issn 0121-2974 resumen un terna importan te en geofísica, es el problema inverso en geomagnetismo. en este trabajo se propone un método para reso lverlo , qu e consiste en hallar la solución de un sistema móvil de ec uacio nes no lineales a lu largo de un perfil magnético. conociendo los datos de ano malías magnéticas y por medio de un programa qu e ejecuta un proceso iterativo efi cient e, se determ ina la estructura bidimensional bajo el perfil muestreado . abstract an ínteresting aspect in geophysics is th e inverse problem in geo magnetism . this paper pro poses a method for so lvin g it. so, th e objective is lo solve a mobi l system of non linear equations along a magnetic profile . from knowed magnetic anoma li es and using an efficient iterativ e process , it is possible lo find a bidimensional stru ctu re below a sample pro file . 1. introduccion encontrar el modelo que mejor interprete la estructura geológica a través de sus manifestaciones magnéticas , equivale a decir que se ha solucionado el problema inverso en geomagnetismo . se prop_one un mét odo para encontrar una solución al problema inverso a partir del modelo propuesto por campbell & hines ( 1981) . el método consiste en solucionar un sistema móvil de ecuaciones no lineales a lo largo de un perfil magnético , en lo posible perpendicular a la estructura. es un proceso iterativo que se repite hasta lograr el modelo que cumpla con las condiciones requeridas (robert & gex, 1985). la aplicación del método a un perfil magnético permite hacer un análisis de la convergencia y de las ventajas que presenta. 2. consideraciones anauticas campbell & hines (1981), proponen el siguiente modelo en interpretación magnética: n : número de vértices del polígono al cual se aproxim a la secc 1on transversal de la fuente magnética t-. t;c: campo magn ético total calcu lado a partir de la estructura magnetizada m = 2 x to (1 -cos2 (1) cos2 (s )] factor que determina las propiedades magnet1cas de la estructura geológica 54 dfaz: una solución al problema inverso en geomagnetlsmo x : contraste de susceptibilidad magnética t 0 : campo geomagnético (campo magnético inductor) 1 : inclinación magnética s : azimut de la estructura pk (zk,z¡.. 1) = sen d[(c-8) sen(u-0) +r cos(u-0)]: factor geométrico del modelo d = tan· 1 [(2:¡.. , -zk)/(xk+, -xk)] c = (ro'2) -tan · 1 (2:¡../xk.,) 8 = (1t/2) -tan·1 (z/xk) u= tan· 1 [tan(l)/ tan(s)l,...-..----,.-. r = ln [ v(z 2 k. 1 + x 2 k. 1)/ {(2:¡. 2 + xk 2 )] el modelo propuesto por campbell & hines (1981 ) , permite calcular la anomalía magnética de la estructura bidimensional sobre un perfil determinado conociendo la posición y geometría de la fuente magnética. el problema así planteado , se denom ina problema d i recto en geomagnetismo. se utiliza el modelo mencionado para dar solución al problema inverso; es decir, conociendo los datos de anomalías magnéticas en un perfil, se pretende determinar la estructura bidimensional. si se tienen n muestras de campo magnético total y se desea encontrar la profundidad de la interfase magnética bajo cada punto de muestreo, se debe pensar en un sistema de n ecuaciones con n incógnitas , lo cual no es conveniente cuando n es muy grande (velleman & hoagiling , 1981 ). para solucionar el problema se propone un sistema móvil de tres ecuaciones que deben recorrer todo el perfil. este proceso se debe repetir hasta que las desviaciones del campo calculado (t:. t;c), con respecto al campo observado en terreno (t:. td.o) , se hagan mínimas. se consi era un sistema de tres ecuaciones q j, o;..,. q . , evaluadas en las estaciones i, 1+ 1, i+2, ~el perfil magnético con tres incógnitas z¡. zi+p z;.,;>· en el recorrido del sistema por el perfil cada punto interviene tres veces en la solución del sistema. cada una de las ecuaciones del sistema se expande en términos de una serie de taylor con el fin de encontrar la relación de recurrencia que se emplea en el proceso iterativo (bruce, 1981) asf: o;= 0°; + (80°/8zj [z; zoj + [80° /8z;. 1] [z;. 1 -zo;. 1] + (80° /8z;.2] [z;.2 -z 0 ;.2] + o, 0;. 1 = 0°;.1 + [80°;./8zj [z;z oj + (80°¡j8z;. 1) (z;. 1 -z 0 ;. 1) + (80°;./8z;. 2] [z;.2 -zo;.2] + o, 0 ;.2 = 0 °;. 2 + [80°;./8zj [z; z oj + (80°¡j8z;. 1) (z;. 1 -z o;. 1) + [80 °;j8z;.2] [z;.2 -z 0 ;. 2] +0, [3] en donde, z; : profundidad real de la estructura zo; : estimador de z; o; : función evaluada en (z;. z;. 1, z;.2) 0 °; : función evaluada en (z"¡, zo;.,, zoi+2) 80 ° /8z; : derivada de la función o;. evaluada en z"¡, zo;.,. z0 ;.2 a, : residuo en la serie de taylor. los valores o;. 0 ;. 1 • 0 ;.2 y 0, deben tender a cero a medida que el campo calculado se aproxima al valor del campo observado ; es decir, el modelo obtenido representa la estructura geológica. bajo estas condiciones el sistema [3] se puede expresar de la siguiente forma: 0 °; = [8qo / 8zj [z 0 ; zj + [80° / 8z;. 1] [z 0 ;. 1 -z;. 1] + [80 ° / 8z;.:j [z 0 ;. 2 -z;. 2) 0°;. 1 = (80°;./8zj [z 0 ; zj + [80°;./8z;. 1) [z 0 ;. 1 -z;. 1) + [80 ";. ,18z;.2) [z 0 ;. 2 -z;.2l 0°;. 2 = [80 °;./8zj [z 0 ; zj + [80°;j8z;. 1) (z"¡., -z;. 1] + [80°;./8z;.2) [z 0 ;.2 -z,. 2) [4] geoffslca colombiana ~ 1, octubre de 1992 a o. 1 de modo que: a o = j (zoz) z = zoj' l ao [5] z : matriz cuyos elementos son los valores que más se aproximan a los z¡ reales zo : matriz que contiene los estimadores de z; ao : matriz de desviaciones j : jacobiano del sistema el proceso se realiza utilizando un programa cuyo algoritmo se plantea en la ecuación de recurrencia [5]. se inicia con una primera aproximación a la estructura geologica y se ajusta el modelo, mediante aproximaciones sucesivas, hasta obtener la estructura que mejor satisface el muestreo magnético del campo total. en el programa se usó una tolerancia de 1 0.1 1 para los 0 ;. 3. convergencia y exactitud del meto do se considera el perfil magnético cuyos datos se consignan en la tabla 1, con origen en el punto a. observando los resultados de los cálculos suministrados por el programa (tablas 2 a 5), se infiere que, a partir de una burda aproximación, en más o menos cinco iteraciones se consigue un modelo ajustado a la estructura geológica, lo cual garantiza su eficiencia. los puntos 1 y 8 son de amarre en el modelo. la topografía se mide como una altura sobre el nivel del mar y los valores de z son profundidades bajo el nivel del mar. 55 tabla 1 datos del perfil magnético estac anomalia(y) di<t. horiz(m) topogr(m) 1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 a 235.7 b -33.4 e -44. 1 o -111 .0 e -64.6 f -66.8 g -52.4 h 288.8 0.0 400.0 1000.0 1600.0 2200.0 3000.0 3800.0 4800.0 tabla 2 datos de la 1 iteración z;' (m) ll t;" (y) o; (y) 0. 1 870.0 -45.6 6.8 860.0 -62. 1 4.7 830.0 -62.9 1.6 800.0 -110.7 0.2 840.0 -43.1 0.9 850.0 -25.1 8 .2 0. 1 tabla 3 datos de la 11 iteración z;' (m) llt;" (y) o; (y) 0.1 927.5 -54.2 -1 .8 1408.4 -66.7 0.1 1162.2 -64.5 0.1 764 .0 110.7 0.2 115.3 -44. 1 0.0 495.9 -33.4 0.0 0.1 0. 1 50.0 40.0 0 . 1 30.0 60. 1 70.1 0 . 1 z; (m) 0. 1 927.5 1408.4 1162.2 764 .0 775.3 495.9 0 . 1 z; (m) 0. 1 917.6 1328. 0 1106.1 690.8 758.7 500.5 0. 1 56 díaz: una solución al problema inverso en geomagnetísmo z;' (m) 1 0.1 2 917.6 3 1328.0 4 1106. 1 5 690.8 6 756.7 7 500.5 8 0.1 tabla 4 4. conclusiones datos de la 111 iteración !l t;" (y) -53.4 -66.8 -64.6 -111 .0 -44. 1 -33.4 tabla 5 o; (y) -1 . 1 0.0 0.0 0.0 0.0 0.0 z; (m) 0.1 900.6 1297.3 1106.7 690.3 753.3 499. 1 0.1 en este trabajo se presenta una solución al problema inverso en geomagnetismo . el algoritmo que se emplea ofrece eficiencia y facilidad para manejar grandes cantidades de información, con un método de análisis numérico sencillo . referencias datos de la v iteración cai!ipbell, d.l. & hines. (1981 ): manua l of geophysical hand calculator programs . seg , usa. z;' (m) 1 0. 1 2 900.7 -52.4 3 1296.4 -66.8 4 1105.5 -64.6 5 690.4 -111 .0 6 753 .2 -44.1 7 499.0 -33.4 8 0. 1 , o; (y) -0.1 0.0 0.0 0.0 0.0 -0.1 z; (m) 0. 1 900.5 1296.4 1105.4 690.4 753.2 499.0 0. 1 bruce, e. (1981 ): introducción al análisis numérico con fortran . univ. ped. tec . colomb., tunja. robert, a. & p. gex. (1985) : inter_prétation geophysique rapide . logiciel ba sic pour micro-<>rdinateur. bull etin de l'lnstitut de géophysique, n26. université de lausanne . velleman, p.f. & d.c. hoaglling. (1981) : applications basics and computin_g of exploratory data analysis . duxbúry fyress, boston , massachuset1s. palabras clave: gas de lutita; exploración y desarrollo; problemas; contramedidas. how to cite item chen, x., liu, z., liang, x., zhang, z., zhang, t., zhu, h., & lang, j. (2018). the shale gas revolution in china—problems and countermeasures. earth sciences research journal, 22(3), 215-221. doi: https://doi.org/10.15446/esrj.v22n3.74390 shale gas development throughout the world has resulted in a revolution in the field of global energy and has also become an important topic in china in recent years. while organic-rich shale is widely distributed in china and the initial commercialization of shale gas has been achieved, the research, exploration, and development of shale gas remain at an early stage. problems exist with crucial technologies, innovation, institutional mechanisms, environmental protection, and other aspects of the industry. the shale gas exploration and development industry in china can learn from the experiences of other countries and strengthen its position in the market, with the support of new government policy. given its unique geological conditions, china should speed up the introduction of technical innovation and establish its unique systems and methods for shale gas exploration and development. abstract keywords: shale gas; exploration and development; problems; countermeasures. the shale gas revolution in china—problems and countermeasures la revolución del gas de lutita en china-problemas y contramedidas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n3.74390 el desarrollo del gas de lutita en el mundo ha significado una revolución en el campo de la energía; en china, durante los útlimos años se ha convertido en un tema importante. las lutitas, ricas organicámente, están ampliamente distribuidas en china y la comercilización inicial de gas de lutita ha sido lograda; sin embargo, la investigación, exploración y desarrollo de este producto permanence en etapas prematuras. los problemas se presentan en las tecnologías clave, la innovación, los mecanismos institucionales, la protección ambiental y otros aspectos de la industria. la exploración del gas de lutita y el desarrollo industrial en china se pueden mejorar a partir de las experiencias de otros países y con su fortalecimiento en el mercado, acompañado del apoyo de políticas gubernamentales. gracias a sus condiciones geológicas únicas, china puede acelerar la introducción de innovación técnica y establecer sus sistemas y métodos únicos para la exploración y desarrollo del gas de lutita. resumen record manuscript received: 14/11/2017 accepted for publication: 26/03/2018 earth sciences research journal earth sci. res. j. vol. 22, no. 3 (september, 2018): 215-221 xiaohui chen1, zhicheng liu2,*, xing liang3, zhao zhang3, tingshan zhang1,2, haihua zhu1, jun lang1 1school of geoscience and technology, southwest petroleum university, chengdu 610500, china 2sichuan institute of land planning and surveying, chengdu 610045, china 3petrochina, zhejiang oil field company, hangzhou, 310013, china *email of the corresponding author: lzc_15@126.com p e t r o l e u m g e o l o g y 216 xiaohui chen, zhicheng liu, xing liang, zhao zhang, tingshan zhang, haihua zhu & jun lang introduction the successful commercial exploration for shale gas in the united states led to an energy revolution, which ended america’s dependence on imported energy and made the us an energy-rich country. this situation brought the shale gas industry under the spotlight throughout the entire world. while china is also aiming to develop its shale gas reserves, it faces challenges with technology and marketing. the international community generally acknowledged that shale gas exploitation revolutionized the global energy landscape, which not only increased the output of natural gas but also enormously influenced the global gas market, global energy supply structure, global climate change policy and even geopolitics. in recent years, shale gas exploration and development has become an important topic in china. organic-rich shales under the right conditions for gas accumulation are widely distributed in china, including marine shales in southern china, northern china, and the tarim basin of xinjiang, and lacustrine shales in north of china, the junggar basin, turpan-hami basin, ordos basin, bohai bay basin, and songliao basin, which represent a considerable resource potential (eia, 2010;zou., et al,2016) it is predicted that the recoverable resources of shale gas are about 22 × 1012 m3 (ministry of land and resources of people's republic of china, 2016), which exceeds the conventional natural gas resources of china. at present, although commercial production has been rapidly achieved in three districts of the sichuan basin, china is still in the early stages of evaluation, exploration, and development of shale gas and key technologies are still not well integrated. large-scale commercialization is challenging due to the commonly complex geological structure and the high cost of drilling and fracturing services. to overcome the bottlenecks in technique, marketing and industrialization that are restricting shale gas exploration and development in china would be necessary the following actions. 1. increasing basic theoretical research. 2. developing shale gas theory by introducing innovative thinking and methodology. 3. improving exploration and development technology as well as relevant environmental technology; 4. establishing a sophysticated marketing system with better relations between industry and local government. 2. the status of shale gas exploration in china the united states has greatly benefited from the development of shale gas and other unconventional petroleum resources. in 2010, the united states had shale gas production of more than 1379 × 108 m3, accounting for about 23% of the national output, and has now surpassed russia to become the world’s largest gas producer (eia, 2010). the united states shale gas revolution has reshaped the global liquefied natural gas market, and its influence will become even more remarkable over time. the enormous success of shale gas development and exploration has attracted considerable attention from governments and energy companies all over the world and set off a boom in exploration for shale gas. it has categorically changed the global energy picture. with the rapid development of the national economy in china, the demand for energy, and exceptionally clean energy continues to increase. according to the ministry of land and resources (mlr) of china, the country’s gas shortage will exceed 1350 × 108 m3 by 2020. therefore, active, systematic exploitation of china’s unconventional gas resources is a significant way to meet the country’s future energy demands. the exploration and development of global non-renewable energy resources show that shale gas is one of the realistic replacements for conventional petroleum. the twelfth five-year plan for the national economy and social development in china stated explicitly “to promote the exploitation and utilization of shale gas and other unconventional petroleum.” on march 16th, 2012, the national energy administration issued the “shale gas development plan (2011–2015)” in which the government requested emphasis on the following strategies: 1. researching into shale reservoir geology and key accumulation factors for shale gas; 2. establishment of the evaluation methods and criteria for shale gas target optimization, based on sedimentary facies, tectonic evolution, burial depth, organic matter content (toc), thermal maturity and gas content. it aims to increase the annual shale gas production rate from a current level of 13 × 108 m3 up to 300 × 108 m3 per year by 2020. china has achieved substantial progress regarding the exploitation of shale gas in recent times. petrochina identified weiyuan, changning, zhaotong, and fushun–yongchuan as the four most favorable districts for shale gas exploration in south sichuan and north yunnan, and more than 20 wells were put into production (figure 1). sinopec (china national offshore oil corporation) has completed several appraisal wells in eastern guizhou, anhui, northeast sichuan and the chongqing fuling area, and the jiaoye 1hf well in the fuling district produced a high-yield commercial shale gas flow of approximately 20.3 × 104 m3 per day. cnooc (china national offshore oil corporation) has carried out pilot work for the exploration of shale gas in the anhui and zhejiang areas, and yanchang petroleum has discovered lacustrine shale gas at yanan in the shanxi province. at present, shale gas reserves in china are estimated to be nearly 500 billion m3 with an annual production rate of approximately 1.5 billion m3. according to the mlr, up until the end of july 2015, cnpc, sinopec, and yanchang petroleum had cumulatively produced 680 million m3 of shale gas. by the end of 2015, the cumulative production reached 6.5 billion m3 (dong et al., 2016). this progress shows that china has great potential when it comes to shale gas exploration. however, although shale gas exploration and research has made some advances, the exploration and development of shale gas in china still has much further potential. compared with the united states, the commercialization of shale gas in china has been challenging, due to the more complex geology, the presence of older strata, more complex tectonism, more complicated thermal histories, and higher levels of thermal maturity. considerable difficulties in specific fields need to be overcome if future shale gas exploration is to be successful, including understanding accumulation mechanisms, laboratory testing, evaluation and forecasting of sweet spots, productivity analysis and prediction, hydraulic fracturing technology, geological steering, and reservoir stimulation. in recent years, china has intensified its efforts into shale gas exploration and development, and some examples of its industry’s promotion are as follows: figure 1. the status of shale gas exploration in china (after dong et al., 2016). 217the shale gas revolution in china—problems and countermeasures (1) investment in researching and development has been increased, and technical systems were formed and improved with the support of the national energy administration, the research institute of petroleum exploration & development of petrochina set up a national energy shale gas r&d center to address the critical technical problems related to shale gas exploration. a specific sub-project for research into the key technologies of shale gas exploration and development has been set up from the national key science and technology special project “large oil–gas fields and coalbed methane development.” also, major oil corporations, as well as local governments, such as sichuan and chongqing, have set up research institutions and laboratories to research the evaluation of gas reservoir potential and the selection of advantageous exploration areas. 2) a diversified investment and multi-type enterprise participation pattern has emerged the state council of china has approved shale gas as being the 172nd independent mineral in china, which means that the oil and gas franchise will no longer restrict shale gas exploration and development, and any companies who have the financial ability and qualifications for gas exploration can bid for it. in the field of natural gas prospecting, only petrochina, sinopec, cnooc and yanchang petroleum had chinese government approval and were eligible for gas prospecting in the past. on september 10th 2012, in the second round of public bidding for shale gas exploration in china, the ministry of land and resources demanded that those companies qualified to explore for oil and gas or gas minerals, had registered capital of more than 300 million rmb, no matter their domestic enterprise or independent legal entity (ministry of land and resources of the people's republic of china, 2012). many private enterprises who had planned on investing in shale gas in the past had been blocked previously because of the requirement for registered capital. on the 30th october 2013, the national energy board issued the “shale gas industry policy”, which removed the registered capital limitation. it only made requests for investment capability, moderate financial status and the ability to be responsible for civil liability, which created new opportunities for private enterprises who were willing to invest in the shale gas industry. the ministry of land and resources has held two open tenders for exploration permits, and private enterprises were welcomed. in the second round of public bidding, through state-owned enterprises won most of the bids for the exploration rights, about one-third of the original 83 bids were from private enterprises, and they won 2 blocks out of the 19 up for bid (ministry of land and resources of the people's republic of china, 2013). this results marked a significant breakthrough in the bidding mechanism for oil and gas exploration rights in china. although foreign firms are not allowed to participate in oil and gas exploration and development directly, the chinese government encourages them to take part in joint ventures. it shows a desire by the chinese government and local enterprises in obtaining specialized foreign technologies and international development experience. 3) a policy system for industrial shale gas development has been formed the central government is a strong supporter of the development of shale gas and lists the shale gas industry as an industry of growing national strategic importance (national energy of the people's republic of china, 2013). to encourage enterprises to accelerate the commercialization of shale gas, the ministry of land and resources, the national development and reform commission, the ministry of finance, the ministry of commerce, the ministry of science and technology, the environmental protection bureau, the central ministries and commissions and other national ministries have launched a series of detailed policy measures for promoting shale gas exploration and development, encouraging foreign investment, increasing financial subsidies, guiding industry development, constructing a demonstration zone, and supporting science and technology research (china geological survey, 2014). at the central government level, the development and reform commission is responsible for formulating general policy and supervises the price of natural gas. the national energy administration is competent for setting up shale gas production goals. the ministry of land and resources is accountable for shale gas mineral exploration and development rights as well as for organizing and implementing the shale gas tendering process for exploration blocks. the ministry of finance is responsible for shale gas exploration subsidies. the science and technology department is qualified for funding support for shale gas related technology research and development. the environment protection bureau is responsible for monitoring and preventing environmental pollution. the ministry of commerce is responsible for examining chinese–international cooperative projects. the state administration of work safety is responsible for monitoring work safety at sites of production. the policies of central government related to shale gas development are mainly regarding to the setting of production goals, calculation of government subsidies, administration of taxes and fees, handling of market access, mineral classification of shale gas, and publishing of the shale gas industry standard (table 1). annual shale gas production in china in 2017 reached 91 × 108 m3 by 2017. the chinese government planning target is for 30 × 108 m3 by 2020 and requests petrochina and sinopec to take a series of measures to ensure they achieve these goals on time. the central government used to subsidize shale gas development enterprises, paying 0.4 rmb per m3 of gas produced. these subsidies expired in 2015, but the ministry of finance has issued a new subsidy policy and will now pay 0.3 rmb per m3 of gas produced from 2016–2018. the subsidy will drop to 0.2 rmb per m3 of gas produced from 2019–2020. the ministry of land and resources lists shale gas as the 172nd independent mineral. the decision has far-reaching significance and gives other private enterprises an opportunity to invest in shale gas development. in august 2014, the sichuan provincial government classified the shale gas industry as being one of the top five highly advanced sectors of the future. the current shale gas policy of china is similar to that of the united states, during the early development stage of shale gas in that country. china has set up procedures for government subsidy, arranged financial support for technology research and development, removed shale gas price control, opened a natural gas pipeline network system, and made other advancements. however, shale gas development in china is still dominated by large stateowned enterprises, and in a situation which differs from the united states, united kingdom, and other western countries, as these entities are not required to release geological information or production data. 3. problems affecting the development of china’s shale gas industry although china’s shale gas industry has achieved initial success, there are still some problems relating to the areas of critical technology, innovation, environment, institutional mechanisms, environmental protection and other aspects which are discussed below. 3.1 a technical system for exploration and development adapted to china’s geological characteristics has not been established although some achievements relating to the exploration and development of shale gas have been made in china, a systematic investigation and evaluation of the national shale gas resources have not yet been completed. the theory of shale gas reservoir geology still needs to be studied in more detail. significantly, more work is required for the study of the accumulation of shale gas, field of sweet spot optimization methods, evaluation of productivity, dynamic mechanisms, characterization of gas flow through micro–nanoscale pore systems, complex multi-field seepage coupling theory, as well as other aspects of this science. 218 xiaohui chen, zhicheng liu, xing liang, zhao zhang, tingshan zhang, haihua zhu & jun lang the rapid development of the united states shale gas industry occurred mainly due to breakthroughs in large-scale horizontal drilling and massive multi-stage hydraulic stimulation, which significantly reduced the cost of exploiting shale gas, and made resources economically viable upon they were initially worthless. china’s service companies are just beginning to acquire these two key technologies, which are partly still only mastered by foreign enterprises. the experience of shale gas development in the united states shows that different mining technology is required in different regions, as reservoir conditions change from place to place. as a latecomer to the shale gas development industry, the number of drilled wells is relatively limited in china and the technical knowledge for drilling and producing needs more time to develop. similarly, equipment and industry standards are still in the process of being developed and implemented. due to this underdeveloped technical environment, only three state-owned oil companies have currently fully mastered the latest technology. new enterprises that are starting to enter the field of shale gas development are finding technical services to be deficient. also, china’s shale gas reservoirs are different from those in the united states. shale gas is currently commercially produced in the sichuan basin from marine shales. the producing zone is the lower silurian longmaxi formation, which is older than shales in the united states. it has had a long geological history and has been transformed into an “over mature, table 1. the policies of the central government to promote shale gas development timing policy name agency main content valid time 2011 outline of the 12th five-year plan for economic and social development. sc. promote shale gas development and utilization of multiple clean energy development planning. 2011–2015 2011 catalog of industry adjustment guidance. ndrc list of shale gas exploration and development projects to be encouraged. long-term 2011 chengdu–chongqing economic zone regional planning. ndrc speed up sichuan shale gas exploration and development long-term 2011, 2015 catalog for the guidance of foreign investment. ndrc. moc list of shale gas exploration and development projects (joint ventures) to be encouraged. long-term 2012 the 12th five-year national strategic emerging industry development planning. sc. class shale gas industry as a critical developing direction and primary tasks. 2012–2015 2012 shale gas development plan (2011–2015). ndrc. mof. molr. nea. clear objectives, tasks, and measures of the shale gas industry. 2012–2015 2012 “12th five-year” energy conservation and environmental protection industry development plan. sc. improve the level of overall utilization of shale gas in critical areas. 2012–2015 2012 energy conservation and emissions reduction “twelfth five-year” plan. sc. promote shale gas development and utilization as the primary task of adjusting and optimizing the energy structure. 2012–2015 2012 notice about strengthening shale gas resources, prospecting, and exploitation, supervision, and administration of relevant work. molr the management of shale gas exploration, mining and the administration of exemptions from mineral resources compensation fees and mineral rights use charges. 2012–2017 2012 opinions about further encouragement and a guide to the private capital investment in the field of land and resources. molr. acfic encourage and guide private capital to participate in shale gas resource prospecting and exploitation. long-term 2012, 2015 notice about shale gas development and utilization of subsidy policy. mof. shale gas definitions and subsidies standards. 2012–2020 2012 advice on encouraging and guiding private capital to further expand the implementation of energy sector investment. nea. list of shale gas exploration developments within the scope of private capital investment. long-term 2012 enforcement advice on utilizing price level to encourage and guide the development of private investment. ndrc. shale gas production price should be determined by mutual consent so that private capital is encouraged into the shale gas field of production. long-term 2013 the shale gas industry policy. nea. shale gas development should be financially supported so that the development of shale gas exploration may be increased. long-term 2013 the announcement about oil and gas enterprises, development of cbm and shale gas vat related issues. sat. provision of a vat tax rate for shale gas production of 17% that is applicable for the development of oil and gas. long-term 2014 the strategy of energy development action plan (2014–2020). sc. aim to enhance shale gas production up to 300 billion m3 by 2020. 2014–2020 2014 open regulatory measures for oil and gas pipeline network facilities (trial). nea promote the oil and gas pipe network facilities. improve the utilization efficiency of pipe network facilities. regulate the behavior of oil and gas pipe network facilities and relevant open markets. 2014–2019 2015 the announcement about the urban land used tax policy for oil and gas production enterprises. mof. sat. oil and gas production construction. suspend leaving urban land use tax. long-term ndrc: national development and reform commission; mof: the ministry of finance. nea: the national energy administration. moc: the ministry of commerce. sat: state administration of taxation. acfic: the national federation of industry and commerce. molr: the ministry of land and resources. sc: the state council. 219the shale gas revolution in china—problems and countermeasures highly reworked” shale. the present burial depth of the shale in the sichuan basin ranges from 2000–4000 m, which is more profound than that in the united states, which is mainly at depths of 800–2600 m furthermore, the mountainous topography of the sichuan shale gas areas poses challenges for ground engineering. although commercialization of shale gas at burial depths of around 2000 m is possible, the drilling, completing and fracturing technologies for depths of 3000–4000 m have not yet been developed. 3.2 the cost barrier for shale gas industrialization based on the american experience, the central controlling factors for the development of the shale gas industry are resource abundance, technology, government policy, and input-output relationships. the rapid expansion of shale gas in the united states benefited from national policy support, an integrated technical system and a well-developed infrastructure (pipe-network); conditions which helped lower cost. the cost of shale gas development remains high in china. immature technology, an imperfect industry chain (which has resulted in a shortage of equipment), and an undeveloped pipe-network have stalled progress within the shale gas industry of china. shale gas development is still at an early stage, and low rates of production are making increases in scale challenging to achieve. compared with other energy sources, shale gas has no advantage when it comes to price. although the government has issued a subsidy policy of 0.4 rmb/m3 for shale gas, it takes a long time to commercialize shale gas projects fully, and most companies have still not yet qualified for the government subsidy. most shale gas districts in china are located in mountainous areas where transportation is difficult, and this further increases the cost of exploration and development. 3.3 a non-competitive environment for industrial development the united states shale gas industry benefited not only from improvements in technology and innovation but also from its mature market system. in contrast, china's shale gas industry still faces problems with the market. a dynamic, competitive market has still not been established. large state-owned enterprises have dominated china's petroleum industry for a long time, and it is mostly a closed market. only the three major state-owned oil companies have experience in exploration and development. moreover, shale gas blocks overlap with the traditional oil and gas blocks and, as a result, the three major oil companies possess most of the shale gas blocks. it has meant a monopoly when it comes to exploration rights. besides, local governments in places such as chongqing, guizhou and sichuan province have begun to set up energy investment companies for shale gas exploration, and this will form local monopolies. as an essential support facility for the development of the shale gas industry, china’s oil and gas pipeline network is also a monopoly and its distribution is not suitable for shale gas transportation because the current shale gas production wells are mostly in remote areas, over which the existing pipeline network does not extend. a new regional pipeline network needs to be built. also, the absence of sharing of shale gas data and technology between these companies has greatly hindered the progress of the industry and technology. the technical services for shale gas are still in their infancy. although some new specialized service companies for shale gas have started up in recent years, the majority of them lack vital technologies and experienced personnel. many of the new shale gas exploration and development enterprises have little experience in the oil and natural gas exploration and development. in the second round of bidding for china’s shale gas blocks in 2012, for example, four of the companies that won bids had only been established three months earlier. as such, the lack of experienced technical and management personnel makes shale gas exploration difficult. development of the shale gas industry in the united states showed that while many companies may begin shale gas exploration, only the fittest survive, which inevitably leads to companies entering and leaving the industry. with the introduction of relevant policies in china, a competitive environment for the shale gas industry will emerge. shale gas exploration and development needs a lot of capital investment in the early stages and has an extended investment recovery period, which means that profits are difficult to realize in the short-term. therefore, financial obstacles are an essential issue for some small and medium enterprises. the shale gas industry is a highinvestment, high-risk sector and development of china’s shale gas industry will be affected without a good investment and financial basis. 3.4 environmental problems in the development of shale gas due to water resource limitations, and even though china possesses the world's largest shale gas reserves, massive exploitation without suitable water protection will be disastrous. according to the united states environmental protection agency (epa), the average water consumption of a single shale gas well in 2010 was 0.76 × 104 to 2.39 × 104 t (depending on the depth, horizontal section length and fracturing scale), of which 20%–85% remained underground (epa, 2012). i the united states, the average water requirements for drilling each well is 200–2500 m3, and the water demand for hydraulic fracturing each well is as high as 7000–23000 m3 (vengosh, 2014). united states counterparts have noted that water protection and conservation is a huge issue for the industry. in the united states shale gas drilling industry, about 70% of the water used comes from hydraulic fracturing water recovery. although china's total water resources are 2.8124 × 1012 m3, accounting for almost 6.7% of total global freshwater resources, the actual amount of water available is only 0.8 × 1012 m3 (wang, 2010). in total water resources, china ranks sixth in the world, after brazil (6.95 × 1012 m3), russia (6.54 ×1012 m3), united states (3.056 × 1012 m3), canada (2.9114 × 1012 m3), and indonesia (2.8113 × 1012 m3), according to wang (2011). however, according to the ranking released by the world water development report (2012) on water resources in 180 countries and regions, china ranked 128th in the world, having only 2260 m3 of water per person/per year, which is just one quarter of the world's average, and only half of that in japan, one quarter of that in the united states, 1/12 of that in russia, and 1/58 of that in canada. according to data from the united states epa, for each horizontal shale gas well, the water used for fracturing accounts for 85–98% of the total water consumption for the well. shale gas wells in different plays have different water consumptions. for example, in the marcellus shale gas field, the average water consumption for a well is about 1.5 × 104 m3, in the barnett shale the average water consumption is about 1 × 104 m3, in the fayetteville shale the average water consumption is about 1.2 × 104 m3, and in the haynesville shale it is about 1.4 × 104 m3 (rahm, 2011). compared with the united states, china’s shale is generally more deeply buried, especially in the changning–weiyuan district where it is usually found at a depth of 2600–3000 m. the reservoir thickness is 20–800 m, which also means that the demand for water for hydraulic fracturing will be more significant. at present, there are no official figures for water volumes used during the drilling of shale gas wells in china, but according to investigations made by researchers, the drilling and fracturing of each shale gas well will consume approximately 25000 m3 of water on average in china, which is far higher than the average value (15000 m3) in the united states (lu, et al., 2014). the data from the weiyuan shale gas district support these findings. water consumption in the wei201-h1 well for a 10-level hydraulic fracturing job reached 22000 m3, and the total water consumption for the entire well was approximately 25000 m3. although water is relatively abundant in the sichuan province (a vital shale gas production region), its distribution is uneven, resulting in regional and seasonal water shortages. the water resources of the sichuan province are principally located in the ganzi, aba and liangshan prefectures, accounting for 50% of the total water 220 xiaohui chen, zhicheng liu, xing liang, zhao zhang, tingshan zhang, haihua zhu & jun lang resource. per capita, the supply there reached 16993 m3 in 2015. however, within the sichuan basin, the most water-poor area of the entire province, the per capita water resource is only 1624 m3 (sichuan provincial water resource department, 2015). agricultural water consumption accounts for more than 60% of the total water use in the sichuan province, and there has been a water shortage there for a long time. several factors contribute to the agricultural water shortage in the sichuan province: (1) the geographical distribution of the water resource and rainfall is uneven, which leads to severe water shortages within the sichuan basin, and water abundance in the mountain plateau areas of western sichuan. (2) historical issues and inadequate investment in the whole province, but especially the basin and hilly regions, has resulted in a shortage of irrigation systems, making water deployment difficult. the efficiency of water use is low. (3) the utilization rate of water for agriculture is low. due to a lack of water management, water wastage is common. in most areas of the province, canal water utilization efficiency is less than 50%, and generally over half of the water cannot be used. (4) the agricultural water resource is commonly used for other activities, due to its lower value when used in the agriculture industry. local governments regularly use part of agrarian water for other industrial activities. the national demonstration shale gas areas of weiyuan, changning, and huangjinba are located in the weiyuan, changning, gong, and xingwen counties, and all suffer long-term, or seasonal, water shortages. for example, weiyuan has had an annual precipitation of around 1000 mm for several years. total accumulated precipitation is 15 × 108 m3, the gross water resource is 7.4227 × 108 m3, and the per capita water resource is just 994 m3, resulting in severe water shortages. agricultural water consumption is more than 1 × 108 m3, accounting for 51% of the total water consumption. the standard water resource of the changning, gong, and xingwen counties is around 10 × 108 m3, the per capita water resource is just 2521 m3, the total agricultural water consumption is less than 3 × 108 m3, and there is a severe water shortage. the national demonstration shale gas areas of the south sichuan basin are all in the main rice cultivation areas, with a rice planting area of approximately 680 km2. the rice planting areas of changning, gong, juanlian, and xingwen counties are around 451.72 km2 in size and about 228.06 km2 in weiyuan. fu et al. (2014) calculated that 1177 m3 of water per year were needed to produce 1000 kg of rice. there are more than 160 horizontal shale gas wells drilled in the sichuan, yibin and weiyuan regions. assuming that the water consumption of each well is 2.5 × 104 m3, the total water consumption during 2014–2015 would have been at least 4 × 106 m3, which is equivalent to the irrigation water used for 26.7 km2 of paddy fields for one year, and therefore accounts for 4% of the five counties’ annual water consumption. taking the total number of wells that will be put into production (about 353) for these three shale plays, it is possible to calculate that the annual water consumption will reach 8.8 × 106 m3, which would be 9% of the five counties’ annual irrigation water consumption. according to the plans made by the local government of the sichuan province, 1000 wells will be added, every year, over the next few years. as a result, the water consumption of shale gas development will increase by 2.5 × 107 m3 every year over the near future, which is equivalent to the irrigation water consumption of an additional 166.7 km2 of rice growing area every year. therefore, in line with the exponential increase in the number of shale gas wells, there will be an exponential increase in water consumption. the rice growing industry, which demands a great deal of water, will suffer a severe water shortage. this conflict between agriculture and the shale gas industry, and who is going to use the dwindling water supply, will become more severe over the next few years. the fracturing fluid, including the fracturing water and a small amount of formation water from the shale, will gradually flow back into the ground after the massive multi-stage hydraulic stimulations of the shale gas wells. shale gas well fracturing flowback fluid (mainly fracturing fluid) differs from conventional fracturing fluid, in that it contains abundant chemical additives used in the hydraulic stimulation process, and poses a considerable risk for the environment if not carefully handled. how to deal with this massive flowback drainage is critical for the successful development of the shale gas industry in china. about 30%–70% of the high-pressure injection fluid flows back to the ground after the hydraulic stimulation of a shale gas well in the united states (united states environmental protection agency, 2012). measures taken to process these flowback fluids include 1. reinjecting them underground using re-injection wells; 2. reusing them for hydraulic stimulations after simple treatment; 3. treating them into fresh water. the abundant chemical additives make the processing of flowback fluid difficult and costly. at present, treatment equipment or factories for the processing of flowback fluid have not been set up in sichuan shale gas production areas. all of the flowback fluid is transported by truck and injected back underground using re-injection wells to reduce the cost. this process prevents the reuse of flowback fluid and increases the overall water consumption of the shale gas developments. it is essential to establish processing facilities for flowback water, so that the water may be reused. currently, the united states has taken measures to encourage the purification and recycling of flowback water. for example, the reused flowback fluid ratio of the marcellus shale gas region has increased from less than 10% in 2008 to more than 70% in 2011, which has greatly reduced the water consumption (rahm et al., 2013) nevertheless, the new york environmental protection ministry recently announced the prohibition of hydraulic fracturing operations in new york state, after a seven-year investigation. it considers hydraulic fracturing to be a significant risk to land, air and water resources, as well as posing a potential public health risk, which may lead to a negative perception that is impossible to remedy. 4. countermeasures for shale gas industrialization in china shale gas industrialization in china should follow the successful model of the united states, taking into account the differences that exist in china. 4.1 improvements in exploration and development technology advanced technology is the key to shale gas development in china. with the rapid economic growth of our country and the additional demands for energy, energy structure adjustment, energy security, energy savings and other factors, it is necessary for shale gas industrialization of our country to speed up. the shale gas development plan (2011–2015) specified the following goals to be achieved by 2015: 1. completion of an assessment of the potential and distribution of shale gas resources; 2. prioritization of prospective areas and sweet spots, establishing several shale gas development zones and achieving commercial production; 3. conducting international cooperation with regards to shale gas technology and increasing research investment. shale gas exploration represents the basis for the rapid development of the shale gas industry in china. with the support of the national policy for the shale gas industry, future work should focus on the geological research of the “over mature, high reworked” shale and accurately assess the potential and distribution of shale gas in china, especially in the sichuan, chongqing, guizhou, hunan, hubei, yunnan, jiangxi, anhui, jiangsu, shanxi, henan, liaoning and xinjiang provinces, and to establish new shale gas development zones. 221the shale gas revolution in china—problems and countermeasures acknowledgments this work was funded by the young scholars development fund of southwest petroleum university (swpu) (201499010046), china postdoctoral science foundation (2015m580797), national natural science foundation of china(41772150) and national science and technology major project (2017zx05063002-009). references china geological survey (2014). china shale gas survey report. http://www. cgs.gov.cn/upload/201506/20150610/20150624093321310.pdf dong, d., wang, y., li, x., zou, c., guan, q., ... qiu, z. (2016). breakthrough and prospect of shale gas exploration and development in china. natural gas industry b., 3(1), 12-26. eia us energy information administration (2010). natural gas gross withdrawals and production. https://www.eia.gov/dnav/ng/ng_prod_ sum_dc_nus_mmcf_m.htm fu, n., liu, x. g., zhang, y., & yang, q. l. (2014). variation pattern of rice irrigation water requirement in southwest of china. chinese journal of ecology, 33(7), 1895–1901. lu, j., gao, w., & liu, x. (2014). environmental influence in shale gas development and considerations. natural gas & oil, 1, 35-43. ministry of land and resources of the people’s republic of china (2012). shale gas prospecting right tender notice. http://www.mlr.gov.cn/ zwgk/zytz/201209/t20120910_1139187.htm ministry of land and resources of the people’s republic of china (2013). the second round of shale gas exploration tender results. http://www.mlr. gov.cn/xwdt/jrxw/201301/t20130122_1177283.htm ministry of land and resources of the people’s republic of china (2016). national oil and gas resources evaluation. http://www.sinooilgas.com/ gzdt/gzdt/bc4de9c456bf3f420156c5b7c9d7001a.shtml national energy of the people’s republic of china (2013). shale gas industry policy. http://www.gov.cn/gongbao/content/2013/content_2547157. htm rahm, b. g., bates, j. t., bertoia, l. r., galford, a. e., yoxtheimer, d. a., & riha, s. j. (2013). wastewater management and marcellus shale gas development: trends, drivers, and planning implications. journal of environmental management, 120, 105-113. rahm, d. (2011). regulating hydraulic fracturing in shale gas plays: the case of texas. energy policy, 39(5), 2974-2981. sichuan provincial water resource department (2015). water resources bulletin. http://www.scwater.gov.cn/main/wqhg/ggl41/80279/index. html u.s. environmental protection agency (2012). study of the potential impacts of hydraulic fracturing on drinking water resources. https://www.epa. gov/hfstudy/study-potential-impacts-hydraulic-fracturing-drinkingwater-resources-progress-report-0 vengosh, a., jackson, r. b., & warner, n. (2014). a critical review of the risks to water resources from unconventional shale gas development and hydraulic fracturing in the united states. environmental science & technology, 48(15), 8334-8348. wang, z. f. (2011). present situation and water resources utilization sustainable utilization countermeasure in china. journal of xi’an university of posts and telecommunications, 16, 122-127. wang, z. q.(2010). studies on the water resource pollution, prevention, and control industry of our nation. china university of geosciences (beijing). zou, c. n., dong, d. z.., wang,y. m., et al., 2015. shale gas in china: characteristics, challenges and prospects (i). petroleum exploration and development, 42(6), 753-767. 4.2 reinforcing the market and reducing the input costs the exploration for shale gas in our country is at an early stage. investment cannot yet reach the well-developed levels seen in the united states, at least in the short term. however, once china’s service companies acquire technology for large-scale horizontal drilling and massive multi-stage hydraulic stimulation, the process of shale gas industrialization will accelerate, and the costs will reduce. this process will result in enormous demand for specialized equipment from domestic suppliers, providing the equipment manufacturers with opportunities for rapid development. gradual improvements in the structure of the industry will stimulate shale gas industrialization in china, and a more competitive industry model will develop. 4.3 forming a greener outlook for the shale gas industry by making water resource protection a core issue and placing equal emphasis on resource development and ecological protection the development of shale gas will not only bring about the challenge of protecting water resources but will also raise environmental issues such as land disturbance, greenhouse gas emissions and the potential leakage of underground water. therefore, during the revolution in shale gas, the implementation of policies relating to water resource protection and environmental pollution prevention should proceed simultaneously. due to the nature of the contaminants involved in shale gas development, research into pollution reduction and treatment should be encouraged and developed. shale gas in china is mainly distributed in mountainous and hilly areas, such as the sichuan and guizhou provinces, which experience water shortages and have transportation issues. hence, launching the development of, and exploration for shale gas in these areas may bring about adverse impacts on local vegetation, aquatic biology, municipal water and the usage of industrial and agricultural water. large-scale river water fracturing only applies to experimental development at an early stage. water shortage and environmental pollution will be a significant issue for mass commercial production of shale gas. it is, therefore, necessary to utilize the combined technologies of horizontal drilling and hydraulic stimulation, to help reduce water consumption, during the development of shale gas. the installation of water treatment facilities will be indispensable for reusing flowback fluid, thereby improving water use efficiency, as well as for groundwater protection. the use of oil-based muds for drilling should be reduced, and the use of water-based muds should be encouraged. drill cuttings should be collected and disposed of properly, which will help to reduce land contamination. noise control in populated areas and functional relationships between corporations and communities are also necessary for the long-term future of the shale gas industry. optimization methods for the drilling process include the use of “cluster wells,” “multi-well pads,” “gas terminal standardization designs” and the rehabilitation of agricultural land after construction. optimization of the site selection process will avoid damaging ecological and environmentally sensitive areas and will reduce tree felling and vegetation destruction. plans to restore vegetation and rejuvenate the land should be designed before construction and executed adequately after construction. use of the methods mentioned above will ensure that shale gas development, and environmental protection progress hand in hand. conclusions shale gas exploration and development in china should learn from the experiences of other countries and cultivate an active market, with the support of government policy. a geological theory applicable to the unique geological conditions found in china is indispensable for the future of shale gas exploration and should be established as soon as possible. shale gas development technology should be improved to protect the water and land resources, as well as prevent pollution. the present paper contains longitudinal profile, hypsometric analysis of river basin and geomorphological studies for future planning of soil and water conservation in anjani and jhiri river basin, jalgaon district, maharashtra, india (20°45’ to 21°08’n latitude and 75°09’ to 75°27’e longitude). the main aim of the present study is to recommend a suitable site for soil and water conservation in anjani and jhiri river basin, northern maharashtra, india. the study area is geologically covered by deccan trap basalts of the cretaceous to the lower eocene age, and some part is covered by thick alluvium of the quaternary age. drainage map of the study area reveals two drainage patterns including dendritic and sub dendritic patterns, indicating that the uniform lithology of the basaltic landforms has uniform resistance to erosion. this condition has caused the anjani and jhiri river get little recharge from the rainwater and the surface drainage flows during only monsoon season and latter becomes dry. therefore, water conservation is a prime need for minimize water scarcity problems in the study area. in the present study, landscape information, such as longitudinal profile and hypsometric integral, has been analysed. the longitudinal profile reveals the presence of nick points at three locations in anjani river and two locations in jhiri river. the nick points indicate that the rapids are formed due to different erosion, relief and presence of lineament (joint). the hypsometric integral values for the anjani and jhiri river are 0.52 and 0.51, respectively. the present research illustrates that both basins fall under equilibrium stage and are in a mature phase of development. el presente artículo contiene el perfil longitudinal, el análisis altimétrico de las cuencas ribereñas y los estudios geomorfológicos para la planeación a futuro de la conservación de suelos y aguas de los ríos anjani y jhiri, en el distrito jalgaon, maharashtra, india (20° 45' a 21° 08' n de latitud y 75° 09' a 75° 27' e de longitud). el objetivo principal de este estudio es la identificación de un lugar apropiado para la conservación de agua y de suelos en las cuencas de los anjani y jhiri, al norte de maharashtra. el área de estudio está geológicamente cubierta por basalto de las traps del decán, en la era del cretáceo al bajo eoceno, y por aluvión espeso de la era cuaternaria. el mapa hidrográfico de la zona de estudio muestra dos modelos de drenaje, el dendrítico y el subdendítrico, lo que indica que la litología uniforme de los accidentes geográficos basálticos tiene una resistencia uniforme a la erosión. esta condición ha causado que los ríos anjani y jhiri tengan una pequeña recarga proveniente de las aguas lluvias y de los drenajes superficiales durante la única estación monsónica del año y que luego se secan. es por esto que la conservación del agua es una necesidad primaria para minimizar los problemas de escasez en el área de estudio. en el presente trabajo se analizó la información del paisaje, como el perfil longitudinal y la altimetría integral. el perfil longitudinal revela la presencia de cambios de gradiente en tres puntos del río anjani y en dos puntos del jhiri. los cambios de gradiente indican que los rápidos se formaron debido a la erosión diferente, al relieve y presencia de lineamiento (convergencias). los valores de altimetría integral para los ríos anjani y jhiri son de 0.52 y 0.51, respectivamente. la presente investigación muestra que ambas cuencas se encuentran en un escenario de equilibrio y en una fase madura de desarrollo. earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 51 58 abstract resumen key words: longitudinal profile, hypsometric analysis, soil and water conservation structures, anjani-jhiri river basin, deccan trap, india palabras clave: perfil longitudinal, análisis de altimetría, estructuras de conservación de agua y suelos, cuencas anjani y jhiri, traps del decán, india. altimetric and hypsometric analysis for soil and water conservation: a case study of anjani and jhiri river basin, northern maharashtra, india golekar r. b.¹*, baride m. v.², patil s. n.³, adil s. h.4 ¹department of applied geology, school of environmental and earth sciences, north maharashtra university, jalgaon 425001 (m.s.), (india) ²department of geology, z. b. patil college, dhule 424002 (m.s.), (india) ³department of applied geology, school of environmental and earth sciences, north maharashtra university, jalgaon 425001 (m.s.), (india) 4department of applied geology, dr. h. s. gour university, sagar – 470003 (m.p.), india *corresponding author: r. b. golekar (e-mail): rbgolekar@gmail.com *present address: department of civil engineering, rajarambapu institute of technology, rajaramnagar, islampur district sangli 415414 (m.s.), (india) record manuscript received: 23/08/2014 accepted for publication: 20/04/2015 geomorphology issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n1.45175 52 golekar r. b., baride m. v., patil s. n., adil s. h. introduction the maharashtra state, with its peculiar geohydrological, geomorphological and climatic conditions, faces multifold problems in exploration, exploitation and development of its groundwater resource in a scientific and planned manner. the operation of groundwater on a large scale for meeting the demands of rapid industrialization, agricultural sector, and domestic purpose has further aggravated the problem. currently, maharashtra, especially in marathwada and northern maharashtra regions, faces drinking water problem and supply for drinking water in this area depends on water tankers. the anjani and jhiri river basin was chosen for catchment hydrogeomorphological studies, both of which are tributaries of girna river. girna is one of the major tributaries of river tapi. anjani and jhiri river basin covers an area of about 1243.65 km². the present study area is located in jalgaon district (fig. 1). it is confined to latitude 20° 45’ n to 21° 08’ n and longitude 75° 09’ e to 75° 27’ e, covering topographic maps no. 46 p/1, 46 p/5 and 46 o/8 of the survey of india. the data on anjani jhiri river basin for the present study can serve as a pilot study based on hydrology and hydrogeomorphology. the anjani jhiri river basin is characterized by an arid to semi-arid climate, and its available water depends on the amount of rainfall in this basin. the water in the basin is used for agricultural and domestic purposes. most of the domestic water supply is through common well in rural areas funded under rural water supply schemes of the state government. people use water from unprotected open pits (dug) well, and/or depend on supply by tankers. the scarcity of water is also one of the main reasons that compel the farmers to cultivate their land under rain fed condition annually. the anjani jhiri river basin enjoys a hot and dry summer, cool winter and fair monsoon with an average annual rainfall of about 750 mm. the anjani river has an elevation difference of 135 m, with the highest elevation (305m) at titvi village located at the southern edge of the catchment, and drops to its lowest elevation (170m) at the northern edge where anjani river joins girna river. the jhiri river has an elevation difference of 125 m, with highest elevation of 300 m amsl near padamalay reserve forest, and the lowest being 175 m. amsl near its confluence with the river girna. the contour map of this area (fig. 2) shows elevation ranging from 300m to about 180m from southern part of the basin to the confluence of girna river. the hydrogeomorphological study helps to explore, plan and manage the water and the natural processes which produce landforms in association with their hydrological characteristics (babar, 2005). figure 1. location map of the study area 53altimetric and hypsometric analysis for soil and water conservation: a case study of anjani and jhiri river basin, northern maharashtra, india figure 2. contour map and river flowing in the study area figure 3. dark grey massive basaltic flows. (location: 20°58’ n 75°27’ e) geologically, the major part of study area is covered by deccan basalt of the cretaceous to the lower eocene age and the rest by alluvium of the recent period (gsi, 2000). the deccan trap lava sequence is grouped under sahyadri group, which was formed during the upper cretaceous to lower eocene. the lava assemblage of sahyadri group consists of alternating sequence of pahoehoe and ‘aa’ flows (gsi, 1976). dark grey massive basaltic exposed in the study area has been depicted in figure 3. those parts that are covered by deccan basalts consist of amygdaloidal and vesicular basalts. alluvial deposits can be observed along the lower reaches of river valleys and watercourses (fig. 4). the thickness of the alluvium ranges from less than a meter to several tens of meters (10 – 100 m). alluvium consists of unconsolidated to consolidated sand, brownish yellow silt and calcrete (locally known as kankar) along the girna, anjani and jhiri rivers. in the study area, however, the alluvium is characterized by unconsolidated clays and silts that are associated with a very limited thickness (0.5-2.0m). figure 4. geological map of study area (after golekar et al., 2013) the soil cover in the study area is derived from the basaltic lava flows and is classified as a) deep black soils b) medium black soils c) sandy soils and d) black alluvial clay (cgwb, 2009). a deep black soil covers the central part of the catchment area. medium black soils occur within the valley and southern hills, and the alluvial basin soil (black alluvial clay) covers the northern part of the study area. sandy soils are observed along the foothills of satmala stretching near padamalay hillocks. the gross and net water availabilities in the anjani jhiri river basin were calculated by using equation 1 (pilgrim and cordery, 1993), available water (in hectare) = area (in ha) × rainfall (in mm) (1) the gross and net water availabilities in the anjani jhiri river basin have been presented in table 1. 54 golekar r. b., baride m. v., patil s. n., adil s. h. table 1 gross and net water availability in the anjani and jhiri river basin methodology the field work was carried out in the summer of 2012 for observing geomorphological units in the study area. after completion of the field work, we carried out the hypsometric analysis of anjani and jhiri river basin by strahler methods (strahler, 1964). the topographic maps numbered 46 p/1, 46 p/5 and 46 o/8 on 1:50000 scale from the survey of india are considered. the soi toposheet is geometrically rectified and georeferenced to world space coordinate system (wgs 1998) using digital image processing software (autocad 2004 and arc gis version 9.0) and digitization work was carried out for entire basin. percentage hypsometric curve (strahler, 1952) that involves a ratio of relative height expressed in percentage (cumulative (hx100)/h) was plotted on ordinate, and the relative area expressed in percentage (cumulative (ax100)/a) was plotted on abscissa. where ‘a’ and ‘h’ denote area and height between successive contours, and ‘a’ and ‘h’ denote total area and height of the basin, respectively. the areas between successive contours and their respective heights are the basic data required for the study of area-height relationship. the areas between successive contours were measured by digital planimeter, and contour heights were obtained from topographic maps used for this work. after plotting the respective value on simple arithmetic graph paper and joining all the points, a smooth line percentage hypsometric curve can be obtained. the hypsometric integral (hi) were calculated mathematically from the graph. result and discussion hydrogeomorphological parameters the delineation of a hydrogeomorphological parameter is aimed at demarcating the areas of groundwater recharge/discharge and potential zone for the development of groundwater (raghu and reddy, 2011). alluvial plain represents the runoff zone while valley belongs to discharge zone, and the denudational hills constitute the infiltration area. it is classified into three zones on the basis of their groundwater potential zone as a) very favorable zones b) good to moderate zones and c) poor zone. the groundwater potential zones are identified with the help of geomorphological units. the morphometric data are enclosed in details and are described below (table 2 and 3). dissected plateau (dpt) an extensive flat top and steep slopes formed over horizontally layered deccan basalts that may be crossed by fractures, joints and lineaments are called as plateaus (babar, 2005). these units can be expressed regarding the slope of the area, runoff characteristics, drainage density, stream frequency and relief ratio of the area (babar, 2005). in the present study area, two plateaus were observed, i.e. moderately dissected plateau and un-dissected plateau. moderately dissected plateau the soils covering in this plateau are moderately thick and well drained. high moisture capacity suggests that the irrigation requirement is moderate in the moderately dissected plateau area. the groundwater potential in these units is moderate to high. un-dissected plateau the land of this unit is dissected by the streams of giving rise to un-dissected terrain consisting of flat-topped hills and steep scarps. the groundwater potential in these units is very poor (babar, 2005). the runoff from these areas can be arrested through the construction of check dams and other strategies. hydrogeomorphological characters of anjani river basin moderately dissected plateau moderately dissected plateau is presented in upper part of the anjani river catchment (fig. 4). this unit is observed between denudational hills and un-dissected plateau. un-dissected plateau un-dissected plateau is distributed in lower part of the anjani river catchment (fig. 4). this unit is observed between moderately dissected plateau and eroded lad by stream. valley fills (vf) valley fills are linear depression presented in between the younger alluvium composed of cobbles, pebbles, gravels and sands. valley fills are also observed in between the denudational hills near galapur, khadke bk, mangrul and palskhede bk villages of upper part of the anjani river catchment. denudational hills a group of hills with basalt as resistant rock are formed due to different erosional and weathering processes, and they occupy the low area. denudational hills are observed in a few patches of the upper part of the anjani river catchment (fig. 4). alluvial plains (younger and older alluvium) alluvial plains are composed of sand slit and clay layer. fine to coarse sand layers form the potential aquifer zone. it is the most prospective hydrogeomorphological unit for groundwater exploration. younger alluvial plain can provide sufficient amount of water than older alluvial plains. older alluvium is distributed along the northern part of the anjani river catchment area. 55altimetric and hypsometric analysis for soil and water conservation: a case study of anjani and jhiri river basin, northern maharashtra, india hydrogeomorphological characters of jhiri river basin moderately dissected plateau moderately dissected plateau is mainly distributed in the upper part of the jhiri river catchment (fig. 4). this unit is observed between denudational hills and un-dissected plateau. un-dissected plateau un-dissected plateau is mainly located in the lower part of the jhiri river catchment (fig. 4). this unit is observed between moderately dissected plateau and eroded lad by stream. valley fills (vf) valley fills are found between the younger alluvium composed of cobbles, pebbles, gravels and sands. valley fills are observed at padmalaya and umarde villages. denudational hills a group of hills with hard rock is formed due to different erosional and weathering processes and they occupy the region where the river jhiri originates (fig. 4). alluvial plains (younger and older alluvium) alluvial plains with fine to coarse sand layers form potential aquifer zone. younger alluvial plain can provide sufficient amount of water than older alluvial plains. older alluvium occurs in the northern part of the jhiri river catchment, especially at vaktuki, anjan vihire and jhurkheda (fig. 5). table 2 linear aspects of the anjani and jhiri river basin (after golekar et al., 2013) where, (nu) = number of streams; (lu) = total length of streams in km; (l) = mean stream length in km; (rb) = bifurcation ratio table 3 aerial and relief aspects of the anjani and jhiri river basin (after golekar et al., 2013) 56 golekar r. b., baride m. v., patil s. n., adil s. h. figure 5. geomorphological map of study area (mdp: moderately dissected plateau, udp: un-dissected plateau) profile analysis profiles drawn from the contour map is an image observation of the real environment of the landscape. the longitudinal section of the valley is called as longitudinal profile. the entire distance from source to the mouth of a particular river is considered. the graph drawn reveals the relief impact of the river course (singh, 1997). the longitudinal profile provides breaks in the longitudinal course of the river flow. these breaks may indicate nick points, and helps in examining the nature and control of landform development (singh, 1997). longitudinal profile gives geomorphologists an insight of relief and topographical impact on river flow (babar, 2005). longitudinal profile of the anjani river basin the anjani river originates from the titvi village at 305 m amsl, and flows almost straight towards ne. the slope is gentle with a drop in elevation from 292m to 240m, spreading over 25 km. at the village pashtane budruk, it turns toward ne. after that, the channel gradient decreases steadily till its confluence with the river girna at village babhulgaon (fig. 6). a study of longitudinal profile reveals the character of anjani river. it suggests the presence of nick points at locations chorvad (292m amsl), pharkande (240m amsl) and waghlud budruk (197m amsl). the difference in elevation suggests that the river has developed “rapids”. it is formed due to different erosion, relief and presence of lineament (joints). lineament map of jalgaon district has shown in figure 8. longitudinal profile of the jhiri river basin jhiri river originates from village padmalaya at 300m amsl, and it runs almost parallel with anjani river and at anjan vihire village at 176m amsl it takes a nw turn and meets girna river. the longitudinal profile curve for jhiri river basin indicates the presence of nick points at vikhran (212m amsl) and vanjari khapat (186m amsl). the jhiri river has a steep gradient in its earlier course with a drop from 284m to 231m up to village umarde (8 kms away from the origin of jhiri river), and thereafter the slope become gentle up to village vikhran and vanjari khapat. there is a drop in elevation at vanjari khapat and the river becomes gently sloping after village anjan vihire. longitudinal profile of the jhiri river channel has shown in figure 7. the nick points indicate that the rapids are formed due to different erosion, relief and presence of lineament (joint). figure 6. longitudinal profile of anjani river figure 7. longitudinal profile of jhiri river figure 8. lineament map of jalgaon district source: central ground water board (cgwb, 2009) hypsometric (area-altitude) analysis the hypsometric curve is a normalized cumulative frequency distribution of elevation (strahler, 1964). this curve is constructed by summing the number of data points higher than incremental thresholds 57altimetric and hypsometric analysis for soil and water conservation: a case study of anjani and jhiri river basin, northern maharashtra, india of elevation, starting at the maximum elevation. classically, hypsometric analysis has been used to differentiate erosional landforms at different stages of their evolution (strahler, 1952; schumm, 1956). the hypsometric analysis is necessary to determine the runoff, recharge, and storage conditions of the ungauged basins and is helpful for hydrogeological investigations. hypsometric curves show the relationship between relative heights (h/h) and relative areas (a/a). the relative area is a ratio of the area above a particular contour to the total area of the basin. according to sarangi et al. (2001), the relative elevation is calculated as the ratio of the height of a given contour (h) to the maximum basin elevation (h). hypsometry reveals the relationship between altitude and basin area to understand the degree of dissection and the stage of the erosional cycle (strahler, 1952; schumm, 1956). area height curve, hypsometric curve, and percentage hypsometric curve are used to obtain the relationship between altitudes and area of the basin. convex-up curves with high integrals indicate typical youth, undissected (disequilibrium stage) landscapes, while smooth, s-shaped curves crossing the center of the diagram represent the characteristics of mature (equilibrium stage) landscapes and concave curves with low integrals exhibit old and deeply dissected landscapes (strahler, 1952). hypsometric (area-altitude) analysis of the anjani river basin the hypsometric curve of anjani river (fig. 9) suggests that the larger part of the area is moderate to gentle sloping. the curve can be characterized as mature/equilibrium stage of landscape development. the curve shows 1) elevation in 15 % of the area ranges between 300m -280m 2) 50 % between 280m -240m and 3) 35 % between 240m-180m. these results indicate that about 85% of the study area is moderate to gently sloping and about 15 % area is steep sloping (fig. 6). figure 9. hypsometric curve (anjani river) hypsometric (area-altitude) analysis of the of the jhiri river basin hypsometric curve of jhiri river (fig. 10) shows 5 % are of steep slopes (300m -260m)), 80 % are of moderate slope (280m220m) and 15 % are on gentle slope (220m-180m), suggesting that jhiri river flows over moderate slope to gentle slopes. this indicates a mature (equilibrium) stage of river development (fig. 7). figure 10. hypsometric curve (jhiri river) hypsometric integral (hi) the hypsometric integral (hi) is the area beneath the hypsometric curve which relates the percentage of total relief to its cumulative percentage. this provides a measure of the distribution of landmass volume remaining beneath or above a basal reference plane. integration of the hypsometric curve gives the hypsometric integral. pike and wilson (1971) has mathematically defined the elevation-relief ratio e as where, e is the elevation-relief ratio equivalent to the hypsometric integral hi; elev mean is the weighted mean elevation of the basin estimated from the identifiable contours of the delineated basin; elev min and elev max are the minimum and maximum elevations within the basin, the hypsometric integral is expressed in percentage units. the hypsometric curves and hi values give valuable information about the geological stage of the development of a basin. according to strahler (1952) and kusre (2013), it can be classified as a) hi ≥ 0.6 = the basin is in-equilibrium (youthful) stage. b) 0.35 ≤ hi < 0.6 = the basin is at equilibrium stage. c) hi< 0.35 = the basin is at monadnock stage. in in-equilibrium (youthful) stage, the basin is still under developed. the equilibrium stage is the mature phase of basin development, and the monadnock phase occurs, particularly when isolated bodies of resistant rock from major hills (monadnock) are found above the subdued surface. it is a rejuvenation stage controlled by structural features (kusre, 2013). narayanpethkar et al. (1991) and sable et al. (2009) used the hypsometric integrals for understanding groundwater recharge in different basins of basaltic terrains. the characteristics of hypsometric integrals of the watershed were also reported by vivoni et al. (2008). they stated that the watershed with higher hi values indicated higher soil moisture, whereas a watershed with lower hi values is characterized by soil moisture being concentrated at the shallow depth. this means that watershed with lower hi values has less total runoff contributed from surface runoff. whereas, watershed with higher hi values shows subsurface runoff is major process contributing the total runoff. 58 golekar r. b., baride m. v., patil s. n., adil s. h. hypsometric integral (hi) for anjani river basin the hypsometric integral value of the anjani river is 0.52. according to strahler classification (strahler, 1952), the value falls under the equilibrium stage and that the basin is in a mature stage of development. hypsometric integral (hi) for jhiri river basin the hypsometric integral values of the jhiri river is 0.51. according to strahler classification (strahler, 1952), the value also falls under the equilibrium stage and is in a mature stage of development. the values for both rivers are similar. the river has attained mature stage since it has been controlled by climate, lithology, and relief. the nick points observed for both of the rivers (figure 5 and 6) do not show any indication of structural control or rejuvenation of the river. conclusion and recommendation drainage morphometry data of the area suggests that the maximum number of streams is of the first order (table 2). it is also observed that there is a decrease in stream frequency as the stream order increases. the total and mean lengths of stream segments are maximal in first order streams and decrease as the stream order increases. an exception to this is anjani river basin that shows the 6th order stream has more length then the 5th order stream (table 2). this is due to the moderate to steep slope. topography and also shows nick points at pharkande where the 5th order stream begins. the average bifurcation ratios of anjani and jhiri river are 3.63 and 4.13, respectively, suggesting low runoff, high recharge and somewhat elongated basin (table 2). drainage density values of anjani river (1.25/ sq.km) and jhiri river (1.49/ sq.km) suggest that the jhiri river is flowing through more weathered basalts than anjani river (table 3). elongation ratios of the both basins are categorized as more elongated suggesting the areas has a high infiltration capacity and low runoff (table 3). both rivers flow through basalts with moderate slope and low relief. the ruggedness number for both river basins show lower value due to the presence of hard resistant basalts (table 3). longitudinal profile results reveal the presence of nick points at chorvad (292m amsl), pharkande (240m amsl) and waghlud budruk (197m amsl) in anjani river and at vikhran (212m amsl) and vanjari khapat (186m amsl) in jhiri river (fig. 7). the hypsometric curve of anjani river suggests that a larger part of the area is moderate to gently sloping. the curve can be characterized as mature/ equilibrium stage of landscape development whereas the hypsometric curve of jhiri river shows 5 % steep slopes (300m -260m)), 80 % moderate slope (280m220m) and 15 % gentle slope (220m-180m) suggesting that jhiri river flows over moderate to gentle slopes. this fact indicates a mature (equilibrium) stage of river development. the hypsometric integral values of both the anjani and jhiri rivers fall into the equilibrium stage and that the basins are in a mature phase of development. the area, therefore, suggests relief, topography and climate controlled development. following are preventive measures recommended for the protection of the groundwater resources in the study area. i. cement concrete bandhara (bunds) is most feasible for water conservation structures in the moderately dissected plateaus and construct it across the 3rd and fourth order streams of both the anjani and jhiri river basin. ii. nala bunding, gabion structures and vegetative bunds feasible structures for soil and water conservation and construct it across the 2nd or third order streams of both the anjani and jhiri river basin. iii. in the alluvial tract of the study area, recharge wells or shafts are suggested, because the groundwater can be tapped through deep tube wells. this type of structure can be constructed in the lower part of the anjani and jhiri river catchment, mainly at ahire bk, pashtane bk and tarde bk villages in anjani river basin and at vaktuki, anjan vihire and jhurkheda village in jhiri river basin. the most feasible artificial recharge structure in such areas are recharge wells/shafts on the river bed of the tributaries. iv. controlling urban expansion in the study area is suggested. the study area is a recharge area for the shallow and deep aquifers, and the urban expansion will reduce the area of recharge. acknowledgements i would like thanks to department of science and technology, government of india for valuable research grant sanctioned to dr. m. v. baride (research guide) and the project fellowship awarded to one of the author rbg under this scheme. the co-operation of north maharashtra university officials is also acknowledged. authors also would like thanks to survey of india for provide a topographic map. references babar md. (2005). a text book hydrogeomorphology fundamentals application and techniques new india publishing agency, new delhi 137 p. cgwb (2009). manual on hydrogeology of jalgaon district, maharashtra cgwb, nagpur ground water information jalgaon district maharashtra 1606/dbr/2009 19 p. geological survey of india (2000). district resource map jalgaon district, maharashtra geological survey of india (1976). geology of the jalgaon district, maharashtra geological survey of india 125th annual celebration pp. 6 golekar r. b., baride m.v., patil s. n. (2013). human health risk due to trace elements contamination in groundwater from anjani and jhiri river catchment of northern maharashtra, india earth science research journal vol 17, no. 1, pp 17-23 golekar r. b., baride m.v., patil s. n. (2013). morphometric analysis and hydrogeological implication: anjani and jhiri river basin maharashtra, india archives of applied science research v. 5 (2) pp 33 41 kusre b. c. (2013). hypsometric analysis and watershed management of diyung watershed in north eastern india journal geological society of india vol.82, pp. 262-270 narayanpethkar a.b., zambre m. k., mallick, k. (1991). geophysical studies for groundwater in adila basin around solapur, maharashtra proc. of seminar association of exploration of geophysist pp 255 to 261 pike r.j., wilson s.e. (1971). elevation-relief ratio, hypsometric integral and geomorphic area-altitude analysis geol. soc. london vol 157, pp.303-316 pilgrim d. h., cordery i. (1993). chapter 9: flood runoff. in: maidment dr (ed.) handbook of hydrology. mcgraw-hill, new york, usa. raghu v., mruthyunjaya reddy k. (2011). hydrogeomorphological mapping at village level using high resolution satellite data and impact analysis of check dams in part of akuledu vanka watershed, anantapur district, andhra pradesh. j. ind. geophys. union, v. 15(1), pp.1-8 sable s. m, ghodake v. r., narayanpethkar a. b. (2009). influence of weathering and fracting on aquifer parameters in the basaltic terrain: a case study in dhubdubi basin, solapur district maharashtra, india gondwana geological magazine vol. no. 24 (2) pp – 131 – 136 sarangi a, bhattacharya a k., singh a., singh, a.k. (2001). use of geographic information system (gis) in assessing the erosion status of watersheds indian jour. soil conservation v. 29, pp.190-195 schumm s. a. (1956). evolution of drainage systems and slopes in badlands at perth ambos, new jersey. geol. soc. amer., v. 67 pp 597-646. singh savindra (1997). geomorphology, prayag pustak bhawan, allahabad, pp 18107 strahler a. n. (1952). hypsometric (area-altitude) analysis of erosional topography geo. soc. am. bull v. 63, pp. 1117-1142 strahler, a.n. (1964). quantitative geomorphology of drainage basins and channel networks, in hand book of applied hydrology (edited by v. t. chow), pp 439-476. vivoni e.r., moreno h.a., mascaro g., rodriguez j.c., watts c.j., garatuza‐payan, j., scott, r.l. (2008). observed relation between evapotranspiration and soil moisture in the north american monsoon region geophysical research letters 35: doi: 10.1029/2008gl036001 deflections of the vertical(dov) are essential data in the geodetic observation reduce, the earth gravity field and geoid model refinement, and the mass earth change. the meridional and prime vertical components of global oceanic dovs are estimated from altimetry data of topex/poseidon, jason-1, and jason-2, through 1992 to 2013 with the crossover method, to analyze the temporal-spatial distribution of marine dovs. comparing with the egm2008-modelled dovs, precisions of meridional and prime vertical components can be up to 0.98” and 1.02”, respectively. the time series of annual mean dovs from 1992 to 2013 are studied to get the annual variation law. annual changes of meridional and prime vertical components are small over most oceans. but the annual changes are significantover oceans with large submarine topography undulations.the spatial distribution of oceanic dovs is also analyzed. the absolute dovs over oceans around lands, islands, and large oceanic trenches are greater than those over the other oceans. the meridional and prime vertical components are consistent withthe longitude and latitude directions, respectively. las desviaciones de la vertical (dov, del inglés deflections of the vertical) son información esencial en la simplificación de la observación geodésica, el campo gravitatorio terrestre y el perfeccionamiento del modelo geoide, y el cambio de la masa terrestre. se estimaron los componentes meridional y principal vertical de las desviaciones de la vertical oceánica global con la información altimétrica de topex/poseidón, jason 1 y jason 2, de 1992 a 2013, con un método de cruce para analizar la distribución espacio-temporal de las desviaciones de la vertical marinas. comparadas con las desviaciones de la vertical obtenidas con el modelo egm2008 (modelo de gravitación de la tierra), la precisión de los componentes meridional y principal vertical puede aumentarse 0.98" y 1.02", respectivamente. se estudiaron las series temporales de la media anual de las desviaciones de la vertical desde 1992 a 2013 para obtener la ley anual de variación. los cambios anuales de los componentes meridional y principal vertical son muy pequeñas en la mayoría de los océanos. sin embargo, los cambios anuales son mayores en aquellos océanos con una topografía submarina de grandes ondulaciones. también se analizó la distribución espacial de las desviaciones de la vertical oceánicas. las desviaciones de la vertical absolutas alrededor de los continentes, islas y fosas oceánicas mayores son superiores que aquellas sobre los otros océanos. los componentes meridional y vertical principal son consistentes con las direcciones de la longitud y la latitud, respectivamente. abstract resumen keywords: deflection of the vertical (dov), satellite altimetry, temporal-spatial distribution, crossover point, marine gravity field. palabras clave: desviación de la vertical (dov), altimetría satelital, distribución espacio-temporal, punto de cruce, campo de gravitación marino. temporal-spatial distribution of oceanic vertical deflections determined by topex/poseidon and jason-1/2 missions record manuscript received: 26/11/2015 accepted for publication: 06/03/2016 how to cite item guo, j. y., shen, y., zhang, k., liu, x., kong, q. and xie f. (2016). temporal-spatial distribution of oceanic vertical deflections determined by topex/poseidon and jason-1/2 missions. earth sciences research journal, 20(2), h1-h5. doi:http://dx.doi.org/10.15446/esrj.v20n2.54402 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.54402 jin yun guo1,2*, yi shen1, kaihua zhang1,3, xin liu1, qiaoli kong1,2, feifei xie1,2 1college of geodesy and geomatics, shandong university of science and technology, qingdao 266590, china 2 state key laboratory of mining disaster prevention and control,co-founded by shandong province and ministry of science & technology, qingdao 266590, china 3tianjin survey and design institute for water transport engineering,tianjin 300456, china * corresponding author: jinyunguo, e-mail: jinyunguo1@126.com earth sciences research journal earth sci. res. j. vol. 20, no. 2 ( june, 2016 ) : h1-h5 m e t e o r o l o g y h2 jin yun guo, yi shen, kaihua zhang, xin liu, qiaoli kong and feifei xie 1. introduction the deflection of the vertical (dov)as one basic research topic in geodesy and geophysics can characterize thedirection of gravity vector and provide one tight local tie among space geodetic techniques (shen et al., 2015). the traditional methods to measure dov includethe astronomical geodetic surveying, gravimetry, combination of astronomical geodesy and gravimetry, and global navigation satellite system (gnss). based on the astronomical geodetic principle, the automatic measurement system integrating gnss and ccd zenith telescope can precisely measure dovs (guo, song, chang, &liu,2011; hirt&seeber, 2008; tian et al., 2014, peng&xia,2015).these measurement methods of the vertical deflectionare applicable to lands (ning,guo, b. wang,& h. m. wang, 2006), but are unable to measure practically the oceanic vertical deflection due to limitations of these techniques under effects of ocean dynamic environment over open oceans (guo,chen,liu, zhong,&mai, 2013;guo, liu, chen, wang, &li,2014). dovs over open oceans are commonly calculated from the earth gravity field model at present (jekeli, 1999; pavlis, holmes, kenyon, & factor,2012). ship-borne gravity data can also be used to estimate the oceanic vertical deflections. but the ship-borne gravimetry can mostly be carried out over coastal seas, and most of theoceans are short of gravity field data. with the rapid development of satellite altimetry, altimeter data can widely be used to estimate the oceanic vertical deflection (fu and cazenave, 2000; guo, hu, wang, chang,&li, 2015). with the development of satellite altimetry technique, oceanic dovs are mainly utilized to determineoceanic gravity field model with high precision. dovs on crossover points can be calculated from altimeter data by the crossover method (sandwell, 1984). altimeter data of ers-1 and geosat were processed to estimate dovs of crossover points from which gravity field over antarctic oceans was estimated by the laplace equation (sandwell, 1992). sandwelland smith (1997) used ers-1 and geosat altimetry data to compute dovs of crossover points which are interpolated to get dovs of grids and then estimated the gravity anomalies over oceans with the fast fourier transformation (fft).single differences of sea surface height along thetrack and corresponding track azimuths are used to calculate oceanic dovs along thetrack (olgiati, balmino, sarrailh,& green, 1995). altimetry data from multi-satellite altimetry missions can be integrally processed to get dovs along tracks which are interpolated to calculate dovs of grids used to estimate marine gravity anomalies. gravity anomalies along tracks with the alongtrack dovs are more accurate than those determined by fft with the inverse stokes formula. the inverse vening meinesz formula is derived to calculate dovs with the grid method (hwang, 1998). the geoid and gravity anomalies over china seashas been calculated from oceanic dovs with the crossover method (li,ning, chen,& chao, 2003). dovs from the crossover method are more accurate than those from the grid method and the along track method, and the grid method is optimal to determine the high-resolution precise marine gravity field using altimeter data (peng and xia, 2004). dovs over china coastal seas are calculated from geophysical data records (gdr) of envisat with the grid method (xing, li &liu, 2006). comparing with egm96 (lemoine et al., 1998), precisions of the prime vertical and meridional components are 6″ and 3″, respectively. multi-altimeter data are processed to calculate global dovs with the grid method. comparing with egm96, precisions of the prime vertical and meridional components over global oceans are 1.97” and 1.12”, respectively. topex/poseidon (t/p) launched in 1992, jason-1 initiated in 2002 and jason-2 initiatedin 2008 can together measure oceans. jason-2 as jason-1 follow on runs the same track of t/p and continuously surveys oceans. theprecision of single observation is estimated to be near to 3 cm (beckley et al., 2010; fu and cazenave, 2000; tunini et al., 2010).t/p and jason-1/2 providealtimetry data with high precision which are processed by the crossover method. global oceanic dovs are estimated and then the temporal-spatial distribution of global oceanic dovs is studied in the paper. 2. satellite altimetry data gdrs version c of t/p, jason-1, and jason-2 released by the physical oceanography distributed active archive center(po.daac) are processed in the paper. time spans of t/p, jason-1, and jason-2 data are from october 1992 to august 2002 with cycles 2 to 364, from august 2002 to january 2009 with cycles 22 to 259, and from january 2009 to may 2013 with cycles 21 to 177, respectively. t/p starts the precise satellite altimetry era, and jason-1 and jason-2 follow t/p. these three altimetry satellites are in the same exact repeated orbit. figure 1 shows global tracks and crossover points of t/p cycle 17. the radial precision of t/p orbit determination is better than 3 cm, and the radial precisions of jason-1 and jason-2 orbit determination can be up to 1 cm based on the on-board satellite tracking systems. sea surface height accuracyfrom these three altimeter data can be lower than 3 cm. to get more precise sea surface heights, altimetry data should be edited based on the criterions provided by aviso (aviso/altimetry, 1996; dumont et al., 2011; picot, case, desai, vincent, &bronner, 2012; shah, sajeev, & gopika, 2015). figure 1. t/p tracks and crossover pointsfor cycle 17 3. determination of oceanic dovs the crossover method is one of thebest waysto determine precise oceanic dovs from satellite altimetry data. altimetry profiles of ascending and descending arcs can be determined by the geoid gradient or the single difference of dov along thetrack, and then the prime vertical and meridional components of dov can be solved by combining thetrack gradients. the meridional componentand ξ the prime vertical component η of dov are calculated from the two directions of geoid gradients (heiskanen&moritz,1967), that is (1) (2) where r is the average radius of the earth. ∂n ⁄ ∂φ =1⁄ 2|φ̇ | (n ̇a-n ̇d)and ∂n ⁄ ∂ λ=1⁄ 2|λ ̇ | (n ̇a+n ̇d) and, in which φ andare the directions of prime vertical and meridional speedsof nadir pointrespectively, is the derivative of geoid heightwith respect to time on thealtimetry data point relative,and subscripts and denotethe ascending and descending arcs respectively. ,, and can be obtained by the time and location information in the altimetric data (hwang& parsons,1995; sandwell& smith,1997).the detailed data processing procedure is following. 1) altimetry data preprocessing. these three-altimeter data are read with the corresponding program based on data formats. these data are edited according to the relatedcriteria published by the data providers. all geophysical corrections are applied to altimetry data. these data including latitude, longitude, sea surface height and time are saved based on the ascending and descending arcs. h3temporal-spatial distribution of oceanic vertical deflections determined by topex/poseidon and jason-1/2 missions 2) determination of crossover point. the quadratic polynomials are used to fit the ascending and descending arcs on geodetic latitude and longitude based on the positioning information in altimetry data, respectively. the quadratic equations of ascending and descending arcs are combined to solve coordinates of thecrossover point. 3) unification of time and space datum. there exist systematic biases between different altimeter data under effects of a different ellipsoid, time-varying sea surface height, satellite orbit determination error, altimeter error and geophysical correction error. corrections of coordinate frame biases for different altimetry data are calculated to unify the time and space datum for t/p, jason-1, and jason-2. 4) sea surface height of crossover point. sea surface height of crossover point is interpolated from the ascending and descending arcs’ data based on coordinates with the distance weight method. 5) dov determination of crossover point. 8-10 altimetry points around crossover point are selected based on the position of thecrossover point. the quadratic polynomial on latitude and longitude, and the quadratic curves of latitude and longitude with respect to time are fitted based on information about sea surface height, latitude, longitude and time of altimetry point, respectively. then derivatives of these quadratic polynomials are made to get , andwhich are substituted for equations (1) and (2) to calculate the prime vertical and meridional components of dov on crossover point. 6) repeat step 5) to calculate all dovs of all crossover points. the same pass may be different for all cycles and the farthest distance of crossover points for the same ascending and descending arcs may be up to 1 km for t/p, jason-1 and jason-2 (fu and cazenave, 2000; luz clara, simionato, d’onofrio, & moreira, 2015). so the mean position of crossover point for the same ascending and descending arcs should be determined. 7) reduce dovs of crossover points to the mean position with the inverse distance weighted interpolation method within the studied time span. we can get oceanic dovs of crossover pointsfrom altimetry data from 1992 to 2013. so the temporal-spatial distribution of oceanic dovs can be obtained. altimetry data should be edited and corrected before dov calculation because of contamination of many kinds of errors. the same editing criteria are used to modifyand correct altimetry data to get precise results. to check the precision of oceanic dovs from altimetry data, itwasselected t/p data of cycle 17 to compute oceanic dovs of crossover points in the global area (-60°s~60°n, 0°~360°), which are compared with the modeled dovs from egm2008 (pavlis, holmes, kenyon,& factor,2012). table 1 lists the statistical results. from table 1, it can be seen that theprecisions of prime vertical and meridional components are 1.02″ and 0.98″ comparing with egm2008-modelled components, respectively, which indicates that oceanic dovs of crossover points from altimetry data are reliable. table 1. statistical results of oceanic dovs from t/p, jason-1, and jason-2 4.temporal variations of oceanic dovs the earth gravity field not only changes in space but also changes in time. the temporal-spatial variation of the earth gravity field is one important part of geodynamics (ding, li, &ding, 2006). the gravity field change containing the abundant geophysical information of interior earth can be studied throughthe dov variations. oceanic dovs of crossover points for each cycle are calculated from altimetry data of t/p, jason-1, and jason-2 with the crossover method. the mean position of crossover point is determined by averaging positions of crossover points from the same ascending and descending arcs in one year. dovs from the altimetry data of about 37 cycles in one year are used to get the mean dov of the averageposition with the inverse distance weighted interpolation method. so itispossibletoget a group of dovs of annual mean crossover pointson a global scale, which can make up a time series of annual mean dovs from 1992 to 2013. the time series can be analyzed to study the interannual variations of oceanic dovs. figures 2 and 3 show annual variations of meridional and prime vertical components of oceanic dovs determined by the crossover method, respectively. figure 2.annual change rate ofmeridional componentof dov figure 3.annual change rate ofprime verticalcomponent of dov from figure 2, it can be found that the annual change of meridional component over most oceans is slight in 1992-2013, and the annual change in most oceans in the north hemisphere is greater than that in the south region. the annual change over open oceans far away from islands and lands is small and apt to convergence, such as the middle pacific ocean and the middle indian ocean. the annual variation over coastal seas is great, such as the bering sea, the indonesian sea, the new zealand sea, and the caribbean sea. on the one hand, the mass over these coastal seas changes greatly. on the contrary, the variation can reflect the lower altimetry data quality over coastal seas than that over open oceans. figure 3 shows the annualchangesin prime vertical components. annual variations are small and apt to convergence over most oceans except for the eastern oceans of the north american, the bengal bay, and the china coastal seas. also, the annual variation is significantover some oceans around islands in the pacific ocean. 5. spatial distribution of oceanic dovs the marine gravity field is mainly determined by the oceanic lithosphere and the deep mass distribution and geological structure. it shows the basic geographic information of oceanic lithosphere structure and the submarine topography (chao, yao, li,&xu, 2002). the uneven distribution of vast oceanic gravity affects the direction of themarine plumb line, which makes the spatial distribution of marine dovs inconsistent. the marine gravity field can be analyzed by studying the spatial distribution of global oceanic dovs. dovs of crossover point for the same ascending and descending arcs in all cycles are interpolated to the mean position with the inverse distance weighted interpolation method. so the averagedov of each crossover point in 1992-2013 can be obtained. figures 4 and 5 show the global distribution of oceanic dovs. h4 jin yun guo, yi shen, kaihua zhang, xin liu, qiaoli kong and feifei xie figure 4.spatial distribution of meridional component of dov figure 5. spatial distribution of prime vertical component of dov from figure 4, it can be foundthat the absolute meridional components of dov are comparatively large over the coastal oceans. the meridional components are large over the west pacific ocean and the middle indian ocean. it is also large over the seas around islands and large oceanic trenches, such as the bering sea, the japan seas, the indonesian seas, and the caribbean sea. the significant meridional components over these oceans are mainly caused by the obvious marine mass changes. the meridional components are positive over the most indian ocean, the northcentral pacific ocean, and the caribbean sea. most meridional components over the other oceans are negative. the meridional components tend to be consistent along the longitude direction over most oceans. from figure 5, it can be found that the spatial distribution of prime vertical components of oceanic dovs is approximately consistent with the distribution of meridional components over most oceans. the absolute prime vertical components are large over the china seas, the west pacific ocean, the indian ocean, and the northwest atlantic ocean. it is also large over the oceans around islands and oceanic trenches, like the bering sea, the japan seas, the indonesian seas, the caribbean sea, and the ocean around the mariana trench. the large prime vertical components over these oceans may be mainly caused by the visiblechange of marine gravity field. the prime vertical components over the most indian ocean, the north-central pacific ocean and the caribbean sea are positive, and these are more negative over the other oceans. in contrast to the meridional component, the prime vertical components tend to be consistent along the latitudedirectionover most oceans. 6. conclusions marine dovs are determined from altimetry data of t/p, jason-1, and jason-2 in 1992-2013 with the crossover method. the temporalspatial distribution of oceanic dovs is studied. comparing with the egm2008-modelled dovs, the precisions of prime vertical and meridional components determined from one-cycle altimetry data of t/p, jason-1, and jason-2 can be up 0.98″ and 1.02″, respectively, which indicates that dovs of crossover pointsestimated from altimetry data are very reliable. annual changes of meridional and prime vertical components of dovs over most open oceans are small. but the annual variations over most coastal seas are significant because mass changes over coastal ocean may be large and thealtimetry precision over coastal seas may be lower. absolute dovs over oceans around lands, islands and vastocean trenches are biggerthan those over the other oceans. this is mainly due to the obvious gravity change of interior earth and the severe submarine topography undulation. the meridional and prime vertical components of dovs are consistent withthe longitude and latitude directions, respectively. the spatial resolution of oceanic dovs determined from altimetry data with the crossover method is limited by theexact repeating period of altimetry satellite. thecombination of more satellite altimetry missions can improve the resolution of oceanic dovs. acknowledgments: this study is supported by the national natural science foundation of china (grant no. 41374009), the shandong natural science foundation of china (grant no. zr2013dm009), the public benefit scientific research project of china (grant no. 201412001), and the sdust research fund (grant no. 2014tdjh101). references: aviso/altimetry (1996). aviso user handbook: merged topex/poseidon products (gdr-ms). avi-nt-02-101-cn, cls & cnes beckley, b., zelensky, n. p., holmes, s. a., lemoine, f. g., ray, r., mitchum, g. t., ... brown, s. (2010). assessment of the jason-2 extension to the topex/poseidon, jason-1 sea-surface height time series for global mean sea level monitoring. marine geodesy, 33(1), 447-471. doi: 10.1080/01490419.2010.491029 chao, d. b., yao, y. s., li, j. c., & xu, j. s. (2002). interpretation on the tectonics and characteristics of altimeter-derived gravity anomalies in china south sea (in chinese). editorial board of geomatics and information science of wuhan university, 27(4), 343-347. ding, y. f., li, y. f., & ding, x. b. (2006). deflection of vertical in modern geodesy (in chinese). journal of geodesy and geodynamics, 26(2), 115-119. dumont, j. p., rosmorduc, v., picot, n., desai, s., bonekamp, h., figa j, ... scharroo, r. (2011). ostm/jason-2 products handbook. edition 1.8. salp-mu-m-op-15815-cn, eumetsat/cnes/noaa/nasa/jpl. fu, l. l., & cazenave, a. (2000). satellite altimetry and earth sciences: handbook of techniques and applications. san diego, academic press. guo, j. y., chen, y. n., liu, x., zhong, s. x., & mai, z. q. (2013). route height connection across the sea by using the vertical deflections and ellipsoidal height data. china ocean engineering, 27(1), 99110.doi: 10.1007/s13344-013-0009-9 guo, j. y., hu, z., wang, j., chang, x., &li, g.(2015). sea level changes of china seas and neighboring ocean based on satellite altimetry missions from 1993 to 2012. journal of coastal research: special issue, 73: 17-21. doi: 10.2112/si73-004.1 guo, j. y., liu, x., chen, y., wang, j., & li, c. (2014). local normal height connection across sea with ship-borne gravimetry and gnss techniques. marine geophysical research, 35(2):141-148.doi: 10.1007/s11001-014-9216-x guo, j. y., song, l. y., chang, x. t., & liu, x. (2011). vertical deflection measure with digital zenith camera and accuracy analysis (in chinese). geomatics and information science of wuhan university, 36(9), 1085-1088. heiskanen, w. k., & moritz, h. (1967). physical geodesy. san francisco: freeman. hirt, c., & seeber, g. (2008). accuracy analysis of vertical deflection data observed with the hannover digital zenith camera system tzk2-d. journal of geodesy, 82(6), 347-356. doi: 10.1007/s00190-007-0184-7 hwang, c. (1998). inverse veningmeinesz formula and deflectiongeoid formula: applications to the predictions of gravity and geoid over the south china sea. journal of geodesy, 72(5), 304-312. doi: 10.1007/s001900050169 h5temporal-spatial distribution of oceanic vertical deflections determined by topex/poseidon and jason-1/2 missions hwang, c., & parsons, b. (1995). gravity anomalies derived from seasat, geosat, ers-1 and topex/poseidon altimetry and ship gravity: a case study over the reykjanes ridge. geophysical journal international, 122(2), 551–568. doi: 10.1111/j.1365-246x.1995.tb07013.x jekeli, c. (1999). an analysis of vertical deflections derived from highdegree spherical harmonic models. journal of geodesy, 73(1), 1022. doi: 10.1007/s001900050213 lemoine, f. g., kenyon s. c., factor, j. k., trimmer, r. g., pavlis, n. k., chinn, d. s., & olson, t. r. (1998). the development of the joint nasa gsfc and the national imagery and mapping agency (nima) geopotential model egm96. nasa/tp-1998-206861 li, j. c., ning, j. s., chen, j. y., & chao, d. b. (2003). geoid determination in china sea areas (in chinese). acta geodaetica et cartographic sinica, 32(2),114-119. luz clara, m., simionato, c. g., d'onofrio, e., & moreira, d. (2015). future sea level rise and changes on tides in the patagonian continental shelf. journal of coastal research, 31(3): 519-535. doi: 10.2112/jcoastres-d-13-00127.1 ning, j. s., guo, c. x., wang, b., & wang, h. m. (2006). refined determination of vertical deflection in china mainland area (in chinese). geomatics and information science of wuhan university, 31(12), 1035-1038. olgiati, a., balmino, g., sarrailh, m., & green, c. m. (1995). gravity anomalies from satellite altimetry: comparison between computation via geoid heights and via deflections of the vertical. bulletin géodésique, 69(4),252-260.doi: 10.1007/bf00806737 pavlis, n. k., holmes, s. a., kenyon, s. c., & factor, j. k. (2012). the development and evaluation of the earth gravitational model 2008 (egm2008). journal of geophysical research, 117, b04406. doi: 10.1029/2011jb008916 peng, f. q., & xia, z. r. (2004). vertical deflection theorem of satellite altimetry (in chinese). hydrographic surveying and charting, 24(2), 5-9. picot, n., case, k., desai, s., vincent, p., & bronner, e. (2012). aviso and podaac user handbook-igdr and gdr jason-1 products. edition 4.2. salp-mu-m5-op-13184-cn, cnes & nasa. sandwell, d. t. (1984). a detailed view of the south pacific geoid from satellite altimetry. journal of geophysical research, 89(b2), 1089-1104. sandwell, d. t. (1992). antarctic marine gravity field from high-density satellite altimetry. geophysical journal international, 109(2), 437448.doi: 10.1111/j.1365-246x.1992.tb00106.x sandwell, d. t., & smith, w. h. f. (1997). marine gravity anomaly from geosat and ers1 satellite altimetry. journal of geophysical research, 102(b5), 10039-10054. doi: 10.1029/96jb03223 shah, p., sajeev, r., & gopika, n. (2015). study of upwelling along the west coast of india a climatological approach. journal of coastal research, 31(5): 1151 1158. doi: 10.2112/ jcoastres-d-13-00094.1 shen, y., you, x., wang, j., wu, b., chen, j., ma, x., & gong, x. (2015). mathematical model for computing precise local tie vectors for cmonoc co-located gnss/vlbi/slr stations. geodesy and geodynamics, 6(1), 1-6. doi: 10.106/j.geog.2014.12.001 tian, l. l., guo, j. y., han, y. b., lu, x. s., liu, w. d., wang, z., ... wang, h. q. (2014). digital zenith telescope prototype of china. chinese science bulletin, 59(17), 1978-1983. doi: 10.1007/s11434-014-0256-z tunini, l., braitenberg, c., ricker, r., mariani, p., grillo, b., & fenogliomarc, l. (2010) vertical land movement for the italian coasts by altimetric and tide gauge measurements. procedures esa living planet symposium, bergen, norway. xing, l. l., li, j. c.,& liu x. l. (2006). study on ocean vertical deflection of envisat satellite altimetry along-track geoid gradient(in chinese). science of surveying and mapping, 31(5), 48-49. continuously operating reference station (cors) networks consisting of multiple gnss systems have been set up and operating in many developed countries. in turkey, cors-tr has been operating. the aim of cors-tr is fast, correct, and reliable collection of all kinds of geographic data, thus, speeding up the activities of cadastre, assuring organized urbanization, constituting the spatial infrastructure for relevant works of e-government, and monitoring plate tectonics. therefore, in this paper, we investigated 788 days of linear and periodic of akhr, beys, ciha, kamn, yuna, aksi and knya cors-tr stations by linear trend analysis and wavelet transform in located in the central anatolia region of turkey. as a result of trend analyses of time series it was determined that with the exception of a single station (kamn), other stations were moving in the southwest direction (17.78 mm/year). this finding is in consistency with the region’s tectonic plate movements. in the periodical analyses, stations’ daily, monthly, seasonal, annual and even biennial movements resulting from atmospheric and hydrologic loading, climatic and other effects was revealed. moreover, detail (d) component, which protect the largest percentage of energy, was determined by calculating the energy spectrum. energies of detail components are generally similar in horizontal and especially d7-d8 have the highest energy. however, the components obtained the maximum energy was varied in up coordinate time series. these results indicate that stations of cors-tr network located in central anatolia were rather stable and able to be providing reliable, accurate and continuous data for national and international studies to be conducted in many fields abstract keywords: cors-tr; gnss; wavelet analysis; tectonic plate; power spectrum. research of behaviours of continuous gnss stations by signal analysis methods issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n1.62552 o. oktar1, h. erdoğan1* 1 geomatics engineering, faculty of engineering, aksaray university, tr-68000, aksaray, turkey. * correspondence: hediye.erdogan@gmail.com. earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018): 19-27 las redes de estaciones de referencia de operación continua (cors) conformadas por el sistema satelital de navegación global (gnss) se han establecido y están operando en muchos países desarrollados. en turquía este proyecto se conoce como cors-tr y tiene el objetivo de hacer una recopilación rápida, correcta y confiable de todo tipo de información geográfica, y por lo tanto de agilizar la actividades de catastro, garantizar la urbanización organizada, constituir la infraestructura espacial de los trabajos relevantes de gobierno electrónico y monitorizar las placas tectónicas. además, en este artículo se analizó la información lineal y periódica durante 788 días en las estaciones akhr, beys, ciha, kamn, yuna, aksi y knya a través del análisis de tendencia lineal y el método de transformación por ondículas para la región central de anatolia en turquía. a través del análisis de tendencias en series de tiempo se determinó que, excepto kamn, todas las estaciones se movieron en la dirección suroccidente (17.78 mm/año). este hallazgo es consistente con los movimientos de las placas tectónicas de la región. en los análisis periódicos se revelaron entre otros los efectos climáticos a partir de los movimientos diarios, mensuales, estacionales, anuales e incluso bianuales de las estaciones por la carga atmosférica e hidrológica. además, el componente detalle (d), que consiste en el mayor porcentaje de energía, se determinó al calcular el espectro de energía. las energías de los componentes detallados son similares en su forma horizontal y especialmente el d7 y el d8 tienen el mayor nivel de energía. sin embargo, los componentes que obtuvieron la máxima energía son diferentes en las coordenadas de las series de tiempo. estos resultados indican que las estaciones de la red cors-tr en anatolia central son estables y pueden proveer información confiable, exacta y continua para estudios nacionales e internacionales en las áreas de conocimiento relacionadas. resumen palabras clave: estaciones de referencia de operación continua; sistema satelital de navegación global; transformación por ondiculas; placas tectónicas; espectro de energía. record manuscript received: 09/02/2017 accepted for publication: 19/01/2018 how to cite item oktar, o., & erdogan, h. (2018). research of behaviors of continuous gnss by signal analysis methods. earth sciences research journal, 22(1), 19-27. doi: http://dx.doi.org/10.15446/esrj.v22n1.62552 investigación de comportamientos en una estación con el sistema satelital de navegación global continua a través de métodos de análisis de señales g e o m a t ic s 20 o. oktar, h. erdoğan introduction a satellite navigation or sat nav system is a system of satellites that provide autonomous geo-spatial positioning with global coverage. it allows small electronic receivers to determine their location (longitude, latitude, and altitude) to within a few meters using time signals transmitted along a line-of-sight by radio from satellites. receivers calculate the precise time as well as position, which can be used as a reference for scientific experiments. a satellite navigation system with global coverage may be termed a global navigation satellite system (gnss) (http://en.wikipedia.org/wiki/satellite_navigation, 2016). the original motivation for satellite navigation was for soldiers and military vehicles, planes, and ships in accurately determining their locations worldwide. today, the uses of gnss have extended to include both the commercial and scientific worlds. commercially, gnss is used as a navigation and positioning tool in airplanes, boats, cars, and for almost all outdoor recreational activities such as hiking, fishing, and kayaking. in the scientific community, gnss plays an important role in the earth sciences. meteorologists use it for weather forecasting and global climate studies (ahmet, 2010), and geologists can use it as a highly accurate method of surveying and in earthquake studies to measure tectonic motions during and in between earthquakes. also, gnss is used in determination of geodetic datum (geoid determination), creating a reference system, geophysicalgeodynamic studies, sea-level determination, engineering, cadastral and mining surveying, precision agriculture studies and machine control systems. such practices caused national cartography institutions in countries, local institutions and the private sector to set up gnss networks created by stable gnss stations which provide continuous and real-time observations. particularly, for tectonic studies, gnss based geodynamic projects were developed at a regional level. we can refer the works from (larsen and reilinger, 1992) in the california region, (straub and kahle, 1995; reilinger et al., 1997) in the mediterranean area, (miyazaki et al., 1996) in japan and (erdoğan et al., 2009; eren et al., 2009; gülal et al., 2013) in turkey. in parallel, projects with a broader geographical (intercontinental) scope also started to be developed. examples are the casa uno (kellogg and dixon, 1990) in central and south america and the tango (bastos et al., 1991 in the azores–gibraltar region, which were among the first gps networks established to support geodynamic studies. a nationwide gnss infrastructure network known as cors-tr was established in turkey by istanbul kültür university (iku), general command of mapping and general directorate of land registry and cadastre during 2006–2009. as is known, cors stations are continuous signal receiving points. after obtaining a time series of mean daily positions derived from cors stations observations, one can estimate the station movements as periodical movements from these data. mostly a linear movement is thus assumed to be equal to the plate movement, also called a linear trend, on which the station is located. furthermore, it has been observed that most gnss position time series exhibit annual, semi-annual, seasonal and daily periodical movements with amplitude of a few millimeters (bos et al., 2010; scherneck et al., 2002). blewitt and lavallée (2002) demonstrated that an annual movement within the data deteriorates the accuracy of the estimated linear trend in time series with an observation time span of a few years, even when this annual movement is taken into account during the estimation process. although the causes of periodical movement are not yet completely understood, the most likely explanations are a combination of atmospheric loading (van dam et al., 1997), hydrological loading (van dam et al., 2001), and thermal expansion of the gps stations (romagnoli et al., 2003). in this study, 788 days of gnss observations using 7 cors-tr stations (1.10.2008–26.11.2010) located in the central anatolia region of turkey were processed with gamit/globk program package. then, trend analysis, continuous wavelet transform (cwt) and discrete wavelet transform (dwt) are applied to coordinate time series and correlations between the original time series and details obtained with dwt are investigated. method fast fourier transform (fft) is widely used and well known method for time series analyzers. it is a mathematical technique that transform signal from time to frequency. fft of signal is defined with where n is the number of the signal y(t), t is time and i is complex unit. for 0≤k≤n-1, the calculation of x(k) in eq. (1) requires n complex multiplications and n-1 complex sums. computing all n of the x(k) values demands n2 complex multiplications and n2-n complex additions. the fft coefficients x(k) are in the complex plane but this representation does not aid interpretation. therefore, the power of the fft coefficient is calculated by power spectrum pxx(k)=|x(k)| 2 and existing frequencies in the series are obtained by determining the density frequencies of the signal (allen and mills, 2004). while transforming time information of the signal is lost. in addition, fft supposes signals to be stationary. however, noisy gnss signals are nonstationary. thus, fft can’t detect specific event in time series. for the reasons stated above, wavelet transform (wt) was developed about three decades ago. wt was first applied to analyze seismic waves in geophysics studies (morlet et al., 1982). then, wavelet analysis has gained huge popularity among scientists and engineers who are interested in analyzing the time-frequency characteristics of signals. dividing the signal into different frequencies and researching each component in terms of scale is the main idea of wt. the wt breaks up the signal into its “wavelets”, which are scaled and shifted versions of the “mother wavelet”. the two basic criteria for a function (t) to be considered a wavelet function are (2) that the average value of the wavelet must be zero and (3) that it must have unit energy. the criteria can also is written mathematically as follows: (1) (2) (3) (4) a number of wavelet functions, which have different features, have been identified such as haar, daubechies, symlets, morlet, and mexican hat etc. different wavelets may be used depending on the application. a couple of mother wavelet as shown in fig. 1 also they were used in this study. wt use a time-scale region as shown in fig. 1, so we can detect when specific events occur. wt can be classified in two types: the continuous wavelet transform (cwt) and the discrete wavelet transform (dwt). cwt is a result of multiplication between analyzed time series and scaled and shifted versions of wavelet function at all time (misiti et al., 2002). for the original signals y(t) and wavelet function ψ(t) cwt is defined as (misiti et al., 2002): cwt (a,b) is the coefficients as a result of transition, a is the scale (dilation) parameter of the wavelet, b is the translation parameter of the wavelet, and t is time. scalogram is a method of using wavelet analysis to obtain spectral information. wavelet function has an importance of detecting frequency location. it can be detect that which frequency components is in the time series and when they join the time series with scalogram. figure 1. a) wavelet transform, b) morl wavelet, and c) db1 wavelet. cwt requires lots of time and memory because of calculating wavelet coefficients at every possible scale. therefore, discrete values, which are depended on powers of two, of scale and positions are chosen in the dwt. qiao (2005) developed an effective method, which utilizes filters, to 21research of behaviours of continuous gnss stations by signal (5) (8) (6) (7) apply this template. for the discrete signal y(k) and the wavelet function ψ(n), the dwt is defined as: the dwt filtering process effectively separates the low and high frequency components. from the point of resolution, the low frequency components will have a large scale and the high frequency components will have smaller scale. the low frequency components called as approximations (a) and the high frequency components called as details (d). approximations contain main features of the original signal. noise is generally in detail components. energy of the a and d are able to be calculated with parseval’s theorem. through this, it is possible to compare components each other to detect which d component has a high correlation with original signal. energy of the d components is defined as: energy of a components is defined as: where edi and eai represent energy of the i decomposition level. dij and aij are discrete wavelet transform coefficient of component at i th decomposition level and j time step. n represents the number of the coefficients (qiao, 2005). study area and data the time series analysis was carried out using the cors positions from 7 gnss stations including akhr, beys, ciha, kamn, yuna, aksi, and knya of the cors-tr network (national gnss network), located in the central anatolia region of turkey (fig. 2). figure 2. the cors-tr network in the central anatolia region of turkey. for the time series analysis, 788 days (1.10.2008–26.11.2010) of gnss observations of the cors-tr stations were first processed with gamit software and produced daily time series of cors-tr stations. gamit is a comprehensive suite of programs for analyzing gnss observations and developed by mit, the scripps institution of oceanography, and harvard university with support from the national science foundation. the initial step in the processing of gnss observations is the determination of the exact absolute position of the receiver. exact positioning can be done by two methods, single point positioning and precise positioning (pany, 2007). these two methods are pre-assessments to determine the station coordinates and clock errors. mathematical models of gnss pseudo range code signals for single point positioning system is given in eq.(8). here, c code is the value, c is the speed of light, k index is the satellite and i index is the receiver while δtk is the clock error of the satellite at the times tr and te ,δion ,δtrop ,δtide, and δrel denote the ionospheric, tropospheric, tidal and relativistic effects respectively. the remaining error is denote as ɛc. in eq. (8), satellite clock errors are corrected by using atomic clocks according to igs precise orbit information and broadcast ephemeris information. after the errors are corrected, there are only four unknowns left in eq. (8), receiver clock errors and receiver coordinates. for an acceptable solution, initial value for receiver clock error is considered as null and the eq. (8) is made linear. this linear equation is given below. (9) (10) (11) (12) (13) equation (9) can be written in a more general form as below. in the above equation, akj describes the coefficients in eq.(9), describes the differences between the coordinates, kv the residual errors and receiver clock errors. if we apply the observations made for all satellites to eq. (10), eq. (10) becomes a general equation and single point positioning equation system with p being the weight matrix is as follows ,l ax v p= + here, l is the observation vector, x is unknowns vector and a is coefficient matrix. in equation (11), unknowns vector x is calculated according to least squares method (lsm) with eq. (12). here, coefficient matrix is created by using initial coordinate values. however, most of the time, the initial coordinates are not exactly known. therefore, for single point positioning problems, iterative solutions are mostly used. this means, final coordinates calculated from initially used station coordinates are again processed as initial coordinates. this process is continued until the difference between initial and final coordinates are reduced to minimum. as mentioned above, there are four unknowns in single point positioning equation. therefore, the solution of the problem requires at least four equations. for the formulation and solution of these equations, observation values from at least four satellites are obtained for single point positioning. when receiver station coordinates are calculated, in eq. (10) becomes null and eq. (10) is written as eq. (13), 22 o. oktar, h. erdoğan here, k indicates the total number of satellites observed in one epoch. receiver clock errors can similarly be calculated by phase measurements (pany,2007). after the observations (coordinates or signals) processing, 27 time series were obtained from 7 stations. because of data loss in aksi and knya time series, they were divided into two separate time series including aksi1, aksi2, knya1, and knya2. thus, 27 time series for 7 stations were obtained. these time series show the changes in the signals in the north, east, and up directions of stations according to coordinate system of the gnss. the time series of the ciha, aksi1 and knya1 stations are shown in fig. 3. in fig. 3, the signal change s(ti) of the time series was acquired assuming that s(t1) =0 for the time initiation of ti (i = 1). in this case, the signal change at any time point during the time series is in relation to the initiation time of the signal. a filtering procedure, third-order floating weighted average (fwa), was applied to the series to partially eliminate the noise effect, also big outlier data in the raw time series (gülal et al., 2013). fwa is )]/4s(t)s(t2)[s(ty 2i1ii ++ ++= . this procedure also has a type of lowpass filtering. since the number of time series is too much, just the raw and filtered time series of the ciha and aksi1 stations are shown in fig. 3. (14) and receiver clock error is calculated with eq. (14). figure 3. time series of ciha and aksi1 stations for raw and filtered signals. results and discussion in this study, in the time series linear changes were determined by linear functions and also periodical movements were investigated by cwt and fft applied for details and approximations obtained with dwt. in addition, approximations and details are obtained by applying dwt to the time series and correlations between approximations details and original signals were investigated. linear trend analysis it was determined that if time series for the signal changes of gnss stations north, east an up component contained any linear trend component by calculating a and b parameters of y=a+bt linear functions and standard deviations of these parameters by lsm. whether the calculated parameters are significant or not is decided by t-distribution test. test value t̂ is calculated from parameter and standard deviation of parameter for each parameter, t̂ =parameter/standar deviation of parameter, and this value is compared with pre-determined 1-α confidence level and tf,1-α/2 confidence limit of t-distribution dependent on the f degree of freedom. if, | t̂ |<tf,1-α/2 parameters are insignificant and if | t̂ |>tf,1-α/2 parameters are significantly different from zero at a 1-α confidence level. for this study, t-distribution table values according to the degree of freedom f at 95% confidence level; (f>500,t500,1-0.05/2=1.96) 1.96 for akhr, beys, ciha, kamn, yuna, and aksi1, (f>200,t200,1-0.05/2=1.97) 1.97 for knya1 and knya2, and (f>60,t60,1-0.05/2=2) 2 for aksi2. for all stations, these values from the t-distribution table are compared with the calculated test values and the results showed that test values are greater than table values at 95% confidence level. 23research of behaviours of continuous gnss stations by signal figure 4. north and east time series and linear functions for akhr thus, it is concluded that the time series of north, east and up coordinates of all stations have a linear trend component (there is a linear movement in the stations) and the results are given in table 1. in addition, the north and east coordinates time series and linear trend models of the akhr station are given in fig.4 respectively. statistically significant linear trend component functions (linear model) of the stations and annual linear changes calculated from this model in other words linear movements of stations are shown in table 1. the b parameter of the trend function in table 1 shows the linear variation of the time series of stations and “t” parameters is day. a change in (+) direction indicates increase of movement, a change in (-) direction indicates decrease of movement. using these parameters it is possible to determine the linear changes or movements in stations together with the direction of movement. on the other hand, the change in the (+) direction shows the movements of the stations in the north and east directions, and the change in the (-) direction shows the movements in the south and west directions horizontally. vertically, (+) change indicates that the height is higher, and (-) change indicates that the height is lower. as shown in table 1, while the cors-tr stations have linear movements in west and south directions, kamn have linear movement in west and north directions (+3.65 mm / year). linear movements of stations in horizontal position: the highest linear movement in the south direction is at the akhr station with 5.84 mm/ year, the lowest linear movement in the south direction was observed at knya1 station with 0.66 mm/year. table 1. linear models of gnss stations, y=a+bt, (trend components) and annual velocities. the highest linear movement in the west direction was detected at the akhr station with 20.81 mm/year, and the lowest linear movement was detected at aksi1 station with 6.57 mm/year. except for kamn station, the average resultant movement obtained from the linear movements in the west and south direction of the stations is 17.78 mm/year and the direction is southwest. yet, the linear movement of the kamn station is 13.46 mm/year and the direction is northwest. the results of the linear movements in the up coordinates of the stations show that at akhr, ciha, yuna and knya2 stations have linear movements in (+) direction, that is, vertical values increased. the highest linear movement was calculated at yuna station with 2.63 mm/year value and the lowest linear movement was calculated at ciha station 0.99 mm/year value. linear movements in up coordinates of beys, kamn, aksi1, aksi2 and knya1 stations are in (-) direction. the highest linear movement was detected at kamn station with -21.39 mm/year value and the lowest linear movement at beys station -0.29 mm/year. in the time series analysis of gnss stations, linear changes or movements obtained from the trend component analysis are extremely important in terms of plate movements. the reason is that bt in the trend component is used to determine the time unit and direction dependent plate movements (or velocities). fig. 5 shows that the majority of turkey is located on the anatolian plate where the arabian, eurasian and the african plates conjunct. the anatolian plate is being squeezed by the movement of the eurasian and arabian plates. thus, anatolia is one of the most seismically active regions of the mediterranean basin. historical records of seismicity, as well as instrumental recordings of earthquakes that have occurred in the past 20 years, reveals the active tectonic features of this area. the north anatolian, east anatolian, aegean (western of southwest anatolia), and burdur-fethiye fault zones, indicate the tectonic diversity and activity of anatolia (gilbert et al., 1994; kahle et al., 1996). between 1988 and 1992, measurements were concentrated during alternate years in western and eastern turkey, with sufficient overlap to allow the integration of the networks, but since the 1994 survey, almost all the stations were reoccupied during each survey (reilinger 1997). in addition to these broad-scale surveys in turkey a small and relatively dense network of stations spanning the various strands of the north anatolian fault in the marmara region was observed in 1990, 1992, 1994 and 1996 (straub et al., 1997). to aid in orbit determination and to strengthen the reference frame, (mcclusky et al., 2000) analyzed data from 50 to 200 globally distributed stations from the international gps service (igs) acquired between june 1992 and april 1999. altogether, their solution included data spanning 9 years from 450 stations, included 189 in the mediterranean region. also, for the southern aegean, the gps velocity field is dominated by a southwest motion relative to eurasia of about 30 mm/year. to better show the possible deformation in the southern aegean (<2 mm/year), and the motion of the southern aegean relative to the surrounding areas, they calculated an euler vector for the anatolia and determined the residual velocities that stations indicated increasing velocities toward the trench and counterclockwise rotation in and around the region. also, gps strain fields and velocity fields for the period 20032006 in the southwest of turkey were indicated by (erdoğan et al., 2009). the measurements, which were performed as periods, were evaluated using daily measures. it was found that the region is affected by a southwestern movement of 15-30 mm/year according to a eurasia-fixed stabilization. strain 24 o. oktar, h. erdoğan vectors show a clockwise rotation in the study area. the velocity field and strain fields are characterized by a system that indicates nondeforming and deforming regions separated by fault zones. in addition, the velocity field is in agreement with location of seismicity, earthquake focal mechanisms, and mapped fault locations. according to the gps velocity vectors there is a shows slower movement in that south compared with the north. this movement of area is agreement with the results obtained in this study (table 1). for the whole of turkey, the last study was made by eren et al., (2009) taking short term observations of the gnss stations and the maximum estimated displacement from the 146 cors-tr stations with respect to the eurasian plate are 19.91 mm/year (southwest direction) and 23.16mm/year (northwest direction) at fini and uder stations. also, tectonic activity inferred from velocity field of gnss measurements for the period 2003-2010 in southwest of turkey were studied by gülal et al., (2013), ıt was found that the anatolian plate is moving to the southwest at a rate of 12–40 mm/yr with respect to the eurasian reference frame. in a different study, 20 cors-tr stations located in the southwest of turkey for period 2008-2010 were studied by gülal et al., (2013). the average movement of the stations with the exception of kamn is 21.7 mm/year in southwest direction. point velocities (vertical) of tusaga-active obtained by eren et al., (2009) support the results obtained in this study. figure 5. tectonic setting of turkey (mcclusky et al., 2000) as reported in this paper, the results of 788 days of linear trend component analyses acquired from the gnss stations are consistent with the results of previous short term, long term and campaign type observations based on tectonic studies on the whole of the anatolian plate and around this region (fig. 5 and table 1). continuous wavelet analysis scalograms were obtained by applying cwt at different scales to gnss coordinate time series of north, east, and up coordinate components. morlet wavelet was used to cwt. the morlet wavelet is both complex and contains more oscillations, and hence the wavelet power combines both positive and negative peaks into a single broad peak. scalogram provides to detect periodicities in time series and this information depends on time. in scalogram, while vertical axis provides scale information, horizontal axis gives time information. percentage of energy for wavelet coefficient increases from blue to red. therefore, while detecting any scale, it should be maxima of energy. it cannot be explained clearly that how many times the periodicity is repeated, nevertheless we ensure the existence in this analysis. also the number of excess periodicity should be regarded as a more important criterion than the high-energy scale. since we have a lot of scalograms, determined periodic movements in them are given in table 2. the scalogram of the up direction of station yuna (a) and north direction of station knya1 (b) coordinate time series using cwt are plotted in fig. 6. figure 6. (a)scalogram of yuna for up signals, (b)scalogram of knya1 for north signals fig. 6 shows that there are 60, 90, and 312 days of periodic movements in up direction of station yuna coordinate time series and 15, 42,76, and 129 days of periodic movements in north direction of station knya1 coordinate time series. in this way, it can be reached that when the period is active when we look at the horizontal axis of the graph. regarding the analysis performed for all gnss stations used in this study, up coordinate time series have the most periodicity whereas east coordinate time series have the least periodicity. table 2. periodic movement times (days) of gnss stations obtained from scalograms. table 2 shows that the up signals of the gnss stations in the central anatolia region of turkey contain more movements in comparison with their southern and northern signals. daily (from 2 to 28 days), monthly (28, 30, 66 days etc.), seasonal (87, 95, 112), semi-annual (150, 170, 190 days etc.), and annual (360, 380, 390 days etc.) periodic movements based on various factors related to the signals have been determined (detail information were given in section periodical analysis). discrete wavelet transform when we analyze gnss daily coordinate time series by cwt, it is not possible to examine in detail the behaviors of the gnss stations. therefore, studying on approximation and detail components of these series obtained from dwt make possible to get more information. also with the help of the dwt, we are able to discuss time series in different scales separately. this process gives us positive opportunities on the analysis of nonstationary gnss coordinate time series. for example, we are able to examine any scale-level component that we want depending on time. dwt was carried out to gnss coordinate time series of north, east, and up coordinates components of stations with multiresolutional analysis. for 25research of behaviours of continuous gnss stations by signal translation, db1 wavelet from daubechies wavelet family was chosen because it is practicable to dwt. considering the number of data, decomposition levels were set 9 for akhr, beys, ciha, kamn, yuna, and aksi1, 6 for aksi2, 7 for knya1, and 8 for knya2. in this way, a and d were obtained with dwt applied for all series. approximations (a) and details (b) of station yuna for up signals are shown in fig. 7. figure 7. approximations (a) and details (b) componenets of station yuna for up signals fig. 7 (a) shows that a1 has a high correlation with the original signal while a9 has a low correlation. in this case correlation between the original signal and approximations is going to be weakened from a1 to a9. besides, a8 and a9 have very low correlation with the original signals. fig. 7 (b) shows that d1 and d2 (especially d1) have vibrations caused from noise at 0. also, these components have the highest frequency changes of the original signal. d1 has the highest frequencies and it decreases from d1 to d9. furthermore, it is clearly seen that d9 has low correlation with the original signal. periodical analysis after the elimination of the linear trend component in the time series for the movements of the cors-tr in the north, east and up directions, the periodical components in the approximation and detail series, which are obtained with dwt, were determined. the periodical components are the long-term periodical changes (low frequencies) which occur in the cors-tr as a result of climatic events, atmospheric and hydrological loading, satellite and reference system errors effecting the stations over a period of time (hours, half a day, semi-annual and annual etc.). in order to obtain the frequencies in the signals of the stations, the series was transformed to frequency domain by fft (eq.1) and the power spectrums pxx(k) were calculated. the frequencies of the periodical movements were calculated for a and d of the north, east and up signals of stations from the power spectrum figures (peak-peaking method). for example, the power spectrum figures of d1-d9 of stations beys are given for the north signals in fig. 8. the calculated frequencies of stations in the north, east and up directions indicated which cors-tr station had many periodical changes or movements. it must be defined that whether a large number of periodic movements, which are shown in fig. 8, is statistically significant. for this purpose, periodical component analysis can be made by modeling the sp(ti) signals with the use of the trigonometric function defined as (erdogan and gülal, 2009). in this equation, the frequency fs of signal is estimated by eq. (1) and (2). these periodical movements were tested for statistical significance. figure 8. power spectrum of d1-d9 of station beys for north signals. during the calculations, the frequencies were used in eq. (15), the as and bs parameters of the periodical changes and covariance matrices of these parameters were estimated by lsm. whether the calculated values are significant is decided by t-distribution/t-test. according to “0” expected value and degrees of freedom f at %95 confidence level, t-distribution values are; 1.96 (f>500,t500,1-0.05/2=1.96) for akhr, beys, ciha, kamn, yuna ve aksi1, 1.97 (f>200,t200,1-0.05/2=1.97) for knya1 and knya2, and 2 (f>60,t60,1-0.05/2=2) for aksi2. the parameters with larger test values than the confidence limit were considered as significant. the insignificant frequencies were excluded from eq. (15) and this procedure was repeated until the significant frequencies were determined. the time of the periodical movements was calculated from significant frequencies and time of significant periodical movements as given in table 3. daily (from 2 to 28 days), monthly (28, 30, 66 days etc.), seasonal (87, 95, 112 days etc.), semiannual (150, 170, 190 days etc.), annual (360, 380, 390 days etc.) and even biennial termed periodic movements based on various factors related to the signals have been determined. it is stated that the gps coordinate time series includes periodic changes (signals) ranging from half-day to years, due to various factors (blewitt et al., 2002; mao et al., 1999; poutanen et al., 2001). (15) 26 o. oktar, h. erdoğan table 3. significant periodical movements for cors-tr stations. seasonal movements are generally related to gravitational excitation, of a thermal origin coupled with hydrodynamics, sources that are indirectly due to geophysical processes or resulting from instrument, or modeling deficiency (dong et al., 2002). gravitational excitation consists of rotational displacements due to seasonal polar motion (ut1), and loading induced displacement caused by solid earth, ocean and atmospheric tides, which are modeled in the raw data processing step. however, the residual ocean tide effects which are still present, and are mainly due to the use of global ocean tide models instead of higher quality local tidal models. the pole tide loading also belongs to gravitational excitation, with the spectrum of mostly annual and chandler wobble periods (409 and 512 days or possibly ~682 days). this long time signal with a large amplitude and nearly 450 days period is described in calais (1999). due to this signal not being seasonal and being larger in east and west components, it is concluded that this is probably caused by errors in the reference systems or satellite orbit or ground guidance parameters according to calais (1999). the deformation caused by pressure field variation, non-tidal sea surface fluctuation, ground water changes in both liquid and solid form, bedrock expansion beneath the gps benchmark, and wind shear are all in the category of being of thermal origin coupled with hydrodynamics. also, other error sources, which generate apparent seasonal variations, such as orbit modeling errors caused by an imperfect reference frame, which is defined through a set of stations that are subject to seasonal variations. there are other important phenomena that can affect the position time series through seasonal signals; these involve tectonicinduced deformation, due to possible seasonal movements in the regional fault slip or regional stress, e.g., slow earthquakes; (miller et al., (2002). the acquired periodical movements and their impact on gnss stations in this region are consistent with other gnss stations throughout the world. this indicates that the recently set up cors-tr network for the whole country is capable of providing stable, reliable, accurate and continuous data for national and international studies that can be conducted in many fields. the results obtained in this section support periodicities acquired from scalogram in section cw analysis. energy spectrums of details and correlations as approximations and details obtained with dwt, there are one or several components to show the best fit to original time series. in this study, detail component, which protect the largest percentage of energy, was determined by calculating the energy spectrum. because the latest level represents the series with the minimum coefficient, high energy at these levels was not considered. the energies of details of north, east, and up coordinate time series of each stations are shown in the same graph. thus, details affecting the coordinate components may be compared to each other. because we have a lot of graphs, only results of yuna and aksi1 are given in fig.9. fig.9 shows that d3 has the highest energy in east and up coordinate time series and d8 has the highest energy in north coordinate time series for yuna stations. considering all energy spectrum graphs, detail component energies of north and east time series are usually similar and the highest energies are obtained in d7 and d8 components. however, the components in which the highest energy obtained have varied in up time series. d3 has the periodicity between 8 and 16 days and d8 has the periodicity between 256 and 512 days. in this way, activities having these periods of specified components effect characteristic of north, east, and up coordinate time series of yuna stations. in conclusion, energies of detail components are generally similar in vertical and especially d7-d8 have the highest energy. however, the components obtained the maximum energy was varied in up coordinate time series. d7 has the periodicity between 128 and 256 days and d8 has the periodicity between 256 and 512 days. figure 9. percentage of energy spectrum for yuna and aksi1 north, east and up signals. 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 level (d1-d9) e n e rg y yuna north, east, and up energies yuna-n yuna-e yuna-u 1 2 3 4 5 6 7 8 9 0 10 20 30 40 50 level (d1-d9) e n e rg y aksi1 north, east, and up energies aksi1-n aksi1-e aksi1-u conclusions in this paper, 788 daily gnss time series were obtained using gamit software to process the gnss observations from seven stations of cors-tr network in turkey. firstly, linear models of north, east and up coordinate components of the stations are formed and the annual velocities are calculated. it was determined that the calculated linear movements were in the north-west direction (13.46 mm/year) for kamn station and in the south-west direction (17.78 mm/year) for the other stations. linear movements in up coordinate time series of the stations are: in (+) direction for akhr, ciha, yuna, and knya2 stations and in (-) direction for beys, kamn, aksi1, aksi2, and knya1 stations. the linear movements obtained for these stations were found to be in agreement with the plate velocity values of the zone where these stations are located. then periodical movements in the time series are investigated by cwt and fft. although we can obtain periodic movements by both cwt and fft, fft provides more clear results. however, the results of both methods support each other. furthermore, correlations between the original time series and details were investigated by using parseval’s theorem. in dwt, different 27research of behaviours of continuous gnss stations by signal decomposition levels are applied to the time series. this provides that we can study and concentrate on which component we want. also, in the analyses of the periodic movements, it was determined that signals of stations contained daily (from 2 days to 28 days), monthly (28, 30, 66 days etc.), seasonal (87, 95, 112 days etc.), semi-annual (150, 170, 190 days etc.), annual (~ 360 days) and even biennial termed periodic movements changing from daily to annual terms. these periodic movements acquired for the stations in this region are consistent with those found for other continuous gnss stations in other parts of the world. in the analyses of correlations, energies of detail components are generally similar in vertical and especially d7-d8 have the highest energy. however, the components obtained the maximum energy was varied in up coordinate time series. the acquired important results from cors-tr for this region are consistent with other gnss stations throughout the world. this indicates that the recently set up cors-tr network for the whole country is capable of providing stable, reliable, accurate and continuous data for national and international studies that can be conducted in many fields. references ahmed, f. (2010). evaluation of gnss as a tool for monitoring tropospheric water vapour. msc thesis, department of earth and space sciences, chalmers university of technology, göteborg, sweden. allen, r. l. & mills, d. w. (2004). signal analysis, time, frequency, scale, and structure. ieee press, usa. bastos, l., osorio, j., landau, h., & hein, g. (1991). the azores gps network, arquipélago. journal of life and earth science, 9, 1–9. blewitt, g., & lavallée, d. (2002). effect of annual signals on geodetic velocity. journal of geophysical research, 107, 2145. bos, m. s., bastos, l., & fernandes, r. m. s. (2010). the influence of seasonal signals on the estimation of the tectonic motion in short continuous gps time-series, journal of geodynamics, 49, 205–209. calais, e. (1999). continuous gps measurements across the western alps, 1996-1998. geophysical journal international, 138(1), 221–230. dong, d., fang, p., bock, y., cheng, m. k. & miyazaka, s. (2002). anatomy of apparent seasonal variation from gps-derived site position. journal of geophysical research, 107. erdoğan, h. & gülal, e. (2009). the application of time series analysis to describe the dynamic movements of suspension bridges. nonlinear analysis: real world applications, 10, 910-927. erdogan, s., sahin, m., tiryakioglu, i., gülal, e., & telli, a. k. (2009). gps velocity and strain rate fields in southwest anatolia from repeated gps measurements. sensors, 9(3), 2017–2034. eren, k., uzel, t., gülal, e., yıldırim, o. & cingoz, a. (2009). results from a comprehensive gnss test in the cors-tr network: case study. journal of surveying engineering, 1-10. gilbert, l. e., kastend, k., hurst, k., paradissis, d., veis, g., billiris, h., hoeppe, w. & schlueter, w. (1994). strain results and tectonics from the aegean gps experiment. eos, transactions, american geophysical union, 75(16). gülal, e., tiryakioğlu, i̇., erdoğan, s., aykut, n. o., baybura, t., akpinar, b., telli, a. k., ata, e., gümüş, k., taktak, f., yilmaz, i̇., öcalan, t., kalyoncuoğlu, ü. y., dolmaz, m. n., elitok, ö. , erdoğan, h. & soycan, m. (2013). tectonic activity inferred from velocity field of gnss measurements in southwest of turkey. acta geodaetica et geophysica, 48(2), 109–121. gülal, e., erdoğan, h., & tiryakioğlu, i̇. (2013). research on the stability analysis of gnss reference stations network by time series analysis. elsevier, 13(8), 1945-1957. kahle, h. g., muller, m. v. & veis, g. (1996). trajectories of crustal deformation of western greece from gps observations 1989–1994. geophysical research letters, 23, 677–680. kellogg, j. n., & dixon, t. h. (1990). central and south america gps geodesy – casa uno, geophysical research letters, 17 (3),195–198. larsen, s. & reilinger, r. (1992). global positioning system measurements of strain accumulation across the imperial valley, california: 1986– 1989. journal of geophysical research, 97(b4), 8865–8876. mao, a., harrison, c. g. a., & dixon, t. h. (1999). noise in gps coordinate time series. journal of geophysical research, 104(b2), 2797–2818. mcclusky, s., bassalanian, a. b., demir, c., ergintav, s., georgiev, i., gurkan, o., hamburger, m., hurst, k., hans-gert, h. g., karstens, k., kekelidze, g., king, r., kotzev, v., lenk, o., mahmoud, s., mishin, a., nadariya, m., ouzounis, a., paradissis, d., peter, y., prilepin, m., relinger, r., sanli, i., seeger, h., tealeb, a., toksoz, m.n., & veis, g. (2000). global positioning system constraints on plate kinematics and dynamics in the eastern mediterranean and caucasus. journal of geophysical research, 105(b3), 5695–5719. miller, m. m., melbourne, t., johnson, d. j. & sumner, w. q. (2002). periodic slow earthquakes from the cascadia subduction zone. science, 295, 5564, 2423. misiti, m., misiti, y., oppenheim, y. & poggi, j. (2002). “wavlet toolbox user’s guide the matworks”, inc. miyazaki, s., tsuji, h., hatanaka, y., abe, y., yoshimura, a., kamada, k., kobayashi, k., morishita, h., & limura, y. (1996). establishment of the nationwide gps array (grapes) and its initial results on the crustal deformation of japan. bulletin of the geographycal survey institute, japan. 42, 27–41. morlet, j., arens, g., fourgeau, e., & giard, d. (1982). wave propagation and sampling theory, 1, complex signal and scattering in multilayered media. geophysics, 47(2), 203–221. pany, t. (2007). navigation signal processing for gnss software receivers. artech house, boston. peter, y., kahle, h.g., cocard, m., veis, g., felekis, s. & paradissis, d. (1998). establishment of a permanent gps network across the kephalonia fault zone ionian, greece. tectonophysics, 294, 253–260. poutanen, m., koivula, h., & ollikainen, m. (2001). on the periodicity of gps time series, proceedings of the iag 2001 scientific assembly, vistas for geodesy in the new millennium. budapest, hungary, 2-7, p.4. qiao, f. (2005). introduction to wavelet. a tutorial in workshop on wavelet application in transportation engineering. texas southern university. reilinger, r. (1997). global positioning system measurements of present-day crustal movements in the arabia–africa–eurasia plate collision zone. journal of geophysical research, 102, 9983–9999. reilinger, r. e., mcclusky, s. c., oral, m. b., king, r. w., toksoz, m. n., barka, a. a., kinik, i., lenk, o., & sanli, i. (1997). global positioning system measurements of present-day crustal movements in the arabia–africa– eurasia plate collision zone. journal of geophysical research, 102(b5), 9983–9999. romagnoli, c., zerbini, s., lago, l., richter, b., simon, d., domenichini, f., elmi, c., & ghirotti, m. (2003). influence of soil consolidation and thermal expansion effects on height and gravity variations. journal of geodynamics, 35(4–5), 521–539. satellite navigation: http://en.wikipedia.org/wiki/satellite_navigation/, 2016 (last accessed may 2016). scherneck, h. g., johansson, j. m., mitrovica, j. x., & davies, j. l. (1998). the bifrost project: gps determined 3-d displacement rates in fennoscandia from 800 days of continuous observations in the swepos network. tectonophysics, 294(3–4), 305–321. straub, c. & kahle, h. g. (1995). active crustal deformation in the marmara sea region, nw anatolia, inferred from gps measurements. geophysical research letters, 22, 2533–2536. straub, c., kahle, h. g. & schindler, c. (1997). gps and geologic estimates of the tectonic activity in the marmara sea region, nw anatolia. journal of geophysical research, 102, 27587–27601. van dam, t. m., wahr, j., & chao, y., leuliette, e. (1997). predictions of crustal deformation and of geoid and sea-level variability caused by oceanic and atmospheric loading. geophysical journal international, 129, 507–517. van dam, wahr, t. j., milly, p. c. d., shmakin, a. b., blewitt, g., lavallée, d., & larson, k. m. (2001). crustal displacements due to continental water loading. geophysical research letters, 28, 651–654. articulo 3.indd earth sciences research journal hydrogeology earth sci. res. sj. vol. 15, no. 2 (december, 2011): 101 108 hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india k. brindha and l. elango department of geology, anna university, chennai600 025, india. corresponding author. e-mail: elango34@hotmail.com, elango@annauniv.edu abstract hydrochemical study was carried out in madhuranthakam located near chennai in tamil nadu, india with an objective of understanding the suitability of local groundwater quality for domestic and irrigation purposes. twenty groundwater samples were collected in february 2002 and analysed for physical and chemical parameters. groundwater in this area was found to be within the desirable bureau of indian standards and world health organisation limits for drinking water. ca-hco 3 was the dominant groundwater type. groundwater in this area was assessed for irrigation purposes on the basis of sodium percentage (na%), magnesium hazard (mh), residual sodium carbonate (rsc), sodium absorption ratio (sar), permeability index (pi) and united states department of agriculture (usda) classification. most of the groundwater samples were suitable for irrigation, except in a few locations (15%) based on mh. overall the groundwater quality was suitable for drinking and domestic purposes and permissible for irrigation activities. resumen un estudio hidroquímico fue llevado en la localidad madhuranthakam cerca chennai en tamil nadu, india, con el propósito de evaluar la calidad de las aguas subterráneas locales para usos domésticos y fines de riego. veinte muestras de agua subterránea fueron recogidas y analizadas en términos de parámetros físicos y químicos en febrero 2002. las aguas subterráneas en esta área fueron encontradas aptas como potables según los limites sugeridos por el consejo de estándares de india y los límites permisibles por la organización mundial de la salud para el agua potable. ca-hco 3 es dominante en las aguas subterráneas. las aguas subterráneas en esta área fueron asignadas para propósitos de riego sobre la base del porcentaje de sodio (na%), magnesio peligroso (mh), carbonato sodico residual (rsc), proporción de absorción de sodio (sar), índice permeabilidad (pi) y la clasificación del departamento de agricultura de los estados unidos (usda). la mayoría de las muestras de las aguas subterráneas son aptas para el riego, excepto en unas pocas ubicaciones (15%), basadas sobre mh. en general, la calidad del agua subterránea fue apta para el consumo, uso doméstico y permisible para las actividades de riego. palabras claves: agua subterraneas, madhuranthakam, calidad de agua potable, calidad de agua de riego, tamil nadu, india. keywords: groundwater, madhuranthakam, drinking water quality, irrigation water quality, tamil nadu, india. record manuscript received: 10/03/2011 accepted for publications: 24/11/2011 introduction groundwater is a precious source of fresh water, being the most distributed form on the earth, excluding the polar icecaps and glaciers. groundwater studies are gaining more importance in the present day as it is used for almost all purposes such as domestic, industrial and agricultural activities in most parts of the world. improper management of this replenishable resource may lead to groundwater contamination and scarcity. when some ions and minerals are present beyond the permissible limit, they become unsuitable for drinking and irrigation purposes, which may be due to natural and also anthropogenic causes. groundwater quality has been given lot of importance and studied worldwide (lahermo and backman, 1999; omo-irabor et al., 2008; baalousha, 2010). several regions in india have encountered degradation in groundwater quality too, due to rapid urbanisation and an exponential increase in population (ramesh and elango, 2005; brindha and elango, 2010; brindha et al., 2011). the present study was carried out with the aim of understanding groundwater quality and its suitability for drinking and irrigation purposes in madhuranthakam, located near chennai in tamil nadu, india. groundwater is the major source for domestic and irrigation practices in this area. there has been an increase in the demand for groundwater due to the growth of the local population. groundwater is usually put to direct use in rural areas without proper monitoring and treatment. groundwater may also become contaminated by the agrochemical products used for irrigation. the groundwater quality in the nearby regions, namely chennai, kancheepuram and chengalpet, k. brindha and l. elango102 has been studied earlier (elango and manickam, 1987; ramesh, 1999; rajmohan et al., 2000; elango et al., 2003a; elango et al., 2003b; kumaresan and riyazuddin, 2006). however, no studies have been carried out in the madhuranthakam region pertaining to groundwater quality. the suitability of groundwater for domestic and irrigation purposes thus had to be determined based on the presence of major ions in the groundwater of this region. the present study, which was carried out in 2002, will serve as baseline data for comparing future groundwater quality. study area the study area is situated 76 km south of chennai city and forms a part of madhuranthakam taluk, kancheepuram district, tamil nadu, india (figure 1). most of the annual rainfall of 1,206 mm is received from october to december and the rest during the southwest monsoon season from june to september. climatic condition in the study area varies from 39oc to 40oc during summer (april to june) and 20oc to 24oc during winter (october to december). the study area is surrounded by a number of tanks which are used for drinking and agricultural purposes. this area is intensively cultivated by pumping groundwater from dug wells as well as surface water resources. geologically this area is characterised by charnockites of precambrian era. outcrops of these rocks are found in many parts of the study area. sandy clay and clayey sand overlay the charnockites with thickness ranging from 1.5 to 5m. groundwater occurs under unconfined condition mainly in the weathered and fractured part of this charnockites in this area. figure 1. location of study area and sampling wells. parameter unit minimum maximum mean ph 7.2 8.2 7.69 ec µs/cm 200 1900 807.6 calcium mg/l 11.84 93.9 52.21 magnesium mg/l 3.65 132.46 24.37 sodium mg/l 27.59 82.76 50.35 potassium mg/l 0 2.74 1.17 carbonate mg/l 6 54.01 28.5 bicarbonate mg/l 30.51 256.27 136.9 chloride mg/l 20.35 32.41 21.91 sulphate mg/l 30.48 44.89 35.64 table 1. statistical summary of physicochemical parameters. hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india 103 sampling and analysis a well inventory survey was carried out in february 2002 to obtain background information about the well population, lithology, well use etc. forty-seven wells were investigated and 20 wells (figure 1) were chosen as representative wells for groundwater sampling based on electrical conductivity (ec). the groundwater samples were collected in clean 500 ml polyethylene bottles. the sampling bottles were soaked in 1:1 diluted hcl solution for 24 hours and washed twice with distilled water before sampling. they were washed again in the field with groundwater sample filtrates. the groundwater samples were collected from bore wells after pumping the water for about 10 minutes. field parameters such as ec and ph were measured in the field using portable digital meters. samples were analysed for major ions in the laboratory using the standard recommended methods (apha, 1998). sulphate concentration in the groundwater samples was analysed using a uv visible spectrophotometer. sodium and potassium content was determined by using a flame photometer and calcium, magnesium, chloride, carbonate and bicarbonate by titration technique. total ions measurement precision was checked by calculating the ion balance error (ibe). the ibe was within ±10%. total dissolved solids (tds) were calculated by using the formula: tds (mg/l) = ec (µs/cm) x 0.64; total hardness (th) was calculated by using: th = 2.497ca + 4.115mg in mg/l. results and discussion table 1 shows the maximum, minimum and mean concentrations of various parameters. the order of cation dominance was ca2+>na+>mg2+>k+ and hco3 ->so4 2->co3 2->clfor anions. the general chemical nature of groundwater can be understood by plotting major cation and anion concentrations on a piper trilinear diagram (piper, 1944). ca-hco3 was the major water type dominant in this area (figure 2). the next dominant water type was mixed ca-na-hco3. drinking water quality groundwater used for domestic purposes, such as drinking and cooking, should be free from toxic chemicals and pathogens. domestic water quality indicates that a particular parameter at a given concentration may be suitable for the human body beyond which it is unsuitable. the concentration of various ions in the groundwater samples was compared with bureau of indian standards (bis, 2003) and world health organisation (who, 1993) standards, which are given in table 2 wherein all the groundwater samples were found to be within the suitable limits. the groundwater in this area was thus seen to be fit for domestic consumption, based on the major ion analysis carried out in this study. the groundwater samples were classified regarding tds and th (tables 3, 4 and 5). most of the groundwater was fresh and suitable for drinking purposes based on tds (freeze and cherry, 1979; davis and deweist, 1966). groundwater in this study area varies from soft to very hard (sawyer and mccarty, 1967). irrigation water quality good quality irrigation water is essential for achieving maximum crop productivity. groundwater suitability for irrigation purpose in this study area was assessed using sodium percentage (na%), magnesium hazard (mh), residual sodium carbonate (rsc), sodium absorption ratio (sar), permeability index (pi) and united states department of agriculture (usda) classification. groundwater in most of the study area was found to be suitable for irrigation. irrigation water having high ec content will affect root area and water flow. groundwater in this area was grouped according to the guidelines figure 2. groundwater type. k. brindha and l. elango104 parameter bis (2003) who(1993) percentage of samples above maximum allowable limit highest desirable limit maximum permissible limit highest desirable limit maximum permissible limit ph 6.5– 8.5 6.5 – 9.2 6.5 – 8.5 6.5 – 9.2 nil th (mg/l) 300 600 100 w500 nil calcium (mg/l) 75 200 75 200 nil magnesium (mg/l) 30 100 50 150 nil chloride (mg/l) 250 1000 200 600 nil sulphate (mg/l) 150 400 200 400 nil sodium (mg/l) 200 nil table 2. comparison of groundwater samples with bis and who standards. table 3. freeze and cherry (1979) classification of groundwater based on tds (mg/l). table 4. david and deweist (1966) classification of groundwater based on tds (mg/l). table 5. sawyer and mccarty (1967) classification of groundwater based on th (mg/l). table 6. suitability for irrigation based on usda classification. tds (mg/l) water type number of samples percentage <1,000 fresh 19 95 1,000 10,000 brackish 1 5 10,000 1,00,000 saline nil nil >1,00,000 brine nil nil tds (mg/l) classification number of samples percentage <500 desirable for drinking 11 55 500 1,000 permissible for drinking 8 40 1,000 3,000 useful for irrigation 1 5 >3,000 unfit for drinking and irrigation nil nil total hardness (mg/l) type of water number of samples percentage <75 soft 2 10 75 150 moderately high 7 35 150 300 hard 7 35 >300 very hard 4 20 ec (µs/cm) salinity class percentage of samples remark on quality <250 c1 5 excellent or low 250-750 c2 45 good or medium 750-2250 c3 50 permissible or high 2250-5000 c4 nil unsuitable or very high hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india 105 established by the united states salinity laboratory (freeze and cherry, 1979) based on ec (table 6). this showed that groundwater in this area was permissible for irrigation. sodium is an important parameter for irrigation water and is denoted as na% which was calculated from the formula given in equation 1 (wilcox, 1955) and all concentrations were expressed in meq/l. the na% in this area is given in table 7. groundwater was suitable for irrigation in 60% of the samples while 40% were permissible to doubtful. ec and na% are plotted in figure 3 which showed that most of the groundwater samples were good for agriculture. the concentration of bicarbonate and carbonate higher than calcium and magnesium will influence the suitability of water for irrigation purposes. the rsc value was computed using the following formula (equation 2) where ions were expressed in meq/l. rsc = (co 3 2+ hco3 -) – (ca2+ + mg2+) however, regarding rsc, all samples fall within the safe category for irrigation. sar is another important parameter for determining the desirability of irrigation water. the sar values were calculated by using equation 3 (richards, 1954), where all the concentrations were given in meq/l. all the groundwater samples collected from this area were excellent on the basis of sar. water used for irrigation can be classified into four typesc1, c2, c3 and c4 based on salinity hazard and s1, s2, s3 and s4 based on sodium hazard. figure 4 shows the plot of groundwater samples grouped on the above basis. most groundwater samples fall under c1s1 and c2s1 which are suitable for irrigation. comparatively few samples were grouped under c3s1 (which is permissible for irrigation). thus the groundwater in this area was seen to be suitable for irrigation based on salinity hazard and sodium hazard. mh for irrigating water was calculated using the formula mh = mg/ (ca + mg) x 100, where concentrations were given in meq/l (szabolcs and darab, 1964). mh above 50 is considered to be unsuitable for irrigation. groundwater had an mh above 50 in 15% of the samples, thus not being fit for irrigation. table 7. suitability for irrigation based on sodium percent. % na suitability for irrigation number of samples percentage <20 excellent 5 25 20-40 good 7 35 40-60 permissible 6 30 60-80 doubtful 2 10 >80 unsuitable nil nil (1) (2) (3) )( 100)( % 22 ++++ ++ +++ + = knamgca xkna na 2 22 ++ + + = mgca na sar figure 3. suitability of irrigation water, based on ec and sodium percent. k. brindha and l. elango106 the suitability of groundwater for irrigation based on pi was calculated using equation 4, where concentrations were in meq/l. class i and class ii waters are considered to be good and suitable for irrigation while class iii water is unsuitable for irrigation (doneen, 1964). figure 5 shows that one groundwater sample was not suitable for irrigation based on pi whereas the rest of the samples were good. in general, the groundwater fall within the permissible category for irrigational use, except for a few locations where it was unsuitable based on mh. conclusion groundwater quality of an area must be studied to understand its suitability for domestic and irrigation purposes. all the groundwater samples collected from the madhuranthakam area, tamil nadu, india, showed that the major ions fall within the permissible range. the dominant groundwater type was ca-hco 3. based on tds, 95% of the groundwater was fresh and permissible for drinking. the groundwater varied from soft to very hard on the basis of th. the suitability of groundwater for irrigation was assessed from na%, ec, rsc, sar, mh, pi and usda classification. the groundwater in this area was seen to be good and suitable for drinking and domestic purposes. however, the groundwater was unsuitfigure 4. irrigation water salinity and alkalinity hazard. figure 5. classification of irrigation water based on permeability index. )( 1003 +++ + ++ + = namgca xhcona pi (4) hydrochemical characteristics of groundwater for domestic and irrigation purposes in madhuranthakam, tamil nadu, india 107 able for irrigation in a few places, based on mh. overall, groundwater in the madhuranthakam area remains usable. the present groundwater quality status must be maintained by taking precautionary measures such as rainwater harvesting, less use of chemical fertilisers and the ongoing monitoring of groundwater quality in this region. acknowledgements the authors would like to thank the department of science and technology-funds for improvement in science and technology (dst-fist) (grant no. sr/fst/esi-106/2010) and the university grants commission-special assistance programme (ugc-sap) (grant no. ugc drs ii f.550/10/drs/2007(sap-1)) for providing financial support which helped in creating facilities to carry out this work. references apha (1998). (1998). standard methods for the examination of water and wastewater, 20th edn, american public health association, washington dc, usa. baalousha, h. (2010). assessment of a groundwater quality monitoring network using vulnerability mapping and geostatistics: a case study from heretaunga plains, new zealand, agricultural water management. 97, 240–246. bis (2003). bureau of indian standards specification for drinking water, is: 10500:91. revised 2003, bureau of indian standards, new delhi. brindha, k. and elango l. (2010). study on bromide in groundwater in parts of nalgonda district, andhra pradesh, earth science india, 3, no. 1, 73-80. brindha, k., rajesh, r., murugan, r. and elango, l. (2011). fluoride contamination in groundwater in parts of nalgonda district, andhra pradesh, india, environmental monitoring and assessment, 172, 481-492. davis, s. n. and dewiest, r. j. (1966). hydrogeology, wiley, new york. doneen, l. d. (1964). water quality for agriculture, department of irrigation, university of calfornia, davis, 48 pp. elango, l. and manickam, s. (1987). hydrogeochemistry of the madras aquifer, india spatial and temporal variation in chemical quality of groundwater, geological society of hong kong bulletin, no. 3, 525-534. elango, l., kannan, r. and senthilkumar, m. (2003a). major ion chemistry and identification of hydrogeochemical processes of groundwater in a part of kancheepuram district, tamil nadu, india, environmental geosciences, 10, no. 4, 1-10.  elango, l., kumar, s. s. and rajmohan, n. (2003b). hydrochemical studies of groundwater in chengalpet region, indian journal of environmental protection, 23, no. 6, 624-632. freeze, r. a. and cherry, j. a. (1979). groundwater, prentice hall inc. englewood cliffs, new jersey, 604 pp. kumaresan, m. and riyazuddin, p. (2006). major ion chemistry of environmental samples around sub-urban of chennai city, current science, 91, no. 12, 1668-1677. lahermo, p. and backman, b. (1999). nitrates in groundwater  in finland: the most endangering quality  problem,  hydrogeology and land use management, 329–333. omo-irabor, o.o.,  olobaniyi, s.b., oduyemi k. and akunna, j. (2008). surface and groundwater water quality assessment using multivariate analytical methods: a case study of the western niger delta, nigeria, physics and chemistry of the earth, parts a/b/c, 33, no. 8-13, 666-673. piper, a. m. (1944). a graphical procedure in the geochemical interpretation of water analysis, transactionsamerican geophysical union, 25, 914-928. rajmohan, n., elango, l., ramachandran, s. and natrajan, m. (2000). major ion correlation in groundwater of kancheepuram region, south india, indian journal of environmental protection, 20, no.3, 188-193. ramesh, k. and elango, l. (2005). groundwater quality assessment in tondiar basin, indian journal of environmental protection, 26, no. 6, 497 –504. ramesh, r. (1999). groundwater quality management: pollution perspectives impacts of urban growth on surface water and groundwater quality, proceedings of iugg 99 symposium hs5, birmingham, july, iahs publ. no. 259, 47-55. richards, l. a. (1954). diagnosis and improvement of saline and alkali soils, us department of agriculture, hand book 60. sawyer, g. n. and mccarty, d. l. (1967). chemistry of sanitary engineers, 2nd edn, mcgraw hill, new york, 518 pp. szabolcs, i. and darab, c. (1964). the influence of irrigation water of high sodium carbonate content of soils, proceedings of 8th isss, trans vol. ii , 802-812. who (1993). guidelines for drinking water quality, vol. 1, 2nd edn, recommendations, who, geneva, 130 pp. wilcox, l. v. (1955). classification and use of irrigation waters, usda, circular 969, washington dc, usa. in this paper, the test-bench for sonic logging tool is proposed and designed to realize automatic calibration and testing of the sonic logging tool. the test-bench system consists of host computer, embedded controlling board, and functional boards. the host computer serves as the human machine interface (hmi) and processes uploaded data. the software running on host computer is designed on vc++, which is developed based on multithreading, dynamic linkable library (dll) and multiple document interface (mdi) techniques. the embedded controlling board uses arm7 as the microcontroller and communicates with host computer via ethernet. the embedded controlling board software is realized based on embedded uclinux operating system with a layered architecture. the functional boards are designed based on field programmable gate array (fpga) and provide test interfaces for the logging tool. the functional board software is divided into independent sub-modules that can repeatedly be used by various functional boards and then integrated those sub-modules in the top layer. with the layered architecture and modularized design, the software system is highly reliable and extensible. with the help of designed system, a test has been conducted quickly and successfully on the electronic receiving cabin of the sonic logging tool. it demonstrated that the system could greatly improve the production efficiency of the sonic logging tool. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): k1 k4 abstract keywords: sonic logging tool, test-bench, embedded controlling board, modularization, layered architecture. research on test-bench for sonic logging tool issn 1794-6190 e-issn 2339-3459 doi:http://dx.doi.org/10.15446/esrj.v20n1.54141 g e o p h y si c s xianping liu, xiaodong ju, wenxiao qiao, junqiang lu, baiyong men, kai zhang, yongchao yao (state key laboratory of petroleum resources and prospecting, china university of petroleum, beijing 102249, p. r. china) email: 2005xinchen2006@163.com record manuscript received: 12/11/2015 accepted for publication: 05/02/2016 how to cite item liu, x., ju, x., qiao, w., lu, j., men, b., zhang, k., & yao, y. (2016). research on test-bench for sonic logging tool. earth sciences research journal, 20(1), k1k4. doi:http://dx.doi.org/10.15446/esrj.v20n1.54140 introduction the test equipment is an essential tool that plays a role in both production and maintenance of corresponding logging tool. in this paper, the test-bench for sonic logging tool is proposed and designed to realize automatic calibration and testing of the sonic logging tool. due to the various structures and functions of different types of logging tools, the test equipment for them are quite differently and must be designed separately. during the past years, early researchers in china has developed several test equipment for logging tools, such as multiple function test tables for well-logging instruments (lu et al., 2005), development of the self-test unit for eilog complete logging equipment (yan et al., 2008). however, this early test equipment is designed to check simply the functioning of corresponding logging tool as a whole, not test for the inner modules of it. with the development of electronic technologies, the functional modules of logging tools are becoming more and more complex. a typical example is the imaging logging tool. therefore, it has become a necessity to develop a new test-bench that verifies not only the tool as a whole but also the various internal modules. against this background, researchers have proposed a new method to develop test-bench systems for logging tools the one that is based on the idea of c/s architecture and building blocks layout (ju et al., 2009). so far, test-bench systems have already been developed in this way for multipole acoustic array logging tool (lu et al., 2010; lu et al., 2012), array induction logging tool (men et al., 2011), micro-resistivity scanning imaging logging tool (yang et al., 2012), and have successfully applied to the test and calibration of tool modules and subs. in this article, the test-bench for sonic logging tool is also designed based on the c/s architecture. the software system for test-bench is developed with highly extensible functional modules considering the hardware configuration of the sonic logging tool and the test items required. 1. structure of the test-bench system as shown in figure 1, the test-bench for logging tools mainly consists of a host computer, an embedded controlling board, and some functional boards. the embedded controlling board is connected to the host computer via ethernet, while to the various functional boards via extended i/o bus (eiob) (ju et al., 2009). in this paper, the functional boards mainly refer to the telemetry communication board, the bus interface board, and the signal acquisition board. among them, the telemetry communication board is provided with two can interfaces, the bus interface board is integrated with the high-speed interconnection k2 xianping liu, xiaodong ju, wenxiao qiao, junqiang lu, baiyong men, kai zhang and yongchao yao tool module bus (tmb) and a signal generator for acoustic signal simulation, the signal acquisition board realizes analog signal acquisition and processing in eight channels. according to the structure of hardware system, the software system for the test-bench is also divided into three parts: the host computer software, the embedded controlling board software, and the functional board software. the host computer software is aimed at managing the internetworking with embedded controlling board and performing a series of routine tasks such as data processing, graphic displaying, data storage. the embedded controlling board software communicates with the host computer and in charges of the control of various functional boards, by the way of command forwarding and data uploading. the functional board software controls the interface circuit of the corresponding board directly according to test assignments. figure 1. structure of the test-bench 2. system design 2.1 software design of host computer the host computer is a high-performance notebook pc, which is connected to the embedded controlling board via ethernet. the host computer sets test parameters and mode by sending commands to the embedded controlling board. by employing microsoft visual c++ 6.0 as the software developing platform, taking advantage of the object-oriented programming model, and using the technique of multiple document interface (mdi), the software for host computer has been developed to include functional modules such as hmi, graphic displaying, network communicating, data processing, plotting and testing. functional modules of the software are shown in figure 2. own to the sophisticated built-in functions for network applications in windows, there is no need to develop tcp/ip protocol stack or network drivers anymore. when testing a certain instrument sub, the software can be switched to corresponding functional module and operation interface by making a choice in the menu. testing yes no yes system initialization establishing a network connection connection successful? data accepted? no start to receive data no yes data upload？ create sending thread sending data distinguish frame type figure 2. functional modules of host computer software 2.2 design of embedded controlling board the embedded controlling board realizes functions such as bottom hardware management, test control, and network communication. it uses arm chip s3c44b0x as the microcontroller, a 10m network interface with isolation in which the ethernet controller rtl8019as is integrated; the 16-bit eiob is designed based on the bank8 memory space of s3c44b0x. the software system of the board mainly consists of hardware drivers, real-time operating system (rtos) based on uclinux (wu et al., 2011) and applications. hardware drivers are used to controlling hardware devices directly; rtos realizes multi-task scheduling, inter-process communication management and provides applications with system functions; applications realize various test functions based on the rtos. the software design for embedded controlling board mainly refers to the transplantation of embedded operating system and the development of drivers and applications. hardware drivers are independent of operating system and stored with system files. upon function calls, they are dynamically loaded into memory in the form of modules. the flow chart of the application program is shown in figure 3. after the system is powered on, global variables are initialized, at first; then device drivers are called, and bottom layer circuit boards are initialized. the network communication module is used to realize socket connection with host computer, where the socket works in the non-block mode and data is sent to host computer by a separate thread. the transmitted data frame is defined to consist of a frame head and data segment. upon arrival of data, the program determines which testing routine the data is sent for by reading and decoding the frame head. then the data frame is sent to target routine to be processed. once the test is completed, the program determines whether there is any data to be returned to the host computer. if no data is to be uploaded, the program returns and waits for new information. otherwise, a new thread will be created to send data to the host computer and after that, the program returns to wait for new information. the newly created thread will run independently to finish data uploading and then exit. figure 3. flow chart of application program for embedded controlling board 2.3 design of functional boards the functional boards provide circuit interfaces for logging tool or subs. figure 4 is the hardware structural of functional boards. all the three functional boards are developed based on fpga and integrated with eiob for communication with embedded controlling board. the telemetry communication board uses cyclone ep1c12q240 by altera as the controller and is integrated with controller area network (can) bus interface. the bus interface board and the signal acquisition board both adopt ep2c20q240c8 as the microcontroller. the bus interface board provides tmb circuit interface and the simulation signal. the signal acquisition board can perform synchronous data acquisition and processing in up to eight channels at the same time. all the functional boards software is designed with vhsic hardware description language (vhdl). from modularization and top-down design strategy, the software is divided into several functional modules that are integrated with the top layer of schematic, making it provided with a simple and intuitional architecture that is easy for maintenance. k3 research on test-bench for sonic logging tool figure 4. hardware structural of functional boards the telemetry communication board, which is integrated with two can interfaces and can be used to control logging tool or electronic master sub in replace of telemetry sub, realizes functions of responding to commands from host computer and storing the uploaded data. figure 5 gives the structural modules of telemetry communication board software. the main functional modules of the software are bus interface logic module, interruption management unit, and can interface module. the bus interface logic module makes an address assignment for the 16-bit eiob and generates the control bus that performs interface module selection as well as read-write control. coordinated by the control bus, the internal multiplexer will connect the chosen module to the eiob. the interruption management unit is used to enable, screen or clear the interrupt request (irq) from can interfaces. the can interface modules convert the bus control timing into the address and data bus multiplex timing of sja1000. besides, they will respond to interruptions from sja1000 and issue irq to the interruption management unit. t_data r_data d_wave switching control eiob bus interface logic module transmitting fifo receiving fifo transmitting-receiving controller dac module serial transmitter serial receiver waveform storage module figure 5. structural modules of telemetry communication board software the bus interface board is integrated with tmb interface for highspeed data transmission between internal modules of the downhole instrument, and an acoustic signal generator. the tmb can be used to simulate master or slave node of internal instrument modules by switching the working mode in software. it mainly performs three tasks. first, it receives and analyzes the serial output command from the electronic master sub and then returns the given data. second, it sends commands to sonic receiving sonde or electronic receiving cabin and returns the acquired data. third, it provides command strings for testing of sonic transmitting sonde where serial commands are needed. the acoustic signal generator uses the digital-to-analog converter (dac) ad5541 as waveform generator and produces simulated acoustic signal for testing of electronic receiving cabin (zhang et al., 2014). the waveform data is stored in fpga in advance. the analog signal generated by the dac is provided to corresponding test module after the processing of filtering and impedance matching. figure 6 gives the software structure of bus interface board, which mainly consists of a bus interface logic module, with functions such as transmitting fifo, receiving fifo, transmitting-receiving controller, serial transmitter, serial receiver, waveform storage module, dac module. the bus interface logic module is mainly used to write in the data to be transmitted, read the received data and control data transmitting and receiving. the transmitting and receiving fifos are configured within fpga, serving as a cache for transmitted or received data storage. the transmitting-receiving controller switches the working mode of the board to make it work in transmitting or receiving mode in turn. the serial transmitter or receiver is constructed of counter and shift register. they perform serial data transmitting and receiving in the charge of the transmitting-receiving controller. the waveform storage module is used to store the simulated waveforms that are prepared in advance. the dac module governs the digital-to-analog conversion process according to the timing of dac chip as well as the clock. figure 6. structural modules of bus interface board software the signal acquisition board can to acquire and process analog signals in the eight channels synchronously, and it is used for voltage acquisition of logging tool power supply in this study. after signals are put in, one of them will be selected by the multi-channel time-sharing multiplexer adg408 to be amplified and filtered and then converted by the analog to digital converter (adc) ad7677. after the conversion, fpga stores the data and uploads it if necessary. as shown in figure 7, the signal acquisition board software consists of bus interface logic module, adc controlling module, fifo. the bus interface logic module receives and analyzes the command string from embedded controlling board, breaking it into information pieces such as channel selection, gain factor, sampling control. the channel selection command is used to specify a channel among the eight channels so as to perform signal acquisition in. the gain factor is the number of times signal being amplified before acquisition begins and it ranges from 0 to 90db with a step of 6db. the sampling control command is sent to the adc controlling module to control the sampling process, e.g. initiating or ceasing the process, using setting sampling depth and interval. in particular, the adc controlling module will assign sampling tasks to ad7677 according to the command received. the fifo is a 16-bit first-infirst-out register that is configured in the fpga. it is used for temporary storage of the acquired data, which will then be uploaded to the embedded controlling board by the bus interface logic module via eiob. adc controlling module fifo eiob . . . signal 2 signal n signal 1 bus interface logic module channel selection gain factor sampling control collected data singal conditioning figure 7. structural modules of signal acquisition board software k4 xianping liu, xiaodong ju, wenxiao qiao, junqiang lu, baiyong men, kai zhang and yongchao yao4 3. test set-up and results the electronic receiving cabin is a critical as well as a widely used module in sonic logging tool, with its performance significantly affects the accuracy of signal acquisition. in the production of the module, the module test procedure should be too much complex and time-consuming if it is performed only by using basic instruments such as an oscilloscope, multimeter, which obviously is not beneficial to the scale production. therefore, in this paper, the test-bench is proposed and developed to realize automatic detection of the sonic logging tool. the cabin is power supplied by the test-bench and provided with simulated acoustic signal and tmb interface by the bus interface board. during the test, the host computer configures test parameters and issues commands to the embedded controlling board via ethernet. after receiving the command, the embedded controlling board forwards it to bus interface board via the eiob. according to the command, the bus interface board will send a simulated acoustic signal to the electronic receiving cabin, read the returned data via tmb interface and finally upload the data to host computer by way of embedded controlling board. figure 8 exhibits the waveform displaying interface of the host computer software that shows the test result of electronic receiving cabin. the waveform tagged d_ wave is the simulated acoustic signal while r_wave the signal returned by the electronic receiving cabin. from the figure, it can be seen that the two waveforms are almost the same in shape despite the existing of a time lag. therefore, the electronic receiving cabin module is considered as qualified, and the user may proceed with other tests. figure 8. test result of electronic receiving cabin conclusions in this paper, a brief introduction is given to the architecture of the test-bench before the system design is discussed in detail. the test-bench system design for sonic logging tool is conducted for the three parts respectively: the host computer software, the embedded controlling board, and functional boards. the host computer software is written in vc++, which is an object-oriented programming language. the embedded controlling board uses arm chip as the microcontroller, its software is designed based on a layered architecture with the uclinux operating system serving as the system kernel. the functional boards are designed based on fpga. the software modules on each board are developed in vhdl language and finally integrated with the top layer of schematic. with the layered architecture and modularized design, the software system is highly reliable and extensible. with the help of designed test-bench, a quick test has been conducted successfully on the particular logging tool module of electronic receiving cabin. to conclude, the test-bench for sonic logging tool has made it easy and time-saving to perform tests on the logging tool or subs, which has a significant meaning in laying foundations for industrial production of the sonic logging tool. acknowledgments: this work is supported by national natural science foundation of china (61102102,11204380,11374371,11134011), national science and technology major project (2011zx05020-009), petrochina innovation foundation (2014d-5006-0307), china national petroleum corporation (2014b-4011,2014d-4105,2014a-3912) and science foundation of china university of petroleum, beijing (2462015yq0516). references: ju, x.d., cheng, x.y., lu, j.q., xu, w., men, b.y. and wu, w.h. (2009). design of test-bench system for logging tools based on embedded structures. well logging technology, 33(3), 270-274. lu, b.p., qin, l. and zhang, q.j. (2005). multiple function test tables for well logging instruments. petroleum instruments, 19(6), 17-19. lu, j.q., ju, x.d., men, b.y. (2012). design of test platform for receiver mandrel of multi-pole array acoustic logging tool. in: fuzzy systems and knowledge discovery (fskd), 2012 9th international conference on ieee, 2660-2663. lu, j.q., ju, x.d., men, b.y., and wu, w.h. (2010). study on test method for transmitter mandrel of acoustic logging tool based on arm. well logging technology, 34(4), 389-392. men, b.y., ju, x.d., qiao, w.x., cheng, x.y. and lu, j.q. (2011). design of test-bench system for the multi-frequency array induction imaging logging tool based on embedded structures. science technology and engineering, 11(11), 2450-2454. wu, w.h., ju, x.d., cheng, x.y., lu, j.q. and deng, l. (2011). design of front-end computer software for logging tool test-bench based on μclinux. journal of china university of petroleum, 35(3), 63-66. yan, z.g., zhang, j.t., qin, l. and zhang, q.j. (2008). development of the self-test unit for eilog complete logging equipment. petroleum instruments, 22(1), 12-14. yang, x.f., ju, x.d. and wu, w.h. (2012). design of the communication software for logging tool test-bench based on b/s model. journal of china university of petroleum, 36(3), 88-92. zhang, k., ju, x.d., lu, j.q. and men, b.y. (2014). design of acoustic logging signal source of imitation based on field programmable gate array. journal of geophysics and engineering, v.11(4), 4500845013. doi: http://dx.doi.org/10.1088/1742-2132/11/4/045008 earth sciences research journal earth sci. res. sj. vol. 17, no. 1 (june, 2013): 5 6 letter from editor dear readers and contributors of the earth sciences research journal, thanks again to be part of this interesting professional journal. with pretty proud i want to present this new issue (17:1, june-2013) with interesting research products coming from antarctic continent, mexico, botswana, india, turkey, nigeria and iran. currently, esrj is covered in the top scientific index and bibliographic database as academic search complete, doaj, fuente académica, georef, scopus, isi web of knowledge, latindex, scielo, ulrich, publindex, e-revistas, and academic onefile. and we expect in the future turn into one of the most referred journal in this field of knowledge. the first article is related to antarctic’s crustal structure. “reference crust-mantle density contrast beneath antarctica based on the vening meinesz-moritz isostatic problem and crust2.0 seismic model” looks for a moho density accurate contrast in spite of the data lackness in this part of the world. then, the authors used a combination of methods to reach their main objective and give us major features of the antarctic’s continental and surronding oceanic tectonic plate configuration. our second paper presents a mexican inventory in seismic specialised software. “contribution of software engineering in earthquake engineering”, which is a suggestion for new development lines, prototypes in seismic software aimed to apply preventive actions related to seismic events in this country. third manuscript comes from botswana and presenting an electrical resistivity study carried out in the thata island, one of the numerous islands in the okavango delta. based on seven data profiles, study indicates the centre of the island had low resistivity values, while those values increase laterally outside the island. results of this work showed how high salinity levels may have formed in near surface aquifers in natural inland delta environments in semi-arid areas. concerned about drinking water in maharashtra province, in india, our authors have deployed an important study about anjani and jhiri rivers catchment area, which provide the jalgaon district. trace elements were analysed in ten groundwater samples and the results were compared to bis drinking water standards. concentration of lead, cadmium and nickel exceed the maximum permissible limit marked by bis. coming from turkey, other study shows the preparation of vulnerability maps to prevent contamination in aquifers. using gis techniques and the drastic method, a vulnerability map is prepared for the basin in which the gümüşhacıköy and merzifon aquifers are located, across many villages characterised by intensive agricultural activity. this analysis ends with the conclusion the basin had a low contamination potential. located in a crude oil prospecting area of ondo state in nigeria, we present a speciation metals study aimed to assess the bioavailability, fate and mobility of as, cd, cr, fe, mn, ni, pb, and v. this paper divides the studied metal in five operationally-defined categories and follows their presence in different places of the area for being concluded that sediment from the crude oil prospecting zone may serve as non-point sources for metals contaminating the coastal system, they have higher metal bioavailability and higher toxicity risk potential than other sites. in nigeria too, an active dumpsite was monitored to test natural attenuation efficiency in reducing contaminants by determining soil and groundwater biogeochemical and engineering properties within and around the site. after measuring unsaturated zone thickness and permeability, and grading the soil as clay or silt according to casagrande plasticity chart, study found that the concentration of contaminants in the soil are very high and inversely proportional to depth. our two last works in this edition were prepared in iran. one of them is focused on the stratigraphy of the permian-triassic boundary in the haftad gholleh area. this boundary of erosional unconformity contains dolomitic limestone and shale and sandstone disconformably located below the triassic sequence and represents a hiatus of about 10 million years. finally, we close this edition with the work “late permian brachiopoda fauna in north-western iran”. this paper found in the area twenty-seven species from 13 brachiopoda genera. the recognised fossil community was compared to brachiopod communities in some regions of iran, and they belong to the orders athyris, rhynconellida, productida and strophomenida. these interesting papers will constrain much better new stratigraphic correlations in this region of the world. carlos a. vargas j. chief editor departamento de geociencias universidad nacional de colombia earth sciences research journal earth sci. res. sj. vol. 17, no. 1 (june, 2013): 41 51 metal speciation in sediments from crude oil prospecting in the coastal area of ondo state, nigeria olu-owolabi bamidele i.1, agunbiade foluso o.2 and adebowale kayode o.3 1 department of chemistry, university of ibadan. email: iromidayobamidele@yahoo.co.uk 2 redeemer’s university, nigeria. email: foagunbiade@gmail.com 3 department of chemistry, university of ibadan. email: adebowale2003@yahooi.com oil prospecting information obtainable from metal speciation is far more valuable regarding the bioavailability, toxicity and fate of metals than information from total metal data. metal speciation on sediment was thus carried out in this study to assess the bioavailability, fate and mobility of as, cd, cr, fe, mn, ni, pb, and v in the crude oil prospecting area of ondo state in nigeria. five operationally-defined metal species were obtained using the sequential extraction method: exchangeable species, carbonate-bound species, iron/manganese-bound species, organic-bound species and residual species. species’ concentration, spread and relative abundance were used to assess their fate, mobility and potential toxicity. the crude oil exploration site and the mouth of the estuary had the highest concentrations of these metals compared to other sites. the exchangeable and the organic-bound high risk species were most predominant in the crude oil exploration sites but their concentration became reduced downstream away from these sites. the residual fraction (the least in the crude oil prospecting site) was the most predominant in the most distant sites downstream. there were indications of self-management of the metals in the coastal system through favourable biogeochemical reactions partitioning metals from high risk species from low risk ones and the non-toxic residual fraction. it could, however, be concluded that sediment from the crude oil prospecting area may serve as non-point sources for metals contaminating the coastal system, they have higher metal bioavailability and higher toxicity risk potential than other sites (which should be curtailed). la información obtenida a través de la especiación de metales es mucho más valiosa en términos de biodisponibilidad, toxicidad y destino que aquella información de sus datos generales. en este trabajo se llevó a cabo una especiación de metales en sedimentos para evaluar la biodisponiblidad, destino y movilidad de as, cd, cr, fe, mn, ni, pb y v en el área de explotación de crudo del estado de ondo, nigeria. se obtuvieron cinco especies de metales operacionalmente definidas con el método de extracción secuencial: especies intercambiables, especies vinculadas al carbono, especies vinculadas al hierro y al manganeso, especies relacionadas a lo orgánico y especies residuales. la relativa abundancia, propagación y concentración de las especies fue utilizada para evaluar su destino, movilidad y riesgo de toxicidad. el sitio de la explotación de crudo y la boca del estuario tienen la mayor concentración de estos metales en comparación con otros lugares. las especies intercambiables y relacionadas a lo orgánico de alto riesgo fueron más predominantes en los sitios de exploración de crudo, pero su concentración se reduce río abajo de estos lugares. la fracción residual (menos en el sitio de exploración de crudo) fue la más predominante en los lugares más distantes aguas abajo. hubo señales de automanejo de los metales en el sistema costero a través de reacciones biogeoquímicas con particiones de metales desde especies de mayor riesgo hacia las de menor riesgo y con la fracción residual no tóxica. podría concluirse que los sedimentos del área de explotación de crudo podrían ser fuentes no puntuales de metales contaminantes en el sistema costero, ya que tienen mayores potenciales de riesgo en biodisponibilidad y alta toxicidad a comparación de otros lugares (lo que debería ser reducido). key words: metal, fate, speciation, biogeochemical reaction, coastal sediment. palabras clave: metales, destino, especiación, reacción biogeoquímica, sedimento costero. record manuscript received: 24/10/2012 accepted for publication: 04/04/2013 abstract resumen introduction sediment contamination poses one of the worst environmental problems in marine ecosystems around the world (chapman et al., 1996; sadiq, 2002; bay et al., 2003; valdes et al., 2005). sediments act as sinks and sources for contaminants in aquatic systems (adams et al., 1992; rowlatt and lovell, 1994; mucha et al., 2003). contaminants may sink in aquatic systems through the precipitation, complexation, adsorption and bioaccumulation of such contaminants which then settle by gravitational force (james, 2002). sediments having high organic matter and fine-grained particles have been reported as showing high affinity for metal sorption, which may serve as better sinks (ip et al., 2006; pekey et al., 2007). however, changes in the environment’s biogeochemical and physical factors can reverse reactions and cause contaminants to be released from sediments. high magnitude contaminant concentrations and low variation regarding sediment composition have been documented in the pertinent literature (shine et al., 1995, beiras et al., 2003, caccia et al., 2003; pekey et al., 2006). these qualities have enhanced the use of sediments as preferred tools for environmental pollution monitoring in aquatic systems (liu et al., 2003). such monitoring always involves investigating metals, organic pollutants and other toxic substances on surface sediments or by coring through sediment depth (szefer and kaliszan 1993; atgın et al., 2000). studies of metals in sediments abound in the literature (vogt, 1990; zwolsman et al., 1996; shin and lam, 2001; korfali and davies, 2004). however, most studies have investigated total metal concentrations in these sediments. this approach has recently been regarded as less appropriate since not all total metal concentrations are bioavailable; only bioavailable metals are regarded as toxic in both soil and sediment (ritchie et al., 2001; korfali and davies, 2004). these bioavailable metal species are the labile ones which are affected by changes in the environment’s biochemistry, thereby representing a departure from using total metals in monitoring the environmental pollution of soil and sediments. the use of metal species in now being adopted (ianni et al., 2000; korfali and davies, 2004). speciation studies of metals have been undertaken by sequential extraction, as initially proposed by tessier et al., (1979), several modifications having mentioned in the literature. the operationally-defined species are: exchangeable species, carbonate-bound species, iron/manganese-bound species, organic-bound species and residual species. the outcome of metal speciation from sequential extraction gives the true pictures of metals’ bioavailability, their potentially adverse effects on the environment, the fate of the metals discharged into the environment and the degree of such metals’ mobility. this study was therefore aimed at evaluating the different species into which metals in coastal sediments from the crude oil exploration area of ondo state have been partitioned. this will serve for measuring the metals’ bioavailability, potential risk, degree of mobility and fate. methodology background this study was conducted in mahin, in the ilaje area of ondo state (lat. 5o 50'n – 6o 09'n and long. 4o 45'e – 5o 05'e); it is an estuary adjacent to the atlantic ocean. the area covered by the study has emerging communities dispersed along throughout the coastal area. the population of the entire state is about 3 million, population growth rate being 2.2% annually. the communities are dependent on water for their economic activities, being a fishing community. several settlements are scattered around the river tributaries that empty into the ocean through this estuary. the area also falls within nigeria’s oil-prospecting states, the so-called the niger delta region, because site 1 (ukua) is an oil-prospecting site hosting the crude oil exploration platform of one of the operators of the nigerian national petroleum corporation (nnpc) joint venture. diverse river courses and streams traverse different settlements and discharge into the ocean (bright of benin, atlantic ocean) through this estuary. ten sampling points were selected (figure 1) along the estuarine part of the coast for evaluation in this study; sediment samples were collected at these sampling points. sampling points were chosen after considering anthropogenic activities in the area representing potential sources of contaminants having a negative impact on the zone. site 2 (awoye), which is the estuary discharge point, also plays a significant role in the chemistry of the coast. several crude oil exploration platforms are located within the vicinity of the discharge point but located in the atlantic ocean. the other eight sites were close to different settlements, at increasing distances from sites 1 and 2. trading and fishing are the major occupations of the people along the coast. all domestic waste is directly discharged via river tributaries onto the coast; there is no waste treatment system in the area. transportation of heavy goods and people in the area and fishing activities are carried out by motorboat, which is a potential non-point source of pollution. sampling and chemical analysis a van veen grab sampler was used for randomly sampling sediments at each sampling site before being composited and kept in nylon. samples were spread on plastic bags and air dried at the laboratory. the samples were ground after air drying and sieved through a 2mm sieve and analysed for as, cd, cr, fe, mn, ni, v and pb. figure 2 gives a schematic representation of the sequential extraction used for this analysis modified from the method proposed by rauret et al., (1989). fractions were shaken in conditions specified in the schematic diagram for each extraction (figure 2) and centrifuged before decanting the supernatants. the exchangeable fraction was obtained after equilibrating for 1 hr at 25oc, after which the residue was used for extracting carbonate-bound metals by shaking the reaction mixture at 25oc for 5 hrs. fe/mn oxide-bound metals were then extracted from the residue with 6 hr equilibration of extractants and sediment samples at 100oc in a hot water-bath shaker. after extracting fe/mn oxide-bound metal species, the organic-bound fractions were obtained by shaking the reaction mixture at 25oc for 30 min. the final fraction (i.e. residual fraction) was obtained by wet oxidation at 100oc for 2hours. the chemicals used for the extraction are represented in the flow-chart (figure 2). all reagent used were analytical grade and the necessary quality assurance steps were observed. the supernatants so obtained were quantitatively transferred into volumetric flasks and made up to 25ml each with 1m hno3 before analysing for heavy metals using an atomic absorption spectrophotometer (aas, buck scientific model 205a). analysis was carried out in triplicate and the mean values calculated. results and discussion metal concentration in sediments table 1 presents total metal species concentration results for the sediment samples from the ten sites in the coastal area. the results showed mn to be the most abundant metal followed by fe, this being consistent with earlier studies carried out along this coast (adebowale et al., 2008a; agunbiade et al., 2010). ni and cr ranked next in concentration. metal concentrations had the highest values at the discharge point into the atlantic ocean at awoye, followed by the crude-oil prospecting site (ukpua) which had the highest as and ni concentration. this trend may have been attributed to crude oil exploration in this region. studies have shown that metal concentrations in the mouth of an estuary are generally higher than in the middle estuary (grabemann et al., 1997; guieu et al., 1998; ip et al., 2007). the spread of metals in the different species obtained from sequential extraction is useful in understanding coastal sediments’ potential toxicity and the impact of activities along the coast, the fate and mobility of metals rather than information olu-owolabi bamidele i., agunbiade foluso o. and adebowale kayode o. sites total metal concentration (µg/g) as cd cr fe mn ni pb v ukpa 0.84±0.06 1.26±0.22 16.4±1.1 349±10 705±28 15.8±0.9 5.46±0.4 5.04±0.6 awoye 0.77±0.05 1.73±0.21 18.7±2.1 418±22 818±33 15.5±0.6 6.81±0.8 6.42±0.7 odonla 0.64±0.07 0.89±0.11 14.4±0.9 343±16 713±34 13.2±1.1 5.04±0.4 3.69±0.4 ikorigho 0.77±0.06 0.89±0.10 14.7±1.5 322±18 749±25 14.5±1.2 4.24±0.3 4.19±0.3 ojumole 0.66±0.09 0.71±0.06 14.1±1.3 296±15 714±21 13.5±0.8 3.66±0.2 4.43±0.1 obenla 0.76±0.10 0.92±0.12 16.8±1.1 287±8 675±27 13.5±1.0 4.14±0.4 5.14±0.5 ilepete 0.62±0.05 0.79±0.11 15.4±1.2 299±11 656±33 14.3±1.2 3.32±0.1 4.16±0.2 ilowo 0.63±0.07 0.86±0.07 14.5±2.0 294±14 659±35 14.3±1.3 3.34±0.2 3.56±0.1 ibijinmi 0.69±0.08 0.99±0.08 16.0±1.9 328±15 727±40 14.6±0.9 4.59±0.1 4.14±0.3 idi ogba 0.74±0.05 0.99±0.07 16.1±1.8 356±19 723±38 15.6±1.1 4.49±0.3 4.24±0.4 table 1: total concentration of metal species in sediment from the ondo coastal area figure 1. map of the coastal area of ondo state and the sampling points. the sequence and labelling of the sites is as follows: site 1 (ukuakaba); site 2 (awoye); site 3 (odonla); site 4 (ikorigho); site 5 (ojumole); site 6 (obenla); site 7 (ilepete); site 8 (ilowo); site 9 (ibijinmi); site 10 (idi ogba). metal speciation in sediments from crude oil prospecting in the coastal area of ondo state, nigeria obtainable from total metal concentrations. box and whisker presentation of data was used for assessing the spread of the following operationally-defined species: exchangeable-species, carbonate-bound species, fe/mn-bound species, organic-bound species and residual species. exchangeable metal species exchangeable metal species as presented by the extraction from sediment method was the most loosely-bound and mobile metal species. it could be extracted from samples having aqueous solutions at ph 7 with only a change in extractant ionic strength; it was therefore the most bioavailable one. at high concentration it is an indicator of high impact, fresh discharge and increased environmental risk for the ecosystem (álvarez et al., 2002). the box and whisker presentation of this specie is presented in figure 3. as and cd average concentration were almost equal, but their spread was different; cd had the wider spread. the two metals ranged from 0.4µg/g to 0.1µg/g in this fraction. ni and cr were other metals having close concentration ranges and nearly equal median values. ni varied more widely than cr, both metals having values ranging from above 2µg/g to less than 3.5µg. the plot for pb and v also showed nearly equal median values for both metals. pb plots skewed towards the lower concentration and had a clearly wider range than v. the plot for fe and mn indicated that their concentration and variation was high in the exchangeable fractions in the coastal sediments from the coastal sediments. exchangeable fraction concentrations of the studied metals indicated increased exposure risk potential from the sediments. the spread was wide and caused by high values at crude oil exploration sites, estuary discharged points and the reduction as distance increased away from these points. this was consistent with the results of earlier studies revealing the role of the environment’s biogeochemistry which has selfpurification ability (agunbiade et al., 2010). such self-purification lowers the concentration of the highly mobile, more toxic fractions from the sediments as distance increases away from the aforementioned sites. carbonate-bound species the extraction procedure for carbonate species in sediments requires a reduction in ph to 5 from the residue after exchangeable species extraction. ph reduction liberates co2 and enhances metal species’ mobility and bioavailability in this matrix. it implies that the fate of carbonatebound species metals is dependent on an ecosystem’s prevailing ph. jarvie et al., (1997) documented that partitioning metals in the carbonate phase is enhanced by increased ph and the possibility of having a river with underlying carbonate, sedimentary rock. reduced ecosystem ph will increase this species’ mobility and bioavailability and the risk of affecting an ecosystem without which a metal is safe. jain (2004) reported that metals in carbonate species are in the medium risk range compared to high risk metal species. the outcome of carbonate-bound species extraction and their spread across this coast is presented in figure 4. as and cd results were closely related, as witnessed in exchangeable fractions, only their concentration and spread being lower than the result for exchangeable species. the lower concentration implied that as and cd were less scavenged by the carbonate. outlier values were observed at awoye (site 2) for the metals which were in the upper extreme, indicating the estuary discharge point’s positive role in cleaning-up metals from the coastal region by partitioning cd discharge into the carbonate fraction and reducing its bioavailability. carbonate species in sediment were less affected by environmental changes such as naturally and commonly occurring microbial activities (korfali and davies, 2004). this species was therefore scavenging some metal contaminants and would stabilise them for a long time once the system’s ph was not lowered. comparing the carbonate species (figure 4) to the exchangeable species (figure 3) showed lower concentrations (median value) and spread of carbonate species than exchangeable ones. however, upper extreme outliers witnessed for the carbonate bound metals at awoye (site 2) represent positive results. it is therefore very evident that the chemistry of the awoye coastal environment (the biogeochemical reactions in it) significantly influenced contaminant status by events occurring in this estuary and were favourable for reducing metal mobility, their consequent bioavailability and toxicity, thereby further confirming this coastal environment’s self-purification ability. fe/mn-bound species metal affinity for fe and mn oxides by the adsorption of such metals to precipitate fe and mn oxides has been documented in the pertinent literature (dekov et al., 1997; guieu et al., 1998; korfali and davies, 2004). fe and mn oxide precipitates have active sites adsorbing metals and reducing their mobility and availability for living tissues (ip et al., 2007). the extraction procedure also reflected a condition which may not occur naturally and frequently, indicating this species’ degree of stability in the ecosystem, medium range exposure risk and their reduced toxicity. the fe/mn-bound metal species trend in this study area’s coastal sediment is presented in figure 5. extreme upper concentration outliers were witnessed for most metals at site 6 (obenla), except for mn, pb and ni. arsenic and nickel had a high concentration outlier at site 9 (ibijinmi), therefore implying that the geochemistry for sites 6 and 9 was enhancing the scavenging of these metals from the ecosystem into this less bioavailable and relatively safe fe/mn bound fraction. however, fe/ figure 2: a schematic diagram of the method used for metal speciation in sediment samples, modified from rauret et al., 1989 olu-owolabi bamidele i., agunbiade foluso o. and adebowale kayode o. 1111n = mnfe 200 180 160 140 120 100 80 60 40 11111111n = vpbnicr 3.5 3.0 2.5 2.0 1.5 1.0 .5 0.0 1111n = cdas .4 .3 .2 .1 0.0 figure 3: box and whisker presentation of 8 exchangeable metal species in sediment samples from all sites. the units of the metals are in µg/g. the vertical bars in the box are median values, the ends of the whiskers are the extreme data while the edges of the boxes present half that data point between the medians and extremes. 1111n = cdas .6 .5 .4 .3 .2 .1 -.0 11111111n = vpbnicr 6 5 4 3 2 1 0 2 2 2 1111n = mnfe 300 200 100 0 2 2 figure 4: box and whisker presentation of 8 carbonate-bound metal species in sediment samples from all sites. the units of the metals are in µg/g. the plot description is as that presented in figure 3. metal speciation in sediments from crude oil prospecting in the coastal area of ondo state, nigeria mn bound metal species and the organic-bound ones in sediment could be remobilised by microbial activity (jones and turki, 1997; korfali and davies, 2004). micro-organisms having reducing ability in sediment were able to reduce insoluble fe (iii) oxides to soluble fe (ii) oxides, thereby releasing the associated metals into solution; this metal fraction’s fate and mobility and its resulting toxicity was thus determined by these microbes’ activities and other prevailing biogeochemical reactions. the concentration and spread of the metal followed a trend similar to that of other fractions but geochemical activity at site 6 and 9 complimented the role played by the estuary discharge point to ensure contamination management in this coastal zone. organic-bound specie figure 6 presents the outcome for organic-bound metal species in coastal sediments. concentration and variation were higher than that for the other species studied thus far. the highest set of organic species was observed at site 1 (ukpa) and site 2 (awoye), such wide variation being due to the reduction in value with increasing distance at these two sites. this would imply that metals discharging into the coast may have formed chelates with organic ligands possibly arising from ukpa crude oil exploration and partition into the sediment phase as organic-bound metal species. the bioavailability of metals bound to organic complexes is affected by microbial activity in the sediment. oxidising micro-organisms can oxidise the organic moiety into carbon dioxide and release associated metals, thereby increasing their mobility and toxicity. likewise, organic complex formation by metals can increase their solubility in a lipophilic system and enhance their transport into living tissues for bioactivity. organic-bound species’ concentration and spread pattern are thus indicators of high risk factors associated with anthropogenic activities in an area, especially in regions associated with the discharge of organic ligands. these organic and exchangeable fractions should be controlled, regardless of a coastal environment’s absorbing capacity so as not to over-burden it. residual fraction the residual metal species is an indicator of metals natural to a site and is neither bioavailable nor mobile; it is thus not a toxic fraction. this fraction is stable and can only be obtained from samples by rigorous acid oxidation which does not occur in nature. this species’ high value relative to that of others is a positive development, indicating safe, non-toxic environmental media and could also be an expression of long-term metal deposition. metals in the non-residual fraction may be indicators of temporary fresh discharge as against distant time release (ip et al., 2007). the results of residual fraction for metals in ondo coastal sediment is presented in figure 7. the concentrations of the metals’ residual fractions increased away from sites 1 and 2. the plot shows higher concentration outliers at sites 9 and 10 except for v and mn. it is evident that the metals in distant sites 9 and 10 were not freshly discharged and were less associated with anthropogenic activity. site 9 and 10 were relatively safer than other sites studied. this was reflected in earlier studies on this coast (adebowale et al., 2008a, adebowale et al., 2008b). the increasing concentration of residual fraction away from sites 1 and 2 lend supported the reported self-management structure found in the coastal environment. metal species’ relative abundance the trend of metal species relative to each other at the sites was evaluated with percentage-based bar charts. figure 8a – 8j presents the results for the relative abundance of metal species within the five fractions investiga1111n = cdas .3 .2 .1 0.0 6 9 6 11111111n = vpbnicr 5 4 3 2 1 0 6 2 8 6 1111n = mnfe 180 160 140 120 100 80 60 40 6 fe/mn-bound figure 5: box and whisker presentation of 8 fe/mn-bound metal species in sediment samples from all sites. the units of the metals are in µg/g. plot description is as presented in figure 3. olu-owolabi bamidele i., agunbiade foluso o. and adebowale kayode o. organic-bound figure 6: box and whisker presentation of 8 organic-bound metal species in sediment samples from all sites. the units of the metals are in µg/g. plot description is as presented in figure 3. 1111n = cdas .7 .6 .5 .4 .3 .2 .1 0.0 11111111n = vpbnicr 6 5 4 3 2 1 0 1111n = mnfe 200 180 160 140 120 100 80 60 40 10 2 11 residual-bound figure 7: box and whisker presentation of 8 residual metal species in sediment samples from all sites. the units of the metals are in µg/g. plot description is as presented in figure 3. 1111n = cdas .6 .5 .4 .3 .2 .1 9 10 9 10 11111111n = vpbnicr 6 5 4 3 2 1 0 9 10 9 10 9 10 1111n = mnfe 200 180 160 140 120 100 80 60 40 9 10 metal speciation in sediments from crude oil prospecting in the coastal area of ondo state, nigeria 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8c: relative abundance of metal species in odo nla overall, the relative abundance results for metal species at sites 1 to 3 revealed that the sediment from these sites had a negative impact on the coastal area. despite this result, however, notable amounts of carbonate-bound fractions of v, cd, cr and mn were witnessed at site 2. this would definitely have reduce these metals’ mobility, bioavailability and resultant toxicity. by contrast, carbonate species of extremely toxic metals like as, cd, and pb were least common at site 1 followed by their fe/mn-bound species. these three metals being toxic could have been preferred in carbonateand fe/mn-bound species, especially at site 1 for discharged metal management. site 3 witnessed a slight increase in residual fraction compared to other sites, thus far discussed particularly for as, cd, cr, fe and ni which were above 25% of total metal concentration. the exchangeable fraction predominated at site 4 (figure 8d), especially for metals like as, cd, fe and pb while the organic fraction was more reduced than the ratio obtained in earlier sites but relatively higher than that for other fractions. these trends regarding exchangeable and organic fractions were both risk-creating situations for this site. the residual fraction in this site was further reduced than that at earlier sites. there was a very high likelihood that there were recent discharges of metals from domestic waste disposal or other related anthropogenic activities. closer distribution of relative abundance of other metal species was witnessed for other fractions at this site than at other sites. a similarly close relationship concerning the relative abundance of metal species was witnessed at site 5 (figure 8e) where the five metals species were closely related for all the metals. the residual fraction however was more increased than at the preceding site (site 4) and was so at all other sites as the distance increased away from the crude-oil exploration site and the estuary discharge point. 0% 20% 40% 60% 80% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8d: relative abundance of metal species in ikorigho ted at the ten sites. it is evident that the organic bound (oxidisable species) and the exchangeable fraction were the most dominant species relative to others in sites 1 to 3 (figure 8a – 8c) while the residual was the lowest in sites 1 and 2 but increased slightly at site 3. this would imply that metals in sediment from sites 1 to site 3 were highly mobile and could cause a high risk to the ecosystem. it indicated fresh or recent discharge of the metals. these two sites have been reported as playing significant roles in the chemistry of the area and the coast’s environmental status (adebowale et al., 2008b). crude oil exploration at site 1 (ukpa) was therefore established as point source of contaminants affecting the coastal environment and the metals from this particular site were found in the free ion state in the water column while the sediment was the readily available exchangeable fraction. all the above increased the toxicity tendency of the metals at this site. the organic fraction was found abundant in these three locations, even the exchangeable fraction which was likely associated with formation of complexes (ligands) between the organic compounds from the crude-oil with metals. the organic fraction was also a potentially risky species to the biota because the formation of complexes by metals would increase the metals’ solubility and their mobility into lipophilic systems which are common in living tissues and enhance the formation of bonds in living tissues by the complexes so formed. thus, organic fraction formation increases metal mobility into living tissues, the food chain and consequently their bioaccumulation and bioactivity. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% as cd cr fe mn ni pb v r el at iv e p er ce nt ag es (% ) metal species exchangeable carbonate fe/mn bound organic residual figure 8a: relative abundance of metal species in ukpua 0% 20% 40% 60% 80% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8b: relative abundance of metal species in awoye olu-owolabi bamidele i., agunbiade foluso o. and adebowale kayode o. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8g: relative abundance of metal species in ilepete the reduction in the trend of exchangeable fraction concentration witnessed at site 7 became further reduced at site 8 (figure 8h) except for pb as was the case with site 7. the residual fraction also increased especially for as, cd, cr, and ni. pb exhibited a high preference for the reducible fraction similar to that obtained at site 7 while the carbonate specie of as in this site was similar to that of site 7. it was evident that this coastal site’s alkalinity was essential for management of as present in the area. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8h: relative abundance of metal species in ilowo the relative percentage abundance of the metal species at sites 9 and 10 (figure 8i and 8j respectively) clearly revealed the favourable roles played by the biogeochemical reactions natural to this environment. the metals discharged into the environment have been demobilised into the residual fraction at these two sites. this showed why the safe, non-bioavailable residual fraction predominated at these coastal sites. the trends observed for all the metal fractions in the sediment from these sites agreed perfectly with the results expressed earlier in this study which supported the selfpurification hypothesis. of particular interest was cd and pb management at these two sites as they have been found in high quantities in other sites referred to by previously discussed results. the organic and exchangeable fractions which were higher at other sites reduced in these sites while the carbonateand fe/mn-bound species maintained their relative abundance, except in cd. thus, the relative abundance of the metals at these two sites 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8e: relative abundance of metal species in ojumole the trend of the species at site 6 (figure 8f ) was completely different from all other sites. the less common fraction at most other sites (fe/mn bound fraction) was the predominant specie at this site. the relative abundance of as, cd, pb and v was notably high in this fraction while that of cr, mn, fe and ni increased more than at other sites discussed earlier. it was therefore evident that this site’s biogeochemistry favoured these metals precipitation into the reducible fraction with resultant favourable reduction of metal mobility, bioavailability and toxicity. the organic and the exchangeable fractions reduced at this site while the carbonate-bound fraction was the least observed for most metals. the residual fraction only increased slightly in the site to more than 20%, except for v. this residual quantity was maintained in site 7 (figure 8g) with increased residual fraction for v over the previous site. organic fractions for cd, cr, fe, mn and ni predominated at the site while as and v existed in the carbonate bound fraction which was positive for managing these metals at that site. the exchangeable fraction in site 7 was more reduced than that observed in earlier sites except for pb which increased more than the preceding site. pb in this site exhibited a preference for the fe/mn-bound at this site while the other metals had the lowest concentration in the fe/mn-bound fraction. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8f: relative abundance of metal species in obenla metal speciation in sediments from crude oil prospecting in the coastal area of ondo state, nigeria indicated that the sites were more risk-averse zones on the entire coast and less stressed in the ecosystem. 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% r el at iv e p er ce nt ag e (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8i: relative abundance of metal species in ibijinmi 0% 20% 40% 60% 80% 100% r el at iv e p ec en ta ge (% ) as cd cr fe mn ni pb v metal species exchangeable carbonate fe/mn bound organic residual figure 8j: relative abundance of metal species in idi ogba conclusion this study has assessed five operationally-defined metal species: exchangeable species, carbonate-bound species, iron/manganese-bound species, organic-bound species and the residual species in sediments from the ondo coastal area using sequential extraction. the crude oil exploration site and the mouth of the estuary had the highest concentrations of these metals compared to other sites. mn and fe had the highest concentrations of all the eight metals investigated. the high risk, exchangeable and organic-bound species were the most predominant metal species in the crude oil exploration and estuary discharge sites. there were reductions in the concentrations of highly mobile species away from these two sites to other sites downstream. the reduction was complimented by increased low risk species. the residual fraction occurring least in the crude oil prospecting site was the most predominant in the farthest downstream sites. this would thus indicate self-management of metals in some of the sites within the costal system through favourable biogeochemical reactions partitioning the metals from the high risk species to the low risk ones and the non-toxic residual fraction. it could thus be concluded that the fate of the metals in the sediments from this coastal sediment is from highly mobile, bioavailable species, exchangeable and organic-bound, to the less bioavailable, less toxic species, carbonate-bound and residual species. the crude oil prospecting area may serve as a non-point source of metal contaminant to the coastal system. it has higher metal bioavailability and higher toxicity risk potential than the other sites and should be curtailed. references adams, w.j., kimerle, r.r., barnett, j.w., 1992. sediment quality and aquatic life assessment. environmental science and technology 26 (10), 1864–1875. adebowale, k. o., agunbiade, f. o., olu-owolabi, b. i. 2008a. impacts of natural and anthropogenic multiple sources of pollution on the environmental conditions of ondo state coastal water, nigeria. electronic journal of environmental, agriculture and food chemistry 7(4): 2797 – 2881. adebowale, k. o., agunbiade, f. o., olu-owolabi, b. i. 2008b. fuzzy comprehensive assessment of metal contamination of water and sediments in ondo estuary, nigeria. nigeria. chemistry and ecology 24 (4), 269 – 283. agunbiade, f. o., olu-owolabi, b. i., adebowale, k. o. 2010. seasonal and spatial variations analysis of pollution status of ondo coastal environment nigeria using principal component analysis. geochemical journal 44 (2), 89 – 98. álvarez, e.a., mochón, m.c., sánchez, j.c.j. and rodríguez, m.t. 2002. heavy metal extractable forms in sludge from wastewater treatment plants. chemosphere 47(7), 765-775 atgın, r.s., el-agha, o., zararsız, a., kocatas, a., parlak, h., tuncel, g., 2000. investigation of the sediment pollution in izmir bay: trace elements. spectrochimica acta part b 55, 1151–1164. bay, s. m., zeng, e. y., lorenson, t. d., tran, k. d., alexander, c. 2003. temporal and spatial distributions of contaminants in sediments of santa monica bay, california. marine environmental research 56, 255 – 276. beiras, r., bellas, j., ferna´ndez, n., lorenzo, j.i., cobela-garcı´a, a., 2003. assessment of coastal marine pollution in galicia (nw iberian peninsula); metal concentrations in seawater, sediments and mussels (mytilus galloprovincialis) versus embryo-larval bioassays using paracentrotus lividus and ciona intestinalis. marine environmental research 56, 531–553. caccia, v.g., millero, f.j., palanques, a., 2003. the distribution of trace metals in florida bay sediments. marine pollution bulletin 46, 1420 – 1433. chapman, p. m., paine, m. d., arthur, a. d., taylor, l. a., 1996. a triad study of sediment quality associated with a major untreated marine sewage discharge. marine pollution bulletin 32, 47 – 64. dekov, v.m.; komy, z.; araujo, f. and van grieken, r., 1997. chemical composition of sediments, suspended matter, river water and ground water of the nile (aswan-sohag traverse). science of the total environment 201, 195 – 210. grabemann, i., uncles, r. j., krause, g., and stephens, j. a., 1997. behaviour of turbidity maxima in the tamar (u.k.) and weser (f.r.g.) estuaries. estuarine, coastal and shelf science, 45, 235-246. guieu, c., martin, j.m., tankere, s.p.c., mousty, f., trincherini, p., bazot, m., 1998. on trace metal geochemistry in the danube river and western black sea. estuarine coast. shelf sci. 47, 471 – 485. ianni, c., magi, e., rivaro, p., ruggieri, n., 2000. trace metals in adriatic coastal sediments: distribution and speciation pattern. toxicol. environ. chem. 78, 73–92. ip, c. c. m., li, x. d., zhang, g., wai, o. w. h., li, y. s., 2007. trace metal distribution in sediments of the pearl river estuary and the olu-owolabi bamidele i., agunbiade foluso o. and adebowale kayode o. surrounding coastal area, south china. environmental pollution 147 (2), 311-323. jain, c.k. 2004. metal fractionation study on bed sediments of river yamuna, india. water research 38(3), 569-578 james, i.d. 2002. modelling pollution dispersion, the ecosystem and water quality in the coastal waters: a review. environmental modelling and software 17, 363 – 385. jarvie, h.p., neal, c., leach, d.v., ryland, g.p., house, w.a., robson, a.j., 1997. major ion concentrations and the inorganic carbon chemistry of the humber rivers. sci. total environ. 194y195, 285–302. jones, b., turki, a., 1997. distribution and speciation of heavy metals in surficial sediments from the tees estuary, northeast england. mar. pollut. bull. 34, 768–779. korfali, s. i., davies, b. e. 2004. speciation of metals in sediment and water in a river underlain by limestone: role of carbonate species for purification capacity of rivers. advances in environmental research 8; 599–612. liu, w.x., li, x.d., shen, z.g., wang, d.c., wai, o.w.h., li, y.s., 2003. multivariate statistical study of heavy metal enrichment in sediments of the pearl river estuary. environmental pollution 121, 377388. mucha, a.p., vasconcelos, m.t.s.d., bordalo, a.a., 2003. macro benthic community in the douro estuary: relations with trace metals and natural sediment characteristics. environmental pollution 121, 169– 180. pekey, h. 2006. the distribution and sources of heavy metals in izmit bay surface sediments affected by a polluted stream. marine pollution bulletin 52; 1197–1208. rauret, g., rubio, r., lopez-sanchez, j. f., casassas, e. 1989. specific procedure for metal solid speciation in heavily polluted river sediments. international journal of environmental analytical chemistry. 35(2), 89 — 100. ritchie, j.m.; cresser, m. and cotter-howells, j. 2001. toxicological response of a bioluminescent microbial assay to zn, pb and cd in an artificial soil solution: relationship with total metal concentrations and free ion activities. environmental pollution 114, 129 136. rowlatt, s.m., lovell, d.r., 1994. lead, zinc and chromium in sediments around england and wales. marine pollution bulletin 28 (5), 324–329. sadiq, m. 2002. metal contamination in sediments from a desalination plant effluent outfall area. the science of the total environment 287, 37 – 44. shin, p.k.s., lam, w.k.c., 2001. development of a marine sediment pollution index. environ. pollut. 113, 281–291. shine, j.p., ika, r.v., ford, t.e., 1995. multivariate statistical examination of spatial and temporal patterns of heavy metal contamination in new bedford harbour marine sediments. environmental science and technology 29, 17811788. szefer, p., kaliszan, r., 1993. distribution of elements in sediment cores of the southern baltic from the point of view of principal component analysis. studia 1 materialy oceanogeczne nr 64. mar. pollut. 1, 95– 102. tessier, a.; campbell, p.g.c. and bisson, m., 1979. sequential extraction procedure for the speciation of particulate trace metals. anal. chem. 51, 844 – 851. valdes, j., vargas, g., sifeddine, a., ortlieb, l., guinez, m. 2005. distribution and enrichment evaluation of heavy metals in mejillones bay (23os), northern chile: geochemical and statistical approach. marine pollution bulletin 50, 1558 – 1568. vogt, n.b., 1990. multivariate ecotoxicological mapping of the relationships between sediment chemical composition and fauna diversity. the sci. total environ. 90, 149–161. zwolsman, j. j., van eck, g. t. m. and burger, g. 1996. spatial and temporal distribution of trace metals in sediments from the scheldt estuary, south-west netherlands. estuarine, coastal and shelf science 43, 55 – 79. metal speciation in sediments from crude oil prospecting in the coastal area of ondo state, nigeria this study uses experimental triaxial tests with monotonic loading to develop empirical relationships to estimate undrained critical shear strength. the effect of the fines content on undrained shear strength is analyzed for different density states. the parametric analysis indicates that, based on the soil void ratio and fine content properties, the undrained critical shear strength first increases and then decreases as the proportion of fines increases, which demonstrates the influence of fine content on a soil’s vulnerability to liquefaction. a series of monotonic undrained triaxial tests were performed on reconstituted saturated sand-silt mixtures. beyond 30% fines content, a fraction of the silt participates in the soil’s skeleton chain force. in this context, the concept of the equivalent intergranular void ratio may be an appropriate parameter to express the critical shear strength of the studied soil. this parameter is able to control the undrained shear strength of non-plastic silt and sand mixtures with different densities. este estudio utiliza evaluaciones experimentales triaxiales con cargas repetitivas para desarrollar relaciones empíricas y estimar la tensión crítica de corte bajo condiciones no drenadas. el efecto de contenido de finos en la tensión de corte sin drenar se analizó en diferentes estados de densidad. el análisis paramétrico indica que, basado en la porosidad del suelo y las propiedades del material de finos, la tensión de corte sin drenar primero se incrementa y luego decrece mientras la proporción de finos aumenta, lo que demuestra la influencia de contenido de finos en la vulnerabilidad del suelo a la licuación. una serie de las evaluaciones se realizó en mezclas rehidratadas y saturadas de arena y cieno. más allá del 30 % de los contenidos finos, una fracción del cieno hace parte principal de la cadena de fuerza del suelo. en este contexto, el concepto de porosidad equivalente intergranular puede ser un parámetro apropiado para expresar la tensión crítica de corte del suelo estudiado. este parámetro nos permite controlar la tensión de corte sin drenar de cieno no plástico y mezclas de arena de densidades diferentes. earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014): 149 156 abstract resumen key words: sand, silt, static liquefaction, undrained, strength, density. palabras clave: arena, cieno, licuación estática, no drenaje, tensión, densidad. record manuscript received: 10/03/2014 accepted for publication: 03/09/2014 critical undrained shear strength of sand-silt mixtures under monotonic loading mohamed bensoula ¹˴², hanifi missoum ², karim bendani ² ¹ department of civil and architecture, university of mostaganem, bat 72 n°03 cité 05 juillet, mostaganem 27000 algeria, bensoulamoh@hotmail.com ² laboratory of construction, transports and environment protection (lctpe), bp 227, route de belhacel, university of mostaganem, cp 27000, mostaganem, algeria geotechnics issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.42492 http://dx.doi.org/10.15446/esrj.v18n2.42492 150 khadija baba, lahcen bahi, latifa ouadif list of symbols cu: coefficient of uniformity cc: coefficient of curvature dr: initial relative density : equivalent relative density e: global void ratio emax: maximum void ratio emin: minimum void ratio es: intergranular void ratio e*: equivalent void ratio fc: fines content gs: specific gravity of soil grain gd: solid dry density p’: effective mean stress p’cr: effective mean stress at critical state : effective mean stress at steady state q: deviatoric stress qcr: deviatoric stress at critical state qs: deviatoric stress at steady state sucr: undrained critical shear strength ea: axial strain di: grain diameter corresponding to i% finer sc: confining pressure ip: plasticity index m: slope of critical state line fs: mobilized angle of inter-particle friction at the steady state introduction northern algeria is located on a tectonic plate boundary (ruptured fault), and consequently, many earthquakes occur in this region. recent seismic activity over the last twenty years in algeria includes the mascara earthquake in 1994 (m: 5.6), the algiers earthquake in1996 (m: 5.7), the ain-temouchent earthquake in 1999 (m: 5.8), the beni-ouartilane earthquake in 2000 (m: 5.4) and the boumerdès earthquake in 2003 (m: 6.9).these earthquakes caused major structural damage to constructions and dramatic losses of human life. these events also included numerous cases of ejection of great masses of sandy soils onto the ground surface level and large displacements of equipment foundations. the region of mostaganem is in the northern part of the country, is situated on thick geological deposits of silty sand, and has a relatively shallow phreatic water table. these conditions enhance the liquefaction phenomena. due to the soil failure phenomenon, there is a need to characterize this granular media as an engineering material. under undrained conditions, the shear resistance of sandy soils may decrease by either monotonic (e.g., erosion at toe of a slope, reservoir filling, or rapid sediment accumulation) or dynamic (e.g., earthquake, vibration, or blast) loading, leading to liquefaction. the pore water in a saturated sandy soil will increase due to volume contraction if the drainage of the water is impeded. this results in a lower effective stress and, consequently, a reduction in soil strength. several liquefaction flow failures can be triggered by static and cyclic mechanisms. liquefaction may lead to catastrophic failure when the shear resistance of a soil vulnerable to liquefaction drops below the existing initial static load (e.g., driving shear stress on a slope). many cases of static liquefaction flow failures have been reported by olson and stark (2003). to evaluate the occurrence of flow deformation, the undrained shear strength of sandy soils during flow failures and liquefaction is an essential parameter in undrained stability analysis. the correct estimation of the undrained liquefied strength is of paramount importance for the design of soil structures, such as earthen dams, bridge supports and building foundations, with the aim of protecting them from liquefaction failure. the understanding of soil liquefaction is continuously being revised due to new field and laboratory observations. obvious evidence has significantly influenced the research trend in investigations of liquefaction in natural sandy soil deposits. the effects of fines on the mechanical behavior of sandy matrix soil are not yet fully understood. therefore, the present challenge is to improve fundamental research on the mechanical behavior of soils susceptible to liquefaction to unify the treatment of such soils. the presence of fines may either increase or decrease liquefaction vulnerability. many contributions have tried to characterize the influence of fines on liquefaction behavior in sandy matrix soils. the chinese criteria (wang, 1979) are the well-known liquefaction susceptibility identification. these criteria are essentially based on field observations. these criteria have undergone many revisions, but they still constitute the basic principles and guidelines for many methodologies. nevertheless, recent findings suggest some inadequacy and deviation associated with the presence of fines, especially plastic fines. due to discrepancies in the research results in the laboratory and post-liquefaction events, many researchers agreed that such procedures should be discontinued and should be reviewed in a general context (prakash and puri, 2010). silty sands are the most common type of soil involved in liquefaction flow events. static and earthquake-induced liquefaction failure events have been described by yamamuro and lade (1998). several studies have been conducted on the influence of fine particles on the liquefaction behavior of sandy soils. seed et al. (1983) observed that certain types of soil with high contents of fines may be highly susceptible to liquefaction vulnerability. after performing several tests, yamamuro and lade (1997) demonstrated that an increase in silt in a sand-silt mixture leads to a decrease in undrained shear resistance of the silty sand at a constant total void ratio. the behavior of clean sand differs from that of sand-silt mixtures. there are contradictory conclusions regarding this difference: some researchers have concluded that silt in the sand matrix reduces the undrained shear resistance of sandysoil mixtures, e.g., chang et al. (1982), whereas other researchers have reached the opposite conclusion, e.g.,troncoso and verdugo (1985), and vaid (1994). therefore, based upon the contradictory results presented in the literature, the fines content of a sandy soil does not lead to uniform behavior, and although recent advances have been achieved in this area of research, our understanding of this behavior is still non-unified and controversial. the stability of structures built on liquefied soil depends on the postliquefaction effects of the soil shear strength. the strength of soils at the phase transition is critical for engineering design (ishihara, 1993). consequently, it is important to clarify the main parameters that exert substantial influence on the shear resistance and to delineate the necessary principles for the design process. yamamuro and lade (1997) showed that the intergranular void ratio (es) likely 151critical undrained shear strength of sand-silt mixtures under monotonic loading controls the undrained resistance of sand-silt mixtures. this parameter, (es), represents the space within sand grains and can be computed as follows: es = (e + fc) / (1 fc) (1) where e is the global void ratio and fc is the fines content (the ratio of the silt weight to the total sample weight). thevanayagam et al. (2002) stated that, beyond a certain value of fc (the threshold value), a fraction of the silt participates in the soil skeleton chain force, and these authors introduced a new parameter defined as the equivalent intergranular void ratio. this parameter is able to influence the undrained shear resistance of non-plastic silt and sand mixtures and is described by the following equation: e* = (e + α.fc)/(1 α.fc) (2) where e is the equivalent void ratio and a represents the silt fraction that participates in the soil skeleton strength. when a = 1, the equivalent granular void ratio is reduced to the intergranular void ratio. this paper presents a parametric study focusing on estimating the undrained liquefied shear strength of silty sand using state parameters and the geotechnical properties of the mixtures. this work investigates the effects of non-plastic silty fines content on the undrained liquefied shear strength of silt-sand mixtures. our main aim in this study is to introduce relationships that express the critical undrained shear resistance in terms of the various soil parameters of mostaganem sandy soil at different relative densities and fines contents (up to 40%). to achieve this aim, a series of monotonic undrained triaxial tests were performed on reconstituted saturated sand-silt mixtures. materials tested soil samples from the coastal region of mostaganem were collected at various depths. in these samples, the silty fines content did not exceed 30%. in our experiments, the sand and silt were separated and used to generate the test specimens. the specific gravities of the sand and silt were 2.67 and 2.69, respectively. the tests were conducted on mixtures of the collected sand and silt. the plastic limit and liquid limit of the silt were 20% and 25%, respectively. because most static liquefaction and earthquake-induced liquefaction occur in silty sand and sandy silt, silt contents ranging from 0% to 40% were tested. additionally, liquefaction vulnerability is highly affected by the relative density of the soil (yamamuro and kelly, 2001; maheshwari and patel, 2010). in this work, a variety of soil density states were also studied. the geotechnical properties of the soil used in this work are presented in table 1. 0 10 20 30 40 50 60 70 80 90 100 0.0010.010.11 sand silt grain size (mm) pe rc en t p as si ng by w ei gh t figure 1. grain size distribution curves of the tested materials. the grain size distribution curves for all sand silt mixtures under investigation are illustrated in figure 1. di corresponds to the soil diameter at which i% of the soil weight is finer. the plasticity index of the silt is ip = 5%, and cu and cc are the coefficients of uniformity and curvature, respectively. according to the astm d2487-11(2011) classification, the sand under study is poorly sorted (sp), and the silt is inorganic (ml). table 1. geotechnical properties of sand-silt mixtures. index properties clean sand (0%fc) silty sand (10%fc) silty sand (20%fc) silty sand (30%fc) silty sand (40%fc) silt (100%fc) fc (%) 0 10 20 30 40 100 gs g/cm 3) 2.670 2.672 2.674 2.676 2.679 2.690 emin 0.519 0.461 0.418 0.401 0.457 0.707 emax 0.844 0.791 0.753 0.727 0.744 1.429 d10 0.350 0.130 0.050 0.020 0.010 0.0012 d30 0.550 0.290 0.210 0.140 0.090 0.017 d50 0.770 0.480 0.400 0.300 0.230 0.035 d60 0.820 0.600 0.520 0.450 0.330 0.044 cu 2.34 4.62 10.40 22.50 33.00 36.67 cc 1.05 1.08 1.70 2.18 2.45 5.47 the parameter emax is the maximum void ratio (lowest density state), and emin is the minimum void ratio (densest density state). these parameters, emax and emin, are obtained by fixing the height drop of the funnel (figure 2). the variation of emax and emin in term of the fines content fc is shown in figure 2. the variation in both indices clearly follows the same trend. the two indices decrease with as the fines content increases to 30% and then increase beyond this value. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 0 20 40 60 80 100 dr = 15% dr = 55% emax emin fines content (%) v oi d r at io s (e m ax , e m in ,e ) emin emax figure 2. extreme and global void ratios of samples for different relative densities versus fines content. laboratory experiment procedures preparation method: sample preparation methods considerably affect the liquefaction behavior of soils (ladd, 1974; mulilis et al., 1977). the sample must be prepared in such a way as to replicate the soil conditions in the field. therefore, the choice of proper sample preparation techniques is important in determining the liquefaction potential of sandy soils. numerous sample reconstitution 152 khadija baba, lahcen bahi, latifa ouadif methods have been discussed for use in the laboratory, including moist tamping, dry funnel pluviation and water sedimentation. the dry pluviation deposition method tends to reproduce field performance well, as shown by vaid et al. (1999). in this study, dry funnel pluviation was used as the sample preparation method to reproduce the soil conditions in the field. the experimental device fabricated for the dry funnel depositional method is shown in figure 3. raising system sand-silt mixture funnel mold deposited sand-silt mixture figure 3. dry funnel deposition system. soil sample preparation: dry sand and silt were mixed according to the different specified weight ratios. the samples were prepared by means of a mold consisting of two semicylindrical shells (figure 4). the two parts of the mold can be easily joined by a hose clamp. figure 4. experimental device. the samples are cylindrical in shape and are 70 mm in diameter (d) and 140 mm in height (h) (h/d=2.0). the mass of sand-silt mixture to be put inside the mold is calculated according to the required relative density (the initial volume is known). the relative density of sample is defined as follows: dᵣ = (emax emin) / (emaxemin) (3) where e is the global void ratio. the tests were carried out on reconstituted saturated samples. degree of saturation during the test: first, the samples were purged with carbon dioxide gas for more than 30mn and subsequently saturated with de-aired water. skempton’s pore water parameter b controls the saturation state. according to astm d 4767-02 (2004), samples can be considered fully saturated if b is equal to or greater than 0.95. in all the undrained tests, this condition was fulfilled. in this investigation, a backpressure of 100 kpa was applied during the experiments to achieve a completely saturation state. consolidation and loading: samples were isotropically consolidated at a mean effective stress of 100 kpa and then subjected to undrained monotonic triaxial loading. a constant strain rate of 5% per hour was applied during all the tests to stabilize the pore water pressure build up throughout all the samples. all the tests reached 25% axial strain. estimation of undrained critical shear strength: the behavior of saturated sand specimens in undrained triaxial tests is presented in figure 5 (vaid and chern, 1983). points a, b, and c represent isotropically compressed specimens. in undrained (constant volume) conditions, the effective mean stress decreases as excess pore water pressure develops. during this process, the soil sample reaches peak undrained shear strength su (the yield strength) at positions a1, b1 and c1. liquefaction is triggered when shear stress applied to the soil specimen exceeds the peak shear strength as a result of static or dynamic loading. beyond that point, an unstable regime (prunier et al., 2009) develops in the specimen until the critical state is reached at point d on the critical state line (csl). this position is called steady state. if the effective stress path reaches the flow liquefaction surface (fls), the soil element becomes unstable and its effective stress state rapidly moves to the steady state, where the soil element undergoes sudden failure with large deformation. (a) figure 5. graphical explanation of the flow liquefaction surface (vaid and chern, 1983) a) stress paths (q-p’) in an undrained test. b) flow liquefaction susceptibility. in undrained situations, the shear strength at steady state is conventionally the critical shear strength. in the framework of critical state soil mechanics theory, the following relationship can be written: qs= m.p's (4) where m is the slope of the critical state line. according to schofield and wroth (1968) and triaxial tests, sin øs = (3.m) / (6 + m) (5) where qs, and fs indicate the deviatoric stress (), the effective mean principal stress , and the mobilized angle of inter-particle friction at steady state, respectively. the critical shear strength sucr can be written as follows: sucr = (qs / 2). cosøs (6) (b) 153critical undrained shear strength of sand-silt mixtures under monotonic loading results and discussion the results of the undrained monotonic compression triaxial tests performed for different fines contents (ranging from 0 to 40%) at an average confining pressure of 100 kpa within two separate density ranges (dr = 15 and 55%) are shown in figures 6 to 10. the stress paths (p’, q) recorded during the tests, as well as the axial strain versus the deviatoric stress q, were recorded and represented graphically. for fc values ranging from 0% to 30%, we observe that the critical deviatoric stress (qcr) is reached for axial strain values ranging from 0.8% to 1.6% (figure 6). when the fine fraction is fc = 0%, the critical deviatoric stress (qcr) values are 34.12 kpa and 38.69 kpa corresponding to effective mean stress (p’cr) values of 57.86 kpa and 63.11 kpa for the initial relative densities dr = 15% and 55%, respectively (m = 0.84, fs = 21.62°). when the fine fraction is 10% (figure 7), the critical deviatoric stress (qcr) values are 32.08 kpa and 35.84 kpa (m = 0.83, fs = 21.38°). for fc = 20% (figure 8), the critical deviatoric stress (qcr) values are 28.54kpa and 31.08 kpa (m = 0.82, fs = 21.14°), and for fc = 30% (figure 9), the critical deviatoric stress (qcr) values are 24.09 kpa and 27.32 kpa (m = 0.80, fs = 20.67°). when fc = 40% (figure 10), the mixture exhibits a dilatant behavior and does not develop a contractile phase. the deviatoric stress (qcr) increases again and has values of 27.52 kpa and 30.85 kpa for dr = 15% and 55%, respectively (m = 0.78, fs = 20.19°). we observed that an increase in the amount of fines of 0% to 30% leads to a decrease in the deviatoric stress. this decrease occurs because of the role of the fines in reducing the soil dilatancy and amplifying the phase of contraction in the sand-silt mixtures, which leads to a reduction in the confining effective pressure and, consequently, a decrease in the peak strength of the mixtures. the stress path in the (p’, q) plane clearly shows the role of the fines in the decrease in the average effective pressure and the maximum deviatoric stress (figures 6.b-10.b). in this case, the effect of fines on the undrained behavior of the mixtures is observed for the fines contents lower than 30%. these results are in good agreement with the observations of shen et al. (1977), troncoso and verdugo (1985), koester (1994) and kokusho et al. (2014). figure 6. a) test results of deviatoric stress versus axial strain (q, ea), fc = 0%. b) stress paths (q, p’) for undrained loading tests, fc= 0%. figure 7. a) test results of deviatoric stress versus axial strain (q, ea), fc= 10%. b) stress paths (q, p’) for undrained loading tests, fc= 10%. figure 8. a) test results of deviatoric stress versus axial strain (q, ea), fc= 20%. b) stress paths (q, p’) for undrained loading tests, fc= 20%. figure 9. a) test results of deviatoric stress versus axial strain (q, ea), fc= 30%. b) stress paths (q, p’) for undrained loading tests, fc= 30%. figure 10. a) test results of deviatoric stress versus axial strain (q, ea), fc= 40%. b) stress paths (q, p’) for undrained loading tests, fc= 40%. when the proportion of fines is 40%, there is an increase in the deviatoric stress, which is due to the role of the fines in increasing the dilatancy in the soil and the absence of a contractile phase. this result shows that % in mixtures with greater than 30% fines, the fines participate in the strength of the mixture and reverse the behavior trend. table 2 presents the summary of the undrained monotonic compression triaxial tests. table 2. undrained monotonic test results for different initial relative densities of sand-silt mixtures. fc (%) e e* e*min e * max dr(%) d*r (%) m sucr/σc clean sand (0%fc ) 0% 0.795 0.795 0.519 0.844 15% 15% 0.84 0.1615 0.665 0.665 55% 55% 0.1811 silty sand (10%fc ) 10% 0.742 0.897 0.591 0.951 15% -16.26% 0.83 0.1506 0.610 0.753 55% 27.98% 0.1682 silty sand (20%fc ) 20% 0.703 0.978 0.648 1.037 15% -41.34% 0.82 0.1347 0.569 0.823 55% 6.56% 0.1466 silty sand (30%fc ) 30% 0.678 1.051 0.712 1.111 15% -63.66% 0.80 0.1149 0.548 0.892 55% -14.62% 0.1303 silty sand (40%fc ) 40% 0.701 1.167 0.856 1.222 15% -99.38% 0.78 0.1288 0.586 1.021 55% -54.37% 0.1444 these results suggest that the soil under investigation exceeds the threshold set by the chinese criteria (wang, 1979) and clearly show that the soil under study is susceptible to liquefaction at a proportion of 30% fines. these results coincide with the actual observations in three recent earthquakes, the northridge (1994), kocaeli (1999), and chi-chi(1999) earthquakes, which all showed signs of liquefaction in soils with greater than 15% fines content. this means the soils 0 20 40 60 80 100 120 140 0 50 100 150 200 250 dr = 15% dr = 55% effective mean pressure, p' (kpa) d ev iat or s tre ss , q (k pa ) sand-silt mixture (0%fc) (b) 0 20 40 60 80 100 120 140 160 0 50 100 150 200 dr = 15% dr = 55% effective mean pressure, p' (kpa) d ev ia to r s tre ss , q (k pa ) sand-silt mixture (10 %fc) (b) 0 20 40 60 80 100 120 140 160 0 5 10 15 20 25 30 dr = 15% dr = 55% axial strain (%) d ev ia to r s tre ss , q (k pa ) sand-silt mixture (10 %fc) (a) 0 20 40 60 80 100 120 140 0 10 20 30 dr = 15% dr = 55% axial strain (%) d ev ia to r s tr es s, q (k pa ) sand-silt mixture (0%fc) (a) 0 20 40 60 80 100 120 140 160 0 5 10 15 20 25 30 dr = 15% dr = 55% d ev ia to r s tre ss , q (k pa ) axial strain (%) sand-silt mixture (20 %fc) (a) 0 20 40 60 80 100 120 140 160 0 50 100 150 200 dr = 15% dr = 55% d ev ia to r s tr es s, q (k pa ) effective mean pressure, p' (kpa) sand-silt mixture (20 %fc) (b) 0 20 40 60 80 100 0 5 10 15 20 25 30 dr = 15% dr = 55% d ev ia to r st re ss , q (k p a) axial strain (%) sand-silt mixture (30 %fc) (a) 0 20 40 60 80 100 120 0 50 100 150 dr = 15% dr = 55% effective mean pressure p '(kpa) d ev ia to r s tr es s, q (k pa ) sand-silt mixture (30 %fc) (b) 0 20 40 60 0 10 20 30 dr =15% dr = 55% sand-silt mixture (40 %fc) axial strain (%) d e vi a to r st re ss , q (k p a ) (a) 0 20 40 60 80 100 120 0 50 100 150 dr =15% dr = 55% sand-silt mixture (40 %fc) d ev ia to r s tr es s, q (k pa ) effective mean pressure, p '(kpa) (b) 154 khadija baba, lahcen bahi, latifa ouadif did not meet the chinese criteria and yet still liquefied. the same observations are evident in the catastrophic static liquefaction failure events reported by kramer and seed (1988) and fourie and tshabalala (2005). in these events, the post-liquefied soil was found to have more than 20% silt content. in this context, holzer et al. (1999) noted that soils that would have been declared safe from liquefaction by the chinese criteria did in fact liquefy and lead to permanent ground deformation, and many recent studies, such as bray and sancio (2006) and boulanger and idriss (2006), have recommended that workers should not rely on the chinese criteria. correlation of critical undrained shear resistance (sucr) and equivalent void ratio (e*): in equation (1), the fine particles are simply assumed to occupy voids in the sand skeleton. the concept of the intergranular void ratio suggests that the fines fill the voids formed between the sand grains; thus, the behavior of sand with a moderate amount of fines should be governed by the intergranular void ratio instead of the global void ratio. however, when the intergranular void ratio exceeds the maximum void ratio of the clean sand, there are sufficient fines to prevent grain-to-grain contact of the sand particles. in this case, the fines constitute the dominant structure and carry the shear forces, while the coarse grains may act as reinforcing elements (thevanayagam and mohan 2000). zlatovic and ishihara (1995) and pitman et al. (1994) reported that fines particles started to interfere with sand particle contacts at 5% fines content and that sand particle contacts vanished completely at approximately 25% silt content. to satisfy these experimental observations (thevanayagam et al. 2002), expression (1) was modified to equation (2). generally, e* is obtained from the correlation between soil sorting properties and back-analyzed values (ni et al., 2004; yang et al., 2006). rahman et al. (2008) analyzed the experimental data of mcgeary (1961) on binary packing studies and concluded that this parameter is dependent on both the particle size ratio and fines content, fc. the size ratio r is defined by where d50(fine) is the median size of the fines and d10(sand) is the particle size at 10% fraction of the host sand. the correlation was suggested in the following form: α = 1 [{1exp([-0.3(fc / fthre ) / k) ]}] (rfthre / fc) r (7) where k = 1 –r0.25 and fthre is threshold fines content or transitional fines content, which is characterized by the predominance of finescontrolled behavior. thus, the value of fthre is defined by the point at which the behavior trend reverses with further increases in fines content. based on our tests, this transition point is located at 30% fines content for all two initial relative densities as shown in figures 6.b to 10.b. for the studied sand-silt mixtures, r and k were calculated to be 0.100 and 0.438, respectively. the variations in the equivalent void ratios versus fines content for the initial relative densities (dr = 15% and 55%) are shown in figure 11. for all studied initial relative densities, a decrease in the global void ratio generates an increase in the equivalent or intergranular void ratio. the absolute rate of variation in both void ratios tends to decrease as long as the fines content increases. 0.0 0.2 0.4 0.6 0.8 1.0 1.2 1.4 0 10 20 30 40 50 dr = 15% dr = 55% fines content (%) eq ui va le nt vo id r at io s (e *) figure 11. variation in the equivalent void ratio (e*) versus fine content fc. figure 12. variation in the undrained critical shear strength versus equivalent void ratio (e*) (σc =100 kpa). figure 12 shows the variation in the undrained critical shear strength versus the equivalent void ratio for the initial relative densities (dr = 15% and 55%). it is obvious from this figure that the undrained critical shear strength decreases linearly as the equivalent void ratio decreases and fines content increases for all densities (dr = 15% and 55% during reconstitution) up to 30% fines content. thus, when the global void ratio decreases and the fines content increases, the undrained critical shear strength also decreases. we conclude that the global void ratio does not represent the real behavior of silty sandy soil with 0-30% fines content. moreover, the equivalent void ratio is a key parameter for characterizing the mechanical state of clean sand and sand-silt mixtures. the following correlation of normalized undrained shear strengths and equivalent intergranular void ratio is obtained: sucr / σc = -0.165 e* + 0.290 (8) correlation of critical undrained shear resistance (sucr) and equivalent relative density (): because the equivalent intergranular void ratio (e*) is presented as an essential representative parameter, the equivalent relative density should consequently be defined as follows (thevanayagam et al., 2002, 2003; shenthan, 2005): d*r = [(emax,cs e* ) / (emax,cs emin,cs )]x100 (9) where emax,cs is the maximum void ratio of clean sand and emin,cs is the minimum void ratio of clean sand. this relative index compares an equivalent granular state of sandy soils with two extreme density states of clean sand. equivalent relative density. dr* (%) figure 13. undrained critical shear strength versus equivalent relative density (sc =100 kpa). 155critical undrained shear strength of sand-silt mixtures under monotonic loading figure 13 shows the variation of the normalized undrained critical shear strength in term of the equivalent relative density. a linear relationship is obtained as follows: sucr / σc = 0.0537dr* + 0.1506 (10) where d*r is the relative density expressed as a decimal ( d*r =d*r /100) the critical undrained shear strength increases with increasing equivalent relative density for fc ≤ 30%. in the range 0-30% fine content, we may deduce that, for both obtained relationships, the equivalent intergranular void ratio and the equivalent relative density are key parameters in characterizing such soil’s susceptibility to instability. conclusions a series of monotonic undrained triaxial tests were performed on reconstituted saturated sand-silt mixtures. two levels of initial relative density (15% and 55%) and an initial effective confining stress of 100 kpa were used in the experimental program. the soil sample’s proportions of fines ranged from 0% to 40%. in the framework of the steady state concept and liquefaction susceptibility of sands, the tested sand-silt mixtures experienced limited liquefaction behavior and dilation in the present undrained triaxial tests under monotonic loading. the obtained results show that the critical undrained shear strength is considerably affected by the fines content. the participation of silty fines in the skeleton forces in the soil depends on their content. consequently, the equivalent intergranular void ratio was introduced to describe the fine participation. in our tests, this parameter accurately represents the soils until a fines content of 30%. the advantage of this parameter is that it may be able to define the inter-grain friction properties of sand-silt mixtures. the critical undrained shear strength decreases in a linear manner with increasing equivalent intergranular void ratios, whereas it increases with increasing equivalent relative densities. however, this behavior is only valid for fc ≤ 30%. the factor α should be related to the mechanical parameters of sand silt mixtures via the critical state soil mechanics framework. in the range 0-30% fine content, we may deduce that, for both obtained relationships, the equivalent intergranular void ratio and the equivalent relative density are key parameters in characterizing such soil’s susceptibility to instability. furthermore, these results mean that the studied soil exceeds the threshold set by the chinese criteria (wang, 1979) and clearly show that this soil is susceptible to liquefaction at fines content of 30%. these results agree with recent observations reported by researchers studying static and cyclic liquefaction failures that have caused catastrophic events. references astm d2487-11(2011) “standard practice for classification of soils for engineering purposes (unified soil classification system)” astm international, west conshohocken, pa. doi: 10.1520/d2487-11. astm d4767-02 (2004). “standard test method for consolidated undrained triaxial compression test for cohesive soils.”american society for testing and materials, vol. 04.08, pp. 1-13. boulanger, r. w., & idriss, i. m. (2006). “liquefaction susceptibility criteria for silts and clays.” journal of geotechnical and geoenvironmental engineering, 132(11), 1413-1426.doi.org/10.1061/(asce)10900241(2006)132:11(1413). bray, j. d., & sancio, r. b. (2006). “assessment of the liquefaction susceptibility of fine-grained soils.” journal of geotechnical and geoenvironmental engineering,132(9), 1165-1177. doi. org/10.1061/(asce)1090-0241(2006)132:9(1165). chang, n. y., yeh, s. t., & kaufman, l. p. (1982). “liquefaction potential of clean and silty sands.” in proceedings of the third international earthquake microzonation conference, vol. 2, pp. 1017-1032. fourie, a. b., & tshabalala, l. (2005). “initiation of static liquefaction and the role of k0consolidation.”canadian geotechnical journal, 42(3), 892-906.dx.doi.org/10.1139/t05-026. holzer, t. l., bennett, m. j., ponti, d. j., & tinsley iii, j. c. (1999). “liquefaction and soil failure during 1994 northridge earthquake.” journal of geotechnical and geoenvironmental engineering, 125(6), 438-452. doi.org/10.1061/(asce)10900241(1999)125:6(438). ishihara,k. (1993). “liquefaction and flow failure during earthquakes.” géotechnique, 43(3),351-415. icevirtuallibrary.com/content/ article/10.1680/geot.1993.43.3.351. koester, j. p. (1994). “the influence of fines type and content on cyclic strength.” in ground failures under seismic conditions (pp. 17-33). asce. kokusho, t., nagao, y., ito, f., & fukuyama, t. (2014). “sand liquefaction observed during recent earthquake and basic laboratory studies on aging effect.” in earthquake geotechnical engineering design (pp. 75-92). springer international publishing.doi :10.1007/978-3-319-03182-8_3. kramer, s. l., & seed, h. b. (1988). “initiation of soil liquefaction under static loading conditions.” journal of geotechnical engineering, 114(4), 412-430.dx.doi.org/10.1061/(asce)07339410(1988)114:4(412) ladd, r. s. (1974). “specimen preparation and liquefaction of sands.” journal of the geotechnical engineering division, 100(10), 1180-1184. maheshwari, b. k., & patel, a. k. (2010). “effects of non-plastic silts on liquefaction potential of solani sand.” geotechnical and geological engineering, 28(5), 559-566.doi.org/10.1007/s10706-010-9310-z. mcgeary, r. k. (1961). “mechanical packing of spherical particles.” journal of the american ceramic society, 44(10), 513-522. dx.doi. org/10.1111/j.1151-2916.1961.tb13716.x. mulilis, j. p., arulanandan, k., mitchell, j. k., chan, c. k., & seed, h. b. (1977). “effects of sample preparation on sand liquefaction.” journal of the geotechnical engineering division, 103(2), 91-108. ni, q., tan, t.s., dasari, g.r., & hight, d.w. (2004). “contribution of fines to the compressive strength of mixed soils.” geotechnique, 54(9): 561-569.refdoc.fr/detailnotice?idarticle=8461334. olson, s., & stark, t. (2003). “ yield strength ratio and liquefaction analysis of slopes and embankments.” j. geotech. geoenviron. eng., 129(8), 727-737. doi.org/10.1061/(asce)1090-0241(2003)129:8(727). pitman, t. d., robertson, p. k., & sego, d. c. (1994). “influence of fines on the collapse of loose sands.” canadian geotechnical journal, 31(5), 728-739.doi/abs/10.1139/t94-084. prakash, s., & puri, v. k. (2010). “recent advances in liquefaction of fine grained soils.” in 5th international conference on recent advances in geotechnical earthquake engineering and soil dynamics, san diego, california, pp. 1-6. prunier, f., laouafa, f. & darve, f. (2009). “material stability analysis based on the local and global elasto-plastic tangent operators.” proceedings of the 1st international symposium on computational, geomechanics (comgeo i), pp. 215-225. rahman, m.m., le, s.r., & gnanendran, c.t. (2008) “on equivalent granular void ratio and steady state behaviour of loose sand with fines” can geotech j 45(10):1439-1455. dx.doi.org/10.1139/t08-064. seed, h. b., idriss, i. m., & arango, i. (1983). “evaluation of liquefaction potential using field performance data”. journal of geotechnical engineering, 109(3), 458-482.doi.org/10.1061/(asce)07339410(1983)109:3(458). shen, c.k. ,vrymoed, j.l. , & uyeno, c.k. (1977). “the effects of fines on liquefaction of sands.” proc., 9th int. conf. on soil mechanics and foundation engineering, tokyo, 381–385. shenthan, t. (2005). “liquefaction mitigation in silty soils using composite stone column.” ph. d. dissertation, university at buffalo,buffalo, ny. schofield, a., & wroth, p. (1968). “critical state soil mechanics.” london, mcgraw-hill. 156 khadija baba, lahcen bahi, latifa ouadif thevanayagam, s., & mohan, s. (2000). “intergranular state variables and stress–strain behaviour of silty sands.” geotechnique, 50(1), 1-23. doi:10.1680/geot.2000.50.1.1. thevanayagam, s., shenthan, t., & kanagalingam, t. (2003). “role of intergranular contacts on mechanisms causing liquefaction & slope failures in silty sands.” in final report, usgs award no. 01hqgr0032 and 99hqgr0021. u.s. geological survey, department of the interior, reston, va. thevanayagam, s., shenthan, t., mohan, s., & liang, j. (2002). “undrained fragility of clean sands, silty sands, and sandy silts.” journal of geotechnical and geoenvironmental engineering, 128(10), 849-859. troncoso, j. h., & verdugo, r. (1985). “silt content and dynamic behavior of tailing sands.” in proc., xi int. conf. on soil mechanics and foundation engineering, pp. 1311-1314. vaid, y.p. (1994). “liquefaction of silty soils.” in ground failures under seismic conditions. geotechnical special publication no. 44. american society of civil engineers. pp. 1–16. vaid, y. p., & chern, j. c. (1983). “mechanism of deformation during cyclic undrained loading of saturated sands.” international journal of soil dynamics and earthquake engineering, 2(3), 171-177.doi. org/10.1016/0261-7277 (83) 90014-1. vaid, y. p., sivathayalan, s., & stedman, d. (1999). “influence of specimen-reconstituting method on the undrained response of sand.” astm geotechnical testing journal, 22(3), 187-195. refdoc. fr/detailnotice?idarticle=11381016. wang, w. s. (1979). “some findings in soil liquefaction.” water conservancy and hydroelectric power scientific research institute, beijing, china. yamamuro, j. a., & kelly, m. c. (2001). “monotonic and cyclic liquefaction of very loose sands with high silt content.” journal of geotechnical and geoenvironmental engineering, 127(4), 314-324. doi.org/10.1061/(asce)1090-0241(2001)127:4(314). yamamuro, j. a., & lade, p. v. (1997). “static liquefaction of very loose sands.” canadian geotechnical journal, 34(6), 905-917. doi. org/10.1139/t97-057. yamamuro, j. a., & lade, p. v. (1998). “steady-state concepts and static liquefaction of silty sands.” journal of geotechnical and geoenvironmental engineering, 124(9), 868-877.doi.org/10.1061/ (asce)1090-0241(1998)124:9(868). yang, s.l., lacasse, s., & sandven, r.f. (2006). “determination of the transitional fines content of mixtures of sand and non plastic fines.” geotech test j 29(2):102-107. hrefdoc.fr/ detailnotice?idarticle=6662871. zlatovic, s., & ishihara, k. (1995). “on the influence of non-plastic fines on residual strength.” proceedings of the first international conference on earthquake geotechnical engineering, tokyo, pp. 14-16. issn 1794-6190 e-issn 2339-3459 earth sciences research journal earth sci. res. j. vol. 26, no. 1 (march, 2022): 39 46 g e o p h y si c s effects of vegetation height and relative submergence for rigid submerged vegetation on flow structure in open channel zhongyu wang1, huayong zhang1*, xing he1, qi jiang2, weigang xu3, wang tian1 1. research center for engineering ecology and nonlinear science, north china electric power university, beijing, china. 2. national major science and technology program management office for water pollution control and treatment, beijing, china. 3. institute of wetland research, chinese academy of forestry, beijing, china. *corresponding author: rceens@ncepu.edu.cn abstract: resumen vegetation commonly existing in natural rivers interact with water flow in a strongly nonlinear manner. in this research, vegetation height and a concept of relative submergence are adopted to experimentally investigate the effects of rigid submerged vegetation on flow structure. the results reveal that the relative submergences decrease along the submerged vegetation section. from the perspective of case-average, the linear relationship between water depth and vegetation height demonstrates a decrease trend of average relative submergence with the increase of vegetation height. the differences between the maximum and minimum water depths of submerged cases can be described by a positive linear function with vegetation height and a negative power function with relative submergence, by which, the raised water depths induced by vegetation block can be predicted under similar hydraulic conditions. the findings of the research, especially the introduction of relative submergence and its functional relationship with the water depth alternation induced by submerged vegetation, may improve the comprehensive understandings of the hydraulic behaviors in vegetated open channel and give some guidance to the practices of river ecological restoration. keywords: rigid submerged vegetation; vegetation height; relative submergence; flow structure; water depth; flow velocity distribution. https://doi.org/10.15446/esrj.v26n1.76187 efectos de la altura de la vegetación y de la inmersión relativa en vegetación rígida en la corriente de un canal abierto palabras clave: vegetación sumergida rígida; altura de la vegetación; inmersión relativa; flujo estructura; profundidad del agua; distribución de la velocidad. la vegetación en las corrientes de un río natural interactúa con el flujo de agua de una manera no lineal. en esta investigación se estudia la altura de la vegetación y un concepto de inmersión relativa para investigar experimentalmente los efectos de la vegetación rígida sumergida en una estructura de flujo. los resultados revelan que las submersiones relativas decrecen a lo largo de la sección de vegetación sumergida. desde la perspectiva del caso promedio, la relación lineal entre la profundidad del agua y la altura de la vegetación demuestra una tendencia a la baja del promedio de la inmersión relativa a medida que se incrementa la altura de la vegetación. las diferencias entre la profundidad mínima y máxima del agua en los casos de inmersión puede describirse a través de una función lineal positiva con la altura de la vegetación y una función de poder negativa con inmersión relativa, por lo que el incremento del nivel del agua ocasionado por el bloqueo de la vegetación se puede predecir con condiciones hidráulicas similares. los hallazgos de la investigación, especialmente la introducción de la inmersión relativa y su relación funcional con la alternación de la profundidad del agua inducida por la vegetación sumergida, pueden mejorar los amplios conocimientos de los comportamientos hidráulicos en un canal abierto con vegetación y pueden ofrecer información para la restauración ecológica de ríos. record manuscript received: 15/11/2018 accepted for publication: 31/03/2022 how to cite item: wang, z., zhang, h., he, x., jiang, q., xu, w., & tian, w. (2022). effects of vegetation height and relative submergence for rigid submerged vegetation on flow structure in open channel. earth sciences research journal, 26(1), 3946. https://doi.org/10.15446/esrj.v26n1.76187. https://doi.org/10.15446/esrj.v25n1.74167 40 zhongyu wang, huayong zhang, xing he, qi jiang, weigang xu, wang tian introduction submerged vegetation growing in shallow water is one of the most important components providing ecological services in natural rivers (okamoto et al., 2014; najafabadi et al., 2015; o’briain et al., 2017). for the purposes of ecological restoration and pollutant reduction, aquatic vegetation planting and management with different configurations have become common approaches in recent years (zhou et al., 2006; callow, 2012; lama et al., 2021). for the rigid vegetation, due to the additional vegetal retardance, it has been noticed that vegetation planted in rivers may raise the water depth and reduce flow velocity, and therefore increase the potential risk of flooding or sediment silting (zhang et al., 2015a; friberg et al., 2017). however, the complexity of the combinations of hydraulic and vegetation conditions hindered the development the determination of the alternations of flow state due to vegetation block (stone and shen, 2002; yagci et al., 2010). a comprehensive understanding of the effects of rigid vegetation on flow structure is therefore significant for river ecological restoration. starting from this point, numerous research works can be found in literature (fathi-maghadam and kouwen, 1997; pasquino et al., 2016). to investigate the effects of rigid vegetation on flow parameters including water depth, flow velocity, flow resistance, turbulent structure and so on, rigid cylinders made of wood, iron, plastic and some other materials were commonly utilized in previous studies to simulate rigid aquatic vegetation in river ecosystems (wu, 1999; järvelä, 2005; musleh and cruise, 2006; huai et al., 2009. temmerman et al. (2005) employed a 3-d model to investigate the flow and sedimentation patterns affected by vegetation and water level fluctuations. flume studies were carried out by järvelä (2005) to investigate the flow structure above flexible vegetation, by which the flow above submerged flexible vegetation was reported to follow the log law. a definition for the shearing velocity which can avoid complex turbulence measurements was proposed in his research. the vertical flow velocity distributions in trapezoidal open channels were found following power and logarithmic functional relationships by chanson (2004) and nezu et al. (2008). with experimental and theoretical methods, a 3-layer model was proposed by huai et al. (2009) to predict the vertical flow velocity distribution with submerged vegetation in open channel. the velocity profiles affected by configurations of cylindrical bamboo stems were investigated by yokojima et al (2015). based on the experimental and numerical results, the applicability of large-eddy simulation method was examined for the canopy modelling. water depth was selected as the hydrological controlling parameter by chauhan (2005) to investigate the vegetation structure and dynamics in a subtropical regulated river, by which, the hydrological regime was reported to dominate the developing of plant community in floodplain wetland rather than the physical connectivity. the importance of vegetation effects on turbulent structure was demonstrated by mcbride et al. (2007) through a flume experiment, in which, the flow velocity, turbulent kinetic energy, reynolds shear stress and boundary shear stress for vegetated and un-vegetated cases were found obviously different. in the studies of nepf (1999) and jack et al. (2011), the vegetation existence was found to alter the turbulent structure by changing the shear and wake productions around vegetation stems. for the research on vegetal resistance, rhee et al. (2008) found that manning’s coefficients were affected by the states of plants, whether green or dormant. from the perspective of biomechanics and streamlining of vegetation, some field experiments were also conducted (sandercock and hooke, 2010). vegetation in river can reduce the force of drag and lift acting on the boundary by increasing the effective roughness height (thorne, 1990). based on that white and nepf (2003) correctly captured the dependence of wake generation on vegetation array density and reynolds number, ben meftah and mossa (2013) established a theoretical relationship between the bulk drag coefficient cd of a random, emergent array of cylinders with vegetation density. however, flow phenomena in open channel will be more complicated with vegetation introduced, especially for that the water depth is larger than the vegetation height which means submerged vegetation (stone and shen, 2002; gualtieri et al., 2018). meanwhile, many interactive factors such as river geometry, hydraulic parameters, vegetation type, stems characteristics, planting density and so on also increase the difficulty of the research in this field (green, 2005; mcbride et al., 2007; zhang et al., 2015b). therefore, it is of great significance to find a dimensionless parameter to well describe the impact of rigid submerged vegetation. in this research, vegetation height and a dimensionless concept of relative submergence are adopted to investigate the effects of rigid submerged vegetation on flow structure. via a series of experiments, the relationship between the vegetation height and relative submergence will be determined, based on which, the variation tendencies and amplitudes of water depth and flow velocity affected by rigid submerged vegetation will be analyzed and compared regarding to both vegetation height and relative submergence. the relative submergence adopted in this study and the relevant analysis results are expected to improve the comprehensive understandings of the interactions between vegetation and water flow and provide some guidance to the engineering practices of river ecological restorations. experimental and analytical methodology experimental setup to investigate the influences of vegetation height and relative submergence of rigid vegetation on flow structure in open channel, a series of experiments are conducted in a rectangular concrete channel which dimensions where length=30 m, width=1.3 m, depth=0.8 m and bed slope i=0.0055 m/m. as shown in figure 1, from the recycling tank, water is lifted to the water flatting tower utilizing a pipeline centrifugal pump (model: 200-200(i) a, maximum discharge: 99.4 l/s). then, water flows through the water flatting tower (with an overflow launder), intake gate combined with an hqc-2 water level and discharge controller, thin-walled rectangular weir, stilling basin and energy dissipation fence in sequence to fulfil the experimental requirements of water stabilization and flow discharge control. as can be seen in figure 1, there is a 3 m long stabilizing section in front of the test section to further stabilize the water flow. within the test section (y=3-11 m), a 4 m long vegetation section (y=5-9 m) is determined. to investigate the detailed flow structure affected by the rigid submerged vegetation blocks of different vegetation heights or relative submergences, the flow discharge for this research is selected as q=90 l/s which is a higher and stable value for the utilized pump. in the experimental open channel, the water flow with free surface remains gradually varied flow states, which are also the hydraulic conditions focused on in this study. figure 1. schematic view of the experimental setup. in the experiments, woody cylinders with a uniform diameter of 7 mm are employed to simulate the rigid vegetation in rivers to avoid the influences from leaves and branches. for planting the simulated vegetation stems with uniformly staggered arrays, punched plates are used to cover the planting pool which is built concave downward the channel bed of the vegetation section (y=59 m) and filled with fine sand. on these plates, small holes of 1 cm in diameter are uniformly staggered drilled, by which, the simulated vegetation stems can be firmly planted (schematized in figure 2(a)). as demonstrated by previous studies, vegetation density alters the characteristic length and time scales of energetic motion in open channel, the standard boundary-layer flows could also be shifted to canopy flows gradually with increasing vegetation density (poggi et al., 2004), in fact, higher vegetation densities are generally used in the practices of river restorations to produce better ecological and environmental benefits (nikora et al., 2013). based on the comprehensive considerations on previous study (velasco et al., 2003, zhang et al., 2015a), field investigations and the reality of punched plates, a relatively dense vegetation density of 202 stems/m2 is used in the present research. to determine the vegetation heights above the channel bed hv which can ensure that the vegetation stems are full-course submerged, a blank test without vegetation is first conducted with flow discharge of 90 l/s. within the test section, the average water depth for the blank test is 10.03 cm. the vegetation heights hv are therefore set as 0, 1.25, 2.5, 3.75, 5, 6.25, 7.5, 8.75 and 50 cm to fully investigate the effects of relative submergence on flow structure 41effects of vegetation height and relative submergence for rigid submerged vegetation on flow structure in open channel (see figure 2(b)). for these 9 vegetation heights, 0 represent the blank test without vegetation, 50 cm means an unsubmerged test for comparisons while the other 7 tests are under submerged conditions. the experimental case with vegetation height of 50 cm is served as a control group, and the data are used for the comparison when analyzing water depth and velocity variations. this will promote our understandings on vegetation effects on water flow without affecting or changing the analysis results about vegetation heights or relative submergence. figure 2. (a) vegetation planting array and (b) vegetation heights. the flow velocities and water depths within the test section (y=3-11 m) are measured by adv (sontek 16-mhz micro acoustic doppler velocimeter) and abf2-2 water level measuring system, respectively. the distance of the probes of adv to the sampling volume is 5 cm. adv is therefore a non-contact velocimeter which has high resolution (0.01 cm/s) and accuracy (±1% of measured velocity) for the velocity measuring. with the sampling frequency of 50hz, adv was reported to get a stable average flow velocity if the measuring duration exceeds 45 s (jing et al., 2010). the sampling frequency and duration are therefore set as 50 hz and 60 s in the experiments. these 3000 instantaneous velocity samples will be used to calculate the time-average flow velocity for each measurement point. during the velocity sampling, the signal-to-noise ratios of the adv measurements maintain above 15 which can guarantee the quality of the collected velocity data (pujol et al., 2010). as can be seen in figure 3(a), flow velocities are measured at the intersections of 9 transverse-sections (y=3, 4, 5, 6, 7, 8, 9, 10, 11 m) and 5 different longitudinal-sections (x=0.05, 0.35, 0.65, 0.95, 1.25 m) within the test section. to make the velocity data more accurate, at each of these intersections, the flow velocities are measured each 0.005 m from the channel bed to the water surface (see figure 3(b)). since that the arithmetic average method is typical and critical to the calculations of cross-sectional hydraulic parameters (chow, 1959; zhang et al., 2015b), the arithmetic mean of the flow velocities collected on each transverse-section is taken as the cross-sectional average velocity, correspondingly. figure 3. schematic view of measuring sections and points: (a) plane view and (b) vertical view. with abf2-2 water level measuring system, the water depths along the measuring bridge can be automatically measured point by point with equidistance (0.1 m for the present research). its resolution and accuracy are 0.01 mm, 0.04 mm, respectively. in the experiments, the longitudinalsections for water depth measurements are same to the velocity measurements. however, along the longitudinal direction, the water depths are equidistantly measured every 0.1 m from y=3 m to y=11 m (see figure 3(a)). similar to the calculating of cross-sectional average flow velocity, the arithmetic average of these 5 values on each transvers direction is the cross-sectional water depth. analytical methodology in this research, the average reynolds number re and froude number fr are used to determine the flow state for each experimental test. as shown in table 1, the average reynolds number re is in the range of 48724-52896 for all the experimental tests which suggests that the flow in the test section is fully developed turbulent and independent of re (kouwen and fathi-moghadam, 2000). meanwhile, the average froude number fr between 0.42 and 0.78 is lower than 1 indicating the slow flow condition in the test section (table 1). the friction factor for such rough channels with vegetation covered is independent of gravitational effects (kouwen and fathi-moghadam, 2000). table 1. re and fr of different experimental cases hv (cm) 0 1.25 2.5 3.75 5 6.25 7.5 8.75 50 re 52628 52896 52858 52363 51965 51848 51255 51241 48724 fr 0.72 0.78 0.77 0.71 0.66 0.65 0.58 0.59 0.42 the relative submergence sr analyzed in the research is defined as the specific value of water depth h and the submerged vegetation height hv (equation (1)). sr = h  ⁄ hv (1) the average relative submergence sr can be got by averaging the sr values at each transverse-section corresponding to each experimental test. with the water depths experimentally measured in this research, the difference between the maximum and minimum water depths corresponding to 9 different experimental tests can be calculated with: hd = hmax hmin (2) similarly, the formula for the difference between the maximum and minimum flow velocity is: vd = vmax vdmin (3) results and discussion effects of vegetation height on relative submergence with the water depths measured in the experiments, the relative submergence at each measurement points can be calculated with equation (1) corresponding to different vegetation heights. for the case without vegetation, the relative submergence is +∞ as the vegetation height hv=0. for the other 8 tests, the variations of the relative submergences along the vegetation section are plotted in figure 4(a). as can be seen in the figure, the relative submergences present decreasing trends along the flow direction for all cases. the decreases of the water depths should be responsible to such decreases of relative submergence as the vegetation heights are constant within the vegetation section. in addition, the minimum value of the relative submergences except for the unsubmerged case (hv=50 cm) is 1.133, which appears at the exit of the vegetation block (y=9 m location) for the case of hv=8.75 cm (figure 4(a)). at this time, the water depth is only 9.9 cm (as shown in figure 5), which indicates that the vegetation stems may be submerged in the fore part of vegetation section and unsubmerged near the exit of vegetation section if another larger vegetation height hv=10 cm is introduced. such partially submerged and partially unsubmerged condition is much more complex and not considered in the research. from figure 4(a), it can also be seen that the gaps between two adjacent curves of relative submergence decrease rapidly at low vegetation heights and slowly when the vegetation heights increase to higher values. corresponding to different experimental cases except the unsubmerged one, the case-averaged water depth havg is analyzed regarding to the vegetation height hv to further realize the effects of vegetation height on average relative submergence sr . as shown in figure 4(b), the relationship between them can be described by a linear function 42 zhongyu wang, huayong zhang, xing he, qi jiang, weigang xu, wang tian havg=a+bhv (4) where a=0.102, b=0.332 are constant factors while the coefficient of determination is r2=0.928. this relationship demonstrates that the average relative submergence sr decrease with the increase of vegetation height. it is obvious that the average relative submergence of the blank test is since the vegetation height can be considered as hv=0. by extending the fitting line in the figure, a special point where both vegetation height and average water depth equals 15.22 cm is found, which makes the average relative submergence equals. that means the vegetation stems are just unsubmerged in average although it may be not rigorous. the vegetation stems are presumably partly unsubmerged while the other stems submerged because of that the variation of water depth is affected by the vegetation block planted in the channel. in this case, the average relative submergence should be higher than 1 according to its definition. the exact value of the vegetation height which can make all the stems just unsubmerged couldn’t be predicted and needs to be experimentally calibrated. but this value will undoubtedly be greater than 15.22 cm. figure 4. (a) variations of relative submergence along the vegetation section, (b) the relationship between vegetation height and average water depth. effects of vegetation height on the variation trends of water depth water depth which indicates the absolute height of water in rivers is one of the important control parameters for the designs of water courses and hydraulic structures (chow, 1959). the variations of water depth affected by rigid submerged vegetation of different vegetation heights or average relative submergences are investigated in the research. figure 5 shows the variations of measured water depths along the flow direction. as can be seen in figure 5, the variation of water depth presents a slight decreasing trend with obvious fluctuations for the experimental case of no vegetation. in this case, the irregularity of the concrete channel bed should be responsible to such fluctuations of water depth. such variations of the other 8 vegetated cases are different. in detail, the curves of water depth present increasing trends before the vegetation block (y=3-5 m) and reach maximum values near the y=5 m location which is the entrance of the vegetation block. on the whole, the maximum water depth increases with the increase of vegetation height. within the vegetation section (y=5-9 m), the water depths decrease gradually and smoothly. near the exit of the vegetation block (y=9 m), the gaps between these curves shrink rapidly to a bunch and this situation maintains after that (y=9-11 m). in summary, the curves of water depth for the vegetated cases become smoother in comparison with the blank test, but the water depths before and within the vegetation block are raised due to the vegetation blockage and retardance. and the raised amplitude of water depth increases with the increase of vegetation heights. at last, these water depths after the vegetation block resume to the level and trend of the blank test. it can be easily found that the water depths drop rapidly near the exit of vegetation block, which is well known and consistent with previous studies (zhang, 2015b). therefore, flow acceleration and fluctuation, as well as high frequency exchange of flow momentum and turbulent kinetic energy are inevitable in this zone, which may cause erosion of river bed (yagci, 2010; zhang, 2015b). enhancing the strength of river bed in this zone and regulating the flow structure by configuring the vegetation patches are undoubtedly two ways to deal with this problem. obviously, some more vegetation is needed to some distance downstream of the vegetation block in the practices of river ecological regulation. in the gap regions between the vegetation patches, the flow velocity will be reduced due to the sheltering effect, which have positive effects on the stability of river bed (zhao and huai, 2016). however, how to determine an optimal configuration of vegetation patches such as the type, height and patches number of vegetation, the distance and number of gaps, etc. needs more case studies to analyze and summarize. figure 5. variations of water depth affected by rigid vegetation of different vegetation heights. for the unsubmerged case (hv=50 cm, sr ), it can be noticed that the curve of water depth is much higher than the other cases before y=9 m location. as predicted in the research, the vegetation height making all the vegetation stems unsubmerged must be much higher than 8.75 cm. the vegetation height and average relative submergence of hv=8.75 cm, sr =1.49 are still far away from the unsubmerged case although the lowest relative submergence at y=9 m location is as low as 1.133. much higher blockage and retardance of the unsubmerged vegetation block can make the water depths before and within vegetation block higher. the effects of submerged vegetation on flow resistance are completely different to that of unsubmerged vegetation, the resistance mechanisms for the submerged and unsubmerged vegetation raising the water depths need further experimental studies (fathi-moghadam et al., 2011; zhang et al., 2014). it is obvious that vegetation blocks of different parameters show different affecting strength at different locations in the channel. for detailed analyzing the effects of rigid submerged vegetation on water depth at different locations, the average water depths at selected locations before (y=3 m), entering (y=5 m), within (y=7 m), outing (y=9 m), and after (y=11 m) corresponding to different experimental tests are plotted in figure 6. in the figure, it can be easily observed that the effects of vegetation heights or relative submergence are significant before and within the vegetation block. figure 6. variations of average water depth at different locations regarding to vegetation heights and average relative submergences. 43effects of vegetation height and relative submergence for rigid submerged vegetation on flow structure in open channel in order to determine the significance of vegetation height impacts on water depth at different locations, the relationship between them is analyzed by correlation analysis method corresponding to the integer transversesections between y=3 m to y=11 m. it should be noticed that the data from the unsubmerged case (hv=50 cm, sr =1) is removed to more accurately reflect the effects of submerged vegetation block. in shapiro-wilk normality test, the data of water depth at the transverse-section y =11 m is found not to obey the normal distribution (table 2). and spearman correlation coefficient is therefore adopted. the results of correlation analysis presented in table 2 are consistent with the conclusion directly overserved in figure 6. the effects of vegetation heights are significant before (y=3-5m) and within (y=5-9 m) the vegetation block. on the transverse-sections within these areas, the water depths increase with the increase of vegetation heights or decrease of relative submergences. for the other transverse-sections at and after the exit of vegetation block, the differences in water depth due to variable vegetation heights or relative submergences are negligible. that is why the curves for water depth in figure 5 contract to a bunch of approximatively overlapping lines at this area (y=9-11 m). although the impacts of vegetation height on the variation trends of water depth are illustrated above, the extent of such effects have not been quantified. as such, further analyses on the effects of vegetation height and relative submergence on the variation amplitude of water depth are performed. effects of vegetation height and relative submergence on the variation amplitude of water depth in recent years, vegetation planting and management have become common techniques for river ecological restoration and regulation. how the submerged vegetation blocks affect the variation amplitude of water depth or raised water depth is therefore significant to investigate. the differences between the maximum and minimum water depths hd corresponding to different vegetation heights and average relative submergence are plotted in figure 7. as can be seen in the figure, the water depth differences hd which can represent the variation amplitudes of water depths increase with the increase of vegetation height. as the water depths for 9 different cases are close near the exit of vegetation block, the variation amplitudes of water depth can represent the raised water depths in a great extent, which illustrate that the introduction of submerged vegetation can significantly alter the water surface profile in rivers. as can be seen in the figure, the water depth differences hd increase linearly with vegetation height. the variation trend can be fitted by linear function of hd = 𝓪' + b' hv (5) where a’=0.0497 and b’=0.4 are constant factors representing the intercept and the slope for the trend line, respectively. the coefficient of determination for this fitting is r2=0.999. considering the relationship between vegetation height and water depth presented in the last section, the water depth difference hd should decrease with the increase of relative submergence of vegetation. as shown in figure 7, the fitted relationship between hd and average relative submergence can be described by hd = c s d r (6) where c=0.097 and d=-0.315 are constant factors for this power function while the coefficient of determination is r2=0.945. in comparison with the linear relationship between the water depth difference hd and vegetation height hv, the power function seems more complex in form for predicting the effects of submerged vegetation on raised water depth. however, the average relative submergence sr is a more general parameter and applicable for different flow discharges in engineering practices, it just needs to calibrate the factors c and d (manes et al., 2007; wilson, 2007; fathi-moghadam et al., 2011). figure 7. variations of water depth difference regarding to vegetation heights and average relative submergences. effects of vegetation height and relative submergence on flow velocity distribution figure 8 shows the variations of flow velocity along the test section. as shown in the figure, the flow velocity of the case without vegetation changes very small along the flow direction and presents a very slight increasing trend as a whole, which is consistent with the slight decreasing trend of water depth presented in figure 5. except for these two lowest vegetation heights hv=1.25, 2.5 cm and largest average relative submergences , the flow velocities of the cases with submerged vegetation basically don’t change along the flow direction before 8 m location. when hv=1.25 and 2.5 cm, the flow velocities seem to decrease slightly in this section, which may be related to the bed roughness alternations due to the submerged vegetation with low height and large relative submergence. for table 2 statistical results of vegetation height impact on water depth location (y =) 3 m 4 m 5 m 6 m 7 m 8 m 9 m 10 m 11 m shapiro-wilk test results statistic 0.912 0.923 0.919 0.904 0.924 0.931 0.959 0.946 0.790 p-value 0.370 0.457 0.422 0.315 0.463 0.526 0.795 0.668 0.023 normal distribution p>0.05 yes yes yes yes yes yes yes yes no correlation analysis results r 1.000** 0.929** 0.905** 0.952** 0.929** 0.976** 0.714* 0.286 -0.214 p-value 0.000 0.001 0.002 0.000 0.001 0.000 0.047 0.493 0.610 note: ** represents significance level 0.01, * represents significance level 0.05. 44 zhongyu wang, huayong zhang, xing he, qi jiang, weigang xu, wang tian these two cases. the frontal vegetation area per unit bed area λ can be calculated as 0.018 and 0.035, which demonstrated bed-shear flow conditions according to the division standard (0<λ<0.04) in previous studies (nepf, 2012; zhao et al., 2019). when hv increases 3.75 cm and above, the flow regimes have changed while λ are larger than 0.04 (0.053 for the case hv=3.75 cm). however, further analysis of the bed roughness and vertical distribution of flow velocity under rigid submerged vegetation conditions are necessary to connect these two issues in the future. after y=8 m location, the curves of flow velocity for all the cases with submerged vegetation increase rapidly towards the level of the blank case. such increases near the exit of vegetation block are reasonable in considering the variations of water depth shown in figure 5. for the case with unsubmerged vegetation (hv=50 cm, sr =1), the flow velocities are obviously lower than the other cases and increase when the water just flows into the vegetation block, which presents a better consistency to the trend of water depth. these large increasing gradients of flow velocity at the interface of vegetation block agree with previous study (green, 2005; zhang, 2015b). furthermore, it can be noticed in figure 8 that the curves of flow velocity are crossed or overlapping at many locations, which makes it difficult to rank the flow velocity. anyway, the distributions of flow velocity are significantly affected by the rigid submerged vegetation of different vegetation heights or relative submergences. a problem that cannot be ignored is that high-velocity water flow may erode the river bed and banks (coon, 1998; qian et al., 2016). as reported in previous studies, vegetation patches planted properly will help to better protect against such erosion (liu et al., 2017). as pointed out by najafabadi et al. (2015), the interactions between accelerating flow velocity and vegetated banks are dominant to the flow characteristics of vegetated open channel. a further study on the special values is therefore necessary to realize the effects of vegetation heights or relative submergences on flow velocity distributions and hydraulic erosion. figure 8. variations of flow velocity along the test section corresponding to different vegetation heights and average relative submergences. as mentioned above, an analysis on the variations of the maximum, minimum flow velocities and the differences between them as well as the average flow velocities are investigated regarding to the vegetation heights or relative submergences. as shown in figure 9, the maximum flow velocities fluctuate slightly and basically don’t change with the increase of vegetation heights or decrease of average relative submergences. the fluctuation for the curve of maximum flow velocity is less than 0.135 m/s. for the curve of minimum flow velocities, the values decrease from 0.816 m/s to 0.378 m/s due to the increase of vegetation heights and therefore retardance by vegetation. the decreasing amplitude of the minimum flow velocities is up to 0.438 m/s. the variation of the average flow velocities is just higher in numerical values than the curve for the minimum flow velocities while the variation tendency is similar. it is obvious that the difference between the maximum and minimum flow velocities should increase with the vegetation heights as the maximum values change slightly and minimum values decrease continuously. such difference is as high as 0.772 m/s when the vegetation stems are unsubmerged (hv=50 cm, sr =1) and significantly higher than 0.566 m/s for the case hv=8.75 cm and sr =1.49, which is consistent to the variations of water depths. figure 9. variations of special values of flow velocities regarding to vegetation heights or average relative submergences. conclusions based on a series of experiments conducted in a laboratory channel, the vegetation height and a concept of relative submergence are adopted to analyze the effects of rigid submerged vegetation on flow structure in this research. with the experimental results of the research, the following conclusions can be drawn. (1) the relative submergences present decreasing trend along the vegetation section except for unsubmerged case of the present research. the linear regression function havg=a+bhv between vegetation height and caseaveraged water depth demonstrates a decrease trend of average relative submergence with the increase of vegetation height. (2) the differences between the maximum and minimum water depths corresponding to the submerged cases follow a linear relationship hd=𝓪'+b' hv with vegetation height and a negative power functional relationship hd=c s d r with the average relative submergence. with these relationships, the raised water depths due to rigid submerged vegetation can be predicted, especially for the relative submergence which is more general than the vegetation height. (3) for most submerged cases, the flow velocities basically don’t change before 8m location, and increase rapidly towards the level of the case without vegetation after 8m location. the differences between the maximum and minimum flow velocities increase with the vegetation heights as the maximum values change slightly and minimum values decrease continuously. some parameters in the empirical relationships obtained in this research need to be calibrated in practical applications due to the complexity of vegetation and hydraulic conditions. the findings of the research may improve the comprehensive understandings of the effects of vegetation heights and 45effects of vegetation height and relative submergence for rigid submerged vegetation on flow structure in open channel relative submergences of rigid submerged vegetation in open channel and may give some guidance to the practices of river ecological engineering design and management. acknowledgments the authors would like acknowledge with great appreciation for the support provided by the chinese national major science and technology program for water pollution control and treatment (no. 2017zx07101-002, no. 2015zx07204-007, no. 2015zx07203-011) and the chinese fundamental research funds for the central universities (no. 2021ms047). references ben meftah, m. & mossa, m. (2013). prediction of channel flow characteristics through square arrays of emergent cylinders. physics of fluids, 25(4), 13-25. doi: 10.1063/1.4802047 callow j. n. (2012). potential for vegetation-based river management in dryland, saline catchments. river research and applications, 28(8), 1072-1092. doi: 10.1002/rra.1506 chanson h. (2004). the hydraulics of open channel flows: an introduction (second edition). elsevier butterworth-heinemann, oxford, united kingdom, 585 pp. doi: 10.1016/b978-0-7506-5978-9.x5000-4 chauhan m. & gopal b. (2005). vegetation structure and dynamics of a floodplain wetland along a subtropical regulated river. river research and applications, 21(5), 513-534. doi: 10.1002/rra.821 chow, v. t. (1959). open-channel hydraulics. mc-graw hill, new york, united states, 680 pp. coon w. f. (1998). estimation of roughness coefficients for natural stream channels with vegetated banks. us geological survey water-supply paper, reston, united states, 133 pp. doi: 10.3133/wsp2441 fathi-moghadam, m., kashefipour, m., ebrahimi, n. & emamgholizadeh, s. (2011). physical and numerical modeling of submerged vegetation roughness in rivers and flood plains. journal of hydrologic engineering, 16(11), 858-864. doi: 10.1061/(asce)he.1943-5584.0000381 fathi-maghadam, m. & kouwen, n. (1997). nonrigid, nonsubmerged, vegetative roughness on floodplains. journal of hydraulic engineering, 123(1), 5157. doi: 10.1061/(asce)0733-9429(1997)123:1(51) friberg, n., harrison, l., o’hare, m. & tullos, d. restoring rivers and floodplains: hydrology and sediments as drivers of change. ecohydrology, 10(5), e1884. doi: 10.1002/eco.1884 green, j. c. (2005). comparison of blockage factors in modelling the resistance of channels containing submerged macrophytes. river research applications, 21(6), 671-686. doi: 10.1002/rra.854 gualtieri, p., felice, s. d., pasquino, v. & doria, g. p. (2018). use of conventional flow resistance equations and a model for the nikuradse roughness in vegetated flows at high submergence. journal of hydrology and hydromechanics, 66(1), 107-120. doi: 10.1515/johh-2017-0028 huai, w. x., zeng, y. h., xu, z. g. & yang, z. h. (2009). three-layer model for vertical velocity distribution in open channel flow with submerged rigid vegetation. advances in water resources, 32(4), 487-492. doi: 10.1016/j.advwatres.2008.11.014 järvelä, j. (2005). effect of submerged flexible vegetation on flow structure and resistance. journal hydrology, 307(1-4), 233-241. doi: 10.1016/j.jhydrol.2004.10.013 jing, y., zhang, h. y., wang, z. y., xu, w. g., ji, c. m. & zhang h. (2010). the change of relative turbulence intensity within the reed population, proceedings of the asme 2010 international mechanical engineering congress & exposition, vancouver, canada, november, 8, 397-401. doi: 10.1115/imece2010-40783 kouwen, n. & fathi-moghadam, m. (2000). friction factors for coniferous trees along rivers. journal of hydraulic engineering, 126(10), 732-740. doi: 10.1061/(asce)0733-9429(2000)126:10(732) lama, g. f. c., crimaldi, m., pasquino, v., padulano, r. & chirico, g. b. (2021). bulk drag predictions of riparian arundo donax stands through uav-acquired multispectral images. water, 13(10), 1333. doi: 10.3390/ w13101333 liu, d., valyrakis, m. & williams, r. (2017). flow hydrodynamics across open channel flows with riparian zones: implications for riverbank stability. water, 9(9), 720. doi: 10.3390/w9090720 manes, c., pokrajac d. & mcewan i. (2007). double-averaged open-channel flows with small relative submergence. journal of hydraulic engineering, 133(8), 896-904. doi: 10.1061/(asce)0733-9429(2007)133:8(896) mcbride, m., hession, w. c., rizzo, d. m. & thompson, d. m. (2007). the influence of riparian vegetation on near-bank turbulence: a flume experiment. earth surface processes and landforms, 32(13), 2019-2037. doi: 10.1002/esp.1513 musleh, f. a. & cruise, j. f. (2006). functional relationships of resistance in wide flood plains with rigid unsubmerged vegetation. journal hydraulic engineering, 132(2), 163-171. doi: 10.1061/(asce)07339429(2006)132:2(163) najafabadi, e. f., afzalimehr, h. & sui, j. y. (2015). turbulence characteristics of favorable pressure gradient flows in gravel-bed channel with vegetated walls. journal of hydrology and hydromechanics, 63(2), 154-163. doi: 10.1515/johh-2015-0019 nepf, h. m. (1999). drag, turbulence, and diffusion in flow through emergent vegetation. water resources research, 35(2), 479-489. doi: 10.1029/1998wr900069 nepf, h. m. (2012). flow and transport in regions with aquatic vegetation. annual review of fluid mechanics, 44, 123-142. doi: 10.1146/annurev-fluid-120710-101048 nezu, i. & sanjou, m. (2008). turbulence structure and coherent motion in vegetated canopy open-channel flows. journal of hydro-environment research, 2(2), 62-90. doi: 10.2208/journalam.10.795 nikora, n., nikora, v. & o’donoghue, t. (2013). velocity profiles in vegetated open-channel flows: combined effects of multiple mechanisms. journal of hydraulic engineering, 139(10), 1021-1032. doi: 10.1061/(asce) hy.1943-7900.0000779 o’briain, r., shephard, s. & coghlan, b. (2017). river reaches with impaired riparian tree cover and channel morphology have reduced thermal resilience. ecohydrology, 10(8), e1890. doi: 10.1002/eco.1890 okamoto, t., nezu, i. & ikeda, h. (2014). vertical mass and momentum transport in open-channel flows with submerged vegetations. journal of hydro-environment research, 6(4), 287-297. doi: 10.1016/j.jher.2012.03.002 pasquino, v., gualtieri, p. & doria, g.p. (2016). on evaluating flow resistance of rigid vegetation using classic hydraulic roughness at high submergence levels: an experimental work. in: rowiński, p. & marion a. (editor). hydrodynamic and mass transport at freshwater aquatic interfaces, springer, cham, switzerland, 269-277. doi: 10.1007/978-3-319-27750-9_22 poggi, d., porporato, a., ridolfi, l., albertson, j. d. & katul, g. g. (2004). the effect of vegetation density on canopy sub-layer turbulence. boundary-layer meteorology, 111, 565–587. doi: 10.1023/b:boun.0000016576.05621.73 pujol, d., colomer, j., serra, t. & casamitjana, x. (2010). effect of submerged aquatic vegetation on turbulence induced by an oscillating grid. continental shelf research, 30(9), 1019-1029. doi: 10.1016/j.csr.2010.02.014 qian, f., cheng, d. b., ding w. f., huang j. s. & liu j. j. (2016). hydraulic characteristics and sediment generation on slope erosion in the three gorges reservoir area, china. journal of hydrology and hydromechanics, 64(3), 237-245. doi: 10.1515/johh-2016-0029 rhee, d. s., woo, h., kwon, b. a. & ahn, h. k. (2008). hydraulic resistance of some selected vegetation in open channel flows. river research and applications, 24(5), 673-687. doi: 10.1002/rra.1143 https://doi.org/10.1146/annurev-fluid-120710-101048 https://doi.org/10.1146/annurev-fluid-120710-101048 46 zhongyu wang, huayong zhang, xing he, qi jiang, weigang xu, wang tian sandercock, p. j. & hooke, j. m. (2010). assessment of vegetation effects on hydraulics and of feedbacks on plant survival and zonation in ephemeral channels. hydrological processes, 24(6), 695-713. doi: 10.1002/ hyp.7508 stone, b. m. & shen, h. t. (2002). hydraulic resistance of flow in channels with cylindrical roughness. journal of hydraulic engineering, 128(5), 500506. doi: 10.1061/(asce)0733-9429(2002)128:5(500) temmerman, s., bouma, t. j., govers, g., wang, z. b., de vries, m. b. & herman, p. m. j. (2005). impact of vegetation on flow routing and sedimentation patterns: three-dimensional modeling for a tidal marsh. journal of geophysical research, 110(f4), f04019. doi: 10.1029/2005jf000301 thorne, c. r. (1990). vegetation and erosion: processes and environments. wiley, chichester, united kingdom, 518 pp. velasco, d., bateman, a., redondo, j. m. & demedina, v. (2003). an open channel flow experimental and theoretical study of resistance and turbulent characterization over flexible vegetated linings. flow, turbulence and combustion, 70, 69-88. doi: 10.1023/b:appl.0000004932.81261.40 white, b. l. & nepf, h. m. (2003). scalar transport in random cylinder arrays at moderate reynolds number. journal of fluid mechanics, 487, 43-79. doi: 10.1017/s0022112003004579 wilson, c. a. m. e. (2007). flow resistance models for flexible submerged vegetation. journal of hydrology, 342(3-4), 213-222. doi: 10.1016/j.jhydrol.2007.04.022 wu, f. c., shen, h. w. & chou, y. j. (1999). variation of roughness coefficients for unsubmerged and submerged vegetation. journal of hydraulic engineering, 125(9), 934-942. doi: 10.1061/(asce)0733-9429(1999)125:9(934) yagci, o., tschiesche, u. & kabdasli, m. s. (2010). the role of different forms of natural riparian vegetation on turbulence and kinetic energy characteristics. advances in water resources, 33(5), 601-614. doi: 10.1016/j. advwatres.2010.03.008 yokojima, s., kawahara, y. & yamamoto, t. (2015). impacts of vegetation configuration on flow structure and resistance in a rectangular open channel. journal of hydro-environment research, 9(2), 295-303. doi: 10.1016/j. jher.2014.07.008 zhang, g. h., liu, g. b., yi, l. & zhang, p. c. (2014). effects of patterned artemisia capillaris on overland flow resistance under varied rainfall intensities in the loess plateau of china. journal of hydrology and hydromechanics, 62(4), 334-342. doi: 10.2478/johh-2014-0035 zhang, h. y., wang, z. y., dai, l. m. & xu, w. g. (2015a). influence of vegetation on turbulence characteristics and reynolds shear stress in partly vegetated channel. journal of fluids engineering, 137(6), 061201. doi: 10.1115/1.4029608 zhang, h. y., wang, z. y., xu, w. g. & dai, l. m. (2015b). effects of rigid unsubmerged vegetation on flow field structure and turbulent kinetic energy of gradually varied flow. river research and applications, 31(9), 11661175. doi: 10.1002/rra.2814 zhao, f. & huai w. x. (2016). hydrodynamics of discontinuous rigid submerged vegetation patches in open-channel flow. journal of hydro-environment research, 12, 148-160. doi: 10.1016/j.jher.2016.05.004 zhao, h. q., yan, j., yuan, s. y., liu, j. f. & zheng, j. y. (2019). effects of submerged vegetation density on turbulent flow characteristics in an open channel. water, 11, 2154. doi: 10.3390/w11102154 zhou, z. c., shangguana, z. p. & zhao, d. (2006). modeling vegetation coverage and soil erosion in the loess plateau area of china. ecological modelling, 198(1-2), 263-268. doi: 10.1016/j.ecolmodel.2006.04.019 time series models are often used in hydrology and meteorology studies to model streamflows series in order to make forecasting and generate synthetic series which are inputs for the analysis of complex water resources systems. in this paper we introduce a new modeling approach for hydrologic and meteorological time series assuming a continuous distribution for the data, where both the conditional mean and conditional variance parameters are modeled. bayesian methods using standard mcmc (markov chain monte carlo methods) are used to simulate samples for the joint posterior distribution of interest. two applications to real data sets illustrate the proposed methodology, assuming that the observations come from a normal, a gamma or a beta distribution. a first example is given by a time series of monthly averages of natural streamflows, measured in the year period ranging from 1931 to 2010 in furnas hydroelectric dam, brazil. a second example is given with a time series of 313 air humidity data measured in a weather station of rio claro, a brazilian city located in southeastern of brazil. these applications motivate us to introduce new classes of models to analyze hydrological and meteorological time series. abstract keywords: hydrology time series, meteorological time series, conditional regression models, bayesian analysis, mcmc methods. seasonal hydrological and meteorological time series issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n2.65577 edilberto cepeda cuervo1, jorge alberto achcar2, marinho g. andrade3. 1. statistics department science faculty universidad nacional de colombia, bogotá, colombia email: ecepedac@unal.edu.co 2. social medicine department fmrp, universidade de são paulo ribeirão preto, s.p., brazil 3. applied mathematics and statistics department icmc, universidade de são paulo sao carlos, s.p., brazil earth sciences research journal earth sci. res. j. vol. 22, no. 2 (june, 2018): 83-90 los modelos de series de tiempo se usan a menudo en estudios de hidrología y meteorología para modelar series de flujos a fin de hacer pronósticos y generar series sintéticas que son insumos para el análisis de sistemas complejos de recursos hídricos. en este artículo presentamos un nuevo enfoque de modelado para series de tiempo hidrológicas y meteorológicas asumiendo una distribución continua para los datos, donde se modelan los parámetros tanto de la media condicional como de la varianza condicional. métodos bayesianos estándares que usan mcmc (markov chain monte carlo) son usados para simular muestras de la distribución a posteriori conjunta de interés. dos aplicaciones a conjuntos de datos reales ilustran la metodología propuesta, asumiendo que las observaciones provienen de una distribución normal, gamma o beta. un primer ejemplo está dado por una serie temporal de promedios mensuales de los caudales naturales, medidos en el período anual que va de 1931 a 2010 en la presa hidroeléctrica de furnas, brasil. un segundo ejemplo considera una serie temporal de 313 datos de humedad del aire medidos en una estación meteorológica de río claro, una ciudad brasileña ubicada en el sureste de brasil. estas aplicaciones nos motivan a introducir nuevas clases de modelos para analizar series de tiempo hidrológicas y meteorológicas. resumen palabras clave: : series de tiempo hidrológicas; series de tiempo meteorológicas; modelos de regresión condicional; análisis bayesiano; métodos mcmc. record manuscript received: 09/06/2017 accepted for publication: 27/04/2018 how to cite item cepeda-cuervo, e., achcar, j. a., & andrade, m. g. (2018). seasonal hydrological and meteorological time series. earth sciences research journal, 22(2), 83-90. doi: http://dx.doi.org/10.15446/esrj.v22n2.65577 series de tiempo hidrológicas y meteorológicas estacionales h y d r o m e t e o r o l o g y 84 edilberto cepeda cuervo, jorge alberto achcar, marinho g. andrade 1. introduction time series models are often used in hydrology studies to model streamflow series in order to make predictions and to generate synthetic series which are inputs for the analysis of complex water resources systems (see, for example, salas et al., 1980, 1982; hosking, 1984; hipel& mcleod, 1994; montanari et al., 1997; hasebe et al., 2000). in many studies, hydrologists also use time series data to display the amount of rainfall that has fallen in a region for the past day, year or a period of 10 years (see for example, guimaraes& santos, 2011, and lee & lee, 2000). modeling hydrological variability is very important in the planning and management of water resources. many aspects of the hydrologic cycle could be described by time series data. researchers, usually use time series data to evaluate the resources of a water basin. important variables related to streamflow and watershed describes streamflow properties such as monthly flows or streamflow parameters. assuming a specified time series model, we usually could estimate the streamflow parameters using a classical or a bayesian inference approach. different time series models as arma and higher orders of ma models have been used by some authors when considering hydrologic regionalization of watersheds (see for example, chiang et al., 2002 a, b; wang et al., 2015). valipour et al.(2013) studied annual runoff time series in dahuofang reservoir, in northeast china, using autoregressive integrated moving average (arima) models coupled with ensemble empirical mode decomposition (wu, z., & huang, n. e., 2009). spectral analysis and forecasting of hydrological time series has also been considered (marques et al., 2006), including geostatistical applications (robin et al., 1993) and climate change investigations (lall and mann, 1995). considering hydrological time series, the monthly streamflow series typically have a periodic behavior in the mean and variance and in general, periodic autoregressive models are used in de analysis of the data (see, for example, modal & wasimi, 2006). in this situation, usually it is assumed that the series flow has a normal or log-normal distribution (see for example, tesfaye et al., 2006; wang et al., 2009). the behavior of meteorological time series is similar to the behavior of hydrological time series. in this case, given that the relative air humidity is a random variable with values given in the open interval (0, 1), we could assume a beta distribution to analyze the data. another possibility to analyze the data set is to consider a transformation of the data and to assume a normal distribution for the transformed data. as a special case, we could assume a logistic transformation. in this paper, a more general model is considered in the analysis of the hydrological or meteorological time series conditional to the historical available information: it is assumed that the data is generated from a normal, a gamma or a beta distribution, with conditional mean and variance, given respectively, by e [yt|yt-1] and v [yt| yt-1], where the index tis related to time. thus a general model is proposed to analyze hydrological or meteorological time series, assuming that the observations are generated from a continuous biparametric exponential family of distributions. as an illustration and motivation for the use of the proposed models, we first consider as a first example, a data set consisting of the time series of monthly averages of natural streamflows, measured in the year period ranging from 1931 to 2010, in furnas hydroelectric dam, located in southeastern brazil. this time series is shown in figure 1. from this figure, we observe that the streamflow series have a periodic behavior in the mean and variance and in this case, general periodic autoregressive models are usually assumed in the analysis of the time series data (modal & wasimi, 2006). to take into account the heteroscedasticity in the time series of streamflows showed in this first example, we propose a new model which assumes seasonal and autoregressive terms in the modeling of the mean and variance parameters. in this way, we propose a periodic and heteroscedastic model, in which the variance also presents an autoregressive structure. as a second example, we consider another time series with behavior similar to the assumed hydrological time series presented in the first example, consisting of a meteorological time series given by weekly averages of air humidity, measured in the city of rio claro, são paulo state, brazil (figure 2). in this case, given that the relative air humidity is a random variable with values in the open interval (0, 1), it is assumed a beta distribution in the analysis of the data. in this case, we propose a jointly modeling for the mean and for the variance of the data considering beta regression models, including seasonal and autoregressive terms in both regression models figure 1. time series of monthly averages of natural streamflows. figure 2. air humidity time series data. the paper is structured as follows: in section 2, it is introduced the seasonality analysis of the time series. in section 3, it is proposed seasonal autoregressive models. in section 4, it is presented the results of the analysis of the hydrological time series obtained using the proposed models assuming normal and gamma distributions. in section 5, it is presented the results of the analysis of air humidity time series. finally, in section 6 it is presented some conclusions and future research topics. 85seasonal hydrological and meteorological time series (1) (2) (6) (7) (3) (4) (5) 2. a period model in this section, we introduce a new modeling approach including seasonality terms in the model which better describes time series of monthly averages of natural stream flows, denoted by yt. in the first case, we consider the time series data related to the monthly averages of natural streamflows in furnas hydroelectric dam (figure 1). a preliminary spectral analysis is developed for this time series to determine the time periods to be considered in the mean and variance model formulations. thus, if in the spectral analysis, the number of observations is t = 2q + 1, where q is a positive integer number, the fourier time series model given by: is to be fitted by the data, where fi = i/t is the ith harmonic of the fundamental frequency 1/t and, α1,i and α2,i, i = 1,...,q, are the related coefficients and et is an error term assumed to have the first and second moments given respectively by e(et) = 0, e (et 2) = σ2 and to be uncorrelated, that is, e(et et+k) = 0 for k ≠ 0. considering the n observations of the time series, the least square estimates of the coefficients α0 and (α1,i ,α2,i), i = 1,...q, are obtained from the equations: based on these estimates, the intensity of each frequency is calculated by: observe that the highest frequency is 0.5 cycle per month (time interval) since the smallest period is 2 months. this preliminary analysis of the intensities of frequencies allows us to reduce the number of harmonics considered in the time series included in the model, considering only those frequencies that have higher intensities (marques et al., 2006). for a series consisting of 80 years of observations (that is 960 months) we would have 480 harmonics, but considering this preliminary analysis, we could reduce the number of harmonics to three or four. usually, in the case of monthly time series, these periods are given by periods of 6 and 12 months. these periods will be discussed in section 4, where the specification of the parameters related to the periods is made simultaneously with the specification of the parameters for the autoregressive models assumed for the mean and variance in the time series formulations to be introduced in section 3. if some of these parameters show to be not significant after a preliminary statistical analysis, the terms related to these parameters are deleted from the model to develop a further analysis. the final decision on which harmonics should be included in the model must be made considering some criteria for model selection or hypothesis test on the fitted coefficients α1,i and α2,i. 3. the proposed seasonal autoregressive model in hydrological or meteorological time series, we usually assume that the observations of interest are generated from a conditional continuous probability distribution function. as special cases, we could assume that the observations are generated from standard normal, gamma, beta or exponential conditional density functions, denoted by f (yt│ht-1 ), t = 1,2,...,t , where ht-1 is the available information up to time t 1 and thus yt has conditional means and variances given respectively by µt = e (yt |ht-1) and ht = var(yt |ht-1), following the models: where β = {α0, α1,1 ,..., α1,q ,α2,1,..., α2,q , ϕ1,...,ϕp ,}, is the vector of parameters for the mean model and γ = {λ0, λ1,1,..., λ1,s ,λ2,1 ,..., λ2,s, θ1,...,θr} is the vector of the parameters for the variance model, fi = i/t the ith harmonic of the fundamental frequency 1/t. in this paper, the parameters of the proposed models are estimated under a bayesian approach. in order to illustrate the proposed methodology, we also include the regression equations relating the mean and variance parameters to the assumed covariates assuming gamma and beta distributions for the data. in this way, we have, the following modeling steps: 1. if yt, t = 1,2,...,t, follows a gamma conditional distribution g(pt , qt), where g(p,q) denotes a gamma distribution with mean pq and variance pq2, the conditional mean and variance are related to the original parameters by the equations µt= pt qt and ht = µt qt. 2. if yt, t = 1,2,...,t, follows a beta distribution function b(pt,qt), we consider a reparameterization of the beta distribution density as a function of the mean and precision, ϕt = pt + qt, which results to be appropriate in order to define the joint mean and precision beta regression models as introduced by cepeda (2001). this reparameterization, where ϕ = p + q, p = µϕ and q = ϕ(1 µ), has been extensively used in the literature following the joint modeling approach for the mean and precision beta parameters introduced by cepeda (2001) and cepeda and gamerman (2005), under a bayesian approach. it is important to point out that ferrari and cribarineto (2004) also introduced a beta modeling approach for the mean but assuming constant precision parameters, under a classical approach. in all of these cases, ϕ can be interpreted as a precision parameter in the sense that, for fixed values of µ, larger values of ϕ correspond to smaller values for the variance of y. this interpretation could be not so simple. in this paper, we use the mean and variance reparameterization of the probability beta distribution in the definition of joint mean and variance beta regression models (cepeda-cuervo, 2015), taking into account that µ(1 µ) > σ2; in this way, samples of the joint posterior distribution for the regression parameters should be simulated in the subspace of parameters that satisfy this property. although this reparameterization results in a 86 edilberto cepeda cuervo, jorge alberto achcar, marinho g. andrade complex expression for the beta distribution, it leads to a best and more easily interpretation for the statistical analysis results in the applications. (8) (10) (11) (9) in this reparameterization, where the autoregressive seasonal beta regression models have the conditional mean and variance model given by the equations (6) and (7). special cases of this model could be easily obtained from this general model. a first model is given by a seasonal regression mean model, with mean given by (6) and autoregressive variance not considering the presence of seasonal terms. a second model, also a seasonal mean model, is given assuming the mean given by (6) and a seasonal variance model not considering the presence of autoregressive terms. a third model, is given by an autoregressive mean model and a variance model, not considering the presence of seasonal terms in the mean and in the variance. a fourth model, is given by an autoregressive model, with constant variance. 4. hydrological time series in this section we consider an analysis of the furnas dam hydroelectric hydrological time series dataset, introduced in section 1, assuming seasonal autoregressive conditional heteroscedastic models. the first step in the proposed analysis is to determine the period for the harmonics of higher intensity in the spectral analysis of the streamflows data. in this way, we note in figure 3 that the harmonics of higher intensity (i) corresponds to the cycle (1/fi) of 6 and 12 months, given that the periodogram shows that this time series contains two cosinesine peaks at these frequencies. there are other very small peaks in this periodogram, possibly caused by noise components. thus, the seasonal terms to be included in the mean equation model of the streamflow series are given by cos(2πt/6), sin(2πt/6), cos(2πt/12) and sin(2πt/12). the periodogram of a time series can be obtained using the r-function periodogram (x, method,..) of the library "genecycle" (ahdesmaki et al., 2012). other statistical software can be used to estimate the periodogram of a seasonal time series, for example, mathlab or the statistical software excel-xlstat. figure 3. periodogram of natural streamflows time series. many autoregressive models could be assumed to analyze this data set. as special cases, we will assume, in section 4.1, a seasonal conditional normal distribution for the data and, in section 4.2, a seasonal conditional gamma distribution for the data. in order to apply the bayesian methodology, independent normal prior distributions n(0,10k), with k = 2, are assumed for the regression parameters associated with the seasonal terms. for the other parameters of the model, we assume independent normal prior distributions n(0,10k), with k = 5. it is important, to point out that non-informative prior could be used (see, for example, geman, 2006), but in our case, the obtained results are very similar. since we are using the free available software winbugs to simulate samples for the joint posterior distribution of interest in all assumed models, which only requires the introduction of the joint distribution for the data and the prior distributions for all parameters of the model, we will not present in this paper the required full conditional posterior distributions for all parameters needed in the gibbs sampling or the metropolis-hastings algorithms. 4.1. normal seasonal time series in this section, we present the results of the bayesian analysis for the time series of monthly averages of natural streamflows, measured in the period ranging from 1931 to 2010, considering the dataset related to a hydroelectric dam introduced in section 1, assuming jointly modeling for the mean and variance, assuming a normal model for the data. from the model given by equations (6) and (7), many autoregressive models were considered in the analysis of the data, and the model with smallest dic (deviance information criterion, introduced by spiegelhalter et al, 2002) value was given by the heteroscedastic normal regression model with conditional mean and variance structures given respectively by, samples of the joint posterior distribution of interest were simulated using standard mcmc (markov chain monte carlo) methods and the free available winbugssoftware (spiegelhalter et al, 2003). in each of the cases, many samples were generated starting from different initial values. all of them showed the same behavior, after a small burn-in period consisting of 3,000 or 5,000 generated samples. convergence of the simulation algorithm was observed from trace plots of the generated gibbs samples. for the model given by equations (10) and (11), the value of the logarithm of the likelihood function evaluated at the obtained estimates for the parameters of the model was given by −2logl=10479.200 and the obtained dic value was dic=10876.500. monte carlo estimates of the posterior means for each parameter based on the generated gibbs samples and their respective standard deviations are given in tables 1 and 2, including both autoregressive and seasonal terms in the conditional mean and variance terms. table 1. normal model: bayesian estimates for the mean parameters. table 2. bayesian estimates of the variance parameters. 87seasonal hydrological and meteorological time series figure 5. normal fitted squared root of the expected volatility. figure 6. monthly averages of natural streamflow. figure 4. monthly averages of natural streamflow and normal fitted mean estimates. as an illustration of the performance of the proposed model, figure 4 shows the agreement between monthly averages of natural streamflows and the fitted mean estimate. at the same time, figure 5, depicting of the fitted squared root of the expected volatility shows periods of high volatility around of the months 200, 400, 600 and 800, as observed in the data behavior. 4.2 gamma seasonal time series in this section we present the results of the bayesian analysis for the time series for monthly averages of natural streamflows, measured in the period 1931 to 2010, in furnas hydroelectric dam, assuming joint modeling for the mean and variance autoregressive gamma models, that is, we assume that the observations of the interest are generated from a conditional gamma density function given by f (yt |ht-1), where ht-1) is the information up to time t − 1 and yt has conditional mean and conditional variance given respectively by µt = e(yt | ht-1) and σt 2 = var(yt | ht-1), and defined by (10) and (11). for this model, the logarithm of the likelihood function evaluated at the estimates for the parameters of interest is given by −2logl=10649 and the dic value is given by 10862.300. using the dic criterion to discriminate the two models (normal seasonal time series and gamma seasonal time series), we observe better fit of the data for the gamma seasonal time series model, since we have smaller dic value for this model. the bayesian estimates of the posterior means for the parameters together with the corresponding standard deviations are given in tables 3 and 4. table 3. gamma model: bayesian estimates for the mean parameters. table 4. gamma model: bayesian estimates for the variance parameters. although the mean parameter estimates given in tables 1 and 3, and variance parameter estimates given in tables 2 and 4, show some agreement between conditional normal and conditional gamma regression parameter estimates, the dic value of the conditional gamma model is smaller than that of the conditional normal heteroscedastic models, showing that the class of gamma models is better to fit the monthly averages of natural streamflow data. in order to determine the forecasting performance of the proposed gamma seasonal model, we fit this model to the first 910 observed values of the furnas dam hydrological time series, in order to predict the temporal behavior over the 50 last months. the parameter estimates agree with those reported in tables 3 and 4. figure 6 shows that this model has good performance in predicting monthly averages of natural streamflows. this figure shows the 90% prediction interval for the variable of interest in these last weeks in dashed lines, and the observed values of natural streamflows in black points. 5. beta mean and variance seasonal models applied to air humidity time series in this section it is assumed that the time series data are generated from a beta distribution, b(pi ,qi), with conditional mean and variance given, respectively, by (6) and (7). to illustrate the application of the proposed model to be fitted by the data set, we consider the time series of weekly averages air humidity, measured in rio claro city, located in southeastern of brazil, from 18/10/2002 to 08/10/2008. this time series, introduced in section 1, is shown in figure 2. as in section 4, the first step in the proposed analysis is to determine the period of harmonics of higher intensity in the spectral analysis of the humidity data. from the corresponding periodogram, figure 7, we note that the 88 edilberto cepeda cuervo, jorge alberto achcar, marinho g. andrade harmonics of higher intensity corresponds to the cycle (1/fi) of 26 and 52 days. as in the analysis of streamflow data, the periodogram shows that this time series contains two cosine-sine peaks at these frequencies, two larger peaks, and other very small peaks, possibly caused by noise components. thus, the seasonal term to be included in the mean equation model of the streamflow series are given by: cos(2πt/26), sin(2πt/26), cos(2πt/52) and sin(2πt/52). figure 8. air humidity time series data (continuous line) and fitted mean (diseased line) for double seasonal model. figure 9. fitted conditional variance time series. 5.1. seasonal mean and conditional variance models in this section, double seasonal beta repression models are proposed to analyze the air humidity time series data introduced in section 1. in this way, we conclude that the best beta seasonal mean and variance model is the one where the mean and dispersion regression structures are given by: (12) (13) assuming independent normal prior distributions n(0,100) for the regression parameters α0 and α1,i, for i = 1,2 and for λ0 and λ2,i, for i = 1,2. to fit the model 20,000 samples of the joint posterior distribution for the parameters of the model were also generated using the mcmc methods and the winbugs software (spiegelhalter et al, 2003). monte carlo estimates for the posterior means of each parameter were obtained from the final simulated gibbs sample, after an initial burn-in sample period of 2000 samples. this “burn-in sample” was discarded to eliminate the effect of the initial values in the iterative procedure. after this “burn-in sample” period we simulated another 20,000 gibbs samples choosing every 20 iteration to get approximately non-correlated samples, which gives a final sample of size 1,000 used to get the posterior summaries of interest. the posterior summaries of interest are given in table 5, for the mean regression parameters, and in table 6, for the variance regression parameters. table 5. beta model: bayesian estimates for mean parameters. table 6. beta model: bayesian estimates for the variance parameter. the logarithm of the likelihood function evaluated at the obtained estimates for the parameters of the model is given by log l = −853.082 and the dic criterion has a value equals to −825.525. in figure 8, we observe a good agreement between data and the fitted mean, showing the good performance of the proposed model to analyze this data set. figure 9, revels a good agreement for the variances, where smaller means are accompanied by smaller variance. this behavior also was observed in the original time series. that is, we conclude that this model fits by the time series data very well. to determine the forecasting performance of the proposed models, we fit the seasonal beta autoregressive model to the first 282 observed values of the air humidity time series data, in order to predict the time behavior in the 30 last weeks. figure 10 shows the 90% prediction interval for these last weeks, in dotted lines and the observed values of natural streamflows in black points. figure 7. periodogram of humidity times series. 89seasonal hydrological and meteorological time series figure 10. forecasting humidity time series. 6. concluding remarks in this paper, we introduce a new class of time series models assuming continuous random variables within the exponential family applied to hydrological and meteorological data. special cases of this proposed methodology were considered assuming, normal, gamma and beta distributions. these new models could be of great interest to analyze hydrological and meteorological data, since the original data usually could not be fitted assuming the standard modeling approach for the means of the data or assuming standard distribution assumptions for time series data. other important practical point: in practical work, usually it is required to model simultaneously the mean and the variance depending on a vector of parameters to get better predictions and to discover the real behavior of a hydrological or a meteorological time series. with this modeling approach, we could get better inferences as it was observed in the illustrations introduced in this paper. this approach usually is not considered by hydrological or meteorological researchers using standard available software for hydrology or meteorology time series. in this way, the use of bayesian methods is a good alternative to get accurate inferences and predictions, especially using mcmc methods, since these inference methods are not based on asymptotical results as its common with standard existing classical approaches based on maximum likelihood estimation procedures. other important point: under the use of a bayesian approach, we could consider informative prior distributions when is available prior opinion of experts in hydrology and meteorology. this means, better predictions. it is also important, to point out that the computational work needed in the simulation of samples of the joint posterior distribution to get the posterior summaries of interest is greatly simplified using standard existing softwares like the winbugs software.the proposed methodology was illustrated considering two brazilian data sets: a hydrological time series and a meteorological time series. a final point of relevance for the results of this paper: new statistical models for time series to model climatic variables leading to better inferences and predictions based on data from hydrology and meteorology can be of great practical interest, especially with the large climate changes that have been observed in the world in recent decades. to determine the performance of the proposed models, an extensive comparison with other models proposed in the literature should be developed. however, in order to compare the performance of the gamma seasonal models with in a model without a seasonal component, we fit the autoregressive model ar(3), for which the dic value was 11990, bigger than the dic value of the fitted gamma seasonal model. the sum of square residuals obtained from applying the ar(3) model is also bigger than that for the proposed models. for the ar(3) model, the sum of square residuals is 65679347 and for the proposed model it is 52976950. finally, we fit the ar(3) model to the humidity data, for which the dic value was -774.550, bigger than −825.525, the bic value of the fitted beta seasonal model. the sum of square residuals obtained from applying the ar(3) model to the humidity data is 17.1565, larger than the dic value (1.3786) of the proposed models. references ahdesmaki m, fokianos k, strimmer k (2012) package ‘genecycle’. r-project website. url: http://cran.r-project.org/web/packages/ genecycle/genecycle.pdf. accessed 2017 dec 5. cepeda-cuervo, e. (2001). variability modeling in generalized linear models, unpublished ph.d. thesis. mathematics institute, universidade federal do rio de janeiro. http://www.docentes.unal. edu.co/ecepedac/docs/modelagemda variabilidade. pdf cepeda-cuervo, e., andrade, m. g., & achcar, j. a. (2012). a seasonal and heteroscedastic gamma model for hydrological time series: a bayesian approach. in: aip conference proceedings, vol. 1490, p. 97. cepeda c. e. & gamerman d. (2005). bayesian methodology for modeling parameters in the two parameter exponential family. estadística, 57, 93-105. chiang, s. m., tsay, t. k. & nix, s. j. (2002a). hydrologic regionalization of watersheds. i: methodology development. journal of water resources planning and management, 128(1), 3-11. chiang, s. m., tsay, t. k. & nix, s. j. (2002b). hydrologic regionalization of watersheds. ii: applications. journal of water resources planning and management, 128(1), 1220. ferrari, s., & cribari-neto, f. (2004). beta regression for modeling rates and proportions. journal of applied statistics, 31, 799-815. fleming, s. w., marsh lavenue, a., aly, a. h., & adams, a. (2002). practical applications of spectral analysis to hydrologic time series. hydrological processes, 16(2), 565-574. guimaraes, r. & santos, e. g. (2011). principles of stochastic generation of hydrologic time series for reservoir planning and design: a case study. journal of hydrologic engineering. in press. hasebe, m., dandou, t., kumekawa, t. & neijou, s. (2000). time series analysis of monthly rainfall, mean air temperature and carbon dioxide. in: w. z. y. & s. x. hu (eds.) proceedings of the eighth international symposium on stochastic hydraulics, 533-537. beijing, china. hipel, k. w. & mcleod, a. e. (1994). time series modeling of water resources and environmental systems. elsevier, amsterdam, the netherlands. hosking, j. r. m. (1984). modeling persistence in hydrological time series using fractional differencing. water resources research, 20(12), 1898-1908. lee, j. y. & lee, k. k. (2000). use of hydrologic time series data for identification of recharge mechanism in a fractured bedrock aquifer system. journal of hydrology, 229, 190-201. lall u, mann m. 1995. the great salt lake: a barometer of low frequency climatic variability. water resources research 31(10), 2503– 2515. marques, c. a. f., ferreira, j. a., rocha, a., castanheira, j. m., melogoncalves, p., vaz, n. & dias, j. m. (2006). singular spectrum analysis and forecasting of hydrological time series. physics and chemistry of the earth, parts a/b/c, 31(18), 1172-1179. modal, m. s., & wasimi, s. a. (2006). generating and forecasting monthly flows of the ganges river with par model. journal of hydrology, 323(1-4), 41-66. montanari, a., rosso, r. & taqqu, m. s. (1997). fractionally differenced arima models applied to hydrologic time series: identification, estimation and simulation. water resources research, 33(1-4), 1035-1044. robin mjl, gutjahr al, sudicky ea, wilson jl. 1993. crosscorrelated random field generation with the direct fourier transform method. water resources research 29(7): 2385–2397. salas, j. d., delleur, j. w., yevjervich, v. & lane, w. l. (1980). applied 90 edilberto cepeda cuervo, jorge alberto achcar, marinho g. andrade modeling of hydrologic time series. water resources publications, littlton, usa. salas, j. d., boes, d. c. & smith, r. a. (1982). estimation of arma models with seasonal parameters. water resources research, 18(4), 1006-1010. spiegelhalter, d. j., best, n. g., carlin, b. p. & van der linde, a. (2002). bayesian measures of model complexity and fit. journal of the royal statistical society: series b, 64(4), 583-639. spiegelhalter, d. j., thomas, a., best n. g., & gilks, w. r. (2003). win-bugs user manual (version 1.4). mrc biostatistics unit, cambridge, u.k. tesfaye, y. g., meerschaert, m. m. & anderson, p. l. (2006). identification of periodic autoregressive moving average models and their application to the modeling of river flows. water resources research, 42(w01419), 1-11. wang, q. j., robertson, d. e. & chiew, f. h. s. (2009). a bayesian joint probability modeling approach for seasonal forecasting of streamflows at multiple sites. water resources research, 45(w05407), 1-18. wang, w. c., chau, k. w., xu, d. m., & chen, x. y. (2015). improving forecasting accuracy of annual runoff time series using arima based on eemd decomposition. water resources management, 29(8), 2655-2675. wu, z., & huang, n. e. (2009). ensemble empirical mode decomposition: a noise-assisted data analysis method. advances in adaptive data analysis, 1(01), 1-41. valipour, m., banihabib, m. e., &behbahani, s. m. r. (2013). comparison of the arma, arima, and the autoregressive artificial neural network models in forecasting the monthly inflow of dez dam reservoir. journal of hydrology, 476, 433-441. geofis. colomb. . santafe de bogota d.c. issn • 0121·2974 fractales y series de datos geofisicos luis alfredo montes vides profesor asistente departamento de geociencias-facultad de ciencias-universidad nacional de colombia montes l. a: fractales y series de datos geofisicos. geofis. colomb. 2:9-12, 1993. issn 0121-2974 resumen la geometria de fractales ha surgido como una herramienta potencialmente util para la caracterizaci6n de datos en geofisica. comunmente, los datos geofisicos conforman series de tiempo, que exhiben un comportamiento aleatorio 0 variaci6n a corto y a largo plazo. un ejemplo tfpico son los registros anuales de temperatura. la traza de un registro es una curva con una dimensi6n fractal d, caracterizada por un exponente h. en el presente trabajo se utiliza el metoda de analisis de rango en cambios de sscata, creado por h. e. hurst, para determinar la dimensi6n fractal de una serie de datos geofisicos, y su medida de auto-afinidad. abstract there is a new geometry which provides a potentially tool for the characterization of geophysical data: the fractal geometry. generally, geophysical data consist of records in time or data series, for example yearly records of temperature, and they show a random behavior or variation on both a short and a long-term time scale. the trace of a record is a curve with a fractal dimension d, and it is characterized by an exponent h. in this paper, the hurt's rescaled range analysis method is used to determine the fractal dimension of a geophysical data serie d and h, his self-affinity measure. 1. introduccion en los ultirnos 10 aries se ha reconocido la importancia de la geometria de fractales en las ciencias naturales, como apropiada para describir las estructuras presentes en la naturaleza. la geometrfa fractal suministra una descripci6n y modelo matematico para la mayorfa de las formas aparentemente compleja halladas en la naturaleza. su propiedad mas importante es la ausencia de cualquier caracteristica de longitud 0 tiempo. esto quiere decir que las funciones de distribuci6n de probabilidad de los fractales, adecuadamente normalizadas, pueden expresarse de manera invariante bajo cambios de escala. ei interes desde el punto de vista de la geoffsica se ha manifestado con ensayos en distintos campos por distintos autores (burrough, 1984; unwin, 1989; jones et al., 1989; hewett, 1986). burrough estim6 la dimensi6n fractal de datos geoffsicos, basado en anal isis de densidad de espectros de frecuencia. una caracterfstica fundamental de los datos geofisicos es el comportamiento estadfstico fuertemente no-gaussiano. jones y sus colegas, afinman que existe una correlaci6n entre los datos geofisicos y las propiedades fractales de sus series de tiempo. en este articulo se utiliza la ley empirica de analisis de rangos de escala planteado por hurst, para determinar la dimensi6n fractal de una secci6n en un registro electrico de pozo, partiendo del calculo empfrico del exponente de hurst (h) como medida del grade de autoafinidad en el mismo y la relaci6n con su dimensi6n fractal. 2. fundamentos teoricos la definici6n formal del concepto de fractal fue planteada por el creador de la geometrfa fractal, benoit mandelbrot, en los siguientes terminos: "un fractal es un conjunto para el cual la dimensi6n hausdorf-besicovith estrictamente 10 montes: fractales y series de datos geofisicos excede la dimensi6n topoi6gica"; y de una. manera menos formal como: "la forma hecha de partes similares de algun modo". un elemento 1-dimensional como un segmento de una linea posee una propiedad de similitud a escala, pudiendo dividirse en n partes identlcas al original, cada una un factor de escala 1/n. un objeto 2-dimensional puede descomponerse en n partes iguales al original con un factor de escala de (1/n)'i2. a su vez un cubo se descompone en n pequeiios cubos con factor de escala (1/n)113. en conclusi6n un objeto d-dimensional puede dividirse en n pequeiias partes, cada una similar a la primera y reducida en un factor de escala r = (n)1/d despejando 0 se obtiene la dimensi6n fractal delobjeto. o = 10g(n) / log(r) a diferencia de la dimensi6n euclidiana, la dimensi6n fractal es estrictamente no entera, bajo magnificaci6n, un segmento de la figura principal luce parecida perc no identica a la primera. las formas que tienen un comportamiento marcadamente aleatorio, se repiten estadisticamente unicamente cuando las dos direcciones (t y f[td se magnifican a escalas diferentes. si t se magnifica en un factor r(r*t), entonces f[t] se magnifica en un factor rh(rh*f). esta escala no uniforme es conocida como autoafinidad y esta ligada fuertemente al concepto de dimensi6,n fractal. la dimensi6n fractal de una forma auto-affn, se relaciona con el para metro de escala h mediante la ecuaci6n: 0= 2 h [1] 3. metodo de analisis de rangos bajo cambios de escala ei metoda estadistico de anallsis de rango bajo cambios de escala, fue inventado por h.e. hurst en 1950 (hurst, 1951; hurst et at, 1965). para su entendimiento se considera el siguiente problema: se debe estimar la capacidad de almacenamiento de una represa bas ado en registros observados de descargas de agua, el valor ideal del volumen de la represa es aquel que nunca permita que esta se rebose 0 que se vacie completamente. en un tiempo t recibe una descarga de agua &(t), liberando en un perfodo de tiempo t la cantidad promedio <&(t»,. ei f1ujo promedio 0 el volumen recibido en un periodo de t aiios es: [2] ese volumen promedio recibido debe ser igual al volumen liberado en ese mismo periodo. a su vez el volumen neto acumulado 0 la diferencia entre el volumen recibido y el liberado en un periodo r es: [3] la diferencia entre el maximo y minima f1ujo acumulado x es el rango r, el cual mide la capacidad de almacenamiento para mantener la descarga media y nunca liegar al vaciado complete. r('r) = max x(t,-r) min x(t,-r) l s t s r l s t s r [4] donde t es un entero y t es el periodo considerado, la desviaci6n estandar se estima de las observaciones asf: i=t s = { (1ft) ~1 {e(t)-<e(t», }2 r [5] investigando muchos fen6menos naturales, hurst encontr6 que la relaci6n rls para tales fen6menos se describe muy bien por la relaci6n: r/s = (-r/2)h [6] que es la liamada ley empfrica de hurst claramente r depende del perlodo t en consideraci6n y se espera que aumente al incrementarse t, es decir al cambiar la escala de tlsmpos. este concepto de anallsis de rango bajo cambios de escala coincide con el concepto de auto-afinidad de un fractal cuando se buscan partes similares al total al realizar cam bios de escala 0 magnificaci6n. 4. proceso de datos en el presente trabajo se tom6 un registro electrico dlgitalizado. para garantizar la calidad de los datos se tomaron aquellos que estaban a una profundidad mayor de 100 m y hasta un limite de 480 m como cota inferior. dicho registro pertenece a una zona de rocas sedimentarias, cuya secuencia refleja el comportamiento del clima en perfodos geol6gicos lejanos. los valores que representan la resistividad de las rocas a diferentes profundidades, estan consignados en la tabla 1. se consideraron 10 periodos diferentes, el primero correspondia a los 10 m iniciales de la secuencia (a partir de los 100 m), para los siguientes se increment6 el valor del perfodo (20 m para el segundo, etc) hasta abarcar la totalidad de la secuencia 380 m (480 m tope de la secuencia). en cada perfodo se calcul6 el valor medio de la resistill 'dad, [2], y el acumulado [3). se observ6 que este valor fuera estrictamente mayor que cero (0). geofisica colombiana n° 2, abril de 1993 11 a continuaci6n se determin6 el valor del range r [4], asf como la desviaci6n sstandar alrededor de la resistividad media, en cada perfodo. luego para cada perlodo se encontr6 el cociente entre el rango (r) y la desviaci6n sstandar (s). finalmente se hizo una regresi6n lineal entre el cociente r1s y el perfodo en una escala loglog, para determinar el exponente h [6] y la dimensi6n fractal del registro [1]. ei resultado de los calculos se muestra en la tabla 2. tabla 1 valores de resistividad, muestreados en el registro a intervalos de 1 m, el primero corresponde al techo de la secuencia y el ultimo al tope inferior 29 29 30 33 41 60 75 85 105 116 124 134 131 129 130 135 144 158 166 171 173 177 178 161 141 149 143 129 122 122 109 106 110 118 125 156 178 191 195 203 211 215 207 195 158 143 136 120 123 122 121 116 114 113 112 113 113 110 106 103 103 107 110 114 115 99 83 71 65 58 54 55 52 56 86 97 111 115 106 95 87 84 89 80 74 65 65 57 61 62 67 66 70 72 85 93 95 135 132 121 114 90 69 61 57 46 46 48 48 52 62 82 104 131 166 202 190 161 135 102 93 94 87 92 80 76 67 64 63 62 64 61 49 42 35 33 31 31 31 32 33 33 34 35 41 50 50 62 53 70 53 55 43 44 43 38 45 36 51 34 52 34 40 34 36 35 37 36 36 37 38 37 42 40 45 41 54 42 60 43 67 41 83 26 93 27 95 27 116 27 150 28 195 29 218 33 247 37 222 39 167 35 139 34 117 35 114 36 124 37 144 40 170 40 190 42 205 49 209 55 213 62 210 64 217 59 215 57 241 63 248 63 248 56 218 53 216 50 221 51 212 48 168 48 130 112 46 71 121 47 63 143 53 57 183 75 58 131 110 55 115 157 52 95 185 52 67 181 53 59 181 55 57 187 58 63 189 64 63 190 82 63 188 113 62 188 128 64 187 133 65 178 137 66 175 115 68 165 97 70 151 86 75 140 83 77 126 84 88 120 86 97 109 100 115 111 120 144 106 155 195 109 146 248 121 128 248 122 125 248 121 125 248 117 130 248 117 108 248 150 109 248 169 116 145 162 121 98 158 118 78 144 121 93 140 94 126 75 132 63 126 57 118 57 111 60 108 56 121 48 127 41 113 41 107 43 106 43 102 43 93 • techo tope superior, •• base tope inferior 5. discusion de resultados un analisis de regresi6n lineal efectuado entre el 10g(r1s) y el log(t) arroj6 los siguientes resultados: se hall6 que el exponente de hurst era de 0.854 con un coeficiente de correlaci6n de 0.997, indicando que la dimensi6n fractal del registro electrico era de 1.146 (fig.1). este resultado coincide con los obtenidos por hewett (hewett at ai., 1986) en el calculo de la dimensi6n fractal por el rnstodo de anatsls de frecuencia, de la porosidad medida en un registro de pozo distinto (1.100 pies), y para el cual el exponente de hurst fue h = 0.855. los datos de la tabla 1 y la fig.1, muestran que la resistividad a 10 largo del registro fluctua fuertemente; sin embargo, la fig.2 indica un comportamiento persistente, reafirmado con el valor del exponente de hurst (h = 0.853), indiscutiblemente mayor que 0.5 (valor correspondiente a una serie de datos completamente aleatoria), 10 cual muestra una tendencia en el proceso que dio su origen. una raz6n por la cual se puede explicar porqus estas secuencias de datos geoffsicos se comportan de acuerdo a una estadfstica de hurst, es que una secuencia sedimentaria refleja el comportamiento del clima en tiempos geol6gicos lejanos y los sedimentos que formaron las rocas actuales fueron transportados por corrientes de agua que erodaron las rocas antiguas, y estos f1ujos se comportan adecuadamente con la estadfstica de hurst. se observa que un registro electrico como serie de datos geoffsicos presenta un comportamiento fractal, el cual puede relacionarse con el proceso que 10 gener6, tales como la erosion y el clima de epocas geol6gicas lejanas. la geometrfa fractal permite observar un orden en las formas naturales de los datos geoffsicos, que a primera vista no se observan. tabla 2 muestra los valores de r, s, (r/s), log(t), log(r/s) y los resultados de la regresi6n lineal entre log(r/s) y log(t) r s log(t) log(r1s)r1s 151.9000 31.49937 477.2500 47.96236 722.6500 46.79476 1283.075 43.84825 1750.500 44.36271 2537.938 47.25159 3493.441 47.65990 5496.175 59.98748 5755.375 59.61874 5927.324 56.07979 4.822319 9.950512 15.44297 29.26171 38.78257 53.71115 73.29937 91.62203 96.53634 105.6945 2.302585 2.995732 3.688879 4.382027 4.787492 5.075174 5.393628 5.634790 5.768321 5.940171 1.573255 2.297624 2.737154 3.376280 3.657971 3.983621 4.294552 4.517672 4.569919 4.660552 regression output: constant -0.35835 std err of y est 0.059437 r squared 0.99721 no. of observations 10 degrees of freedom 8 x coefficient(s) 0.0853886 std err of coef. 0.01597 12 monte.: fractale. y .erle. de datos geofi.ico. '"n .. -0 o -0. -0 c ::j '+-e a.. 200 400 (ohm-m) figura 1. registro electrico de pozo. cota superior: 100 m (profundidad) o resistividad o n o"loq. -l'"ci 2.0 3.0 4.0 5.0 6.0 log (periodo) figura 2. curva de regresi6n. datos de la tabla 2. y = 0.8538x • 0.358. coeficiente de determinaci6n = 0.9972 bibliografia burrough. p.a. (1981): fractal dimensi6n of landscapes and other enviromental data. nature 294 (5838). burrough. p.a. (1984): the application of fractals to geophysical phenomena: institute of mathematics and its applications 20 (3-4). feder, j. (1988): fractals. plenum press. new york. heinz ...o. peltgen & p.h. ritcher. (1987): the beauty of r-ractals. springer-verlag, new york. heinz, 0., peitgen et al. (1988): the science of fractal images. spnnger-verlag, new york. hewett, l.a. (1986): fractal distributions of reservoirs hetereoqenity and their influence on fluid transport. technical conference of the society of petroleum engineers. new orleans. hurst et al. (1965): long-term storage. an experimental sfudy. constable, london. jones, j.g. et al. (1989): fractal properties of computer-generated and natural geophysical data. computers & geosciences 15 (2):227-235. mahdelbo~ b. (1982): fractals: form, chance and dimension. w.h. freeman, san francisco. unwin, d. (1989): fractals and the geosciences. computers & geosciences 15 (2). pergamon pres, oxford. statistical embedded markov chain processes were used to analyze facies transitions and to determine the stacking pattern of the lithofacies of the guaduas formation. twelve lithofacies were found and characterized based on lithology and sedimentary structures in four stratigraphic sections. the findings were compared with a previous assemblage of lithofacies, interpretations of sedimentary environments, and depositional systems. as a result, four depositional systems were established. through the statistical analyses of facies transitions it was found that tidal facies are prevalent in the socota section, more specifically in the upper part. whereas in the sogamoso, umbita and peñas de sutatausa sections, fluvial facies are prevalent in the upper part of the sections, and follow a regressive sequence with more continental deposits around the upper part of the sections. for each of these sections the markov chain transition matrices illustrate a strong interaction between tidal facies and fluvial facies, especially in the peñas de sutatausa matrix, where facies 6, made up of tidal deposits, appears several times. from the facies model and markov chain analyses, it is evident that the guaduas formation is a cyclic sequence in which the markov facies repetitions are consistent with the lithofacies analyses conducted in previous stratigraphic studies. the results reveal that the markov chain statistical process can be used to predict stratigraphy in order to correlate contiguous geologically unexplored areas in the guaduas formation, where much work relating the correlation and the continuity of coal beds has yet to be done. se utilizaron los procesos estadísticos de las cadenas de markov para analizar las transiciones de facies y para determinar el patrón de apilamiento de las litofacies de la formación guaduas. se encontraron y caracterizaron doce litofacies en cuatro secciones estratigráficas según litología y estructuras sedimentarias. los resultados fueron comparados con un ensamblaje anterior de litofacies, interpretación de ambientes sedimentarios y sistemas deposicionales. como resultado, se establecieron cuatro sistemas deposicionales. a través de los análisis estadísticos de las transiciones de la facies se encontró que la facies de marea son frecuentes en la sección de socotá, más concretamente en la parte superior. mientras que en las secciones de sogamoso, umbita y peñas de sutatausa, las facies fluviales son frecuentes en la parte superior de las secciones y siguen una secuencia regresiva con depósitos continentales hacia la parte superior de las secciones. para cada una de estas secciones, las matrices de las cadenas de markov presentan una fuerte interacción entre facies mareales y facies fluviales, especialmente en la matriz de penas de sutatausa, donde la facie 6, compuesta de depósitos mareales, aparece varias veces. a partir de los análisis de facies y de las cadenas de markov, es evidente que la formación guaduas es una secuencia cíclica en la que las repeticiones de facies de markov son consistentes con el análisis de litofacies en estudios estratigráficos anteriores. los resultados revelan que el proceso estadístico de las cadenas de markov se puede utilizarse para predecir la estratigrafía y para correlacionar áreas contiguas y geológicamente inexploradas en la formación guaduas, donde aún hay mucho trabajo que hacer en la correlación y la continuidad de los yacimientos de carbón. earth sciences research journal earth sci. res. j. vol. 20, no. 3 ( september, 2016 ) : b1 b9 abstract resumen keywords: litofacies, lithofacies, reservoir environment, facies maps, guaduas, lithofacies maps. palabras clave: : litofacies, ambientes de depósito, mapas de facies, formación guaduas, mapas de litofacies. lithofacies cyclicity determination in the guaduas formation (colombia) using markov chains determinacion de la ciclicidad de las facies en la formacion guaduas (colombia) usando las cadenas de markov record manuscript received: 15/07/2014 accepted for publication: 20/09/2016 how to cite item mariño, j., & amaya, e. (2016). lithofacies cyclicity determination in the guaduas formation (colombia) using markov chains. earth sciences research journal, 20(3), b1-b9. doi:http://dx.doi.org/10.15446/esrj.v20n3.44429 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n3.44429 mariño m. jorge e.1, amaya erika2 1 universidad pedagógica y tecnológica de colombiaescuela de ingeniería geológica; jorge.marino@uptc.edu.co 2 universidad pedagógica y tecnológica de colombiaescuela de ingeniería geológica; amayaer@gmail.com s t r a t ig r a p h y mariño m. jorge e. and amaya erikab2 introduction the guaduas formation is the main source of thermal and coking coal for central colombia and is located on the eastern cordillera of the andes mountain range. according to certain authors, the cyclic patterns found in this formation allow the use of statistical methods. sarmiento (1994) identified the repetition of transitional environments during the transition between the cretaceous sea invasion and the early tertiary with its transitional deposits, and in 2010 amaya et al. (2010) found a prevalent muddy composition with some key sandstones (guias) and coal beds, in which such coal beds are interbedded with mudstone and sandstone in a pattern beginning with tidal deposits followed by coastal deposits (such as lagoons) and lastly fluvial deposits. markov chains statistical analyses have been used by several authors to differentiate geological processes and construct facies models based on transition frequency (harper, 1984) and can also be used to determine which lithofacies are generally found above or below other lithofacies. each lithofacies has an assigned number according to its facies description (miall, 1992). using markov chain analyses, mariño and morris (1996) found 37 transitions between eolian and marginal facies, which helped in the development of the depositional model. and additionally found that most of these transitions follow the depositional model, although some random facies could indicate sudden events such as storms. the purpose of this research is to use markov chains to confirm the patterns of repetition found in the guaduas formation by other authors using standard lithofacies analyses (sarmiento, 1994; amaya, 2009; amaya et al., 2010). amaya et al. (2010) described and reviewed 10 stratigraphic sections in outcrops and cores, and found 12 lithofacies that helped to establish four depositional systems (figure 1). four of such stratigraphic sections were chosen with the intent of discovering a pattern by using the statistical method known as the markov chain, which orders successions of mutually exclusive states (i.e. facies) wherever a change of state occurs to find the significance of the succession. a statistical pattern is suspected in this formation as a result of the cretaceous cyclicity that has been well documented in colombia (etayo-serna et al, 1976). these patterns could help to understand the dynamics that allow the sediment accumulation in this part of the eastern cordillera by the end of cretaceous and beginning of paleocene, and can also be used to improve the correlation and learn more about the continuity of the coal beds in the guaduas formation. figure1. location of the stratigraphic sections in central colombia which cover parts of the boyacá and cundinamarca provinces. cyclicity in sedimentary rocks and markov chains long ago people began to evidence that the sea once covered areas that are currently dry land. episodic sea invasions were found to be periodical and important through the study of sedimentary successions and as a result, cyclic patterns were found. in any stratigraphic succession one can see changes in lithologies, called facies, and several statistical methods have been used to differentiate geologic processes and construct facies models on the basis of transition frequency (harper, 1984). in this study, the term facies is called lithofacies because it represents a rocky unit defined by its lithologic characteristics, which include composition, grain size, and sedimentary structure (miall, 1990). each lithofacies represents an individual depositional event, and is grouped according to characteristics of a depositional environment based on facies association (miall, 1990; amaya, 2009). the cyclic patterns, represented by cyclical lithofacies, go beyond relative and periodical changes in sea level, and include eustatic or global changes in sea level, non eustatic changes in sea level related to local and regional events, subsidence, compaction, climate, tectonics amongst other factors (miall, 1990). the statistical method best suited for the analysis of facies transitions data in stratigraphic sections is the markov chain transition matrix. markov chain analyses can be used to determine which lithofacies are usually found above and below other lithofacies, and thus is one of the methods used to differentiate geologic processes and construct facies models based on transition frequencies, some of which can be used as patterns to predict events or facies related to minerals resources and so on (harper, 1984; walker and james, 1990). a markov chain transition matrix groups lithofacies into a two dimensional array which tabulates the number of times that all possible vertical lithofacies transitions occur in a given stratigraphic succession. the analysis becomes more meaningful with increasing numbers of transitions. each lithofacies has an assigned number or code according to the facies description (miall, 1990). there are two ways to make a transition count matrix: transitions may be recorded at each new bed when lithofacies change in character, regardless of bed thickness, or transitions may be constructed by sampling the section at fixed vertical intervals. in this study, the first method, or “embebed markov chain analysis,” was chosen because it emphasizes changes and focuses on depositional processes. repeating lithofacies are measured as one, hence the row of zeros that spans from the top-left to the bottom-right corner of the transition count matrix (table 1). table 1. example of a 6x6 embedded transition matrix in which the transition numbers in the diagonal are all zero. geologic setting of the guaduas formation the guaduas formation is part of the eastern cordillera basin in east central colombia and has a muddy composition with some sandy units and coal beds; such coal beds are of economic importance in colombia due to their high rank, with some of them being coking coals (luna et al., 2004). the formation presents a systematic thinning to the north and to the east. in the southwest, its thickness can reach 1100 m (checua-lenguazaque lithofacies cyclicity determination in the guaduas formation (colombia) using markov chains b3 syncline), to the north 250-400 m, and to the east 80 m (sarmiento, 1994; cooper et al., 1995; amaya et al., 2010: duarte y mariño, 2000). the guadalupe group, chipaque formation, and une formation underlie the guaduas formation, and the cacho or socha sandstone overlies the guaduas. the northern part of the eastern cordillera basin was bisected by the santander high (the santander and floresta massives) until the emergence of the cretaceous, which allowed for the development of two separated sub-basins: the cocuy and the magdalena-tablazo. this barrier controlled deposition during the lower cretaceous, with the accumulation of marine shallow facies mainly (fabre, 1983), to the south, these sub-basins merged as the cundinamarca sub-basin, where the cretaceous sequence reaches maximum thickness (sarmiento-rojas et al., 2006). the sub-basins tablazo and cocuy combined as one basin during hauterivian due to the flooding of the santander-floresta high (fabre, 1983). during the aptian, a major transgression, followed by a relative sea-level rise, caused the flooding of the current area of the eastern cordillera, including the cundinamarca sub-basin. by that time, the santander high was not a barrier that would prevent the sediments movement (cooper et a.l, 1995). in the albiancenomanian, a relative fall in the base level favored the progradation of deltaic and littoral sands une formation (fabre, 1983; sarmiento-rojas et al., 2006). during the cenomanian-turonian the eustatic level reached its maximum and allowed for the deposition of the chipaque formation (sarmiento rojas et al., 2006). by the santonian, campanian, maastrichtian and paleocene, a regression and general progradation lead to the accumulation of littoral and transitional facies during the deposition of the guadalupe group; it was followed by coastal and alluvial flats and coal facies during the deposition of the guaduas formation (sarmiento, 1994). finally, the entire sequence was covered by the fluvial deposits of the cacho-socha sandstone. the guaduas formation was deposited throughout the upper cretaceous and lower tertiary during the regression of the cretaceous seas. superimposed upon this big-tendency scale there are several smaller transgressive-regressive cycles that allow the repetition of the lithofacies and made the formation suitable for statistical analyses. the subsidence of the basin was fast, and the predominance of shallow muddy marine facies also suggests a quick deposition. (cooper et al., 1995). the 12 lithofacies found in the guaduas formation can be grouped in 4 lithofacies associations (see tables 2 and 3). lithofacies association a consists of interbeddings and interdigitation of sandstones, siltstones, claystones, and at times thin coal beds, which are interpreted as lagoon deposits based on the association, lateral continuity, and organic content of the lithofacies. lithofacies association b is characterized by the interbedding of sandstones, and claystones with plant debris and coal, and is interpreted as being formed by tidal flats based upon heterogeneous stratification, and bidirectional cross-stratification. lithofacies association c consists of claystones, coal, and certain sandstones. this association represents a great percentage of the stratigraphic record and is classified as an alluvial flooding flat. lithofacies association d consists mainly of sandstones and is related to meandering channels based on sedimentary structures, especially big scale cross-stratification (see table 3) (amaya et al., 2010). data and methods the purpose of this study is to use markov chain analysis to verify the facies stacking pattern discovered in the guaduas formation using conventional lithostratigraphic facies analysis (amaya et al., 2010; sarmiento 1994). figure 1 depicts the places where markov chain analysis was performed in the four stratigraphic sections of socota, sogamoso, umbita, and peñas de sutatausa. these sections are well distributed and represent guaduas formation stratigraphy in the provinces of boyacá and cundinamarca. facies determination and distribution was conducted in a five step process: (1) stratigraphic data recollection, (2) facies definition and description (3) depositional environment determination, (4) markovian analysis of lithofacies, and (5) stratigraphic correlation and analysis. each facies description was based on grain size and rock type (folk, 1974), sedimentary structures, bed thickness, vertical and lateral change relationships based upon panoramic sections (miall, 1990), as well as color and texture. after describing the stratigraphic sections of the guaduas formation in the studied area, 12 lithofacies were found: 1) sandstone with bidirectional cross-stratification, 2) sandstone with irregular wavy bedding, 3) rhythmic successions of sandstones and claystones, 4) sandstones with heterolithic bedding, 5) claystones with sandy lenses, 6) claystones with thin lamination, and plant debris, and coal, 7) claystones and siliceous limestones with paleosols, 8) sandstones with through-cross stratification, 9) sandstones with planar cross-stratification, 10) sandstone with planar stratification, 11) sandstone with climbing cross-stratification, and 12) massive sandstones (table 2). table 2. lithofacies number and lithofacies according to the facies analysis, the repetition of different attributes was identified throughout the sections and along the basin. this association of lithologic attributes is presented in table 3, which shows a prograding sequence that characterizes the transition between transitional and continental deposits. the lithofacies association allowed the determination of the depositional environments and the depositional systems. the facies model, by which the depositional environments were interpreted, indicates that tidal flat deposits were followed by coastal lagoons and topped by a fluvial system with alluvial flats interbedding with meandering channels. description and interpretation of depositional environments was supported by research done by reineck y singh (1980), walker y james (1990), boggs (1987), kvale y archer (1990), miall (1990), walker and james (1990), fichter (1993), choi and dalrymple (2004), and mariño and morrison (1996). for the construction of the markov chains, each section was labeled with facies number defined during the facies study (table 2); this procedure eased the transition counting between one lithofacies and an adjacent one (above or below). transitions were recorded based on the changes in character of lithofacies regardless of bed thickness. through this process the transition between the same types of facies was not taken into account, this is why the elements of the main diagonal in each matrix are all zero, which follows the method known as “embedded markov chain analysis” (ethier, 1975). these transitions were tabulated in a two-dimensional arrangement where the columns represent the overlying lithofacies while the rows represent underlying ones (table 1). markov chains are based on two considerations, the first is the supposition that lithology at any point n depends upon the lithology of the proceeding point n-1. this supposition is possible due to the fact that the guaduas formation apparently behaves as a cyclic sequence, in which the same lithology is observed at different points in a stratigraphic section. the second is to consider a transition as a change between two different lithologies where the transition in the same lithology is not taken into account, thus the elements of the main diagonal in the matrix are all zero (table 1). mariño m. jorge e. and amaya erikab4 results the facies analysis using markov chains in the four stratigraphic sections was based upon that carried out by amaya (2009) and amaya et al. (2010), which determined 12 repeating lithofacies throughout the sections (table 2). these facies were associated in order to establish a facies model with depositional environments and depositional systems for the guaduas formation. four depositional systems were established, from coastal lagoons, passing through tidal flats, to fluvial systems with meandering channels interbedded with alluvial flooding flats (table 3). the change between transitional and continental environments seems to be oscillatory, which reflects sea level variations for that time. the markov chain analyses were used to test the stacking pattern found during the previous facies study. the stratigraphic sections and the resulting matrices of socota, sogamoso, umbita, and peñas de sutatausa are as follows: socota stratigraphic section. figure 2 presents the stratigraphic section and the markov chain matrix for the socota area. in this matrix, the presence of lithofacies 6 (claystones with thin lamination, plant debris, and coal) stands out, as it appeared 8 times in the section. facies 6 is followed 4 times by lithofacies 1 (sandstones with bidirectional crosssection), and 3 times by lithofacies 5 (claystones with sandy lenses). sogamoso stratigraphic section. from the matrix completed for the sogamoso stratigraphic section (figure 3), lithofacies 6 (claystones with thin lamination, plant debris, and coal) is the most abundant, appearing 6 times, 4 of which lithofacies 8 is underlain by lithofacies 6, and 3 of which by lithofacies 12 (massive sandstones). lithofacies 5 was found below lithofacies 12 in three instances. lithofacies 12 was found 4 times below lithofacies 6. umbita stratigraphic section. from figure 4 in the umbita’s section, it is clear that lithofacies 6 (claystones with thin lamination, plant debris, and coal) is the most abundant and usually relates with facies 1 (sandstone with bidirectional cross-stratification) and facies 8 (sandstones with through-cross stratification). peñas de sutatausa stratigraphic section. figure 5 presents the stratigraphic section and markov chain matrix for the peñas de sutatausa area. from the matrix, it is concluded that the most related facies are 6 (claystones with thin lamination and plant debris and coal), 1 (sandstone with bidirectional cross-stratification), 5 (claystones with sandy lenses), and 8 (sandstones with through-cross stratification). lithofacies 6 and 8 are closely related and follow each other 13 times (below and above). table 3. prograding lithofacies model for the guaduas formation indicating continentalization of the basin. the lithofacies association result in a deposition environment that is related to a depositional system (amaya et al. (2010). lithofacies cyclicity determination in the guaduas formation (colombia) using markov chains b5 figure 2. stratigraphic section of the guaduas formation in socota with lithofacies numbers and markov chain transition matrix. mariño m. jorge e. and amaya erikab6 figure 3. stratigraphic section of the guaduas formation in sogamoso with lithofacies numbers, and markov chain transition matrix. lithofacies cyclicity determination in the guaduas formation (colombia) using markov chains b7 figure 4. stratigraphic section of the guaduas formation in umbita with lithofacies numbers and markov chain transition matrix. discussion and conclusions from the highlighted areas or more frequent transitions in each of the matrices, certain conclusions can be made: 1) for the socota area, the presence of lithofacies 6 stands out due to its repetition (claystones with thin lamination, plant debris, and coal), followed by lithofacies 1 and 5. the transitions of lithofacies 6 to other lithofacies are tenfold in this section. from this matrix, the presence of muddy areas with coal and the strong relationship between facies 1 (sandstone with bidirectional cross-stratification) and 6 (claystones with thin lamination, plant debris, and coal) is evident, and indicates the prevalence of transitional facies to the top, and a lateral relationship between fluvial and tidal flats, supporting the conclusions of the previous model in which the coal in that area was originated in tidal swamps (amaya et al, 2010; sarmiento, 1994). the socota area is the only section where the transitional facies seem to be to the top of a section with tidal deposits at the base and top of the section, and alluvial facies in the middle. 2) from the sogamoso matrix, it is evident that lithofacies 6 (claystones with thin lamination and plant debris and coal) is the most abundant and is most related to lithofacies 8 (sandstones with through-cross stratification) and 12 (massive sandstones). this indicates the interbedding of clays and coal from swamps (which are part of an alluvial flat), with sandy units that represent the fluvial meandering channels. the prevailing lithofacies analysis in the sogamoso section indicates continental dominance towards the top of the section, unlike that of the north-east section (socota section) where tidal and fluvial environments prevailed. this could be explained by the progressive thinning of the formation to the north and to the east (amaya et al, 2010). 3) for the umbita area, the strongest associations are between the tidal facies (facies 6, 5 and 1), which could indicate that the coal beds in this area are more related to tidal flats and have some interaction with fluvial meandering channels, lateral bars, and fluvial channels (8 and 9 lithofacies). this follows the conclusions of previous studies that have shown a balance between tidal and fluvial environments, with the fluvial deposits prevailing to the top of sections (amaya et al, 2010). mariño m. jorge e. and amaya erikab8 figure 5. stratigraphic section of the guaduas formation in sogamoso with lithofacies numbers, and markov chain transition matrix. 4) in the peñas de sutatausa area two associations are important: the first one relates tidal lithofacies (1-6), and the second associates tidal and fluvial facies (6 and 8), supporting the facts that in this area, fluvial, tidal, and fluvial with tidal facies influence were the prevailing lithofacies during the coal deposition. the fluvial influence is stronger to the top of the peñas section (sarmiento, 1994; amaya, 2009). in peñas, the fluvial influence is stronger than in the umbita area. from the facies model and markov chain analyses, it is evident that the guaduas formation is a cyclic sequence in which the markov facies repetitions are consistent with the lithofacies analyses done in the stratigraphic sections. the statistical facies analyses proved that the tidal facies prevails in the socota section, especially to the upper part; the markov chain transition matrix found little interaction of the tidal facies with the fluvial facies, indicating fewer fluvial facies in that section (figure 2). note that in figure 2 the facies repetitions are between tidal lithofacies 1 and 6, supporting the results of previous facies analyses (amaya et al., 2010). on the contrary, the stratigraphic correlations of the sogamoso, umbita and peñas de sutatausa sections revealed fluvial facies prevailing to the upper part of the sections, following a regression sequence with more continental deposits toward the upper part of the sections (table 2). in these latter sections, the markov chain transition matrix illustrates a strong repetition of fluvial facies following tidal facies, indicating strong interaction between them, as is evident in the peñas de sutatausa´s matrix where facies 6 related to tidal deposits repeats 13 times with facies 8 of fluvial origin (figure 5). lithofacies cyclicity determination in the guaduas formation (colombia) using markov chains b9 the finding of transitional tidal facies located most often to the bottom, followed by fluvial flats and fluvial channel deposits shows a shallowing upward sequence with continental deposits more frequent near the top that was developing at the upper cretaceous and early paleocene epoch as a result of the regression forced by the andean orogeny (cooper et al., 1995). however, in each transition, smaller cycles are present, allowing the cyclic deposition of different lithologies like key sandstones (guias) and 12 to 14 coal beds. this cyclicity allowed the deposition of a great deal of coal, especially in the south, making of this area one of the best coal sources in colombia (mainly coking coals) (luna et al., 2004). not only the vertical changes are evident in this area, but also the horizontal changes, due to the thinning of the stratigraphic succession to the north and east of the study area. the northward thinning of the guaduas formation may be explained from a smaller subsidence to the north (fabre, 1983, amaya et al, 2010). the thinning may also be explained by erosion (sarmiento, 1994). these hypotheses would explain why the guaduas formation is thicker in the cundinamarca province and has two coal bed members. on the contrary, to the north (boyacá province), the formation is thinner and has just one coal member, as the upper one possibly eroded. this along with erosional unconformity would better explain the thinning of the guaduas formation to the north (sarmiento, 1994; duarte y mariño, 2000). the discovery of more continental facies found to the top by use of markov chains is in agreement with facies maps, which indicate an upward shallowing of the environments with an increased presence of fluvial facies to the top, especially in the south (amaya et al., 2010; amaya, 2009). initially the sea regression-transgression had a sw-ne tendency, which later became more e-w. these changes could be related to the reactivation and stratigraphic control of big trust faults like soapaga that have caused differential subsidence (krammer, 1999; duarte y mariño, 2000). from the facies association in the guaduas formation, a progressive shallowing in the environments due to fluvial facies dominance is evident near the top of the section, with three main west to east transgressive episodes with tidal and fluvial environments prevailing to the top and ne of the basin, which could be explained by the progressive thinning of the formation to the north and east as a result of the sea receding to the west. this follows previous findings that used conventional facies analyses (amaya et al, 2010). these results demonstrate that the markov chain statistical process can be used as a quantitative tool to distill lithofacies transitions and define lithofacies trends with precision, as well as to predict stratigraphy in order to correlate contiguous unexplored geologic areas in the guaduas formation where much work on correlation and coal bed continuity is still to be done. acknowledgments the authors are grateful for, the financial support from the smithsonian tropical research institute and to the research office (din) of the universidad pedagógica y tecnológica de colombia (sgi 978), the technical support from: kyunk choi, cedric carpentier, angela leguizamón, and josh reed; the support and information during data collection and access to mines and wells from the companies, carbones de los andes, carbones del caribe, coprocarbon, minergéticos. thanks to sandra pedraza for her help in drawing figures, and matthew mcgee and cynthia marino for reviewing the manuscript. references amaya, e. (2009). litofacies y ambiente de acumulación de la formación guaduas (maastrichtiano tardío – paleoceno). proyecto de grado. uptc sogamoso – ingeniería geológica. biblioteca uptc. amaya, e., mariño, j. & jaramillo, c. (2010). litofacies y ambientes de acumulación de la formación guaduas en la parte central de la cordillera oriental – implicaciones paleo geográficas. boletín de geología, 32(1). boggs, s. (1987). principles of sedimentology and stratigraphy. columbus: merrill publishing co., 784 p. choi k. s., & dalrymple, r. (2004). recurring tide-dominated sedimentation in kyonggi bay. marine geology, 81-96. cooper, m., addison, f., alvarez, r., coral, m., graham, h., hayward, b., howe, s., martinez, j., naar, j., peñas, r., pulham, a. j., & taborda, a. (1995). basin development and tectonic history of the llanos basin, eastern cordillera, and middle magdalena valley, colombia. a.a.p.g bulletin, 1421-1443. duarte, c. & mariño, j. (2000). utilización de las areniscas guías en la búsqueda y exploración de mantos de carbón. caso formación guaduas – sogamoso (boyacá). memorias v congreso nacional de ciencia y tecnología del carbón. valledupar, cesar. etayo-serna, f., renzoni, g., & barrero, d. (1976). contornos sucesivos del mar cretácico en colombia. in: primer congreso colombiano de geología, mem., bogotá, pp. 217–252. ethier, v. (1975). application of markov analysis to the banff formation (mississipian). journal of mathematical geology, 47-65. fabre, a. (1983). la subsidencia de la cuenca del cocuy (cordillera oriental colombiana) durante el cretaceo y el terciario inferior. geología norandina, 8. fichter, l. p. (1993). ancient environments and the interpretation of geologic history. ontario, canada: macmillan publishing company. folk, r. l. (1974). petrology of sedimentary rocks. austin: hemphill, 182 p. harper, c. w. (1984). improved methods of facies sequence analysis, in: r.g. walker (editor) facies models, second edition. geoscience canada, reprint series 1, 11-14. krammer, a., (1999). observaciones acerca de un origen transpresivo de la cordillera oriental. geologia colombiana, 24: 29-53. kvale, e., & archer, a (1990). tidal deposits associated with low-sulfur coals, brazil fm. (lower pennsylvanian), indiana. journal of sedimentary petrology, 563-574 luna, l., rodríguez, e., sánchez, c., renzoni, g., hernández, h., maldonado, j., alba, c.& sanabria, a. (2004). el carbón colombiano recursos, reservas y calidad. publicaciones geológicas especiales. ingeominas, bogotá. mariño, j. & morris, t. (1996). erg margin and marginal marine facies analysis of the entrada sandstone, utah. museum of northern arizona special publication, 483-496. miall, a. (1992). alluvial deposits. in: walker, r. facies models: response to sea level change (pp. 119-142). geological association of canada. miall, a. d. (1990). principles of sedimentary basin analysis. new york: springer -verlag, 483 p. reineck, h., & singh, i. (1980). depositional sedimentary environments. springer, berlin, 551 p. sarmiento, g. (1994). estratigrafía y medios de depósito de la formación guaduas. boletín geológico, 32 (1-3). ingeominas, bogota. sarmiento-rojas, l., van wess, j., & cloetingh, s. (2006). mesozoic transtensional basin history of the eastern cordillera. south american earth sciences, 21, 383411. walker, r., & james, w. (1990). facies models: response to sea level change. canada: geological association of canada. 530p. search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 nota editorial con un gran esfuerzo conjunto de investigadores nacionales se ofrece este nuevo numero de geofisica colombiana, con la esperanza de que satisfaga el interes de la comunidad geocientffica y con la firrne convicci6n que redundara en un desarrollo prolffico de trabajos en la geoffsica y en su aplicaci6n a la soluci6n de los problemas que la industria y el pais demandan. estamos alentando a la nueva generaci6n de cientificos de la tierra a mostrar sus trabajos y producciones, quienes de una manera tesonera y patri6tica ie siguen apostando aeste pais que tanto necesita de sus nuevas generaciones y del compromiso de las mismas en su destino hist6rico. ese compromiso manifiesto de las nuevas generaciones se evidencia cuando en esta edici6n se plasman los trabajos de dos egresados del programa de magister en geoffsica de la universidad nacional de colombia. son ellos pedro vega y felix montano, quienes hacen su contribuci6n en instrumentaci6n sismol6gica en tiempo real y al analisis de parametres geoffsicos en el piedemonte, respectivamente. de la misma manera, el estudio de variaci6n solar acometido por rosa ladino, yaqueline figueredo y emna valdez expresa la panicipaci6n de nuevas geocientfficas. grato es tarnbien reconocer la intervenci6n en esta publicaci6n de tres nuevos docentes vinculados al posgrado en geofisica: en primer lugar el profesor luis castillo hace un aporte a la soluci6n del problema de migraci6n en medios con gradientes de velocidad; la profesora sandra sanchez presenta su estudio de sondeos electromagneticos y su aplicaci6n en la soluci6n del problema inverso; finalmente, los diversos tratados sobre transmisi6n y procesarniento de seiiales en sismologia puestos a disposici6n por el profesor carlos vargas, un grupo de colaboradores de distintas instituciones, carla perez, augusto capera, julio cuenca de la unam, j. florez y j. g6mez. otras entidades muestran sus nuevos resultados a traves de investigadores como elkin salcedo y maria bermudez en lngeorninas, hugo monsalve en la universidad del quindio, jaime le6n y sergio rinc6n en la universidad nacional, sede manizales. nuevamente registramos la continua participacion en la construcci6n de conocimiento de instituciones como ingeominas, universidad del quindio, universidad aut6noma de mexico, universidad del valle, universidad distrital «francisco jose de caldas» y universidad nacional de colombia a traves de sus distintos prograrnas y sedes. esta resefia es un reconocirniento al esfuerzo que tales entidades asumen diariamente y un aliciente al mismo, por 10que el estado debe revertir en ellas recursos que garanticen la continuaci6n perrnanente de esa tarea. son estos resultados obtenidos por estos nuevos autores de nuestra revista 10 que nos hace mantener la fe y la confianza en un futuro prornetedor para las generaciones venideras, y 10que nos sostiene con el indeclinable animo de participar en esta tarea mancomunada de la producci6n y continuidad de esta, su revista geoftsica colombiana. finalmente, un agradecimiento especial a la universidad nacional de colombia por su apoyo constante a esta publicaci6n. 77 the present research work examines three aspects of great importance for the future supply of oil and gas in colombia. 1) the geographic and geological distribution of hydrocarbon resources for use in estimations of the contribution of different field sizes to the total supply of oil and gas in colombia. 2) the current state of the hydrocarbon reserves in colombia and factors that influence their increase or decrease; statistical tools are used to calculate the volume of hydrocarbons that may be found in the future. 3) the observed decline of hydrocarbon production in colombia, which is used to estimate the amount of resources that may be extracted in the future once peak production has been reached. based on these three aspects, we constructed different scenarios to forecast the potential production of oil and gas in colombia. in addition, we calculated the amount of hydrocarbon reserves in these scenarios. el presente trabajo resume tres aspectos considerados de gran importancia para el futuro suministro de petróleo y gas en colombia. estos son: 1) la distribución geográfica y geológica de los recursos hidrocarburiferos como herramienta para conocer la contribución de los diferentes tamaños de campo a la producción total de hidrocarburos en colombia; 2) el estado actual de las reservas de hidrocarburos en colombia y los factores que inciden en el incremento o disminución de las reservas, que son dos asuntos temas que inciden en el futuro suministro de estos recursos. al respecto, por medio de herramientas estadísticas se calcula el volumen de hidrocarburo que puede ser encontrado en el futuro. 3) las tasas de declinación en la producción de hidrocarburos observados observadas en colombia, permiten modelar los recursos que pueden ser recuperados en el futuro una vez alcanzados los picos de producción. bajo esta hipótesis, se construyen diferentes escenarios que resumen pronostican la posible producción de petróleo y gas en colombia para los próximos años, además se calculan las reservas de hidrocarburos para los mismos escenarios. earth sciences research journal eart sci. res. j. vol. 18, no. 1 (june, 2014): 77 83 abstract resumen key words: ultimately recoverable resources, oil production, gas production, production peak, oil reserves, gas reserves, decline rates. palabras clave: rfr, producción de petróleo, producción de gas, pico de producción, reservas de petróleo, reservas de gas, tasas de declinación. hydrocarbon production scenarios in colombia. review of field sizes, hydrocarbon reserves and expectations of conventional and unconventional resources. beltrán rivas julian daniel y vargas jiménez carlos alberto universidad nacional de colombia mail: jdbeltranr@unal.edu.co cvargas@ciencias.unal.edu.co record manuscript received: 30/09/2013 accepted for publication: 31/05/2014 hydrocarbon potential 78 beltrán rivas julian daniel & vargas jiménez carlos alberto introduction hydrocarbons currently represent the primary source of energy in the world and are expected to remain the primary energy source despite discoveries of alternative energy sources. therefore, the supply of hydrocarbons is a topic of great political, economic and academic interest. hydrocarbon reserves are determined by a variety of geological, technological, economic and political factors that create uncertainty for the future of hydrocarbon supply and demand. the goal of this work is to present the current state of three aspects that will influence the supply of hydrocarbons in colombia in the coming years. 1 the size distribution of hydrocarbon fields, which is important for determining the distribution of hydrocarbon resources in colombia and studying the relative importance of large and small fields for the production of oil and gas in the country. the state of hydrocarbon reserves, which is a fundamental topic for the prediction of oil and gas supplies. an estimation of the size of hydrocarbon reserves in colombia must consider all factors that may influence increases or decreases of these reserves. 2 an analysis of the decline in hydrocarbon production. the methodology developed in this article allowed us to estimate oil and gas production for specific fields or regions over time. 2. field size distribution the methods for evaluating recoverable resources and hydrocarbon supplies develop possible future scenarios that are usually based on the size distribution of oil and gas fields for a given region. the field sizes to which we refer in this article correspond to the ultimately recoverable resources (urr) of each field. urr = accumulated production + recoverable reserves. the calculation of oil and gas urr is conducted by adding the accumulated production and recoverable reserves. most urr in a region are concentrated in a small number of large fields and are typically discovered in the early stages of exploration. subsequent discoveries usually correspond to smaller fields that are more difficult to find. table 1 shows the top 5 largest oil fields in the world up to the year 2008 and their year of discovery, which indicates their advanced productive age. only 1% of the oil fields represent 75% of the accumulated discovered resources in the world. table 1. top five largest oil fields in the world up to 2008 and their year of discovery. note their advanced productive age (ivanhoe and luckie, 1993). field year of discovery ghawar 1948 cantarell 1977 safaniyah 1951 rumaila 1953 greater burgan 1938 2.1 the importance of large fields in colombia the size distribution of oil and gas fields in colombia has been established based on the proposal by ivanhoe and luckie (1993) in which fields are grouped into ten categories according to their estimated urr (tables 2 and 3). table 2. size distribution of oil fields in colombia by urr. category estimated urr number in colombia (mmbbl) number in colombia 0 super giant 5.000 50.000 0 giant 500 5.000 5 very large 100 500 20 large 50 100 11 medium 25 -50 23 small 10 25 52 very small 1 – 10 131 minuscule 0.1 – 1 103 insignificant < 0.1 101 as shown in table 2, large fields dominate the supply of hydrocarbons in colombia2 (acp, 2012), figs. 4 and 5, which is consistent with global trends. the five fields with a urr larger than or equal to 500 million barrels (mmbbl) are only 1% of the total fields, but they represent 35% of all accumulated discoveries. the fields with urr between 100 mmbbl and 500 mmbbl correspond to 5% of all fields and represent 70% of the accumulated discoveries. table 3. size distribution of gas fields in colombia by urr (acp, 2012). category estimated urr (mmcf) number in colombia very large 1,000,000 (tera)-10,000,000 4 large 100,000 – 1,000,000 15 medium 10,000 – 100,000 50 small 1,000 – 10,000 74 very small 100 – 1,000 63 minuscule 10 – 100 54 insignificant < 10 42 for gas reserves, there are four fields classified as “very large” (table 3) that exceed a urr of one tera-cubic feet (tcf), which represent 1% of all fields and are responsible for 75% of the accumulated resources. during the development of this work, we primarily relied on demonstrated oil and gas reserves with the exception of a few fields for which the number of probable reserves was used. figure 1. estimated contribution of the giant and very large fields to the total colombian oil production for the last two decades. the definitions of giant and very large fields are presented in table 2. the giant and very large oil fields in colombia contribute on average 60% of the daily oil production in the country (fig. 1). regarding gas production, 88% of the gas delivered to the pipeline originates in large fields (fig. 5), which indicates the strong production control exerted by large fields in colombia. 0   100.000   200.000   300.000   400.000   19 90   19 95   20 00   20 05   20 10   m bb l   daily  oil  produc/on  (1990  -­‐  2012)     producción  total   super  grande   gigantes   79hydrocarbon production scenarios in colombia. review of field sizes, hydrocarbon reserves and expectations of conventional and unconventional resources. figure 2. estimated contribution from very large and large gas fields to the total colombian gas production for the last 25 years. the definition of very large and large fields is presented in table 3. 3. oil and gas reserves in colombia the increase in oil and gas reserves is a fundamental topic for the prediction of future hydrocarbon supplies. therefore, a review of the following concepts is necessary. reserves are the amounts of oil and/or gas whose extraction is considered to be technologically accessible and economically feasible under given conditions. the estimation of reserves is susceptible to change because of accumulated production, revisions to the geological scenario, improvements of extraction techniques, the introduction of new technologies, or the modification of social, political and economic conditions in a given region. considering these variables, the reserves are presented within different confidence levels, which are explained below (salager, 2005). • demonstrated reserves (1p) are the amounts of oil and/or gas that have been determined (by the analysis of geological, geophysical and engineering data) to be commercially recoverable from a given date, from known deposits and under well-defined economic conditions, operational methods and government regulations. in probabilistic terms, there must be a probability of at least 90% that the real extracted quantities will be equal to or superior than the estimates. • probable reserves (2p) are additional reserves whose extraction has been determined (by the analysis of geological and engineering data) to be less likely than that of 1p but more likely than that of possible reserves. in probabilistic terms, there must be at least a 50% probability that the real extracted quantities will be equal to or superior than the estimates. • possible reserves (3p) are reserves that are less likely to be extracted than 2p. there must be a probability of at least 10% that the real extracted quantities will be equal to or superior than the estimates. (anh, 2008). 3.1 sources of reserve increases or decreases the factors influencing the increase or decrease of hydrocarbon reserves can be grouped into three categories. geological factors: these factors represent an increase or decrease in the estimations of oil originally in place (ooip) and gas originally in place (goip) because of increased knowledge of the geology of a field or region. new geological knowledge may lead to the discovery of new reservoirs or extensions of known reservoirs and directly affects the volume of ooip. technological factors: these factors represent activities that increase the rate of extraction of ooip and goip, increasing the feasibility and economic viability of extractions from certain fields. two techniques that stand out are related to increases the effective porosity and the injection of pressurized fluids for the mobilization of hydrocarbons, which concentrate hydrocarbons in a given zone and facilitate their extraction (salager, 2005). definition factors: this term encompasses all legal definitions, economic and political factors that are independent of the hydrocarbons present at the site and extraction rates. these factors include changes in the classification schemes for the reserves, calculation of the reserves by different institutions, calculation methods and oil prices. figure 3. cumulative oil discoveries from 1921 to 2012, which indicate an increase during the 1980s and 1990s (vargas, 2012). ecopetrol s.a. (2005) the increase in oil cumulative discoveries (mmbbl) between the years 1980 and 2000 corresponds to the discovery of fields such as caño limón (1986), cusiana (1988), cupiagua (1993) and rubiales (1988) (fig. 6). the fields of cusiana, cupiagua, chuchupa and ballenas (1973) represent significant gas reserves in colombia, which also increased the cumulative discoveries in the country. 3.2 contribution of colombian oil and gas reserves to the world reserves the hydrocarbon industry in colombia represents close to 60% of all export revenue thus, it is of great importance for the country’s economy (dane, 2012). the columbian hydrocarbon reserves relative to the world’s reserves are shown in table 4. table 4. comparison between the world oil reserves (anh, 2009; bp, 2012), and colombian oil reserves (anh, 2008; acp, 2012) for the last three decades. decade world reserves (mmbbl) colombian reserves (mmbbl) 1980 683.4 0.554 1990 1027.5 1.991 2000 1257.9 1.972 2010 1622.0 2.058 2012 2000.0 2.377 in the last three decades, colombia’s contribution to the world’s oil reserves has been 0.17% on average. the maximum contribution occurred in the 1990s when colombia’s reserves were 0.25% of the total world oil reserves (table 4). the contribution by colombian gas reserves was lower, and over the last 25 years, colombia’s gas reserves have amounted to 0.11% on average of the world’s reserves. this percentage does not vary significantly for the considered time interval (table 5). table 5. comparison between the world’s gas reserves ( ) british petroleum (2012)and colombian gas reserves acp (2012) for the last 25 years. decade world reserves (tcf) colombian reserves (tcf) 1987 3687 3.426 1990 4438 3.577 2000 5447 6.235 2010 6926 5.405 2012 7815 6.460 0   200   400   600   800   1.000   1.200   1.400   19 87   19 92   19 97   20 02   20 07   20 12   m m pc   daily  gas  produc0on  (1987  -­‐  2012  )   total  produc5on   large  +  very  large   very  large   0   2.000   4.000   6.000   8.000   10.000   1. 90 0   1. 92 0   1. 94 0   1. 96 0   1. 98 0   2. 00 0   2. 02 0   m m bb l   año   cumula,ve  oil  discoveries     cumula/ve  discoveries  (mmbbl)   80 beltrán rivas julian daniel & vargas jiménez carlos alberto 3.3 estimation and forecast of the oil and gas reserves in colombia in this section, we discuss a fractal model used to estimate the oil and gas reserves that could potentially be found in colombia. with this model, we are able to forecast the hydrocarbon reserves that may be extracted based on data for the current reserves and production. in addition, this model allows us to project the amount of hydrocarbons that may be found under a given set of technological, political, social and environmental conditions (vargas, 2012) the fractal method assumes that the reserves of oil and gas fields are described by an ideal distribution, given by a parabolic or linear function at a log-log scale. this function is determined from a trend analysis that considers the field sizes and ranges to group a number of major fields in a given size [mmbbl and billions of cubic feet (bcf)]. under these conditions, the function that best fits the observed data for the size and field reserves suggests the total amount of resources that may be discovered. based on the urr distribution, it is possible to estimate the “amount of hydrocarbon to be discovered,” which corresponds to the total amount of resources still underground, defined by the fitting function minus the accumulated resources found to date, or the urr (fig. 4, vargas, 2012) thus, the reserves to be discovered in a region are estimated by the difference between the curve of the observed data and their theoretical fit (parabolic or linear). we used annual data on production and reserves from oil and gas fields to forecast the discovery of reserves using the fractal method. these data correspond mainly to fields located near the basins of the upper magdalena valley (umv) and middle magdalena valley (mmv), the eastern llanos and the caguán-putumayo, catatumbo and guajira offshore basins. this information was provided by companies operating in colombia as well as several institutions, including the colombian association of petroleum engineering (acipet), colombian petroleum association (acp), ecopetrol, and ministry of mines and energy. fields located at other basins from which lower densities of data are available were also included in this study. figure 4. fractal representation for undiscovered resources in a region. the blue line represents the distribution of fields with a given urr at a given time. the green line represents the data fit. the gray region is the difference between these two curves and represents the resources to be discovered (vargas, 2012). 1. 4 methodology for reserve calculation from the fractal method 1 the data were organized into different ranges, and each range corresponds to the number of fields with urr larger than or equal to a given number of interest. 2 the ranges are plotted on the horizontal axis, and their corresponding urr are plotted on the vertical axis (in mmbbl for oil and bcf for gas) at a loglog scale, thus providing the size distribution of the oil and gas fields in colombia. 3 the function that best fit the data was estimated. figure 5. fractal analysis of the undiscovered oil resources in colombia. the blue line represents the distribution of oil fields in colombia in 2012; the green line is the data fit; and the difference between the two curves corresponds to the undiscovered reserves. according to the production and reserve data from the basins mentioned above and additional data from other basins, it is estimated that under the current technological, political, social and environmental conditions, the total volume of oil to be discovered is 21.486 mmbbl (fig. 5). it is important to clarify that these values can be interpreted as new discoveries or reevaluations of currently known fields. the calculation of the gas volume to be discovered was conducted following the same methodology. according to the production and reserve data from the basins mentioned above and data from other basins, it is estimated that under the current technological, political, social and environmental conditions, the total volume of gas to be discovered is approximately 49,834 bcf (fig. 6). figure 6. the blue line represents the observed field distribution; the red line represents a parabolic fit to this data; and the green line represents a linear fit. 4. decline rates the production behavior of the fields over time is the determining factor for the trends of hydrocarbon supply. the hydrocarbon supply can be forecast from the decline rates observed in production. in this section, we examine the decline in hydrocarbon production, and we use this information as a tool for the prediction of future production behavior in colombia. 4.1 analysis of the decline in the hydrocarbon production decline rate: by “decline”, we are referring to the decrease in the production capabilities of a field. the decline rate represents the percentile decrease in production at a given time with respect to an earlier time period (fig. 8). for colombia, a decline rate of 9% has been proposed for oil, whereas the corresponding number for gas is 15% (anh, 2009) the estimated productive life of a field and the expected production during that time are two questions that can be answered in terms of volumetric calculations. however, the necessary information for an accurate prediction may or may not be available for a given field. therefore, generating empirical models for the extrapolation of variables, such as the production levels of a 1   10   100   1000   10000   100000   1   10   100   1000   rf r   (m m bb l)   rango   observed  data   data  fit   1   10   100   1000   10000   100000   1   10   100   1000   fr f   (b pc )   rango   observed  data   parabolic  fit   linear  fit   81hydrocarbon production scenarios in colombia. review of field sizes, hydrocarbon reserves and expectations of conventional and unconventional resources. given resource in time, can provide answers to these questions. one of the simplest and most readily available characteristics of a well is its production rate. forecasts of production can be extrapolated from a plot of known production vs. time by means of an empirical equation. from this extrapolation, it is possible to predict the hydrocarbon production at a given time in the future. the intersection of the extrapolated curve and economic limit indicates the remaining productive life of a field or region; in addition, it provides information on the volume of oil that may be extracted in the future. in other words, an equation can be generated that predicts the future hydrocarbon production trends of a region from known production data under the statistical principles of extrapolation. however, this empirical method may underestimate the values predicted by volumetric methods. the hydrocarbon production of a field may vary according to the geological characteristics, region and technology employed for its exploitation. nonetheless, the production cycle of a field or region can be described as a function of time, which is described below. when extraction begins, there is a rapid increase in production until peak production is reached. the peak production is then sustained for a given time period referred to as the plateau. the duration of this plateau is proportional to the field size and may be extended or reduced by the incorporation of new extraction technologies. at a certain point, the period of peak production comes to an end, and production rates begin to decline (fig. 7; höök, 2009). initially, the decline in production is very steep, but it gradually slows as time passes. figure 7. stylised production cycle of an oil field empirical equations that model hydrocarbon production within a field or region after their production peak show that production decreases geometrically at a constant rate (fig. 8). this statistical model assumes that the production rate at any given time is a fraction of its value at a previous instant. therefore, the production rate forms a geometric progression at regular intervals ( arps, 1944). a geometric progression includes a constant ratio between successive terms. (swokowsky, 1989) this also explains why the decrease in production during a given time interval is given by a fixed fraction or percentage of its value at the time immediately before. this model is used for the estimation of oil and gas production in colombia for the coming years. figure 8. production decline for different decline rates. at first, a large decline in production is observed, which tends to decrease with time. 4.2 empirical model for hydrocarbon production decline the analysis of decline curves is a method of predicting the future hydrocarbon production of a well, field or region based on production records. (arps, 1944) introduced empirical curves based on the following principal equation: , equation 1 where q(t) is the production of oil or gas at time t during the decline stage. the production peak is given by q(t0), where t0 is the year corresponding to peak production. the decline rate is given by λ. 4.3 model for the estimation of future oil and gas production in colombia the model described above is applicable for the stage after the production peak. therefore, the periods of peak oil and gas production in colombia must be established. the unit for mining and energetic planning (upme) has proposed three scenarios for hydrocarbon supply in colombia based on five sources for the projected incorporation of reserves and the development of production profiles for the period 2012 to 2030 (upme, 2012). for the construction of these scenarios, the upme used variables specific to every aspect (technical, operational, economic, political, social and environmental) that has an impact on hydrocarbon exploration and the prospective extraction of resources. description of scenarios the proposed scenarios are scarcity, base and abundance. for the construction of these three scenarios, five sources of projected incorporation of reserves and production profile development were considered. the factors to be considered are listed below. • demonstrated reserves currently producing: these are existing fields where oil and gas are currently being extracted. • additional reserves resulting from improved extraction: added reserves and increased production because of improvements in the extraction rate resulting from the application of new technologies. • undeveloped reserves: currently known resources, re-estimations of specific fields, and probable reserves in other fields. • conventional resources: incorporation of other potential conventional resources, including heavy crude oil and off-shore gas. • unconventional resources: incorporation of resources such as shale oil, shale gas, tar sands and coal bed methane (cbm). it is important to clarify that the abundance and base scenarios encompass all five items, whereas the scarcity scenario does not consider unconventional resources. with the aforementioned factors, the upme estimated a time of peak production for oil and gas for each of the proposed scenarios (table 6). these production peaks are of vital importance for the development of decline models. table 6. estimation of the production peaks of oil and gas for each of the proposed scenarios. scenario hydrocarbon abundance base scarcity oil 2037 2018 2016 gas 2028 2027 2016 once the peak production period has been established (table 6), the parameter t0 of eq. (1) is known (fig. 9) for each scenario. the production magnitude at peak production corresponds to q(t0) and is known because of the scenarios constructed by upme. therefore, we have all the information needed to develop the production model. after the results are obtained, a production vs. time plot is constructed (figs. 9 and 10), where the results are presented graphically so that the three 82 beltrán rivas julian daniel & vargas jiménez carlos alberto scenarios can be compared. this type of plot also allows us to evaluate the oil and gas production models for colombia in the coming years. figure 9. daily production of crude oil in colombia for different scenarios. the dashed curves represent the scenarios proposed by the upme, and the solid curves represent the scenarios proposed in the present work. these scenarios include production between the years 2000 and 2012. 4.4 results the average oil production in colombia during 2012 was 944 mmbbl/ day (acp, 2012) and the base and scarcity scenarios show a rapid increase to peak production. the scarcity scenario reaches peak production in 2014, with a net production of 1,022.8 mmbbl/day, which is predicted to be sustained for a period of 3 years and will subsequently decline. in 2027, the production will have been reduced to half the value at peak production, and the estimated production for 2050 is 241.8 mmbbl/day. during the first 10 years of the decline stage, the scarcity scenario proposed by the upme fits well to the model proposed in the present work. however, the upme proposes an increase in crude oil production after 2022 based on the discovery and development of yet-to-find (ytf) oil fields. our prediction model forecasts a continued decline after that year. the base scenario reaches peak production in 2016, with a production of 1,273.5 mmbbl/day. no significant difference is observed with respect to the scarcity scenario once the decline stage has begun. the production in 2030 will have been reduced to half its peak value, and the estimated production for 2050 is 315.2 mmbbl/day, which is only 70 mmbbl/day better than in hat the scarcity scenario. the proposal by the upme shows a correlation with the base scenario proposed in this work up to 2022. at that point, the prognoses diverge because of the increase in crude oil production proposed by the upme. the abundance scenario considers unconventional resources, and it presents us with a different projection compared with the base and scarcity scenarios. in this case, a growth in the production rate is predicted up to 2033, where a peak production of 3,296.2 mmbbl/day will be reached and sustained for 5 years. after this stable stage ends and the decline stage begins, the production will not return to current values for nearly three decades. the production in the year 2070 is estimated to be 797.1 mmbbl/day under this scenario, which is much larger than the value obtained in the scarcity and base scenarios for the previous 20 years. for the abundance scenario, the upme only presents results up to 2035; therefore, our results cannot be compared. the average decline rate used for modeling the oil production in all three proposed scenarios in this work is 9% (anh, 2009) and a similar number is expected from the upme. we now present our results for the developed model applied to the production of gas in colombia (fig. 10). the average gas production in colombia during 2012 was 1,169.2 mmcf/ day (millions of cubic feet per day (acp, 2012) the scarcity scenario shows that the production peak of 1,322.8 mmcf/day is quickly reached in 2014 and sustained for 3 years. production then drops to half its peak value in 2023, and in 2050, the production is estimated to be 216.9 mmcf/day. the production proposed by the upme for the scarcity scenario once peak production has been reached shows production up to 2029 that is larger than what we found in the current work. subsequently, however, the situation is inverted, and our decline model shows higher production levels than those predicted by the upme. the base scenario shows an increase in production up to 2025, when a peak production of 1,743.8 mmcf is reached and sustained for 3 years. once the decline has begun, it is predicted that production will be reduced to half the peak value after only 7 years, and in 2060, the production estimate is 293.1 mmcf/day. for the base scenario, the gas production proposed by the upme after peak production is always smaller than the estimates from the decline model presented here. the abundance scenario, which considers unconventional resources, presents production growth up to 2025, when a peak production of 3,209.0 mmcf/day is reached and sustained for 4 years. both the base and abundance scenarios reach peak production in the same year; however, they differ because the growth rate for the abundance scenario is on average twice that of the base scenario. once the 4-year period has passed and the decline stage has begun, the production predicted for 2040 shows similar numbers to current production levels. by 2070, the production is estimated to be 439.6 mmcf/day. the decline rate for gas production in colombia used for the production model is 15% (anh, 2009). the proposal by the upme shows a larger decline rate; thus, the production values predicted by the upme decrease faster than our results. according to the upme, gas production will be exhausted much sooner. figure 10. daily gas production in millions of cubic feet under each scenario. the dashed curves correspond to the scenarios proposed by the upme, and the solid lines correspond to the scenarios proposed in the present work. these scenarios include production for the period from 2000 to 2012. once these production curves were constructed, it was possible to estimate the oil and gas reserves in colombia according to the decline model presented in this work. this estimation is illustrated by the area under the curve corresponding to each production scenario. conclusions the results and forecasted data presented in this research work correspond to state-of-the-art and highly relevant topics in the hydrocarbon industry. in addition, we presented results obtained from the application of various statistical techniques and models, which allowed us to generate an estimation of the hydrocarbon resources in colombia. in colombia, 0.1% of all oil fields contain 35% of the total exploitable resources, and only 0.5% of all fields represent 70% of all ultimately recoverable oil resources. in the case of gas, 0.1% of all fields contain 75% of all exploitable resources. these numbers confirm that colombia follows the same trend as the rest of the world, where most resources are concentrated in a small number of fields. for colombia, the hydrocarbon industry represents a significant economic resource, although colombia’s contribution to the world’s reserves is only 0.17% for oil and 0.11% for gas. these numbers show that colombia does not have a significant effect on the world’s reserves and supply of hydrocarbons. from decline models and based on the production curves constructed for this work, colombian oil reserves were calculated for each scenario. for the scarcity scenario, a total reserve of 7,018.0 mmbbl up to 2050 was calculated. 83hydrocarbon production scenarios in colombia. review of field sizes, hydrocarbon reserves and expectations of conventional and unconventional resources. for the base scenario, the reserves calculated for the same year were 9,315.0 mmbbl, and for the abundance scenario, our calculations predicted reserves of 39,046.0 mmbbl up to 2070. using the aforementioned modeling methodology, colombian gas reserves were estimated for each scenario. for the scarcity scenario, the total gas reserves calculated up to 2050 were 7.9 tcf. for the base scenario, total reserves of 15.4 tcf were calculated for 2060, and for the abundance model, the estimation was 27.6 tcf up to 2070. the forecasted yet-to-find oil volume, which was calculated in section 3.3 using the fractal method, was 21,486.0 mmbbl. if we add to this the 2,377.0 mmbbl corresponding to the demonstrated oil reserves up to december 31, 2012, we obtain a total of 23,863.00 mmbbl. this number lies within the range generated by the maximal reserves calculated by the base and abundance scenarios, which are 9,315 mmbbl and 39,046 mmbbl, respectively. for the case of gas exploitation, the estimated yet-to-find volume was 49.834 tcf. if we add to this the 6.760 tcf corresponding to the demonstrated gas reserves up to december 31, 2012, we obtain a total volume of 56.6 tcf. this value is double that of the reserves estimated under the abundance scenario and much larger than the gas reserves calculated for the base and scarcity scenarios. finally, the exploitation of hydrocarbons in colombia must be incentivized, and the implementation and development of new technologies must occur so that significant increases in extraction rates can be achieved. acknowledgments the authors would like to thank the institutions that provided valuable information for the development of the present work: the colombian association of petroleum engineering (acipet), colombian association of petroleum (acp), ecopetrol, and ministry of mines and energy. references agencia nacional de hidrocarburos anh (2008): sistema gerencial para el manejo de los recursos petroleros [management system for the management of oil resources] – acuerdo 11 de 2008 de la anh, 19 – 20, bogotá. agencia nacional de hidrocarburos -( anh) (2009): colombia petróleo y futuro [colombia oil and future]. 37p., bogotá. arps, .j.j., (1944): analysis of decline curves. 229 – 230 p. asociación colombiana del petróleo (acp), (2012): informe estadístico petrolero [statistical review on oil], bogotá. british petroleum. (2012) statistical review of world energy 2012. departamento administrativo nacional de estadística. dane (2013). colombia, exportaciones de café, carbón, petróleo y sus derivados, ferroníquel y no tradicionales según valores y pesos netos. ecopetrol s.a. (2005) vicepresidencia de producción. bogotá. höök, m., 2009. depletion and decline curve analysis in crude oil production (licentiate thesis, uppsala university) 112p. ivanhoe lf, luckie gg. global oil, gas fields, sizes tallied, analyzed. oil and gas journal 1993: 87 91p. salager, j., (2005): . recuperación mejorada del petróleo [enhanced oil recovery]. – 33 p., universidad de los andes; facultad de ingeniería; escuela de ingeniería química, mérida, venezuela. swokowsky earl w. (1989): calculo (cálculo) con geometría analítica [calculus (calculus) with analytic geometry] 132p., grupo editorial iberoamericana. upme unidad de planeación minero energética (unidad de planeación minero energética upme) (2012). escenarios de oferta y demanda de hidrocarburos en colombia [scenarios for supply and demand for hydrocarbons in colombia]. 9 -216p., la imprenta editores s.a, bogotá. colombia, exportaciones de café, carbón, petróleo y sus derivados, ferroníquel y no tradicionales según valores y pesos netos. [colombia, coffee exports, coal, petroleum and petroleum products, ferronickel and non-traditional products according to their values and net weights] departamento administrativo nacional de estadística. dane (2013) vargas, c. a., 2012. evaluating total yet-to-find hydrocarbon volume in colombia. earth sci. res. j., vol. 16, special issue (april, 2012): 1-246. agencia nacional de hidrocarburos (anh) (2009): colombia petróleo y futuro [colombia oil and future]. 37p., bogotá. höök, m (2009: (2009) depletion and decline curve analysis in crude oil production (licentiate thesis, uppsala university) 112p. arps .j.j. (1944): analysis of decline curves. 229 – 230 p. earl w. swokowsky (1989): calculo (cálculo) con geometría analítica [calculus (calculus) with analytic geometry] 132p., grupo editorial iberoamericana. unidad de planeación minero energética (upme) (2012). escenarios de oferta y demanda de hidrocarburos en colombia [scenarios for supply and demand for hydrocarbons in colombia]. 9 -216p., la imprenta editores s.a, bogotá. issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n3.100754 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (september, 2022): 205 210 c o m pu t e r s c ie n c e a pp l ie d t o g e o ph y si c s rigorous spherical bearing with soldner coordinates and azimuth angles on sphere sebahattin bektaş ondokuz mayis university, turkey sbektas@omu.edu.tr abstract meridian systems, called soldner coordinates (parallel coordinate) systems, have found wide application in geodesy. in particular, the meridian system constitutes a suitable base for the gauss-kruger projection of the ellipsoid and the sphere. soldner coordinates can be used in cassini-soldner projection without any processing. as it is known, the directions of the edges are shown with azimuth angles in the geographic coordinate system and the bearing angles in the soldner coordinate system. bearing or azimuth angles are frequently used in geodetic calculations. these angles give the direction of sides in the clockwise direction from a certain initial direction. both angle values range from 0 to 360 degrees and are usually calculated from the arctan function. but the arctan function returns an angle value between -90 and +90 degrees. therefore, it is necessary to analyze the quarter for the angle found. for practical computations, the quadrants of the arctangents are determined by the signs of the numerator and denominator in the tangent formulas. determining the quarter of the angles is done with if…, then…, end..., blocks on the computer. it should be noted that each comparison requires a separate processing time. this study will be given how to calculate both bearing and azimuth angles with direct formulas without any need to examine them. in addition, a solution proposal will be given against the division by zero errors in the bearing and azimuth angles calculations. keywords: bearing angles; azimuth angles; geographical coordinates; soldner coordinates; meridian convergence; zero division error; rodamiento esférico riguroso con coordenadas soldner y ángulos azimut en la esfera resumen los sistemas meridianos, llamados coordenadas de soldner (coordenadas paralelas), han gozado de una amplia aplicación en geodesia. en particular, el sistema meridiano constituye una base adecuada para la proyección de gauss-kruger del elipsoide y la esfera. las coordenadas de soldner se pueden usar en la proyección cassini-soldner sin ningún procesamiento. como es conocido, las direcciones de los bordes se muestran como el ángulo acimut en el sistema geográfico de coordenadas y los ángulos rumbo en el sistema de coordenadas de soldner. los ángulos rumbo y acimut se usan frecuentemente en los cálculos geodéticos. estos ángulos dan la dirección de los lados en el sentido del reloj desde una dirección inicial específica. ambos ángulos van de 0 a 360 grados y usualmente se calculan desde la función arctan. el resultado de la función arctan, sin embargo, está en el rango de -90 y +90 grados. además, es necesario analizar el cuartil del ángulo resultado. para computaciones prácticas, los cuadrantes de los arcotangentes se determinan por los signos del numerador y denominador en las fórmulas tangentes. la determinación de los cuartiles de los ángulos se hace en el computador con los bloques si..., entonces..., y fin... se debe tener en cuenta que cada comparación requiere un tiempo de procesamiento por separado. este estudio presenta cómo calcular los ángulos rumbo y acimuth con formulas directas sin necesidad de examinarlas. adicionalente, se dará una propuesta de solución frente a la división por cero errores en el cálculo de los ángulos de rumbo y acimut. palabras clave: ángulos rumbo; ángulos azimuth; coordenadas geográficas; coordenadas soldner; convergencia meridiana; error de división cero. record manuscript received: 25/02/2022 accepted for publication: 27/10/2022 how to cite item: bektaş, s. (2022). rigorous spherical bearing with soldner coordinates and azimuth angles on sphere. earth sciences research journal, 26(3), 205-210. https://doi.org/10.15446/esrj.v26n3.100754 https://doi.org/10.15446/esrj.v26n3.100754 mailto:sbektas@omu.edu.tr https://doi.org/10.15446/esrj.v26n3.100754 206 sebahattin bektaş 1. introduction the reciprocal bearing angles are exactly 180 degrees different on the plane. unlike the plane on the sphere surface, the difference between reciprocal bearing or azimuth angle is approximately 180 degrees. therefore, while a single bearing angle is sufficient in plane calculations, it is desired to find reciprocal bearing or azimuth angles separately on surfaces such as spheres and ellipsoids. classic the bearing calculation or the azimuth calculation formulas only work correctly if the side is in the 1st quarter. if the side is located in the other quarters, the angles of the bearing or azimuth should be examined. for example, if the angle value given by the arctan function is positive, the bearing or azimuth angle is in the 1st or 3rd quarter. if the angle value is negative, the angle is in quarter 2 or 4. in order to find the value of the angle as univocal, the angle values found from the classical formulas need to be examined. bearing or azimuth angles are critical in geodetic calculations. it is impossible to produce correct position information with wrong bearing or azimuth angles. calculation of angles is done with the inverse (second) geodetic problem. while solving the inverse geodetic problem in order to find reciprocal bearing or azimuth angles, first, the angle from 1 to 2 is found. as a result of the analysis, if the 1st bearing or azimuth angle is wrong, it is likely to be wrong in the opposite (from 2 to 1) bearing or azimuth. in the proposed method, reciprocal bearing or azimuth angles are obtained independently and directly. in our literature research, we came found some studies only for determining plane bearing angle (antes, 1990; breuer et al., 1985; gellert et al., 1972) demonstrated direct bearing angle determination in the plane without quarter analysis. hekimoglu (1991) also examined the condition problem and the direct formulas. we could not find any studies that give a direct bearing and azimuth angle on the spherical surface. formulas that give a direct azimuth angle on the sphere surface without examining the quarter with geographical coordinates are shown by the author in (bektas, 2019; bektas, 2021). for the first time, in this study, formulas that give a direct bearing angle without examining the quarter with soldner coordinates on the sphere surface will be discussed. on the other hand, a solution proposal will be given against the division by zero errors in the bearing and azimuth angles calculations. in this paper, firstly, bearing and azimuth angle, meridian system and soldner coordinates on sphere will be mentioned in order to form a basis for the subject. after the relationship between azimuth, bearing angle and meridian convergence is given, direct azimuth angle determination with geographical coordinates, then direct bearing angle determination with soldner coordinates will be discussed. 2. direct azimuth angles with geographical coordinates figure 1. problem on sphere classically, the reciprocal azimuth angles (a12, a21) from two points whose geographic coordinates p1(φ1, λ1), p2(φ2,λ2) are given obtained from the following formulas (fig.1). (1) (2) it is important to remember that these classic formulas only work correctly if the side is in the 1st quarter. if the side is located in the other quarters, the angles of the bearing angle should be examined. the necessary additions should be made for correct angles according to table-1 seen below. table 1. fixed value to add for azimuth angles quadrant ∆φ=φ2−φ1 ∆λ=λ2− λ1 fixed value to add for a12 fixed value to add for a21 1.quadrant + + 2.quadrant + +180o +180o 3.quadrant +180o -180o 4.quadrant + +360o in this proposed method, formulas are given for obtaining the azimuth angle directly without any examination. for direct determination of azimuth by geographic coordinates, we provide the below formulas. in this proposed method, formulas are given to obtain the azimuth angle directly without any examination. the proposed method can calculate direct azimuth angles on the sphere and ellipsoid surface (bektas, 2019). i=sinδ λ ii=tan 2cos 1 − sin 1cosδ iii=tan 1cos 2 − sin 2cosδ a 12 = 2.arctan( −√ 2 + 2 ) + 180 a 21 = 2.arctan(− +√ 2 + 2 ) + 180 (3)i=sinδ λ ii=tan 2cos 1 − sin 1cosδ iii=tan 1cos 2 − sin 2cosδ a 12 = 2.arctan( −√ 2 + 2 ) + 180 a 21 = 2.arctan(− +√ 2 + 2 ) + 180 (4) i=sinδ λ ii=tan 2cos 1 − sin 1cosδ iii=tan 1cos 2 − sin 2cosδ a 12 = 2.arctan( −√ 2 + 2 ) + 180 a 21 = 2.arctan(− +√ 2 + 2 ) + 180 (5) i=sinδ λ ii=tan 2cos 1 − sin 1cosδ iii=tan 1cos 2 − sin 2cosδ a 12 = 2.arctan( −√ 2 + 2 ) + 180 a 21 = 2.arctan(− +√ 2 + 2 ) + 180 (6) i=sinδ λ ii=tan 2cos 1 − sin 1cosδ iii=tan 1cos 2 − sin 2cosδ a 12 = 2.arctan( −√ 2 + 2 ) + 180 a 21 = 2.arctan(− +√ 2 + 2 ) + 180 (7) the given formulas in equations 6 and 7 do not work if the denominator equals zero. the program will be interrupted because division by zero is done. this problem also arises in the use of classical formulas. in order not to interrupt the program, very small numbers are added to the numerator and denominator at a rate that will not affect the result. if μ1=1e – 15,... μ2 =1e – 27,... are selected, ten digit sensitive bearing angle values are obtained (hekimoğlu, 1991). the division by zero error will be avoided if the following formulas (eq. 8 and eq. 9) are used. a12 = 2.arctan ( +µ1 µ2+ −√ 2 + 2 ) + 180 a21 = 2.arctan ( +µ1 µ2− +√ 2 + 2 ) + 180 (8) a12 = 2.arctan ( +µ1 µ2+ −√ 2 + 2 ) + 180 a21 = 2.arctan ( +µ1 µ2− +√ 2 + 2 ) + 180 (9) 207 rigorous spherical bearing with soldner coordinates and azimuth angles on sphere 3. direct bearing angles with soldner coordinates in meridian system on sphere this meridian system, as the name suggests, is based on the meridian. the meridian that defines the system is called the prime meridian, and a meridian near the study area is taken as the prime meridian (heck, 2003).osoldner coordinates system, have found wide application in geodesy. in particular, the soldner coordinates system constitutes a suitable base for projection of the ellipsoid and the sphere. soldner coordinates can be used in cassini-soldner projection without any processing. in this system, the x-axis is the selected prime meridian (figure 2). the coordinates of a point p1 on the sphere in the meridian system are defined as follows. the spherical distance (arcs of a great circle) from the p1 point to the prime meridian (p1f1) is the y value of the p1 point (y1 = p1f1). the distance of the perpendicular point f1 to the starting point po on the prime meridian also gives the x coordinate (x1 = pof1). the starting point of po can also be taken on the equator. figure 2. meridian system and soldner coordinates (bektas, 2021) these sides used in the calculations on the sphere s, y1, x1, y2, x2 in figure 2 must be arcs of a great circle. point o in figure 2 becomes the pole of the prime meridian circle. so it is always of1= of2 = opo = π/2. in the soldner coordinate system, the reference direction of the bearing angles of the sides is the parallel y=constant circle drawn to the prime meridian at that point (figure-2). the prime meridian determines the soldner system. each meridian system creates a separate coordinate system. the ordinates of the points (y) to the left of the prime meridian are negative. likewise, the abscissas of points (x) below the starting point take negative values. the azimuth angle is used for the direction values of the sides in main problem solutions with geographical coordinates. the azimuth angle (a12, a21) is the angle from the north wing of the meridian at a point to the clockwise side (fig. 2). the bearing angle is used for the direction values of the sides in main problem solutions with soldner coordinates meridian system. the bearing angle (α12, α21) is from the north wing of the parallel to the prime meridian at a point to the clockwise side (fig. 2). figure 3. azimuth angle, bearing angle, meridian convergence (bektas, 2021) there is a meridian convergence ( γ ) between bearing and azimuth angle (fig. 3). meridian convergence can be easily calculated from the geographic or soldner coordinates of the point. if the bearing angle (α) is given, it is converted to the azimuth angle by using the meridian convergence at that point a = α + γ (10) for example, γ1 meridian convergence on the p1 point tan γ1 = sin φ1 tan( λ1 -λprime) from geographic coordinates (11) tan 1 = 1 1 𝛾𝛾 from soldner coordinates (12) reciprocal angles between two points whose coordinates are given in geodesy are obtained by solving the inverse (second) geodetic problem. classically, the reciprocal bearing angles (α12, α21) from two points whose soldner coordinates p1(y1, x1), p2(y2, x2) are given obtained from the following formulas (fig. 4). solving p1p2o spherical triangle figure 4. inverse geodetic problem in the meridian system on the sphere (bektas, 2021) dy = y2 y1 (13) α12 = arctan ( ) ( 1 ) ( )− ( 2 ) ( 1 ) α21 = arctan ( ) ( 1 ) ( 2 )− ( 2 ) ( ) (14) α12 = arctan ( ) ( 1 ) ( )− ( 2 ) ( 1 ) α21 = arctan ( ) ( 1 ) ( 2 )− ( 2 ) ( ) (15) the bearing angle values found from the classical formula need to be examined. according to the quarter where the angle is located, the additions in the table 2 below should be made. table 2. fixed value to add for bearing angles quadrant ∆y=y2− y1 fixed value to add for α12 fixed value to add for α21 1 and 2 + 90o 270o 3 and 4 270o +90o 208 sebahattin bektaş the following values of i, ii and iii are calculated to find the direct bearing angle without examining the sphere surface. dx = x2 x1 (16) = ( 1 ) ( 2 ) − ( 1 ) ( ) = ( ) = ( 2 ) ( 1 ) − ( 2 ) ( ) α12 = 2.arctan ( −√ 2 + 2 ) + 180 α21 = 2.arctan(− −√ 2 + 2 ) + 180 (17) = ( 1 ) ( 2 ) − ( 1 ) ( ) = ( ) = ( 2 ) ( 1 ) − ( 2 ) ( ) α12 = 2.arctan ( −√ 2 + 2 ) + 180 α21 = 2.arctan(− −√ 2 + 2 ) + 180 (18) = ( 1 ) ( 2 ) − ( 1 ) ( ) = ( ) = ( 2 ) ( 1 ) − ( 2 ) ( ) α12 = 2.arctan ( −√ 2 + 2 ) + 180 α21 = 2.arctan(− −√ 2 + 2 ) + 180 (19) = ( 1 ) ( 2 ) − ( 1 ) ( ) = ( ) = ( 2 ) ( 1 ) − ( 2 ) ( ) α12 = 2.arctan ( −√ 2 + 2 ) + 180 α21 = 2.arctan(− −√ 2 + 2 ) + 180 (20) = ( 1 ) ( 2 ) − ( 1 ) ( ) = ( ) = ( 2 ) ( 1 ) − ( 2 ) ( ) α12 = 2.arctan ( −√ 2 + 2 ) + 180 α21 = 2.arctan(− −√ 2 + 2 ) + 180 (21) the given formulas in equation 20 and equation 21 do not work if the denominator equals zero. the program will be interrupted because division by zero is done. this problem also arises in the use of classical formulas. in order not to interrupt the program, very small numbers are added to the numerator and denominator at a rate that will not affect the result. if μ1=1e – 15,... μ2 =1e – 27 are selected, ten-digit sensitive bearing angle values are obtained. the division by zero error will be avoided if the following formulas equation 22 and equation 23 are used. α12 = 2.arctan ( +µ1 µ2+ −√ 2 + 2 ) + 180 α21 = 2.arctan( +µ1 µ2− −√ 2 + 2 ) + 180 (22) α12 = 2.arctan ( +µ1 µ2+ −√ 2 + 2 ) + 180 α21 = 2.arctan( +µ1 µ2− −√ 2 + 2 ) + 180 (23) 4. numerical example in order to show the validity of the given formulas, numerical applications were made separately in the east and west of greenwich in both the northern and southern hemispheres. inverse problem was solved between twenty-point pairs in total. the radius of the earth is taken as r=6370000m, the prime meridian is taken as λprime=30 o to the east of greenwich and λprime= -30 o to the west of greenwich. in numerical applications (y, x) soldner coordinates and (γ) meridian convergences of twenty-point pairs are calculated whose (φ, λ) geographic coordinates are given. then, the bearing (α) and the azimuth angles (a) calculations were made separately with direct formulas. in order to control between the bearing and azimuth angles, the azimuth angle was found by adding the meridian convergence to the bearing angle (aij=αij +γi) between each pair of points. numerical application results can be seen in table 3. table 3. the bearing and azimuth angles calculations from direct formula φ1= 30.0 λ1= 30.5 y1= 48141.1054 x1= 3335429.2308 γ1= 0.2500047597 φ2= 32.0 λ2= 31.0 y2= 94282.5003 x2= 3558115.1919 γ2= 0.5299579646 a12= 11.9669826036 a21=192.2245420693 α12= 11.7169778439 α21=191.6945841047 a12=α12+γ1= 11.9669826036 φ1= 30.0 λ1= 30.5 y1= 48141.1054 x1= 3335429.2308 γ1= 0.2500047597 φ2= 29.0 λ2= 32.0 y2= 194466.7324 x2= 3225792.8907 γ2= 0.9699205898 a12=127.0797659802 a21=307.8184614062 α12=126.8297612205 α21=306.8485408164 a12=α12+γ1=127.0797659802 φ1= 30.0 λ1= 30.5 y1= 48141.1054 x1= 3335429.2308 γ1= 0.2500047597 φ2= 28.0 λ2= 29.0 y2= -98162.7839 x2= 3113371.4602 γ2= -0.4695087290 a12=213.6299774554 a21= 32.9026204737 α12=213.3799726956 α21= 33.3721292026 a12=α12+γ1=213.6299774554 φ1= 30.0 λ1= 30.5 y1= 48141.1054 x1= 3335429.2308 γ1= 0.2500047597 φ2= 32.0 λ2= 29.0 y2= -94282.5003 x2= 3558115.1919 γ2= -0.5299579646 a12=327.6475721969 a21=146.8748649794 α12=327.3975674372 α21=147.4048229439 a12=α12+γ1=327.6475721969 φ1= 30.0 λ1= 30.5 y1= 48141.1054 x1= 3335429.2308 γ1= 0.2500047597 φ2= 30.0 λ2= 31.0 y2= 96281.2941 x2= 3335744.3496 γ2= 0.5000380801 a12= 89.8749994050 a21=270.1250005950 α12= 89.6249946453 α21=269.6249625149 a12=α12+γ1= 89.8749994050 φ1= 0.0 λ1= 30.5 y1= 55588.7367 x1= 0.0000 γ1= 0.0000000000 φ2= 0.0 λ2= 32.0 y2= 222354.9467 x2= 0.0000 γ2= 0.0000000000 a12= 90.0000000000 a21=270.0000000000 α12= 90.0000000000 α21=270.0000000000 a12=α12+γ1= 90.0000000000 209 rigorous spherical bearing with soldner coordinates and azimuth angles on sphere φ1 = 30.0 λ1= 30.5 y1= 48141.1054 x1= 3335429.2308 γ1= 0.2500047597 φ2 = 29.0 λ2= 30.5 y2= 48618.8595 x2= 3224249.5773 γ2= 0.2424095173 a12=180.0000000000 a21=360.0000000000 α12=179.7499952403 α21=359.7575904827 a12=α12+γ1=180.0000000000 φ1 = 30.0 λ1= 0.0 y1= -2852692.3684 x1= 3745576.5846 γ1=-16.1021137520 φ2 = 29.0 λ2= 0.0 y2= -2883094.7939 x2= 3626800.4993 γ2=-15.6371970759 a12=180.0000000000 a21=360.0000000000 α12=196.1021137520 α21= 15.6371970759 a12=α12+γ1=180.0000000000 φ1= 30.0 λ1= -30.5 y1= -48141.1054 x1= 3335429.2308 γ1= -0.2500047597 φ2= 32.0 λ2= -31.0 y2= -94282.5003 x2= 3558115.1919 γ2= -0.5299579646 a12=348.0330173964 a21=167.7754579307 α12=348.2830221561 α21=168.3054158953 a12=α12+γ1=348.0330173964 φ1= 30.0 λ1= -30.5 y1= -48141.1054 x1= 3335429.2308 γ1= -0.2500047597 φ2= 29.0 λ2= -32.0 y2= -194466.7324 x2= 3225792.8907 γ2= -0.9699205898 a12=232.9202340198 a21= 52.1815385938 α12=233.1702387795 α21= 53.1514591836 a12=α12+γ1=232.9202340198 φ1= 30.0 λ1= -30.5 y1= -48141.1054 x1= 3335429.2308 γ1= -0.2500047597 φ2= 28.0 λ2= -29.0 y2= 98162.7839 x2= 3113371.4602 γ2= 0.4695087290 a12=146.3700225446 a21=327.0973795263 α12=146.6200273044 α21=326.6278707973 a12=α12+γ1=146.3700225446 φ1= 30.0 λ1= -30.5 y1= -48141.1054 x1= 3335429.2308 γ1= -0.2500047597 φ2= 32.0 λ2= -29.0 y2= 94282.5003 x2= 3558115.1919 γ2= 0.5299579646 a12= 32.3524278031 a21=213.1251350206 α12= 32.6024325628 α21=212.5951770561 a12=α12+γ1= 32.3524278031 φ1= -30.0 λ1= 30.5 y1= 48141.1054 x1= -3335429.2308 γ1= -0.2500047597 φ2= -32.0 λ2= 31.0 y2= 94282.5003 x2= -3558115.1919 γ2= -0.5299579646 a12=168.0330173964 a21=347.7754579307 α12=168.2830221561 α21=348.3054158953 a12=α12+ γ1=168.0330173964 φ1= -30.0 λ1= 30.5 y1= 48141.1054 x1= -3335429.2308 γ1= -0.2500047597 φ2= -29.0 λ2= 32.0 y2= 194466.7324 x2= -3225792.8907 γ2= -0.9699205898 a12= 52.9202340198 a21=232.1815385938 α12= 53.1702387795 α21=233.1514591836 a12=α12+ γ1= 52.9202340198 φ1= -30.0 λ1= 30.5 y1= 48141.1054 x1= -3335429.2308 γ1= -0.2500047597 φ2= -28.0 λ2= 29.0 y2= -98162.7839 x2= -3113371.4602 gam2= 0.4695087290 a12=326.3700225446 a21=147.0973795263 α12=326.6200273044 α21=146.6278707973 a12=α12+ γ1=326.3700225446 φ1= -30.0 λ1= 30.5 y1= 48141.1054 x1= -3335429.2308 γ1= -0.2500047597 φ2= -32.0 λ2= 29.0 y2= -94282.5003 x2= -3558115.1919 γ2= 0.5299579646 a12=212.3524278031 a21= 33.1251350206 α12=212.6024325628 α21= 32.5951770561 a12=α12+ γ1=212.3524278031 φ1= -30.0 λ1= -30.5 y1= -48141.1054 x1= -3335429.2308 γ1= 0.2500047597 φ2= -32.0 λ2= -31.0 y2= -94282.5003 x2= -3558115.1919 γ2= 0.5299579646 a12=191.9669826036 a21= 12.2245420693 α12=191.7169778439 α21= 11.6945841047 a12=α12+ γ1=191.9669826036 φ1= -30.0 λ1= -30.5 y1= -48141.1054 x1= -3335429.2308 γ1= 0.2500047597 φ2= -29.0 λ2= -32.0 y2= -194466.7324 x2= -3225792.8907 γ2= 0.9699205898 a12=307.0797659802 a21=127.8184614062 α12=306.8297612205 α21=126.8485408164 a12=α12+ γ1=307.0797659802 φ1= -30.0 λ1= -30.5 y1= -48141.1054 x1= -3335429.2308 γ1= 0.2500047597 φ2= -28.0 λ2= -29.0 y2= 98162.7839 x2= -3113371.4602 γ2= -0.4695087290 a12= 33.6299774554 a21=212.9026204737 α12= 33.3799726956 α21=213.3721292026 a12=α12+ γ1= 33.6299774554 φ1= -30.0 λ1= -30.5 y1= -48141.1054 x1= -3335429.2308 γ1= 0.2500047597 φ2= -32.0 λ2= -29.0 y2= 94282.5003 x2= -3558115.1919 γ2= -0.5299579646 a12=147.6475721969 a21=326.8748649794 α12=147.3975674372 α21=327.4048229439 a12=α12+ γ1=147.6475721969 210 sebahattin bektaş 5. results and discussion in this study, formulas are given for how to obtain the bearing and the azimuth angle directly without any examination. the numerical examples also show no loss of precision in the use of direct formulas. it has been observed that the classical formulas (equations 1, 2, 14, and 15) and the formulas we propose (equations 6, 7, 20, and 21) give the same results in the calculations made with ten digits after the decimal point. it is suggested to use equations 8, 9, 22, and 23) to avoid division by zero error. the advantage of the method is that no examination is required. in computer calculations, if…, then..., end…, blocks are not used when direct formulas are used. the if..., then…, end…, blocks reduce the execution speed in computer calculations. the classical formulas and the direct formulas we proposed were compared in terms of processing time. all were coded in matlab. the main characteristics of the hardware were windows 10 pro, intel(r) core(tm) i52450m cpu @ 2.50ghz, 6 gb of ram. the same bearing and the same azimuth angle were calculated 500 million times and the processing times are shown in table 4 below. table 4. processing time bearing angle azimuth angle classic formulas 4.642 sec eq. 14-eq. 15 5.492 sec eq. 1-eq. 2 proposed formulas 4.445 sec eq. 20-eq. 21 4.413 sec eq. 6-eq. 7 it has been observed that the processing time is slightly shorter in the direct formulas we recommend. this is because classical formulas need quadrant comparison. 6. conclusion in this proposed method, formulas are given for how to obtain the bearing and azimuth angle directly without any examination. the need for comparison in classical formulas increases both the processing time and the possibility of error. direct formulas are always more useful. while saving time on the one hand, the probability of making mistakes will also be seriously reduced on the other hand. we hope that the accuracy and non-examination of the direct formulas we proposed will increase the usability of our formulas. the formulas on the sphere surface, which are used to calculate the direct bearing and azimuth, can also be adapted to the ellipsoid surface. for future studies, researchers are advised to find more simple direct formulas. references antes, e. (1990). noch ein beitrag zur berechung des richtungwinkels. verm. wesen, 115. bektaş, s. (2019). direct bearing angles determination on globe. moj civil engineering, 5(4), 78–80. bektaş, s. (2021). jeodezi-1, atlas akademi press. isbn 978-605-7839-87-9. breuer, p., hirle, m., & joeckel, r. (1985). basicprogrammierbare taschenrechner und handcomputer. mitt.des deutscher verein für vermessungswesen-bayern, 37, jahr. sonderheft 1/1985, s.218-219 gellert, w., küstner, h., hellwich, m., & kastner, h. (1972). kleine enzyklopadie: mathematik. verlag harri deustch, frankfurt/ m. und zürich. heck, b. (2003).  rechenverfahren und auswertemodelle der landesvermessung: klassische und moderne methoden. 3. neu bearbaitete und erweiterte auflage. heidelberg: herbert wichmann, 473 pp. hekimoğlu, ş. (1991). açıklık açısını bölge sorgulaması yapmadan hesaplayan yeni formüller ve kondüsyon sorunu. hkmo dergisi, sayı, 70, trabzon theissen, r. (1990). berechnung des richtungswinkels t ohne quadrantenabfrage. z.fg. verm. wesen 115,193-195 _hlk97057942 _hlk97058323 _hlk97058734 _hlk97059392 _hlk97059724 rock burst is one of main engineering geological problems greatly threatening the safety of construction. prediction of rock burst is always an important issue concerning the safety of workers and equipments in tunnels. in this paper, a novel pnn-based rock burst prediction model is proposed to determine whether rock burst will happen in the underground rock projects and how much the intensity of rock burst is. the probabilistic neural network (pnn) is developed based on bayesian criteria of multivariate pattern classification. because pnn has the advantages of low training complexity, high stability, quick convergence, and simple construction, it can be well applied in the prediction of rock burst. some main control factors, such as rocks’ maximum tangential stress , rocks’ uniaxial compressive strength , rocks’ uniaxial tensile strength and elastic energy index of rock are chosen as the characteristic vector of pnn. pnn model is obtained through training data sets of rock burst samples which come from underground rock project in domestic and abroad. other samples are tested with the model. the testing results agree with the practical records. at the same time, two realworld applications are used to verify the proposed method. the results of prediction are same as the results of existing methods, just same as what happened in the scene, which verifies the effectiveness and applicability of our proposed work. abstract keywords: probabilistic neural network (pnn); rock burst; prediction pnn-based rock burst prediction model and its applications modelo de predicción de fractura de rocas basado en una red neuronal probabilística y sus aplicaciones issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n3.65216 el fracturamiento o explosión de rocas es uno de los principales problemas en ingeniería geológica que amenaza significativamente la seguridad de una construcción. la predicción del fracturamiento de rocas es importante para la seguridad de los trabajadores y el equipamiento en túneles. en este artículo se propone un nuevo modelo de predicción de fracturamiento de rocas basado en una red neuronal probabilística (pnn por sus siglas en inglés) para determinar la posible ocurrencia e intensidad de uno de estos eventos en proyectos subterráneos. la pnn se desarrolló con base en un criterio bayesiano para la clasificación multivariada de patrones. debido a que la pnn tiene las ventajas de una menor complejidad de adiestramiento, estabilidad, rápida convergencia y simplicidad en su construcción, se puede adecuar en la predicción del fracturamiento de rocas. algunos factores principales de control, como la fuerza máxima tangencial de rocas, la resistencia de compresión uniaxial, la fuerza de tensión uniaxial, y el índice de energía elástica de las rocas fueron escogidos como los vectores característicos de la pnn. el modelo se obtuvo a través del adiestramiento de datos sobre fracturamiento de rocas en proyectos subterráneos en diferentes localidades. otras datos también se analizaron con el modelo. los resultados de la evaluación se ajustan a los registros observados. simultáneamente, se utilizaron dos aplicaciones prácticas para verificar el método propuesto. los resultados de la predicción son similares a los de métodos existentes, un factor que además se presentó en las pruebas de campo, lo que demuestra la efectividad y la aplicabilidad de la metodología propuesta. resumen palabras clave: red neuronal probabilística; fracturamiento de rocas; predicción. record manuscript received: 24/05/2017 accepted for publication: 23/08/2017 how to cite item zhou, y., & wang, t. (2017). pnn-based rock burst prediction model and its applications. earth sciences research journal, 21(3). 141 146. doi: http://dx.doi.org/10.15446/esrj.v21n3.65216 earth sciences research journal earth sci. res. j. vol. 21, no. 3 (september, 2017): 141 146 yu zhou*, tingling wang school of electric power, north china university of water resources and electric power, zhengzhou 450045, china zhouyumobile@163.com g e o l o g ic a l e n g in e e r in g 142 yu zhou, tingling wang 1. introduction a rock burst is a sudden and violent expulsion of rock from the surrounding rock mass. rock burst is considered a dynamic instability phenomenon of surrounding rock mass of un-derground space in high geostatic stress and caused by the violent release of strain energy stored in the rock mass. rock burst occurs during excavating underground space in the form of a stripe of rock slices or rock fall or throwing of rock fragments, sometimes accompanied by crack sound. rock bursts are related to the fracture of rock in place and require two conditions for their occurrence: stress in the rock mass sufficiently high to exceed its strength, and physical characteristics of the rock which enable it to store energy up to the threshold value for sudden rupture. rocks which yield gradually in plastic strain under load usually do not generate rock bursts. the likelihood of rock bursts occurring increases as the depth of the mine increases. rock bursts are also affected by the size of excavation, becoming more likely if the excavation size is around 180m and above. induced seismicity such as faulty methods of mining can trigger rock bursts. other causes of rock bursts are the presence of faults, dikes, or joints (dong et al., 2013). because rock burst occurs suddenly and intensely, it usually causes injury including death to workers, damage to equipment, and even substantial disruption and economic loss of under-ground space excavation. therefore, there is a need for the development of suitable computa-tional methods for the prediction and control of rock bursts particularly for a safe and economic underground excavation for construction or mining in the burst-prone ground. for this case, numerous related research works, concerning about the mechanism, characteristics or type, the cause of formation, the critical conditions and preventive methods of rock burst have been con-ducted by many researchers. many researchers have suggested various theories, many predic-tion methods, and empirical correlation, such as fuzzy-base evaluation method (wang et al., 1998; amoussou et al., 2013), distance discriminant analysis (wang et al., 2009; wang et al., 2017), support vector machine (svm) (zhao, 2005; zhou et al., 2012, extension-theory-based method (xiong et al., 2007), rough-set-based method (yang, 2010), unascertained measurement method(shi et al., 2010), numerical simulation (zhen & gao, 2017; zhu et al., 2010) and case study (mansurov, 2001). these studies offered new ideas and approaches for rock burst prediction. however, each method discussed above has its advantages and disadvantages, and understanding, predicting and controlling the rock bursts still pose a considerable challenge for underground engineering. as an important means, the ann-based method for prediction of rock burst has been adopted by many researchers gradually in recent years (bai et al., 2002; zhang et al., 2012). ar-tificial neural network technique is considered the most effective and reliable artificial intelli-gence methods for solving classification, prediction and recognition problems. currently, back propagation (bp) and radial basis function (rbf) networks are used in the field of prediction of robust classification. however, for network training, they are all easily trapped in local minimum values. probabilistic neural network (pnn), on the other hand, is a feedforward neural network. it is derived from the bayesian network and a statistical algorithm called kernel fisher discriminant analysis. it was introduced by specht and donald(1990). be-cause pnn has the advantages of low training complexity, high stability, quick convergence, and simple construction, it has a wide range of application in model classification, identification, prediction, as well as fault diagnosis and other fields(adeli and panakkat, 2009; song et al., 2007; ataa et al., 2017; rutkowski, 2004 ). in this work, according to the practice of complicat-ed problems of the rock burst prediction, the pnn is applied to predicting rock burst classifica-tion. 2. material and methods 2.1. criteria and indexes of rock burst and rock burst classification 2.1.1. criteria considering stress in surrounding rock the criteria listed in table 1 were proposed early, and only considered the stress level in surrounding rock. furthermore, different scholars chose different parameters as an evaluation index of criterion for rock burst, and the classification of rock burst intensity also differed from each other. so it is difficult to use these criteria in the construction of underground engineering. table 1. criteria only considering stress in surrounding rock. 2.1.2. comprehension criteria considering stress, properties of surrounding rock and energy 1) the following criterion is presented with rock burst tendency index and energy condition of surrounding rock (dong et al, 2013). (1) (2) (3) (4) where w qx is the rock burst tendency index; σ 1 and σ2 are the major and middle principal stress in surrounding rock, respectively; μ is the poisson ratio. 2) it is stipulated that rock burst could occur if σ ө / σ c ≤ k s in which the value of k s related to σ t / σ c criterion. 3) kidybinski (1981) proposed an elastic energy index w et . no rock burst activity, moderate rock burst activity and strong rock burst activity, meet the conditions w et <2.0, 2.0≤ w et ≤5.0, and w et >5.0, respectively. 2.1.3 input characteristic vector for pnn the indexes of criterion should reflect the main factors of rock burst the properties and stress of surrounding rock. at the same time, they should be obtained easily and can be compared with each other for different cases. in this work, the compressive rock strength σ c , tensile strength σ t , elastic energy 143pnn-based rock burst prediction model and its applications index w et and the maximum tangential stress σ ө are chosen as the indexes of criterion. compressive rock strength σ c , tensile strength σ t , and elastic energy index w et can indicate the properties of surrounding rock, and the tangential stress σ ө can reflect the virgin geostatic stress condition and the influence of the shape and dimension of the underground space on rock burst. in this work, σ ө , σ c , σ t and w et are selected as the input index for pnn model to predict the degree of rock burst activity. hence, the input characteristic vector for pnn is [σ ө , σ c , σ t , w et ]. 2.1.4 classification for intensities of rock burst according to the extent and intensity of the characteristics of the rock burst phenomenon in the underground openings, the grade of rock burst is divided into four degrees, namely none rock burst, light rock burst, moderate rock burst, strong rock burst, respectively. so the pnn model output is rock burst degree, output = [none rock burst, light rock burst, moderate rock burst, strong rock burst]. also, the division of rock burst degree can be described in table 2 (wang et al., 1998; zhou et al., 2012; zhang et al., 2004) table 2. standard of classification for intensities of rock burst. number of category ө q . additionally, σ is the smoothing parameter. figure 1 depicts a schematic diagram of the multi-class classification pnn. x is a vector to be classified as the neurons in the input layer. it is passed to the corresponding neurons in the hidden layer with no change. the hidden layer then transmits each neuron in the accumulated layers. at this point, the output obtained by the accumulation layer is the estimation of the probability density function of each pattern for the test vectors. accordingly, the category of the occurrence of the maximum probability of the current test vector is the one that corresponds to the largest probability density function. this function is the output of the accumulation layer. the neuron output with the probability density maximum is 1; the corresponding category is the one to which h belongs. 2.2. pnn-based rock burst prediction model 2.2.1 outline of pnn probability neural networks are a tool for handling uncertainty for improving learning per-formance. probability uncertainty and fuzziness uncertainty processing play a key role in boost-ing classification systems including extreme learning machines and decision trees (wang et al., 2015; lu et al., 2015). pnn is essentially a classifier that places the bayes estimate in a feed-forward neural network. the central concept of bayes criterion classification is the minimal ‘‘predictable risk’’ of the bayes decision. the bayes decision is based on the non-parametric estimation of the probability density function; accordingly, it obtains the classification results. based on its advantages, such as rapid training time, a stable and simple neural structure, and good convergence, it is suitable for use in defect recognition. for a multi-class problem with σ 1 , σ 2 , , σ q , σ s , we apply to the above issues two types of classification problems in bayes decision classification. for p-dimensional vector x = {x 1 , x 2 , , x p } based on the bayes decision rule, we determine the status of өϵө q with its measurement set, (5) in the equation (5), h q represents the priori probability of ө = ө q , d(x) as the bayes decision of test vector x , h k is the priori probability of ө = ө k , and l q and l k are incorrectly classified into other categories of losses. the latter should belong to ө q and ө k . besides, f q (x) and f q (x) are probability density function of ө q and ө k , respectively. (6) where x is a sample of the input vector to be classified, p is the dimension of sample vectors, x qj is the j-th sample of the category ө q , and mq is the sample figure 1. schematic diagram of the multi-class classification pnn. 2.2.2 pnn modeling for prediction of rock burst the number of input layer neurons of pnn is the same as the dimensionality of the input characteristic vector. based on discussed above, σ ө , σ c , σ t , and w et are selected as the input characteristic vector for pnn model and the input vector of pnn = [σ ө , σ c , σ t , w et ]. hence, the number of input layer neurons of pnn is 4. the number of output neurons of pnn is the same as the number of classification of rock burst activity. according to the section 2.4, the grade of rock burst is divided into four degrees, and the pnn model output is rock burst degree and the output vector = [none rock burst, light rock burst, moderate rock burst, strong rock burst]. so the number of output layer neurons is 4. the number of hidden layer neurons is determined by the training data set. the number of hidden layer neurons is equal to the sum of the number of each category of training sample the number of accumulation layer neurons is the same as the number of classification of rock burst activity. here, the number of accumulation layer neurons is 4. the design of pnn modeling for prediction of rock burst includes the following aspects: a collection of data sets, data preprocessing, build a pnn, network model, training pnn, testing pnn etc. the design process is shown in figure 2. 144 yu zhou, tingling wang figure 2. the design process of pnn modeling for prediction of rock burst. 3. results rock burst samples which come from underground rock projects in domestic and abroad are collected as training data set (show in table 3) and testing data set (show in table 4) to verify the rationality of our posed method. the relationship among the indexes of criteria, the occurrence of rock burst and its intensity is very complicated. for the sake of the capability of pnn for pattern recognition, we attempt to predict the rock burst activity by using pnn. four degrees of rock burst activity, including none rock burst, light rock burst, moderate rock burst and strong rock burst are indicated by 1, 2, 3 and 4, respectively. pnn model and criterion are obtained through training data sets of rock burst samples which come from underground rock projects in domestic and abroad. the training effect and the training error are shown in figure 3. it is noted from figure 3 that the misjudgment ratios of training samples using pnn model is 0, which prove that the pnn has a good learning performance. figure 4 is the perdition results of testing samples. from the figure 4, we can find that the prediction accuracy of pnn model is 100%. the results show that the prediction results agree well with the practical records, which prove the pnn-based rock burst model is useful and available and can be applied to the prediction for the possibility and classification of rock burst in underground engineering. table 3. collected samples of rock burst. table 4. testing samples. figure 3. the training of pnn. 145pnn-based rock burst prediction model and its applications figure 4. the perdition results of testing samples. the results of the pnn-based method are compared with that svmbased method, bp-based method, and lvq-based method. the calculated results of pnn, svm, bp and lvq are listed in table 3. from table 3, we can find that the misjudgment ratios of tested samples using svm, bp, lvq, and pnn are 10%, 20%, 20% and 0, respectively. the compared predicted results show that it is feasible and appropriate to use pnn model for rock burst prediction. table 5. comparison of calculation results by different methods. to study the effectiveness and feasibility in engineering practice applications, two real-world examples are analyzed by using our posed pnnbased rock burst prediction method. case 1: tongyu tunnel engineering tongyu tunnel is currently one of the most deep-lying and longest tunnels in chongqing, china. its geological conditions are incredibly complex. the measured data of rock burst in depth 900m at a cross section of k21+680 of tongyu tunnel are listed in table 6 (he et al., 2008). applying our proposed prediction model to rock burst prediction of this engineering, the result of prediction is light rock burst activity. the results agree well with the practical records. table 6. prediction of rock burst in depth of 900 m at the cross section of k21+680 of tangyu tunnel. case 2: qinnling tunnel engineering the qingling tunnel is the longest railway tunnel in china and takes the third place in the world at present. reference (thaldiri et al., 2017; li and wang, 2009) provides rock burst measured data for of qingling tunnel engineering. some measured data listed in table 7. table 7 compared the performance of the proposed method with existing bp-based method (bai et al., 2002) and unascertained measurement method(shi et al., 2010). from the table 7, the result of prediction is same as the results of existing bpbased method and unascertained measurement method, just same as what happened in the scene. this case further confirms that the pnn-based method is effective and practical in the application of prediction of rock burst. on the other hand, pnn-based method has the advantages of low training complexity, high stability, quick convergence, and simple construction. table 7. a case of qinling tunnel is analyzed by using our proposed method. 4. discussion in this paper, a novel pnn-based rock burst prediction model is proposed to determine whether rock burst will happen in the underground rock projects and how much the intensity of rock burst is. pnn model is obtained through training data sets of rock burst samples which come from underground rock project in domestic and abroad. other samples are tested with the model. the testing results agree with the practical records. at the same time, two real-world applications are used to verify the proposed method. the results of prediction are same as the results of existing methods, just same as what happened in the scene, which verifies the effectiveness and applicability of our proposed work. 5. conclusions because pnn has the advantages of low training complexity, high stability, quick convergence, and simple construction, it can be well applied in the prediction of rock burst. in this work, a pnn-based prediction model of rock burst is presented. according to the mechanism of rock burst, rocks’ maximum tangential stress σ ө , rocks’ uniaxial compressive strength σ c , rocks’ uniaxial tensile strength σ t and elastic energy index w et are defined as the criterion indices for rock burst prediction in the proposed pnn-model. some collected rock burst samples which come from underground rock projects in domestic and abroad and two real-world engineering in china are used to verify the new model. the prediction results demonstrated that the developed pnn-based prediction model is effective and efficient approach to predict rock burst potential grade. acknowledgements the authors gratefully acknowledge the project supported by the national natural science foundation of china under grant no. u1504622, the project supported by the scientific research foundation of henan province for returned chinese scholars, china. the authors are grateful for the anonymous reviewers who made constructive comments. references adeli, h. & panakkat, a. (2009). a probabilistic neural network for earthquake magnitude prediction, neural network, 22, 1018-1024. adoko, a. c., gokceoglu, c., wu, l., & zuo, q. j. (2013). knowledgebased and data-driven fuzzy modeling for rockburst prediction. international journal of rock mechanics and mining sciences, 61, 86-95. doi: https://doi.org/10.1016/j.ijrmms.2013.02.010 146 yu zhou, tingling wang ataa, s., hazmi, i. r., & samsudin, s. f. (2017). insect’s visitation on melastoma malabathricum in ukm bangi forest reserve. environment ecosystem science, 1, 20-22. cai, s. j. zhang, l. h., & zhou, w. l. (2005). research on prediction of rock burst in deep hard-rock mines. journal of safety science and technology, 1, 17-20. du, z. j., xu, m. g., & liu, z. p. (2006). laboratory integrated evaluation method for engineering wall rock rock-burst. gold, 11, 26-30. he, z., li, x. h., lu, y. y. (2008). application of bp neural network to the prediction of rockburst in tongyu tunning. chinese journal of underground space and engineering, 4, 494-498. hoek, e., & brown, e. t. (1997). practical estimates of rock mass strength, international journal of rock mechanics and mining sciences, 34, 1165-1186. hou, f. l., liu, x. m., & wang, m. c. (1992). re-analysis of rockburst mechanism and discussion on the gradation of the rockburst intensity. proceedings of the third national conference on rock dynamics. wuhan: wuhan university of mapping technology press, 448-457. jian, z., xibing, l., & xiuzhi, s. (2012). long-term prediction model of rockburst in underground openings using heuristic algorithms and support vector machines. safety science, 50, 629-644. kidybinski, a. (1981). bursting liability indices of coal. journal of rock mechanics and mining sciences, 18, 295-304. li, d. q., & wang, l. g. (2009). theory and technology of the large-scale mining in hard-rock and deep mine-a case study of dongguashan copper mine. beijing: metallurgical industry press. mansurov, v. a. (2001). prediction of rockbursts by analysis of induced seismicity data. international journal of rock mechanics and mining sciences, 38, 893-901. rutkowski l, "adaptive probabilistic neural networks for pattern classification in time-varying environment", ieee trans neural netw, 15 (2004): 811-827. specht, d. f. (1990). probabilistic neural networks. neural networks, 3, 109-118. song, t., jamshidi, m. m., lee, r. r. & huang, m. (2007). a modified probabilistic neural network for partial volume segmentation in brain mr image. ieee transactions on neural networks, 18, 1424-1432. tang, b. y. (2000). rockburst control using distress blasting. ph.d. dissertation. mcgill university. tao, z. y. (1988). support design of tunnels subjected to rockbursting. in: romana (ed.), isrm international symposium, rock mechanics and power plants, 407-411. thaldiri, n. h., hanafiah, m. h., & halim, a. a. (2017). effect of modified micro-sand, poly-aluminium chloride and cationic polymer on coagulation-flocculation process of landfill leachate. environment ecosystem science, 1, 17-19. turchaninov, i. a., markov, g. a., gzovsky, m.v., kazikayev, d. m., frenze, u. k., batugin, s. a., & chabdarova, u. i. (1972). state of stress in the upper part of the earth’s crust based on direct measurements in mines and on tectonophysical and seismological studies. physics of the earth and planetary interiors, 6, 229-234. doi: https://doi.org/10.1016/0031-9201(72)90005-2 wang, x. z., aamir, r. & ai-min, f. (2015). fuzziness based sample categorization for classifier performance improvement. journal of intelligent & fuzzy systems, 29, 1185-1196. wang, j., zeng, y., xu, y., & feng, k. (2017). analysis of the influence of tunnel portal section construction on slope stability. geology, ecology, and landscapes, 1, 56-65. wang, y. h., li, w.d., & li, q. g., (1998). method of fuzzy comprehensive evaluations for rockburst prediction. chinese journal of rock mechanics and engineering, 15, 493-501. wu, d. x. & yang, j. (2005). prediction and countermeasure for rockburst in cangling mountain highway tunnel. chinese journal of rock mechanics and engineering, 24, 3965-3971. yang, t., li, g. w. w. (2000). study on rockburst prediction method based on the prior knowledge. chinese journal of rock mechanics and engineering, 19, 429-431. zhang, x. z. (2005). prediction of rock burst at underground works based on artificial neural network. yangtze river, 36, 17-18. zhang, z. y., song, j. b., & li, p. f. "rock burst comprehensive forecasting method for the chamber group of underground power house", advance in earth sciences, 19 (2004): 451-456 . zhen, s., & gao, w. (2017). geological tourist route planning of henan province based on geological relics zoning. geology, ecology, and landscapes, 1, 66-69. piedrahita and montana.indd earth sci. res. j. vol. 11, no. 1 (june 2007): 35-43 methodology implemented for the 3d-seismic modelling using gocad and norsar 3d software applied to complex areas in the llanos foothills 1 carlos piedrahita. instituto colombiano del petróleo, ecopetrol. piedecuesta colombia. e-mail: carlos.piedrahita@ecopetrol.com.co phone number: (577) 6847104 2 clara l. montaña. instituto colombiano del petróleo, ecopetrol. piedecuesta – colombia. e-mail: clara.montana@ecopetrol.com.co phone number: (577) 6847199 abstract a methodology has been applied in order to use seismic modeling and generate synthetic data which are compared with results from field data. this methodology has been applied to the sirirí area and subsequently it has been extended to other geographic areas. to fully develop this methodology, the gocad and 3d norsar software packages were used to build respectively the geological and 3d seismic models. this methodology involves making structural modeling by edition of the different objects (horizons and faults) in the structural module of gocad, with the help of available seismic and well data, in order to be consistent with reality. from this result a valid seismic model is imported to norsar 3d for making the seismic modeling, using ray tracing to get synthetic data. finally, this information is compared with field data that has been processed. key words: seismic modeling, structural modeling, sirirí, gibraltar. resumen se ha elaborado una metodología con el fin de integrar el modelamiento sísmico y generar datos sintéticos que se comparan con los resultados a partir de datos de campo. esta ha sido aplicada al área de sirirí y posteriormente se ha extendido a otras áreas geográficas. para desarrollarla se utilizó el software gocad y norsar 3d en la construcción de modelos y el modelamiento sísmico 3d, respectivamente. esta metodología consiste en la elaboración del modelamiento estructural mediante la edición de los diferentes objetos (horizontes y fallas) en el módulo estructural de gocad, ayudados de la información existente de sísmica y pozos para que los datos sean coherentes con la realidad. de este resultado se importa un modelo sísmico válido dentro del norsar 3d, en el que procedemos a hacer el modelamiento sísmico utilizando trazamiento de rayos para obtener datos sintéticos. finalmente esta información es comparada con datos reales que se hayan adquirido y procesado anteriormente. palabras claves: modelamiento sísmico, modelamiento estructural, sirirí, gibraltar. 35 earth sciences research journal manuscript received july 10 2006. accepted for publication may 2 2007. introduction the exploratory activity in the country must add new reservoirs to satisfy the national consumption beyond 2008 and to obtain incomes by exploitation of hydrocarbons to help stabilize the national economy. all the efforts are concentrated in increasing the productivity and the lower risk factor that the new prospects must have. new locations have been proposed and already explorated areas have been reinterpreted. in the same manner, new seismic 2d and 3d programs have bee acquired implementing the latest technology regarding design, acquisition, and processing. seismic modeling has turned into the new option for designing acquisition parameters, being able to simulate the whole seismic experiment before acquisition. this is very convenient for testing many alternative designs, in which the lowest exploratory risk and the best seismic quality are expected, especially in such complex areas like the llanos foothills in colombia, characterized by its abrupt topography and geological complexity. the degree of information available in the foothill area is appropriated to perform seismic modeling tests, because deep interpreted structural-crossings and some previously performed seismic 2d lines are available. this methodology was implemented, in this case, on the siriri block located in the area known as the sarare region between arauca, boyacá, and norte de santander departments, specifically in the gibraltar structure (figure. 1). in a regional and geological context, the siriri block is located in the folding and the thrust fault area of the foothills of the colombian eastern mountain range. the current morphology and geology is the result of several phases of changes in the tectonic style varying from extensive to compressive. the siriri block is an area affected by the subduction of the nazca plate in the west and the caribbean plate in the north. the mirador formation’s quartz sands are the main reservoir in the area. rocks in the structural section, ranging from quaternary to cretaceous (campanian), are found in faulting areas, in which parts of the paleocene and cretaceous sections are repeated. the main generator rocks in the region are shale and clay of the la luna formation (cretaceous and turonian). the tertiary section is considered as having some possibilities as a generator rock. methodology figure 2 is showing a flow-chart of the methodology followed to realize the modeling a structural model was realized based on the knowledge of the area and the available seismic data using the time interpretation on the software landmark’s seiswork 2d. figure 3 shows the gibraltar structural model, which was imported into paradigm’s geosec 2d geophysical software, to be converted from time to depth. the exploratory objective was to illuminate the gibraltar structure in the mirador formation. initially, the horizons and interpreted faults were imported and associated to the different thin skins of the mirador formation, in the earth science decisions’ gocad software. gocad is one of the latest state-of-the-art computational integrated applications, in which advanced capacities for 3d modeling, visualization, and geoanalysis are offered. this software has a huge advantage in the integration of information from external sources, and possesses the capacity to export into other applications such as the norsar’s norsar 3d software, where seismic modeling is realized (earth science decisions, 2005). the construction of the 3d model begins by importing the different horizons and faults, editing all the surfaces, assigning the category of fault to each corresponding fault and horizon, and using the algorithms of gocad´s structural software module to obtain the correct relationships between the fault-horizon intersections and the skips of the fault. the final product is a valid three-dimensional model, which will be exported to norsar 3d software (figure 4). norsar 3d is a software for seismic modeling in three dimensions. the method used for modeling is the ray tracing to wave front construction. this algorithm was developed by norsar´s researchers (vinge et al., 1993). methodology implemented for the 3d-seismic modelling using gocad and norsar 3d software applied to complex areas in the llanos foothills 36 piedrahita and montaña, esrj vol. 11, no. 1. june 2007 37 figure 1. location of the area of study containing some seismic lines acquired in the area. figure 2. flow chart that summarizes the methodology followed to realize the seismic modeling 38 figure 3. gibraltar structural model (surfaces m1 to m7 correspond to different thin skin of the mirador formation; and f1-f3, fd1, fd2, fa, ft and fg are the different fault associated to el mirador.). (ecopetrol, 2003). figure 4. 3d model made in gocad and visualized in norsar 3d. the m2 and m4 surfaces correspond to the different scales of the mirador formation. methodology implemented for the 3d-seismic modelling using gocad and norsar 3d software applied to complex areas in the llanos foothills 39 it has the capacity of exchanging data models with other software and importing or exporting in different formats (norsar, 2003a). a description of the ray tracing method is presented by lecompte et al. (2004). the first step to start the modeling is to define the working environment (e.g., choosing the project’s name and defining its extension and coordinate system). the 3d model constructed with gocad is imported to proceed with the geometrical modeling, in which, the blocks are constructed taking into account that each surface or interface has a positive side, specified in this case for the direction of the basement, and a negative side (atmosphere). the system performs this procedure automatically, meaning that false blocks will be generated, making necessary to redraft the model to adapt it to the reality another important step in the validation of the model in norsar 3d is the assignment of the properties of the rocks such as the velocity. these velocities were taken from the vertical seismic profile (vsp) of the gibraltar-1 well and from geological interpretation as well. this allows the model to become suitable for the ray tracing by the wave front method. with a valid seismic model for the ray tracing, it proceeds to generate the seismic design for each one of the lines of the sarare seismic program that passes near the exploratory objective (six 2d lines in total). in the application of the common shot wave front tracer of norsar 3d, all the shots of the design were modeled for each line, obtaining events as a result (databases of all the propagations in the common shot domain), which register the attributes of each ray traced. the common shot wave front tracer is based on a concept developed by norsar called wave front construction. this modeling method is based on ray tracing to interpolate a wave front in intervals of a given time. the purpose of this method is to simulate the seismic wave field and placing a number of shots on the surface, with a certain number of receivers associated to each shot. to execute this tool, three elements were used: (1) a valid seismic model generated in norsar’s model building tool, which should be validated and have assigned properties such as the velocity and the density; (2) a survey, designed from the seismic parameters used, and (3) the ray code, which specifies the trajectory of the rays through the model (norsar, 2003b). the tracer was configured to simulate the wave propagation of the mirador horizon (m4 and m2 mainly) and the deeper (figure 5). this configuration is called signature and usually uses the velocities of the p waves. only waves reflected by this horizon are used. results the final product of the seismic modeling was the generation of illumination maps and synthetic seismograms. the illumination maps evaluate the design of seismic acquisition on the horizon of interest (e.g., the mirador formation). these maps are built in the norsar 3d using the events registered in the seismic modeling. several functions can be visualized on the illumination maps like the hitmat function, equivalent to the seismic fold, which shows the effectiveness of the seismic design according to the parameters required (figure 6). another of the final products of the seismic modeling is the generation of synthetic seismograms. the result of the wave front tracer is stored in files corresponding to events. a trace type and the frequency can be assigned to these events such as a wavelet (ricker of 30 hz). the software generates two types of seg-y files, one internal, for display locally in the project and another external that can exported to landmark´s promax 2d software for processing. a subsoil image is obtained, which can be compared with acquisitions already processed or to propose new seismic prospects (norsar, 2003b). figure 7 shows a comparison among real data (sarare 2004 seismic program) and the synthetic data in the seismic lines on the sirirí block. it can be observed that in the seismic line with real data there is no continuity on the reflectors, being notorious in the right side of the seismic section. this deficiency in the quality of the data is product of the structural complexity presented in the llanos foothills. this is one of the most important uses of this methodology for complex piedrahita and montaña, esrj vol. 11, no. 1. june 2007 40 figure 5a. shot 72 (143 in total) on the horizon m4 of the mirador formation at seismic line sa-95-22. figure 5b. time vs distance graphic (travel time) shot 72. the event indicated with the arrow represents the m4 horizon. methodology implemented for the 3d-seismic modelling using gocad and norsar 3d software applied to complex areas in the llanos foothills 41 areas because some information of the reflectors in the synthetic lines is more notorious, and information of the field data can be enhanced, which is not visible in the field data by itself, because it can be mixed with many types of events. the seismic modeling permits to simulate the effects of the illumination for lines previously acquired and for new lines to be acquired as well. these illumination studies will be utilized for the design of future 3d projects to be executed in the siriri block. conclusions seismic modeling using the methodology described is an excellent tool for the validation of the 2d or 3d seismic designs and it is adequate to use it on complex areas, as the colombian llanos foothills. the complexity of this area makes necessary to use this method. an example of this is observed on the comparison between the images obtained from field data and the ones obtained from synthetic data (figure 7). ray tracing is a tool of great computational efficiency, a very important aspect to be considered, because of the great deal of data that has to be calculated, especially 3d data. the poor knowledge of the seismic velocities on the deep structures is one of the key issues that have to be improved to use the model without limitations. the different variations on the seismic design offer a lot of possibilities of experimentation like testing the illumination of the horizons of interest, especially on geologically complex areas such as the ones located on the colombian foothills (alvarez et al., 2004). figure 6. the illumination map is showing the hitmap function in the mirador horizon in the seismic line sa-95-22. piedrahita and montaña, esrj vol. 11, no. 1. june 2007 42 f ig ur e 7. r ea l d at a (l ef t) a nd s yn th et ic d at a (r ig ht ) c or re sp on di ng to th e lin e sa -9 522 o f t he s ir ir í a re a. methodology implemented for the 3d-seismic modelling using gocad and norsar 3d software applied to complex areas in the llanos foothills 43 acknowledgments the authors are deeply grateful with the research division (uin) of the instituto colombiano de petroleo (icp), for the opportunity of having access to the different types of tools and data required on the research. references alvarez, g., pereira, v., and carcione, l. (2004). model-based 3-d seismic survey design as an optimization problem. seg int´l exposition and 74th annual meeting. denver, colorado. earth sciences decision (2005). introducción al módulo base de gocad. houston. lecompte, i., gjoystdal, h., and drotining, a. (2004). efficient and flexible seismic modeling of reservoirs: a hybrid approach. the leading edge. 23 (5). norsar (2003a). norsar 3d ray modeling. user’s guide version 4.0. kjeller, norway. norsar (2003b). norsar 3d ray modeling. interactive ray modeling by norsar-3d, version 3.0.2. course manual. kjeller, norway. piedrahita, c. and salinas, t. (2005). 3d seismic modeling on the sirirí block, colombia: case history. memorias del congreso colombiano del petróleo. acipet. bogotá-colombia. vinje, v; iversen, e., and håvar, g. (1993). traveltime and amplitude estimation using wavefront construction. geophysics. 58 (8). 1157-1166. piedrahita and montaña, esrj vol. 11, no. 1. june 2007 in this study, site-specific earthquake spectra for bitlis province in lake van basin has been obtained. it is noteworthy that, in probabilistic seismic hazard assessment, as a first stage data from geological studies and records from the instrumental period were compiled to make a seismic source characterization for the study region. the probabilistic seismic hazard curves for bitlis were developed based on selected appropriate attenuation relationships, at rock sites, with a probability of exceedance 2%, 10% and 50% in 50 year periods. the obtained results were compared with the spectral responses proposed for seismic evaluation and retrofit of the building structure in turkish earthquake code, section 2. at the end of this study, it is apprehended that the code proposed earthquake response spectra are not sufficient for the performance evaluation of the existing structures and the current estimations show that the potential seismic hazard research of the turkey is underestimated in the code. therefore, sitespecific design spectra for the region should be developed, which reflect the characteristics of local sites. en este estudio se obtuvieron espectros de terremoto de sitio específico para la cuenca del lago de van, en la provincia de bitlis, al este de turquía. la primera fase del trabajo consistió en una evaluación probabilística de riesgo sísmico donde se compilaron los estudios geológicos y registros del período instrumental para hacer una caracterización de fuente sísmica en la región de estudio. las curvas de amenaza sísmica para la provincia de bitlis se desarrollaron con base en las relaciones de atenuación apropiada seleccionadas en los sitios rocosos, con una probabilidad de exceso de 2 %, 10 % y 50 % durante 50 años. los resultados obtenidos se compararon con las respuestas de espectro propuestas para la evaluación sísmica y modernización de estructuras contempladas en el código de terremoto de turquía, en la sección 2. en la parte final de este trabajo se comprende que las respuestas de espectros de terremoto propuestos en el código no son suficientes para la evaluación de desempeño de las estructuras existentes y que las estimaciones actuales muestran que la investigación de amenazas potenciales sísmicas en turquía está subestimada en el código. por lo tanto, el diseño de espectros de sitio específico para la región se debe desarrollar, ya que permitiría conocer las singularidades locales. abstract resumen keywords: site-specific, earthquake spectra, site response analysis, codes, probabilistic. palabras clave: sitio específico, espectros de terremoto, análisis de respuesta de sitio, códigos, probabilística. record manuscript received: 12/04/2015 accepted for publication: 17/06/2015 how to cite item isik, e. and kutanis, m. (2015). determination of local site-specific spectra using probabilistic seismic hazard analysis for bitlis province, turkey. earth sciences research journal, 19(2), 129-134. http://dx.doi.org/10.15446/esrj.v19n2.50101 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n2.50101 determinación de espectros de sitio específico locales a través del análisis probabilístico de amenazas sísmicas para la provincia de bitlis, turquía earth sciences research journal earth sci. res. j. vol. 19, no. 2 (december, 2015): 129 134 se is m o l o g y determination of local site-specific spectra using probabilistic seismic hazard analysis for bitlis province, turkey ercan işık1*, mustafa kutanis2 1 bitlis eren university, civil engineering dept. tr-13100 bitlisturkey 2 sakarya university, civil engineering dept. tr-54100 sakaryaturkey * correspondence author: ercan işık, phone:+90 (434) 222 00 97 ext:9701; fax: +90 (434) 222 91 01; e-mail: ercanbitliseren@gmail.com 130 ercan işık, mustafa kutanis introduction seismic hazard analysis of the earthquake-prone eastern anatolia region of turkey has become more important due to its growing strategic importance as a global energy corridor and closer integration with the european union. in this study, bitlis province is selected as the study area. the town of bitlis has a population of 70,000 (including the surroundings) as of the year 2000. the town is located 15 km from lake van along the steep slopes of the bitlis river valley at an elevation of 1,400m. the seismicity of bitlis has been evaluated using a performancebased earthquake engineering approach in this study. performance-based earthquake engineering seeks to improve seismic risk decision-making through assessment and design methods that have a strong scientific basis and that reveal options in terms that enable stakeholders to make informed decisions. given the inherent uncertainty and variability in seismic response, it follows that a performance-based methodology should be formalized on a probabilistic basis. the framework has four main analysis steps: hazard analysis, structural/nonstructural analysis, damage analysis, and loss analysis. the first assessment step entails a hazard analysis, through which one evaluates one or more ground motion intensity measures (im). for standard earthquake intensity measures (such as peak ground acceleration or spectral acceleration) is obtained through conventional probabilistic seismic hazard analysis. typically, im is described as a mean annual probability of exceedance, which is specific to the location and design characteristics of the facility (moehle and deirlein, 2004). in performance-based design and assessment method, it is possible to determine in quantities the damage levels that may arise from the design ground motion within the system structural system elements. it is checked whether this damage stays under the acceptable damage levels for each related component. acceptable damage limits are defined in a way to be consistent with the foreseen performance targets at various earthquake levels (aydınoğlu, 2007; doran et al. 2011; kutanis and boru, 2014). sitespecific design spectra for the region have great importance to determined building’s performance under an earthquake hazard. according to section 2 of the turkish earthquake code (tec’07), the demand spectra used to determine seismic performance of an existing building based on a probability of exceedance of 10% in 50 years (fig 1). figure 1. the earthquake spectrum curves for different exceedance probability the assessment procedure aims to estimate the earthquake force demand at which the building would sustain the performance objectives. demand spectrum, which is used in determining the performance of the building’s system, shows the maximum response that a building gives against seismic activities during an earthquake (i̇lki and celep, 2011). the assessment calculations were done based on a simple technique called the “equivalent displacement rule”. the equivalent displacement approximation is based on the assumption that inelastic spectral displacement is the same as that which would occur if the structure remained perfectly elastic (atc40, 1996). for the flexible structures, where the natural vibrational periods are greater than the corner periods, this rule yields acceptable results. in other cases, particularly in short period (rigid) structures, where the natural vibrational periods are shorter than the corner periods, the displacements obtained from this approximation method might be significantly different from the actual results (fig. 2 and fig. 3). in such cases, elastic spectral displacement is modified by multiplying it by a spectral displacement amplification factor (cr1) to obtain inelastic spectral displacement. figure 2. equivalent displacement rule: flexible structures figure 3. equivalent displacement rule: rigid structures 2. local geology soil conditions change the characteristics of surface seismic response. it is a known fact that this may cause damage to the existing structures built on these grounds (borcherdt, 1990). the lake van basin, which contains bitlis, is located in the region known as the bitlis thrust zone. it is a collapsed tectonic basin which is related to the eastern taurus region (özkaymak et al., 2003). orogenic movements have occurred in the field until the third phase of miocene. volcanic and tectonic events have caused many faults to form, as well as depressions and large lakes in this period (facenna et al., 2006; köse, 2004). metamorphic rock in the region belonging to the bitlis massif include the upper cretaceous ahlatadilcevaz mélange and ahlat conglomerate, miocene adilcevaz limestone, pliocene-quaternary volcanic rocks and alluvial deposits from the surface in bitlis and surrounding region (report 1, report 2). a geological map of lake van basin is shown in figure 4. ki spectral displacement, sd sp ec tra l a cc el er at io n, s a capacity curve elastic spectrum with 5% damping sde sdi d(p) sae where d(p), target displacement sdi , inelastic displacement sde , elastic displacement ki, initial stiffnes cr, spectral amplification factor (cr=1.0) d(p)=sdi=1.0·sde ki spectral displacement, sd sp ec tra l a cc ele ra tio n, s a elastic spectrum with 5% damping sde sdi d(p) where d(p), target displacement sdi , inelastic displacement sde , elastic displacement ki, initial stiffnes cr, spectral amplification factor (cr>1.0) d(p)=sdi=cr·sde sae sde capacity curve 131determination of local site-specific spectra using probabilistic seismic hazard analysis for bitlis province, turkey figure 4. geological map of the lake van region. n – nemrut volcano, s – süphan volcano near the lake (litt et al., 2009). 3. tectonic setting and seismicity of bitlis and surrounding areas the general tectonic setting of eastern anatolia is mainly controlled by the collision of the northerly movable arabian plate against anatolian plate along a deformation zone known as bitlis thrust zone (fig. 5). the collision leads to the westward extrusion of the anatolian plate along the two notorious transform faults with a different sense of slip, the dextral north anatolian fault, and the sinistral east anatolian fault zones, which join each other in karlıova triple junction (ktj) in the eastern anatolia (fig. 5). in the eastern side of ktj: however, the collision deformation is largely accommodated within the eastern anatolian block through distributed nwse trending dextral faults and ne–sw trending sinistral faults representing escape tectonics, and shortening of the continental lithosphere along the caucasus thrust zone. east–west trending mush-lake van and pasinler ramp basins constitute other conspicuous tectonic properties within the eastern anatolian border (şengör et al. 1985; barka and kadinsky, 1988; mc clusky et al. 2000; reilinger et al. 2006; utkucu et al. 2013). the east anatolian fault zone is a 550 km-long, approximately northeast-trending, sinistral strike-slip fault zone (fig. 5) that comprises a series of faults arranged parallelly, sub-parallelly or obliquely to the general trend. the bitlis suture is a complex continent-continent and continent-ocean collisional boundary that lies north of fold-and-thrust belt of the arabian platform and extends from south-eastern turkey to the zagros mountains in iran (homke, 2007; bonnin et al. 1988; piper et al. 2008; stern et al. 2008 and lyberis et al. 1992). the area to the east of karlıova triple junction is characterized by an n–s compressional tectonic regime (fig. 6). conjugate strike-slip faults of dextral and sinistral character paralleling to north and east anatolian fault zones are the dominant structural elements of the region. some of these structures include ağrı fault, bulanık fault, çaldıran fault, erciş¸ fault, horasan fault, iğdır fault, malazgirt fault, süphan fault, balıklıgölü fault zone, başkale fault, çobandede fault zone, dumlu fault zone, hasan timur fault zone, kavakbaşı fault, kağızman fault zone, doğubayazıt fault zone, karayazı fault, tutak fault zone, yüksekova–şemdinli fault zone and the northeast anatolian fault zone (fig. 6) (bozkurt, 2001). aydemir et al. (2014) investigated the faults and possible structural elements that may have caused the devastating earthquake that occurred on october 23, 2011. this potential extended fault zone starting from the nemrut mountain in the west may exist through to the east of lake van (aydemir et al. 2014). the faults are seismically active and form the source for many earthquakes. some of the major earthquakes in the 20th century are 13 september 1924 pasinler (m = 6.8), 1975 lice (m = 6.6), 24 november 1976 çaldıran (m = 7.3), 1988 van (m= 5.0), 2000 van (m=5.7), 23 october 2011 van (m=7.2) and 09 november 2011 van (m=5.6) earthquakes. figure 5. tectonic map of turkey including major structural features (from bozkurt, 2001) figure 6. active faults of eastern anatolian province (bozkurt, 2001). the examination of historical and instrumental earthquakes in bitlis and its surroundings proves that this region is constantly under the influence of micro and macro earthquakes. thus, bitlis remains under an enormous influence of micro and macro earthquakes (işık et al., 2012). therefore, its buildings must be constructed, especially in lake van basin where earthquake resistant design has always been neglected, according to earthquake codes. 4. seismicity parameters on any given fault within any particular region, earthquakes occur at irregular intervals in time, and one of the primary activities in seismology has long been the search for meaningful patterns in the time sequences of earthquake occurrence (dowrick, 2003). among some recurrence laws have been proposed, in this study, gutenberg and richter’s law was used because there is no available evidence to determine whether the gutenberg –richter or some other recurrence laws are correct. during any given interval in time, the general underlying pattern or distribution of size of events is that first described by gutenberg and richter (1944), who derived an empirical relationship between magnitude and frequency of the form log n = a − b.m (1) where n is the number of shocks of magnitude at least m per unit time and unit area, and a and b are seismic constants for any given region (dowrick, 2003). in a seismic hazard modeling study of bitlis, recurrence rates are 132 ercan işık, mustafa kutanis estimated by using historical and digital records. after the compilation of collected data, a plot of “m” against “log n” was constructed and the bestfit line of the form of eqn. 1 was determined by regression analysis (fig. 7). in probabilistic seismic hazard analysis, beside magnitude–frequency relationship that is calculated for bitlis province as logn = 5.6247 – 0.7794 m. figure 7. gutenberg-richter magnitude–frequency relationship for earthquakes from bitlis and surrounding data 5. site-specific design spectra for bitlis province the seismic hazard analysis approach is based on the model developed originally by cornell (1968) who quantified it regarding the probability of exceedance of the design level peak ground acceleration (pga). the procedure for conducting a probabilistic seismic hazard analysis includes seismic source characterization, size distribution and rate of occurrence determination for the source, ground motion estimation and, lastly, probability analysis. in the current study, since the neotectonic faults are not identified in the research area clearly, earthquake sources are characterized as area source zones. area seismic sources are often defined where specific fault data are not known, but seismicity does exist. area sources assume that the rate of occurrence is uniform throughout. therefore, every location within the area has equal probability that an event will occur (ez-frisk; anton and gibson, 2008). all seismic sources, that can generate strong ground shaking in bitlis and surroundings, are classified into 7 areal seismic zones (fig. 8): (1) bitlis-zagros suture zone; (2) northern bitlis thrust fault zone (dhont and chorowicz, 2006); (3) kavakbaşı fault zone; (4) malazgirt fault zone; (5) ahlat and surrounding fault zone; (6) suphan fault zone ; and (7) southern van faults (erçek fault, kalecik fault, edremit fault and southern boundary fault (utkucu, 2006). figure 8. earthquake areal zones (bitlis suture, van south, bitlis north suture, kavakbaşı, ahlat, malazgirt and suphan) in bitlis and surroundings in eastern anatolia region, previously recorded strong ground motion acceleration records are limited. therefore, in the current analysis, worldwide applicable three empirical attenuation relationships are utilized to perform the seismic hazard analysis. attenuation relationships for rock sites employed in this study are abrahamson-silva (1997), ambraseys et al. (2005), boore et al. (1997), campell (2003) and idris (2008) (fig. 9). figure 9. abrahamson-silva (1997), ambraseys et al. (2005), boore et al. (1997), campell (2003) and idriss (2008) attenuation relationships for rock sites. after the compilation of the seismic hazard analysis data, the procedure for conducting a probabilistic seismic hazard analysis, by using ez-frisk (mcguire, 1995). the software was employed to produce the pga as a function of return periods (fig. 10) and uniform probability response spectra for selected return periods (fig. 11). the results of probabilistic seismic hazard analysis for bitlis are presented regarding spectral responses at 5% damping for the return periods of 72, 474.6 and 2474.9 years (fig. 11). the results are compared with the spectral responses proposed for seismic evaluation and retrofit of building structure in turkey earthquake code (2007) section 7. figure 10. peak ground acceleration (pga) at bitlis with varying return periods. 133determination of local site-specific spectra using probabilistic seismic hazard analysis for bitlis province, turkey figure 11. spectral responses at 5% damping for the return period of 72, 474.6 and 2474.9 years the results of probabilistic seismic hazard analysis revealed peak acceleration values for a typical rock site as 0.76g for 50% probability of exceedance in 50 years, 1.61g for 10% probability of exceedance in 50 years and 2.68g for 2% probability of exceedance in 50 years. the obtained results are compared with the spectral responses proposed for seismic evaluation and retrofit of building structure in turkey earthquake code, section 7 (fig. 12). figure 12a. comparison of spectral responses at 5% damping for the return period of 2474.9 years in bitlis figure 12b. comparison of spectral responses at 5% damping for the return period of 474.6 years in bitlis figure 12c. comparison of spectral responses at 5% damping for the return period of 72 years in bitlis 6. results and conclusions by utilizing available data and the use of improved methods, a probabilistic seismic hazard of bitlis province in turkey was determined. as a first step of the performance-based earthquake engineering, it is well understood that the code proposed spectra are not sufficient to represent earthquake demand in the performance evaluation. the results of this work will form the basis for the replacement of the existing earthquake design spectra in the assessment of earthquake performances of the existing buildings in bitlis province. in this study, since active faults are not identified clearly, regional areas were used as an earthquake source zones. future work will increase the resolution of the seismotectonic model by adding specific active faults. the obtained results are compared with the spectral responses proposed for seismic evaluation and retrofit of building structure in turkish earthquake code, section 7, and the amplitude and frequency range was different from each other (fig.13). using the response spectrum obtained from probabilistic seismic hazard analysis will make obtained data for bitlis and other regions which are under a threat of earthquakes. figure13. comparison of demand displacement of the earthquake therefore, sitespecific design spectra for the region should be developed, which reflect the characteristics of local sites. references abrahamson, n.a. and silva. (1997). empirical response spectral attenuation relations for shallow crustal earthquakes, seismological research letters, 68(1), 94-127. anton, l. and gibson, g. (2008). analysing earthquake hazard in papua new guinea. earthquake engineering in australia, aees2008, ballarat victoria, australia, 21-23. 134 ercan işık, mustafa kutanis ambraseys, n.n., douglas, j., sarma s.k. and smith, p.m. (2005). equations for the estimation of strong ground motions from shallow crustal earthquakes using data from europe and the middle east: horizontal peak ground acceleration and spectral acceleration. bulletin of earthquake engineering, 3(1), 1-53. applied technology council (atc), atc-40 (1996). seismic evaluation and retrofit of concrete buildings, report no. atc-40, seismic safety commission, state of california. aydemir, a., ates, a., bilim, f., büyüksaraç, a., and bektaş, o. (2014). evaluation of gravity and aeromagnetic anomalies for the deep structure and possibility of hydrocarbon potential of the region surrounding lake van, eastern anatolia, turkey, surveys in geophysics, 35(2), 431-448. aydınoğlu, m. n. (2007). a response spectrum-based nonlinear assessment tool for practice: incremental response spectrum analysis (irsa), iset journal of earthquake technology, 44(1), 169-192. barka, a. and kadinsky-cade, k.(1988). strike-slip fault geometry in turkey and its influence on earthquake activity, tectonics, 7(3), 663–684. bonnin, j., cara, m. and cisternas, a. (1988). seismic hazard in mediterranean regions, proceedings of the summer school organized in strasbourg, france. boore, d.m., joyner, w.b. and t.e. fumal (1997). equations for estimating horizontal response spectra and peak acceleration from western north american earthquakes: a summary of recent work, seismological research letters, 68(19), 128-153. borcherdt, r.d. (1990). influence of local geology in the san fransisco bayregion california on ground motions generated 1990, by the loma prieta earthquake of october 17, 1989, proceedings of international symposium on safety of urban life and facilities, tokyo, japan, november 1-2. bozkurt, e. (2001). neotectonics of turkey –a synthesis, geodinamica acta, 14(1), 3-30. cornell, c.a. (1968). engineering seismic hazard analysis, bull. seismil soc. am., 59 (5), 1583-1606. dhont, d. and chorowicz, j. (2006). review of the neotectonics of the eastern turkish – armenian plateau by geomorphic analysis of digital elevation model imagery, int. j. earth sciences ( geol rundcsh), 95, 34-49. doran, b., akbaş, b., sayım, i̇., fahjan, y. and alacalı, s.n. (2011). uzun periyotlu bir yapıda yapısal sağlık izlemesi ve deprem performansının belirlenmesi, 1. turkey conference on earthquake engineering and seismology, ankara, turkey, october. dowrick, d. (2003). earthquake risk reduction, wiley, england. ez-frisk tm, software for earthquake ground motion estimation. http://www.ez-frisk.com/tech/seismichazard/areadb.html online date: 05.11.2015. facenna, c., bellier, o., martinod, j., piromallo, c. and regard, v. (2006). slab detachment beneath eastern anatolia: a possible cause for the formation of the north anatolian fault, earth and planetary science letters, 242, 85-97. homke, s. (2007). timing of shortening and uplift of the pusht-e kuh arc in the zagros fold and thrust belt (iran); a combined magnetostratigraphy and apatite thermochronolgy analysis, universitat de barcelona, facultad de geologia, departamento de geodinámica y geofísica, 213p., barcelona, spain. gutenberg, b. and richter, c.f. (1944). frequency of earthquakes in california, bulletin of seismology society america, 34,185-188. idriss, i.m. (2008). an nga empirical model for estimating the horizontal spectral values generated by shallow crustal earthquake, earthquake spectra, 24(1), 217-242. ilki, a. and celep, z. (2011). betonarme yapıların deprem güvenliği, 1. türkiye deprem mühendisliği ve sismoloji konferansı, ankara, turkey, october işık, e., aydın, m.c., bakış, a. and özlük, m.h. (2012). the faults near bitlis and seismicity of the region, bitlis eren university, beu journal of science, 1(2), 153-169. köse, o. (2004). van gölü yakın çevresinin coğrafyası, van gölü havzası jeotraversleri çalıştay kitapçığı, dajeo-2004, 1-6. kutanis, m. and boru, o.e. (2014). the need for upgrading the seismic performance objectives, earthquakes and structures, 7(4), 401-414. litt, t., krastel, s., sturm, m., kipfer, r., örçen, s., heumann, g., franz, s.o., ülgen u.b. and niessen f. (2009). paleovan, internatıonal continental scientific drilling program (icdp): site survey results and perspectives, quaternary science reviews 28 (2009) 1555–1567. lyberis, n., yürür, t., chrowicz, j., kasapoğlu, e. and gündoğdu, n. (1992). the east anatolian fault: an oblique collisional belt, tectonophysics, 204(1), 1-15. mcclusky, s., balassanian, s., barka, a., demir, c., ergintav, s., georgiev, i., gurkan, o., hamburger, m., hurst, k., kahle, h., kastens, k., nadariya, m., ouzouni, a., paradissis, d., peter, y., prilepin, m., reilinger, r., sanli, i., seeger, h., tealeb, a., toksöz, m.n, and veis, g. (2000). gps constraints on plate kinematics and dynamics in the eastern mediterranean and caucasus, journal of geophysical research: solid earth, 105(b3), 5695–5719. mcguire r. (1995). ez-frisk, user’s manual, risk, risk engineering, boulder, co. moehle, j. and deirlein, g.g. (2004). a framework methodology for performance-based earthquake engineering, 13th world conference on earthquake engineering, vancouver, b.c., canada, august 1-6. özkaymak, ç., sağlam, a. and köse, o. (2003). van gölü doğusu aktif tektonik özellikleri, atag-7 aktif tektonik araştırma grubu 7. toplantısı bildiri özleri, yüzüncü yıl üniversitesi jeoloji mühendisliği bölümü, van , 22-23. piper j., tatar, o., gürsoy, h., mesci, l., koçbulut, f. and huang, b. (2008). post-collisional deformation of the anatolides and motion of the arabian indenter: a paleomagnetic analysis, iop publishing, donald d harrington symposium on the geology of the aegean, iop conf. series: earth and environmental science 2. reilinger, r., mcclusky, s., vernant p., lawrence, s., ergintav, s., cakmak, r., ozener, h., kadirov, f., guliev, i., stepanyan, r., nadariya, m., hahubia, g., mahmoud, s., sakr, k., arrajehi, a., paradissis, d., al-aydrus, a., prilepin, m., guseva, t., evren, e., dmitrotsa, a., filikov, s.v., gomez, f., al-ghazzi, r. and karam, g. (2006). gps constraints on continental deformation in the africa-arabia-eurasia continental collision zone and implications for the dynamics of plate interactions, journal of geophysical research: solid earth, 111(b5). report 1, jeo massif (2003). bitlis i̇li merkez ilçesi değişik parselleri için jeoteknik raporları, jeo-masif enjeksiyon ankraj i̇nş. ve taah. ltd.şti. (in turkish). report 2, (2006). bitlis rahva 2. bölge 272 adet konut adaiçi ve genel altyapı ileçevre düzenlemesi inşaatı alanının zemin ve temel etüdü raporu, t.c.başbakanlık toplu konut i̇daresi başkanlığı, (in turkish). sengör, a.m.c., görür, n. and saroglu, f. (1985). strike-slip deformation, basin formation and sedimentation: strike-slip faulting and related basin formation in zones of tectonic escape: turkey as a case study, society of economic paleontologists and mineralogist, special publication,, 37, 227–264. stern, r.j. and johnson, p.r. (2008). do variations in arabian plate lithospheric structure control deformation in the arabian-eurasian convergence zone, iop publishing, donald d harrington symposium on the geology of the aegean, iop conf. series: earth and environmental science 2. turkish earthquake code (2007), turkish earthquake code-specification for structures to be built in disaster areas, turkey utkucu, m., durmuş, h., yalçın, h., budakoğlu, e. and işık, e. (2013). coulomb static stress changes before and after the 23 october 2011 van, eastern turkey, earthquake (mw = 7.1): implications for the earthquake hazard mitigation, natural hazards and earth system science, 13(7), 1889–1902. utkucu, m. (2006). implications for the level change triggered moderate ( m≥4.0) eartquakes in lake van basin, eastern turkey, journal of seismology, 10, 105-117. this paper proposes the concept of variable-permeability effect and sets up the one-dimensional and two-dimensional non-darcy well testing models. the finite difference algorithm is employed to solve the differential equations of the variable-permeability model, and the non-convergence of the numerical solutions is solved by using the geometric mean of permeability. the type curves of pressure and pressure derivative with variable-permeability effect are obtained, and sensitivity analysis is conducted. the results show that the type curves upturn in the middle and late sections, and the curves turn more upward with the severer of the variable-permeability effect. the severer the non-darcy effect is, the less obviously the curve upturns caused by boundary effect. furthermore, the boundary effect is increased by increasing the number of impermeable boundaries or decreasing the distance between the well and boundary. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): j1 j6 abstract keywords: variable-permeability effect, lowpermeability reservoirs, non-darcy flow, well testing, pressure gradient, boundary effect. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.54144 h y d r o l o g y naichao feng1, shiqing cheng1*, weixing lan2, guoquan mu3, yao peng4, haiyang yu1 1. moe key laboratory of petroleum engineering china university of petroleum beijing beijing 102249, china 2. e&d research institute of liao he oil field company, panjin, liaoning 124010, china 3. e&d institute of changqing oil company, xi’an, shanxi 710021, china 4. department of petroleum engineering, university of texas at austin, austin 78712, usa *corresponding author shiqing cheng. e-mail: chengsq973@163.com variable-permeability well-testing models and pressure response in low-permeability reservoirs with non-darcy flow record manuscript received: 12/11/2015 accepted for publication: 05/02/2016 how to cite item feng, n., cheng, s., lan, w., mu, g., peng, y., & yu, h. (2016). variable-permeability well-testing models and pressure response in low-permeability reservoirs with non-darcy flow. earth sciences research journal, 20(1). j1 j6. doi: http://dx.doi.org/10.15446/esrj.v20n1.54144 1. introduction to date, darcy’s law (darcy, 1856) is fundamentally used in petroleum engineering, soil science, underground hydrology and hydrotechnies to mathematically describe the flow through porous media. for decades, the classic percolation theory based on darcy’s law has been developing and becomes an important branch of fluid mechanics. however, limitations of darcy’s law were found both from the laboratory experiments and also reported in field observations. many investigators have reported the presence of a threshold gradient for certain reservoir formations, so that there is no flow through porous media when the pressure gradient is below this value, and the relationship between flow rate and pressure gradient was found to be approximately a straight line that does not pass through the origin. when water flows through porous media under small pressure gradient, the connate water will not flow in narrow pores and will block the flow of free water in the adjacent bigger pore. the block of the connate water will be removed, and the free water will become to flow, only when the pressure gradient increased to a critical value, that is, threshold pressure gradient. the threshold pressure gradient effect in the non-steady flow through a porous media was investigated, and the effect on the pressure and flow rate distributions in a flow system was evaluated in oil reservoir (pascal, 1981). the existence of threshold pressure gradient during fluid transport in a porous medium has also been proved by other researchers (yan et al., 1990; das, 1997), and there is a need for correcting darcy’s law for other conditions. the threshold pressure gradient may be related to the flow boundary layer (huang, 1998). in low-permeability reservoirs, the differences of pore sizes and interfacial forces at solid-liquid interface affect boundary layers, resulting in different threshold pressure gradient. its value is inversely proportional to formation permeability, that is, lower permeability leads to higher threshold pressure gradient. however, the power law correction was obtained by regression of three core flood experiments by using brine (prada and civan, 1999). in 1973, the consolidation problem with initial gradient was investigated, and the approximate solution of the moving boundary model with threshold pressure gradient was obtained (schmidt and westman, 1973; pascal, 1973). then several modified expressions of darcy’s law were developed (huang, 1998), based on the flow rate-pressure gradient curves of core flood experiments. feng and ge (1985, 1988) investigated the low-velocity non-darcy flow mechanisms in single-porosity and dualporosity reservoirs and derived the well testing solution with wellbore storage and skin effect based on regular constant-production model of j2 naichao feng, shiqing cheng, weixing lan, guoquan mu, yao peng and haiyang yu single phase flow. by employing the moving boundary method of the thermal conduction, wu (1990) presented the second order integral solution of bingham fluid flow with threshold pressure gradient. based on the gringarten type curves, cheng and xu (1996) further introduced the solution and type curves of apparent wellbore radius model. many other researchers have studied on well testing problem and got many modified mathematical models, solutions and interpretation methods (li and liu, 1997; warren, 1993; hegeman et al., 1993; liao and lee, 1993; luo and wang, 2011; guo et al., 2005; liu et al., 2014). all the researchers discussed above considered the existence of threshold pressure gradient in low-permeability reservoirs. however, the practical experiments indicate that the flow curves are non-linear at very low velocity, and the existing low-velocity non-darcy model is only an approximation, and it cannot accurately represent the flow behavior in low permeability reservoirs (zhou et al., 2002). so developing a well testing interpretation model and the analytical method, by actual flow, will improve the accuracy of the well testing analytical interpretations significantly. this paper presents the concept of variable-permeability in the well testing of low-permeability reservoirs; meanwhile, wellbore storage effect and skin effect are taken into consideration. 2. one-dimensional variable-permeability well-testing model 2.1. variable-permeability effect the non-darcy flow process in low permeability reservoir is shown in figure 1. the linear section (above point jn) obeys darcy’s law, and there are mainly three methods to describe the curved section (huang, 1998): (1) simplifying into a straight line passing through the origin with a slope different from the straight line above point jn, which is simple but cannot reflect the flow accurately especially in low permeability reservoirs; (2) power law expression, which is accurate but difficult for mathematical simulation; (3) converting into linear relationship with threshold pressure gradient, reflecting the threshold pressure but resulting in lower velocity in big pores under low pressure gradient condition. the third method, threshold pressure gradient model, is used in most cases in recent years; however, there exist some deviations based on this model. this paper develops a variable-permeability model by dividing the curved section from j1 to jn into many linear sections with different slopes and assuming that darcy’s law can be applied in these linear sections. in this way, the simulation can describe the percolation more accurately, which is close to power law method but easy to calculate. according to the definition of permeability, the slope of the curve at each point stands for the permeability at the corresponding pressure gradient. the permeability rises with the increase of the pressure gradient under a low pressure gradient condition while it maintains constant when the pressure gradient is higher than a critical value. this phenomenon is called the variable-permeability effect. figure 1. percolation curve of low-velocity non-darcy flow in figure 1, the slope at any point stands for the permeability at the correlated pressure gradient. in this way, the relationship between permeability and pressure gradient can be obtained, as shown in figure 2. thus, permeability is converted from a physical meaning into a mathematical meaning, and the flow simulation can be solved by the mathematical method. the processing procedures are as follows: figure 2. permeability curve of low-velocity non-darcy flow (i) through the initial experimental data, plot the flow velocity versus pressure gradient curve v-∆p, then plot the k-∆p curve presenting the relationship between permeability and pressure gradient; (ii) divide the curve into several linear sections, and consider the permeability of each linear section as a constant; (iii) the finite difference algorithm is employed to discretize the nonlinear flow differential equations, and then set up a non-linear equation system in which the pressure is the unknown, and the permeability is a function of pressure gradient, k= k( dp/dr). an alternating iterative method is used to solve the equations: during the iteration process, initialize the permeability first and make it linear, obtain the pressure solution using iteration method to solve the equations. get the pressure gradient in each section, and then obtain the permeability of each section according to the plot of permeability versus pressure gradient. substitute this permeability into the non-linear flow equations and make it linear again. circulate the alternating iterative process until the pressure tends to be stable and satisfies the accuracy requirement. this result is the pressure at the correlated time step, and then go to calculate the next time step. 2.2. well testing model 2.2.1 assumptions (i) a well is located in the center of an infinite, horizontal, homogeneous, isopachous and low-permeability reservoir with constant production rate; (ii) single phase compressible fluid flow in the low-permeability reservoir. the flow does not obey darcy’s law, and variable-permeability effect is taken into consideration; (iii) wellbore storage effect and skin effect are considered; (iv) the effect of gravity and capillary pressure is negligible; (v) the total compressibility is constant. 2.2.2 mathematical model according to the assumptions discussed above, the condition equations are as follows: (1) j3variable-permeability well-testing models and pressure response in low-permeability reservoirs with non-darcy flow state equations: the flow equation was discretized as: where where where inner boundary condition was discretized as: external boundary condition was discretized as:mass conservation equation: substitute equations 1-3 into equation 4: by considering the effects of wellbore storage effect and sk in effect, the correlated initial condition, and boundary conditions are: initial condition: external boundary condition: inner boundary conditions: kinematic equation: equations 5-9 compose the general non-darcy flow well-testing model in the low-permeability reservoir with variable-permeability effect under radial coordinate. 2.3. numerical simulation 2.3.1 model discretization for one-dimensional radial flow, the pressure gradient is higher near the wellbore. so, non-uniform grid, with denser grids near the wellbore, is employed for precise calculation, which can solve the non-convergence problem and decrease iteration time. set r = rwex , time step ∆t = t /(m-1), and (7) (8) (9) (10) (11) (12) (13) (2) (3) (4) (5) (6) implicit difference method is employed to discretize equations 5-9. forward elimination and backward substitution algorithm were employed to solve the non-linear equation system composed by equation 10-13, and pressure distribution and bottom hole pressure (bhp) in reservoirs with time were then obtained. 2.3.2 iterated misconvergence solution during numerical simulation, the approximate solution may not converge and may oscillate near the exact solution. the main reason for the iterated misconvergence is that the non-linear flow curve is not divided into proper sections. at tn, the pressure gradient is ∆pi(tn) , and the correlated permeability obtained is ki+1(tn), then pi+1(tn) and ∆pi+1(tn) can be calculated. after that, calculate the new permeability ki+2(tn) , similarly ki+3(tn), ki+4(tn),..., kn+1(tn) can be calculated. however, some problems are likely to appear: the oscillation of the solution makes it difficult to solve the variablepermeability model, and the geometric mean of permeability is used to solve this problem. calculate the permeability ki+1(tn) using the pressure gradient ∆p i+1(tn) obtained after the ith iteration. (tn) for next iteration is: then keep doing this iteration process until max | km+1 (tn) k (tn)| ≤ ε. 2.4. well testing analysis and interpretation the relationship between flow velocity and the pressure gradient is presented based on three groups of experimental data to investigate the influence of non-darcy effect (fig. 3a), where the non-darcy effect increases gradually in curve 1, 2, 3. figure 3b shows the relationship between the corresponding permeability and pressure gradient. the model discussed above is used to interpret the field test data (table 1), and the type curves are further obtained, as shown in figure 4. table 1. field test data and reservoir parameters 1 ' 1 1 ' 2 ( ) ( ) ( ) ( ) ( ) i i i n n n i i n n k t k t k t k t k t + + + = + j4 naichao feng, shiqing cheng, weixing lan, guoquan mu, yao peng and haiyang yu figure 3. darcy and non-darcy flow curves: flow velocity/permeability vs. pressure gradient figure 4. influence of non-darcy effect on non-darcy type curves in the early section of figure 4, the type curves in log-log scale coincide as a 45o line, indicating the influence of wellbore storage effect. in the transition section, the pressure derivative curves drop down after reaching the peak values. the peak value depends on the value of cde2s. larger value of cde2s results in larger peak value and later appearance of the peak. in the later section, the type curves upturn to various degrees. the longer duration of the variable-permeability effect, the higher the type curves upturn; meanwhile, the pressure derivative curve achieves horizontal for darcy flow since there is no variable-permeability effect. compare the following three kinds of type curves: (i) the gringarten type curves of darcy flow (gringarten et al., 1979); (ii) the type curves with threshold gradient model (cheng and xu, 1996); (iii) the type curves based on the variable-permeability model of this paper, using the data in curve 1 in figure 3a. the comparison of the type curves of these three different models is shown in figure 5. horizontal sections appear in the pressure derivative curves of infinite and homogeneous reservoirs (type i), and this is the feature of typical radial flow. compared with the type curve based on gradient threshold model (type ii), the type curves based on variable-permeability model (type iii) result in gentle upturn in the later section, and its calculation method is more stable than that based on moving boundary model with threshold pressure gradient. figure 5. type curves based on three methods 3. two-dimensional variable-permeability well-testing model to take the boundary effect into consideration, we discuss the solution of the percolation problem in two-dimensional reservoirs. 3.1. well testing model and numerical simulation taking boundary conditions into account, and the two-dimensional well testing model with variable-permeability effect is as follows: control equation: initial condition: inner boundary conditions: (14) (15) (16) (17) external boundary condition with one impermeable boundary: flow control in equation14 was discretized as: where where (18) (19) (20) other boundary conditions corresponding to other forms can be obtained successively: qi,j=q for well point grids, and qi,j=0 for non-well-point grids. consider the grids within wells as sink & source items. since the pressure gradient near the well is large, linearize the non-boundary grids. the inner boundary condition was discretized as: for rectangle grids, and re = 0.208∆x for square grids. j5variable-permeability well-testing models and pressure response in low-permeability reservoirs with non-darcy flow (21) (22) (23) (24) external boundary conditions were discretized as: pressure gradients were discretized as: a non-linear system is formed by equations 19-22. gauss-seidel method was employed to solve the above equations, and pressure distribution and bottom hole pressure in reservoirs with time were further obtained. 3.2. type curves of well testing models with boundaries considering different boundary types, we obtained the type curves based on the model discussed above, as shown in figures 6-8. figure 6 demonstrates the influence of non-darcy effect on type curves in reservoirs with one impermeable boundary and shows that the pressure derivative curve will become horizontal with a value of 0.5 before reaching the impermeable boundary for darcy flow. after reaching the impermeable boundary, the pressure derivative curve initially upturns and then become horizontal again with a value of 1. for non-darcy flow, the pressure derivative curves upturn during radial flow period. the severer the non-darcy effect results in the larger amplitude of the ‘concave’, the earlier the derivative curves upturn, the longer duration of the upturned radial non-darcy section, and the weaker the effect of the boundary (nondarcy effect covers the boundary effect). figure 6. influence of non-darcy effect on type curves with one impermeable boundary the impact of impermeable boundaries with different shapes and distances on type curves are shown in figure 7, including parallel faults, orthogonal faults, three faults, and closed boundaries (the dimensionless distances from the well to the faults all equal 500). the four darcy flow type curves at the bottom of figure 7 reveal that how fast the pressure derivative curves upturn in the later section is related to the number of the faults. if the distances from the well to the faults are equal, the more the boundaries are, the earlier the curves upturn and the larger the amplitudes are. the pressure derivative curves of closed boundary reservoirs rise with a slope unity line in the later section. if there are two faults near the well, the orthogonal faults show the second radial flow plateau and the pressure derivative curve of the parallel faults rise with a 0.5 slope line. the non-darcy flow well testing derivative curves show that pressure derivative curves in the radial flow section upturn due to non-darcy effect, lying above the 0.5 line of darcy flow. in the later section, the pressure derivative curves upturn further due to the impact of the faults, later than the darcy flow. more faults result in larger amplitude upturn. if there are closed boundaries near the well, the derivative curves of darcy flow and non-darcy flow coincide with each other and show a slope unity line. the variable-permeability well testing curves of four different faults reveal that more faults lead to the earlier upturn of the non-darcy well testing curves. the non-darcy pressure derivative curve of orthogonal faults upturns less obviously. therefore, to detect the distance from the well to the faults, the testing time should be appropriately extended when there are fewer faults near the well. figure 7. influence of different boundaries on type curves figure 8 demonstrates the influence of non-darcy effect on type curves in a reservoir with a constant pressure boundary. in the later section, the pressure derivative curves initially upturn and then drop down after the second hump. it drops down faster and faster until the decline rates reach that of darcy flow. the larger the amplitudes of the concave parts are (the more severe the non-darcy effect is), the higher the second humps are and the larger the rate of decline of the derivative is at the end of the curves. j6 naichao feng, shiqing cheng, weixing lan, guoquan mu, yao peng and haiyang yu figure 8. influence of non-darcy effect on type curves with a constant pressure boundary 4. conclusions this paper employs the concept of variable-permeability effect and develops one-dimensional and two-dimensional non-darcy well testing models with variable-permeability and boundaries in low permeability reservoirs. forward elimination and backward substitution algorithm are used to address the difference equation system of the models. the geometric mean of permeability was employed to solve the oscillation of the solution, and the reliable solutions are obtained. the general non-darcy flow well-testing model based on variable-permeability effect and its calculation method are more stable and more practical than those based on threshold pressure gradient. the type curves of well testing with variable-permeability effect are further investigated. in the transition section, the pressure derivative curves drop down after reaching the peak values, and the peak value depends on the value of 2c sde . a larger value of 2c sde results in larger peak value, sharper dropping down rate and later appearance of the peak. in the later section, the type curves upturn. the longer duration of the variablepermeability effect, the higher the type curves upturn. for the well-testing model with one impermeable boundary, the radial flow section upturns again after first upturning under the influence of boundary. the pressure derivative curves upturn along with the increasing of the number of boundaries under the influence of non-darcy effect. when the impermeable boundary is far away from the well, the curves are close to the non-darcy type curves of the infinite reservoir and the boundary effect is not evident. the more the boundaries are, and the less the distance from the well to the boundary is, the earlier the non-darcy curves upturn and the large the amplitudes are. for constant pressure boundary, the fast the derivative curves drop down after the second hump. acknowledgements this work was sponsored by national science and technology major projects (no. 2016zx05009-004, no. 2016zx05013002-005) and national natural science foundation of china (no. 51304223). references cheng, s. q. and xu, l. x. (1996). type curve matching of well test data for non-darcy flow at low velocity (in chinese). petroleum exploration & development, 23(4), 50-53. darcy, h. (1856). les fontaines publiques de la ville de dijon. victor dalmont, 304–311. das, a.k. (1997). generalized darcy’s law including source effect. journal of canadian petroleum technology, 36(6), 57-59. doi: http:// dx.doi.org/10.2118/97-06-06 feng, w. g. and ge, j.l. (1985). non-darcy flow at low velocity through single medium and double media (in chinese). petroleum exploration & development, 1, 56-62. feng, w.g., and ge, j.l. (1988). influence of after flow and skin effect for the non-darcy flow at low velocity through single medium (in chinese). petroleum geology & oilfield development in daqing, 7(2), 45-50. gringarten, a.c., bourdet, d.p., landel, p.a. and kniazeff, v.j. (1979). a comparison between different skin and wellbore storage typecurves for early-time transient analysis. spe annual technical conference and exhibition, society of petroleum engineers, las vegas. doi: http://dx.doi.org/10.2118/8205-ms guo, d.l., zeng, x.h., zhao, j.z. and liu, c. q. (2005). model and method of well test analysis for wells with vertical fracture. applied mathematics and mechanics (english edition), 26(5), 571-578. doi 10.1007/bf02466330 hegeman, p.s., hallford, d.l. and joseph, j.a. (1993). well-test analysis with changing wellbore storage. spe formation evaluation, 8(3), 201-207. huang, y.z. (1998). the mechanism of low permeability reservoir seepage (in chinese). petroleum industry press, beijing, 80-88. li, f.h., and liu, c.q. (1997). pressure transient analysis for unsteady porous flow with start-up pressure derivative (in chinese). well testing, 6(1), 1-11. liao, y.z, and lee, w.j. (1993). new solutions for wells with finite-conductivity fractures including wellbore storage and fracture-face skin. eastern regional conference & exhibition, society of petroleum engineers, pittsburgh. doi: http://dx.doi.org/10.2118/26912-ms liu, y.w., ou-yang, w.p., zhao, p.h., lu, q. and fang, h.j. (2014). numerical well test for well with finite conductivity vertical fracture in coalbed. applied mathematics and mechanics (english edition), 35(6), 729-740. doi: 10.1007/s10483-014-1825-6 luo, e.h. and wang, x.d. (2011). a study on transient flow under threshold pressure gradient in dual-pore media with low permeability (in chinese). china offshore oil and gas, 23(5), 318-321. pascal, h. (1973). consolidation with threshold gradients. international journal for numerical and analytical methods in geomechanics, 5, 1201-1215. pascal, h. (1981). nonsteady flow through porous media in the presence of a threshold gradient. acta mechanica, 39, 207-224. prada, a. and civan, f. (1999). modification of darcy’s law for the threshold pressure gradient. journal of petroleum science and engineering, 22(4), 237-240. doi: 10.1016/s0920-4105(98)00083-7 schmidt, j.d. and westman, r.a. (1973). consolidation of porous media with non-darcy flow. journal of the engineering mechanics division, 99, 1201-1215. warren, g.m. (1993). numerical solutions for pressure transient analysis. spe gas technology symposium, society of petroleum engineers, calgary. doi: http://dx.doi.org/10.2118/26177-ms wu, y.s. (1990). theoretical studies on non-newtonian and newtonian fluid flow through porous media. ph. d. dissertation, university of california, berkeley, 174-208. yan, q.l., he, q.x. and wei, l.g. (1990). a laboratory study on percolation characteristics of single phase flow in low-permeability reservoirs (in chinese). journal of xi’an petroleum institute, 5(2), 1-6. zhou, y.y., peng s.m. and li y. (2002). fully implicit simulation model for low-velocity non-darcy flow (in chinese). petroleum exploration & development, 29(2), 90-93. palabras clave: imágenes sísmicas; cálculo del tiempo de viaje; método de marcha rápida; construcción de frente de onda; procesamiento sísmico. how to cite item sun, h., meng, f., zhang, z., gao, c., & liu, m. (2018). high-precision joint 2d traveltime calculation for seismic processing. earth sciences research journal, 22(4), 327-334. doi: https://doi.org/10.15446/esrj.v22n4.77362 fast marching method (fmm) boasts high calculation efficiency and strong adaptability and stability while being applied to seismic traveltime. however, when it is applied to the largescale model, the calculation precision of fmm is insufficient. fmm has poor calculation precision near the source, which is an essential reason for the low accuracy of the whole algorithm. this paper puts forward a common traveltime calculation method to address the problem. wavefront construction (wfc) with a relatively high calculation accuracy rather than fmm is adopted for calculation of the grid nodes’ traveltime near the source. after that, fmm is used to calculate the seismic traveltime in the remaining area. joint traveltime calculation method greatly improves the calculation accuracy of the source’s surrounding area and the calculation accuracy of fmm. according to the new method, fmm is still adopted for the calculation of most grid nodes in the model, so the high calculation efficiency of fmm is maintained. multiple numerical models are utilized to verify the above conclusions in the paper. abstract keywords: seismic imaging; traveltime calculation; fast marching method; wavefront construction; seismic processing. high-precision joint 2d traveltime calculation for seismic processing cálculo de alta precisión del tiempo de viaje integrado bidimensional en procesamiento sísmico issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n4.77362 se presupone que el método de la marcha rápida (fmm, del inglés fast march method) tiene una alta capacidad de calcular y una fuerte adaptabilidad y estabilidad cuando se aplica en el tiempo de viaje sísmico. sin embargo, cuando se aplica a un modelo de gran escala su precisión de cálculo es insuficiente. el fmm tiene una baja precisión de cálculo cerca de la fuente, lo que explica su poca exactitud en todo el algoritmo. en este artículo se presenta un método de cálculo común de tiempo de viaje para acercarse al problema. la construcción de frente de onda (wfc, del inglés wavefront construction), con una mejor precisión que el fmm, se utilizó para el cálculo de la cuadrícula de nodos del tiempo de viaje cerca de la fuente. después de esto, el fmm se utilizó para calcular el tiempo de viaje sísmico en el área restante. el método conjunto aumentó considerablemente la exactitud en el cálculo del área alrededor de la fuente y la exactitud en el cálculo del mecanismo fmm. de acuerdo con este nuevo método, el fmm se utiliza en el cálculo de la cuadrícula de nodos, con lo que se mantiene su alta eficiencia de cálculo. varios modelos numéricos se utilizaron para verificar las conclusiones de este trabajo. resumen record manuscript received: 24/07/2018 accepted for publication: 29/11/2018 earth sciences research journal earth sci. res. j. vol. 22, no. 4 (december, 2018): 327-334 hui sun1,2,3, fanchang meng4*, cheng gao5, zhihou zhang1,3, mingchen liu6 1moe key laboratory of high-speed railway engineering, southwest jiaotong university, chengdu, 610031, china 2shandong provincial key laboratory of depositional mineralization and sedimentary minerals, college of geological science and engineering, shandong university of science and technology, qingdao, 266590, china 3faculty of geosciences and environmental engineering, southwest jiaotong university, chengdu, 611756, china 4institute of geology and geophysics, the chinese academy of sciences, beijing, 100029, china 5school of mines, inner mongolia university of technology, huhhot, 010051, china 6college for geoexploration science technology, jilin university, changchun, 130026, china *corresponding author: mfc@mail.iggcas.ac.cn se is m o l o g y 328 hui sun, fanchang meng, cheng gao, zhihou zhang, mingchen liu introduction as the difficulty of finding oil and gas resources in shallow seismic exploration increases, in-depth seismic exploration is gaining more and more attention (liang et al., 2016). however, deep seismic exploration imposes higher requirements on seismic exploration methods because of the larger area of calculation and the complexity of the structure. seismic traveltime has a broad application in seismic exploration methods, including migration, demigration, tomography, and forward modeling (wang et al., 2018; sun et al., 2017; wan, wei & wu, 2017; yang & zhu, 2017; wang, li & chen, 2017; yang et al., 2018; huang & sun, 2018; wang et al., 2016; sun et al, 2018). it is of great significance to study seismic traveltime calculation methods for in-depth seismic exploration. currently, traveltime calculation methods can mainly be divided into three kinds, including ray-based methods, finite difference methods, and new ray tracing methods based on graph processing techniques. regarding ray-based methods, they can be divided into two kinds. the first kind is traditional ray tracing methods (mainly including the trial ray method and the shooting method); the second kind is wavefront construction (wfc). the trial ray method and the shooting method (julian, 1977) were widely applied to seismic exploration techniques in the past, but these two methods face with prominent problems, such as low calculation efficiency and low adaptability to complex models. wfc was first put forward by vinje (1993) and improved later by various scholars (vinje, 1997; coman, r., & gajewski, 2005). chambers and kendall (2008) presented a practical wfc implementation method in 3d isotropic media. hauser et al. (2008) obtained the multiarrival traveltime by the wfc method. han et al. (2009) proposed a new positioning method for grid points in wfc. bai et al. (2011) studied the seismic wavefront evolution of multiply reflected, transmitted, and converted phases in 2d/3d triangular cell model. the method does not break new ground regarding principles. it is still a new technique to realize traditional ray tracing methods in essence, but the technique succeeds in playing an important role in popularizing the application of ray theories to seismic exploration. wfc boasts a good calculation precision, but its calculation efficiency is low because lots of rays are required to insert into the calculation process to ensure the ray density of the model space. combining graph processing techniques and traveltime calculation, modern ray tracing methods mainly include: interpolation method (asakawa & kawanaka, 1993; cardarelli & cerreto, 2002; nie & yang, 2003); shortest ray path tracing method (moser, 1991; bai, greenhalgh & zhou, 2007; sun z., sun j., & han, 2009). these methods can help obtain both traveltime and ray paths, but their calculation process calls for the screening of authentic ray paths from numerous useless ones. thus, methods of the kind are low in calculation efficiency. a finite difference method for traveltime calculation was first put forward by vidale (1988). methods of the kind obtain the viscosity solution of the eikonal equation through numerical analysis. their calculation efficiency is high, but their defect is striking, in that, their calculation accuracy is low especially in the area near the source. fast marching method (fmm) is one of the finite difference methods, which was raised by sethian (1996). the technique was originated from the level set method (sussman, smereka & osher, 1994). it integrates the narrow-band technology and the heapsort technique into its calculation process to achieve high efficiency, stability, and adaptability. due to salient advantages, fmm was favored by many scholars (sun & fomel, 1998; alkhalifah & fomel, 2001; rawlinson & sambridge, 2004) not long after its emergence, but it still failed to solve inadequate calculation accuracy of traveltime near the source. for the strong adaptability of fmm, it was often used for calculating models under rugged topography (sun, z., sun, j. & han, 2010; sun, z., sun, j. & han, 2011; sun, j. & han, 2012). liu et al. (2014) proposed a tti traveltime computation method based on a dynamic programming approach. ma and zhang (2014) obtained the ray information from the first arrival traveltime by solving the topography dependent eikonal equation. sun et al. (2016) analyzed the effects of seawater velocity variation on deep-water seismic migration and found that the traveltime difference was a significant factor. huang et al. (2016) solved the complex eikonal equation to obtain the traveltime by a nonuniform grid-based finite-difference method. sun et al. (2016) traced the ray in the random media with complex topography by employing the gmm-ulti method. sun et al. (2017) calculated the seismic traveltime under complex conditions using the group marching upwind hybrid method. huang and greenhalgh (2018) further proposed an approximate solution for the seismic complex traveltime. lan and chen (2018) presented an upwind fast sweeping scheme for seismic traveltime calculation under the irregular free surface condition. based on the advantages and disadvantages of fmm and wfc, this paper puts forward a joint traveltime calculation method for deep-sea geology. according to this method, wfc is adopted for traveltime calculation in the surrounding of the source; and fmm is for the traveltime calculation of the remaining area. by improving the traveltime calculation accuracy near the source, the joint traveltime calculation method achieves the goal of enhancing the overall calculation accuracy while maintaining the high calculation efficiency of fmm. the above conclusions are verified through the error analysis of the homogeneous model and the trial of the gradient model, the layered model, the marmousi model, and sigsbee model. methods fmm is an efficient and stable method to calculate seismic traveltime. its calculation of grid nodes’ traveltime relies on the information of the surrounding nodes already derived. since there is little node information near the source, the calculation accuracy of the area is relatively weak. inadequate calculation accuracy in the area also influences the overall calculation accuracy. to improve the efficiency of fmm near the source is of vital importance to the improvement of the method. wfc is also an effective method for traveltime calculation. however, with the increase of the calculation area, the blind calculation zone will appear, and the calculation efficiency is not high, but its calculation accuracy near the source is relatively high. based on the characteristics of the two traveltime calculation methods, this paper puts forward a joint traveltime calculation method based on the two calculation methods. as is shown in figure 1, the new method employs wfc for traveltime calculation in the surrounding of the source, and fmm for traveltime calculation of the remaining area. before introducing the joint traveltime calculation method, this part first gives a brief introduction of basic principles of fmm and wfc; then puts forward strategies to realize the combination of the two ways, and finally provides the schematic flow diagram. principles of fmm by solving the eikonal equation through the upwind method, fmm obtains the seismic traveltime. before the introduction of the method, the 2d eikonal function is first introduced (bai, greenhalgh, zhou, 2007). figure 1. fmm and wfc calculation regional diagram in the joint calculation method. wfc calculates the gray region, fmm calculates the white part. 329high-precision joint 2d traveltime calculation for seismic processing ∇ =t s (1) where, t is the seismic traveltime; s is the slowness;  is the gradient symbol. in (1), the upwind difference of the gradient item is approximate to (sethian, j. a., & popovici, 1999): ∇ = + + + − + − t d t d t d t d i j x i j x i j z i max( , ) min( , ) max( , ) min( , , , , 0 0 0 2 2 2 jj zt+          , ) / 0 2 1 2 (2) where, d ti j x , − , d ti j x , + , d ti j z , − and d ti j z , + are the forward and backward difference operator in the direction of x and z. below are the specific forms: d t t t x d t t t x d t t t z d i j x i i i j x i i i j z i i i j , , , , , , − − − − = − ∆ = − ∆ = − ∆ 1 1 1 + + +xx i it t t z = − ∆ +1 (3) (2) could be reduced to a simpler form: ∇ = −( ) + −( )   − + − +t d t d t d t d ti j x i j x i j z i j zmax , , , max , ,, , , ,0 0 0 2 2   1 2/ (4) during the realization process, fmm adopts the narrow-band technique as the wavefront expansion method. the first step of the narrowband procedure is to establish the first narrow-band. usually, the grid points near the source were selected as the elements of the first narrowband, which is shown in figure 2. during the narrow-band initialization, traveltime of the source (is, js) is zero. the traveltime of the points near the source can be expressed as t i j h v t i j h v t i j h v s s i j s s i j s s s s s s ±( ) = ±( ) = ± ±( ) = ± ± 1 1 1 1 1 1 , / , / , / , , ii js s± ±1 1, (5) where, t stands traveltime;v for velocity; h for grid space. when the traveltime calculation of the grid nodes is complete, they will be inserted to the narrow band to establish the initial narrow band. figure 2. narrow-band initialization. the red star is the source; gray points are the initial elements of the narrow-band. after the narrow-band initialization is finished, the wavefront will be expanded as is shown in figure 3. in this step, the minimum traveltime grid node within the narrow band need to be found, and the heapsort technique is employed to improve the calculation efficiency. figure 3. narrow band extension schematic diagram of fmm. a is the minimum traveltime grid node within the narrow-band. principles of wfc the theoretical basis of wfc for traveltime calculation is to solve the kinetic radial tracing (krt) equations. the krt equations are shown below: dx d v p dp d v v x i i i i i   = = − ∂ ∂       = − 2 1 1 2 , (6) where, xi stands coordinates in space; v for velocity; pi for slowness component product;  for the value of traveltime. figure 4. conditions for inserting a new ray. by solving the above equations, traveltime of discrete nodes along the ray paths can be obtained. for the reason that wfc is only employed near the source, just the distance factor is considered to insert a new ray to ensure the ray coverage. as is shown in figure 4, new rays are added only when the distance is higher than the critical value. in the joint traveltime calculation method, traveltime of regular grid nodes is needed. below is a brief introduction of calculating traveltime of regular grid nodes based on the information along the ray paths. figure 5. wfc traveltime calculation schematic diagram. 330 hui sun, fanchang meng, cheng gao, zhihou zhang, mingchen liu the adjacent rays and the wavefronts form irregular quadrangles. (see fig. 5) concerning any irregular quadrangle, abcd, regular grid nodes within the quadrangle are first chosen out, the process of which is discussed in detail by han et al. (2009). then, based on the information of four vertexes, namely a, b, c and d, the traveltime of regular grid nodes within them can be interpolated (see fig. 6). figure 6. sketch map of the vector product method. once the position of the target point is determined, its traveltime could be calculated by the information of the quadrilateral vertices. as is the shown in figure 7, r is the target point; a, b, c and d are the quadrilateral vertices. oa, ob, oc and od are the virtual sources, which are determined the information of quadrilateral vertices. the traveltime of r can be expressed as: t t t t t t v t v t v t r ar br cr dr ar r br r cr r = + + + = = = 0 25. * ( ) / , / / , ro ro ro a b c ddr rv= rod / (7) where, vr stands the velocity at point r. figure 7. sketch map of the vector product method. principles and realization strategies of joint traveltime calculation based on fmm and wfc the joint traveltime calculation method is proposed to combine the advantages of fmm and wfc. the new method calculates the grid nodes’ traveltime with wfc in a small area near the source. the maximum single ray tracing distance is chosen between 5% and 10% of the model side length. three aspects were taken into consideration when the scope was selected. firstly, wfc takes relatively more time, so the calculation area will influence the computation efficiency of the new algorithm. secondly, when the wfc calculation area becomes larger, it’s easy to appear a blind area in the model and may need to insert new rays to satisfy the precision of the algorithm, which will further reduce the efficiency of the new method. finally, it can be seen from the accuracy of the analysis below is that the calculation accuracy of the technique has been greatly improved when wfc is used within that scope. though the combination of the two traveltime calculation methods can make the best of -both in different model areas, fmm and wfc are independent of each other and have little association, so the key to realizing the joint traveltime method lies in how to combine the two ways into a whole. this process could be realized through the narrow-band technique of fmm. narrow-band technique plays an essential role in fmm. as the wavefront expansion mode of fmm, it endows fmm with robust stability and flexibility. before the use of the narrow-band technique, grid nodes in the whole space are divided into three kinds, namely accepted nodes, narrowband nodes, and far-away nodes. accepted nodes refer to grid nodes which have been calculated and confirmed. narrow-band nodes refer to grid nodes which have been calculated but have not yet been finally confirmed. faraway nodes refer to grid nodes which have not been calculated. narrowband refers to the collection of all narrow-band nodes. in fmm, it is used to stay close to the wavefront. the above process of forming the narrow band is called narrow-band initialization. in the joint traveltime calculation method based on fmm and wfc, the grid nodes’ traveltime calculation sequence is: first, wfc is used to finish the traveltime calculation of the area near the source; then, fmm is used to calculate traveltime of the remaining area. therefore, the narrow-band initialization in the joint method is a step after wfc finishing the traveltime calculation. as is shown in figure 8, the imaginary line stands for the length of the single ray; the nodes within the enclosed full line graph are grid nodes already calculated by wfc. as to the narrow-band initialization in the joint method, the grid nodes outside the enclosed full line graph are defined as far-away nodes. (see white nodes in fig. 8) nodes within the enclosed full line graph are a collection of accepted nodes and narrow-band nodes. therefore, they should be further divided. as to grid nodes already calculated by wfc, if a grid node’s surrounding (back, front, left and right) nodes are all calculated by wfc, it is defined as the accepted node (see black nodes in fig. 8); otherwise, it is the narrow-band node (see gray nodes in fig. 8). move all narrow-band nodes into the narrow band to finish the step of narrow-band initialization in the joint traveltime method. the completion of the above step means that wfc and fmm finish the connection. below is an introduction of how to use fmm to achieve traveltime calculation of the remaining grid nodes. the process is defined as the narrow-band expansion, which can generally be divided into the following three steps: (1) choose the minimum traveltime node in the narrow band which have been initialized through wfc, and change the attribute of the node from the narrow-band node to the accepted node; (2) judge the characteristic of the surrounding nodes of the newly-confirmed accepted node. if the node is a far-away node, the finite difference method is employed to work out its traveltime and change its attribute to the narrowband node. if the node is a narrow-band node, its traveltime value should be updated, but its characteristic should be maintained; if the node is an accepted node, the attribute and the traveltime value of the node should both be maintained; (3) repeat step 1 and step 2 until the narrow band is empty. through the above analysis, it can be seen that the key to the joint traveltime calculation method lies in the flexible application of the narrowband technique. the technique is a bridge to connect fmm and wfc. below is a realization flow diagram of the joint traveltime calculation method put forward in this paper (see fig. 9) 331high-precision joint 2d traveltime calculation for seismic processing figure 8. graphical description of narrow band’s initialization in the new method. experiments and analysis this part uses the homogeneous model to analyze the accuracy of the joint traveltime calculation method based on fmm and wfc, and multiple complex numerical models to verify the stability and adaptability of the new approach. the size of the homogeneous model is 601×601; the grid spacing is 10m×10m; the velocity is 1,000m/s. in the joint traveltime calculation method, the maximum length of the single ray is 500m. figures 10a, 10b and 10c show the relative error of the first-order fmm, the second-order fmm, and the fmm and wfc joint traveltime calculation method on the model. (the fmm used in the joint traveltime calculation method is the first order.) it can be seen that the calculation accuracy of the fmm and wfc joint traveltime calculation method has greatly improved compared with the first-order fmm, and is even higher than that of the second-order fmm. (a) (b) (c) figure 10. homogeneous medium relative error schematic diagram: first-order fmm relative error (a); second-order fmm relative error (b); fmm and wfc joint traveltime calculation method relative error (c).figure 9. flow diagram of joint traveltime calculation program.. 332 hui sun, fanchang meng, cheng gao, zhihou zhang, mingchen liu table 1 shows the calculation accuracy of the joint traveltime calculation method put forward in this paper in different areas. the traveltime calculated by the joint method can be divided into two parts: one part is traveltime obtained through wfc, and the collection of the regular grid nodes in the area is called a; the other part is traveltime obtained through fmm, and the group of nodes of this part is called b. table 1 shows that the average relative error of the three methods in part a, part b, and all grid nodes. regarding velocity models of different sizes, the ratio of the maximum ray tracing length to the model’s unilateral length is 1:12. in this way, the number of grid nodes in part a steadily accounts for around 1.1% of the total. from table 1, it can be seen that the accuracy of the joint traveltime calculation method based on the first-order fmm and wfc registers an obvious improvement compared with the first-order fmm. the joint way improves the calculation accuracy not only in the area of part a through wfc, but also in the area of part b. though part b is calculated through the first-order fmm, it can be seen that, due to the improvement of the traveltime calculation accuracy in part a, the calculation accuracy of the first-order fmm in part b is higher than that of the second-order fmm. in the former part, the calculation of the joint traveltime calculation method is analyzed. the following section is an analysis of the new method’s calculation efficiency. table 2 shows the time consumed by the calculation of homogenous models of different sizes through the first-order fmm, the second-order fmm, and the joint traveltime calculation method. it can be found the calculation efficiency of the new approach is a little lower than that of the first-order fmm but is far higher than that of the second-order fmm. generally speaking, the fmm and wfc joint traveltime method has good accuracy and efficiency. complex models are adapted to verify the stability and adaptability of the fmm and wfc joint traveltime calculation method. figure 11a is the isochronal graph of the gradient model, whose model size is 401 x 401; the grid spacing is 10m x 10m ; the velocity distribution is v = 1000m + 0.4z. figure 11b is a layered medium, whose model size is 801 x 601; grid spacing is 10m x 10m; the up-down layer velocity is 1500m/s, 2000m/s and 2500m/s, respectively. figure 11c is the isochronal graph of marmousi model, whose model size is 2085 x 603; grid spacing is 5m x 5m. figure 11d is the isochronal graph of sigsbee model, whose model size is 3210 x 1201; grid spacing is 25m x 25m. based on the calculation results of figure 11, it can be found that the new method can obtain the traveltime results met the laws of seismic wave propagation and boast good stability and adaptability to complex models. (a) (b) table 1. the average relative error comparison between fast marching methods and the method combining the fast marching method with the wavefront construction method. model size average relative error(%) region a region b entire model 1st order fmm 2nd order fmm new method 1st order fmm 2nd order fmm new method 1st order fmm 2nd order fmm new method 401×401 2.411 0.659 0.002 0.471 0.078 0.054 0.493 0.085 0.040 501×501 2.110 0.554 0.002 0.396 0.062 0.043 0.415 0.068 0.038 601×601 1.899 0.481 0.002 0.344 0.053 0.036 0.360 0.057 0.036 701×701 1.717 0.421 0.001 0.304 0.045 0.031 0.320 0.049 0.033 801×801 1.576 0.376 0.001 0.273 0.040 0.027 0.288 0.043 0.031 table 2. the computational cost(s) of different methods model size computational cost(s) 1st order fmm 2nd order fmm new method 1001×1001 0.718 2.187 0.824 2001×2001 2.649 9.962 3.033 3001×3001 6.168 21.006 7.057 4001×4001 9.812 36.763 11.217 333high-precision joint 2d traveltime calculation for seismic processing (c) (d) figure 11. traveltime results of different models: isochronal graph of the gradient model (a); isochronal graph of the layered model (b); isochronal graph of the marmousi model (c); isochronal graph of the sigsbee model (d). conclusions fmm boasts high stability and efficiency while calculating the seismic traveltime, but the calculation accuracy near the source is low, which might influence the accuracy of the whole calculation method. in response to the problem, this paper puts forward a joint traveltime calculation method based on fmm and wfc. the core idea of the new way is to calculate the small area near the source with wfc and calculate the major remaining area with fmm. the calculation accuracy and calculation efficiency of the joint calculation method are analyzed, and the complex geological models are adapted to verify that the joint traveltime calculation method put forward in this paper is a seismic traveltime calculation method accommodating to both calculation efficiency and accuracy and boasting strong stability and adaptability. the new approach can serve as a powerful traveltime calculation tool for seismic exploration techniques, such as migration, demigration, and tomography. acknowledgments this work was supported in part by the natural science foundation of china under grant 41804100, in part by the fundamental research funds for the central universities under grant 2682018cx36, in part by the china postdoctoral science foundation under grant 2018m640910, in part by the science research project of inner mongolia university of technology under grant zd201622, in part by the science research project of institutions of higher learning in inner mongolia under grant njzz18079, and in part by the china geological survey project under grant 121201011000150013-05. the authors would like to thank prof. sun jianguo, prof. sun zhangqing, prof. han fuxing and prof. gong xiangbo in jilin university for their kindness help. references alkhalifah, t., & fomel, s. (2001). implementing the fast marching eikonal solver: spherical versus cartesian coordinates. geophysical prospecting, 49(2), 165-178. asakawa, e., & kawanaka, t. (1993). seismic ray tracing using linear traveltime interpolation. geophysical prospecting, 1993, 41(1): 99-111. bai, c., greenhalgh, s., & zhou, b. (2007). 3d ray tracing using a modified shortest-path method. geophysics, 72(4), t27-t36. bai, c., li, x., tang, x (2011). seismic wavefront evolution of multiply reflected, transmitted, and converted phases in 2d/3d triangular cell model. journal of seismology, 15(4), 637. cardarelli, e., & cerreto, a. (2002). ray tracing in elliptical anisotropic media using the linear traveltime interpolation (lti) method applied to traveltime seismic tomography. geophysical prospecting, 50(1), 55-72. chambers, k., & kendall, j. m. (2008). a practical implementation of wave front construction for 3-d isotropic media. geophysical journal international, 173(3), 1030-1038. coman, r., & gajewski, d. (2005). traveltime computation by wavefront¯orientated ray tracing. geophysical prospecting, 53(1), 23-36. han, f., sun, j., & sun, z. (2009). positioning of grid points in wave front construction. applied geophysics, 6(3), 248. hauser, j., sambridge, m., & rawlinson, n. (2008). multiarrival wavefront tracking and its applications. geochemistry, geophysics, geosystems, 2008, 9(11). huang, x. & greenhalgh, s. (2018). linearized formulations and approximate solutions for the complex eikonal equation in orthorhombic media and applications of complex seismic traveltime. geophysics, 83(3), c115-c136. huang, x., & sun, h. (2018). numerical modeling of gaussian beam propagation and diffraction in inhomogeneous media based on the complex eikonal equation. acta geophysica. doi: 10.1007/s11600018-0154-x. huang, x., sun, j., & sun, z. (2016). local algorithm for computing complex travel time based on the complex eikonal equation. physical review e, 93(4), 043307. julian, b. r. (1977). three-dimensional seismic ray tracing. journal of geophysics, 43, 95-113. lan, h., & chen, l. (2018). an upwind fast sweeping scheme for calculating seismic wave first¯arrival travel times for models with an irregular free surface. geophysical prospecting, 66(2), 327-341. liang, s., liang, x., & yang, x. (2016). development of seismic exploration technologies and its significance for the discovery of gas field clusters in the buried anticlines, kuqa foreland basin. china petroleum exploration, 21(6), 98-109. liu, s., wang, h., & yang, q. (2014). traveltime computation and imaging from rugged topography in 3d tti media. journal of geophysics and engineering, 11(1), 015003. ma, t., & zhang, z. (2014). calculating ray paths for first¯arrival travel times using a topography¯dependent eikonal equation solver. bulletin of the seismological society of america, 104(3), 1501-1517. moser, t. j. (1991). shortest path calculation of seismic rays. geophysics, 56(1), 59-67. nie, j., & yang, h. (2003). quadratic/linear travel time interpolation of seismic ray-tracing. journal of tsinghua university, 43(11), 1495-1498. rawlinson, n., & sambridge, m. (2004). wave front evolution in strongly heterogeneous layered media using the fast marching method. geophysical journal international, 156(3), 631-647. sethian, j. a. (1996). a fast marching level set method for monotonically advancing fronts. proceedings of the national academy of sciences, 93(4), 1591-1595. 334 hui sun, fanchang meng, cheng gao, zhihou zhang, mingchen liu sethian, j. a., & popovici, m. a. (1999). three dimensional traveltimes computation using the fast marching method. geophysics, 64(2), 516-523. sun, y., & fomel, s. (1998). fast-marching eikonal solver in the tetragonal coordinates. stanford exploration project report, 97, 241-250. sun, z. q., sun, j. g., & han, f. x. (2009). traveltimes computation using linear interpolation and narrow band technique under complex topographical conditions. chinese journal of geophysics, 52(11), 2846-2853. sun, z., sun, j., & han, f. (2010). traveltime computation using fast marching method under complex topographical conditions. chinese journal of computational physics, 27(2), 281-286. sun, j., sun, z., & han, f. (2011). a finite difference scheme for solving the eikonal equation including surface topography. geophysics, 76(4), t53-t63. sun, z., sun, j. & han, f. (2012). traveltime computation using the upwind finite difference method with nonumiform grid spacing in a 3d undulating surface condition. chinese journal of geophysics, 55(7), 2441-2449. sun, h., sun, j., & han, f. (2016). research into the effects of seawater velocity variation on migration imaging in deep-water geology. earth sciences research journal, 20(3), h1-h7. sun, z., sun, j., & yue, y. (2016). ray tracing using gmm-ulti method in the random media with complexity topography. chinese journal of geophysics, 59(7), 2615-2627. sun, h., sun, j. g., & han, f. x. (2017). synthetic analysis of model smoothing’s influence on ray path, traveltime and migration imaging. applied ecology and environmental research, 2017, 15(3), 443-452. sun, z., sun, j., & wang, x. (2017). computation of multiple seismic traveltime in mountainous areas with complex 3d conditions using the multistage group marching upwind hybrid method. chinese journal of geophysics, 60(5), 1861-1873. sun, h., zhang, z., & hu, g. (2018). kirchhoff beam migration based on compressive sensing. ieee access, 6(1), 26520-26529. sussman, m., smereka, p., & osher, s. (1994). a level set approach for computing solutions to incompressible two-phase flow. journal of computational physics, 114(1), 146-159. vidale, j. (1988). finite-difference calculation of travel times. bulletin of the seismological society of america, 78(6), 2062-2076. vinje, v., iversen, e., & gjøystdal, h. (1993). traveltime and amplitude estimation using wavefront construction. geophysics, 58(8), 1157-1166. vinje, v. (1997). a new interpolation criterion for controlling accuracy in wavefront construction. 1997 seg annual meeting. society of exploration geophysicists. wang, x., bradley, k. e, & wei, s. (2018). active backstop faults in the mentawai region of sumatra, indonesia, revealed by teleseismic broadband waveform modeling. earth and planetary science letters, 483, 29-38. wang, x., chen, q. f., & li, j. (2016). seismic sensor misorientation measurement using p¯wave particle motion: an application to the necsaids array. seismological research letters, 87(4), 901-911. wang, x., li, j., & chen, q. f. (2017). topography of the 410 km and 660 km discontinuities beneath the japan sea and adjacent regions by analysis of multiple¯scs waves. journal of geophysical research: solid earth, 122(2), 1264-1283. wang, x., wei, s., & wu, w. (2017). double-ramp on the main himalayan thrust revealed by broadband waveform modeling of the 2015 gorkha earthquake sequence. earth and planetary science letters, 473, 83-93. yang, j., & zhu, h. (2017). a practical data-driven optimization strategy for gaussian beam migration. geophysics, 83(1), s81-s92. yang, j., zhu, h., mcmechan, g. & yue, y. (2018). time-domain least-squares migration using the gaussian beam summation method. geophysical journal international, 214(1), 548-572. doi: https://doi.org/10.1093/ gji/ggy142 issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n3.103981 earth sciences research journal earth sci. res. j. vol. 26, no. 3 (september, 2022): 197-204 g e o ph y si c s seismic properties of the permafrost layer using the hvsr method in seymour-marambio island, antarctica carlos a. vargas1*, juan m. solano1, adriana m. gulisano2,3,4, sergio santillana2, edwin a. casallas5 1. departamento de geociencias, universidad nacional de colombia at bogotá, colombia 2. instituto antártico argentino, dirección nacional del antártico, argentina 3. instituto de astronomía y física del espacio, conicet, universidad de buenos aires, argentina 4. departamento de física, facultad de ciencias exactas y naturales, universidad de buenos aires, argentina 5. fuerza aérea colombiana *corresponding author: cavargasj@unal.edu.co abstract authors have calculated the h/v spectral ratios using seismic-noise recordings in the uppermost layers north of the seymour-marambio island, antarctic. sixty-seven seismic site-response measurements near and far from the argentinean marambio base runway suggest geotechnical works on the uppermost sedimentary layers due to maintenance, landing, and taxi of large loads and aircraft during decades could contribute to changes in their seismic dynamic response. two horizontal images of vp, vs, and vp/ vs ratios at 1.0 m and 35.0 m depth show lateral variations in the permafrost properties. authors interpret that permafrost is emplaced in rocks with different porosities and contrasting fluids saturation at those depths. in shallow strata, the saturation of gases affects mainly the elastic properties. in deeper strata, where the location of water reservoirs is detected, the primary mechanism of seismic dissipation is anelastic. keywords: h/v spectral ratio; seismic response; permafrost; marambio; antartic; fluids' saturation propiedades sísmicas del permafrost a través del método de relaciones espectrales horizontal/vertical en la isla seymour-marambio de la antártida palabras clave: relaciones espectrales h/v; respuesta sísmica; permafrost; marambio; antártida; saturación de fluidos record manuscript received: 29/07/2022 accepted for publication: 16/08/2022 how to cite item: vargas, c. a., solano, j. m., gulisano, a. m., santillana, s., & casallas, e. a. (2022). seismic properties of the permafrost layer using the hvsr method in seymour-marambio island, antarctica. earth sciences research journal, 26(3), 197-204. https://doi.org/10.15446/esrj.v26n3.103981 en este trabajo se presentan estimaciones de relaciones espectrales h/v utilizando registros de ruido sísmico en las capas superiores al norte de la isla seymour-marambio, en la antártida. sesenta y siete mediciones de respuesta sísmica del sitio cerca y lejos de la pista de la base marambio, argentina, sugieren que los trabajos geotécnicos en las capas sedimentarias superiores debido al mantenimiento, aterrizaje y rodaje de grandes cargas y aeronaves durante décadas podrían contribuir a cambios en su respuesta dinámica sísmica. dos imágenes horizontales de relaciones vp, vs y vp/vs a 1,0 my 35,0 m de profundidad muestran variaciones laterales en las propiedades del permafrost. interpretamos que el permafrost se emplaza en rocas con diferentes porosidades y saturación de fluidos a esas profundidades. en estratos someros, la saturación de gases afecta principalmente a las propiedades elásticas. en estratos más profundos, donde se esperan reservorios de agua, el principal mecanismo de disipación sísmica es inelástico. 198 carlos a. vargas, juan m. solano, adriana m. gulisano, sergio santillana, edwin a. casallas 1. introduction permafrost is frozen ground in high latitudes formed by low sea-level stands during the ice ages (overduin et al., 2019; martens et al., 2020). the permafrost, the active layer on its top, and the underneath stratum hold complex ecosystems and are leading players in the complex study of climate change due to the release of greenhouse gases during its thaw. the data required for its research is typically time-consuming due to elaborated laboratory essays, sometimes field drilling, or complicated sampling collection. estimating permafrost’s elastic properties is a scientific challenge that involves layers with in-depth contrasting behaviors, lateral variations in each layer’s composition, inversion of velocities and densities, and diverse seismic inelastic performances. besides, the permafrost seismic performance is involved in damages to the men-made infrastructure (see, e.g., melvin et al.,2017; raynolds et al., 2014; hong et al., 2014), showing that there is a significant lack of information to diagnose its comportment appropriately and to promote proper hazard management. for instance, in some places, cryoturbation related to thawing produces seismic liquefaction, often accompanied by sand volcanoes (vandenberghe et al., 2016), but sometimes with large-scale subsidence and potential changes in the drainage systems (thienen-visser et al., 2015). in recent years, the permafrost’s thickness and seismic properties have been mapped with seismic noise measurements. a good example is presented by picotti et al. (2017), who corroborated estimations using radio-echo sounding, geoelectric, and active seismic methods in glacial environments such as the adamello and ortles-cevedale massifs (italy), the bernese oberland alps (switzerland), and the whillans ice stream (west antarctica). their results suggested that the resonance frequency in the horizontal-to-vertical spectral ratio (hvsr) correlates well with the ice thickness at the site, in a wide range from a few tens of meters to more than 800 m, even allowing reliable estimations of the seismic properties. on the other hand, in areas where large loads repeatedly impact the permafrost, some reports recognize their elastic and inelastic performance or allow the inference of the local seismic structure oriented to the infrastructure design care. good examples constitute the aerial runways deployed near the scientific bases in the antarctic continent. thus, to provide valuable information regarding the seismic structure in these places, the authors collected seismicnoise information on seymour-marambio island, approximately 80 km far the antarctic peninsula. the easy deployment, fast processing, and trusty results from similar sites suggest that the hvsr analyses may contribute to the knowledge of the antarctic permafrost’s elastic and inelastic properties (bignardi et al., 2018; köhler & weidle, 2019). sixty-seven seismic noise stations were deployed on the island’s plateau, where the runway is located (figure 1). the study area corresponds to a topographic contrast of ~200 m altitude, located about 2 km from the shoreline. high slopes are observed with the shallower layers’ surficial activity that triggers abundant landslides. in this figure 1. location of the study area. a) seymour-marambio island, approx. 80 km from the antarctic peninsula. the blue square represents the north part of the island. b) topographic image of the area where 67 seismic noise stations were recorded (blue squares). c) marambio scientific base's runway is located on a narrow plateau. the map shows the scientific station's location (red houses) with 67 seismic-noise stations measured in the colombian argentinean antarctic summer field campaign in january 2020 (yellow pins). d) perpendicular topographic profiles crossing the runway suggested in b). 199 seismic properties of the permafrost layer using the hvsr method in seymour-marambio island, antarctica scenario, the permafrost and the topographic contrast may offer information about the combined effect’s seismic response. in this work, the authors report the results of seismic site-response experiments made during the colombian and argentinean scientific teams’ campaign in january 2020 near the marambio scientific base of argentina to evaluate the permafrost structure on the island’s plateau, the site response along the runway, and contrast these results with previous observations based on other geophysical approaches. 2. geological setting marambio island (also known as seymour island) is one of the 16 islands encompassing major islands on the antarctic peninsula. this island is located in the westernmost weddell sea, and according to fukuda et al. (1992), its ground has been frozen since the holocene. at least three geomorphological terraces are recognized in the north of the island, with depths of the permafrost estimated to be 200m (upper terrace), 105m (middle terrace), and 35m (lower terrace). according to montes et al. (2013), the north of the island outcrops miocene to quaternary formations related to surficial and glacier processes such as the weddell and hobbs formations that rest on la meseta formation of eocene–oligocene age. this sequence reaches 600 m in thickness. tectonic faults with vergence eastward and westward limit those sedimentary sequences with older sedimentary units dated from maastrichtian to paleocene, such as snow hill island, lopez de bertonado, sombral, and cross valley-wiman formations (figure 2). figure 2. geologic map of the seymour-marambio island (after montes et al., 2013). the hvsr study was done at the northern sector of the island, located on the marambio scientific base’s runway. profile a-a’ is a representative section of the uppermost basin in the study area, suggesting a graben structure filled with paleogene to neogene sediments. the eocene to quaternary strata underneath the airport runway suggests a graben structure filled at different sedimentary rates. besides, annual climate seasons favor permanent erosive activity, mainly at the runway’s north, west, and east, where surficial runoff is observed. along the runway, the active layer was removed or compacted. this layer of < 1 m thick is only observed around the runway. 3. materials and methods nogoshi and igarashi (1971) and later nakamura (1989, 2000) introduced the hvsr or ‘h/v’ method, which is based on the processing of microtremors or the ambient seismic noise generated under natural or anthropogenic perturbations, e.g., by the incidence of wind, sea waves, traffic of vehicles or industrial processes on sedimentary sequences. acquisition and processing of the seismic noise’s orthogonal components are used to estimate the power spectra ratio of the horizontal (h) over vertical (v) components to infer the strata’s elastic structure below the subsurface. thus, resonance peaks of various frequencies are then related to the thickness (h) and shear wave velocity (vs) of each layer under the hypothesis of lateral homogeneity. under this assumption, the following expression is used to analyze the direct problem of the soft sedimentary cover resting on a rigid basement (nakamura, 1989): = (2 −1 ) 4ℎ , = 1 ,2, … ∞ (1) equation 1 suggests that different physical properties in the sedimentary sequence incorporate one harmonic in the spectra. therefore, the hvsr curve presents as many peaks as layers, with amplitudes linked to each impedance contrast. diverse strategies have been proposed to estimate the inverse problem to infer the elastic structure of the soft sedimentary cover resting on a rigid basement using the hsvr curve (see, e.g., castellaro and mulargia, 2009; herak, 2008; sánchez-sesma et al., 2011; priolo et al., 2012; abu-zeid et al., 2014; lunedei & albarello, 2015; mantovani et al., 2015; bignardi et al., 2016; carcione et al., 2017). in this work, the authors have used the program openhvsr (bignardi et al., 2018), based on the previous program modelhvsr (herak, 2008). this code computes theoretical transfer functions for layered soil models based on the fast recursive algorithm proposed by tsai (1970) and modified to take frequency-dependent attenuation and body-wave dispersion into account, as was suggested by tsai and housner (1970). in this case, the soil model consists of any number of horizontal, homogeneous, isotropic, and viscoelastic layers stacked over a halfspace, each of which is defined by thickness (h), the velocity of propagation of the body wave (vp and vs ), density, poisson’s ratio, and the frequency-dependent seismic q-factor (qp and qs), which controls the anelastic properties. the theoretical transfer functions are calculated using a simple and guided monte carlo search in the model space. those estimations are compared with the observed hvsr curves to minimize a misfit function. details of this procedure can be consulted by tsai (1970), tsai and housner (1970), and herak (2008). the initial model for starting the inversion (table 1) was inferred from the detailed stratigraphy reported by montes et al. (2013). after several tests, the best fit was reached using a one-layer model with parameter values based on the average of the seven layers of table 1. the range of values used as input parameters in the openhvsr code was: vp = 80 4500 m/s; vp/ vs= 1.6 – 4.0; density = 1.4 – 2.8 g/cm3, thickness =0.5 – 999 m,qs = 5 500, and qp/ qs =1.5 – 4.5. table 1. the initial model of elastic and anelastic parameters used for stating the inversion. the model is based on seven sedimentary layers on a rigid basement inferred from the stratigraphy reported by montes et al. (2013). after several tests, a best-fit was reached using a one-layer model with parameter values based on the average of the seven layers. layer (m/s) (m/s) ρ (g/ cm3) thickness (m) 1 881 509 1.69 12 59 30 2 855 494 1.71 82 30 10 3 961 556 1.74 45 58 31 4 1006 581 1.75 58 30 10 5 1032 597 1.79 177 32 11 6 1031 596 1.82 60 58 31 7 1332 770 1.87 359 37 15 average 1014 586 1.77 113 43 20 the study area’s operational and access restrictions limited the experiment to be carried out between january 10 and 24, 2020. during this period of the austral summer, the intensity and direction of winds were stable. once the authors recognized the terrain, functionality of the seismic instrument, and field logistics, 67 continuous seismic noise stations, 20-30 min-long, were recorded. signals recorded were not discriminated from or excluded by possible noise sources, which were expected to be mainly related to wind and oceanic tides. signals were acquired using a three-component seismometer with a natural frequency fn= 0.45hz. the seismometer was directly installed 200 carlos a. vargas, juan m. solano, adriana m. gulisano, sergio santillana, edwin a. casallas on the ground using spikes under the base to guarantee a proper soil-sensor coupling. the seismographic system is composed of a 24-bit digitizer to 200 sps and was set up with an anti-aliasing filter at 50 hz, which, in this particular case, guaranteed sample frequencies below the noise associated with the electrical system frequency that powers the infrastructure of the scientific base in the seymour-marambio island. the seismograph is synchronized by a gps with 5 microseconds resolution and receives analogic signals from a triaxial seismometer at 4.5 hz. those analogic signals were coupled to an operational amplifier with negative feedback and a buffer amplifier to extend the bandwidth to 0.45 hz (e.g., vitale et al., 2018). according to parolai and galiana-merino (2006), transients into the seismic noise record do not affect the h/v ratio. hence, original signals need not be filtered or have anti-triggering procedures. geopsy was the open software used for signal processing (wathelet et al., 2020). figure 3 shows a flow diagram of the procedure used, starting with 1) acquisition of >20 min three-components seismic record; 2) split the record in windows of 30 s with overlaps of 50%; 3) estimation of a weighted hanning window and then fft for each 30 s record; 4) smoothed of the spectrum of each component using the konno–ohmachi filter with a b-value of 40 (konno & ohmachi, 1998); 5) the horizontal specter was computed by euclidian norm using the north and east components (h=√n2 + e2 ) and then estimated the h/v spectral ratio. 6) stacking those spectral ratios for all the 30-second segments allowed the calculation of an average and standard deviation in each station for the frequency range 0.5-10.0 hz. the same figure presents an example of one >20 min three-component seismic record, its h/v spectral ratios for all 30 second-long windows, the stacked spectral ratio, and the estimations of the elastic and anelastic properties using the openhvsr code. in general, the authors observed that even having momentary and abrupt changes in the seismic noise records’ energy, the best 1d model found is like that estimated without the abrupt changes. two conditions were assumed along with all estimations: 1) the seismic ambient noise field is homogeneous across all azimuths, and 2) the soil model has several viscoelastic layers laterally homogeneous stacked over a halfspace. strictly those restrictions are not satisfied owing to: 1. the diverse intensity of the antarctic weather in the study zone, with wind trend nne throughout the year (yu et al., 2020); 2. varied interactions between ocean currents and the island; 3. the effect of material damping on the expected noise wavefield as a consequence of sources relatively distant from the receivers (e.g., lunedei and albarello, 2009); 4. complex geology inside a steep topographic contrasting (figures 1 and 2); 5. inversion of viscoelastic parameters due to an active upper layer, the rigid permafrost, which is over sedimentary layers with and without water, and the deeper bedrock. all those factors make the hvsr curves challenging to invert and interpret. carcione et al. (2017) studied soft-layers and bedrock anelasticity on the s-wave amplification to evaluate the impact of attenuation and elasticity of bedrock on the amplitude and frequency of the resonance peaks in two glaciers located in northern italy and the antarctic continent. they concluded that the attenuation and bedrock elasticity must be considered to obtain reliable layer thickness estimations. hence, the combined interpretation of the thickness and viscoelastic parameters (v p, vs, qp and qs ) is necessary to understand the permafrost structure in the study area. figure 3. procedure used to estimate elastic and anelastic parameters using stacked hvsr records. (a) flow diagram describing different preprocessing stages, processing, and estimation of elastic and anelastic parameters. (b) original seismic-noise recording of the sn-09 station (>20 min duration). (c) amplitude spectrum every 30-seconds. (d) the average h/v ratio and its standard deviation for a frequency range of 0.5-10.0 hz. (e) example of best-fit of the h/v averaged spectral ratio, used to calculate changes in depth for elastic (vp, vs, and vp/vs), and anelastic (qp, qs, and qp/qs) parameters using the program openhvsr (bignardi et al., 2016, 2018). 201 seismic properties of the permafrost layer using the hvsr method in seymour-marambio island, antarctica 4. results due to the h/v ratio does not discriminate the horizontal incidence direction of the perturbations, and each station may be analyzed from the point of view of a 1d model with limited seismic energy influencing its surrounding area, the authors hypothesize that results may show trends related to the seismic response of the underneath sedimentary structure. figure 4 shows the 67 hvsr stations classified in three main trends: (a) with almost flat hvsr response, (b) presenting a growing-up hvsr response, and (c) showing a decreasing hvsr response. only two stations offer a dominant peak in the frequency range analyzed (sn-2 and sn-3). comparatively, the spectral response along the runway does not present dominant peaks, with trend responses (a) and (b) suggesting that the permanent interaction of these layers throughout impacts, movements of high loads (e.g., the permanent landing and taxi of large aircraft on the runway), and maintenance works contributed to permanent changes in the dynamic response of this ground. the dataset impedes exploring if this activity may contribute to permanent changes in the seismic dynamic response beyond the first meters. those stations with dominant peaks or related to trend (c) are on the eastern steepest side of the runway, near active landslides, indicating a seismic response linked with not well-consolidated materials. a related observation concerning changes in the ground’s dynamic response by human activity is reported by abu zeid et al. (2017). those authors suggest that long-term trampling results in sediment stiffening, increasing the density and velocity of seismic shear waves. the authors found that vp and vs have a broad range of values (figure 5 and table s1). several authors have suggested that in permafrost, vp reaches high values (e.g., timur, 1968; carcione & seriani, 1998; krautblatter & draebing, 2014). according to these authors, high velocities of the seismic wave’s propagation respond to the pore space occupied by ice and the velocities of the rock matrix, interstitial ice, and available water in an extensive range of temperatures. nevertheless, the present study’s authors observe that our fitting of observed and theoretical spectra at low frequencies is poor (also reported in other works, e.g., talha qadri et al., 2015; abu zeid et al., 2017; bignardi et al., 2018). this uncertainty may suggest artifacts in the solution of the velocity structure, for instance, large velocity values. the authors want to encourage developing future works to contrast the estimated range of velocities reported in this paper. it draws attention that plotting values of qp vs vp and qs vs vs at depths ranges of z ≥ -35m, -35 m ≤ z < -100 m, and z ≤ 100 m appear to be trends of low attenuation–high velocity in shallow depths, and converse, high attenuation-low velocities in deeper layers (figure 5). following barton (2007), this trend could be interpreted as increasing fluid saturation in deeper layers. indeed, the authors consider that the shallow frozen and rigid permafrost observed in the field presents low attenuation, high velocity, and, consequently, low fluid saturation, except in the thinner active layer. they also interpret that the incremental geothermal gradient promotes melting ice, generating high attenuation and low velocities in deeper layers. prasad (2002) suggested that the relationship between ratios (vp/ vs) 2 and qp/ qs offers an idea of increasing pore pressure by reducing effective stress in saturated or partially saturated rocks. the dataset (figure 5) indicates a broad range of pore pressure conditions that dominate in surficial layers, probably due to permafrost but are still observed in deeper layers. besides, in the wide range of densities detected, around -100 m, it was found find high vp/ vs ratio with low-density values indicating highpressure fluids’ potential presence (see, e.g., vargas & torres, 2015; koulakov & vargas, 2018), probably related to the high concentration of gases in the permafrost. figure 5. elastic and anelastic parameters were estimated in the uppermost strata north of the seymour-marambio island. the parameters were interpolated at three depths from the surface: z ≥ -35 m (cyan circle), -35 m ≤ z < -100 m (blue star), and z ≤ -100 m (light blue square). arrows suggest general trends of parameters. figure 4. hvsr curves of 67 seismic noise stations were recorded in the seymour-marambio island, classified in three main trends: (a) with almost flat hvsr response (blue frame), (b) presenting a growing-up hvsr response (green frame), and (c) showing a decreasingly hvsr response (purple frame). only two stations offer a dominant peak in the frequency range analyzed (sn-2 and sn-3). the map shows the spatial distribution of these patterns. the decreasingly hvsr response stations and the two previously mentioned stations with a dominant peak are close to active landslides. 202 carlos a. vargas, juan m. solano, adriana m. gulisano, sergio santillana, edwin a. casallas 5. discussion in the fieldwork activities during the austral summer of 2020, the authors observed high mobility of the active layer, accompanied by runoff in the northern sector of seymour-marambio island, probably linked to the weddell, hobbs, and la meseta formations. these processes made it possible to detect landslides, mainly in some areas of high slopes. thanks to these events and the accessibility to two surficial trenches, the permafrost’s presence at approx. 50100 cm depth was verified. this upper permafrost surface may offer keys to its spatial distribution. fukuda et al. (1992), using one reference site in seymourmarambio island with a mean annual temperature of approx. -10°c at the surface, a thermal profile up to 2.5 m depth, and assuming a yearly temperature gradient of 0.19°c/m in steady-state conditions, estimated the permafrost base with 0°c at a depth of ~34 m. they also contrasted other geoelectrical resistivity measurements acquired by fournier et al. (1990), which reported the permafrost thickness between 28.6 m and 127.5 m. beneath the rigid permafrost base, the abundance of melted water and other precipitated minerals may promote viscoelastic inversions in the stratigraphy profile, at least from the runway (summit at approx. 200 m.a.s.l.) up to sea level. besides, kato et al. (1990) found high concentrations of caco3 and na2so4 in ice wedges, indicating percolation processes of molten water on this island under an extremely arid environment. all those observations may explain the diverse viscoelastic profiles and inversions of properties presented in this paper’s supplementary material. for instance, according to the vp profiles, in general, the authors observed the first inversion at depths ranging between 5 and 110 m. there are up to three inversions in some places, suggesting a variable thick of the permafrost and the possibility of lateral and vertical variations of porosity that allow more significant ice concentrations, mainly in the neogene quaternary formations. on the other hand, the inversions of and do not coincide with those of and , suggesting that seismic attenuation mechanisms may be observed in the permafrost in response to the presence of gases or under the permafrost base, probably related to formations with water deposits. thereby, vp, vs , and vp/ vs ratios were mapped at two depths: at 1.0 m to observe the lateral variations of the permafrost under the active layer and at 35.0 m in the first permafrost base reported (fukuda et al., 1992). as the most profound resolution reached in this work is approx. 200 m, it guarantees to observe the possible cases of permafrost up to the sea level and the role of the paleogene to quaternary formations (almost flat) in the lateral distribution of this layer (supplementary material). figure 6 presents the horizontal sections, showing lateral variations of the seismic velocity structure in the study area. laboratory and field experiments have associated high values of vp to permafrost layers (krautblatter & hauck, 2007; krautblatter & draebing, 2014; skvortsov et al., 2014), meaning in this case that permafrost is emplaced at 35 m-depth and highlighting patches of high values of elastic properties. high p-wave velocities are widespread except in high-slope areas. similar patterns are observed with the s-wave velocities (vs). n-s and e-w profiles of the same figure suggest lateral variation in the permafrost layer’s elastic properties. consequently, the vp/ vs ratio is distributed in a broad range of values (1.6 – 4.0), indicating rocks with various porosities. many are probably saturated with fluids in different phases and/or critical points (e.g., lee, 2003). the authors do not discard that these high vp/ vs ratio patches represent the permafrost with anomalous saturated accumulations of gases such as ch4 and/or co2. hence, they inferred that paleogene to quaternary formations reported by montes et al. (2013) have high petrophysical properties heterogeneity. consequently, they also should be associated with high lateral variation in ice concentrations. beyond observations of the hvsr curves in three trends (figure 4), it is difficult to discern the topographic effect on the distribution of the mapped elastic parameters. low values of and in the peripheral part of the maps (figure 6) could be interpreted as the seismic response of non-compacted material associated with landslides. future work could collect data on hillside areas, even by more extended sampling periods and instruments with more bandwidth, to verify if the combined effect of topography and permafrost presence introduces imprints in the seismic response. discarding the topographic effect in the study area, the almosthorizontal stratigraphy in the northern sector of the island offers an interesting constraining to validate the hvsr method’s hypothesis regarding a subsurface structure composed of viscoelastic layers laterally homogeneous. in addition, the trends of vp vs. vs showed in figure 5 illustrate that in shallow strata persist saturation of gases, something not observed in the plot qp vs.qs . it means that gas saturation mainly affects the elastic properties and, in a lesser proportion, the anelastic properties. in contrast, in deeper strata, under the permafrost base, where the water reservoir’s location is calculated (kneisel & hauck, 2008), the primary mechanism of seismic dissipation is anelastic. the plot of (vp/vs) 2 vs. qp/qs shows these effects nicely. according to borzotta and trombotto (2004), the abundance of unfrozen water in frozen soils promotes variation in subsurface geophysical properties. hence, having detected the possible interface where permafrost’s basement has a temperature assumed as 0°c, the authors speculate as an additional worth of the hvsr method applied in the study area that the spatial distribution of elastic and anelastic properties may reflect the thermal structure of this layer in the study area. figure 6. maps of vp, vs , and vp,/ vs ratios estimated at 1.0 m and 35.0 m depth. high values of could vp, be related to permafrost layers. permafrost could be associated at 35 m-depth with patches of high values of elastic properties. 6. conclusions using the estimates of h/v ratios on 67 seismic-noise stations, we calculated elastic and anelastic properties in the uppermost layers at the north sector of seymour-marambio island. the h/v spectral ratios distribution suggests the permanent interaction on the uppermost sedimentary layers due to maintenance works, trampling, impacts, and high loads (e.g., the permanent landing and taxi of large aircraft on the runway during decades) may contribute to changes in the seismic dynamic response. the mapping of elastic properties at 1.0 m and 35.0 m depth allowed identifying lateral variations of the study area’s permafrost seismic properties. at those depths, we hypothesize that permafrost is emplaced in rocks with patches of porosities saturated with fluids in different phases and critical points, e.g., with abnormal saturated accumulations of gases like ch4 and co2. in shallow strata, the saturation of gases affects mainly the elastic properties. in contrast, in deeper strata, where the location of water reservoirs is expected, the primary mechanism of seismic dissipation is anelastic. 7. data and resources this study is based on seismic microtremors’ data acquired during the colombian and argentinean scientific teams’ campaign in january 2020 near the marambio scientific base of argentina. for more information and access to the dataset, contact to authors of this paper. all data processing and plotting were done using the open software geopsy (wathelet et al., 2020), openhvsr (bignardi et al., 2016, 2018), and the academic license of matlab® r2020b of mathworks. acknowledgments cav and amg thank instituto antártico argentino and comando conjunto antártico for the logistic support. cav and jms thank universidad nacional de colombia, departamento de geociencias for helping support this research. the authors are grateful to staff members of the laboratorio antártico multidisciplinario marambio (lambi) for taking care of the permanent instrumentation. the authors also thank officers, crew, technicians, and the 203 seismic properties of the permafrost layer using the hvsr method in seymour-marambio island, antarctica fuerza aérea argentina science party that take care of the marambio station. amg is a member of the carrera del investigador científico, conicet. references abu zeid, n., bignardi, s., caputo, r., mantovani, a., tarabusi, g., & santarato, g. (2014). shear-wave velocity profiles across the ferrara arc: a contribution for assessing the recent activity of blind tectonic structures. in: proceedings of the 33rd gngts national convention, vol. 1. pp. 117–122. abu zeid, n., corradini, e., bignardi, s., nizo, v., & santarato, g. (2017). the passive seismic technique ‘hvsr’ as a reconnaissance tool for mapping paleo-soils: the case of the pilastri archaeological site, northern italy. archaeological prospection, 24(3). https://doi.org/10.1002/ arp.1568. barton, n. (2007). rock quality, seismic velocity, attenuation, and anisotropy. taylor & francis/balkema, 729 p. isbn 0-415-39441-4. bignardi, s., mantovani, a., & abu zeid, n. (2016). openhvsr: imaging the subsurface 2d/3d elastic properties through multiple hvsr modeling and inversion. computers & geosciences, 93, 103-113. https://doi.org/10.1016/j.cageo.2016.05.009 bignardi, s., yezzi, a. j., fiussello, s., & comelli, a. (2018). openhvsr processing toolkit: enhanced hvsr processing of distributed microtremor measurements and spatial variation of their informative content. computers & geosciences, 120, 10-20. https://doi.org/10.1016/j.cageo.2018.07.006 borzotta, e., & trombotto, d. (2004). correlation between frozen ground thickness measured in antarctica and permafrost thickness estimated on the basis of the heat flow obtained from magnetotelluric soundings. cold regions science and technology, 40(1-2), 81-96. https://doi.org/10.1016/j.coldregions.2004.06.002 carcione, j. m., & seriani, g. (1998). seismic and ultrasonic velocities in permafrost. geophysical prospecting, 46(4), 441–454. https://doi.org/10.1046/ j.1365-2478.1998.1000333.x carcione, j. m., picotti, s., francese, r., giorgi m., & pettenati, f. (2017). effect of soil and bedrock anelasticity on the s-wave amplification function. geophysical journal international, 208(1), 424-43. https://doi. org/10.1093/gji/ggw402 castellaro, s., & mulargia, f. (2009). the effect of velocity inversions on h/v. pure and applied geophysics, 166, 567–592. https://doi.org/10.1007/ s00024-009-0474-5 fournier, h.g., buk e.m. & corte a.e. (1990). three permafrost conditions indicated by geophysical soundigns in tertiary sediments in seymour island, antarctic peninsula. cold regions science and technology 17: 301-307fukuda, m., strelin, j., shimicawa, k., takahashi, n., sone, t. & trombotto, d. (1992). permafrost ocurrence of seymour island and james ross island, antarctic peninsula. recent progress in antarctic earth science, 1, 745-750. tokyo. herak, m. (2008). modelhvsr—a matlab® tool to model horizontal-to-vertical spectral ratio of ambient noise. computers & geosciences, 34(11), 15141526. https://doi.org/10.1016/j.cageo.2007.07.009 hong, e., perkins, r., & trainor, s. (2014). thaw settlement hazard of permafrost related to climate in alaska. arctic, 67(1), 10. kato, k., corte, a.m. & fukuda, m. (1990). chemical and isotopic characteristics of ice from an ice-wedge in seymour island (isla v. com. marambio), antarctic peninsula region. proc. nipr symp. antarct. geosci., 4, 181–190. konno, k., & ohmachi, t. (1998). ground-motion characteristics estimated from spectral ratio between horizontal and vertical components of microtremor. bulletin of the seismological society of america, 88(1), 228– 241. https://doi.org/10.1785/bssa0880010228 köhler, a., & weidle, c. (2019). potentials and pitfalls of permafrost active layer monitoring using the hvsr method: a case study in svalbard. earth surface dynamics, 7, 1–16. https://doi.org/10.5194/esurf-7-1-2019 kneisel, c., & hauck, c. (2008). electrical resistivity tomography. in c. hauck & c. kneisel (eds.). applied geophysics in periglacial environments. cambridge university press. online isbn: 9780511535628. https://doi. org/10.1017/cbo9780511535628 koulakov, i. & vargas, c.a. (2018). evolution of the magma conduit beneath the galeras volcano inferred from repeated seismic tomography. geophysical research letters 45 (13), doi: 10.1029/2018gl078850. krautblatter, m., & hauck, c. (2007). electrical resistivity tomography monitoring of permafrost in solid rock walls. journal of geophysical research, 112:f02s20. doi:10.1029/2006jf000546. krautblatter, m., & draebing, d. (2014). pseudo 3-d p wave refraction seismic monitoring of permafrost in steep unstable bedrock. journal of geophysical research, 119: 287–299. doi:10.1002/2012jf002638 lee, m. w. (2003). velocity ratio and its application to predicting velocities. u.s. geological survey bulletin 2197. https://pubs.usgs.gov/bul/b2197/ b2197-508.pdf lunedei, e. & albarello, d. (2009). on the seismic noise wavefield in a weakly dissipative layered earth. geophysical journal international, 177, 10011014. doi:10.1111/j.1365-246x.2008.04062.x. lunedei, e. & albarello, d. (2010). theoretical hvsr from the full wave field modelling of ambient vibrations in a weakly dissipative layered earth. geophysical journal international, 181, 1093–1108. https://doi.org/10.1111/j.1365-246x.2010.04560.x lunedei, e. & albarello,d. (2015). horizontal-to-vertical spectral ratios from a full-wavefield model of ambient vibrations generated by a distribution of spatially correlated surface sources. geophysical journal international 201(2), 1140–1153 . http://dx.doi.org/10.1093/gji/ggv046.mantovani, a., abu-zeid, n., bignardi, s., & santarato, g. (2015). a geophysical transect across the central sector of the ferrara arc: passive seismic investigations part ii. proceedings of the 34th gngts national convention, 1. pp. 114–120. martens, j., wild, b., muschitiello, f., o’regan, m., jakobsson, m., semiletov, i., dudarev, o. v., & gustafsson, ö. (2020). remobilization of dormant carbon from siberian-arctic permafrost during three past warming events. science advances, 6, eabb6546. doi: 10.1126/sciadv.abb6546 melvin, a. m., larsen, p., boehlert, b., neumann, j. e., chinowsky, p., espinet, x., martinich, j., baumann, m. s., rennels, l., bothner, a., nicolsky, d. j., & marchenko, s. s. (2017). climate change damages to alaska public infrastructure and the economics of proactive adaptation. proceedings of the national academy of sciences of the united states of america, 10;114(2), e122-e131. https://doi.org/10.1073/pnas.1611056113 montes, m., nozal, f., santillana, s., marenssi, s. & olivero, e. (2013). mapa geológico de isla marambio (seymour); escala 1:20.000. 1ª edición. serie cartográfica geocientífica antártica. con texto complementario. madrid-instituto geológico y minero de españa; buenos aires, instituto antártico argentino. nakamura, y. (2000). clear identification of fundamental idea of nakamura’s technique and its applications. proceedings of the 12th world conference on earthquake engineering, new zealand, 8 pp. nakamura, y. (1989). a method for dynamic characteristics estimation of subsurface using microtremor on the ground surface. quarterly report of railway technical research, 30, 25–33. nogoshi, m., & igarashi, t. (1971). on the amplitude characteristics of microtremor (part 2). journal of the seismological society of japan, 24, 26–40 [in japanese]. overduin, p. p., schneider von deimling, t., miesner, f., grigoriev, m. n., ruppel, c., vasiliev, a., lantuit, h., juhls, b., & westermann, s. (2019). submarine permafrost map in the arctic modeled using 1‐d transient heat flux (supermap). journal of geophysical research-oceans, 3490-3507. https://doi.org/10.1029/2018jc014675 parolai, s. & galiana-merino, j. j. (2006). effect of transient seismic noise on estimates of h/v spectral ratios. bulletin of the seismological society of america, 96(1), 228–236. doi: 10.1785/0120050084. https://doi.org/10.1016/j.cageo.2016.05.009 https://doi.org/10.1016/j.cageo.2016.05.009 https://doi.org/10.1046/j.1365-2478.1998.1000333.x https://doi.org/10.1046/j.1365-2478.1998.1000333.x https://doi.org/10.1007/s00024-009-0474-5 https://doi.org/10.1007/s00024-009-0474-5 https://doi.org/10.5194/esurf-7-1-2019 https://doi.org/10.1017/cbo9780511535628 https://doi.org/10.1017/cbo9780511535628 https://pubs.usgs.gov/bul/b2197/b2197-508.pdf https://pubs.usgs.gov/bul/b2197/b2197-508.pdf https://doi.org/10.1111/j.1365-246x.2010.04560.x https://doi.org/10.1111/j.1365-246x.2010.04560.x https://doi.org/10.1073/pnas.1611056113 https://doi.org/10.1029/2018jc014675 204 carlos a. vargas, juan m. solano, adriana m. gulisano, sergio santillana, edwin a. casallas picotti, s., francese, r., giorgi, m., pettenati, f., & carcione, j. m. (2017). estimation of glacier thicknesses and basal properties using the horizontal-to-vertical component spectral ratio (hvsr) technique from passive seismic data. journal of glaciology, 63(238), 229(248), https://doi. org/10.1017/jog.2016.135 prasad, m. (2002). acoustic measurements in unconsolidated sands at low effective pressure and overpressure detection. geophysics, 67(2), 405–412. https://doi.org/10.1190/1.1468600 priolo, e., romanelli, m., barnaba, c., mucciarelli, m., laurenzano, g., dall’olio, l., abu zeid, n., caputo, r., … di bartolomeo, p. (2012). the ferrara thrust earthquakes of may-june 2012: preliminary site response analysis at the sites of the ogs temporary network. annals of geophysics, 55(4), 591–597. doi:10.4401/ag-6172. raynolds, m. k., walker, d. a., ambrosius, k. j., brown, j., everett, k. r., kanevsky, m., kofinas, g. p., romanovsky, v. e., shur, y., & webber, p. j. (2014). cumulative geoecological effects of 62 years of infrastructure and climate change in ice-rich permafrost landscapes, prudhoe bay oilfield, alaska. global change biology, 20(4), 1211-24. doi: 10.1111/ gcb.12500. sánchez-sesma, f. j., rodríguez, m., iturrarán-viveros, u., luzón, f., campillo, m., margerin, l., garcía-jerez, a., suarez, m., santoyo, m. a., & rodríguez-castellanos, a. (2011). a theory for microtremor h/v spectral ratio: application for a layered medium. geophysical journal international, 186, 221–225. https://doi.org/10.1111/j.1365-246x.2011.05064.x skvortsov, a. g., sadurtdinov, m. r., & tsarev, a. m. (2014). methods for permafrost studies seismic criteria for identifying frozen soil. earth’s cryosphere, 18(2), 75–80. talha qadri, s. m., nawaz, b., sajjad, s. h. & ahmad sheikh, r. (2015). ambient noise h/v spectral ratio in site effects estimation in fateh jang area, pakistan. earthquake science, 28(1), 87–95. https://doi.org/10.1007/ s11589-014-0105-9 thienen-visser, k., pruiksma, j. p., & breunese, j. n. (2015). compaction and subsidence of the groningen gas field in the netherlands. proceedings of the international association of hydrological sciences, 372, 367–373. doi: 10.5194/piahs-372-367-2015 timur, a. (1968). velocity of compressional waves in porous media at permafrost temperatures. geophysics, 33(4), 584–595. https://doi. org/10.1190/1.1439954 tsai, n. c. (1970). a note on the steady-state response of an elastic halfspace. bulletin of the seismological society of america, 60, 795–808. tsai, n.c. & housner, g.w. (1970). calculation of surface motions of a layered halfspace. bulletin of the seismological society of america 60, 1625– 1651. vandenberghe, j. (2016). from planation surfaces to river valleys. bulletin de la société géographique de liège, 67, 93–106. doi: 10.25518/07707576.4364 vargas, c. a. & torres, r. (2015). three-dimensional velocity structure of the galeras volcano (colombia) from passive local earthquake tomography. journal of volcanology and geothermal research 301, 148-158. https:// doi.org/10.1016/j.jvolgeores.2015.05.007. vitale, g., greco, l., d’alessandro, a., & scudero, s. (2018). bandwidth extension of a 4.5 hz geophone for seismic monitoring purpose. ieee international conference on environmental engineering, 1-5. doi:10.1109/ ee1.2018.8385253. wathelet, m., chatelain, j. l., cornou, c., di giulio, g., guillier, b., ohrnberger, m., & savvaidis, a. (2020). geopsy: a user-friendly open-source tool set for ambient vibration processing. seismological research letters, 91(3), 1878-1889. doi: 10.1785/0220190360. yu, l., zhong, s., & sun, b. (2020). the climatology and trend of surface wind speed over antarctica and the southern ocean and the implication to wind energy application. atmosphere 11(1), 108. https://doi. org/10.3390/atmos11010108 https://doi.org/10.1017/jog.2016.135 https://doi.org/10.1017/jog.2016.135 https://doi https://doi.org/10.1111/j.1365-246x.2011.05064.x https://doi.org/10.1007/s11589-014-0105https://doi.org/10.1007/s11589-014-0105https://doi.org/10.3390/atmos11010108 https://doi.org/10.3390/atmos11010108 _hlk62231714 _hlk59629760 _hlk62723850 _hlk62250359 _hlk62830182 _hlk62327353 _hlk62317979 _hlk62740614 _hlk90238602 _hlk90237768 _hlk54906197 _goback earth sci. res. j. vol. 11, no. 2 (december 2007): 155-162 self-potential signals caused by eruptions of the galeras volcano – colombia greinwald, s.1; ortega, a.2; rugeles, a.3; v. seth, ips, m.1 1 bundesanstalt für geowissenschaften und rohstoffe, stilleweg 2, 30655 hannover germany 2 instituto colombiano de geología y minería (ingeominas) colombia 3 universidad de nariño – colombia abstract the national institute of geology and mining, ingeominas – colombia, and the federal institute for geosciences and natural resources, bgr – germany, perform since 1997 as a joint venture multi-parameter measurements at the galeras volcano, in the southwest of colombia. since the end of 1998 the multi-parameter station at galeras (estación multiparámetro del galeras – emg) includes the continuous monitoring of electromagnetic variations. the electromagnetic (em) station is located at the north-eastern foot walls of the central cone inside the caldera. during almost six years of electromagnetic monitoring, the data did not show significant variations of the electromagnetic field, which could be related to the volcanic activity. in july 2004 a new active period of galeras began with two strong ash emissions. during both emissions strong self potential(sp) signals were recorded lasting for several hours. the present paper will present the data which could give some indications on the movements of liquids during ash emissions. key words: ingeominas; bgr; galeras; electronic monitoring. resumen el instituto nacional de geología y minas, ingeominas – colombia, y el instituto federal para las geociencias y recursos naturales, bgr – alemania, funcionan desde 1997 como una empresa de riesgo compartido tomando medidas multiparamétricas en el volcán de galeras, al sudoeste de colombia. desde finales de 1998, la estación del multiparamétrica del galeras (emg) incluye el monitoreamiento continuo de las variaciones electromagnéticas. la estación electromagnética (em) está situada en las paredes nororientales del cono central dentro de la caldera. durante casi seis años de supervisión electromágnetica, los datos no demostraron variaciones significativas del campo electromagnético, que se podrían relacionar con la actividad volcánica. en julio del 2004 un nuevo período de actividad del galeras comenzó con dos emisiones fuertes de ceniza. durante ambas emisiones, fuertes señales auto-potenciales (sp) fueron registradas por varias horas. este artículo presentará datos que podrían dar algunas indicaciones sobre los movimientos de los líquidos durante las emisiones de ceniza. 155 earth sciences research journal manuscript received october 13 2007. accepted for publication december 20 2007. palabras claves: ingeominas; bgr; galeras; monitoreamiento electrónico. introduction galeras volcano is located in the southwestern part of colombia, in the province of nariño, about 10 km to the west of the city of pasto, the capital of the province of nariño with about 350 000 inhabitants (fig. 1). galeras is a stratovolcano with a caldera of about 4.5 km diameter. a cone shaped crater with about the same height as the caldera rim is located at its center. the caldera is broken and open to the west. galeras is one of the most active volcanoes of colombia. in 1997 a cooperation project between bgr of germany and ingeominas of colombia started. aim of the project was a constant geophysical monitoring of the galeras volcano. the last eruption before occurred in 1993 and caused the death of 9 people which were inside the caldera at the time of the eruption. for that purpose a multi-parameter station was installed at the volcano comprising three broadband seismic stations, an em station inside the caldera, a gas-chemical station, a weather station and regular overflights with a thermalcamera. more information on the concept of the multi-parameter station can be found in seidl et al. (2003). from the start of the emg the volcanoes activity was very low and was further decreasing in the following years. since 2004, however, the volcanic activity is rising. strong external indications began in july 2004 with two ash fall emissions and since that time the activity level has increased. the data, which will be presented in this paper, are connected with the july 2004 emissions. description of the electromagnetic station inside the caldera of the galeras the em station was set up toward the north of the cone of the crater (fig. 2). it was located in a relative flat part at the foot wall of the active cone, at an elevation about 150 m below the rim of the caldera and about 120 m below the peak of the cone. the instruments are powered by batteries and solar panels and therefore independent of regular visits. the electric field of the ground is measured in two directions, north south (ns) and east – west (ew). the ns direction runs more or less perpendicular to the contour lines and in direction of the inclination of the cone walls at the point of the em station while the ew direction runs more or less parallel to the contour lines. the electrodes used were non-polarizable copper-sulphate electrodes. the electrode separation for each component was 100 m. in addition to the electric field, the three components of the magnetic field are recorded, ns, ew and vertical, using a fluxgate magnetometer, with a sensitivity of 0.1 nanot. the data of the two electric channels and the three magnetic channels are digi-tized with a 24 bit guralp digitizer with 50 samples per second for each channel (fig. 3) and are automatically transmitted to the observatorio volcanológico de pasto (ovp) and registered there on hard disc. description of the data figure 4 shows a continuous record of spdata from july 4th to july 25th 2004. only one interruption occurs, lasting from midday july 18th july 19th and was caused by a power failure due to ash fall after the first eruption. after clearing the solar panels the station worked again. the end of the measurements on july 25th was caused by the ash falls on the second eruption. at that time it was too dangerous to visit the station and clean it again. later on the station was completely destroyed by rock fall. the sp data of galeras contain a wide spectrum of time variations. the main sources for these variations are changes of the magnetic field in the ionosphere caused by solar activity. due to the position of the galeras, just about 1° north of the equator, the magnetic field shows constantly and very strong the 24 hour period of the equatorial electro-jet. many other effects induced by the ionosphere are also clearly visible. the average amplitudes of the daily variations are less than 100 µv/m. all short period variations (shorter than 3 minutes), however, are eliminated in the self-potential signals caused by eruptions of the galeras volcano – colombia 156 greinwald et al. esrj vol. 11, no. 2. december 2007 157 figures presented, which are based on data with a reading frequency of one sample per minute. other signals visible all the time are cultural effects caused by technical installations inside the city of pasto in a distance of less than 10 km or by a power line which leads up to the top of the galeras to feed a tv transmitter and some communication systems of the colombian army and police. these signals show constantly irregular variations. during night time the amplitudes are much smaller than during the day due to the reduced human activity. figure 5 represents the changes of the magnetic field for the same time period. here the all dominating effect is the equatorial electro jet. all cultural effects are much smaller than in the electric field. one effect deserves being mentioned: the sudden and strong change of the amplitude in all components on july 16 about one hour before the beginning of the volcanic eruption at that day. it signals the start of a world wide magnetic storm. figure 6 shows a comparison of the magnetic field of this day recorded at the galeras em-station and at the magnetic observatory in kourou (french guyana). this sudden magnetic effect is visible clearly in both stations. while some authors, e. g. yoshino and tomizawa (1989) have reported of electromagnetic precursors to volcanic eruptions, this is an effect clearly without any connection to the eruption of galeras. signals during the ash fall emissions emission of july 16th (fig. 7) on july 16th 2004 the first ash fall emission occurred after a quiet period of more than 11 years. following a rise in the seismic activity, the first ash emissions started on july 16th at 14:01 utc. the seismic activity continued with a constant tremor for several hours. it is difficult to determine the exact end of the tremor, but most probably occurs at about 18:00 utc. in figure 7 the sp-signals are shown together with the signals of the seismic activity, calculated from the vertical component of the seismic station anganoy (location see fig. 2). this information is given to demonstrate the correlation be-tween the seismic and the electric effects. the sp-signals start at the same time as the seismic tremor, in both compo-nents of the electric data. at the ns-component follows a rise in signal amplitude of about 450 μv/m. it continues with a flat top for about 4 hours and finally decays for 8 hours. in the ew-component we see continuous changes of the signal amplitudes up to 100 μv/m. these are followed by a decay of voltage of 120 μv/m. at the same time (17:45 utc) both components of the electric field start to move towards there original level before eruption. the ew-component ends at the same level while the ns-component remains at an offset of about 200 μv. while the short-period features in both components look similar, the figure 1. map of colombia with the location of the galeras volcano in the southwest of colombia. main difference are the much stronger short time movements (periods less than 1.0 hours) in the ew-component. the sp-signals indicate an end of the emission of liquids and perhaps also of the ash eruption at about 17:45 utc. emission of july 21th (fig. 8) the second ash fall on july 21th starts with a sudden rise in the seismic activity at 21:16 utc. the seismic activity stays at a high level until it stops at 3:25 utc on july 22th. this seismic activity consists of a continuous series of tremor pulses. the behaviour of the sp-channels during this ash fall emission appears quite different from the first one. while the start of the seismic activity does not seem to influence visibly the nscomponent, the ew-component shows a strong spike like feature. immediately after the spike the ew-component starts with small short period variations (periods of less than 0.5 hour). at about 22:00 utc additional medium frequency signals are visible (period 1.0 – 1.5 hours). these voltage variations stop about one hour after the end of the high seismic activity level. the total effect in the ew-component appears similar to the one of this component during the first eruption, beginning at a certain level followed by a series of up and down movements and finally ending at more or less the same level as before the event. this time the ns-component shows smaller amplitudes than the ew-component(about half the size compared with the the first eruption). further the component does not mark an increase of the measured voltage during the first part of the eruption but stays at a relatively low level, however, showing medium period voltage changes (period about 1 hour). only at relatively late time during the eruption a sudden rise in voltage occurs. at the end of the high seismic activity level the voltage begins to decay to low values for about five hours down to a level similar to the one at the beginning of the ash fall. results sp-data from a period of almost one month is shown. during this time only two data intervals have been recorded which show in a very obvious way the influence of the volcanic activity. during several years of measurements at the galeras such effects never have been observed on the quiet volcano. these anomalies show very strong amplitudes up to 500 µv/m (compared with about 100 µv/m without such effects). they are strongly correlated with the level of seismic activity. they start at the same time, however end with a certain delay to the seismic signals. the behavior of the sp-signals can possibly be ex-plained by the movements of liquids during the emission. one indication of this is that the shorter sp-variations are associated with the high seismic activity levels. the delay of the spsignals beyond the end of the seismic activity level could be due to the time delay until the movements of liquids has settled after the last transportation of water or vapor to the surface. the differences in the sp-signals between the two ash fall emissions are probably due to the different seismic activity levels which we can seen in the seismic data. at the moment we have very little information on the course of the eruptions and its correlation with the seismic activity. an interesting feature is the different behaviour of both sp-components. this could be caused by the direction of the two components of the sp-measurements in relation to the inclination of the walls of the cone through which the eruption rises and which influences the flow of water during and after the eruption. while the ns-component is more or less in the direction of the inclined cone wall at its lower end, the ew-component runs more or less perpendicular to this geological feature. and therefore the movements of liquids come to a stop after the last emission of liquids earlier. discussion and conclusions it is too early to draw final conclusions from the data we have seen. firstly the evaluation of the seismic and sp-data has not reached its final stage, secondly the correlation between the course of the eruptions and the seismic activity and the spsignals is not too well known. further research in this field is necessary. however, at first sight it seems that the sp-signals could contribute to a better representation of the course of an eruption in the measured data than the seismic activity data alone. aims of our further investigation will be the search for strong correlations between the volcanic activity and the changes of the sp 158 self-potential signals caused by eruptions of the galeras volcano – colombia 159 greinwald et al. esrj vol. 11, no. 2. december 2007 figure 2. map of the galeras caldera with the location of the em station and the seismic station in relationship with the main crater of the volcano. fluxgatemagnetometer+ calibration digitizer+ data transmitter em-station antenna em station antenna ovp data receiver and data storage ovp b ns b vert e ew e ns figure 3. schematic diagram of the em station at the galeras. the distance between the station and the observatorio volcanológico de pasto (ovp) is approximately 10 km. 160 figure 4. sp-signals from the em station inside the caldera of galeras. the upper curve represents the nscomponent, the lower one the ew-component. between july 4th and july 25th of 2004 only two periods show strong signals. both are related to ash emissions from the galeras. figure 5. variations of the horizontal magnetic field at the em station at galeras. the constant strong variations are caused by the equatorial electro-jet. the upper curve represents the ns-component, the lower one the ew-component. the amplitudes are given in nanot. self-potential signals caused by eruptions of the galeras volcano – colombia 161 greinwald et al. esrj vol. 11, no. 2. december 2007 figure 6. magnetic variations from the galeras em-station and the kourou observatory on july 16th , 2004. the strong sudden changes at about 13:00 utc represent the beginning of a worldwide magnetic storm and are not related to volcanic activity. the amplitudes are given in nanot. figure 7. . detailed representation of the sp-effects during the eruption of july 16th. the upper diagram shows the ns-component, the center diagram the ew-component of the sp-measurements. the lower diagram contains the seismic activity level for the time presented. remarkable is the very different behaviour of the two sp-components and the good correlation between the begin of the seismic and electric activity. 162 figure 8. plot of the sp-effects during the eruption of july 21th. the diagram follows the same order as in figure 7 with the ns-component at top, the ew-component at the centre and the seismic activity level in the lowest plot. again the two components show a characteristically different behaviour. self-potential signals caused by eruptions of the galeras volcano – colombia signals as they are demonstrated in the literature e.g. byrdina et al. 2003. references byrdina, s.,friedel, s., wassermann,j., zlotnicki, .j, 2003. self-potential varia-tions associated with ultra-long-period seismic signals at merapi volcano. geo-physical research letters, vol. 30, no.22 zlotnicki, j.,le mouel, j.l., sasai, y., yvetot, p., ardisson, m. h.; 2001. self-potential changes associated with volcanic activity. short-term signals associated with march 9, 1998 eruption on la fournaise volcano (réunion is-land). annali di geofisica, vol. 44, no. 2., april 2001. seidl, d., hellweg, m., calvache, m., gomez, d., ortega, a., torres, r., böker, f., buttkus, b., faber, e., greinwald, s. ; 2003. the multiparameter station at galeras volcano (colombia): concept and realization. journal of volcanology and geothermal research 125, is. 1-2. yoshino, t. and tomizawa, i., 1989: observation of low-frequency electromagnetic emissions as precursors to the volcanic eruption at mt. mihara during november, 1986. physics of the earth and planetary interiors, vol. 57, nos 1-2. the poor water retention capacity of sandy soils commonly aggregate soil erosion and ecological environment on the chinese loess plateau. due to its strong capacity for absorption and large specific surface area, the use of nanocarbon made of coconut shell as a soil amendment that could improve water retention was investigated. soil column experiments were conducted in which a layer of nanocarbon mixed well with the soil was formed at a depth of 20 cm below the soil surface. four different nanocarbon contents by weight (0%, 0.1%, 0.5%, and 1%) and five thicknesses of the nanocarbonsoil mixture layer ranging from 1 to 5 cm were considered. cumulative infiltration and soil water content distributions were determined when water was added to soil columns. soil water characteristic curves (swcc) were obtained using the centrifuge method. the principal results showed that the infiltration rate and cumulative infiltration increased with the increases of nanocarbon contents, to the thicknesses of the nano carbon-soil mixture layer. soil water contents that below the soil-nano carbon layer decreased sharply. both the brooks-corey and van genuchten models could describe well the swcc of the disturbed sandy soil with various nano carbon contents. both the saturated water content (θs), residual water content (θr) and empirical parameter (α) increased with increasing nano carbon content, while the pore-size distribution parameter (n) decreased. the available soil water contents were efficiently increased with the increase in nanocarbon contents. abstract keywords:soil-nano carbon mixture layer; infiltration process; soil water characteristic curves; available water content effect of nano-carbon on water holding capacity in a sandy soil of the loess plateau issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n4.66104 earth sciences research journal earth sci. res. j. vol. 21, no. 4 (december, 2017): 189-195. beibei zhou* and xiaopeng chen state key laboratory base of eco-hydraulic engineering in arid area, xi’an university of technology, xi’an 710048, china la poca capacidad de retención de agua en suelos arenosos es un factor determinante en la erosión del terreno y en el entorno ecológico de la meseta de loes, en china. debido a su capacidad de absorción y a su amplia superficie específica, se investigó el uso de materiales nanocarbonados hechos de cáscara de coco de forma que con la estabilización del suelo se pueda mejorar la retención de agua. se realizaron experimentos con muestras tomadas en el área de estudio en las cuales se mezcló una capa de materiales nanocarbonados a una profundidad de 20 cms. los análisis se realizaron a partir de cuatro parámetros establecidos por el contenido en peso de los materiales nanocarbonados (0 %, 0.1 %, 0.5 % y 1 %) y de cinco parámetros basados en el espesor de la mezcla de nanocarbonados con el suelo, que van de uno a cinco centímetros. las distribuciones de infiltración acumulada y contenido de agua frente a material sólido se determinaron al añadir agua a las muestras de suelo. a través del método de centrifugación se obtuvieron las relaciones agua-suelo (swcc, del inglés soil water characteristic curves). los principales resultados muestran que el índice de infiltración y la infiltración acumulada aumenta con proporción directa a los contenidos nanocarbonados, al igual que crece el espesor de la capa mezclada del suelo con materiales nanocarbonados. los contenidos de agua y suelo en la capa de suelo nanocarbonada se redujeron pronunciadamente. los modelos de brooks-carey y van genuchten describen las relaciones de agua-suelo en suelos arenosos irregulares con varios contenidos de nanocarbón. tanto el contenido de saturación de agua (θs), como el contenido de agua residual (θr) y el parámetro empírico (α) aumentaron proporcionalmente con el contenido de nanocarbón, mientras que el índice de distribución de los poros (n) bajó. los contenidos disponibles de agua y suelo se incrementaron eficientemente con el aumento de los contenidos nanocarbonados. resumen palabras clave: mezcla de suelo nanocarbonado; proceso de infiltración; relación agua-suelo; contenido disponible de agua. record manuscript received: 05/07/2017 accepted for publication: 16/11/2017 how to cite item: zhou, b. & chen, x. (2017). effect of nanocarbon on water holding capacity in a sandy soil of the loess. earth sciences research journal, 21(4). 189-195. doi: http://dx.doi.org/10.15446/esrj.v21n4.66104 efecto de materiales nanocarbonados en la capacidad de retención de agua en suelos arenosos de la meseta de loes g e o c h e m is t r y 190 beibei zhou and xiaopeng chen introduction the loess plateau, located in northwest china, is a key agricultural area for the country. however, it is also susceptible to some of the most severe soil and water losses on the slope in the world (bo et al., 2011; chen et al., 2007; li et al., 2015; mi et al., 2014). as a result, numerous studies have investigated the soil and water losses on the hillslope from the plateau (she et al., 2010; el-jakee et al., 2016). these studies have mainly focused on ecosystem management to reduce the soil and water losses by optimizing land use structure, which could involve steep terracing slopes, converting cultivated steep slopes to forests and grasslands, banning grazing, constructing reservoirs and check dams, etc. therefore, there is an urgent need for developing novel and effective measures to improve soil structure at the loess plateau. the implementation of these measures facilitated vegetation restoration and ecological conservation. however, even though the soil layer on the plateau is very thick, its weak structure and/or coarse texture often leads to a low soil water holding capacity. furthermore, surface runoff generated during rainstorms, which can be intense, can still reduce the effectiveness of local soil erosion and water controls. in the past decade, there has been a rapid growth in the manufacture and use of soil amendment in helping with the soil water capacity, such as biochar and charcoal. but high rates of this amendment, always 100 and 200 t/ha, which is not practically feasible at farmer level (zhou, 2006; liu and liu, 2010). nanomaterials have been used in a wide range of industrial and agricultural fields (kammann et al., 2013; nayak et al., 2013). nanocarbon has been found to improve plant growth. khodakovskaya et al. (2009) found that nanocarbon was able to penetrate the thick coat of tomato seeds and enhance water uptake into the seed, which improved seed germination and the growth of the tomato seedlings in water culture experiments. subsequently, kammann et al. (2013) reported that tomatoes produced two times more flowers and fruit when grown in soil containing carbon nanotubes than when grown in soil without carbon nanotubes. fan et al. (2012) concluded that the application of different levels of nitrogen fertilizer and nanocarbon significantly affected nitrogen use in a rice paddy field with a saline-alkali soil, and that yield increased when nitrogen was added with nanocarbon. when wu (2013) applied nanocarbon contained within two slowreleased fertilizers, rice yields and nitrogen use efficiency both increased while nitrogen losses in the surface water runoff from the paddy soil for both slowreleased fertilizers decreased when compared to the fertilizers alone. the capacity of nanocarbon to adsorb other elements is also of importance. liu and cheng (2012) found that cu2+ and cd2+ were better adsorbed onto a modified oxidized nano-carbon black sorbent. wang et al. (2009) applied nanocarbon near the roots of ryegrass and found that there were reductions in the cu and zn concentrations in the ryegrass and that the ryegrass biomass increased. khaydarov et al. (2012) reported that nanocarbon-polymer compositions were effective in removing metal ions from contaminated groundwater and soil. there have been many studies about uses of nanocarbon in agriculture. however, most of them focused on the ability of nanocarbon to adsorb heavy metals and improve crop yields. there is a scarcity of published studies on the effects of nanocarbon on soil water holding capacity, especially in sandy soils. soil water holding capacity is a major factor that can affect soil erosion by water, as well as plant growth (heathman et al., 2003; she et al., 2014).tan et al. (2014) found that nano carbon could increase the water contents effectively in columns of the disturbed soil column. these findings are important since they imply that nanocarbon would also reduce water losses due to percolation to deeper soil layers and that soil layers containing nanocarbon can hold more water for a longer time, for example, in soil near to plant roots. one of the advantages of nanocarbon made of coconut shell, when compared to metallic nanomaterials, is that carbon naturally exists which is also good for controlling carbon emission (nair, 2010). consequently, the use of nano-carbon materials in agriculture is more acceptable and has been widely propagated. therefore, in this paper, the effects of different nano-carbon contents of layers containing nanocarbon-soil mixtures on soil water infiltration and soil water characteristic curves (swcc), as well as on soil water content distributions in the soil profile after infiltration ceases were examined. since nanocarbons are expensive, we also investigated the effect of reducing the thickness of the soil-nanocarbon layers, which would, in turn, reduce the costs, on these soil water properties. figure 1. columns of sandy soil with soil-nanocarbon mixture at different nanocarbon contents and thicknesses. materials and methods soil and nano carbon the soil used in the study was an aeolian sandy soil, typical of soils on the loess plateau. it developed from loessial parent material and was characterized by a loose structure and high erodibility. the soil was collected from the shenmu soil water conservation field station, the chinese academy of sciences (110°11′-110°23e, 109°19′23″e, 38°46′-38°50e), yulin county, shaanxi province. the collection site has a semiarid area climate with a mean annual rainfall of 437 mm, of which about 70% falls in intense storms between june and september that can cause high soil and water losses. to reduce the influence of organic matter found in the topsoil, the upper 10 cm soil layer was removed first, and only soil from the 10~25 cm layer of the profile was collected. debris, which included dead leaves and animals, was removed, and the soil was airdried and passed through a 2-mm sieve. particle size analysis was determined by sieving and the pipette method (lv, 1992).the particle size distribution is as follows, sand content, 0.482 kg kg−1; silt content, 0.255 kg kg−1, which belong to the aeolian sandy soil according to the usda soil system (hillel, 1982). the nanocarbon used in this study was supplied by the hainuo nano material company (shanghai), china and the mean specifications of the particles were: 40 nm diameter; 0.99 g cm-3 density; purity 99%. infiltration experiment in this part, the effect of nanocarbon contents and thicknesses of soilnanocarbon mixture layer on infiltration was studied. the nanocarbon was mixed well with the air-dried soil samples to give four different gravimetric contents (0%, 0.1%, 0.5%, and 1.0%). then the mixtures with different thickness (1 cm, 2 cm, 3 cm, 4 cm and 5 cm) were applied to the packed soil columns (10 cm in diameter, 50 cm in length) 20 cm below the surface where is near most of the plant roots (shown in figure 1). 191effect of nano-carbon on water holding capacity in a sandy soil of the loess plateau the bases of all of the columns were perforated and were covered by a coarse filter paper, to permit water to drain and prevent air entrapment while retaining the soil. the surfaces of the soil mixtures were also covered with a circle of filter paper to reduce disturbance from the inflowing water. infiltration experiments were carried out by rapidly establishing and then maintaining a constant 2 cm head of deionized water on the surface of the soil using a mariotte bottle. the cumulative infiltration rate was initially determined from the volume of water supplied from the mariotte bottle measured at timed intervals. the infiltration experiments were carried out in three replications in a laboratory where the mean temperature was 28±3 °c. relative humidity was not controlled but remained at 40%±10%. the top 40cm of the soil columns at every 3cm depth increment was sampled just after the infiltration process, and the gravimetric soil water content was determined by mass loss during ovendrying at 105 °c. soil water characteristic curve (swcc) to understand the effect of nanocarbon on swcc, the air-dried sandy soil was thoroughly mixed well with the nanocarbon with five different gravimetric contents (0%, 0.1%, 0.5%, 0.7% and 1%). all the mixtures were packed into a cutting ring (100 cm3 in volume) to give a bulk density of 1.4 g cm-3. the rings with the combinations were placed in distilled water to become slowly saturated within about 12 hours. each saturated mixture was placed in a centrifuge tube. the tubes were centrifuged in a state make and model of centrifuge. to measure the water contents of the mixtures at different pressure heads, centrifugation was carried out at specific rotation speeds corresponding to known pressure heads. after attaining equilibrium at each given pressure head, the tubes were removed from the centrifuge to determine the gravimetric water content of the mixture. excess water was poured off, and the mass of the mixture was measured using an electronic balance (precision of 0.01 g). the procedure was repeated sequentially as the pressure head was increased to the highest level of 10200 cm. centrifugation was carried out at 20. at the end of the experiment, the gravimetric water content of the mixture samples was put into the ovendrying at 105°c for 24 hours. subsequently, the volumetric water contents of the different mixtures at each pressure head were determined from the dry bulk density and gravimetric water content. the swcc was then obtained by plotting the pressure head (cm) against the volumetric water content (cm•cm-3). water repellency of the tested nano carbon-soil mixture samples the soil samples were mixed well with nanocarbon contents at five different gravimetric contents (0%, 0.1%, 0.5%, 0.7% and 1%). a wdpt test (water droplet penetration test) was also used to measure water repellency (king, 19981 in this study to understand the effect of nanocarbon contents on water repellency. the penetrate time was recorded. each treatment has ten replicates. and the results are shown as follows with the mean values (in table 1). table 1. effect of nanocarbon contents on water repellency. models brooks-corey model (bc) according to brooks and corey (1964), swcc functions are given by: (1) where se is the degree of saturation; q is the volumetric water content (cm3 cm-3); qs is the saturated volumetric water content (cm3 cm-3); qr is the residual volumetric water content (cm3 cm-3); a is an empirical parameter (cm1) whose inverse is often referred to as the air entry value or bubbling pressure; h is the soil water pressure head (cm); and n is a pore-size distribution parameter affecting the slope of the retention curve. van genuchten model (vg) van genuchten (1980) found relatively simple functions could describe the swcc well: (2) where se, q, qs, qr, a, h, n, and ks are the same as in equation (1), and m is an empirical parameter. according to the statistical pore-size distribution model of mualem (1976), the parameter m is expressed as m=1-1/n (n>1). burdine’s model burdine (1953) proposed a model that described the hydraulic conductivity (k). when the bc is used in burdine’s model, the hydraulic function is given by the expression: where, l is a pore-connectivity parameter, which was assumed to be 2.0 in the original study of brooks and corey (1964); se and n are the same as in equation (1), and ks is the same as in equation (3). mualem’s model mualem (1976) described a new model for predicting the hydraulic conductivity. combined with vg, the hydraulic conductivity is given by: 1/2 ++= lnes skk (3) (4) (5) where ks, l, and se are the same as in equation (4), and m is the same as in equation (2). the pore-connectivity parameter, l, in the hydraulic conductivity function was estimated to be about 0.5 and was considered as a mean value that could be valid for many soils (mualem et al., 1976). to analyze the cumulative infiltration data, we fitted the experimental data with philip’s infiltration equation (equation (5)), the parameters of which are presented in table 2: (5)where i is the cumulative infiltration (l), t is the time (t), and s is the sorptivity (l t-1/2), a is the steady-state infiltration rate (l t-1). 192 beibei zhou and xiaopeng chen table 2. experimental treatments. results and discussion effect of soil-nano carbon mixture layers on soil water infiltration processes effect of different nano carbon contents on cumulative infiltration figure 2. effect of nanocarbon contents on soil water infiltration for 5 soil-nano carbon layer thicknesses. figure 2 presents the effect of nanocarbon contents on the soil water infiltration process. smooth cumulative infiltration curves were obtained for all nanocarbon contents and followed a similar pattern to those typical for homogeneous soils without nanocarbon. this indicated that the fine soil matrix was continuous in all cases. during the early stage of the infiltration process, there was no noticeable effect of nanocarbon content on cumulative infiltration since the wetting front had not yet entered the soil-nanocarbon mixture. when the wetting front reached the soil-nanocarbon mixtures, 20 cm below the soil surface, the nanocarbon content had significant effects on the infiltration rates of all four soilnanocarbon mixtures in each of the five-layer thicknesses. the cumulative infiltration increased with increasing nanocarbon content. this was the opposite effect of that reported by tan et al. (2014) for various mixture compositions in different soil columns that were each filled by a particular homogenous soil-nanocarbon mixture. cumulative infiltration increased at significantly faster rates when nanocarbon was present than when it was not present. this indicated that the presence of a soil-nanocarbon layer at a depth of 20 cm within the soil profile could accelerate infiltration. there was no noticeable difference between the effects of the 1% and 0.5% nanocarbon contents on cumulative infiltration. soil water infiltration could be affected greatly by hydrophobic materials which are either intermixed with the soil or coat the mineral soil particles. the infiltration curves for wettable and water repellent soils presented earlier reflect this effect (figure 2). the practical implications of water relations in these hard-to-wet soils (such as the effects on plants, runoff, and erosion) have led to numerous studies on water movement. the effect of hydrophobic substances on water movement in soils provides a basis for characterizing water repellency. effect of different thicknesses of nano carbon-soil layers on soil water infiltration incorporating nanocarbon into the soil profile at depth incurs high costs and requires more labor. therefore, it is desirable to apply the most costeffective amount of nanocarbon possible to reduce the expense of the treatment. this raises the important question as to how much nanocarbon is needed in the sandy soil. therefore, we investigated the effect of changing the thicknesses of the soil-nano carbon layers on soil water infiltration (figure 3). smooth cumulative infiltration curves were obtained for all soil-nanocarbon layer thicknesses, which followed a pattern similar to those typical for homogeneous soils without nanocarbon. this indicated that the fine soil matrix was continuous in all cases. cumulative infiltration increased at a significantly faster rate when the thickness of the soil-nanocarbon mixture increased. this also indicated that the presence of soil-nanocarbon at 20 cm depth within the soil column could accelerate infiltration of all the layer thicknesses investigated. figure 3. effect of soil-nanocarbon layer thickness on soil water infiltration for three nanocarbon contents. the effect of the three thicker soil-nano carbon layers (3, 4, and 5 cm) on cumulative infiltration was considerably greater than that of the two thinner layers (1 and 2 cm). statistically, there were no significant 193effect of nano-carbon on water holding capacity in a sandy soil of the loess plateau differences for cumulative infiltration among the mixture layers with thicknesses of 3, 4 and 5 cm. this would suggest that a layer of the soil-nanocarbon mixture that was 3 cm thick would be sufficient to improve cumulative infiltration and that increasing the thickness of the layer would not result in further improvement. to analyze the cumulative infiltration data, we fitted the experimental data with philip’s infiltration equation, the values of the fitted parameters for each of the 20 treatments are presented in table 3. philip’s infiltration equation considers both the downward movement of water, due to gravity, and advection due to matric forces. the coefficient of determination, r2, values were all above 0.99, which indicated that the equation effectively described the relation between cumulative infiltration and time during the course of our experiment. both the sorptivity and the steady state infiltration were affected similar by either increasing the nanocarbon contents or the thicknesses of the soil-nanocarbon mixture layer (table 3). figure 4. effect of nanocarbon layer content and thickness on soil water contents at ten 3-cm depth increments in the soil profile. table 3. parameters for the philip infiltration equation determined for columns with soil-nanocarbon mixture layers having various nanocarbon contents and layer thicknesses. note: r2 is the coefficient of determination for the fit of i = st 1/2 to the experimental data, where i is the cumulative infiltration, t is the time, s is the sorptivity, and a is the steady-state infiltration rate, with the experimental data the main effect of the presence of the nanocarbon is to act as a strong absorptive for water transmission. the values of sorptivity determined by philip’s equation were higher for nanocontents of 0.5% and 1% than for the soil alone and may be an indication of this increased soil water infiltration ability. during the initial stage of the infiltration process, water movement into the dry soil was faster due to the presence of the matrix potential that existed at the wetting front between the wet and dry soils. the presence of the soil-nanocarbon layer might be expected to represent a discontinuity in the soil pore system. however, the matrix potential might also be expected to increase with increasing nanocarbon contents when the wetting front reached the soil-nanocarbon layer. however, from our data, it appears that there was a critical nanocarbon content of about 0.5%, for the 3-cm thick soil-nanocarbon layer. we attributed this to a greater degree of special characteristic of nanocarbon in the soil profile. effect of different thicknesses of soil-nano carbon layers and nano carbon contents on soil water contents in the soil profile the findings of this study presented above have shown that the soilnanocarbon could increase soil water infiltration rates. this might help reduce the amount of runoff considerably from the slopes on the loess plateau. this suggestion should be investigated further in rainfall simulation experiments. however, it is not clear that the nanocarbon layer could hold sufficient soil water for crop growth on the semi-arid loess plateau. after the infiltration experiments ended, the soil in the column was sampled at every 3-cm depth increment to find out where the infiltration water was stored. the measured soil water contents of these soil samples are shown in figure 5. the soil water content in the control soil column varied little with depth, and the soil water content was relatively high even at the lower end of the soil column; this distribution of soil water contents was similar to that of the column containing the layer with a nanocarbon content of 0.1%. with the increases in the nanocarbon content and in the thickness of the soil-nanocarbon mixture layer, the soil water contents below 20 cm were considerably reduced. this was especially the case when the thickness of the soil-nanocarbon mixture was 3-5 cm. these results showed that the sandy soil with nanocarbon could hold more available water for plant growth as well as reducing runoff. a positive correlation between applied nanocarbon contents and mixture thickness was obtained. therefore, nanocarbon has the potential to serve as a soil amendment that could increase the plant available soil water reservoir. effect of nano carbon contents on swcc in the disturbed sandy soil the swcc represents the relationship between the matric potential and the water content or the degree of saturation. it is related to pore-size distribution, structure, and the nature and extent of the soil surfaces. figure 5. soil water characteristic curves for a disturbed sandy soil with different contents of nanocarbon. 194 beibei zhou and xiaopeng chen194 as the matric potential is less than zero, the soil energy state was better expressed by the pressure head (the negative value of the metric potential) to give a more readily interpreted illustration of the swcc. figure 5 presents volumetric water content-pressure head equilibrium data points. smooth swccs were obtained for the five nanocarbon contents, which followed similar patterns to that of the homogenous sandy soil without nanocarbon. at the lower pressure heads (0-102 cm), since most of the drainage water was due to gravitational forces, the differences in volumetric water content changes for soil containing the various nanocarbon contents were not evident. changes began to exhibit differences in the high-pressure head section (>102 cm) of the swcc. the volumetric water content for all treatments presented a sudden reduction, but the ranges were different. in this study, the slope dh/dq at 102 cm pressure head was used to describe the intervals. the slopes for the five nanocarbon contents between 0% and 1% were -2.23×10-3, -1.73×10-3, -1.49×10-3, -1.40×10-3, and -1.34×10-3, respectively. similarly, with increasing pressure head, reductions in volumetric water content for soils containing nanocarbon were less than those for the disturbed sandy soil alone. the phenomenon manifested itself in that the slope increased with increasing nanocarbon content and when the soil contained nanocarbon the slope became steeper. furthermore, these data points in figure 5 indicated that the range of pore sizes was greater for the soil samples that contained nanocarbon than for the soil without nanocarbon and increased with the nanocarbon content. due to its large specific surface area and ultrafine size, nanocarbon tended to increase the fine pores between soil particles by partially filling those spaces. this caused a shift in the pore-size distribution by decreasing the number of larger pores and increasing the number of smaller pores in the system. at low-pressure heads, water tended to pass through larger pores. as the pressure head increased (>100 cm), water in the macropores was emptied and only remained in the smaller pores. there were larger suctions in the finer pores. thus, water discharge for soil with nanocarbon was less than for the homogeneous soil without nanocarbon. consequently, nano carbon could improve soil water retention capability to some extent. table 4. parameters of the soil water characteristic curve for a disturbed sandy soil with various nanocarbon contents fitted by the brooks–corey (bc) and van genuchten (vg) models. note: r2 is the coefficient of determination for the fit of the vg or bc model to the experimental data; θs, saturated water content; θr, residual water content; α, empirical parameter; n, pore-size distribution parameter the minimum data requirement for swcc generation is at least five q – h pairs (nimmo and akstin, 1988). thus, to determine the effect of nanocarbon on the swcc shape and related physical properties of the disturbed sandy soil, this study used the brooks-corey (bc) (equation (2)) and van genuchten model (vg) (equation (4)) to describe the dehydration process of the swcc occurring under centrifugation. the fitted parameters of these two models are given in table 3. the r2 values for the vg and bc models relating the predicted values to the measured data values were above 0.99 and 0.98, respectively. this indicated that, as with the homogenous sandy soil alone, the swccs for the sandy soil with different nanocarbon contents could be described well by both models although the vg was slightly better fitted. from table 4, it was clear that both qs and qr were affected in a similarly increasing manner by the increases in nanocarbon contents. moreover, the increased available water contents due to nanocarbon in the soil could be calculated by qsqr, which further indicated that nanocarbon could improve the water capacity. table 5. the relationship between nano carbon content and parameters of the brooks–corey (bc) and van genuchten (vg) models. note: * p<0.05; ** p<0.01; α, empirical parameter; n, pore-size distribution parameter comparing the two models, a and n in vg were about 1.5 and 3 times larger than in bc, respectively. in both the vg and bc models, the value of a increased, except for the 0.7% nanocarbon content, and n decreased with growing nanocarbon contents. the value of a expresses the inverse of the air entry value. the increasing values of a suggested that air entry values decreased and soil water capacity increased with the increases in nanocarbon content. this further confirmed that sandy soil containing nanocarbon had more available water for plant growth. the parameter n is related to pore-size distribution. the values of n decreased with increasing nanocarbon content which indicated that the fine pores increased (table 4). due to the special characteristics of nanocarbon (i.e., the large specific surface area and high adsorption capacity), adding nanocarbon to the soil could reduce the proportion of larger pores while increasing that of the small pores. consequently, the saturated water volumetric content increased with the increases in nanocarbon content. also, it was clear that the parameters of the two models were linearly related to the nanocarbon content. conclusions this study investigated the effects of infiltration and the soil water holding capacity of adding nanocarbon to a sandy soil. the addition of nanocarbon to the sandy soil had a significant effect on soil water holding capacity, as well as on infiltration. based on the analysis of swccs, the presence of nanocarbon enhanced water retention and increased the proportion of fine pores. both the brooks-corey and van genuchten models could well describe the swccs for the disturbed sandy soil with different nanocarbon contents. in general, qs, qr and a increased with increases in nanocarbon content, while n decreased. nano carbon content was linearly related to the parameters of the vg and bc models. infiltration experiments investigated four different nanocarbon contents of a soil layer with five different thicknesses within a soil profile. cumulative infiltration increased with increasing nanocarbon contents and with the thickness of the soil-nanocarbon mixture layer. this was especially evident when the nanocarbon content was higher than 0.5%, and the thickness of the mixture layer was between 3 and 5 cm. storage of infiltrated water increased, which indicated that nanocarbon could be beneficial for plant growth and could reduce losses of water due to percolation to deep soil layers. this study provides some basic information about the effects of nanocarbon on the water holding capacity of a sandy soil on the loess plateau. it suggests that nanocarbon may provide a means by which to improve the poor structure of loess plateau soils. however, more detailed and systematic research is required to determine optimal levels of nanocarbon and in what manner it is added to the soil, as well as the effectiveness ofnano carbon over time. acknowledgements the work in this paper is supported by the national natural science foundation of china (51239009; 41371239), the science and technology planning project of shaanxi province (2013kjxx-38), the shaanxi water conservancy project (2013slt-04), and the doctoral fund of xi’an university of technology (106-211301), which are all gratefully acknowledged. 195effect of nano-carbon on water holding capacity in a sandy soil of the loess plateau 195195 references abbasi, n., & yasin, m. (2017). petrography and diagenetic history of nagri formation sandstone in district bagh and muzaffarabad, pakistan. pakistan journal of geology, 1(2), 21–23. afzal, j., ullah, n., hussain, z., rukh, s., ayaz, m., akbar, a., & zaman, a. (2017). phytochemical analysis andantibacterial potential of leaf extract of bauhinia linn.: an ethnomedicinal plant. matrix science pharma, 1(2), 17-19. ali, s.s., ijaz, n., aman, n., & noor, m. (2017). feasibility study of low density waste plastic in non-load bearing asphalt pavement in district faisalabad. earth sciences pakistan, 1(2), 14-15. brooks, r. h., & corey, a. t. (1964). hydraulic properties of porous media. hydrology papers, colorado state university. burdine, n. t. (1953). relative permeability calculation from pore size distribution data. transactions of the american institute of mining and metallurgical engineers, 198, 71-78. el-jakee, j. k., ali, s. s., el-shafii, s. a., hessain a. m., al-arfaj, a. a., & mohamed, m. (2016). comparative studies for serodiagnosis of haemorrhagic septicaemia in cattle sera. saudi journal of biological sciences, 23, 48-53. doi: https://doi.org/10.1016/j.sjbs.2015.06.011 fan, l., wang, y., shao, x., geng, y., wang, z., ma, y., & liu, j. (2012). effects of combined nitrogen fertilizer and nano-carbon application on yield and nitrogen use of rice grown on saline-alkali soil. journal of food, agriculture & environment, 10, 558-562. heathman, g. c., starks, p. j., ahuja, l. r., & jackson, t. j. (2003). assimilation of surface soil moisture to estimate profile soil water content. journal of hydrology, 207, 42–55. hillel (1982). introduction to soil physics. new york, academic press. ismail, i., husain, m. l., zakaria, r. (2017a). attenuation of waves from boat wakes in mixed mangrove forest of rhizophora and bruguiera species in matang, perak. malaysian journal geosciences, 1(2), 32-35. ismail, m.n., rahman, r., & tahir, s.h. (2017b). wave-dominated shoreline deposits in the late miocene sedimentary sequence in the miri formation north sarawak, malaysia. geological behavior, 1(2), 14–19. kammann, c., linsel, s., johannes, g. w. (2011). influence of biochar on drought tolerance of chenopodium quinoa: wild and on soil-plant relations. plant and soil, 9, 115-123. khaydarov, r. a, khaydarov, r. r, gapurova, o., & malish, r. (2012). remediation of metal ion-contaminated groundwater and soil using nanocarbon-polymer composition. in: clean soil and safe water. (springer netherlands), 167-182. khodakovskaya, m., dervishi, e., mahmood, m., xu, y., li, z., watanabe, f., & biris, a. s. (2009). carbon nanotubes are able to penetrate plant seed coat and dramatically affect seed germination and plant growth. acs nano 3, 3221-3227. doi: 10.1021/nn900887m king, p. m. (1981). comparison of methods for measuring severity of water repellence of sandy soils and assessment of some factors that affect its measurements. australian journal of soil research, 19, 275–285. doi: 10.1071/sr9810275 li, d., peng, w., ge, s., li, s., mo, b., & ohkoshi, m. (2015). groups characteristics of bioactivator extractives in three poplar woods. wood research, 60, 755-762. liu, h. b., & liu, z. l. (2010). recycling utilization patterns of coal mining waste in china. resources, reservation and recycling, 12, 1331-1340. liu, y. z, & cheng, j. m. (2012). adsorption kinetics and isotherms of cu (ii) and cd (ii) onto oxidized nano carbon black. advanced materials research, 529, 579-584. lv, r. k. (1992). nanjing agricultural university, soil agrichemical analysis. beijing: agriculture press, 37-103. (in chinese). maryam, a., aslam, s., saif, s., aslam, t., tusleem, k., qamar, m.t.u., abdullah, i., mushtaq, a., khalid, r.r., & siddiqi, a.r. (2017). statistical analysis of risk factors affecting the prognosis of biliary atresia in infants. matrix science pharma, 1(2), 20-24. mi, c., huang, y., liu, z., mi, w., zhang, z. (2014). a novel experimental teaching approach for electrical engineering based on semiphysical simulation. world transactions on engineering and technology education, 12, 779-783. mualem, y. (1976). a new model for predicting the hydraulic conductivity of unsaturated porous media. water resources research, 12, 513-522. mualem, y., & dagan g. (1976). methods of predicting the hydraulic conductivity of unsaturated soils. technion israel institute of technology. nayak, t. r., & pastorin, g. (2013). nano-carbon-based systems for the delivery of bioactive agents:. pros and cons. nanopharmaceutics: the potential application of nanomaterials, by liang xing jie (published by world scientific publishing co. pte. ltd., 2013. isbn# 9789814368674, 1, 535-569. nair, r., varghese, s. h., nair, b. g. maekawa, t., yoshida, y. & kumar, d. s. (2010). nanoparticulate material delivery to plants. plant science, 179, 154-163. nimmo, j. r., & akstin, k. c. (1988). hydraulic conductivity of a sandy soil at low water content after compaction by various methods. soil science society of america journal, 52, 303-310. shahzad, a., munir, m.u.h., yasin, m., umar, m., rameez, s., samad, r., altaf, s., & sarfraz, y. (2017). biostratigraphy of early eocene margala hill limestone in the muzaffarabad area (kashmir basin, azad jammu and kashmir). pakistan journal of geology, 1(2), 16–20. she, d. l., shao, m.a., timm, l. c., sentis, i. p., reichardt, k., & yu, s. e. (2010). impacts of land-use pattern on soil water-content variability on the loess plateau of china. acta agriculturae scandinavica, section b — soil & plant science, 60, 369–380. she, d. l, tang, s. q, shao, m. a, yu, s. e, & xia, y. q. (2014). characterizing scale specific depth persistence of soil water content along two landscape transects. journal of hydrology, 519, 1149-1161. tahir, s., siong, k.v., musta, b., & asis, j. (2017). facies and sandstone characteristics of the kudat formation, sabah, malaysia. geological behavior, 1(2), 20–25. tan, s., zhou, b., wang, q. j. (2014). effect of nano-carbon on soil water infiltration. acta pedologica sinica 51, 263-269. (in chinese) tunggolou, j., & payus, c. (2017). moringa oleifera as coagulant used in water purification process f or consumption. malaysian journal geosciences, 1(2), 29-31. usman, m., yasin, h., nasir, d.a., & mehmood, w. (2017). a case study of groundwater contamination due to open dumping of municipal solid waste in faisalabad, pakistan. earth sciences pakistan, 1(2), 12-13. van genuchten, m. t. (1980). a closed-form equation for predicting the hydraulic conductivity of unsaturated soils. soil science society of america journal, 44, 892-898. wang, h. w, wang, y. j, chen j.h, wang, s. q., cheng, j. m., zhou, d. m. (2009). application of modified nano-particle black carbon for the remediation of soil heavy metal pollution, 29, 431-436. (in chinese) wu, m. y. (2013). effects of incorporation of nano-carbon into slow-released fertilizer on rice yield and nitrogen loss in surface water of paddy soil. intelligent system design and engineering applications (isdea), 2013 third international conference on. ieee, 676-681. zhou, z. c., shangguan, z. p., & zhao, d. (2006). modeling vegetation coverage and soil erosion in the loess plateau area of china. ecological modelling 198, 263-268. earth sci. res. j. vol. 11, no. 2 (december 2007): 97-107 mapping subsurface formations on the eastern red sea coast in jordan using geoelectrical techniques: geological and hydrogeological implications awni t. batayneh. department of geology, king saud university , po box 2455 , riyadh 11451 , saudi arabia corresponding author. tel.: ++966-56-8086395. e-mail address: awni_batayneh@yahoo.com (a. t. batayneh) abstract during 2006, geoelectrical measurements using the vertical electrical sounding (ves) method were conducted on the eastern red sea coast in jordan, using the syscal-r2 resistivity instrument. the objectives of the study were (i) to evaluate the possibility of mapping of quaternary sediments medium in areas where little is known about the subsurface geology and to infer shallow geological structure from the electrical interpretation, and (ii) to identify formations that may present fresh aquifer waters, and subsequently to estimate the relationship between groundwater resources and geological structures. data collected at 47 locations were interpreted first with curve matching techniques, using theoretically calculated master curves, in conjunction with the auxiliary curves. the initial earth models were second checked and reinterpreted using a 1-d inversion program (i.e., resix-ip) in order to obtain final earth models. the final layer parameters (thicknesses and resistivities) were then pieced together along survey lines to make electrical cross sections. resistivity measurements show a dominant trend of decreasing resistivity (thus increasing salinity) with depth and westward toward the red sea. accordingly, three zones with different resistivity values were detected, corresponding to three different bearing formations: (i) a water-bearing formation in the west containing red sea saltwater; (ii) a transition zone of clay and clayey sand thick formation; (iii) stratas saturated with fresh groundwater in the east disturbed by the presence of clay and clayey sand horizons. deep borehole (131 m) drilled in the northwestern part of the study area for groundwater investigation, has confirmed the findings of the resistivity survey. key words: red sea coast; geoelectrical measurements; geology; hydrogeology; jordan. resumen durante 2006, mediciones geoeléctricas que usaban el método eléctrico vertical de sondeo (evs) fueron realizadas en la costa oriental del mar rojo en jordania, usando el instrumento de resistividad syscal-r2. los objetivos del estudio fueron (i) evaluar la posibilidad de cartografiar los sedimentos cuaternarios en áreas donde poco se sabe sobre la geología subterránea y deducir 97 earth sciences research journal manuscript received july 8 2007. accepted for publication december 10 2007. estructuras geológicas someras a partir de interpretación eléctrica, y (ii) identificar formaciones que pueden presentar acuíferos, y posteriormente estimar la relación entre recursos de agua subterránea y estructuras geológicas. los datos recolectados en 47 sitios fueron interpretados primero con técnicas para emparejar las curvas, usando curvas principales teóricamente calculadas, conjuntamente con las curvas auxiliares. los modelos iniciales terrestres fueron comprobados y reinterpretados en segundo lugar usando un programa de inversión 1-d (i.e., resix-ip) para obtener modelos finales terrestres. los parámetros finales de la capa (espesores y resistividades) entonces fueron ensamblados a lo largo de líneas de medición para hacer secciones transversales eléctricas. las mediciones de resistividad demuestran una tendencia dominante de disminución de la resistividad (además incremento de la salinidad) con la profundidad y hacia el oeste del mar rojo. por consiguiente, tres zonas con diversos valores de resistividad fueron detectadas, correspondiendo a tres diferentes formaciones portadoras: (i) una formación acuífera en el oeste que contiene el agua salada del mar rojo; (ii) una zona de transición entre arcilla y arena gruesa arcillosa; y (iii) estratos saturados con agua subterránea fresca en el este con presencia de arcilla y horizontes arcillosos arenosos. pozos (131 m) perforados en la parte noroccidental del área del estudio para la investigación de agua subterránea, han confirmado los resultados. palabras claves: costa del mar rojo; mediciones geoeléctricas; geología; hidrogeología; jordan. introduction the problem of the salination of groundwater aquifers arises in coastal areas, where the excessive pumping of unconfined coastal aquifers by water wells leads to the intrusion of sea water. this negative effect of human activity has been recorded in many areas of the world. hence, this problem is likely to arise in areas like jordan that has poor water resources (low precipitation and high evapotranspiration) and has mismanagement of water resources (e.g., batayneh, 2006; batayneh and qassas, 2006). jordan is considered as one of the ten poorest countries in water in the world. an arid climate, high natural growth rates, and forced migrations have conspired to push available water resources to the limit. annual rainfall in jordan ranges from 600 mm in the northwestern highlands to less than 100 mm in the eastern and southern regions. it is estimated that 80.6% of jordan receives less than 100 mm of rainfall per year (salameh and bannayan, 1993). assuming that the average rainfall in this area is 70 mm, dry areas in jordan receive 5 billion m3 a year. most of this water flows in small drainage basins or wadis to end up in playas (qa’s), and ultimately are lost to evaporation. due to the scarcity of boreholes in the east area of the red sea coast in jordan which could provide information on the configuration of the different water bodies, vertical electrical sounding (ves) survey, utilizing a schlumberger array configuration, syscal-r2 resistivity instrument (iris instruments, france), were performed on 47 sites for several purposes: (1) verification of the presence of the different water-bearing formations and estimation of their depth and thickness; (2) finding the relationship between the resistivity variations and the different configurations of the water-bearing formations; and (3) mapping the water table in the shallow coastal aquifer and selecting new location(s) for drilling. data from a single shallow (33 m) borehole (k1, fig. 1) drilled by the jordan phosphate mines company showed saline water (tds > 30,000 mg/l) at 32 m deep, and a 131 m deep borehole (k3028, fig. 1) drilled by the water authority of jordan for groundwater investigation encountered saline water (tds > 27,000 mg/l) at 127.5 m deep. the data were analyzed and used to correlate the results of the geoelectrical surveys (see section field measurements and methods of interpretation). the k1 shallow borehole penetrated alternating bands of gravel and sand down to a depth of about 4 m which is underlain by a unit composed by approximately 25 m clay and clayey sand sediments. the third unit corresponds to the sand/sandy clay containing saline water (saturated). data from mapping subsurface formations on the eastern red sea coast in jordan using geoelectrical techniques: geological and hydrogeological implications 98 batayneh. esrj vol. 11, no. 2. december 2007 99 the k3028 deep borehole shows that three units of sediments were found. the upper unit has approximately a thickness 36.5 m and consists of medium to coarse grained size gravel and sand. the underlying 91 m thick unit is mainly composed by clay and clayish sand. the third unit is composed by sand/sandy-clay sediments containing saline water (saturated). study area the area under study is approximately located at 30 km to the south of aqaba city in the southwestern part of jordan. it is limited to the south by the jordanian-saudi border and to the west by the red sea (fig. 1). the study area is located in an extremely arid environment with an annual average precipitation of 70 mm. rainfall generally occurs during the winter months (november to january). however, there are years where the rainfall is absent; while in other years, ephemeral floods of short duration may occur. the climate of the region is very hot in summer (april to august) with temperatures exceeding 38 ºc in summer (april to august). the area under investigation was included in the 1:50,000 national geological mapping project carried out by the geological mapping division of the natural resources authority of jordan (al khatib, 1987). this map shows that metamorphic rocks of the late proterozoic aqaba complex dominate the eastern side of the study area (fig. 1). it varies in composition from monzogranite to alkali feldspar granite. the coastal plain between the metamorphic rocks and the red sea consists mainly of quaternary continental sediments. these constitute clastics (clay, sand, and gravel) deposited in fan deltas, with some intercalations of lacustrine sediments (clay, gypsum, and aragonite) of pleistocene age (al khatib, 1987). the coastal plain area (fig. 1) is approximately 7 km long and 9 km wide and is accessible from a modern highway joining jordan with saudi arabia. the alluvial shallow aquifer is the primary source of water for domestic, municipal, and industrial use in the region. the recharge to this aquifer takes place either along the elevated areas in the east and northeast sides, or due to local surface water infiltrations. field measurements and methods of interpretations figure 1. location of ves sites (black triangle) in relation to geology. also shown is the location of two profile lines a-a’ and b-b’ and two boreholes. the elevation contours are in meters. the inset map shows the location of the investigated area on jordan. palestine grid are used. surface resistivity methods have been used for groundwater research for many years. earth resistivities are related to important geologic parameters of the subsurface including types of rocks and soils, porosity, and degree of saturation (keller and frischknecht, 1966). it was shown by parasnis (1956 and 1966) that the electrical resistivity of rocks and minerals, except for massive sulfides and graphite, vary in a wide range between 1 to 107 ohm-m, whereas coastal aquifers that are prone to saline water are identified by relatively low resistivity values. thus, saltwater can be easily distinguished from almost any combination of lithological types. resistivity methods are used to map the freshwater-saltwater interface and for studying conductive bodies of hydrogeological interest (e.g., zohdy and jackson, 1969; ayers, 1989; barongo and palacky, 1991; khair and skokan, 1998; gnanasundar and elango, 1999; mukhtar et al., 2000; batayneh, 2006). in general, the resistivity method involves measuring the electrical resistivity of earth materials by introducing an electrical current into the ground and monitoring the potential field developed by the current. the most commonly used electrode configuration for geoelectrical soundings, which was used in this field survey, is the schlumberger array. four electrodes (two current a and b and two potential m and n) are placed along a straight line on the land surface such that the outside (current) electrode distance (ab) is equal to or greater than five times the inside (potential) electrode distance (mn). vertical sounding, in schlumberger array, were performed by keeping the electrode array centered over a field station while increasing the spacing between the current electrodes, thus increasing the depth of investigation. the potential difference (δv) and the electrical current (i) are measured for electrode spacing and the apparent resistivity (ρa) is calculated by the equation: where is the geometrical factor that depends on the electrode arrangement for the schlumberger array. a total of 47 ves stations were established across the study area. the data were collected using a syscal-r2 resistivity instrument (iris instrument, france). the layout of the survey stations is superimposed on the geological map in figure 1. the locations of the ves sites were considerably restricted by logistical difficulties. the presence of narrow valleys and topography prevented a wider coverage. the maximum ab/2 spacing of the schlumberger array ranged from 15 m to 900 m. the separation of the current electrodes was = 3, 4, 6, 8, 10, 12, 16, 20, 24, 30, 40, 50, 60, 80, 100, 150, 200, 250, 300, 350, 400, 500, 600, 800, 1000, 1200, 1400, 1600, and 1800 m. the potential electrode separation was = 1, 10, 20 and 40 m. the increase of the potential electrode separation mn allowed that readings from the same current electrode spread ab with the previous and expanded mn were taken. the sounding curves were subjected to a preliminary interpretation using the partial curve matching technique by zohdy (1965), and orellana and mooney (1966). based on this preliminary interpretation, initial estimates of the resistivities and thickness (layer parameters) of the various geoelectric layers were obtained. in a second analysis method, the layer parameters derived from the graphical curve matching was then used to interpret the sounding data in terms of the final layer parameters through a 1-d inversion technique (e.g., resix-ip, interpex limited, golden, co., usa). inversion analyses of the sounding curves have been made with an average fitting error of about 5%. quantitative interpretation of geoelectrical sounding curves is complicated due to the well known principle of equivalence (van overmeeren, 1989). data from the k3028 borehole (fig. 1) was used to minimize the choice of equivalent 100 mapping subsurface formations on the eastern red sea coast in jordan using geoelectrical techniques: geological and hydrogeological implications    1m-ohm ι δv κρ a   2 mn an.am πκ  101 batayneh. esrj vol. 11, no. 2. december 2007 models by fixing thicknesses and depths to certain levels, and allowing the adjustment of resistivity. correlation between ves stations 4 and 5 and lithology from the k3028 borehole was performed (fig. 2) in order to determine the electrical characteristics of the rock units with depth. based on the lithological log from the k3028 borehole, the geological interpretation of the geoelectrical model for ves 4 and ves 5 (fig. 2) is: (i) a resistive layer (500-1700 ohm-m) with variable thickness and moisture content at the surface, which consists of a mixture of gravel, silt, and sand; (ii) a thick clay and clayish sand layer having resistivity values of 30 to 40 ohm-m; and (iii) a saturated sandy layer with saline water having resistivity values of 1 to 10 ohm-m. field results analysis of ves curves due to the distinctive characteristics features in the field of the apparent resistivity curves, the ves stations show four types of curves: type i, type ii, type iii, and type iv (figure 3). these types were defined in terms of the number of geoelectrical layers and their respective resistivity relationships. among the four types of curves, type i and type ii curves show similar shape of field curves with layer resistivities decreasing with depth such that ρ1>ρ2>ρ3. such curve behaviour undoubtedly proves the presence of a low-resistivity layer at the bottom of the section. of the 47 field curves, 19 field curves at ves stations 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 20, 21, 22, 28, 29, 30, 38, and 39 were classified as type i (fig. 3a), and 15 of the field curves at ves stations 12, 13, 18, 19, 24, 25, 26, 27, 36, 37, 43, 44, 45, 46, and 47 were classified as type ii (fig. 3b). eight of the field curves at ves stations 15, 16, 17, 31, 32, 33, 34, and 35 were classified as type iii and reflect the presence of five geoelectric layers where the layers resistivity relationship is ρ1>ρ2<ρ3>ρ4<ρ5 (fig. 3c). the remaining five field curves at ves stations 14, 23, 40, 41, and 42 made in the eastern and southeastern parts of the study area were classified as type iv and reflect the presence of three geoelectric layers where the layers resistivity relationship is ρ1>ρ2<ρ3 (fig. 3d). it is to be noted that the apparent resistivity values representing the surface layer in the four types of field curves vary considerably at different ves locations, mainly due to the fact that resistivity depends on the soil moisture content. the analysis of each type of curve in relation to groundwater in the study area is discussed in detail in the following sections. type i the type i curve (fig. 3a) generally shows: (1) a relatively thin surface layer of coarse grained loose sand, gravel, and sand dune existing below the ground surface having apparent resistivity of 300 to 1800 ohm-m, followed by (2) the presence of sand, clay, sandy clay, and gravel of varying grain size (fine to medium) of 50300 ohm-m apparent resistivity; and (3) a lowresistivity third layer with apparent resistivity < 12 ohm-m, indicating saline water. the type i curves were obtained from soundings measured in the western side of the study area (fig. 1). type ii the type ii curve (fig. 3b) is composed by (1) a thin layer of medium to coarse grained sand and gravel (surface layer) with an apparent resistivity of about 250 to 1300 ohm-m, (2) a second fine to medium grain size sand and gravel layer (100500 ohm-m), and (3) a relatively low-resistivity third layer (20-100 ohm-m) indicating a clay and clayish sand formation. the fields curves of type ii are obtained in the north-northeast trending narrow zone located to the east of the type i curve (fig. 1). type iii the type iii curves (fig. 3c) were obtained from soundings taken from profiles ii and iv. apparent resistivity values of these curves are relatively low, when compared to the other three types. the type iii field curves describe qualitatively a model composed of five layers where the layer resistivity relationship is ρ1>ρ2<ρ3>ρ4<ρ5. 102 mapping subsurface formations on the eastern red sea coast in jordan using geoelectrical techniques: geological and hydrogeological implications figure 2. correlation between sounding curves and borehole well lithology. figure 3. apparent resistivity curves 103 from the top to the bottom, these are: (1) a sandy, clayish sand, and gravel of fine grained size having apparent resistivity of 60 to 250 ohm-m, followed by (2) the presence of clay and clayish sand formation (40-80 ohm-m), a (3) saturated sand with good groundwater quality (100 to 200 ohm-m), (4) a clay and clayish sand formation (40-80 ohm-m), and (5) a sandy formation saturated with fresh groundwater (80200 ohm-m). type iv the type iv sounding curves (fig. 3d) describe qualitatively a model composed by (1) a relatively thin surface layer with an apparent resistivity of 200 to 1300 ohm-m of medium to coarse grained material with a variable moisture content, followed by (2) the presence of a relatively low-resistivity second layer (4080 ohm-m), indicating a clay and clayish sand formation, and (3) a sandy formation saturated with fresh groundwater (100-200 ohm-m). the apparent resistivity values increase with depth from 100 to 200 ohm-m depending on the saturated condition. figure 4 is a typical sounding data plot and bestfit model for four sounding data corresponding to the four types of field curves (i, ii, iii, and iv; fig. 3). it is of interest to note that the soundings of fig. 4a were derived from the western part (site 1, type i); fig. 4b is based on sounding (site 37, type ii); fig. 4c is based on sounding (site 33, type iii); fig. 4d is based on sounding in the southeastern (site 41, type iv) side. the best-fit models are derived along geoelectrical soundings points with lower and upper bound models (1-5% error). to the left of fig. 4, it is shown the schlumberger apparent resistivity curve with data (points) superimposed on the best match 1-d inversion (solid line). to the left of fig. 4, the interpreted results in terms of resistivity and depth together with the allowable of equivalence (dashed lines) are shown. electrical cross sections the results from the 1-d inversion of the geoelectrical soundings were compiled and plotted along two typical profiles: a-a’ and b-b’ (fig. 1). in these sections, the thickness of the subsurface layers and their geological structures are given. two interpreted resistivity vs. depth cross sections are shown in figs. 5 and 6. cross section a-a’ (fig. 5) represents the behaviour of the northern side of the studied area. the distribution of the resistivity along the geoelectrical profile indicates the presence of two zones with different properties: the first zone is from sounding point 1 to 11 corresponding to type i field curves and the second is from 12 to 13 representing type ii field curves. as seen in fig. 5, the first zone shows resistivity values in the range 400-1800 ohm-m at the surface, which are considered caused by coarse grained sand, gravel, and sand dune at 3-50 m depth. in general, the layer thickness increases with increasing distance from the coast, followed by the presence of a relatively conductive second clay and clayish sand layer, ranging from 20 to 100 ohm-m, at shallower depths and thickening landwards. the top of this layer is above mean sea level (msl), by several hundred meters at sites 7, 8, 9, 10, and 11 (fig. 5), followed by conductive layer with resistivity values of 1 to 11 ohm-m corresponding to saline water. the depth of this conductor is closed to msl at coastal sites and show increasing depth with increasing distance from the coast. in the second zone shown in fig. 5 (sites 12 and 13), the range is from 400-600 ohm-m and 5 to 15 m thick at the surface, which are considered caused by sand and gravel of varying grain size, followed by the presence of sand and sandy clay having resistivity values of 200 to 300 ohm-m and 10 to 30 m thick, followed by the presence of a relatively conductive third clay and clayish sand layer that has a resitivity value of 40 ohm-m at shallower depths. cross section b-b’ (fig. 6) represents the behaviour of the southern part of the studied area. the cross section shows three different zones with different properties: the first zone is from sounding point 39 to 38 (type i field curves), while the second zone from 37 to 36 (type ii field curves) and the third zone from 35 to 31 (type iii field curves). the first zone shows resistivity values in the range 500-600 ohm-m at the surface, considered to have been caused batayneh. esrj vol. 11, no. 2. december 2007 104 mapping subsurface formations on the eastern red sea coast in jordan using geoelectrical techniques: geological and hydrogeological implications by coarse grained sand, gravel and sand dune at 10-20 m depth, followed by the presence of a relatively conductive second clay and clayish sand layer, ranging from 30 to 90 ohm-m, at shallower depths and thickening landwards, followed by conductive layer with resistivities of 2 to 3 ohm-m corresponding to saline water. in the second zone (fig. 6, sites 37 and 36), the range is from 200-600 ohm-m and 40 to 50 m thick at the surface, considered to have been caused by sand and gravel of varying grain size (fine to medium), followed by the presence of sand and sandy clay having resistivity values of 80 to 100 ohm-m and about 100 m thick, followed by the presence of a relatively conductive third clay and clayish sand layer that has a 30 ohm-m resistivity value. in the third zone (fig. 6, sites 35, 34, 33, 32 and 31), the main features of the derived structure, from the surface downward, may be summarized as follows: (1) a relatively thin surface layer of fine to medium sand, gravel, and sand clay that has a 100-300 ohm-m resistivity value and 5-15 m thick, followed by (2) the presence of clay and clayish sand having resistivity values of 50 to 90 ohm-m and 30 to 40 m thick, followed by (3) saturated sand with good groundwater quality having resistivity values of 100 to 270 ohm-m and 40 to 80 m thick; (4) a clay and clayish sand formation (50-80 ohm-m resistivity values and 40-90 m thick), and (5) a sandy formation saturated with fresh groundwater (100-350 ohm-m). as seen from fig. 6, the cross section is characterized by low to moderate resistivity values at the surface when compared with the cross section (fig. 5). this decrease in resistivity values from about several hundred ohm meters on the surface along profile a-a’ to less than 650 ohm-m on the surface along profile b-b’ can be explained by more homogenous gravel and clayish sand nature of the pleistocene sediments in the southern parts than the northern parts. geological and hydrogeological discussion as seen before, the geoelectrical stratification of the study area comprises three to five units of high-, relatively high, relatively low-, and low-resistivity values from the surface down. the western part of the study area (ves sites 1-11, profile a-a’, fig. 5) can be easily interpreted directly in terms of geological and hydrogeological structures: a thin resistive coarse grained sand and gravel underlain by some tens of meters thick strata of fine to medium grain sand, gravel, clay, and clayish sand, followed by saltwater. as indicated by the cross section of profile b-b’ (fig. 6), the basic structure beneath the central part of the study area (ves sites 3736 profile b-b’, fig. 6) comprises three units of moderate high, relatively low, and low resistivity values from the surface down. the surface resistive layer represents dry sand and gravel. the second layer, marked by resistivities of some tens of ohm meters could indicate the presence of sand and sandy clay. the low resistivities values (~30 ohm-m) in the third layer may be related to the presence of clay and clayish sand layer. in addition, the eastern part of the study area (ves sites 35-31, profile b-b’, fig. 6) comprises a succession of clay and clayish sand with fresh groundwater. conclusions the use of ves surveying technique was proved useful to map the subsurface of structurally complex area where little geologic information is available due to lack of drillholes. ves surveys can also be used to locate bodies of groundwater and zones with anomalous electrical properties. the apparent resistivities measured on the study area can be explained by resistivity distributions involving four extensive electrically distinctive zones (i-iv, fig. 7). the study area can then be subdivided in four compartments (fig. 7). electrical differences between the four zones cannot be explained by water content, as different geological contexts have to be considered. as a result, two different hydrogeological behaviours can be distinguished: (i) a saline groundwater coastal aquifer in the western part, and (ii) a fresh groundwater aquifer in the eastern part. an important general result of this study is the existence of different coastal hydrogeological zones directly linked to geological subsurface structures, despite a monotonic surface cover 105 batayneh. esrj vol. 11, no. 2. december 2007 figure 4. typical electrical resistivity sounding data and best-fit three to five-layer model interpretations corresponding to the four types of field curves. (a) site 1 (type i field curves); (b) site 37 (type ii field curves); (c) site 33 (type iii field curves); and (d) site 41 (type iv field curves) figure 5. composite electrical cross section inferred from geoelectrical sounding data along a-a’ profile (see fig. 1 for profile location). resistivities are in ohm-m 106 mapping subsurface formations on the eastern red sea coast in jordan using geoelectrical techniques: geological and hydrogeological implications figure 6. composite electrical cross section inferred from geoelectrical sounding data along b-b’ profile (see fig. 1 for profile location; see fig. 4 for legend). resistivities are in ohm-m figure 7. map of the four electrical distinct zones i-iv in relation to location of the ves sites and geology. the elevation contours are in meters. palestine grid are used 107 batayneh. esrj vol. 11, no. 2. december 2007 the surface alluvial and wadi sediments. the geological structures of coastal areas within an active plate may significantly control the groundwater behaviour and knowledge of the tectonic history of the region; therefore, it is necessary to deal with its hydrogeology. the second important result is that the area is characterized by a thick clay and poorly permeable substratum that occurs in the central part. this result provides new ideas about the hydrogeology of the area. the results obtained are in a good agreement with the borehole data available. acknowledgments the author wishes to acknowledge the support received by the natural resources authority of jordan, iris syscal-r2 resistivity equipment. facilities provided by the geophysics division of the natural resources authority of jordan are acknowledged. references al khatib, f., 1987. the geology of jabal al mubarak & al yamaniyya. map sheet nos. 3048 iv & 2948 i. geological mapping division, natural resources authority, amman, jordan. ayers, j., 1989. conjunctive use of geophysical and geological data in the study of an alluvial aquifer. ground water 27: 625-632. barongo, j., and palacky, g., 1991. investigations of electrical properties of weathered layers in the yala area, western kenya, using resistivity soundings. geophysics 56: 133-138. batayneh, a., 2006. use of electrical resistivity methods for detecting subsurface fresh and saline water and delineating their interfacial configuration: a case study of the eastern dead sea coastal aquifers, jordan. hydrogeology journal 14: 1277-1283. batayneh, a., and qassas, h., 2006. changes in quality of groundwater with seasonal fluctuations: an example from ghor safi area, southern dead sea coastal aquifers, jordan. journal of environmental sciences 18: 263-269. gnanasundar d, elango l (1999) groundwater quality assessment of a coastal aquifer using geoelectrical techniques. journal of environmental hydrology 7: paper 2 keller, g., and frischknecht, f., 1966. electrical methods in geophysical prospecting, pergamon, ny. khair, k., and skokan, c., 1998. a model study of the effect of salination on groundwater resistivity. journal of environmental geophysics 2: 223-231. mukhtar, a., sulaiman, w., ibrahim, s., abdul latif, p., and hanafi, m., 2000. detection of groundwater pollution using resistivity imaging at seri petaling landfill, malaysia. journal of environmental hydrology 8: paper 3. orellana, e., and mooney, h., 1966. master tables and curves for vertical electrical sounding over layered structures. interciencia, madrid. parasnis, d., 1956. the electrical resistivity of some sulphide and oxide minerals and their ores. geophysical prospecting 4: 249-279. parasnis, d., 1966. mining geophysics. elsevier science, amsterdam. salameh, e., and bannayan, h., 1993. water resources of jordan, present status and future potentials. fredrich eberet stiftung, amman. van overmeeren, r., 1989. aquifer boundaries explored by geoelectrical measurements in the coastal plain of yemen. a case study of equivalence. geophysics 54: 38-48. zohdy, a., 1965. the auxiliary point method of electrical sounding interpretation, and its relationship to the dar zarrouk parameters. geophysics 30: 644-660. zohdy, a., and jackson, d., 1969. application of deep electrical soundings for groundwater exploration in hawaii. geophysics 34: 584-600. articulo 4.indd earth sciences research journal meteorology earth sci. res. sj. vol. 15, no. 2 (december, 2011): 109 114 detecting low frequency cycles in rainfall series from colombian coffee-growing area by using descriptive methods andrés j peña q.1, álvaro jaramillo r.2 and maría j paternina q.3 1 disc. de agroclimatología. centro nacional de investigaciones de café, cenicafe. km 4 vía antigua chinchiná-manizales. e-mail: andres.pena@cafedecolombia.com 2 disc. de agroclimatología. centro nacional de investigaciones de café, cenicafe. km 4 vía antigua chinchiná-manizales. e-mail: alvaro.jaramillo@cafedecolombia.com 3 estudiante de especialización en estadística aplicada. fundación universitaria los libertadores. bogotá. e-mail: mjpaterninaq@unal.edu.co abstract descriptive statistical methods were used for improving climatic variability scenarios regarding rainfall using time series from five representative pluviometric stations (miguel valencia, naranjal, cenicafé, la bella and paraguaicito); such scenarios are used to make decisions regarding coffee-growing. the purpose was to find signs of cyclic behaviour besides those associated with el niño southern oscillation (enso) which happens every 3.5 to 4.5 years. signals were found of decadal to interdecadal cycle (10 to 30 years), modulating known enso effects related to periodic changes in the pacific ocean and solar activity. resumen con el objeto de mejorar la generación de escenarios de variabilidad climática, que son utilizados para tomar decisiones en el cultivo del café, se exploran en forma descriptiva las series temporales de lluvia medidas en cinco estaciones pluviométricas representativas de la zona cafetera colombiana (miguel valencia, naranjal, la bella y paraguaicito) para encontrar señales de comportamiento cíclico, además del asociado con la oscilación del sur el niño (enso), que es de 3.5 a 4.5 años. se encontraron señales de ciclos de periodo decadal a interdecadal (10 a 30 años) que modulan los efectos ya conocidos del enso y que están asociados a cambios periódicos en el océano pacífico y la actividad solar. palabras claves: el niño oscilación del sur (enos), oscilación decadal del pacífico (odp), series de tiempo, semivariograma. keywords: el niño oscilación del sur (enos), oscilación decadal del pacífico (odp), series de tiempo, semivariograma. record manuscript received: 20/04/2011 accepted for publications: 05/12/2011 introduction climate cycle detection is an ancient topic; our ancestors thought that our life was governed by the seasons’ rhythms; it would thus be logical to think that it would be the same in the long-term (burroughs, 2004). one can now speak about climate cycles being associated with changes in the sun (climate engine) and other cycles related to the ocean (climatic regulator). the most studied cycles from the former group are the 11-year cycle associated with sunspots and the 22-year cycle related to changes in the sun’s magnetic field (burroughs, 2004). the most important cycle in the second group is the 3to 4-year cycle related to el niño southern oscillation (enso). mesa et al., (1997) found that hydrology and rainfall in colombia are usually affected by cycles related to changes in the tropical pacific ocean’s surface temperature than by any other cycle. enso’s local climatic effects have a defined pattern; rainy years are associated with la niña (negative phase) and dry years with el niño (positive phase), as described by guzmán & baldión (1997 and 1999). nevertheless, some very dry or rainy years do not correspond to la niña or el niño, i.e. dryness or wetness intensity are not associated with oceanic niño index (oni) intensity (figure 1). some authors have found that cycles having longer periods than enso could modulate this phenomenon’s effect on climate; several authors have found cycles having a 10-40 year rainfall series period since the earliest part of the 20th century. the most important work in colombia has shown 16-year (peña, 1982) and 35-year (poveda et al., 2002; poveda, 2004) related to special phases in solar motion (landscheidt, 2001) and pacific decadal oscillation (pdo) peaks which could be affecting enso’s effects on climate (verdon & franks, 2006). the purpose of this paper was therefore to assess rainfall time series at five locations having an annual dataset going back more than 46 years in the colombian coffee-growing area (a rain-fed crop system) to detect other cycles, apart from enso, which could be used as predictable patterns for annual rainfall forecasting. andrés j peña q., álvaro jaramillo r. and maría j paternina q.110 simple methodological approximations based on moving averages and semivariograms were used as an alternative to periodograms, which have been shown to have problems when used for determining the presence of cycles and oscillations (hernandez, 1999), and the autocorrelation function whose results would not have been appropriate in exploratory analysis because correlation level significance would not have been high and the aim was to identify signals not to quantify them. moving averages with burroughs filter (eq.1) were used as in the mathematical expression shown for (eq.2). using running means or moving averages was common practice in climatology; although this was a powerful tool for reducing short-term fluctuations in meteorological time series, it became unused because unweighted moving averages could be inappropriate for explaining periodicity (lewis, 1960). therefore, burroughs (1978) proposed moving averages with filters to improve the smoothness of meteorological series. function (eq.1) is known as a seven-year triangular filter and its great advantage lies in effectively suppressing high frequency elements. kutzbach & bryson (1974) explained the importance of non-high frequency components in a climatological time series. weighted moving averages can show cyclical time series’ behaviour. however, the semivariance function was also used here for improving results. this function is used in geostatistics to evaluate spatial correlation (structural analysis); it was used here as a temporal correlation estimator. this function was thus estimated with the experimental semivariogram calculated according to wackernagel (1995), as shown in (eq.3). figure 1. the relationship between oceanic niño index (oni) and rainfall anomalies at the cenicafé pluviometric station. there was an inverse correlation; however, some dry (wet) years did not correlate with el niño (la niña) and oni mean intensity did not correspond to rainfall intensity. figure 2. rain gauge locations and inter-annual (monthly) patterns. time series longitudes are shown beside monthly pattern profiles; i.e. paraguaicito has had a historical dataset dating back from 1963 up to 2008, while miguel valencia has a historical data set dating from 1956 up to 2008 data and methodology five rain gauges were chosen having historical annual data sets going back a long way; they formed part of the colombian federation of coffee growers’ meteorological network. the rain gauges were located in the coffee-growing area (c.g. zone), between 4°24´ n (paraguaicito station) and 5°36´ n (miguel valencia station) at different heights above sea level (in meters, masl) (figure 2). = 2 ( 4 / ) 16 2 ( / ) �� = 1 16 ( 3 +2 2 +3 1 +4 +3 +1 +2 +2 + 3 )� (1) (2) detecting low frequency cycles in rainfall series from colombian coffee-growing area by using descriptive methods 111 where z(x) is the value in x time, z(x+h) is a value separated by h time from the previous value and n is the pair’s number separated by this time (and it is not constant). semivariograms were drawn using a semivariance/ sample variance ratio on the y-axis to have the same scale at all locations. as opposed to a geostatistical approach, where the aim is to fit a model to the experimental semivariogram to be interpolated, semivariograms were used in this work to detect signs of cyclical behaviour, i.e. it was used as a descriptive tool. semivariance function could thus be inversely interpreted regarding the correlation function; when there is high semivariance there is a negative correlation, when semivariance is close to one there is no correlation, whereas there is positive correlation if there is low semivariance. thus when there is cyclical behaviour, a semivariogram having high to low semivariance oscillations is expected when distance (time, in this case) becomes increased. figure 3. moving averages for annual rainfall time series. there was a cyclic pattern; a dry period followed a wet period at the five stations analysed here. figure 4. semivariogram for time series. minimum semivariance / sample variance ratio value denotes signs of cyclic behaviour in the period shown on the y axis. (3) andrés j peña q., álvaro jaramillo r. and maría j paternina q.112 figure 5. hypothetical time series generated with the experimental semivariogram illustrated in figure 4 (left-hand column). the right-hand column gives an experimental semivariogram for each series. 11-year period signal (a), 21-year period signal (b), 33-year period signal (c) and reconstructed series representing mean annual rainfall pattern in the area (d). (a) (b) (c) (d) detecting low frequency cycles in rainfall series from colombian coffee-growing area by using descriptive methods 113 figure 6. reconstructed and running mean for a time series for the cenicafe and la bella locations. discussion and conclusions figure 3 shows the annual rainfall’s cyclic pattern recorded at five stations. consecutive wet years alternated with consecutive dry years; the wettest periods in the area were 1971-1977 and 1996-2002 and the driest periods were 1977-1983 and 1989-1995. there was an exceptional period having greater annual rainfall in the longer time series: 1951-1957. overall, 12 to 18 years between crests (rainy years) were noticed in weighted time series. in spite of the fact that there were more “la niña” months (and more intense cooling in the pacific ocean) between 1971 and 1977 than between 1996 and 2002, the latter period was rainier regarding 1971–1977. it was not related to enso effects, at least not completely because when “la niña” occurred there was greater rainfall in the area (guzman & baldión, 1999). similar results were obtained when the driest periods were analysed; there were less “el niño” months between 1977 and 1983 than between 1989 and 1995; however, the dry period was longer after 1977. figure 4 shows signs of periodical variability in annual rainfall time series. three signals (decadal, bi-decadal and interdecadal) occurred at almost all stations which other authors have associated with sunspots (burroughs, 2004; hiremath, 2006), solar magnetic cycles (almeida, 2004) and ocean phenomena such as pdo and north atlantic oscillation (nao) (poveda, 2004; tourre et al., 2001). three hypothetical time series (n=65) were calculated for each signal (11, 21 and 33 years), having values between -1 and 1 to obtain a reconstructed series by summing representing mean annual rainfall pattern in the area (figure 5). the reconstructed series representing mean annual rainfall pattern (figure 5d) was compared to the smoothed time series (running means with burroughs filter) for la bella and cenicafé locations. a relationship was found between running mean series and bit-lagged reconstructed series (figure 6). the results showed that signs of low frequency cycles could be detected in annual rainfall series by using simple statistical methods related to a description of variability and that there were relevant cycles in rainfall figure 7. semivariogram for sea surface temperature time series in the equatorial pacific ocean. andrés j peña q., álvaro jaramillo r. and maría j paternina q.114 series which could have been modulating enso effects. this occurred because sea surface temperature time series’ semivariance function (equatorial pacific ocean) was related to enso phenomenon, showing signs of cycles having oneand three-decade periods as rainfall series (figure 7), whereas a two-decade cycle would mostly be related to an 18.5 year cycle associated with the pdo’s moon-tidal cycle (currie, 1991; yasuda, 2009). references almeida, a., gusev, a., mello, m., martin, i., pugacheva, g., pankov, v., spjeldvik, w. and schuch, n. (2004). geofísica internacional. 43, no. 2, 271-279. burroughs, w. (1978). on running means and meteorological cycles, weather. 33, no. 3, 101–109. burroughs, w. (2004). weather cycles: real or imaginary?, cambridge university press, cambridge, uk, 317 p. currie, r. g. (1991), luni-solar 18.6-year signal in tree-rings from argentina and chile, pure and applied geophysics. 137, no. 3, 281– 300. guzman, o. and baldión, v. (1997). el evento cálido del pacífico en la zona cafetera colombiana, cenicafé. 48, no. 3, 141-155. guzman, o. and baldión, v. (1999). influencia del evento frío del pacífico en la zona cafetera colombiana, cenicafé. 50, no. 3, 222-237. hernandez, g. (1999). time series, periodograms and significance, journal of geophysical research. 104, no. a5, 10355-10368. hiremath, k. (2006). the influence of solar activity on the rainfall over india: cycle-to-cycle variations, journal of astrophysics and astronomy. 27, no. 2-3, 367-372. kutzbach, j. and bryson, r. (1974). variance spectrum of holocene climatic fluctuations in the north-atlantic sector, journal of atmospheric science. 30, no. 8, 1958-1963. landscheidt, t. (2001). trends in pacific decadal oscillation subjected to solar forcing, internet site of the institute for research in cycles of solar activity, http://www.john-daly/theodor/pdotrend.htm (last accessed november 2010). lewis, p. (1960). the use of moving averages in the analysis of time series, weather. 15, no. 4, 121–126. mesa, o., g. poveda and f. carvajal. (1997). introducción al clima de colombia, universidad nacional de colombia, medellín, colombia, 390 p. peña, d. (1982). algunas características de la precipitación en palmira (colombia), turrialba. 32, no. 3, 219-228. poveda, g., velez, j., mesa, o., hoyos, c., mejía, f., barco, o. and correa, p. (2002). influencia de fenómenos macroclimáticos sobre el ciclo anual de la hidrología colombiana: cuantificación lineal, no lineal y percentiles probabilísticos, meteorología colombiana. 6, 121-130. poveda, g. (2004). la hidroclimatología de colombia: una síntesis desde la escala inter-decadal hasta la escala diurna, revista academia colombiana de ciencias. 28, no. 107, 201-222. tourre, y., rajagopalan, b., kushnir, y. and barlow, m. (2001). patterns of coherent decadal and interdecadal climate signals in the pacific basin during the 20th century, geophysical research letters. 28, no. 10, 2069-2072. verdon, d. and franks, s. (2006). long-term behaviour of enso: interactions with the pdo over the past 400 years inferred from paleoclimate records. geophysical research letters, 33, l06712, doi:10.1029/2005gl025052. wackernagel. h. (1995). multivariate geostatistics. an introduction with applications, springer-verlag, berlín, germany, 389 p. yasuda, i. (2009). the 18.6-year period moon-tidal cycle in pacific decadal oscillation reconstructed from tree-rings in western north america. geophysical research letters, 33, l05605, doi:10.1029/2008gl036880. although scholars have conducted numerous researches on content-based image retrieval and obtained great achievements, they make little progress in studying remote sensing image retrieval. both theoretical and application systems are immature. since remote sensing images are characterized by large data volume, broad coverage, vague themes and rich semantics, the research results on natural images and medical images cannot be directly used in remote sensing image retrieval. even perfect content-based remote sensing image retrieval systems have many difficulties with data organization, storage and management, feature description and extraction, similarity measurement, relevance feedback, network service mode, and system structure design and implementation. this paper proposes a remote sensing image retrieval algorithm that combines co-occurrence region based bayesian network image retrieval with average high-frequency signal strength. by bayesian networks, it establishes correspondence relationships between images and semantics, thereby realizing semantic-based retrieval of remote sensing images. in the meantime, integrated region matching is introduced for iterative retrieval, which effectively improves the precision of semantic retrieval. abstract keywords: the bayesian network; co-occurrence region; remote sensing image retrieval. a co-occurrence region based bayesian network stepwise remote sensing image retrieval algorithm issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n1.66107 rui zeng1, 2, *, yingyan wang1, wanliang wang2 1school of electro-mechanical and information technology, yi wu industrial &commercial college, yiwu 322000, china 2college of computer science & technology, zhejiang university of technology, hangzhou 310014, china * email of corresponding author: jamse007@126.com earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018): 29-35 a pesar del gran número de trabajos en recuperación de imagenes basadas en contenido, poco se ha avanzado en la recuperación de imágenes por teledetección. tanto la teoría como la aplicación de los sistemas son inmaduros. ya que las imágenes por teledetección se caracterizan por un gran volumen de información, amplia cobertura, temas difusos y semántica abundante, los resultados de las investigaciones en imágenes naturales e imágenes médicas no pueden ser usados directamente en la recuperación de imágenes por teledetección. incluso en una consulta perfecta de imágenes basadas en contenido, los sistemas tienen muchas dificultades con la organización de información, almacenamiento y manejo, descripción de características y extracción, medición de similitudes, retroalimentación relevante, modo de servicio de red y diseño e implementación del sistema estructural. este artículo propone un algoritmo de recuperación de imágenes por teledetección que combina la co-ocurrencia local de una red bayesiana de recuperación de imagénes con el promedio de potencia de la señal de alta frecuencia. por las redes bayesianas, se establecen las relaciones de correspondencia entre imágenes y semántica, además de permitir la recuperación de imágenes de teledetección a través de la semántica. mientras tanto, se desarrolló el módulo de región integrada para la recuperación repetitiva, lo que mejora efectivamente la precisión de la recuperación semántica. resumen palabras clave: red bayesiana; región de coocurrencia; recuperación de imágenes por teledetección. record manuscript received: 05/07/2017 accepted for publication: 19/01/2018 how to cite item zeng, r., wang, y., & wang, w. (2018). a co-occurrence region based bayesian network stepwise remote sensing image retrieval algorithm. earth sciences research journal, 21(1), 29-35 doi: http://dx.doi.org/10.15446/esrj.v22n1.66107 región de coocurrencia basada en una red bayesiana gradual hacia un algoritmo de recuperación de imágenes por teledetección e n g in e e r in g g e o l o g y 30 rui zeng, yingyan wang, wanliang wang 1. introduction studies on natural and medial image retrieval have achieved remarkable results. meanwhile, there have been numerous research projects designed specifically for remote sensing image retrieval. however, both theoretical and application systems are immature and need to be further researched (samadzadegan et al., 2012; hejazi et al., 2017). for example, the kim system developed by dlr explores implied semantic features of images by bayesian networks. nanyang technological university conducts multi-sensor (rs)2i thoroughly research feature description and extraction, multidimensional indexing technology and system structure design (mountrakis, 2012). the study compares existing image retrieval systems and focuses on image storage, network transmission model, feature extraction and description, segmentation algorithm, reasonable segmentation, similarity measurement and relevance feedback. the results show that studies on remote sensing image retrieval have some deficiencies, which are mainly manifested as follows: (1) although studies on content-based image retrieval have achieved brilliant results, they are mostly focused on natural and medical images. rarely do scholars study remote sensing images (han, 2012; xu, 2015; radan et al., 2017). (2) currently, remote sensing image retrieval systems are mainly based on texts, strip numbers, latitudes, and longitudes, rather than contents (abdi, 2013). (3) scholars have made great use of bayesian networks to study semantic-based remote sensing image retrieval. retrieval precision of existing algorithms can be further improved. (4) the retrieval precision is inversely related to time efficiency for most image retrieval algorithms. in other words, high-precision image retrieval tends to consume lots of time (zhai, 2014; simon et al., 2017). this paper proposes a stepwise bayesian network algorithm for retrieving remote sensing images, aim to address these problems; the schema combines co-occurrence region-based bayesian network image retrieval with average high-frequency signal strength, and adopts integrated region matching for iterative retrieval, thereby efficiently improving the precision of semantic retrieval and significantly reducing the retrieval time. in the meantime, semantic-based feature vectors are introduced to enhance the precision of image retrieval. 2. material and methods in content-based image retrieval, scientific and reasonable similarity measurement models are the keys to achieving the precision of the entire retrieval system (zhong et al., 2012; bata et al., 2017). bayesian network image retrieval systems consist of three parts (boser et al., 1992; roslee et al., 2017): (1) segment images with a simple image segmentation algorithm and extract color or texture features of each sub-image to describe image contents, which is referred to as image segmentation and feature extraction. (2) employ unsupervised classification to classify the extracted features, generate codebooks and encode images following the codebooks, thereby producing encoded image library. (3) obtain the final retrieval results by selecting training samples and calculating related probabilities, which is known as semantic inference. the paper proposes two concepts: code co-occurrence matrix and semantic score function to simplify the process of semantic inference. 2.1 co-occurrence region based bayesian network algorithms for remote sensing images retrieval 2.1.1 features of ikonos images ikonos is the world’s first high-resolution commercial satellite, with an orbit altitude of 681km, a revisit period of 3 days and a nadir image swath width of 11.3km. it has one panchromatic band and four multi-spectral bands (pedergnana, 2013; rahman et al., 2017). wherein, the spectral resolution of the panchromatic band and multispectral bands is 1m and 4m respectively. fundamental parameters of wave bands are shown in table 1. table 1. wave bands of iknos imagesz2.1.2 spectral feature extraction this study adopts a 256-dimensional color histogram to describe spectral features of images. in fact, swain et al. proposed to describe color features of images with color histograms in 1991. since color histogram, which presents the number of image pixels of each color afterimage colors is quantified, has transition invariance and strong robustness to rotation and scale changes, it is extensively used (stumpf, 2014; ruiz, 2014; maleki-ghelichi and sharifi, 2017). normalization processing shall be adopted, and color features shall be described by the proportion of (1) different colors to keep the scale invariant; as shown in formula (1). in the formula, n is the total number of image pixels; l is the dimension of the color histogram; nk is the number of image pixels in the k-th color dimension. 2.1.3 image segmentation in the experiment, multi-spectral konos images are segmented into non-repetitive sub-images, and 256-dimensional color histograms of the sub-images are extracted and considered as their spectral features. however, the sub-images are in large quantities, and objects with similar characteristics are segmented into different sub-images, which is not conducive to image retrieval (amiri et al., 2017). to better describe major features of images and reduce the number of feature vectors, the k-means algorithm is introduced to cluster segmented sub-images. the clustering results are deemed as the final segmentation results. sub-image is taken as the minimum unit to judge whether image regions are spatially connected, preventing segmented regions from being spatially unconnected (tariq et al., 2017). if not, the clustering results shall be segmented so that the images are spatially combined after segmentation. 2.2 codebook and image coding 2.2.1 codebook image features are accurately extracted for quantification and classified by the k-means algorithm to realize many-to-many matching between image features and semantic meanings. assuming that all image areas in the database are expressed as , n is the total number of image segmentation regions, f(d) is the region feature extraction function and f(d) represents features of all image areas in the database. f(d) is classified into l categories by the k-means algorithm, and each cluster is a set of features. let cu be the cluster center of λu, then there is a mapping relation . wherein, cbi is referred to as codebook and the index number between cu and λu is referred to as code. for the image it, its region code rt can be expressed as ci(rt ). wherein, ci is the code function of the region. since codebooks generated by the k-means algorithm are all of the fixed length and appropriate codebook sizes are difficult to be identified, this paper adopts tree structured coding to solve the problem. tree-structured codebooks are composed of different levels of (n represents the number of codebook levels). figure 1 demonstrates a schematic drawing 31a co-occur rence region based bayesian network stepwise remote sensing image retrieval algorithm of structured tree coding. figure 1. a schematic drawing of three-level tree structured codebooks according to figure 1, top-level codes have more regional features, which flexibly and efficiently allows users to select codebooks required at different levels of . based on the features of image areas, the chosen codebooks shall be used to encode images, thereby generating image library (encoded image library). 2.2.2 image coding it is necessary to select proper codebooks from tree-structured codebooks to encode images. the coding function of image areas is defined as (n represents the number of tree-structured codebook level), and the process of image coding is described as follows: step 1: select a m-level tree-structured codebook for in it and its region features , and let the optimal codebook level l be 1. step 2: traverse each cluster λu, identify the cluster center cu that contains clusters of regional features and encode the region as u, then, . step 3: if , the program stops. otherwise, it will go to step 2. step 4: encode into a code strand (each code represents clusters of image areas at different codebook levels), encode each image area and obtain encoded images, thereby forming encoded image library. 2.3 semantic inference 2.3.1 mathematical models , and coexist in the image it , and are referred to as cooccurrence regions of . semantic inference is based on codes of and its co-occurrence regions ( and ). spectral or texture features of the triad constituted by , and are described by corresponding code triad (region code). assuming that the codebook has l codes, it may generate l3 code combinations from the perspective of the combination. in this way, spectral or texture feature symbols of the region increased from l to l3, thereby improving the precision of semantic inference. when inferring semantic meanings of images, the correlation between it and is measured by the posterior probability of cooccurrence region based bayesian networks. co-occurrence region based bayesian networks are shown in figure 2: figure 2. co-occurrence region based bayesian networks formula (2) is the formula for calculating the posterior probability of co-occurrence region based bayesian networks. (2) (5) (7) (6) (3) according to bayesian formula, formula (2) can be expressed as: where in, is the prior probability of ; the conditional probability reflects random links between codes and semantic concepts and can be obtained from user feedback or training samples. assuming that the prior probability follows a uniform distribution, it can be ignored when calculating . the calculation formula is as follows: when interpreting the image , importance of the code triad can be described with and calculated by the product of region importance function and co-occurrence region importance function . assuming that and its co-occurrence regions are equally important to interpreting , then and . where in, the function is element data in the set. if all co-occurrence regions are taken into account, the following formula is valid: formula (6) prevents the bias toward regions with more co-occurrence regions when calculating . based on formula (5) and the calculation formula of , formula (4) can be simplified as: 2.3.2 code co-occurrence matrix the codes are defined to describe the importance and generate code cooccurrence matrix to simplify the calculation of , the correlation between and is reflected by the code triad importance function . it is possible to determine the importance of the code triad to by virtue of prior probability of . the calculation formula is as follows: (8) wherein, u, v and w are codes that jointly constitute code cooccurrence matrix mk. assuming that the prior probability follows a uniform distribution, formula (7) can be simplified as: (4) 32 rui zeng, yingyan wang, wanliang wang 2.3.3 semantic score function semantic scores can be divided into region semantic scores and semantic image scores. region semantic score function is used to calculate the correlation between and : (9) (10) (12) (13) (14) (15) (17) (18) (19) (16) (11) the normalization coefficient prevents bias of semantic scores toward regions with more co-occurrence regions. according to the region semantic score function, the image semantic score function can be defined as: the image semantic score function is used to measure the correlation between it and . if is greater than the predefined thresholds, it is deemed to contain the semantic concept at least. according to formula (9) and formula (11), it should be noted that this is a special form of formula (3), which indicates that the image semantic score function is consistent with cooccurrence region based bayesian networks. 2.3.4 learning with bayesian networks the prior probability can be obtained from a group of training samples that reflect semantic concept and consist of and its cooccurrence regions to calculate the semantic score of images. the probability of the code triad can be calculated through formula (13): since and may not belong to tk, the choice function ε is defined as: the prior probability can be approximately calculated through formula (15): if users submit a new training sample and merge it with the original training sample tk, the prior probability shall be recalculated through formula (16): if there are intersections between the new training sample and the original training sample, the calculation formula of the prior probability is . 2.4 stepwise remote sensing image retrieval traditional content-based remote sensing image retrieval systems tend to calculate the distance between low-level feature vectors. due to semantic gaps, however, images with similar features and irrelevant semantics are always retrieved (ali et al., 2017). in the process of image retrieval, time efficiency is also a key factor that determines the performance of the retrieval system. users require not only high precision and recall but also demand high time efficiency. 2.4.1 integrated region matching integrated region matching (irm), which has the strong robustness to possible segmentation errors, determines the overall similarity of images by comprehensively calculating the similarity of all image regions. mathematically speaking, image similarity can be determined by calculating the distance between two point sets in high-dimension space (feature space). although there are many methods of calculating the distance, such as euclidean distance, between points in high-dimension space, they cannot be used to calculate the distance between two point sets. the biggest problem of defining the distance between two point sets in feature space is to maximize the difference between overall similarity and people’s subjective feelings of image similarity. by comprehensively considering significance and distance between image regions, integrated region matching allows one region to match multiple regions, thereby minimizing retrieval errors caused by inaccurate image segmentation (yasin et al., 2017). according to the research results of wang et al.(2016), the algorithm is described as follows: image 1 and image 2 are represented by and , and feature vectors of the segmentation region i and j are represented by and respectively. also, the distance between and is .when calculating the distance between r1 and r2, it is necessary to match all regions of the two images, calculate the distance between all regions and give corresponding weights to significance factors si,j. since significance factor represents the matching degree between and , is referred to as significance matrix. the significance matrix s of integrated region matching is determined by significance factors si,j – the area ratio between one region and the whole image. assuming that the area ratio of in image 1 and in image 2 is and separately, then under standard conditions, . a reasonable matching mechanism is required to involve all image regions and give priority to the matching of the most similar regions. under this matching mechanism, regions in image 1 will only match with similar regions in image 2. in other words, the distance between the two images is 0. by distributing significance factors, integrated region matching connects the most similar regions in two images with the shortest distance, which is referred to as the most similar highest priority (mshp). it is assumed that d(i’,j’) is the minimum distance, and to obtain significance matrix with an iterative method. when , i’ and j’, which are regions with the minimum matching distance, are successfully matched together by significance factors of i’. the significance factor of j’ is calculated through . according to mshp, the matching problem of can be solved when the following four formula requirements are met. 33a co-occur rence region based bayesian network stepwise remote sensing image retrieval algorithm (20) (21) (22) (23) (24) (25) the iteration stops when all significant factors are calculated. by co-occurrence region based bayesian networks, the algorithm selects a group of candidate images that are highly correlated with the query semantics from large-scale remote sensing image database and adopts integrated region matching to reorder candidate images and return top ranking images, thereby realizing rapid image retrieval. since co-occurrence region based bayesian network image retrieval is characterized by low time complexity, it can retrieve a group of candidate images that are highly correlated with the query semantics within a short time (shahzad et al., 2017; basarian and tahir, 2017). integrated region matching has high time complexity. in comparison with the entire database, candidate images are small in quantities, which effectively reduce image retrieval time. flowchart of the whole process is shown in figure 3. figure 3. a flowchart of stepwise region retrieval stepwise region retrieval is mathematically described as follows: step 1: select a group of candidate images that are highly correlated with the query semantics. there are two alternatives: (1) users choose candidate images by searching images by the image semantic score function trained from co-occurrence region based bayesian networks. let be the query image and be the threshold of semantic image score; then the candidate images can be expressed as: (2) users select candidate images by searching semantic information of pictures by semantic score function. assuming that has two semantic meanings, and are the two semantic score functions, and and are thresholds of the two functions, then the candidate images selected can be expressed as: under normal circumstances, the threshold shall be set to 0, to ensure the precision of image retrieval. step 2: adopt integrated region matching and query to reorder the candidate image gallery a, return top ranking images and finish the search. the more the returned images are similar to the query images in spectral and texture features, the more top-ranked the images will be. 3. results this experiment involves a total of 28 multi-spectral remote sensing images photographed by ikonos and synthesizes standard true-color images with blue, green and red wave bands. each image is segmented into 256x256 sub-images, and the maximum repeatability between two subimages is 50%. after deleting unqualified images, the database has a total of 12,000 sub-images, covering nearly 40,000km2 earth surfaces.images in the database are classified into six categories following semantic meanings: farmland (644), city (3,547), water body (3,243), vegetation (2,937), bare earth (1,033) and rock (1,730). wherein, the figures in brackets represent numbers of images related to corresponding semantic concepts, and one image may be correlated with multiple semantic concepts. this paper measures performance of the retrieval system with precision and recall. in the experiment, the author segments images into 32x32 subimages extracts color histograms of regions with spectral features and employs the k-means algorithm for image clustering. when generating encoded image library, the color feature vectors are classified into tree codes by the k-means algorithm, thereby generating codebooks (hussin et al., 2017). each image area is encoded into corresponding codes following its feature vectors, thereby making encoded image library. this experiment creates three-level codebooks: 150, 300 and 600. experiments are designed to verify the retrieval algorithm proposed, which focuses on co-occurrence region based bayesian network stepwise remote sensing image retrieval, verifies precision and time efficiency of the algorithm. 3.1 experiment design and results considering that the retrieval time of integrated region matching may be influenced by the segmentation region numbers of query images, the experiment selects images with 4 to 13 image regions, with integrated region matching and stepwise image retrieval as controlled trails. each retrieval experiment is repeated for three times, to ensure the accuracy of time, and the average of the retrieval time is deemed as the experiment result. since stepwise image retrieval experiment records the time consumed in step 1 and step 2, it is possible to obtain the total retrieval time. this experiment measures performances of retrieval systems by retrieving time and precision. the retrieval time of integrated region matching and stepwise image retrieval is shown in table 2, and the comparison results are shown in figure 4(a). since integrated region matching belongs to sorting algorithm, rather than classification algorithm, it selects the first 100 images for precision calculation. as for stepwise image retrieval, it will fetch the first 100 images if it retrieves more than 100 images; otherwise, it will be subject to the number of images returned. the final retrieval precision is demonstrated in figure 4(b). table 2. a comparison of retrieval time between integrated region matching and stepwise image retrieval 34 rui zeng, yingyan wang, wanliang wang figure 4. a comparison of between integrated region matching and stepwise image retrieval 4. discussion 4.1 time efficiency analysis according to table 2, the retrieval time of integrated region matching increases gradually with the increase in region numbers, which is mainly because the number increase adds difficulties to the computation. nevertheless, the increase in region numbers has little impact on the retrieval time of stepwise image retrieval. it fluctuates around 1.3s, 1.4s and 2.35s at 150, 300 and 600 yards respectively. with the increase of codebook numbers, however, the retrieval time increases accordingly. on average, the retrieve time rises by 0.126s from 150 yards to 300 yards and by 0.943s from 300 yards to 600 yards. due to the increase of codebook numbers, code co-occurrence matrices appear. the codebooks at 300 yards are eight times those at 150 yards, and the codebooks at 600 yards are eight times those at 300 yards, which indicates that the ratio of retrieval time difference is 1:8. through experiment, the ratio is proved to be 1:7.4841, which is consistent with the speculation. the retrieval time of step 2 is related to the number of candidate images returned in step 1. it can be seen from figure 4(a) that the retrieval time of stepwise image retrieval is significantly shorter than that of integrated region matching, which is because co-occurrence region based remote sensing image retrieval has lower time complexity. 4.2 recall analysis according to figure 4(b), the precision of integrated region matching is 0.609, which is significantly higher than that of stepwise image retrieval (0.853, 0.926 and 0.93 at 150, 300 and 600 yards respectively). the reasons are as follows: images retrieved by integrated region matching are merely similar to each other in respect to lowlevel features, without considering semantic similarity. stepwise image retrieval selects candidate images highly correlated with query images and retrieves images with similar characteristics to query images by integrated region matching, which effectively improves the precision of image retrieval. 4.3 retrieval cases this paper selects an image of city region for retrieval analysis and 600 yards for stepwise image retrieval. the first ten images retrieved are shown in table 3: arranged from left to right and from top to bottom. table 3. a comparison of retrieval results between stepwise image retrieval and integrated region matching 35a co-occur rence region based bayesian network stepwise remote sensing image retrieval algorithm 5. conclusions by considering remote sensing image features and remote sensing image retrieval, the paper proposes a co-occurrence region based bayesian network stepwise remote sensing image retrieval algorithm that takes into account both time efficiency and retrieval precision. this algorithm mainly draws reference from stepwise remote sensing image retrieval proposed by moustakidis et al.(2012) and based on csbn and integrated region matching. it consists of two parts: co-occurrence region based remote sensing image retrieval and integrated region matching. select a group of candidate images that are highly correlated with query images in semantics by co-occurrence region based remote sensing image retrieval and reorder the candidate images with integrated region matching, thereby obtaining the final retrieval results. from the perspective of time efficiency, stepwise remote sensing image retrieval consumes 1.5171s, 1.5425s and 2.4454s at 150, 300 and 600 yards respectively. regarding the precision, integrated region matching enjoys an accuracy of 0.609 while stepwise image retrieval significantly improves the precision to 0.853, 0.926 and 0.93 at 150, 300 and 600 yards respectively. acknowledgments this research was financially supported by science and technology project of zhejiang province (grant no. 2015c31171) and national education information technology research plan project(grant no. 146241806) (grant no. 136241559). references abdi, m. j., & giveki, d. (2013). automatic detection of erythematosquamous diseases using pso-svm based on association rules. engineering applications of artificial intelligence, 26(1), 603–608. ali, s. s., ijaz, n., aman, n., nasir, d. a., anjum, d. l., & randhawa, d. i. a. (2017). clinical waste management practices in district faisalabad. earth sciences pakistan, 1 (2), 01–03. amiri, m., arabhosseini, a., kianmehr, m. h., mehrjerdi, m. z., & mirsaeedghazi, h. (2017). environmental impact assessment of total alkaloid extracted from the atropa belladonna l. using lca. geology, ecology, and landscape, 1(4), 259-263. basarian, m. s., & tahir, s. h. (2017). groundwater prospecting using geoelectrical method at kg gana, kota marudu, sabah. earth science malaysia, 1(2), 7-9. bata t., samaila, n. k., maigari, a. s., abubakar, m. b., & ikyoive, s. y. (2017). common occurences of authentic pyrite crystals in cretaceous oil sands as consequence of biodegradation processes. geological behavior, 1(2), 26–30. boser, b. e., guyon, i. m., & vapnik, v. n. (1992). a training algorithm for optimal margin classifiers. in: proceedings of the fifth annual workshop on computational learning theory, pittsburgh, pp. 144–152. han, m., zhu, x., & yao, w. (2012). remote sensing image classification based on neural network ensemble algorithm. neurocomputing, 78 (1), 133–138. hejazi, s. m., lotfi, f., fashandi, h., & alirezazadeh, a. (2017). serishm: an eco-friendly and biodegradable flame retardant for fabrics. environment ecosystem science, 1(2), 05-08. hussin, h., fauzi, n., jamaluddin, t. a., & arifin, m. h. (2017). rock mass quality effected by lineament using rock mass rating (rmr) – case study from former quarry site. earth science malaysia, 1(2), 13-16. maleki-ghelichi, e., & sharifi, m. (2017). prioritize and choose the best process of anaerobic digestion to produce energy from biomass using analytic hierarchy process (ahp). geology, ecology, and landscapes, 1(4), 219-224. moustakidis, s., mallinis, g., koutsias, n., theocharis, j. b., & petridis, v. (2012). svm-based fuzzy decision trees for classification of high spatial resolution remote sensing images. ieee transactions on geoscience and remote sensing, 50(1), 149–169 . pedergnana, m., marpu, p. r., dalla-mura, m., benediktsson, j. a., & bruzzone, l. (2013). a novel technique for optimal feature selection in attribute profiles based on genetic algorithms. ieee transactions on geoscience and remote sensing, 51(6), 3514–3528. radan, a., latifi, m., moshtaghie, m., ahmadi, m. & omidi, m. (2017). determining the sensitive conservative site in kolah ghazi national park, iran, in order to management wildlife by using gis software. environment ecosystem science, 1(2), 13-15 . rahman, n. a., tarmudi, z., rossdy, m., & muhiddin, f. a. (2017). flood mitigation measures using intuitionistic fuzzy dematel method. malaysian journal geosciences, 1(2), 01-05. roslee, r., mickey, a. c., simon, n., norhisham, m. n. (2017). landslide susceptibility analysis (lsa) using weighted overlay method (wom) along the genting sempah to bentong highway, pahang. malaysian journal geosciences, 1(2), 13-19. ruiz, p., mateos, j., camps-valls, g., molina, r., & katsaggelos, a. k. (2014). bayesian active remote sensing image classification. ieee transactions on geoscience and remote sensing, 52(4), 2186–2196. samadzadegan, f., hasani, h., schenk, t. (2012). simultaneous feature selection and svm parameter determination in classification of hyperspectral imagery using ant colony optimization. canadian journal of remote sensing, 38(2), 138–156 . shahzad, a., munir, m. h., yasin, m., umar, m., rameez, s., samad, r., altaf, s., & sarfraz, y. (2017). biostratigraphy of early eocene margala hill limestone in the muzaffarabad area (kashmir basin, azad jammu and kashmir). pakistan journal of geology, 1(2), 16–20. simon, n., roslee, r., & lai, g. t. (2017). temporal landslide susceptibility assessment using landslide density technique. geological behavior, 1(2), 10–13 . stumpf, a., lachiche, n., malet, j. p., kerle, n., puissant, a. (2014). active learning in the spatial domain for remote sensing image classification. ieee transactions on geoscience and remote sensing, 52(5), 2492–2507. tariq, w., hussain, s. q., nasir, d. a., tayyab, n., gillani, s. h., & rafiq, a. (2017). experimental study on strength and durability of cement and concrete by partial replacement of fine aggegate with fly ash. earth sciences pakistan, 1(2), 07-11. wang, q., lin, j., & yuan, y. (2016). salient band selection for hyper spectral image classification via manifold ranking. ieee transactions on neural networks and learning system, 27 (6) , 1279–1289. xu, l., li, y. p., li, q. m., yang, y. w., tang, z. m., & zhang, x. f. (2015). proportional fair resource allocation based on hybrid ant colony optimization for slow adaptive ofdma system. information sciences, 293, 1–10 . yasin, m., shahzad, a., abbasi, n., ijaz, u., & khattak, z. (2017). the use of stratigraphic section in recording quagmire of information for the fluvial depositional environment – a worked example in district poonch, azad jammu and kashmir, pakistan. pakistan journal of geology, 1(2), 01-02. zhai, s., & jiang, t. (2014). a novel particle swarm optimization trained support vector machine for automatic sense-through-foliage target recognition system. knowledge-based systems, 65, 50–59. zhong, y., & zhang, l. (2012). an adaptive artificial immune network for supervised classification of multi-/hyperspectral remote sensing imagery. ieee transactions on geoscience and remote sensing, 50(3), 894–909. in this study, experimental investigations were carried out on polypropylene fiber (pf), steel mesh (sm) and pf+sm reinforced shotcrete (rs) panels to evaluate performance characteristics of energy absorption and load capacity. in addition to this, material cost evaluation of shotcrete mixtures for unit energy absorption and load capacity has been given. the panel tests, along with the european specification for sprayed concrete (efnarc), were done on 18 prismatic specimens having the same mix designs and were cured for 28 days but reinforced at varying fiber dosages. test results indicate that in terms of unit cost performance for pf, the best dosages were 2 kg/m³ and 5 kg/m³ respectively. when compared with sm, it was determined that pf is more expensive within the range of 37 53%. en este estudio se realizaron investigaciones experimentales en fibras de polipropileno, en mallas de acero y en paneles de hormigón proyectado reforzados con fibra de polipropileno y con mallas de acero para evaluar las características de desempeño en absorción de energía y en capacidad de carga. adicionalmente, se provee una evaluación del costo de materiales en mezlas de hormigón proyectado por unidad de absorción de energía y por capacidad de carga. los exámenes en paneles, con las especificaciones europeas para hormigón proyectado (efnarc, inglés), se realizaron en 18 muestras prismáticas con los mismos diseños de mezcla y se curaron por 28 días con refuerzos de diferentes dosis de fibras. los resultados de la evaluación indican que, en términos de desempeño por unidad de costo para fibras de polipropileno, las mejores dosis fueron 2 kg/m3 y 5 kg/m3, respectivamente. cuando se comparó con las mallas de acero, se determinó que las fibras de polipropileno son entre el 37 % y el 53 % más caras. earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 81 84 abstract resumen key words: shotcrete, polypropylene fiber, energy absorption, panel test. palabras clave: hormigón proyectado; fibras de polipropileno; absorción de energía; evaluación de paneles. determining optimal polypropylene fiber dosages in sprayed concrete for mining and civil engineering applications bayram kahraman dokuz eylül university, department of mining engineering, turkey bayram.kahraman@deu.edu.tr geotechnics record manuscript received: 27/11/2013 accepted for publication: 30/09/2014 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n1.40995 82 bayram kahraman82 introduction sprayed concrete is a mixture of cement, aggregate, and water projected pneumatically from a nozzle into place to produce a dense homogeneous mass. sprayed concrete normally incorporates admixtures and may also include additions or fibers or a combination of these (efnarc, 1996). s h o t c r e t e (sprayed concrete) is widely used as rock support in mines and civil engineering projects. it is applied through a process by which concrete or mortar is sprayed onto a surface to produce a compacted self-supporting and load-bearing layer. the main design principle for shotcrete as well as for other rock support elements is to help the rock to carry its inherent loads (malmgren, 2005). the use of shotcrete for the support of underground excavations was pioneered by the civil engineering industry. reviews of the development of shotcrete technology have been presented by morgan (morgan et al., 1989). rabcewicz was largely responsible for the introduction of the use of shotcrete for tunnel support in the 1930s, and for the development of the new austrian tunneling method for excavating in weak ground. in recent years, the mining industry has become a major user of shotcrete for underground support (hoek, 2007). in mining, flexible linings are preferred because large displacements of underground openings are allowed. it often results in extensive cracking of the shotcrete. therefore, the toughness of shotcrete is very important which is enabled by the use of reinforcement (morgan et al., 1989). toughness is the amount of energy that is absorbed before and after fracture. ductility and high-fracture strains are also important characteristics of fiber-reinforced shotcretes (frs) because the main reason for incorporating fibers in concrete and shotcrete is to impart ductility to an otherwise brittle material. in addition, fiber reinforcement improves the energy absorption and crack resistance of shotcrete (morgan et al., 1995). with an increasing load on the composite, fibers will tend to transfer the additional stress to the matrix through bond stresses until they fail or pull out (mehta and monteiro, 1993). in this way, they enable shotcrete to continue to carry the load after cracking, the so-called post cracking behavior (vandewalle, 1997). in this study, rs panels were made with pf, sm and pf+sm. these rs panels, moreover, examined for load capacity and energy absorption tests. this study shows that the use of pf in rs, with sufficient fiber content can greatly improve energy absorption and ultimate load capacity. pf, sm and pf+sm concrete mixture estimation costs, furthermore, were compared in terms of unit load and energy absorption cost. 2.materials and experimental work in previous studies, malmgren discussed the comparison of steel fibers and pf and its effects on shotcrete. results of malmgren’s study show us the shotcrete reinforced with steel fibers and the pf had almost the same energy absorption. in this study, to investigate the load capacity and the energy absorption of reinforced shotcrete, efnarc panel test was used. firstly energy absorption and peak load of pf at different dosages and sm is examined. then the pf content was evaluated in terms of unit cost. for this aim, 27 shotcrete mixtures are prepared and then the energy absorption of mixtures is tested. 2.1.materials in the design of test panels, turkish portland cement was used for the production of concrete mixtures (pc 42.5 r, astm type i). crushed rock having a particle size between 0-3 mm and 3-7 mm were equally used as aggregates. the sieve analysis of aggregate is given in table 1. as a high range water reducing admixture, rheobuild® t90, was used in shotcrete mixtures. the specifications of pfs are given in table 2. the steel mesh (diameter: 5.5 mm; intervals: 100 mm) was used as a reinforcement. table 1. sieve analysis of aggregates in shotcrete table 2. specification of pf (rheobuild® t90) 2.2. shotcrete mixtures the shotcrete mix proportions used in this experimental study are given in table 3. shotcrete mixes having 0.47 water/cement ratio were prepared and put into molds. 83determining optimal polypropylene fiber dosages in sprayed concrete for mining and civil engineering applications materials (kg/m3) pf-0* pf-2 pf-3 pf-4 pf-5 pf-6 pf-8 pf-4+ sm** sm cement 450 450 450 450 450 450 450 450 450 coarse aggregate 825 825 825 825 825 825 825 825 825 fine aggregate 825 825 825 825 825 825 825 825 825 water 210 210 210 210 210 210 210 210 210 admixture (wra) 4 4 4 4 4 4 4 4 4 pf 0 2 3 4 5 6 8 4 +sm sm *:pf dosages **:steel mesh table 3.base concrete mix proportions 2.3. experimental work the efnarc panel test was used to compare the toughness of shotcrete panels with different reinforcement and amount of reinforcement. the toughness was calculated as the absorbed energy corresponding to the area under the load-displacement curve between 0 and 25 mm (efnarc, 1996). the test panel 600 mm x 600 mm x 100 mm was supported on its four edges rigidly (figure 1-b). the free face of the panel is 500 mm x 500 mm because of the thickness of the support (figure 1-a). the load was applied through a contact surface of 100x100 mm at the center of the panel by 100 kn testing machine at 1.5 mm/min deformation rate. during the test load and displacement at the center are recorded up to of 25 mm deflection has occurred at the center of the panel. fig. 1. a. efnarc panel test (efnarc, 1996), b. testing machine (100 kn) used for panel test (a) (b) 2.4. shotcrete cost the required amount of shotcrete, to cover 10 cm thick, and 1 m long shotcrete of a decline of 5x5 m cross-section was taken as basis for calculating the shotcrete cost. in these calculations, rebound was assumed to be % 20. accordingly the amount of shotcrete was found to be 1.93 m3 for 1 m advancement. the cost of 1.93 m3 shotcrete was found for every mixture using the current prices given in table 4, these values were divided by load capacity and energy absorption of each mixture, and thus the unit load and energy cost performance of the mixture were calculated (table 4). table 4.shotcrete costs concrete mix materials unit unit cost aggregate (0-7 mm) $/t 9,90 cement, cem 1 42,5 $/t 56,81 rheobuild® t90 $/kg 0,57 fiber meyco® fib sp 540 $/kg 6,01 steel mesh (5x2 m) $/unit 31,03 3. results and discussion the load–deformation curves (figure 2) were obtained as an average of the results of the tests performed on nine groups having three panels each. pf-0, pf-2, pf-3, pf-4, pf-5, pf-6, pf-8, pf-4+sm, sm panels. fig. 2. load deformation curves then the energy absorption of shotcrete lining is calculated by integrating the area under the load-displacement curve. the load– deformation curves of tests are given in figure 2. the area under the load displacement curve between 0 and 25 mm calculated by the aim of cad based program and the results are given in table 5. also unit costs per peak load and unit costs per absorbed energy are calculated and given in table 5. 84 bayram kahraman table 5. results of panel tests the test results show us the dosage of pf has a moderate effect on the peak load of panel samples. on the other hand, energy absorptions of samples have considerable differences by the changes of fiber dosages. the results from panel tests are summarized in figure 3. as inferred from the table 4 and figure 3, use of both pf and sm has the maximum energy absorption but it should also be evaluated in terms of costs. however, in any case both pf and sm will be used for high-stress areas like as intersections of underground spaces. 0 0,1 0,2 0,3 0,4 0,5 0,6 0,7 0,8 0 10 20 30 40 50 60 70 pe ak l oa d (k n ) peak load (kn) energy absorption (kj) en er gy a bs or bt io n (k j) fig. 3. peak load and energy absorption of test samples when the effect of pf in a mixture is discussed, the tests show us increasing of pf dosage have a positive effect on energy absorption of shotcrete lining. another issue that must not be forgotten is the effect of mixture ratios on shotcrete. the results are discussed in terms of cost per load and cost per energy absorption. as seen from the table when the fiber dosage increased from 2 kg/m3 to 8 kg/m3 the peak load increases 19% whereas the unit cost per peak load increases 26%. on the other hand, the energy absorption increases 29% but the cost per energy absorption increases only 15%. similarly when the sm and sm + pf are discussed the cost per peak load and the cost per energy absorption decrease up to 50%. 4. conclusion the experimental investigation was performed on polypropylene fiber, steel mesh, and polypropylene fiber + steel mesh panels to evaluate performance characteristics of energy absorption and load capacity. in addition, material cost evaluation of shotcrete mixture for unit energy absorption and load capacity were discussed. the obtained results are summarized as follows; 1. the dosage of pf has not considerable effect on peak load of shotcrete linings, but the use of sm or sm+pf together has a positive effect on peak load. 2. according to the achieved results, by the increasing of pf dosage a linear rise of energy absorption of shotcrete linings is observed. especially compared with zero dosage of fiber, the existence of pf has a favorable effect on shotcrete lining. 3. if the cost of pf discussed in terms of energy absorption the study shows when the energy absorption increases 29% the cost per energy absorption increases 15%. however, if the peak load discussed the cost per peak load increases with the increase of fiber dosages, because the fiber dosages have no remarkable positive effect on peak load 4. the uses of pf and sm together have the maximum energy absorption and minimum cost per energy absorption. although the sm and sm + pf has a few disadvantages such as time-consuming installation, difficult installation, etc., especially in the unstable formations the sm and sm + pf linings should be preferred 5. it should be kept in mind that the change of energy absorption is not related to only the fiber dosages. it is also the function of mixture component ratios. references [1] european federation of producers and applications of specialist products for structures, european specification for sprayed concrete, efnarc (1996). loughborough university, pp. 8 – 10. [2] malmgren, l. (2005). interaction between shotcrete and rock experimental and numerical study, doctoral thesis, luleå university of technology, department of civil and environmental engineering, division of mining and geotechnical engineering rock mechanics and rock engineering, sweden. [3] morgan, d.r., mcaskill n., richardson b.w., zellers r.c. (1989). a comparative evaluation of plain, polypropylene fiber, steel fiber and wire mesh reinforced shotcrete, transp. res. rec. 1226, washington, d.c., 78 – 87. [4] hoek, e. (2007). practical rock engineering. chap. 12, rock-support interaction analysis for tunnels in weak rock masses, pp 1-19, www.rockscience.com. [5] morgan d.r., chen l., beaupre d. (1995). toughness of fibre reinforced shotcrete, asce shotcrete underground support vii, austria, 66– 87. [6] mehta p.k., monteiro p.j.m. (1993). concrete, microstructure, properties, and materials, 2nd ed., the mcgraw-ill companies, inc., p. 405. [7] vandewalle m. (1997). tunneling the world, bekaert s.a., belgium. palabras clave: zoeppritz; variación de la amplitud con el desplazamiento; inversión; ecuaciones de fatti y connolly; registro de pozos; valor k. how to cite item abbasi, s. s., liu, j., hameed, n., & ehsan m. (2018). a modified approach for elastic impedance inversion due to the variation in the value of k. earth sciences research journal, 22(3), 205-213. doi: https://doi.org/10.15446/esrj.v22n3.61135 elastic impedance inversion is the latest development in the field of hydrocarbon exploration and production. the present research focuses on the improvement of the use of elastic impedance inversion, easing exploration of hydrocarbons. the seismic velocities change with variation in geological constraints. constant k, which is s-wave to p-wave ratio of the nth layer and n+1 layer across the interface, it must be changed accordingly. this research focuses on testing the effects of k as a constant in the elastic impedance equation. as using the same value of k for all types of formations can give rise to severe errors in the interpretation of data. the importance of the value of k for particular amplitude variation with offset avo type (i-iv) is studied using different elastic impedance equations. the reflection coefficient (rc) curves for each avo class are generated using zoeppritz approximation and elastic impedance equations. the comparison of rc curves shows significant variations at far offsets in each avo type using the constant value of k. when k calculated is used, avo type i and type ii shows a good match at near, mid and far offsets. type iii does not change due to the changing value of k. type iv gives good agreement at near and intermediate offsets. this variation in curves, with the change in the value of k, indicates that it is a significant factor of interpretation using elastic impedance. the application of findings on well logs has given a satisfactory confirmation of the present results. this research can be helpful to resolve severe errors in the interpretation due to the constant value of k. abstract keywords: zoeppritz; amplitude variation with offset (avo) classification; inversion; fatti ei and connolly ei, well-log; k value. a modified approach for elastic impedance inversion due to the variation in the value of k enfoque modificado de la inversión de impedancia elástica con variación del valor k issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n3.61135 la inversión de impedancia elástica es el último desarollo en los campos de la producción y exploración de hidrocarburos. la presente investigación se enfoca en el mejoramiento del uso de la inversión de impedancia elástica para facilitar la exploración de hidrocarburos. las velocidades sísmicas cambian con la variación de las condiciones geológicas. la constante k, que es el resultado de la velocidad de la onda s y la onda p en una capa n y en la capa n+1 a lo largo del punto de contacto, debe ser cambiada de acuerdo con estos condicionantes. este trabajo busca medir los efectos de k como una constante en la ecuación de impedancia elástica. si se usa el mismo valor de k para todos los tipos de formaciones se puede incurrir en errores severos al interpretar la información. se analizó el valor de la constante k en la variación de la amplitud con el desplazamiento (avo, del inglés amplitude variation with offset) tipos i al iv a través de diferentes ecuaciones de impedancia elástica. las curvass del coeficiente de reflexión para cada tipo avo se generaron a traves la ecuaciones de zoeppritz y impedancia elástica. la comparación de las curvas de reflexión muestran variaciones determinantes en los desplazamientos más lejanos en cada tipo avo si se usa un valor constante de k. cuando se utiliza el valor modificado de k, los tipos i y ii de avo muestran coincidencia en los desplamientos cercanos, medianos y lejanos. la avo tipo iii no cambia cuando se utiliza el valor k constante o modificado. la avo tipo iv concuerda en los desplazamientos cercanos e intermedios. esta variación en curvas, con los cambios en el valor de k, indican que esta constante es un factor significante en la interpretación a traves de la impedancia elástica. la aplicación de los hallazgos de este trabajo en registro de pozos han confirmado satisfactoriamente los resultados. esta investigación puede ser de ayuda para resolver errores en la interpretación de información que tenga un valor de k constante. resumen record manuscript received: 21/11/2016 accepted for publication: 30/05/2017 earth sciences research journal earth sci. res. j. vol. 22, no. 3 (september, 2018): 205-213 saiq shakeel abbasi1,*, jiangping liu1, nayima hameed2, muhsan ehsan3 1institute of geophysics and geomatics china university of geosciences, wuhan. 2halliburton co, houston, tx, usa 3department of earth and environmental sciences, bahria university islamabad. *corresponding author: saiqshakeel@gmail.com g e o p h y si c s 206 saiq shakeel abbasi, jiangping liu, nayima hameed & muhsan ehsan introduction seismic velocities play a vital role in the determination of lithology, porosity and fluid content. seismic velocities vary horizontally and vertically depending upon the geological constraints, the depositional environment, overburden pressure, lithological variations, and others. elastic impedance inversion is based on the assumption of using s-wave to p-wave ratio of the nth layer, and n+1 layer across the interface which is known as k and its value is 0.25. in this research, the effects of using k as constant in elastic impedance inversion have been studied. as the seismic velocities change with the variation in both the elastic properties and other geological constraints, k also changes accordingly. conventional acoustic impedance (ai) inversion, which is the product of p-wave velocity and density, assumes that the p-wave strikes the subsurface interfaces at normal incidence (lindseth, 1979). this assumption remains valid when the offset range is small, and the reservoir is deep. however, in the presence of avo at larger offsets the reflection coefficients and polarities may change from those at normal incidence (ma, 2004). the elastic impedance approach provides better inversion results than the standard intercept-gradient approach which can be influenced by noise (cambois, 2000). elastic impedance inversion is the generalization of acoustic impedance for the variable angle of incidence. elastic impedance inversion (ei) is the latest development in the field of geophysics. the idea of elastic impedance was presented by connolly (1999). he describes the importance of elastic parameters inversion to identify the fluid and lithology. aki and richards (2002) approximation of zoeppritz (1919) equations and connolly (1999) defined equation 3 (appendix i) for elastic impedance. it is the product of p-wave, s-wave velocities, and density. elastic impedance enables geophysicists with an additional tool for the reservoir prediction. acoustic impedance alone cannot distinguish and identify the fluid type and lithology. the acoustic impedance log ignores avo effects, which enhances the chances of information loss (verwest et al., 2000). as the p-wave velocity is a function of the mineral constituents, it can only give information regarding the geometry of pores and volume of fluid (cosban et al., 2002). elastic impedance uses the necessary elastic parameters of the earth such as p-wave, and s-wave velocities and density. poisson ratio is lithologically sensitive and helps in distinguishing sandstone from mudstone, in reservoir characterization, and fluid identification. verwest et al. (2000) presented another form of elastic impedance inversion by introducing a planar boundary between two layers of acoustic media, which depends on ray parameters. whitcombe (2002) gave a normalized form of elastic impedance equation 7 (appendix i). other researchers presented different approximations of the zoeppritz equation to provide several kinds of elastic impedance equation. whitcombe et al. (2002) introduced the term extended elastic impedance (eei) to identify both fluid and lithology using the elastic impedance. duffaut et al. (2000) gave the shear-elastic impedance to estimate the ratio of p-wave velocity and s-wave velocity using joint p-p and p-s elastic impedance inversion. to deal with the dimensionality problem, savic et al. (2000) defined the elastic impedance as the function of p-wave, s-wave velocities, and ray parameters. martins (2002)set out the elastic impedance in weak anisotropic media. based on fatti approximation of the zoeppritz equation, wang et al. (2008) proposed an elastic impedance equation by using fatti approximation equation 9 (appendix i). other elastic impedance relations are presented using different media or parameterization. in this research, the importance of the value of k for each avo type has been used. as seismic velocities increase with depth, then their effect at large offsets in depth also varies. the principal purpose of this study is to check the impact of variation in the value of k when used for specific data sets. usually, k= 0.25 is used in elastic impedance inversion. however, due to lateral and horizontal variation in the velocity, k should also change accordingly, instead of using a constant value. avo classes have been evaluated by applying the elastic impedance equations for a continuous and calculated k. this study will be helpful for geoscientists as well as exploration and production companies to identify the lithological reservoirs more accurately without erroneous results. constant k connolly’s elastic impedance equation assumes using s-p wave velocity ratio k as a constant for the whole data set, which is an unrealistic approach. it is the average v vs p 2 2 of the nth layer and n+1 layer across the interface which gives different values of k at the top and bottom of a layer. hence giving different elastic impedance values. k v v v v sn pn sn pn= + + 2 2 1 2 1 2 2  (1) however, for the multilayer strata, its value is taken as an average of for each layer. mostly the rocks have v vs p 2 2 = 0.5, i.e., k = 0.25 (connolly, 1999). it is not reasonable value as it assumes that the velocity ratio of p and s waves of each medium remains the same which is not possible because seismic velocity generally increases with depth (ma, 2004). lu and a. mcmechan (2004) used a variable value of k based on empirical approximations. elastic impedance inversion with different k values gives different values of poisson’s ratio estimation, which shows that it is sensitive to the choice of k (mallick, 2001). ma (2004), and santos and tygel (2004)introduced the term reflection impedance (ri), which is derived by integrating the reflectivity along the ray path to overcome the limitations of constant v vs p ratio and normalization of the elastic impedance. in their study, the density is assumed to be related to the s-wave velocity(potter et al., 1998). most of the avo/ei equations assume the angle of incident and value of k as a constant which generates errors with the increase in angle, prominently seen from the avo/ei equations. the value of k should be used as a constraint rather taking as constant to minimize the errors that arise in the inversion (jin et al., 2015). the interpretation of the results of the normalized form of the elastic impedance depends on k and the normalization parameters (zhang et al., 2012). classification of avo anomalies the intercept/gradient method is not only a good indicator of reservoirs, but this approach can also be used to classify avo anomalies. rutherford and williams (1989) categorized various avo trends as they diversified with increasing incident angle for the top of gas-saturated reservoirs. based on field studies, they developed an avo anomaly classification scheme for gas-saturated reservoirs. the rutherford and williams (1989) classification is as follows: class i. gas-sand reservoir has larger impedance than the impedance of the encasing shale. typically, normal-incident values (ni) are higher than +0.03. class ii. gas-sand reservoir has nearly the same impedance as the encasing shale. ni typically ranges between ±0.03. class iii. gas-sand reservoir has a lower impedance than the encasing shale. ni is usually less than -0.03. class iv. gas-sands for which the reflection coefficients decreases with increasing offset. rutherford and williams (1989) also indicated that the slope of the reflection coefficient curve is usually negative for all classes and magnitude of the gradient decreases from class i to class iii anomalies. j.p castagna et al. (1998) added class iv to the rutherford and williams classification of avo. they concluded that certain class iii gas-saturated anomalies have slowly decreasing amplitudes with offset and may even reverse polarity 207a modified approach for elastic impedance inversion due to the variation in the value of k within the common depth point cdp gather. it is a reservoir with a negative avo intercept and positive avo gradient. their classification demonstrates that reflection coefficients do not always have negative slopes for gas sands. however, they cautioned that the avo gradient for class iv brine-saturated sand might be almost identical to the avo gradient from class iv gassaturated sand. hence the gas may be difficult to detect by partial offset stacks. methodology in this paper, two-layer models based on four avo classes which included avo type i, type ii, type iii, and type iv were used. avo type is a continuously distributed property of the seismic data (young & lopiccolo, 2003). the input parameters of all four models are given in table 1. from each derived elastic impedance equation, different elastic parameters by inversion can be obtained. connolly’s elastic impedance equation and fatti based elastic impedance equations were used for checking the behavior of value of k and determining which value of k should be utilized. reflection coefficient (rc) curves based on the zoeppritz equation and elastic impedance equations are used for comparison of the results. matlab software tool is utilized to generate the plots between the reflection coefficient and offset angle. the reflection coefficient is plotted on y-axis while offset angle of 0°-40° is plotted on the x-axis. sayers and rickett (1997) provided the parameters for the avo’s type i to type iii, while john p. castagna and swan (1997) provided the type iv. the value of k is calculated for each avo class using equation 1. this value varies with the change in the p-wave and s-wave velocities as shown in table 1. these estimated values of k are used during generation of reflection coefficient curves for elastic impedance. two steps are involved in seismic inversion. first is the estimation of the subsurface reflectivity as a function of the angle of incidence for each point. second is the inversion of reflectivity to estimate the corresponding earth parameters (grossman, 2003). the elastic impedance allowed to calibrate far offset angle stack like acoustic impedance logs are used to calibrate zero-offset data. there are various approaches for quantitative estimation of reservoir properties from seismic inversion. in this research, the generalized matrix inversion was used. well-x is used in the application of the findings from which the synthetic angle traces at 10º, 15º, 25º were obtained using fatti’s three-term elastic impedance formula. after generation of elastic impedance values at different angles, inversion procedure was applied; p-impedance, s-impedance, and density of the following equation were inverted. (equation 2 in appendix 1 will give rise to the equation.). a a a a b b b b c c c c t n t n t θ θ θ θ θ θ θ θ θ θ θ 1 2 1 2 1 2 ( ) ( ) ( ) ( ) ( ) ( ) • • ( ) ( ) ( ) ( ) ( ) θθn p t p s t s in i i in i i ( )                                  ( ) ( ) 0 0                                      = ( ) ( ) in in ie t i tρ ρ ,θ 0 1 pp p t p n in ie t i in ie t i in ie t 0 0 0 2       ( )      • • ( )      ( ,θ ,θ ,θ ))                                           ip0 (2) the 5% of random noise with gauss distribution is added to the synthetic angle traces to test the elastic impedance inversion method. calculated log of k is generated for each value. results and discussion type 1 avo here, it is observed the elastic impedance equations of connolly and fatti with zoeppritz approximation reflection coefficient curves with constant k = 0.25 and calculated k = 0.333. the behavior of both curves is shown in figure 1 along with near and far angles gather along type 1 avo. for avo type i when k = 0.25, the plot between reflection coefficients derived from zoeppritz approximation and both elastic impedance equations from connolly and fatti start with a similar trend. the curves start to deviate from each other from 10° and continue to deviate as they go towards far angles. by the time the curve reaches 30° both the curves have clear space among them. it is evident from the figures that at far offsets the separation of curves is prominent. the reflection coefficient is generated using calculated k =0.3393 using equation 6, and the plot between both the reflection coefficient starts with a similar trend and matches correctly. both the curves begin to deviate from each other from 30°, and the deviation is slight. avo type ii avo type ii shows in figure 2 the reflection curves using connolly elastic impedance equation for the constant and calculated value of k. it starts from near to zero reflection coefficient and the trend is more negative table 1. parameters used for each avo type for elastic impedance inversion, modified from sayers and rickett (1997)and john p. castagna and swan (1997) avo type lithology p-wave velocity km/s s-wave velocity km/s density gm/cc k v v v v sn pn sn pn= + + 2 2 1 2 1 2 2  i shale 3.30 1.70 2.35 0.33932 sand 4.20 2.70 2.49 ii shale 2.96 1.38 2.43 0.32395 sand 3.49 2.29 2.14 iii shale 2.73 1.24 2.35 0.28921 sand 2.02 1.23 2.13 iv shale 3.24 1.62 2.34 0.3432 sand 1.65 1.09 2.07 208 saiq shakeel abbasi, jiangping liu, nayima hameed & muhsan ehsan with the increase in the angle. both curves begin to deviate from each other from 10° and continue to more negative as they go towards the far angles when the constant value of k is used. connolly reflection coefficient curves match mostly but show slight separation as illustrated in figure 2. when calculated k = 0.32395 is used, it was noticed that the elastic impedance curves become more negative than the zoeppritz curve. both elastic impedance curves closely match the zoeppritz reflection coefficient curve. avo type iii the elastic impedance reflection coefficient curve for avo type iii gives a peculiar behavior for both constant and calculated value of k. the deviation of reflection coefficient curves of zoeppritz and the elastic impedance is similar, and it increases with the angle. as the estimated value of k is 0.28921, very close to k constant 0.25, connolly and fatti elastic impedance equations are only valid up to 20° for both constant k and calculated k. figure 1. reflection coefficient curves avo type i using connolly and fatti elastic impedance equations with constant and calculated value of k. figure 2. reflection coefficient curves avo type ii using connolly and fatti elastic impedance equations with constant and calculated value of k. 209a modified approach for elastic impedance inversion due to the variation in the value of k avo type iv type iv is gas sands for which reflection coefficients decrease with increasing offset figure 4. they are noticeable when the s-wave velocity in the gas sand is lower than in the overlying shale(john p. castagna & swan, 1997). the reflection coefficient curves for connolly and fatti based elastic impedance equations tends to deviate from the zoeppritz curve from 10 while using k = 0.25. while when the calculated k 0.3432 is used, the deviation starts from ° 27 for both equations. from the results, it can be seen that the value of k is a significant factor in elastic impedance inversion and its variation also affects the results. avo type i and type ii give excellent agreement with the zoeppritz when calculated k is used. avo type iii, on the other hand, does not change with the change in the value of k. with computed k type iv agrees with the zoeppritz equation until the angle of 27°; this happens when the value of k becomes significant upon the use on a considerable scale. so, for a constant value of k, the results are good for the near-offset data, but an figure 3. reflection coefficient curves avo type iii using connolly and fatti elastic impedance equations with constant and calculated value of k. figure 4. reflection coefficient curves avo type iv using connolly and fatti elastic impedance equations with constant and calculated value of k. 210 saiq shakeel abbasi, jiangping liu, nayima hameed & muhsan ehsan accurate shear wave and density information from using near-offset cannot be gotten. thus, it can be deduced that if the calculated value of k is used, a good agreement at far offset will give good control over a shear wave and density information. there exists a contradiction which can be resolved using calculated value of k. avo type i and type ii will give better results with calculated k even at far offsets. while avo iii is more accurate at near angles for constant and calculated k. type iv has a range from close offset for constant and near to mid offset for an estimated value of k and these are more accurate at near angles table 2. application of the well data the present study has been tested using 70-meter part of well log x in the study area. figure 5a shows the comparison of the well-logs with p-impedance, and s-impedance, p-wave velocity, s-wave velocity, and density. the inverted p-impedance, s-impedance, p-wave velocity, s-wave velocity, and density is shown as ip1, is1, vp1, vs1, and p1. results are generated using fatti’s three-term elastic impedance inversion for the angle gathers with 5% noise. the error in the real and inverted value is shown in figure 5b. it is evident that when k constant is used, the inverted logs deviate more from the normal well-logs. these errors arise because the constant k, as k = 0.25, or average k of well-log data has been taken; it means each layer has the same value of vs2/vp2. when some segments deviate from this assumption, significant errors are introduced into inversion. the inverted and real curves using the calculated value of k are shown in figure 6a. when calculated k is used, it is noticed that the error in s-impedance and s-wave velocity is reduced considerably, while in p-impedance and p-wave velocity error is also reduced to some extent as seen in figure 6b. density in both cases is very unstable while inverting. it gives large errors even for the slight presence of noise. these results agree with the findings that k is a significant factor in the interpretation of data in elastic impedance inversion, and it should not be considered as a constant value. the calculation of elastic impedance log is a significant step towards the elastic impedance inversion of seismic data. if the elastic impedance logs are not generated with proper care, it can deprive us of some valuable information in the interpretation of data. conclusions elastic impedance equations coupled with the zoeppritz equation can successfully be utilized to identify lithological reservoirs as well as avo table 2. maximum angles for agreement of zoeppritz and elastic impedance reflection coefficient curves avo type maximum matching curve constant k maximum matching curve calculated k i 10° 40° ii 20° 40° iii 20° 20° iv 10° 27° figure 5. (a) real and inverted logs p-impedance s-impedance p-wave, s-wave and density logs using a constant value of k. (b) error in the real an inverted logs. 211a modified approach for elastic impedance inversion due to the variation in the value of k cosban, t., helgesen, j., & cook, d. (2002). beyond avo: examples of elastic impedance inversion. paper presented at the offshore technology conference, houston, texas. duffaut, k., alsos, t., rognø, h., al-najjar, n. f., & landrø, m. (2000). shearwave elastic impedance. the leading edge, 19(11), 1222-1229. jin, z., huaishan, l., siyou, t., xing, l., xiangpeng, c., & chaoguang, s. (2015). estimation of elastic parameters using two-term fatti elastic impedance inversion. journal of earth sciences, 26(4), 556-566. lindseth, r. o. (1979). synthetic sonic logs-a process for stratigraphic interpretation. geophysics, 3-26. lu, s., & a. mcmechan, g. (2004). elastic impedance inversion of multichannel seismic data from unconsolidated sediments containing gas hydrate and free gas. geophysics, 69, 164-179. ma, j. (2004). the extending of elastic impedance and generalized elastic impedance. paper presented at the expanded abstracts, cps/seg geophys. conf, beijing, china. mallick, s. (2001). avo and elastic impedance. the leading edge, 20, 10941104. martins, j. l. (2002). an approach for elastic impedance in weakly anisotropic media seg technical program expanded abstracts 2002 (pp. 185188): society of exploration geophysicists. potter, c. c., dey, a. k., & stewart, r. r. (1998). density prediction using p-and s-wave sonic velocities: geotriad. paper presented at the cspg, cseg, cwls joint convention: http://www. crewes. org/foroursponsors/ conferenceabstracts/1998/cseg-1998/den_pred. pdf. rutherford, s. r., & williams, r. h. (1989). amplitude-versus-offset variations. geophysics, 54, 680-688. santos, l. t., & tygel, m. (2004). impedance-type approximation of the p-p elastic reflection coefficient: modeling and avo inversion. geophysics, 69, 592-598. classes. the value of elastic impedance equations constant k and its impact on avo classes and computed reflection coefficient curves were discussed and analytical studied. sometimes, a fixed value of constant k generates an error in lithological reservoirs characterizations. the constraint k value was successfully applied to study avo classes. results indicate that it improves the matching with zoeppritz curve to near, mid and far angle for avo type i and ii, and near and mid-offset for avo type iv. avo type iii does not show significant change with the change in the value of k. the decrease in the percentage of error in the inversion of the well's logging segment using calculated value of k also agrees with the findings. thus, it can be induced the findings that keeping k constants would give errors in the results, but it is necessary to have proper control over the value of k to make the results accurate acknowledgments authors would like to thanks the china university of geosciences for giving adequate facilities for this research. they would also like to extend the gratitude to mr. ahsan shafi of the china university of geosciences for his moral support and help in improvement of the manuscript. references aki, k., & richards, p. g. (2002). quantitative seismology (2nd ed.): university science books. cambois, g. (2000). avo inversion and elastic impedance. seg technical program expanded abstracts 2000, 142-145. castagna, j. p., & swan, h. w. (1997). principles of avo crossplotting. the leading edge, 16(4), 337-344. doi:10.1190/1.1437626 castagna, j. p., swan, h. w., & foster, d. j. (1998). framework for avo gradient and intercept interpretation. geophysics, 63, 948-956. connolly, p. (1999). elastic impedance. the leading edge, 18, 438-452. figure 6. (a) real and inverted logs p-impedance, s-impedance, p-wave, s-wave and density logs using a calculated value of k. (b) error in the real and inverted logs. 212 saiq shakeel abbasi, jiangping liu, nayima hameed & muhsan ehsan savic, m., verwest, b., masters, r., sena, a., & gingrich, d. (2000). elastic impedance inversion in practice seg technical program expanded abstracts 2000 (pp. 689-692): society of exploration geophysicists. sayers , c. m., & rickett, j. e. (1997). azimuthal variation in avo response for fractured gas sands. geophysical prospecting, 45, 165-182. verwest, b., masters, r., & sena, a. (2000). elastic impedance inversion. paper presented at the 70th annual seg 2000. wang, b.-l., yin, x.-y., zhang, f.-c., & li, a.-s. (2008). elastic impedance equation based on fatti approximation and inversion. progress in geophysics, 23(1), 192-197. whitcombe, d. n. (2002). elastic impedance normalization. geophysics, 67(1), 60-62. whitcombe, d. n., connolly, p. a., reagan, r. l., & redshaw, t. c. (2002). extended elastic impedance for fluid and lithology prediction. geophysics, 67(1), 63-67. young, r. a., & lopiccolo, r. d. (2003). a comprehensive avo classification. the leading edge, 22(10), 1030-1037. zhang, f., wang , y., & li, x. (2012). viabilities of seismic ray impedance and elastic impedance for hydrocarbon-sand discrimination. geophysics, 58, 1350-1360. zoeppritz, k. (1919). erdbebenwellen viiib, on the reflection and propagation of seismic waves. göttinger nachrichten, i, 66-84. 213a modified approach for elastic impedance inversion due to the variation in the value of k appendix i connolly elastic impedance equation elastic impedance is a generalization of acoustic impedance for the variable angle of incidence. connolly obtained this expression given in equation 1 by the integration and exponentiation of three-terms of aki and richards (2002) approximation of zoeppritz equation, and it is more sensitive than acoustic impedance for narrow distributions. with p-wave velocity (vp), s-wave velocity (vs), and density (ρ), connolly gave the following formula. ei v vp tan s ksin ksinθ ρθ − θ − θ( ) = +1 1 4 2 2 2 (3) where a tan= +1 2 b ksin= − θ8 2 c ksin= 1 2−4 θ k in the equation is constant; it is defined as the average v vs p 2 2 . for a two-layered model, authors get the following equations. ei v vp tan s ksin ksin 1 1 1 1 4 1 2 1 2 2 θ ρθ − θ − θ( ) = + (4) ei v vp tan s ksin ksin 2 1 2 1 4 2 2 2 2 2 θ ρθ − θ − θ( ) = + (5) the elastic impedance for a given angle of incidence θ, the reflection coefficient r as a function of  is given as, θ − + ( ) =rc ei ei ei ei 2 1 2 1 (6) ei and ei1 2 are the elastic impedances of the first and second layer, respectively. the normalized form of elastic impedance elastic impedance and acoustic impedance cannot be displayed together because the elastic impedance values change dramatically with the change in incident angle. to overcome this problem, whitcombe (2002) introduced constants v vp s0 0, and  0 which allow elastic impedance values to remain constant for a layer and does not change with the angle  and modified the elastic impedance function as following. ei v v v v p p a s p b c θ ρ ρ ( ) =                   0 0 0 (7) furthermore, whitcombe introduced a scale factor vp0 0  by which authors could predict the correct value of acoustic impedance ai  from ei () at =0. ei v v v v vp p p a s p b c θ ρ ρ ρ ( ) =                  0 0 0 0 0 (8) equation 8 has enabled more control over the elastic impedance values as they will not vary rapidly with the change in the angle . it also sets the elastic impedance function to normalized due to the addition of the constants for a layer. fatti based elastic impedance equation based on fatti approximation of zoeppritz, wang et al. (2008) proposed the elastic impedance equation as shown below ei i i i i ip p p a s p b c θ ρ ρ ( ) =                             0 0 0 0 (9) where a tan= +1 2 b ksin= − θ8 2 c ksin tan= 4 2 2θ− θ i ip s0 0 0, ,  are constants for normalization and can be taken as the p-wave impedance, s-wave impedance, and density of the upper layer, respectively. weathering involves important processes that alter the original structure, texture and chemical components of rocks and minerals. bulk changes produced by weathering in granitic rocks have been studied by several methods, including chemical weathering indices. they are based on the assumption that some ions are more easily leached from minerals in relation to others. such methods have also been briefly tested on building stone but not in urban environments, where fast stone weathering rates are typically observed, mainly due to interaction of minerals with several pollutants and where other specific processes, such as salt weathering, can occur. the aim of this work is to discuss the use of weathering indices in the study of weathering rates on granitic stones applied in four historical buildings of an urban area. results suggest that some factors can cause scatter of results in the relation indices vs. exposition age in the built environment, namely previous weathering degree in rock massifs and the quarry, different orientation of façades and different exposure to urban pollutants. indices that consider some highly mobile cations that are present in other building materials (namely ca that is related to leaching from mortar joints) should be avoided due to the uptake of such elements by the stone pores or used to assess this some types of weathering. moreover, the use of some indices shows more reliable results. la meteorización implica importantes procesos que alteran la estructura, textura y componentes químicos originales de las rocas y minerales. los cambios producidos por la meteorización en rocas graníticas han sido estudiados por varios métodos, incluyendo los índices de meteorización química. éstos se basan en la suposición de que algunos iones son lavados de los minerales más fácilmente que otros. tales métodos también se han probado, aunque en pocos casos, en rocas de edificios, pero no de ambientes urbanos, en donde se han observado típicamente elevadas tasas de meteorización, principalmente por la interacción de los minerales con varios compuestos contaminantes y en dónde otros procesos específicos, tales como la erosión por sales, pueden producirse. el objetivo de este trabajo es discutir el uso de índices de meteorización en el estudio de las tasas de meteorización de rocas usadas en cuatro edificios históricos de un área urbana. los resultados sugieren que algunos factores pueden causar la dispersión de los resultados obtenidos a través de los índices en relación al tiempo de exposición en el ambiente construido, principalmente por el grado de meteorización previo de la roca en los macizos rocosos y en la cantera, la orientación de las fachadas y la diferente exposición a compuestos contaminantes. los índices que consideran algunos cationes altamente móviles que están presentes en otros materiales de construcción (principalmente el ca que se relaciona con la meteorización de morteros en las juntas) deben ser evitados debido a que penetran en los poros de las rocas de edificios o deben ser usados para evaluar algunos tipos de meteorización. por otra parte, el uso de algunos índices muestra resultados más fiables. earth sciences research journal earth sci. res. j. vol. 20, no. 3 ( september, 2016 ) : f1f13 abstract resumen keywords: weathering indices, weathering rates, urban environment, granitic rocks, historical buildings palabras clave: índices de meteorización, tasas de meteorización, ambientes urbanos, rocas graníticas, edificios históricos. assessing the weathering of granitic stones on historical urban buildings by geochemical indices evaluación de la meteorización de rocas graníticas en edificios históricos urbanos por medio de indices geoquímicos record manuscript received: 09/03/2015 accepted for publication: 27/05/2016 how to cite item sanjurjo-sánchez, j., alves, c., vidalromaní, j.r. (2016). assessing the weathering of granitic stones on historical urban buildings by geochemical indices. earth sciences research journal, 20(3), f1-f13 doi:http://dx.doi.org/10.15446/esrj.v20n2.49560 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.49560 jorge sanjurjo sánchez1, carlos alves2 and juan ramón vidal romaní1 1instituto universitario de xeoloxía “isidro parga pondal”, campus de elviña, universidade da coruña, 15071 a coruña, spain. e-mail:jsanjurjo@udc.es 2lab2pt-landscape, heritage and territory laboratory school of sciences, university of minho, braga, portugal. g e o c h e m is t r y jorge sanjurjo sánchez, carlos alves and juan ramón vidal romaníf2 1. introduction stone surfaces suffer alteration processes when exposed to surface agents (water, atmospheric chemical components, etc) resulting in the formation of coatings and/or weathering. such alteration occurs at variable rates on natural outcrops but faster on urban environments (amoroso and fassina, 1983; sanjurjo-sánchez et al., 2012; sanjurjo-sánchez and alves, 2012a. 2012b), being necessary to take conservation measures (alves and sanjurjosánchez, 2015). weathering produces modifications in the structure, texture, mineral and chemical components of rocks. it results in increasing stone porosity (and weakening) and the disintegration of the rock and minerals. these changes have been extensively observed by petrological, physical and chemical studies (meunier et al., 2007; moses et al., 2014). chemical weathering leads to the decomposition of the original minerals of fresh (unweathered) rocks. some authors (topal, 2002) have postulated that weathering can occur on the earth surface in geologically short time periods (even decades or centuries). depending on the rock composition, environmental factors (temperature, ph, eh, ionic potential) and the previous weathering degree, some ions can be leached from minerals preferably in relation to others. the leaching processes are linked to other effects such as mineral transformations, porosity increase and mechanical strength decrease (meunier et al., 2007; chiu and ng, 2014; moses et al., 2014). chemical variations have been considered by many researchers in the definition of chemical weathering indices to assess quantitatively the degree of weathering (reiche, 1943; ruxton, 1968; irfan, 1996; ng et al., 2001; duzgoren-aydin et al., 2002; otha and arai, 2007). according to gerrard (1998), miller proposed in 1961 the following mobility sequence for ions of igneous rocks: ca>mg>na>ba>k>si>-fe=mn>ti>al. however, other different sequences have been proposed for granites as that of gupta and rao (2001): na>ca>k>mn>mg>si>fe>al>ti. several weathering indices have been successfully applied to assess the weathering degree on igneous rocks (ng et al., 2001). some indices have also been tested on building stones (topal and sözmen, 2003; lee and yi, 2007; lee et al., 2009), where surface weathering occurs and can be assessed for periods of decades or centuries (meunier et al., 2007; smith, 2003; moses et al., 2014). however, to our knowledge, such indices have not been tested on stones applied in buildings located in the urban environment, despite the fast decay observed on them mainly due to interaction of minerals with different pollutants (amoroso and fassina, 1983; sanjurjo-sánchez and alves, 2012a). fast decay in buildings is observed in urban areas as well as high weathering rates that could result in chemical and physical erosive processes. in addition, the built environment entails specific weathering processes namely salt weathering. contributing sources to salt contamination and weathering are variable leading to different geochemical signatures (sanjurjo-sánchez and alves, 2012a). the possible sources can be related to surrounding natural sources such as sea spray (whose impact is enhanced in some portions of the built environment according to erosion) and human-related pollution either from the ground (capillary rising), atmospheric (air pollution) and other building materials such as joint mortars (sanjurjo-sánchez et al., 2010). these sources can contribute diverse substances such as carbonates, sulphates, chlorides and nitrates or alkaline and earth alkaline elements (sanjurjo-sánchez and alves, 2012a). previous studies on buildings regarding chemical weathering indices have been focused on the surface and subsurface weathering (comparing weathering vs. depth), but not in the weathering rates (comparing weathering vs. time). to assess such rates, it is important to check the suitability of different weathering indices to assess the relation age/weathering for these stones and buildings. for this purpose, the behaviour of major and minor components of rock minerals was previously studied on building granite in an urban area (sanjurjo-sánchez et al., 2011a). depending on the considered elements, different weathering indices provide variable results in the leaching of some mobile elements. the aim of this paper is to compare the suitability of weathering indices to study the weathering rates on granite surfaces in a polluted environment and to assess them (relation age/weathering) for granitic stones taken from different parts of four historical buildings (built in different historical time periods) of an urban area. such comparison must be taken carefully as the weathering process could be variable in different time periods. in this sense, the exposition of the rock ashlars to strong air pollution conditions has been limited, above all, to the last century or half-century. 2. determination of weathering indices several chemical weathering indices have been proposed for quantification of weathering (table 1) and tested to assess weathering depth profiles in natural outcrops or weathering rates after exposure to artificial weathering tests. in general, a good result has been observed for most indices to assess weathering (irfan, 1996; kim and park, 2003), but they usually provide scattered data depending on the type of weathering (chemical or not), heterogeneity of the studied rocks and the rate of weathering. the indices that have been proposed can be classified as either absolute or relative. relative indices allow determining the weathering degree of different samples from the same rock type by comparing the proportion of some mobile and immobile oxides. absolute indices are based on the assumption that the fresh rock composition is known. they assume that some elements (i.e. al, ti) remain constant during the weathering process. thus, the degree of weathering can be measured by the loss of mobile oxides from the analysis of a reference non-weathered rock sample (kim and park, 2003). figure 1. pictures of the studied granitic rocks: (a) san pedro leucogranite quarry front (left fresh rock. right weathered rock). (b) w façade of colegiata de santa maría built with san pedro leucogranite. (c) detail of a wall of the s façade of colegiata de santa maría built with san pedro leucogranite (bottom) and bregua granodiorite (top). (d) bregua granodiorite ashlars of casas de paredes (note erosive features on the window lintel). assessing the weathering of granitic stones on historical urban buildings by geochemical indices f3 multiple relative indices have been proposed and tested from the first insight of reiche (1943) who proposed several weathering indices to assess weathering, taking into account a decrease in mobile cations. relative indices are based on the decrease of some mobile cations (ca, na, k) compared with immobile ones (al, fe, si), expressed as ratios (parker, 1970; irfan, 1996; gupta and rao, 2001; ng et al., 2001; arel and tugrul, 2001; kim and park, 2003). in other cases, the indices compare the loss of major cations (harnois and moore, 1988; jayaverdena and izawa, 1994; sing et al., 1998). alternative indices consider the loss on ignition (loi) or the water content, reflecting the increase of weathering caused by hydration of clays, and they are considered as sensitive to weathering under humid conditions (ng et al., 2001; kim and park, 2003). other authors have proposed absolute indices such as the lixiviation index (β), insensitive to mild chemical weathering processes, but useful to compare the evolution of mobile cations to al (ng et al., 2001). irfan (1996) identified the different behaviour of mobile and immobile elements during weathering and devised an index that is considered a good indicator of weathering, known as the mobiles index (imob) (guan et al., 2001). the use of weathering index on building stones has been scarce. esaki and jiang (2000) used the loi index in the study of the weathered condition of tuffs (vs. depth) applied in historical bridges, inferring that chemical weathering attained around 10 cm of depth. topal and sözmen (2003), in a study of tuffs in the midas monument (carved in tuffs in a little village of central anatolia, turkey) compared diverse chemical indices and refer that loi and wpi indices were good indicators for weathering depth quantification. lee and yi (2007) studied the weathering degree of diverse igneous rocks (tuffs, granites, andesitic and basaltic rocks) used in a stone pagoda located in a rural area of south korea (near a medium-size city), concluding that the chemical index values were high for all the rock types. they also observed that weathering indices showed very active potential for studying chemical weathering in all the stone materials presented (basaltic and andesitic rocks being the most sensitive). lee et al. (2009), in a study of granite stones in an historical building (located 10 km far from the city of gyeongju, south korea), stated that the weathered stones presented slightly higher weathering index than the fresh granite. the scarcity of studies performed on historical buildings is due to the difficulties to get and analyse samples from materials considered of high cultural value. in the last years, non-destructive techniques such as portable xrf have allowed obtaining chemical analyses on the surface of stones table 1. summary of chemical weathering indices used in the study. of historical buildings. however, the non-destructive techniques do not allow obtaining information in depth or they can be difficult to apply in heterogeneous rocks. to consider reference values (unweathered samples) to compare the indices, we have used quarry samples as an alternative to surface-depth comparison. thus, the study of quarry samples provides essential information to assess the weathering degree of building stones. despite some weathering indices were tested on building rock surfaces some aspects must be considered in urban environments that not occur either on natural outcrops or buildings in non-urban environments. these are pollution sources and façade orientations. several pollution sources lead to different geochemical signatures (sanjurjo-sánchez and alves, 2012a) being the sources correlated to natural sources, such as sea spray or anthropic sources such as air pollution due to human activities (e.g., traffic exhaust, industrial emissions, heating system emissions). moreover, water inputs in building materials can come either from the ground (capillary rising), atmospheric water (air pollution) or other building materials, being the pollutants present in the water different, but dissolving and transporting carbonates, sulphates, chlorides or nitrates to the building rocks (sanjurjo-sánchez and alves 2012a, 2012b). these effects must be considered in the interpretation of the results of the use of weathering indices. moreover, some aspects must be taken into account referring rock weathering in buildings. the direct or not direct exposure to sunlight of could result in more or less marked thermal and wetting/drying cycles (wendler et al., 1990). wetter f4 conditions can cause stronger dissolution and leaching of some mobile elements in the rock while more marked thermal cycles can cause more physical weathering. besides, the exposure of façades to pollutants can be variable depending on their distance and orientation to air pollution sources (e.g. sea spray, atmospheric pollution) as it has been observed (sanjurjo-sánchez et al., 2009, 2011a, 2012). 3. study area the studied building stones are located in a coruña, a town on the nw coast of spain with approximately 250 000 inhabitants and about 600 000 in the metropolitan area. the downtown is located on a peninsula with the seacoast to the north, east and west and dominant winds came from the ssw and n. thus, the climate of the city is mild, sub-humid mediterranean with atlantic trend and strong seaside influence, average annual temperatures of 13.9°c and average annual rainfall of 1000 mm. air pollution in the town is not very high but important in some urban areas due to industrial activities and car traffic. previous studies have referred a marked occurrence of gypsum-rich black crusts (sanjurjo-sánchez et al., 2009, 2011b), a feature generally considered as indicator of the effect of air pollution. heritage buildings of the city have been constructed with two types of granitic rocks from local quarries: san pedro leucogranite and bregua granodiorite. the san pedro leucogranite is the most used rock in the historic buildings of the town. it is a white-yellowish fine-grained granite with quartz (forming heterogranular aggregates), feldspars (with plagioclase > microcline) and muscovite, while biotite is absent (igme, 1975). feldspars and muscovite grains are slightly deformed and oriented. in the quarry, near the town centre (3 km w) and abandoned in the middle 20th century, two granitic varieties can be observed in relation to natural rock weathering. in the nw part of the quarry, the leucogranite is white and hard (fresh leucogranite) and few joints are observed. in the se part, the granite is white-yellowish, weathered and joints are very frequent. both parts of the quarry were historically used for buildings. it has been observed that weathering sharply decreases the mechanical strength of the rock surface and increases the porosity from ~1.1% in the non-weathered leucogranite to ~4% in weathered specimens (sanjurjo-sánchez and alves, 2006). the bregua granodiorite (igme. 1975) is a milky gray coarse-grain rock with biotite and muscovite. plagioclase is dominant but it shows major microcline megacrystals. it also shows biotite as well as muscovite, apatite, zircon, epidote and opaque minerals. when weathered, the granodiorite shows a reddish colour. this can be observed in most ashlars of buildings and most areas of the present quarry front, located about 8 km s from the town centre, and abandoned in the early 20th century. the mechanical strength of the weathered rock is lower than that of the weathered leucogranite (in both cases building ashlars and quarry surfaces) and porosity is around ~4% (sanjurjo-sánchez and alves. 2006). the studied buildings are las capuchinas church (cap), the colegiata de santa maria del campo (csm), the church of santiago (sc) and the casas de paredes (p). the first one was built using only san pedro leucogranite ashlars and the last one only using bregua granodiorite, while the others were built with both rock types. the studied buildings were constructed in different historic periods and re-buildings were made with the same rock type (table 2). this fact allows us to get rock samples exposed to surface weathering during different periods from one to more than seven centuries. they are also exposed to different air pollution intensities, at least in the last decades, due to traffic exhaust and industrial air pollution, as data from previous studies have shown (sanjurjosánchez et al., 2009, 2011b, 2012). exposure to sea spray should be similar as the studied buildings are between 50 and 500 m from the seaside. to define a weathering geochemical baseline, samples from both quarries were taken. samples csp-1 and csp-2 correspond to the fresh (unweathered) and weathered rock of san pedro leucogranite in the quarry, respectively. the sample bre-1 corresponds to the bregua granodiorite quarry. in this last quarry the rock shows a red pale colour indicating some weathering effects. table 2. studied samples and orientation of the façades and time exposure in the built environment (a coruña. spain). jorge sanjurjo sánchez, carlos alves and juan ramón vidal romaní assessing the weathering of granitic stones on historical urban buildings by geochemical indices f5 ta bl e 3. x r f re su lts (i n % ) o f t he s am pl es . f6 4. methods to compare the weathering indices and the age of exposition of building stones, samples were taken from façades of known age (construction period) and this information is presented in table 1. for our study, samples taken from the quarries are considered as no exposed in the built environment (that is zero age). sampling was carefully carried out on the studied buildings causing the least possible damage to the façades. material was removed from the surface of ashlars (about 1 cm in depth) with a chisel by gentle hammering. quarry samples were taken by taking rock blocks in order to use the sample for other analyses (water absorption) being results previously published (sanjurjosánchez and alves, 2006). the samples were taken at least 10 cm far from ashlars joints, to minimise contamination from nearby joint mortars. about 40 samples were taken from the façades. major and trace elements were estimated by x-ray fluorescence (xrf). thirty grams of every sample were taken and two 10 g aliquots were powdered and analyzed in a fluorescence spectrometer s4 pioneer of wavelength dispersion bruker-nonius. two other 5 g aliquots were fired to 975°c to obtain loss on ignition (loi). 5. results the results obtained must be analysed considering the behaviour of all the analysed elements with increasing weathering and the suitability of the used indices to assess such weathering. values of different oxides (table 3) show table 4. pearson’s correlation coefficient (r) of the elements in oxides for samples with the time exposure in the building environment. table 5. weathering indices calculated for the samples. jorge sanjurjo sánchez, carlos alves and juan ramón vidal romaní assessing the weathering of granitic stones on historical urban buildings by geochemical indices f7 some overlapping between samples of both considered lithologies, bregua samples present higher values of iron, calcium, titanium, potassium, zirconium and magnesium (which has null results in most of the san pedro samples) while san pedro samples present higher sodium and rubidium content. these differences might be explained by primary mineralogical differences between the considered rock types, with bregua being richer in mafic minerals while san pedro having higher albite content as is frequent in leucogranitic rocks. the oxide with higher variation is so3 (between 0% and 4.59%), probably reflecting the effects of weathering processes in the built environment. the possible correlations between oxides and the time elapsed since the sampled ashlars have been placed in the building façades are discussed considering the plots presented in figures 2 and 3 as well as tables 3 and 4. for the san pedro leucogranite changes in the content in most of the studied oxides did not showed correlation with the exposure time. thus, no clear correlation between oxides and the exposition time in the built environment is observed to assess if leaching progress with time at a more or less constant rate. however, a decrease in si is observed with time for san pedro samples (correlation coefficient of -0.56), being clearer in façades not oriented to south (table 3 and figure 2). there seems to be some positive trend with time in relation to so3 (correlation coefficient = table 6. pearson's correlation coefficient (r) and squared coefficient (r2) of the weathering indices for the san pedro leucogranite. table 7. pearson’s correlation coefficient (r) and squared coefficient (r²) of the weathering indices for the bregua granodiorite. 0.49), a result that might represent the effect of sulphate contamination in more recent works. soluble elements such as cl and ca seem to show increase with time, above all in façades not oriented to south. it must be considered that, depending on the orientation of the studied façades, different erosive patterns are commonly observed on buildings due to the different weathering rates because the different exposure to possible leaching effects by rainwater (and/ or capillary rise), wind (with the consequent transport of pollutants), hydrationdehydration and thermal cycles (depending on direct exposure to sunlight). in relation to the bregua granodiorite samples, it is observed that some oxides, such as al2o3 and p2o5, show a negative correlation with age (correlation coefficients of -0.85 and -0.81, respectively), that is, it seems that there is a leaching of such cations with increasing exposure and weathering. in this case, this seems not to be correlated with the orientation of façades. however, the correlation of other ions with age is positive, as occurs again in the case of cl, so3 and this time with mno (correlation coefficients of 0.61, 0.45 and 0.81, respectively). some scatter can be observed in most of these correlations, significantly increased due to the deviation of some samples (above all the quarry sample bre-1 and the sample from casas de paredes, p-08), but f8 evident in the figure 3 and table 3. despite the low number of samples in relation to the considered orientations of façades, it is noteworthy the good correlation of some ions with age in both rocks, such as cl and so3, as this could result from salts typically found in the building environment (arnold and zehnder. 1989), namely in coastal regions and in the urban environment (and uptake with time). as expected, the calculated weathering indices showed different trends. there is some overlapping of values between the two rock types and the only one that shows a clear separation is the alk index for which san pedro presents lower values (between 45.7 and 57.1) than the bregua type (between 59.00 and 69.23). regarding the san pedro leucogranite samples, it is observed a good correlation (correlation coefficients > 0.90) for some indices (figure 4, tables 5 and 6), especially because several of them consider some same elements. the observed differences can be attributed to the variation in the concentration of some oxides. for the bregua granodiorite (tables 5 and 7. figure 5), absolute weathering indices (imob and lc) show a good correlation with most relative weathering indices as the pearson’s correlation coefficient shows (it can achieve 0.99 between imob and lc). thus, absolute indices seem to be more consistent and considerable to assess weathering in such complex environment. the relative indices based on the consideration of al2o3 and/or loi as reference parameter (hc, acn, akn, loi, vr and cia) show several low correlation values in this rock type. this can be indicative of leaching of some elements in the built environment (and not previously in the quarry). moreover, the results indicate that hydration effects related to weathering of minerals are negligible. regarding the possible correlation between weathering degree and exposure time in the built environment, only some indices show a weak correlation (r=0.48 to +0.43) for the san pedro leucogranite (imob. mwpi. lc and sio2/r2o3 index). in addition, the rate of variation of the indices observed in south façades seems to be lower than for other façades, indicating a slower weathering rate and leaching of some elements (figure 6). if we consider that south façades are more exposed to direct sunlight and dry environments (and water remains in the porous systems of the rock surface and subsurface) than north façades, this weaker trend would indicate that south façades could be affected to lower weathering rates. moreover, it must be remarked that the granite of this buildings comes at least from two varieties of the rock with different weathering degrees. this certainly affects to the obtained result and it can be observed by visual inspection of the ashlars surfaces on the buildings that show very variable states of decay. for the bregua granodiorite, a very good positive correlation (correlation coefficient of 0.90) is found for alk (it considers k2o and na2o) and a weak (more scattered) correlation for pi. sa. soc hc. loi. rr. r2o3 and sio2/r2o3 index. the correlation of some indices is also different if we consider south and other façades (figure 7), as occurs with the leucogranite. thus, this consideration must be taken into account for the discussion of the results. it is remarkable that the number of samples considered for this granodiorite is lower (n=13) than the number of samples used for the san pedro leucogranite (n=21). in this case, the use of an only rock type (an only weathering degree observed in the quarry) is an important factor for the final results that provide the weathering indices. from these, index that consider k, al and h2o seem to provide a weathering vs. age correlation. 6. discussion 6.1. variability of the chemical composition of samples although some substances are not typically considered to assess most weathering indices, it is very interesting the apparent correlation between cl and so3 with age in the bregua granodiorite samples (and weaker for cao). for the san pedro leucogranite, a good correlation is observed for cao and so3, and weaker for cl. this cannot be correlated with the distance to the seaside, as can be observed in the table 1. in the case of the built environment, such substances can be considered indicators of weathering effects, and thus should be included in the weathering assessment, as they are due to uptake from the atmospheric environment (seaside air and air pollution in the case of cl and so3, respectively) or other sources (joint mortars and capillarity rise for so3). in the case of building walls made with the san pedro leucogranite it has been observed that gypsum coatings are present sometimes due to the migration of calcium from mortars (sanjurjo-sánchez et al., 2008; sanjurjo-sánchez et al., 2009). considering that the porosity of the weathered granite increases about four times (with reference to the fresh granite), the uptake of calcium and sulphate from joint mortars is possible through the porous system. air pollution in the area also supports the presence of sulphates (sajurjo-sánchez et al., 2011). thus, it is plausible that joint mortars are an important source of salts and weathering, but difficult to assess by most weathering indices. it would be considered that the ca and s content must be considered in similar cases (building stones in urban environments). the porosity of the bregua granodiorite is similar to the weathered leucogranite. thus, these elements (ca and s) could also be used to assess the uptake of pollutants in granite rocks in buildings and to assess the risk of damage due to salt growth, responsible for some typical decay features in these kinds of stone and environments (amoroso and fassina, 1983; sanjurjo-sánchez et al., 2009; sanjurjo-sánchez and alves, 2012b). 6.2. variability of the chemical indices the weathering indices calculated for a same rock type should reflect the degree of weathering of every sample. comparing the most common indices in the studied samples, a good correlation (among them) is observed for the most part (tables 6 and 7). however, an important scatter is observed. for the san pedro leucogranite, we cannot dismiss the leaching of some elements due to in situ chemical weathering (as observed in the case of si) and two probable causes are considered for the scatter. the first one is related to the previous weathering of the rock in the massif, as can be observed in the quarry. as referred above this rock quarry shows two varieties regarding weathering (fresh and weathered). both varieties show slight differences in some minor and major components (table 3, csp-1 and csp-2). this makes difficult the identification of both types within the buildings. however, the porosity observed in the studied samples implies that some of the sampled rock ashlars probably come from the weathered variety of the quarry and some others from the fresh variety (sanjurjo-sánchez and alves, 2006; sanjurjo-sánchez and vidal vázquez, 2013). thus, differences would be more significant in the weathering indices than in the oxides. this is due to the different starting point of the weathering process when the rock ashlars are placed in the façades. the higher differences found in the relationship between weathering indices and age for this rock type could be due to the rate of weathering not being high enough to affect significantly the weathering state of the rock in comparison with the quarry rock (either for an initial weathered or fresh variety). moreover, the samples taken from both las capuchinas church (cap) and from the oldest part of the colegiata de santa maria de campo (csm) are more scattered than the others. this may be attributed to two reasons. on one hand, the two different varieties found in the quarry and the different weathering progress, as related above. this can be observed in the present state of conservation of las capuchinas church, with strong decay of most ashlars and as the building was plastered with gypsum after construction as it has been demonstrated (sanjurjosánchez et al., 2009). on the other hand, some samples have been taken from façades of different orientation. while the samples taken from younger façades (one to five centuries) come from north façades, the older samples come from south (las capuchinas church and some colegiata de santa maria del campo samples) or east (only the csm-13 sample) façades and this is a typical cause of different rates of decay (sanjurjo-sánchez and alves, 2012a, 2012b). the most important source of scatter for bregua granodiorite samples are the samples taken from the quarry (bre-1) and from casas de paredes (p-08). in the first case, the quarry front shows that the rock has suffered some weathering, as can be observed in its pale red colour. the casas de paredes were built some years after the other considered buildings, and thus some scatter can be expected as the rock used was taken from other parts of the quarry, where some jorge sanjurjo sánchez, carlos alves and juan ramón vidal romaní assessing the weathering of granitic stones on historical urban buildings by geochemical indices f9 figure 2. plots of oxides content vs. exposition time in the built environment for san pedro leucogranite. key: open circles (south façades). filled squares (other façades). filled circles (quarry). differences in the weathering processes could occur. moreover, the weathering of a façade can also vary depending on its relation to the sources of air pollution. in this sense, p-08 corresponds to a building located in a street which supports intense car traffic (sanjurjo-sánchez et al., 2009, 2011b), being more exposed to some pollutants (higher concentration of air pollutants) that cause decay (amoroso and fassina, 1983; sanjurjo sánchez and alves. 2012a, 2012b). this is the most probable cause of the gypsum crusts observed in that building (sanjurjo-sánchez et al., 2008), despite analysed sample p-08) does not show an important uptake of ca and s. however, another source of scatter is the different orientation of façades. the weathering degree seems to occur at different rates depending on the orientation of façades that is different in several studied buildings (figures 2 and 3). despite this, some of the used indices provided good results (figure 4) and a good weathering vs. exposure age correlation, due to the leaching of some elements (al, k) and the increase of porosity (h2o content). thus, these elements must be considered to choose reliable indices to assess weathering on building rock surfaces in urban environments. in buildings, the direct exposition to the sun of south façades could result in marked wetting/drying cycles (wendler et al., 1990). wetter conditions can cause stronger dissolution and leaching of some mobile elements in the rock. however, the concentration of some mobile elements, such as ca, increase with exposition time to weathering (figure 2). the increase of cao with exposition time could be due to the dissolution of ca from lime mortars (joint mortars) and uptake within the rock ashlars’ porous system. this hypothesis is also supported by the uptake of so3 observed on samples, as the lime of mortars easily gives origin to the neoformation of gypsum under polluted air conditions in buildings of a coruña (sanjurjo-sánchez et al., 2009). therefore, indices that do not consider cao as a mobile oxide would be independent from influence of dissolution of ca form mortars. other mobile elements do not show this or any other clear trend. f10 figure 3. plots of oxides content vs. exposition time in the built environment for the bregua granodiorite. key: open circles (south façades). filled squares (other façades). filled circles (quarry). figure 4. plots of selected weathering indices (indices that show similar trend) for the san pedro leucogranite. key: open circles (south façades). filled squares (other façades). filled circles (quarry). jorge sanjurjo sánchez, carlos alves and juan ramón vidal romaní assessing the weathering of granitic stones on historical urban buildings by geochemical indices f11 figure 5. plots of selected weathering indices (indices that show similar trend) for the bregua granodiorite. key: open circles (south façades). filled squares (other façades). filled circles (quarry). figure 6. plots of weathering indices vs. exposition time in the built environment for the san pedro leucogranite. open circles (south façades). filled squares (other façades). filled circles (quarry). f12 figure 7. plots of weathering indices vs. exposition time in the built environment for the bregua granodiorite. open circles (south façades). filled squares (other façades). filled circles (quarry). regarding again the scatter of indices, despite the considered questions it seems that relative weathering indices using a high number of elements will usually be more consistent than those that consider a few oxides (tables 6 and 7). moreover, indices that consider the leaching of some elements (si, al, k) or the water content (that can be directly correlated with the increase of porosity) seem to be more consistent. thus, it must be considered in future studies that the use of weathering indices to assess weathering in buildings constructed with granite rocks in urban environments should consider: (i) any causes of the scatter observed in this study (façade orientation, exposure to direct air pollution, inherited weathering of rock in the quarry) previously to sampling, (ii) indices that include alkaline elements (mainly, ca, s and cl) will show the effects of pollution and interaction with building materials that will cause some physical and chemical weathering processes (e.g. salt weathering), and (iii) indices that include some elements (such as si, al and k) seem to be generally more consistent in these complex environments. 7. conclusions the applicability of weathering indices to the weathering of rocks used as building materials of historic buildings in urban environments has been assessed considering two igneous rock types in buildings of a coastal region with a certain amount of atmospheric pollution. theoretically, the progress of weathering on the stone surfaces can be assessed due to the leaching of some mobile elements. however, this effect is not clearly observed in the studied granite rocks from the considered urban environment in most elements. this is due to different factors that cause scatter of results. in this case, previous rock weathering in the quarry, the different orientation of façades and the different exposure to pollutants are proposed as the cause of such scatter. also, at the considered timescale (from one to seven centuries), weathering rates in south and other façades seem to be different. thus, future studies should consider these questions before sampling. if possible it is also recommended the use of a significant number of samples, as the use of 13-21 samples in this case is not very conclusive due to the referred scattering factors. in addition, the interference of some mobile ions that migrate from other building materials, as occurs with ca dissolved from lime joint mortars, does not advise the use of weathering indices that consider cao for the assessment of weathering losses. some other mobile oxides, such as so3 or cl, commonly related with other building materials, air pollution or other environmental factors should be monitored and used to assess the influence of such factors on the decay and weathering in the built environment. despite scattering, the use of weathering indices can be useful to assess the degree of surface weathering of granitic rocks used as construction materials in historical buildings of urban environments, as it has been observed in one of the studied rocks. moreover, indices that include some elements (such as si, al and k) seem to be more consistent. acknowledgments this work is included in the activities of the project lab2pt landscapes, heritage and territory laboratory aur/04509, which has financial support of the fct through national funds and when applicable of the feder co-financing, in the aim of the new partnership agreement pt2020 and compete2020 poci 01 0145 feder 007528. the authors are also grateful for funding from xunta de galicia, spain through projects r2014/001 and gpc2015/024. 8. references alves, c., sanjurjo-sánchez, j. (2015). conservation of stony materials in the built environment. environ. chem. lett. 13, 413-430. amoroso, g.g., fassina, v. (1983). stone decay and conservation: atmospheric pollution. cleaning. consolidation and protection. elsevier, amsterdam. arel, e., tugrul, a. (2001). weathering and its relation to geochemical properties of cavusbasi granitic rocks in northwestern turkey. b. eng. geo. environ. 60, 123-133. jorge sanjurjo sánchez, carlos alves and juan ramón vidal romaní assessing the weathering of granitic stones on historical urban buildings by geochemical indices f13 arnold, a., zehnder, k. (1989). salt weathering on monuments. 1st international symposium on the conservation of monuments in the mediterranean basin. bari, 13-23. chiu, c.f., ng, c.w.w. (2014). relationship between chemical weathering indices and physical and mechanical properties of decomposed granite. eng. geol. 179, 76-89. duzgoren-aydin, n.s., aydin, a., malpas, j. (2002). re-assessment of chemical weathering indices: case study on pyroclastic rocks of hong kong. eng. geol. 63, 99-119. esaki, t., jiang, k. (2000). comprehensive study of the weathered condition of 10 welded tuff from a historic stone bridge in kagoshima. japan. eng. geol. 55, 121-130. gerrard, a.j. (1988). rocks and landforms. ulwin hymen ltd. london. guan, p., ng, c.w.w., sun, m., tang, w. (2001). weathering indices for rhyolitic tuff and granite in hong kong. eng. geol. 59, 147-159. gupta, a.s., rao, k.s. (2001). weathering indices and their applicability for crystalline rocks. b. eng. geo. environ. 60, 201-221. harnois, j., moore, j.m. (1988). geochemistry and origin of the ore chemistry formations. a transported paleoregolith in the grenville province of southern ontario. canada. chem. geol. 69, 267-289. igme (1975). mapa geológico de españa 1:50000. la coruña. instituto geológico y minero. irfan, t.y. (1996). mineralogy. fabric properties and classification of weathered granites in hong kong. q. j. eng. geol. 29, 3-35. jayawardena, u.s., izawa, e. (1994). application of present indices of chemical weathering for precambrian metamorphic rocks in sri lanka. b. ass. eng. geol. 49, 55-61. kim, s., park, h.d. (2003). the relationship between physical and chemical weathering indices of granites around seoul. korea. b. eng. geol. environ. 62, 207-212. lee, c.h., yi, j.e. (2007). weathering damage evaluation of rock properties in the bunhwangsa temple stone pagoda. gyeongju. republic of korea. environ. geol. 52, 1193-1205. lee, d.s., lee, c.h., kim, j., yang, h.j. (2009). geochemical characteristics of surface efflorescence on the seventh century stone pagoda in republic of korea. environ. geol. 58, 197-204. meunier, a., sardini, p., robinet, j.c., prêt, d. (2007). the petrography of weathering processes: facts and outlooks. clay miner. 42, 415-435. moignien, r. (1966). review of research on laterites. natural resources research w. united nations educational. scientific and cultural organization, paris. moses, c., robinson, d., barlow, j. (2014). methods for measuring rock surface weathering and erosion: a critical review. earth-sci. rev. 135, 141-161. nesbitt, h.w., young, g.m. (1982). early proterozoic climates and plate motions inferred from major element chemistry of lutites. nature 299, 715-717. ng, c.w.w., guan, p., shang, y.j. (2001). weathering mechanisms and indices of the igenous rocks of hong kong. q. j. eng. geol. hydroge. 34, 133-151. ohta, t., arai, h. (2007). statistical empirical index of chemical weathering in igneous rocks: a new tool for evaluating the degree of weathering. chem. geol. 240, 280-297. parker, a. (1970). an index of weathering for silicate rocks. geol. mag. 107, 501-504. reiche, p. (1943). graphic representation of chemical weathering. j. sediment. petrol. 13, 58-68. rocha-filho, p., antunes, f.s., falcao, m.f.g. (1985). quantitative influence of the weathering degree upon the mechanical properties of a young gneiss residual soil. in: proceedinfs of the 1st international conference on geomechanics in tropical lateritic and saprolitic soils. brasilia. publ 1, 281-294. ruxton, b.p. (1968). measures of the degree of chemical weathering of rocks. the j. geol 76, 518-527. sanjurjo-sánchez, j., alves, c.a.s. (2006) degradación de rocas graníticas empleadas como material de construcción en edificios históricos de a coruña (nw españa). cadernos lab. xeol. laxe. 31, 11-28. sanjurjo-sánchez, j., alves, c. (2012a). decay effects of pollutants on materials applied in the built environment. in environmental chemistry for a sustainable world. volume 2: remediation of air and water pollution (lichtfouse.,e.; schwarzbauer. j.; robert. d., eds) springer. berlin, 47-121. sanjurjo-sánchez, j., alves, c. (2012b). decay effects of pollutants on stony materials in the built environment. environ. chem. lett. 10, 131-143. sanjurjo-sánchez, j., trindade, m.j., blanco-rotea, r., benavides garcia, r., fernández mosquera, d., burbidge, c., prudêncio, m.i., dias, m.i. (2010). chemical and mineralogical characterization of historic mortars from the santa eulalia de bóveda temple, nw spain. journal of archaeological science, 37, 9, 2346-2351 sanjurjo sánchez, j., vidal romaní, j.r., fernández mosquera, d., alves, c.a.s., (2008). study of origin and composition of coatings in a monument built with granitic rocks by sem, xrd and xrf and dta–tga. x-ray spec 37, 346–354. sanjurjo-sánchez, j., vidal-romani, j.r., alves, c.a.s., fernándezmosquera, d. (2009). origin of gypsum-rich coatings on historic buildings. water air soil poll. 204, 53-68. sanjurjo-sánchez, j., vidal-romaní, j.r., alves, c.a.s. (2011a). assessing the weathering degree of granite ashlars of heritage buildings by geochemical weathering. in: rock geochemistry (macias. b; guajardo. f., eds). novapublishers, new york, 103-120 sanjurjo-sánchez, j., vidal-romaní, j.r., alves, c.a.s. (2011b). deposition of particles on gypsum-rich coatings of heritage buildings in urban and rural environments. constr. build. mater. 25, 813-822. sanjurjo sánchez, j., vidal romaní, j.r., simoes alves, c.a. (2012). comparative analysis of coatings on granitic substrates from urban and natural settings (nw spain). geomorphology 138, 231-242. sanjurjo-sánchez, j., vidal vázquez, e. (2013). characterizing weathering of granite buildings by multifractal analysis of mercury intrusión porosimetry. journal of vadose zone. 12, 1-10 singh, b., roy, s.k., sharma, k.p., goswami, t.k. (1998). role of acidity of pillared inter-layered clay (pilc) for the synthesis of pyridine bases. j. chem. technol. biot. 71, 246-252. smith, b.j. (2003). background controls on urban stone decay: lessons from natural rock weathering. in: the effects of air pollution on the built environment. air pollution reviews. 2 (brimblecombe. p., ed.). imperial college press. london, 31-61. suoeka, t., lee, i.k., huramatsu, m., imamura, s. (1985). geomechanical properties and engineering classification for decomposed granite soils in kaduna district. nigeria. in: in: proceedinfs of the 1st international conference on geomechanics in tropical lateritic and saprolitic soils. brasilia. publ 1, 175-186. topal, t. (2002). quantification of weathering depths in slightly weathered tuffs. environ. geol. 42, 632-641. topal, t., sözmen, b. (2003). deterioration mechanisms of tuffs in midas monument. eng. geol. 68, 201-223. vogel, d.e. (1975). precambrian weathering in acid metavolcanic rocks from the superior province. villebon township. south-central québec. can. j. earth sci. 12, 2080-2085. wendler, e., klemm, d.d., snethlage, r. (1990). contour scaling on building facades dependence on stone type and environmental conditions. advanced workshop analytical methods for the investigation of damaged stone, pavia. as hydrological models become more prevalent in water resources planning and management, increasing levels of detail and precision are needed. currently, reliable models that simulate the hydrological behavior of a basin are indispensable; however, it is also necessary to know the limits of the predictability and reliability of the model outputs. the present study evaluates the influence of uncertainty in the main input variable of the model, rainfall, on the output uncertainty of a hydrological model. using concepts of identifiability and sensitivity, the uncertainty in the model structure and parameters was estimated. then, the output uncertainty caused by uncertainties in i) the rainfall amounts and ii) the periods of the rainfall was determined. the main conclusion is that uncertainty in rainfall estimation during rainy periods produces greater output uncertainty. however, in non-rainy periods, the output uncertainty is not very sensitive to the uncertainty in rainfall. finally, uncertainties in rainfall during the basin filling and emptying periods (apr. – jun. and sep. – nov., respectively) alter the uncertainty in subsequent periods. therefore, uncertainties in these periods could result in limited ranges of model predictability. los modelos hidrológicos se han vuelto cada vez más necesarios en la planificación y gestión de recursos hídricos, donde un aumento en los niveles de detalle y precisión es necesario. actualmente disponer de modelos para simular el comportamiento hidrológico de una cuenca resulta indispensable, sin embargo, también es necesario conocer los límites de predictibilidad y de confiabilidad de las salidas de un modelo. en este estudio se evalúan la influencia de la incertidumbre en la principal entrada de un modelo, la precipitación, sobre la incertidumbre de las salidas de un modelo hidrológico. utilizando conceptos de identificabilidad y sensibilidad se estima la incertidumbre de los parámetros y estructura de un modelo. luego, la incertidumbre en las salidas causadas por incertidumbre en i) los montos de precipitación, ii) los períodos de precipitación fue calculada. como conclusiones se obtuvo que la incertidumbre en la estimación de la precipitación en períodos de lluvia produce mayor incertidumbre sobre las salidas. en períodos no lluviosos, la incertidumbre de las salidas es poco sensible a incertidumbre sobre las precipitaciones. finalmente, incertidumbres en periodos de llenado y vaciado (abril-junio y septiembre-noviembre respectivamente) afectan la incertidumbre en las salidas en los períodos subsecuentes. por lo tanto incertidumbres en aquellos períodos pueden resultar en rangos limitados de predictibilidad de un modelo. earth sciences research journal eart sci. res. j. vol. 18, no. 1 (june, 2014): 69 75 abstract resumen key words: uncertainty, hydrological predictability, surface hydrology, water resources. palabras clave: incertidumbre, predictibilidad hidrológica, hidrología superficial, recursos hídricos. uncertainty in rainfall input data in a conceptual water balance model: effects on outputs and implications for predictability enrique muñoz, pedro tume, gabriel ortíz1 1. alonso de ribera 2850, concepción. departamento de ingeniería civil, universidad católica de la santísima concepción, concepción, chile mail: emunozo@ucsc.cl, ptume@ucsc.cl, gaortiz@ing.ucsc.cl meteorology record manuscript received: 16/07/2013 accepted for publication: 29/04/2014 70 enrique muñoz, pedro tume, gabriel ortíz introduction considering that i) water is a limited resource, ii) its demand increases with development and population growth, iii) its availability changes year to year due to local, global, natural and anthropogenic phenomena, iv) the impact of climate change on the water supply is of fundamental importance and v) a large part of the world’s population is already experiencing water stress (vörösmarty et al., 2000), it is necessary to develop tools for efficient water resources management and to support the prediction of future conditions under different scenarios caused by local (orographic), regional (el niño southern oscillation) or global effects (climate change). one alternative for efficient water resources management and planning is the use of hydrological models. a model attempts to reproduce a physical phenomenon that occurs in an object or area. therefore, in hydrology, a model seeks to represent an area defined by a watershed, the phenomena of rainfall-runoff processes and the associated water movement. the objective of reproducing these processes is to simulate and predict future conditions for management, administration and optimization of water uses. consequently, it is essential to have reliable models that adequately represent the hydrological behavior of a basin. for example, when using predictive models or alternative sources of information (e.g., global datasets) (mahe et al., 2008, muñoz, 2010), it is necessary to know the predictability limits of the model outputs and their uncertainty. normally, a conceptual hydrological model requires at least two input variables (potential evapotranspiration and rainfall) to quantify the inputs and water losses in the water balance. rainfall is the main input variable for a hydrological model (olsson and lindström, 2008). therefore, the potential predictability and the range of the predictability of the basin flows depend on the uncertainty associated with the input variables and their quality. uncertainty is defined as the degree of the lack of knowledge of or confidence in a certain process or result (caddy and mahorn, 1995). therefore, in a hydrological model, the sources of uncertainty are associated with the input variables (lack of knowledge of the quality of the measurements and predictions) and with the model structure and calibration parameters (lack of knowledge and simplification of the simulated hydrological processes in a basin). the present study aims to quantify how uncertainty in the outputs of a conceptual hydrological model is affected by uncertainty in the main input, rainfall, while keeping the uncertainty associated with the model structure and calibration parameters fixed. methods case study the polcura river basin (figure 1) is located in the temperate zone of south-central chile, between 37°20’s 71º31’w and 36°54’s 71°06’w. it is bounded by the andes to the east, the nevados de chillán volcanic complex to the north, and lake laja to the south. it comprises an area of 914 km2 that is between 700 and 3,090 masl; the area is characterized by steep slopes (≈ 26° on average) that are mainly composed of partially eroded volcanosedimentary sequences (om2c, ppl3 and m3i) (sernageomin, 2006). figure 1: location, boundaries and characteristics of the polcura river basin. in addition, the area is located in an extratropical zone that is greatly affected by el niño southern oscillation (enso) (grimm et al., 2000), which introduces seasonality and interannual variability to the local hydrometeorological patterns (grimm et al., 2000; montecinos and aceituno, 2003). the average annual precipitation in the basin is 2300 mm, with a pluvial period during winter and an ice-melt and snowmelt period in spring and early summer. the average monthly temperature is 9° c, ranging from 2.5° c in winter to 16.5 ° c in summer. because of the location of the basin, its mountainous nature and its geomorphology (figure 1), it exhibits high temporal variability with respect to hydro-meteorological characteristics, where the orographic effect on the eastern slope of the andes produces an increase in rainfall (garreaud, 2009; vicuña et al., 2011). additionally, the area is affected by enso. enso is a coupled ocean-atmosphere phenomenon that is characterized by irregular periodicity (2 to 7 years), where the alternation between el niño and la niña episodes is associated with above and below average rainfall and warmer and colder than normal air temperatures, respectively (garreaud, 2009). therefore, the characteristics of the basin, such as high rainfall variability and enso effects, provide an ideal case study in which uncertainty in precipitation could be the main source of uncertainty in predictions. description of the water balance model the model used in this paper is the snow-rain conceptual water balance model presented by muñoz (2010) and muñoz et al. (2014). this model separately simulates pluvial and snow-melting processes and includes external alterations, such as irrigation or transfer canals, by adding or subtracting flows. this model has been successfully implemented in andean basins in southcentral chile (e.g., zúñiga et al., 2012, arumí et al., 2012). therefore, it is an adequate option for analyzing the study area. the pluvial component is modeled using a lumped monthly rainfall-runoff model that treats the watershed as a double storage system: subsurface (ss) and underground storage (us). ss represents the water stored in the unsaturated soil layer as soil moisture. us is the water that covers the saturated soil layer. the model requires two inputs: rainfall (pm) and potential evapotranspiration (pet). the model output is the total runoff (etot) at the watershed outlet, and it includes both subterranean (es) and direct runoff (ei). these amounts are calculated by six calibration parameters and two input modifiers (useful in case of non-representative pm and pet data). the snowmelt model calculates snowfall (psnow) based on precipitation above the 0 (°c) isotherm. psnow is stored in the snow storage system (sn), for which the melting calculations are performed using the degree-day method (rango and martinec, 1995). thus, potential melting (psp) is estimated, and depending on the snow stored, real melting (ps) is calculated. then, ps is distributed in the pluvial model through calibration parameter f. every calibration parameter has a conceptual meaning for integrating spatial and temporal variability. table 1 presents a brief description of the model parameters and their influence on the model. the external alterations module permits the incorporation of alterations, such as irrigation or transfer canals, and simulates inflows and/or outflows to/ from the basin by adding or subtracting flows as follows: (1) where the basin outflow (qout) at time t is the basin runoff (etot) plus the contributions (qcontributions) and less the extractions (qextractions) during the same period. for a further explanation of the model, refer to muñoz (2010) and muñoz et al. (2014). q out (t) = etot (t) + qcontributions (t) qextractions(t) 71uncertainty in rainfall input data in a conceptual water balance model: effects on outputs and implications for predictability model inputs to run the described model, it is necessary to obtain precipitation, temperature and potential evapotranspiration series, as well as the geomorphological characterization of the basin to compute the monthly 0°c isotherm elevation. for the geomorphological characterization of the basin, a digital elevation model was constructed using data from the shuttle radar topography mission (srtm) of 3 arc-seconds (90 m). for the inputs, rainfall series from the rain gauge stations of las trancas, san lorenzo and trupán were obtained (administered by the dirección general de aguas, dga). temperature series from the center for climatic research of the university of delaware (ud) (willmot and matsuura, 2008) were collected. climatological model inputs were constructed using both sources of information. additionally, the potential evapotranspiration series were calculated using the thornthwaite method and the ud temperature data series. the spatial distribution of these variables in the basin was determined using thiessen polygons. based on prior experience, both the thornthwaite and thiessen polygons methods are adequate for estimating potential evapotranspiration and for determining the meteorological spatial distribution of monthly data in basins of south-central chile (e.g., muñoz, 2010, zúñiga et al., 2012). therefore, both methods were used in this study. because of the availability and quality of the input data, the analysis was performed on a monthly scale for a period of 13 years (1990 – 2002). the stream gauge station for controlling the basin outflows is the polcura antes de descarga central el toro station (figure 1). uncertainty analysis to quantify the uncertainty, the monte carlo analysis toolbox (mcat) (wagener and kollat, 2007) was used. mcat is a tool that operates under the generalized likelihood uncertainty estimation methodology (beven and binley, 1992), and it contains a group of analysis tools to explore the identifiability of a model and its parameters. mcat was chosen because it is an easy-to-use tool and is an adequate and simple option for evaluating model behavior and uncertainty. to evaluate identifiability and quantify the uncertainty in a model, mcat operates by performing repetitive simulations using a set of randomly selected parameters within a range defined by the user. the program stores the outputs and values of the objective function(s) for subsequent analyses. in this case, due to the lack of knowledge of the parameter distribution, a prior uniform distribution of the model parameters was used. parameters of hydrological models cannot be identified as unique sets of values. this is mainly because changes in one parameter can be compensated by changes in one or more other parameters due to their interdependence (bárdossy, 2007) and because the processes simulated in a hydrological model are commonly interrelated. in the case of the muñoz (2010) model, for example, direct runoff depends on cmax; therefore, the processes that occur in the subsurface storage layer depend on the amount of rainfall that is not transformed into runoff (i.e., 1-cmax); consequently, the processes that occur in this storage layer depend on the identifiability of cmax. this generates an interconnection among the identifiability of the calibration parameters of a model. therefore, it is necessary to perform various iterations as a means of restricting the range of the validity of the parameters that first show identifiability and to observe identifiability in the remaining (dependent) parameters as a means of reducing the identifiability range of these parameters. because the monte carlo method is based on random trials, it normally requires a large number of simulations that cover a wide spectrum of possible simulations. in this case, the number of monte carlo simulations was estimated via trial and error. the “stop” criterion was met when the correlation (according to the pearson correlation coefficient) between uncertainty bands (calculated as a linear correlation between the time series of the upper and lower limits of the uncertainty bands) of two trials with the same number of simulations was equal to or greater than 0.999. under this criterion, it was determined that the adequate number of simulations for this study was 25,000. the procedure to estimate the rainfall uncertainty and to differentiate it from the model structure and calibration parameters was as follows: through an identifiability analysis of the factors for the modification of the inputs (a comparison of these factors with the value of the objective function), these factors were limited to precise values. it is necessary to set these factors to unique values, as they ensure the long-term water balance in which the precipitation input volumes are equal to the evapotranspiration and flow output volumes during the entire period of the simulation. with the values of these fixed factors, new simulations were performed, in which the ranges of variation assigned to each parameter were reduced in table 1: description of the model parameters, adjustment factors, and conceptual physical range for the pluvial and snow-melting model. parameter description influence on range pl uv ia l m od ul e pa ra m et er s cmax maximum runoff coefficient when the sub-surface layer is saturated. ei 0.00 – 0.60 plim (mm) limit of rainfall for which ppd exists. ppd 0.00 500 d percentage of rainfall for which plim transforms into ppd. ppd 0.00 – 0.60 hmax (mm) maximum capacity of the soil layer to retain water. cmax and er 180 500 porc fraction of hmax that defines the soil water content restriction of the evaporation processes. hcrit and er 0 100 ck subterraneous runoff coefficient. es 0.05 – 0.60 a adjustment factor of the precipitation data. pm 0.80 – 2.50 b adjustment factor of the potential evapotranspiration data. pet and er 0.80 – 2.50 sn ow m od ul e pa ra m et er s m (mm °c-1) fraction of snowmelt over a base temperature (tb) at which melting starts. psp, ps 1 – 12 dm minimum rate of melting when tm < tb. psp, ps 0.00 – 0.50 f fraction of the real snowmelt that contributes to ei. ei 0.0 – 1.00 fgt adjustment factor of the thermic gradient data. psp, ps 1.00 – 5.00 72 enrique muñoz, pedro tume, gabriel ortíz accordance with the observed identifiability. the established stop criterion was defined as when identifiability was not observed in the new defined range (the reduced range) of each parameter. after the identifiability analysis, the resulting uncertainty in the model outputs was quantified. this uncertainty was defined as the uncertainty associated with the model structure and calibration parameters. the confidence level was 90% (a range between 5 and 95% of the outputs for each time step of the series). the established rejection criterion was set according to the kling-gupta efficiency objective function (kge) (gupta et al., 2009). the kge function (eq. 2) is an improvement of the nash-suctliffe efficiency index (nse) (nash and sutcliffe, 1970) in which correlation, deviation and variability are equally weighted to resolve systematic underestimation of the maximum values and low variability identified in the nse function (gupta et al., 2009). the kge varies between and 1, where the best model is closest to 1. because the glue methodology is based on likelihood, the kge was transformed into 1-kge. with this change, 1-kge represents a measure of likelihood in which the best model is closest to 0 and the worst model is + . where r is the linear correlation between the simulated and observed flows, α is the ratio between the standard deviation of the simulated and observed flows and represents the variability in the values, and β is the ratio between mean simulated and mean observed flows (i.e., the bias). then, with the uncertainty associated with the model structure and parameters quantified, the rainfall series were modified for different magnitudes and periods, as shown in table 2. it is important to note that this variation corresponds to a punctual variation of the rainfall series of different magnitudes and periods, and uncertainty is not added as a rainfall range in each time step. this analysis aims to identify how output uncertainty would be affected by an error in the precipitation prediction. table 2: precipitation percentage change in different time periods variation period variation (%) mean rainfall season (mm) entire year (jan. – dec.) ±5 ±5 ±15 ±20 ±25 2253 winter (jun. – aug.) ±5 ±10 ±15 ±20 ±25 805.4 summer (dec. – feb.) ±5 ±10 ±15 ±20 ±25 126.6 basin-filling period (apr. – jun.) ±5 ±10 ±15 ±20 ±25 808.4 basin-emptying period (sep. – nov.) ±5 ±10 ±15 ±20 ±25 368.7 results and discussion using the previously described methods, uncertainty associated with the model structure and parameters was found; then, the uncertainty related to rainfall data were computed and analyzed. the major findings and a discussion are presented as follows. uncertainty associated with model structure and calibration parameters figure 2 shows the identifiability and sensitivity analysis performed for the model parameters. the graphs show the probability distribution function (pdf) and the cumulative distribution function (cdf) of each parameter for the best 10% of the simulations according to a determined objective function (kge in this case). as observed, five parameters presented identifiability (according to the pdf): a, b, cmax, hmax and ck. according to the cdf, a, cmax and ck were the most sensitive (i.e., greater slope within a determined range) to the model outputs. a 1 1.5 2 0.5 1 d 0.2 0.3 0.4 0.5 0.5 1 plim 200 400 600 800 0.5 1 porc 20 40 60 80 0.5 1 dm 0.05 0.1 0.15 0.5 1 f 0.2 0.4 0.6 0.8 0.5 1 m 2 4 6 8 10 0.5 1 fgt 3.5 4 4.5 0.5 1 b 1.3 1.4 1.5 1.6 1.7 1.8 1.9 0.5 1 cmax 0.3 0.35 0.4 0.45 0.5 0.5 1 hmax pd f k g e; c df k g e 250 300 350 400 450 0.5 1 ck 0.1 0.2 0.3 0.4 0.5 0.5 1 [0.5 2.5] [0.5 2.5] [200 500] [0.0 0.6] [50 1000] [0 100] [0.05 0.60] [0.05 0.80] [0.0 0.2] [0 1] [1 12] [3 5] figure 2: analysis of the identifiability and sensitivity of the model parameters. the black line and the histogram in each graph represent the cdf and pdf of the best 10% of the simulations, respectively. the range indicated in the upper left corner is the initial range of the analysis for each parameter. because of the identifiability analysis, the ranges of variations in these five identifiable model parameters were reduced. a complementary analysis was performed in which the value of each parameter versus the value of the objective function was plotted. based on this analysis, it was determined that the factor that modifies the inputs (a) has an optimal value of 1.58. therefore, the differences between the simulated and observed flows, according to the kge, are minimal when a is 1.58. this analysis allows the value of the parameter to be fixed and the influence of the model structure and parameters on the output uncertainty to be separated from later analyses. the resulting value of parameter a suggests that the distributed rainfall over the basin was underestimated. the underestimation is caused by the orographic effect and the related enhancement of rainfall in mountainous areas (garreaud, 2009); these effects are not well recorded by meteorological stations, mainly because they are located in the lower part of the basin. finally, the aforementioned parameters were limited to the following ranges: i) b [1.2 – 1.7], ii) cmax [0.3 – 0.52], iii) hmax [220 – 500], and iv) ck [0.25 – 0.52]. the remaining parameters, which exhibited neither greater sensitivity nor identifiability, retained their initial ranges in the simulations. then, considering the aforementioned parameter ranges and using mcat, the output uncertainty associated with the model structure and calibration parameters was estimated. next, using this data, the influence of the input variables on the uncertainty in the output was estimated. uncertainty associated with the input rainfall data by incorporating the variations in the rainfall in accordance with table 2, new ranges of uncertainty were estimated. to identify these new ranges and observe the differences generated by the modification of the input variables, figures 3 to 7 show the uncertainty bands associated with the model structure and calibration parameters (white area between the black lines), the uncertainty associated with the positive variations in the rainfall (blue area), and the uncertainty associated with the negative variations in the rainfall (red area). additionally, to quantify the propagation of rainfall uncertainty effects, table 3 presents a summary of the average width of the uncertainty bands (in m3/s) for the different variations in the rainfall data. additionally, table 4 presents the percentage of variations in these ranges. figure 3 shows that the model outputs are sensitive to changes in the kge = 1 − (r −1)2+(α −1)2+(β −1)2 (2) 73uncertainty in rainfall input data in a conceptual water balance model: effects on outputs and implications for predictability rainfall amounts, where the range of uncertainty increases by an average of 15% for positive variations and 14% for negative variations in the rainfall (table 4). upon comparing these results with figure 3, it can be observed that overestimations of the rainfall amounts result in an increase in the uncertainty toward greater flows, while underestimations of the rainfall amounts result in an increase in the uncertainty of the outputs toward greater or reduced flows. figure 3: uncertainty bands associated with the model structure and calibration parameters (white area between the black lines) and uncertainty bands for the positive (blue) and negative rainfall variations (red) of 5, 10, 15, 20 and 25% (plots a, b, c, d and e, respectively) during the entire simulated period. figure 4 shows that errors in the prediction of the rainfall amounts during the winter period generate an effect on the flows of the recession curve by increasing the range of the uncertainty of the outputs of the model in the spring-summer (oct.-feb.) and early fall (mar.-apr.) (see the periods prior to the peak flows in figure 4) and by increasing the discharge amounts estimated by the model. (b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (c) q ( m 3 / s) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (d) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (e) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 tiempo (meses) figure 4: uncertainty bands associated with the model structure and calibration parameters (white area between the black lines) and uncertainty bands for the positive (blue) and negative rainfall variations (red) of 5, 10, 15, 20 and 25% (plots a, b, c, d and e, respectively) during the winter months (jun. – aug.) figure 5 shows that errors in the estimation of the rainfall in summer do not significantly affect uncertainty in the model outputs, most likely because the earlier and later flows of the period mainly depend on the base flow and water storage in the basin during winter, rather than the runoff that can be produced in summer. (a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (c) q (m 3 / s) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (d) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (e) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 tiempo (meses) figure 5: uncertainty bands associated with the model structure and calibration parameters (white area between the black lines) and uncertainty bands for the positive (blue) and negative rainfall variations (red) of 5, 10, 15, 20 and 25% (plots a, b, c, d and e, respectively) during the summer months (dec. – feb.) figure 6 shows that overestimations in the prediction of the rainfall during the basin-filling period produce a displacement of the output uncertainty bands toward greater flows but maintain a relatively constant uncertainty band width (tables 3 and 4). this result suggests that the model is highly sensitive to overestimations of the rainfall data in such a period, most likely because the model is not capable of redistributing the water in the basin processes; therefore, higher flows and similarly wide uncertainty bands are obtained. in contrast, underestimations of the rainfall in this period do not result in greater differences in the uncertainty band widths, suggesting that the uncertainty is related to different processes and/or parameters, such as the underground storage. (a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (c) q (m 3 / s) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (d) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (e) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 tiempo (meses) figure 6: uncertainty bands associated with the model structure and calibration parameters (white area between the black lines) and uncertainty bands for the positive (blue) and negative rainfall variations (red) of 5, 10, 15, 20 and 25% (plots a, b, c, d and e, respectively) during the basin-fill months (apr. – jun.). figure 7 shows that overestimations of the rainfall in the recession period of a basin produce an increase in the output uncertainty bands, increasing the bands toward greater flows while keeping the lower limit constant. (a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (c) q (m 3 / s) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (d) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (e) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 tiempo (meses) 74 enrique muñoz, pedro tume, gabriel ortíz (b) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (c) q (m 3 / s) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (d) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (e) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 (a) 1990 1991 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 0 100 200 tiempo (meses) figure 7: uncertainty bands for the positive (blue) and negative rainfall variations (red) of 5, 10, 15, 20 and 25% (plots a, b, c, d and e, respectively) during the basinrecession months (sep. – nov.). figures 3 to 7 show that errors in the estimation of the rainfall in a given period would affect the uncertainty in the model in subsequent periods. additionally, errors in the estimation of the rainfall during rainy periods, especially during the filling period of the basin, generate greater uncertainty in the model outputs. this is reasonable because the greatest water intakes occur in winter, when the basin produces water storage that influences the flows of the recession curve, in addition to surface runoff. these results are also consistent with the pluvial-snow regime of the basin, in which the generated flows are directly dependent on the rainfall inputs. figure 7, which analyzes the influence of the rainfall in the recession period of the basin, presents similar results for the uncertainty in the estimation of the rainfall in the summer months (figure 5); this result supports the concept of the important dependence of the model and its outputs on the inputs of water to the basin in rainy periods. tables 3 and 4 show that the model structure and parameters generate a 13 (m3/s) range of output uncertainty on average. after estimating how the model uncertainty is affected by uncertainty in the rainfall amounts during different periods, it is observed that the outputs are more sensitive to rainy periods. specifically, greater ranges of uncertainty and greater percentage changes in the uncertainty bands are observed. a smaller influence is observed for low water periods (recession and summer) when the model outputs are less sensitive to errors or uncertainties in the estimation of the rainfall amounts. however, in the filling period of the basin, it is observed that the width of the uncertainty bands does not significantly increase (8% on average), but it is displaced toward greater flows when rainfall is overestimated. table 3: average range (m3/s) of the uncertainty associated with the different variations (positive/negative) in precipitation. the range indicated for a variation of 0% corresponds to the range of uncertainty associated with the model structure and parameters. period 0% 5% 10% 15% 20% 25% entire year (jan. – dec.) 13 14/14 15/14 15/15 16/14 15/16 winter (jun. – aug.) 13 13/14 14/14 14/15 14/15 15/16 summer (dec. – feb.) 13 14/14 15/14 14/14 14/14 14/14 basin-filling period (apr. – jun.) 13 13/13 14/14 14/13 14/13 15/13 basin-emptying period (sep. – nov.) 13 14/14 14/14 15/14 15/18 15/13 table 4: flow percentage change in the range of the uncertainty produced by positive/negative variations in precipitation period 0% 5% 10% 15% 20% 25% entire year (jan. – dec.) 0 8/8 15/8 15/15 23/8 15/23 winter (jun. – aug.) 0 0/8 8/8 8/15 8/15 15/23 summer (dec. – feb.) 0 8/8 15/8 8/8 8/8 8/8 basin-filling period (apr. – jun.) 0 0/0 8/8 8/0 8/0 15/0 basin-emptying period (sep. – nov.) 0 8/8 8/8 15/8 15/18 15/0 conclusions the model is more sensitive to rainy periods; therefore, greater uncertainty in the estimation of rainfall during these periods directly influences uncertainty 75uncertainty in rainfall input data in a conceptual water balance model: effects on outputs and implications for predictability in the model outputs and increases the range of the related uncertainty bands. the output uncertainty is highly sensitive to rainfall. therefore, in dry periods, when the rainfall amounts are relatively small, the model is less sensitive to uncertainties in the rainfall than in the wet periods. currently, it is common to estimate flows and their uncertainty bands using a hydrological model. in the present study, it was observed that the uncertainty associated with the main water input (rainfall) in an andean basin can generate a significant increase in the uncertainty bands of the outputs. therefore, when using alternative sources of information to feed a hydrological model (e.g., global gridded data or data interpolated at a global scale), it is necessary to reduce the uncertainty associated with the predictions by ascertaining and reducing the deviation of the rainfall data in rainy periods (with respect to the observed or ground values) and/or by ascertaining the quality of the presented predictions. additionally, for predictive hydrological models, high uncertainty in periods of low rainfall does not greatly influence uncertainty in the output. however, when predicting flows in rainy periods or in recession periods, the uncertainty associated with pluviometric predictability should be evaluated. acknowledgements this research has been supported by the eu erasmus mundus external cooperation window, the conicyt programa nacional de formación de capital humano avanzado scholarship, the bmbf/conicyt project 2008099 impacto de la variabilidad climática en la disponibilidad de recursos hídricos y requerimientos de riego en la zona central de chile, and the fondecyt 11121287 project hydrological process dynamics in andean basins. identifying the driving forces, and implications in model predictability and climate change impact studies. the authors thank the dirección general de aguas for providing the rain gauge and stream gauge data. references arumí, j.l., d. rivera, e. muñoz, m. billib. (2012). interacciones aguas superficiales y subterráneas en la zona central de chile, obras y proyectos, 12, 4-13. bárdossy a. (2007). calibration of hydrological model parameters for ungauged catchments, hydrology and earth system sciences, 11(2), 703–710. beven, k., and binley, a. (1992). the future of distributed models: model calibration and uncertainty prediction, hydrological processes, 6, 279-298. caddy, j. and mahor, r. (1995). reference points for fisheries management. fao fisheries technical paper, 346, food and agriculture organization of united nations, rome, italy. garreaud, r. d. (2009). the andes climate and weather, advances in geosciences, 22, 3–11. grimm, a. m., barros, v. r., and doyle, m. e. (2000). climate variability in southern south america associated with el niño and la niña events, journal of climate, 13, 35-58. gupta, h. v., kling, h., yilmaz, k. k., and martinez, g. f. (2009). decomposition of the mean squared error and nse performance criteria: implications for improving hydrological modeling, journal of hydrology, 377, 80-91. mahe, g., girard, s., new, m., paturel, j-e., cres, a., dezetter, a., dieulin, c., boyer, j-f., rouche, n., and servat, e. (2008). comparing available rainfall gridded datasets for west africa and the impact on rainfallrunoff modelling results, the case of burkina-faso. water sa, 34(5), pp. 529-536. montecinos, a. and aceituno, p. (2003). seasonality of the enso-related rainfall variability in central chile and associated circulation anomalies, journal of climate, 16(2), 281-296. muñoz e. (2010). desarrollo de un modelo hidrológico como herramienta de apoyo para la gestión del agua. aplicación a la cuenca del río laja, chile. msc. thesis, departamento de ciencias y técnicas del agua y del medio ambiente, universidad de cantabria, españa. muñoz, e., rivera, d., vergara, f., tume, p., arumí, jl. (2014). identifiability analysis: towards constrained equifinality and reduced uncertainty in a conceptual model, hydrological sciences journal, 59(9), 1690-1703. nash, j. and sutcliffe, j. (1970). river flow forecasting through conceptual models part i: a discussion of principles, journal of hydrology, 10(3), 282–290. olsson, j., and lindström, j. (2008). evaluation and calibration of operational hydrological ensemble forecasts in sweden, journal of hydrology, 350, 14-24. rango, a. and martinec, j. (1995). revisting the degree-day method for snowmelt computations, journal of the american water resources association, 31(4), 657-669. sernageomin. (2006). mapa geológico de chile, servicio nacional de geología y minería, santiago de chile. vicuña, s., garreaud, r. d., and mcphee, j. (2011). climate change impacts on the hydrology of a snowmelt driven basin in semiarid chile, climatic change, 105, 469–488. vörösmarty c., fekete f., meybeck m., and lammers r. (2000). global system of rivers: it’s role in organizing continental land mass and defining land-to-ocean linkages, global biogeochemical cycles, 14(2), 599–621. wagener, t., and kollat, j. (2007). numerical and visual evaluation of hydrological and environmental models using the monte carlo analysis toolbox, environmental modelling and softwtware, 22, 10211033. wagener, t., lees, m., and wheater, h. (2001). a toolkit for the development and application of parsimonious hydrological models. in singh, frevert, and meyer, editors, mathematical models of small watershed hydrology, volume 2. water resources publications, llc, usa. willmott, c., and matsuura, k. (2008). terrestrial air temperature and precipitation: monthly and annual time series (1900–2008) version 1.02. university of delaware web site, http://climate.geog.udel.edu/ climate (last accessed february 2012). zúñiga, r., e. muñoz, j.l. arumí. (2012). estudio de los procesos hidrológicos de la cuenca del río diguillín. análisis mediante un modelo hidrológico conceptual, obras y proyectos, 11, 69-78. keywords: overhead mining; floor roadway; mining-induced stress; roadway stability. palabras clave: minería aérea; superficie de calzada; esfuerzo inducido por la minería; estabilidad de calzada. how to cite item wang, p., jiang, l., ma, c., & yuan, a. (2020). evolution laws of floor stress and stability of floor roadway affected by overhead mining. earth sciences research journal, 24(1), 45-54. doi: https://doi.org/10.15446/esrj. v24n1.67925 based on the geological conditions of a drainage roadway in the xinglongzhuang coalmine of shandong province (china), a mechanical model of a working face for overhead mining over a roadway was established, and the laws influencing the mining-induced stress were obtained. the evolution of mining-induced stress in the floor with different horizontal distances between working face and floor roadway was examined using universal distinct element code (udec) numerical simulation; the stability of the floor roadway was analyzed. the results of the numerical simulation were verified via on-site tests of the deformation of surrounding rocks and bolt pull-out readings from the drainage roadway. the results indicated that the floor roadway was significantly affected within a certain depth and then the effect weakened with the depth increasing. in view of the roadway stability, the roadway was observed with worse stability of surrounding rocks during the period of the working face mining near the roadway or just passing the roadway; moreover, this exploitation was not safe and needed to strengthen the support for the roadway. the results may provide a scientific basis for determining reasonable locations and support for roadways in xinglongzhuang coalmine and other coal mines with similar coal measures strata. abstract evolution laws of floor stress and stability of floor roadway affected by overhead mining leyes de evolución en el esfuerzo del suelo y de estabilidad de la superficie de una calzada afectada por minería aérea issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v24n1.67925 este artículo estableció un modelo mecánico de un frente de trabajo aéreo sobre una calzada con base en las condiciones geológicas de la calzada de drenaje de la mina de carbón xinglongzhuang y se obtuvieron los factores determinantes del esfuerzo inducido por la minería. parte del estudio examinó la evolución del esfuerzo inducido por minería en el suelo con diferentes distancias horizontales entre el frente de trabajo y la superficie de la calzada a través de una simulación numérica basada en el código de elemento distintivo universal (udec, del inglés universal distinct element code). la estabilidad del suelo también fue analizada. los resultados de la simulación numérica se verificaron a través de pruebas de campo de la deformación de las rocas circundantes y de las muestras de la calzada de drenaje. los resultados indican que la superficie de la calzada fue afectada significativamente hasta cierta profundiad y luego el efecto se debilitó a medida que la profundidad se incrementa. los resultados muestran que la estabilidad de la calzada registró el peor índice de las rocas circundantes durante el tiempo en que el frente de trabajo estuvo cerca de la calzada o tuvo que pasar por la calzada; lo que significó que esta explotación no es segura y necesitó el fortalecimiento de los soportes de la calzada. los resultados podrían proveer una base científica para determinar las ubicaciones razonables y los soportes adecuados para calzadas en la mina xinglongzhuan y en otras minas de carbón con medidas de capas similares. resumen record manuscript received: 24/08/2018 accepted for publication: 25/10/2019 earth sciences research journal earth sci. res. j. vol. 24, no. 1 (march, 2020): 45-54 pu wang1, lishuai jiang1*, changqing ma1, anying yuan2 1department of resources and civil engineering, shandong university of science and technology, tai’an, 271019, china 2state key laboratory of mining disaster prevention and control co founded by shandong province nd the ministry of science and technology, shandong university of science and technology, qingdao 266590, china * corresponding author: skwp0701@hotmail.com g e o l o g ic a l e n g in e e r in g 46 pu wang1, lishuai jiang, changqing ma, anying yuan introduction given the increase in the demand for coal, there is an increase in the depth and intensity of mining activities. in order to improve the recovery rate of coal resources and increase their economic benefits, overhead mining technology is widely used. it includes several advantages, such as a low drivage ratio, rational exploitation of resources, and stress-relaxed state for the roadway, and is especially applied to short-distance coal seams and thick coal seams (kang et al., 2011; wang et al., 2017). coal mining induces a stress readjustment and stress transfer to the deeper part of the floor. overhead mining significantly affects the floor roadway, and high mining stress concentrates in the surrounding rocks of the roadway to a certain depth, which induces serious deformation such as roof falls, floor heaves, and rib spalling (wang et al., 2015; wang et al., 2017; liu et al., 2017). several on-site photos of roadway instability are shown in figure 1. the stress and deformation of the floor roadway induced by mining activities differ with the position during overhead mining. hence, it is important to examine the evolution laws of the mining stress on the floor, and the stress distribution and failure of the floor roadway to select the location and roadway parameters. several studies have used different methods to focus on the distribution of the mining stress on the floor and the deformation-failure of floor roadway. the stress distribution and its propagation laws on the floor were studied by means of elastic mechanics theory, and the deformation characteristics of surrounding rocks of a floor roadway were obtained (yin et al., 1992; yuan et al., 2016). the distribution laws of the mining stress in floor and its additional stress were analyzed by utilizing numerical simulations and physical simulations, and reasonable parameters for a floor roadway layout were obtained (gao et al., 2015; ma et al., 2015; liu et al., 2016; feng et al., 2016). in terms of the deformation and stability of a roadway, the deformation characteristics of the surrounding rocks, such as the floor heave and roof fall of the roadway, were examined using on-site observations, physical simulations, and numerical simulations (islam and shinjo, 2009; li et al., 2016; jiang et al., 2016, 2017a). given the instability of roadways with a soft interlayer, the physical and chemical properties, along with the engineering mechanical properties of the soft interlayer were studied using rock mechanical tests, physical simulations, numerical simulations, and theoretical analysis. moreover, the instability characteristics and failure mechanism of a roadway with a soft interlayer were analyzed (li, 2011). because of the heterogeneity and anisotropy of the coal measures strata in different coal mines, as well as discrete factors such as interface joints, the commonly used method (i.e., theoretical analysis or empirical formula) can not accurately reflect the engineering problems that we intend to study. hence, based on the geological conditions of a drainage roadway in xinglongzhuang coalmine of shandong province (china), a mechanical model of a working face for overhead mining over the roadway is established. the influence of mining stress on the surrounding rocks of the roadway for different spatial positions of the working face and floor roadway are studied in detail by utilizing a universal distinct element code (udec) numerical simulation, and the danger area of the roadway layout under this coal measures strata condition is determined. this may serve as a scientific basis for floor roadway selection, which is affected by overhead mining in xinglongzhuang coalmine or its surrounding coal mines. it may also provide some roadway support experience and supporting parameters for similar coal measures strata. stress analysis of floor roadway affected by overhead mining geological conditions of overhead mining the drainage roadway under coal seam no. 3 of the shanxi group in the 10th district of the xinglongzhuang coalmine has a depth of approximately 606 m, which is close to coal seam no. 3. the surrounding rocks are primarily carbonaceous mudstone, and the internal fissures are extremely well developed. serious damage, including cracking and failure of the surrounding rocks of the roadway, had been induced by the overhead mining of panel 10302, which was in need of repair and support prior to reuse. mechanical analysis of floor stress affected by overhead mining given the relationship between the advancing direction of working face and floor roadway, the overhead mining was divided into mining with the advancing direction of the working face parallel to the roadway and mining in which the advancing direction was perpendicular to the roadway. hence, the effects of mining on the floor roadway differed. on-site observations indicated that the mining stress in the surrounding rocks of the floor roadway was affected by overhead mining when the advancing direction perpendicular to the roadway exceeded that of the other type of overhead mining, and the mining influence of the former was more evident (dai et al., 1999; yuan et al., 2016). hence, a mechanical model of overhead mining when the advancing direction was perpendicular to the roadway was established, and the laws influencing the mining stress of a floor roadway with different layers were analyzed. front abutment stress can be formed ahead of the coal wall because of the mining activities. in order to solve the problem conveniently, the front abutment stress could be simplified and deemed as a uniform load according to the reference written by lu et al. (2017). the distribution zone of the front abutment stress approximately corresponded to a rectangle with a length l and width w. the overlying load was viewed as a uniform load q, as illustrated in figure 2a. when the peak value of the abutment stress was just above the middle axis of the roadway, the mechanical model was simplified as a plane stress model, as shown in figure 2b (dai et al., 1999), and the vertical stress component σh of the surrounding rocks of the roadway was obtained as follows: σ π h q l h lh l h = − + +    2 4 4 2 2 arctan (1) where q denoted the average value of front abutment stress (mpa); l denoted the length of the working face (m), according to the geological report of the working face, l is 160 m; and σh denoted the vertical stress component of surrounding rocks of the roadway (mpa); h denoted the vertical distance between the working face and the floor roadway. figure 1. the deformation of the floor roadway is affected by mining activities (li, 2011). http://fanyi.baidu.com/#auto/auto/drivage ratio http://fanyi.baidu.com/#auto/auto/drivage ratio http://fanyi.baidu.com/#auto/auto/rectangle 47evolution laws of floor stress and stability of floor roadway affected by overhead mining maintained at about 23mpa. moreover, the curve slope that could reflect the attenuation rate of the stress in this range was very small, so it was in the low attenuation rate state and the roadway was affected by mining activities notably due to the high stress. with an increased in h, the σ0h linearly increased, while the σh decreased when 40 m < h ≤ 160 m. moreover, the attenuation rate of the vertical stress decreased notably due to the curve slope obviously increasing. hence, the difference between σh and σ 0 h decreased. these indicated that the effect of mining activities on the floor roadway weakened. the relationship between the vertical stress σh and vertical distance h between the working face and the floor roadway was obtained by curve fitting, as shown in eq. (3), and a high fitting accuracy was observed. the expression was as follows: h h h r= − × − × + =− −1 58 10 1 28 10 23 20 0 9774 2 3 2. . . . (3) evolution of mining stress in floors affected by overhead mining establishment of udec numerical model the numerical simulation methods commonly used in engineering and theoretical analysis can effectively solve complex engineering problems. the udec is a two-dimensional discrete element calculation program for based on the geological report of the working face, the mining depth h was approximately 566 m. thus, the in situ stress σ corresponding to level 566 m was calculated by eq. (2) as follows: σ γ= = × =h 25 5 566 14 43. .kn/m m mpa3 (2) where γ denoted the unit weight of the rock mass. in the present study, the unit weight of the overlying rock mass was assumed to be 25.5 kn/m3 (sainoki and mitri, 2015; wang et al., 2018). hence, the average value of the front abutment stress q was 23.09 mpa because the average value of the abutment stress concentration factor was 1.6, as indicated by the geological report of the working face. thus, the vertical stress σh of the surrounding rocks of the roadway with different positions was calculated by eq. (1), and the values were listed in table 1. table 1. vertical stress σh of surrounding rocks of the roadway with different positions. h(m) σh (mpa) σ 0 h (mpa) 10 23.09 14.69 20 23.08 14.94 40 22.96 15.45 60 22.66 15.96 80 22.16 16.47 100 21.49 16.98 120 20.69 17.49 140 19.81 18.00 160 18.90 18.51 180 17.99 19.02 as shown in figure 3 and table 1, when the vertical distance between the working face and floor roadway was less than 160 m (i.e., 0 m < h ≤ 160 m), the vertical stress σh of the surrounding rocks for the roadway always exceeded the corresponding in situ stress σ0h (i.e. the in situ stress of level which corresponds to different positions of the floor roadway). this indicated that the mining activities in this range affected the floor roadway and the propagation depth of the mining stress was about 160 m. when 0 m < h ≤ 40, the value of σh was greatly larger than its corresponding in situ stress σ 0 h and figure 2. a mechanical model for overhead mining: (a) spatial model and (b) planar model. (a) (b) figure 3. vertical stress σh of surrounding rocks of the roadway with different positions. 48 pu wang1, lishuai jiang, changqing ma, anying yuan discontinuous media that allows large deformation sliding, rotation, and detachment along the joint surface of a discrete rock mass. the structure of the overlying strata and distribution of the mining stress in the roof and floor are clearly reproduced. thus, it satisfied the needs of the present study (shnorhokian et al., 2014). based on the geological conditions of panel 10302 in xinglongzhuang coalmine, a udec numerical model was established with respect to the overhead mining and floor roadway. the thicknesses of the roof and floor of panel 10302 were simplified and adjusted to satisfy the research requirements. given the boundary effect, a 340 m (length) × 150 m (high) calculation model was established, as shown in figure 4. the mining depth and thickness of the coal seam in the model were 500 m and 10 m, respectively. the vertical distance between the coal seam and floor roadway was 5 m, and the horizontal distance from the left boundary of the model was 150 m. in the mine-wide model, meshes were more densely discretized around the coal from which the stopes were extracted compared to those in areas near the model boundary. thus, the discretization method made it possible to simulate the stress state in an active mining area as accurately as possible using the available computation capability (hofmann and scheepers, 2015; sainoki and mitri, 2015). figure 4. a numerical model for calculation. the mechanical properties assigned to the blocks and joints of the model were conventionally derived from laboratory testing programs (yu et al., 2016; jiang et al., 2017b). the physical and mechanical parameters of the rock mass derived from the geological report and rock mechanical test report of the working face employed in the model were listed in table 2. the mohrcoulomb model was used to simulate the continuous part of the rock because of the elastic and plastic deformation of the overlying strata, and the joint surface contact–coulomb slip model was used to simulate the joint calculation (itasca, 2005; sainoki and mitri, 2014). the full displacement constraint boundary was used at the bottom of the model, and the horizontal displacement constraint and vertical displacement free conditions were used at the left and right sides, respectively. the top corresponded to a free boundary. the simulated depth of the coal seam was 500 m, and the roof height of the model was 30 m. thus, the simulated failure of the overlying strata with a height of 470 m was assumed to be a result of the vertical stress that acted on the top of the numerical model. hence, the compensated load σv could be calculated by eq. (2) as follows: σ γv vh= = × =25 5 470 11 99. .kn/m m mpa3 where hv denoted the height of the failed simulated strata. in the present study, the height of the failed simulation was set at 470 m. according to the geological report from the xinglongzhuan coalmine, the value of the horizontal stress applied to the two sides of model with the trapezoidal distribution was half of the value of vertical stress. evolution of mining stress in floors affected by overhead mining the floor roadway was significantly influenced by overhead mining within the range of 0 m < h ≤ 40 m, as discussed in section 2, but the effect decreased with an increase in h. hence, in this study, based on the condition of the floor roadway located below a working face of 5 m, the laws influencing the mining stress in floor of the roadway (as affected by the overhead mining) were examined. the horizontal distance between the working face and floor roadway was defined as ld, corresponding to 60 m, 40 m, 20 m, 0 m, -20 m and -60 m (where the negative value indicates the working face had passed the roadway) as shown in figures 5 – 10. abutment stress was not only concentrated in front of the coal wall, but also would transfer to the bottom of coal floor affected by mining activities. as shown in figures 5 and 6, when the horizontal distance between the working face and floor roadway were 60 m and 40 m (i.e., ld = 60 m, 40m), the abutment stress was formed ahead of coal wall with 56 m and 47 m, respectively. however, the floor roadway was beyond this range when ld = 60 m which indicated that it was not affected by overhead mining; thus the surrounding rocks of floor roadway did not fail. when ld =40 m, the floor roadway was in the margin of the abutment stress-affected zone with the small stress concentration factor which reflected the degree of stress concentration (stress concentration factor was the ratio of concentration stress to in situ stress); thus, only the roadway roof failed. hence, it indicated that the effect of overhead mining on the floor roadway was slight and the roadway support did not need to be strengthened when ld≥40 m. from figures 7 and 8, as the working face gradually approaching the roadway, the effect of overhead mining on the roadway enhanced and the floor roadway was located in the stress significant influence area with a high-stress concentration factor of 1.6; thereby the surrounding rocks of roadways seriously destroyed, and plastic failure distributed around the roadway. hence, when 40m>ld≥0 m, the effect of overhead mining on the floor roadway enhanced notably by the high concentration stress and the roadway support should be paid more attention to. from figure 9, when 0 m>ld≥-20 m (the negative value indicated that the working face had passed the floor roadway), the working face was passing through the roadway and the roadway was beneath the goaf of the working face; thereby locating in a stress-relaxed zone with a small stress concentration factor of 0.2 when ld=-20 m. it meant that the effect of overhead mining on the floor roadway decreased gradually due to the small vertical stress. however, during this progress of mining operation, the roadway had experienced the effect of table 2. mechanical properties of the rock mass. lithology thickness (m) density (kg.m-3) bulk (gpa) shear (gpa) poisson ratio tensile strength (mpa) friction angle (º) siltstone 18 2530 1.34 0.76 0.2 3.08 37 medium sand 10 2520 1.83 1.15 0.18 7.00 41 mudstone 2 2450 0.98 0.71 0.24 1.80 31 coal seam 10 1350 1.05 0.65 0.28 1.50 20 siltstone 15 2530 1.34 0.76 0.28 3.08 37 medium sand 20 2520 1.83 1.15 0.18 7.00 41 mudstone 75 2340 0.98 0.71 0.24 1.80 31 49evolution laws of floor stress and stability of floor roadway affected by overhead mining figure 5. floor stress affected by overhead mining when ld = 60 m: (a) distribution of floor stress; (b) plastic zone of coal roof and floor. figure 6. floor stress affected by overhead mining when ld = 40 m: (a) distribution of floor stress; (b) plastic zone of coal roof and floor. figure 7. floor stress affected by overhead mining when ld = 20 m: (a) distribution of floor stress; (b) plastic zone of coal roof and floor. figure 8. floor stress affected by overhead mining when ld = 0 m: (a) distribution of floor stress; (b) plastic zone of coal roof and floor. (a) (b) (a) (b) (a) (b) (a) (b) 50 pu wang1, lishuai jiang, changqing ma, anying yuan the largest stress and then the effect decreased with the stress dropping; thereby inducing serious failure of surrounding rocks of roadway. hence, in this stage, even though the roadway located in the goaf-affected zone gradually, it had experienced the notable effect of the largest stress, hence the serious failure was observed and the range of plastic failure increased as shown in figure 9(b). as shown in figure 10, when -20 m>ld≥-60 m, as the goaf zone increased, the vertical stress in the goaf recovered and increased with constant compaction of migrated overlying strata. the effect of the stress on the floor roadway increased gradually and the stress concentration factor increased to 1.2 which indicated the floor roadway re-located in the stress increasing zone when ld=-60 m. hence, the vertical stress in the goaf re-affected the roadway. this effect by the vertical stress in recovery zone of the goaf was often neglected although it might influence the roadway significantly, the failure of surrounding rocks of roadway aggravated again as shown in figure 10(b). hence, when -20 m>ld≥-60 m, the roadway was affected by the recovered stress in the goaf and the failure of roadway aggravated further which indicated that the effect of the recovered stress on the roadway should not be ignored and necessary support should be considered with respect to the roadway. hence, according to the aforementioned analysis, the evolution laws of the vertical stress and the effect on the floor roadway were obtained, as shown in table 3 and figure 11. the entire process of overhead mining with respect to the effect on the floor roadway was divided into four stages. stage i: 40 m ≤ ld ≤ 60 m. the stress concentration factor of the rocks surrounding the roadway exceeded 1 and rose slightly with a decrease in ld. hence, the effect of overhead mining on the floor roadway was slight and the roadway did not need strengthened support. stage ii: 0 m ≤ ld < 40 m. the stress concentration factor of the rocks surrounding the roadway significantly increased and tended to be stable with a decreased in ld, and the maximum value increased to 1.6. hence, the effect of overhead mining on the floor roadway enhanced notably affected by high concentration stress and the roadway support should be paid more attention to. stage iii: -20 m ≤ ld < 0 m. the roadway was in the stress-relaxed state and the stress concentration factor significantly decreased with the working face passing through the roadway. however, it had experienced the notable effect of the largest stress during this stage, hence the serious failure of roadway was observed and the range of plastic failure increased. thus the floor roadway support should be more strengthened. stage iv: -60 m ≤ ld < -20 m. the abutment stress in the goaf corresponded to recovery and increased with the constant compaction of the migrated overlying strata. the floor roadway re-located in the stress increasing zone. hence, the roadway was affected by the recovered stress in the goaf and the failure of roadway aggravated further which indicated that the effect of the recovered stress on the roadway should not be ignored and necessary support should be considered with respect to the roadway. figure 9. floor stress affected by overhead mining when ld = -20 m: (a) distribution of floor stress; (b) plastic zone of coal roof and floor. (a) (b) figure 10. floor stress affected by overhead mining when ld = -60 m: (a) distribution of floor stress; (b) plastic zone of coal roof and floor. (a) (b) table 3. stress concentration factor of rocks surrounding of floor roadway. ld (m) stress concentration factor 60 1 40 1.2 20 1.6 0 1.6 -20 0.2 -40 0.3 -60 1.2 51evolution laws of floor stress and stability of floor roadway affected by overhead mining engineering case a significant effect of overhead mining on the floor roadway was observed in the above analysis of the theory and numerical simulation. hence, onsite observations of the deformation and failure of a drainage roadway in xinglongzhuang coalmine were used to verify the previously mentioned results. deformation monitoring of roadway. in this study, the deformation monitoring of the roadway mainly included the relative displacement of roof to floor, which reflected the vertical deformation, and the relative displacement of two ribs, which represented the horizontal deformation. hence, seven sections of the roadway (marked as 1–7), which corresponding positions were shown in figure 12, were surveyed at 30 m intervals along the roadway with the mining range of 300 m. make a sectional plane at the midpoint of working face, and obtain the three states between the working face and the floor roadway, including ahead of floor roadway, right above floor roadway and rear of floor roadway, as shown in figure 13. figure 14a shows the the monitoring points of surrounding rock deformation for the roadway section. table 4 and figure 14 showed the final results of deformation survey for the floor roadway within the mining range of 300 m when the working face had passed through the roadway and the roadway was no longer affected by mining operation. hence, it can be seen the vertical and horizontal deformations of the drainage roadway in sections 1–4 were large, with both values approximately 1100–1300 mm. this was because the roadway in this testing range was notably affected by the highly vertical stress of panel 10302 mining, and a large-scale plastic failure zone was formed in the rocks surrounding the figure 11. relationship between ld and stress concentration factor. figure 12. plan of panel 10302 and the layout of roadway sections. (a) ahead of roadway (b) right above roadway (c) rear of roadway figure 13. three states related the working face and the floor roadway. table 4. results of roadway deformation. sequence of sections relative displacement of roof to floor convergence (mm) relative displacement of two ribs convergence (mm) no.1 1156 1203 no.2 1321 1078 no.3 1210 1135 no.4 1098 1254 average value 1196 1167 no.5 701 743 no.6 621 703 no.7 598 664 average value 640 703 52 pu wang1, lishuai jiang, changqing ma, anying yuan roadway. thus, the strength of the surrounding rocks decreased, and large deformation was observed in the roadway. moreover, many fractures formed in the floor strata because of the mining activities. this resulted in seepage from the goaf penetrating the floor, which further reduced the strength of the surrounding rocks. however, the deformations of sections 5–7 were relatively small, with values of 600–700 mm. according to the geological report for the xinglongzhuang coal mine, this testing range (sections 5–7) was affected by mining activities slightly compared to that of sections 1–4, and the deformation was mainly contributed by the weakening effect of the seepage from the goaf. thus, the deformations in sections 5–7 were much smaller than those of sections 1–4. hence, it can be concluded that the effect of overhead mining on the floor roadway was significant, and the deformation of the surrounding rocks increased notably. so this exploitation is not safe and the support for the roadway should be strengthened. in order to study the relationship between the roadway deformation and the mining activities in detail, section 2 which was notably affected by mining were selected; and then the deformation of roof-floor and two ribs with the mining range of 300 m was surveyed. figure 15 showed the variation curves of deformation for the roadway with mining activities. as can be seen from figure 15, the variations of roof-floor convergence and two ribs convergence of section 2 were similar and showed obvious stage characteristics during the face progress. when the horizontal distance between working face and floor roadway exceeds 47 m, the roadway was affected by mining activities slightly, so both the roof-floor convergence and two ribs convergence were small and had no much change, as shown in figure 15 with stage (a), which the pressure appearance from the filed observation was similar to the simulation results with 40 m ≤ ld ≤ 60 m in section 3.2. when the horizontal distance between working face and floor roadway was in range of -34 m ~ 47m (the negative value indicated that the working face had passed the floor roadway), the deformation of roadway occurred significantly due to the violet effect of mining activation, resulting in both the displacement of rib-floor and two ribs convergence increasing sharply, as shown in figure 15 with stage (b). hence, compared with the fore-mention numerical simulation, the pressure appearance of filed observation was similar to the simulation results with -20 m ≤ ld < 40 m in section 3.2. when the working face passed through the floor roadway exceeds 34 m, the floor roadway was under the goaf of panel 10302, and the effect of mining operations on the roadway reduced gradually where roadway failure was no longer aggravated. hence, both the roof-floor convergence and two ribs convergence rose slightly and tended to be stable, as shown in figure 12 with stage (c). however, the both displacements were large and peak values in survey range were 1321 mm and 1078 mm (shown in table 4) respectively. at this stage, the pressure appearance of filed observation was similar to the simulation results with -60 m ≤ ld < -20 m in section 3.2. pull-out test analysis of bolts. based on the stress characteristics of the bolts, it was possible to assess the stability of the surrounding rock and the rationality of the bolt support of the roadway. hence, pull-out tests of the existing bolts used for the drainage roadway were performed using a pull-out device (zy-30), in which the maximum pull-out force was 300 kn. three bolts (distributed in the roof and two ribs) were arranged in each section for the test, as shown in figure 16a. the results of the pull-out tests are shown in table 5 and figure 16b. as shown in table 5 and figure 16, the pull-out force for most of the bolts was unable to satisfy the minimum requirement of 50 kn as indicated in the geological report of xinglongzhuan coalmine. the qualified rates for the bolts in the roof and two ribs corresponded to 42.9%, 28.6%, and 42.9%, respectively, with an average value of only 38.1%. this indicated that the existing support of the roadway had a poor performance during the overhead mining of panel 10302. based on the study of lu et al. (1998), an important reason for a low bolt pull-out force was an inability to fully utilize the anchoring capacity of the bolt system (i.e., in fractured surrounding rock, the existence of more cracks lead to bolt failure) or the existence of serious damage to the surrounding (a) layout of monitoring points (b) deformation of roadway figure 14. results of roadway deformation. figure 15. deformations of section 2 of roadway with the working face mining. 53evolution laws of floor stress and stability of floor roadway affected by overhead mining rock, which prevented the bolt from reaching its anchorage capacity. however, according to the results of the roadway deformation monitoring, the overhead mining of panel 10302 significantly affected the floor roadway, many fractures formed in floor strata, and the deformation of the surrounding rocks increased notably. the strength of the surrounding rocks decreased. hence, the bolts might not have fully reached their anchoring capacity as a result of the overhead mining, which produced a low pull-out force for the bolts. thus, the on-site results for the surrounding rock deformation and the pull-out tests of the existing bolts of the drainage roadway indicated that the floor roadway was affected by overhead mining significantly and its deformation-failure were serious; thus more attention should be paid on the similar coal measures strata. additionally, the observation results and the numerical simulation results showed a good agreement which can further explain rationality of numerical model. conclusions in this study, the laws influencing the mining-induced stress in different position of a roadway and the evolution of the mining-induced stress in floor with different horizontal distances between the working face and the floor roadway were examined to investigate the influence of overhead mining on a floor roadway. a mechanical model of a working face for overhead mining over a roadway was established, and a numerical model was generated using udec software. it was found that the depth influencing the mining stress on the floor corresponded to 160 m. thus, the mining stress was high and slowly decreased within a range of less than 40 m, through which the floor roadway was significantly affected, while the stress significantly decreased within a range exceeding 40 m, through which the effect of the stress on the roadway evidently decreased. the mining stress on the floor gradually increased, which led to an increase in the failure of the rocks surrounding the floor roadway within 0 m ≤ ld ≤ 60 m, and the effect was particularly evident in a range of 0 m ≤ ld ≤ 40 m. although the floor roadway was in a stress-relaxed state, the worse stability of the surrounding rock was observed in a range of -20 m ≤ ld < 0 m. the floor roadway was affected by the recovery of the abutment stress in the goaf when -60 m ≤ ld < 20 m, and thus attention should be focused on strengthening the support. finally, the surrounding rock deformation and bolts pull-out tests of a drainage roadway in xinglongzhuang coalmine were used to verify the results of the numerical simulation. the results could be utilized to serve as a scientific basis for floor roadway selection, which was affected by overhead mining in the xinglongzhuang coalmine or its surrounding coal mines and might also provide some roadway support experience and supporting parameters for similar coal measures strata. acknowledgments: this study was funded by the national natural science foundation of china (no. 51574155, no. 51504145, no. 51704182 and no. 51704185), the natural science foundation of shandong province (no. zr2017bee050 and no. zr2017bee045), the shandong provincial key r&d plan (public welfare special program) of china (no. 2017ggx20125), scientific research foundation of shandong university of science and technology for recruited talents (no. 2015rcjj057), tai’an science and technology development plan of shandong province (no. 2015zc1058), science and technology innovation fund of the college of the mining and safety engineering, shandong university of science and technology (no. kykc17008). references dai, j., li, h., li, f. h., & liu, t. x. (1999). the influence of working face advancing direction on the maintenance of overhead mining. mine pressure and roof management, 3–4, 190–193. feng, q., jiang, b. s., wang, g. & hu, c. p. (2016). analytical solution for a circular roadway considering the transient effect of excavation unloading. international journal of mining science and technology, 26(4), 543–549. (a) layout of test points (b) results of the pull-out tests of the bolts figure 16. layout of test points and results of the pull-out tests of the bolts. table 5. results of the pull-out tests. sequence of sections the pullout force of roof (kn) the pullout force of left rib (kn) the pullout force of right rib (kn) note: disqualification position 1 46 54 74 roof 2 52 28 38 two ribs 3 39 45 28 roof and right rib 4 36 61 53 roof 5 52 54 32 right rib 6 29 33 61 roof and left rib 7 51 41 28 two ribs qualified rate 42.9 % 28.6 % 42.9 % average: 38.1 % http://www.sciencedirect.com/science/article/pii/s2095268616300106 http://www.sciencedirect.com/science/article/pii/s2095268616300106 http://www.sciencedirect.com/science/article/pii/s2095268616300106 54 pu wang1, lishuai jiang, changqing ma, anying yuan gao, f. q., stead, d. & kang, h. p. (2015). numerical simulation of squeezing failure in a coal mine roadway due to mining-induced stresses. rock mechanics and rock engineering, 48(4), 1635–1645. hofmann, g. f. & scheepers, l. j. (2015). simulating fault slip areas of mining induced seismic tremors using static boundary element numerical modeling. transactions of the institution of mining & metallurgy, 120(1), 53–64. islam, m. r. & shinjo, r. (2009). numerical simulation of stress distributions and displacements around an entry roadway with igneous intrusion and potential sources of seam gas emission of the barapukuria coal mine, nw bangladesh. international journal of coal geology, 78(4), 249–262. itasca. (2005). universal distinct element code user’s guide (version 4.0). itasca consulting group, minneapolis, usa, 16 pp. jiang, l. s., sainoki, a., mitri, h. s., ma, n. j., liu, h. t. & hao, z. (2016). influence of fracture-induced weakening on coal mine gateroad stability. international journal of rock mechanics & mining sciences, 88, 307–317. jiang, l. s., zhang, p. p., chen, l. j., hao, z., sainoki, a., mitri, h.s. & wang, q.b. (2017a). numerical approach for goaf-side entry layout and yield pillar design in fractured ground conditions. rock mechanics and rock engineering, 50(11), 3049–3071. jiang, l. s., wang, p., zhang, p. p., zheng, p. q. & xu, b. (2017b). numerical analysis of the effects induced by normal faults and dip angles on rock bursts. comptes rendus mecanique, 345(10), 690–705. kang, q. r., tang, j. x., hai, h. u. & zhang, w. z. (2011). stress distribution rule of roadway affected by overhead mining in gently inclined coal seams group. transactions of nonferrous metals society of china, 21(s3), 529–535. li, g. c. (2011). study on the surrounding rock stability and safety control of roadways roof embedded weak intercalated seam. ph.d. thesis, china university of mining and technology, beijing, china. li, q. h., shi, w. p. & qin, z. c. (2016). effect of bolting on roadway support in extremely weak rock. springerplus, 5(1), 1355–1372. li, y. y. (2011). study on the technology of surrounding rock control in floor roadway under repeated overhead mining. ph.d. thesis, china university of mining and technology, beijing, china. liu, x. j., li, x. m. & pan, w. d. (2016). analysis on the floor stress distribution and roadway position in the close distance coal seams. arabian journal of geosciences, 9(2), 1–8. liu, c., li, s. g., cheng, c. & cheng, x. y. (2017). identification methods for anomalous stress region in coal roadways based on microseismic information and numerical simulation. international journal of mining science and technology, 27(3), 525–530. lu, h. f., yao, d. x., hu, y. b., sun, j. (2017). elasticity solution for failure depth of mining floor under water pressure. journal of mining & safety engineering, 34(3), 452–458. lu, s. l., tang, l., yang, x. a. (1998). anchorage force of bolts and anchorage technology.china coal industry publishing house, beijing, 34p. ma, r., li, g. c., zhang, n., liu, c., wei, y. h. & zhang, m. (2015). analysis on mechanism and key factors of surrounding rock instability in deeply inclined roadway affected by argillation and water seepage. international journal of mining science and technology, 25(3), 465–471. sainoki, a. & mitri, h. s. (2015) evaluation of fault-slip potential due to shearing of fault asperities. canadian geotechnical journal, 52, 1417–1425. sainoki, a. & mitri, h. s. (2014b). dynamic modelling of fault-slip with barton’s shear strength model. international journal of rock mechanics & mining sciences, 67(6), 155–163. shnorhokian, s., mitri, h. s. & thibodeau, d. (2014). numerical simulation of pre-mining stress field in a heterogeneous rockmass. international journal of rock mechanics & mining sciences, 66(9), 13–18. wang, b., liu, s. d., zhou, f. b., zhang, j. & zheng, f. k. (2017). diffraction characteristics of small fault ahead of tunnel face in coal roadway. earth sciences research journal, 21(2), 95–99. wang, f. t., duan, c. h., tu, s. h., liang, n. n. & bai, q. s. (2017). hydraulic support crushed mechanism for the shallow seam mining face under the roadway pillars of room mining goaf. international journal of mining science and technology, 27(5), 853–860. wang, p., jiang, l. s., jiang, j. q., zheng, p. q. & li, w. (2018). strata behaviors and rock burst-inducing mechanism under the coupling effect of a hard, thick stratum and a normal fault. international journal of geomechanics, 18(2), 04017135. wang, w., cheng, y. p., wang, h.f., liu, h. y., wang, l., li, w. & jiang, j. y. (2015). fracture failure analysis of hard-thick sandstone roof and its controlling effect on gas emission in underground ultra-thick coal extraction. engineering failure analysis, 54, 150–162. yin, h. f., wang, h. & zhu, h. x. (1992). applying the space elastic theory to analyse and compute the stress distribution in the floor affected by extract. journal of fuxin mining institute, 11(1), 19–24. yu, b., zhao, j., fang, k., tan, y. l. & ning, j. g. (2016). rock strength evaluation during progressive failure process based on fractural characterization. marine georesources & geotechnology, 34(8), 1–5. yuan, a. y., yang, z. y. & yang, y. m. (2016). across mechanical analysis and control technology of surrounding rock of roadway under the influence of dynamic pressure. metal mine, 45(2), 47–50. http://www.sciencedirect.com/science/article/pii/s1631072117301249 http://www.sciencedirect.com/science/article/pii/s1631072117301249 http://www.sciencedirect.com/science/article/pii/s1631072117301249 the fenghuangshan copper deposit, in the tongling area of anhui province, china, is genetically associated with the fenghuangshan granodiorite, which contains amphiboles that can be used to determine the pressure and temperature history of the magmas that formed this intrusion. here, we investigate the petrography and mineralogy of the fenghuangshan granodiorite, focusing on variations in amphibole composition and determining the conditions of amphibole crystallization. the data are used to determine the petrogenetic history and metallogenic significance of the hosting granodiorite. the amphiboles within the intrusion can be divided by texture into phenocryst, matrix, and poikilocrystal amphiboles. electro model, combined with the barometry and thermometry and the stratigraphy of the study area, indicates that the fenghuangshan copper deposit has not undergone any significant changes since its formation, thereby indicating that both the granodiorite and the associated mineralization are well preserved. this also suggests that the final depth of emplacement of the magma (3–6 km) represents the depth of formation of the associated copper deposit. this inference, combined with the fact that the current depth of exploration and mining in the study area is <1 km, indicates the great potential for discovering significant copper mineralization by deep exploration in this area. el depósito cuprífero de fenghuangshan, en el área de tongling, provincia de anhui, china, está genéticamente asociado con la granodiorita de fenghuangshan, que contiene anfíboles que pueden ser usados para determinar la presión y la temperatura histórica de los magmas que formaron esta intrusión. en este trabajo se investigo la petrografía y la mineralogía de la granodiorita de fenghuangshan enfocados en las variaciones de composición de anfíboles y en determinar las condiciones de cristalización de estos. los datos se utilizaron para determinar la historia petrogenética y la significación metalogénica de la granodiorita anfitriona. los anfíboles al interior de la intrusión pueden dividirse por texturas en fenocristales, cristales en matriz y cristales moteados. los análisis de microsonda electrónica indican que los anfíboles son de características magnesiohornblenda y tschermakita. las plagioclasas en la granodiorita son andesinas y oligoclasas. la barometría del aluminio en los anfíboles y la termometría de los anfíboles plagioclásticos indican que los anfíboles moteados cristalizados se fundieron a presiones entre 4,46 y 4,74 kbar y temperaturas de 1066 °c y 1071 °c, y los anfíboles en la matriz se formaron bajo condiciones de 1,00-1,91 kbar y 784-823 °c. los datos indican que el magma que formó la granodiorita fue generado dentro del manto superior y migró a través de fracturas antes de ubicarse en la corteza media-baja a una profundidad de ~15 km, lo que formó una profunda cámara magmática donde el magma sufrió una cristalización y asimilación fraccionada. el magma migró luego a través de estructuras geológicas (como fallas y fracturas) antes de ubicarse en la corteza a poca profundidad, entre 3 y 6 km, y formando eventualmente la granodiorita fenghuangshan. este modelo combinado con la barimetría, la termometría y con la estratigrafía del área de estudio indica que el depósito cuprífero de fenghuangshan no ha experimentado cambios sustanciales desde su formación, lo que indica que tanto la granodiorita como la mineralización asociada están bien preservadas. esta condición también nos sugiere que la profundidad final de la ubicación del magma (3-6 km) representa la profundidad de formación del depósito cuprífero asociado. esta inferencia, combinada con el hecho de que la actual exploración y explotación en el área de estudio es de <1 km, señala que el gran potencial para descubrir mineralización cuprífera con una exploración profunda en el área. petrology and mineralogy of the fenghuanshan granodiorite, anhui province, china: implications for petrogenesis and metallogenesis implication aiping zhang, yangsong du, yi cao, zhenshan pang² , gan luo¹ , kangkang xu¹ 1. school of the earth sciences and resources, china university of geosciences, beijing 100083, china 2. development and research center of china geological survey, beijing 100037, china correspondence should be addressed to aiping zhang, e-mail: zhangaiping_cugb@163.com earth sciences research journal earth sci. res. j. vol. 19, no. 1 (june, 2015): 31 37 abstract resumen key words: fenghuangshan granodiorite, mineralogy, geothermometry–barometry, emplacement depth, denudation, tongling, anhui province, china. palabras clave: granodiorita de fenghuangshan, mineralogia, geotermometría-barometría, profundidad de emplazamiento, denudación, tongling, provincia de anhui, china. petrology record manuscript received: 25/08/2014 accepted for publication: 04/02/2015 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n1.45202 32 aiping zhang, yangsong du, yi cao, zhenshan pang , gan luo , kangkang xu introduction amphibole is widespread in igneous rocks, can form in a number of tectonic settings, and plays an important role in the magmatic evolution of igneous systems. many studies have indicated that al substitution in amphibole is controlled by both temperature and pressure (e.g., anderson, 1980; hammarstrom et al., 1986; johnson et al., 1989; schmidt, 1992; ernst, 1999; davidson et al., 2007; martin, 2007; larocque et al., 2010; ridolfi et al., 2010; dessimoz et al., 2012; krawczynski et al., 2012), indicating that amphiboles can accurately record the pressure and temperature evolution of a magma (zen, 1989; tulloch et al., 2000; stein et al., 2001; lu et al., 2011). the middle–lower yangtze river polymetallic belt (mlyrb) is one of the most economically important metallogenic belts in china, and contains more than 200 cu, fe, au, mo, zn, pb, and ag deposits (li, 1989; chang et al., 1991; pan and dong, 1999). the tongling ore district is located within anhui province, within the central mlymb (fig. 1), and hosts widespread igneous rocks that temporally and spatially related to mineralization. dozens of metallic ore deposits have been discovered in this ore district, including the tongguanshan, fenghuangshan, baocun, chaoshan, jiguanshan, and dongguashan deposits, among others. the fenghuangshan deposit is a skarn cu deposit that is typical of such mineralization in this area (chang et al., 1991; zhai et al., 1992; fig. 1). the copper mineralization within this deposit is closely related to the fenghuangshan intrusion, with mineralization hosted by both the intrusion and the contact between the intrusion and the surrounding triassic limestone. in addition, the mineralization in this area (141.71 ± 0.82 ma) was formed at the same time as the fenghuangshan intrusion (143.1 ± 1.6 ma; li et al., 2014). this means that research into the fenghuangshan intrusion can provide insights into the process associated with the formation of magma and mineralization in this district. previous geochemical and geochronological research in this area has focused on the origin of the magma that formed the fenghuangshan intrusion, the processes involved in the emplacement of the magma, and the ages of intrusion and mineralization. however, little research has been undertaken on the crystallization conditions of minerals within the granodiorite, in turn meaning that little is known about this crucial mineralogical evidence of the magmatic conditions and ore-forming processes associated with the fenghuangshan deposit. here, we describe variations in amphibole composition within the fenghuangshan granodiorite and use these data to estimate the magmatic conditions and reveal the petrogenetic and metallogenic significance of the fenghuangshan granodiorite. these data enable the determination of the depth of emplacement and the degree of denudation that this area has undergone, providing useful information on the preservation history of the fenghuangshan copper deposit and the prospectivity of the surrounding region. fig.1 sketch map of the middle–lower yangtze river polymetallic belt, east china. the inset is a simplified structural map of china. tlf-tancheng-lujiang fault, xgf-xiangfan-guangji fault, ycf-yangxingchangzhou fault. modified from chang et al. (1991), mao et al. (2011), tang et al. 1998), and zhai et al. (1992). 2. geological setting the fenghuangshan deposit is located in the eastern part of the tongling ore district, 35 km southeast of tongling. the geology of this area is dominated by the fenghuangshan granodiorite, which intruded the core of a ne–sw-trending synclinorium (fig. 2). the country rocks are dominantly triassic limestones that have metamorphosed into marble, forming a 400 to 1200-m-wide thermal metamorphic aureole around the intrusion. the fenghuangshan intrusion formed during the late yanshanian, is high-k and calc-alkaline, and is dominated by a granodiorite phase (qu et al., 2010; li et al., 2014). the geochemistry of the granodiorite indicates that it formed from a deeply sourced alkaline magma that assimilated crustal material; this crustal contamination may have occurred during partial melting (i.e., contamination of the source) or during ascent and the emplacement of the magma. as such, this suggests the magma that formed the intrusion underwent typical assimilation and fractional crystallization (afc) processes (liu et al., 2002; mao et al., 2004; wang et al., 2008; qu et al., 2010; li et al., 2014). the fenghuangshan granodiorite, which is the largest intrusion in the area, is circular in cross-section, and is exposed over an area of about 10 km2. it is dominated by a granodiorite phase and is cross-cut by some later trachytic, doleritic, and lamprophyre dikes. six cu–fe–au deposits are located near the contact zone of the fenghuangshan granodiorite: the fenghuangshan, jiangjiachong, qingshuitang, xianrenchong, tieshantou, and baoshantao deposits (fig. 2). the granodiorite samples analyzed in this study were collected from the –440 m level of the fenghuangshan copper deposit. fig.2 location of the fenghuangshan ore field in tongling, anhui province (a) and a sketch geological map of the ore field (b). modified from 321 geological team of bureau of geology and mineral exploration of anhui province (1989). 33petrology and mineralogy of the fenghuanshan granodiorite, anhui province, china: implications for petrogenesis and metallogenesis sample fh011 fh100 spots 1a 2a 3a 4a 5a 1b 2b sio2 49.56 48.32 48.11 49.21 48.30 43.32 43.00 al2o3 5.38 6.40 6.40 5.31 6.09 9.44 10.84 feo* 12.79 12.96 13.53 14.84 15.32 19.95 17.74 mno 0.49 0.50 0.45 0.50 0.58 0.55 0.63 mgo 16.37 15.88 16.15 15.10 14.33 11.44 11.95 cao 11.66 11.59 11.36 11.51 11.84 11.58 11.44 na2o 1.26 1.59 1.60 0.91 1.15 1.61 1.83 k2o 0.05 0.07 0.07 0.04 0.05 0.13 0.12 total 97.56 97.31 97.67 97.42 97.66 98.02 97.55 si t 7.15 7.02 6.97 7.16 7.05 6.43 6.37 al (ⅳ) t 0.85 0.98 1.03 0.84 0.95 1.57 1.63 total t 8.00 8.00 8.00 8.00 8.00 8.00 8.00 al (ⅳ) b+c 0.07 0.12 0.06 0.07 0.09 0.08 0.27 fe 3+ b+c 0.41 0.40 0.51 0.51 0.53 1.00 0.81 fe 2+ b+c 1.13 1.18 1.12 1.29 1.34 1.48 1.39 mn b+c 0.06 0.06 0.06 0.06 0.07 0.07 0.08 mg b+c 3.52 3.44 3.49 3.27 3.12 2.53 2.64 ca b+c 1.80 1.80 1.76 1.79 1.85 1.84 1.82 total c + b 7.00 7.00 7.00 7.00 7.00 7.00 7.00 na a 0.35 0.45 0.45 0.26 0.33 0.46 0.53 k a 0.01 0.01 0.01 0.01 0.01 0.02 0.02 total a 0.36 0.46 0.46 0.26 0.33 0.49 0.55 p(kbar) 1.01 1.64 1.91 1.00 1.53 4.46 4.74 d(km) 3.35 5.42 6.31 3.32 5.05 14.72 15.63 3. petrology the fenghuangshan intrusion is granodioritic and has a porphyraceous texture (fig. 2a) comprising phenocrysts of plagioclase (30%), amphibole (10%), quartz (15%), k-feldspar (10%), and biotite (3%) in a fine-grained matrix of the same minerals with accessory zircon, apatite, and sphene (fig. 3a, b). the amphiboles within the intrusion can be texturally divided into phenocryst, matrix, and poikilocrystal amphiboles, and plagioclase within the intrusion is divided into phenocryst and matrix groups. amphibole phenocrysts are 1–3 mm in size, are euhedral, have rhombus-shaped cross-sections, and are rimmed by opaque magnetite (fig. 3c, d). these phenocrysts coexist with plagioclase phenocrysts (fig. 3c), with the latter having oscillatory zoned overgrowths and partially resorbed textures (fig. 3a, b). matrix amphibole within the intrusion is present as small (<0.5 mm) subhedral–anhedral crystals (fig. 3b, d), some of which are optically twinned. nearly all of the matrix amphibole also coexists with matrix plagioclase (fig. 3d). the amphibole poikilocrystals within the intrusion are small (<0.2 mm) and euhedral, and are enclosed within the cores of plagioclase phenocrysts (fig. 3b), indicating that these poikilocrystals crystalized earlier than or at the same time as the amphibole phenocrysts. this indicates that the amphibole poikilocrystals were the first amphiboles to crystalize from the magma, whereas the matrix amphibole was the last amphibole to crystalize out, indicating that these differing amphiboles represent two different phases of the evolution of the magma that formed the fenghuangshan intrusion. fig. 3 microphotographs of amphibole and plagioclase from the fenghuangshan granodiorite. (a)-granodiorite showing typical porphyraceous texture (crossed polarized light). (b)-phenocryst of zoned plagioclase containing an inclusion of amphibole (crossed polarized light). (c)-phenocryst of euhedral amphibole with a hexagonal cross-section and with nearby magnetite grains (plane polarized light); (d)-amphibole and plagioclase in the matrix are <0.2 mm in size (plane polarized light); pl-plagioclase, am-amphibole, bt-biotite, q-quartz; mt-magnetite. 4. mineralogy electron microprobe analyses (epma) were undertaken at the beijing research institute of uranium geology, beijing, china, using a jxa-8100 instrument operated at a 20 kv accelerating voltage, 10 na beam current, and using a 10 mm beam diameter, giving a detection limit of 0.002 wt%. analyses were calibrated using albite (na), sanidine (si, al, and k), diopside (ca and mg), almandine (fe), rutile (ti), fluorapatite (p), and rhodonite (mn) standards. table 1 results of compositional analyses (wt%) of amphiboles fig.4 classification of amphibole (a) and plagioclase (b) in the fenghuangshan granodiorites after the international mineralogical association (leake et al., 1997) and smith (1974), respectively. a-amphibole in the matrix, b-amphibole poikilocrystals 34 aiping zhang, yangsong du, yi cao, zhenshan pang , gan luo , kangkang xu 34 sample fh011 spots 1c 2c 3c 4c 5c 1d 2d sio2 58.73 58.26 58.49 61.68 60.37 59.15 60.60 tio2 0.07 0.08 0.08 0.00 0.00 0.00 0.00 al2o3 25.08 25.31 25.48 23.38 24.51 25.78 25.31 feo 0.19 0.33 0.24 0.27 0.24 0.20 0.24 mgo 0.00 0.00 0.00 0.00 0.00 0.00 0.02 cao 7.15 7.11 7.48 4.98 6.07 6.61 6.16 na2o 7.39 7.90 7.28 9.19 8.49 7.49 7.63 k2o 0.06 0.04 0.03 0.02 0.02 0.27 0.42 total 98.67 99.06 99.08 99.52 98.28 99.52 100.39 si 2.66 2.63 2.64 2.75 2.70 2.65 2.69 ti 0.00 0.00 0.00 0.00 0.00 al 1.34 1.35 1.35 1.23 1.29 1.36 1.32 fe 2+ 0.01 0.01 0.01 0.01 0.01 0.01 0.01 mg 0.00 0.00 0.00 0.00 0.00 0.00 0.00 ca 0.35 0.34 0.36 0.24 0.29 0.32 0.29 na 0.65 0.69 0.64 0.80 0.74 0.65 0.66 k 0.00 0.00 0.00 0.00 0.00 0.02 0.02 cr 0.00 0.00 0.00 0.00 0.00 ab 64.94 66.64 63.67 76.87 71.60 66.15 67.42 an 34.72 33.14 36.15 23.02 28.29 32.26 30.12 or 0.35 0.22 0.17 0.11 0.11 1.59 2.47 p(kbar) 1.01 1.64 1.91 1.00 1.53 4.46 4.74 t(°c) 798.50 812.29 822.73 784.26 803.41 1066.39 1070.55 fh100 table 2 results of compositional analyses (wt%) of plagioclase c-plagioclase in the matrix; d-core of plagioclase phenocrysts coexisting with amphibole poikilocrystals the classification of amphiboles is based on the composition of the standard amphibole formula ab � ivc5ivt8 o � � (oh) � . the composition and stoichiometrically calculated chemical formulae of representative amphiboles from the fenghuangshan granodiorite are given in table 1, with mineral formula calculations based on 23 oxygens and standardized on 13 cations. the calculation method of droop gtr (1987) was used to adjust fe²+ and fe³+ concentrations, and the electrovalency balance principle was used to calculate crystal formulae. amphibole stoichiometry and nomenclature followed the international mineralogical association (ima; leake et al., 1997; fig. 4). all of the amphiboles analyzed during this study are calcic amphiboles, defined by (ca + na)b values of ≥1.00, with na values of <0.50 and cab values of ≥1.50 (leake et al., 1997). consequently, they are classified as magnesiohornblende and tschermakite amphiboles, with the matrix amphiboles being classified as magnesiohornblende and the lower-si amphibole poikilocrystals being classified as tschermakite. magnesiohornblende within the intrusion has si values of 6.97– 7.16, al(iv)t contents of 0.84–1.03, cab values of 1.96–1.85, (na + k)˄ values of 0.27–0.46, and ca˄ values of <0.5. in comparison, tschermakite within the intrusion has si values of 6.37–6.43, al(iv)t values of 1.57– 1.63, cab values of 1.82–1.84, (na + k)˄ values of 0.48–0.55, ca˄ values of <0.5, and mg/(fe2+ + mg) values of 0.63–0.66 (table 1). the compositions and calculated formulae of the plagioclase analyzed during this study are given in table 2, with structural plagioclase formulae calculated on the basis of 8 oxygens and 5 cations. these data indicate that the plagioclase has an values of 23.02–36.15 and that the granodiorite contains oligoclase (an = 23–28) and andesine (an = 30–36) plagioclase (fig. 4). both matrix and phenocryst plagioclase have similar compositions, with na2o and cao concentrations increasing from 7.49 to 8.49 wt% and decreasing from 6.61 to 6.07 wt%, respectively, from core to rim in zoned plagioclase. 5. barometry and thermometry amphibole is stable over a wide range of pressure (p) and temperature (t), making it useful for determining the geothermometry and geobarometry of calc-alkaline intrusive magmatic systems (blundy et al., 1990). the total al content of amphibole can be used to estimate the crystallization pressure of a magma (hammarstrom et al., 1986; hollister et al., 1987; johnson et al., 1989; ghent et al., 1991; vyhnal et al., 1991; ague et al., 1992; schmidt, 1992). the first empirical al-in-amphibole barometer was proposed by hammarstrom (1986; eq. 1), who used this barometer with hornblende compositions to estimate the pressure of formation of intermediate calc-alkaline plutons. hollister (1987) reduced the error of this barometer by using a larger dataset for calibration (eq. 2) with a similar correlation. johnson and rutherford’s (1989) calibra tion (eq. 3) was obtained by performing experiments on natural samples under set co² + h � o volatile pressures, with further experimental calibration of this barometer under water-saturated conditions (eq. 4) by schmidt (1992). p (+3 kbar) = –3.92 + 5.03 a1t (1) p (+1 kbar) = –4.76 + 5.64 a1t (2) p (±0.5 kbar) = –3.46 + 4.23 a1t (3) p (±0.6 kbar) = –3.01 + 4.76 a1t (4) the last two calibrations are more commonly used because they are experimentally rather than empirically derived and have lower uncertainties that the first two calibrations. here, we use the calibration of schmidt (1992; eq. 4) to calculate the pressures obtained from amphibole analyses. the present amphiboles yield pressures of 1.00–4.76 kbar, with the highest values (4.46–4.74 kbar) being derived from the amphibole poikilocrystals. blundy (1990) published an empirical thermometer based on the edenite– tremolite reaction that can only be used on quartz-bearing, intermediate to felsic igneous rocks containing plagioclase with compositions of an ≤92% and amphibole with si values of ≤7.8, as follows: where k = xab, xab = the composition of plagioclase coexisting with amphibole, and using the pressure values derived from the al-inamphibole barometry described above. the fenghuangshan granodiorite has sio2 concentrations that vary between 60.82 wt% and 65.97 wt% (shao et al., 2003; li et al., 2014), contains plagioclase with an values between 23.02 and 36.15, and contains amphibole with si values of 6.37–7.16. all of these features mean that these samples are suitable for amphibole–plagioclase thermometry (blundy et al., 1990), yielding crystallization temperatures of 784°c–1071°c, with amphibole poikilocrystals yielding both higher pressures (as detailed above) and temperatures (1066°c–1071°c). 6. discussion 6.1 crystallization conditions the fact that poikilocrystals and matrix amphiboles within the fenghuangshan granodiorite have different compositions suggests that they formed in two different magmatic environments. the poikilocrystals most likely formed in a deeper and hotter setting, in a deep magma chamber, whereas the matrix amphiboles probably formed within a shallow magma chamber under the conditions of the final emplacement of the magma, and these amphiboles therefore record the depth and temperature of emplacement. the barometry and thermometry undertaken during this study indicate that the amphibole t= 35petrology and mineralogy of the fenghuanshan granodiorite, anhui province, china: implications for petrogenesis and metallogenesis poikilocrystals crystallized from the melt at pressures of 4.46–4.74 kbar and temperatures of 1066°c–1071°c, whereas matrix amphiboles formed at 1.00–1.91 kbar and 784°c–823°c. the pressures are probably equivalent to the depth of emplacement of the magma, which, when combined with the fact that the average crustal density of the mlyrb is 3.3 g/cm3, yields a depth of 14.72–15.63 km (average of 15.16 km) and a temperature of 1066°c–1071°c (average of 1068°c) for the deep magma chamber, and an emplacement depth and temperature for the shallow magmatic chamber of 3.32–6.32 km (average of 4.7 km) and 784°c–823°c (average of 804°c), respectively. the geochemistry of the fenghuangshan granodiorite suggests that the intrusion has an adakitic affinity similar to that of other intrusions in the tongling area (du et al., 2004, 2007; cao et al., 2008, 2010 ; wang et al., 2008; mao et al., 2009; zhai et al., 2010; wu et al., 2014), which have initial 87sr/86sr ratios of 0.709074 – 0.709965 and end(t) values of –9.26 to –16.25 (xing et al., 1996). this suggests that the magma that formed the fenghuangshan granodiorite originated from the upper mantle, and assimilated crustal material during its evolution. these data have allowed the development of the following model for the fenghuangshan granodiorite: an upper-mantle-derived magma migrated along faults and was emplaced within the middle–lower crust (at a depth of ~15 km), forming a deep magma chamber where the magma underwent assimilation and fractionation crystallization processes. the region then underwent tectonic deformation, causing the magma within the deep magma chamber to migrate along tectonic fractures to the shallow crust (depths of 3–6 km), forming the fenghuangshan granodiorite. 6.2 estimate of denudation amount ore deposits undergo significant changes after formation, and post-formational changes and preservation of orebodies provide vital information for targeting during mineral exploration (harrison, 1994; kesler et al., 2006; uchida et al., 2007; wilkinson et al., 2007; wang et al., 2008; kirstein, 2011; lu et al., 2011; zheng et al., 2011). the degree of denudation of an ore deposit is an important factor in determining the preservation potential of orebodies and hence the prospectivity of an area. the copper mineralization in the present study area is closely related to the fenghuangshan granodiorite, indicating that the denudation of the deposit can be inferred from the emplacement depth of the intrusion, the timing of mineralization in the area, and the thickness of overlying sediments. the samples analyzed in this study were obtained from the –440 m level of the fenghuangshan deposit. this, combined with the shallowest emplacement depth obtained during this study (3.23 km), suggests that this area has experienced at least ~2.8 km of uplift since 141 ma. the intrusion in the study area was emplaced into sediments at 141 ma; the total thickness of all of the sediments overlying the intrusion is >2.6 km (321 geological team, 1989). this indicates that all of these sediments must have been eroded away, although the intrusion itself has not undergone significant erosion. as such, the fenghuangshan copper ore deposit, which is closely related to the fenghuangshan granodiorite, is well preserved and has undergone few postmineralization changes. this suggests that the emplacement depth (3.32–6.32 km) of the intrusion could also represent the depth of exhumation of the copper mineralization in this area. therefore, exploration for copper mineralization at depths of <1 km, which is the current limit of exploration and mining in this area, may identify significant amounts of previously unknown mineralization . 7. conclusions analyses of minerals from the fenghuangshan granodiorite yielded the following conclusions. amphibole poikilocrystals contain lower amounts of si than other amphiboles in the intrusion and are classified as tschermakite, whereas matrix amphibole in the intrusion is magnesiohornblende. plagioclase within the deposit has oligoclase and andesine compositions. amphibole poikilocrystals crystallized from the melt at pressures of 4.46– 4.74 kbar and temperatures of 1066°c–1071°c, whereas matrix amphibole formed at pressures of 1.00–1.91 kbar and 784°c–823°c, representing conditions within deep and shallow magma chambers, respectively. the fenghuangshan granodiorite formed from upper-mantlederived magmas that initially migrated along fault zones before being emplaced in the middle–lower crust (at a depth of ~15 km), forming a deep magma chamber where these magmas underwent fractionation crystallization and assimilation. tectonism then caused the magma within the deep magma chamber to migrate along faults to the shallow crust (depths of 3–6 km), forming the fenghuangshan granodiorite. the barometry and thermometry undertaken during this study, combined with the stratigraphy of this region, indicates that the fenghuangshan copper deposit is well preserved and has undergone few changes since formation. this indicates that the emplacement depth of the intrusion (3.32–6.32 km) also represents the initial burial depth of the copper deposit. the current depth of exploration and mining in this area is <1 km, meaning that this area is highly prospective for deeper exploration for copper mineralization similar to that within the fenghuangshan copper deposit. acknowledgements this study was supported financially by the china geological survey (grant 12120113069900) and the national natural science foundation of china (grant 40672045). senior engineers chenghuo zhang and shuping tang provided valuable assistance in the field. we are grateful to huiyan zhu and li liu of china universitiy of geosciences, beijing, and jingqiu ren of peking universitiy,for their kind help during this work. we thank ruiping liu, qiang fu, and two anonymous reviewers for their critical and constructive reviews, which greatly improved the paper. 36 aiping zhang, yangsong du, yi cao, zhenshan pang , gan luo , kangkang xu references ague, j.j.; brandon, m.t.(1992). tilt and northward offset of cordilleran batholiths resolved using igneous barometry. nature, 360(6400), 146-149. anderson, j.r.(1980). significance of hornblende in calc-alkaline andesites and basalts. american mineralogist, 65, 837-851. blundy, j.d. and holland, t.j.(1990). calcic amphibole equilibria and a new amphibole-plagioclase geothermometer. contributions to mineralogy and petrology, 111(2), 269-272. cao, y.; du, y.s.; cai c.l.; qin, x.l.; li, s.t.; xiang w.s.(2008). mesozoic a-type granitoids and xenoliths in the lujiang-zongyang area, anhui province: significance in post-collisional magmatic evolution. geological journal of china universities, 14(4):565-576. cao, y.; du, y.s.; gao, f.p.; hu, l.f.; xin, f.p.; pang, z.s.(2010). origin and evolution of hydrothermal fluids in the taochong iron deposit, middle–lower yangtze valley, eastern china: evidence from microthermometric and stable isotope analyses of fluid inclusions. ore geology review, 48:225-238. chang, y.f.; liu, x.p.; wu, y.c.( 1991). the copper-iron belt of the lower and middle reaches of the changjiang river. geological publishing house, beijing, 379pp. davidson, j.; turner, s.; handley, h.; macpherson, c.; dosseto, a.( 2007) amphibole“sponge” in arc crust. geology, 35(9), 787-790. dessimoz, m.; müntener, o.; ulmer, p.( 2012). a case for hornblende dominated fractionation of arc magmas: the chelan complex (washington cascades). contributions to mineralogy and petrology,163(4), 567-589. droop, g.t.( 1987). a general equation for estimating fe3+ concentrations in ferromagnesian silicates and oxides from microprobe analyses, using stoichiometric criteria. mineralogical magazine, 51(361), 431-435. du, y.s.; lee, h.; qin, x.l.( 2004). underplating of mesozoic mantlederived magmas in tongling, anhui province: evidence from megacrysts and xenoliths. acta geologica sinica, 78(1), 131-136. du, y.s.; li, s.t.; cao, y.; qin, x.l; lou, y.e.( 2007). uafc-related origin of the late jurassic to early cretaceous intrusions in the tongguanshan ore field, tongling, anhui province, east china. geoscience, 21(1), 71-77. ernst, w.g.( 1999). hornblende, the continental maker-evolution of h2o during circum-pacific subduction versus continental collision. geology, 27(8), 675-678. ghent, e.d.; nicholls, j.; simony, p.s.; sevigny, j.h.; stout, m.z.( 1991). hornblende geobarometry of the nelson-batholith, southeastern british-columbia-tectonic implications. canadian journal of earth sciences, 28(12), 1982-1991. hammarstrom, j.m. and zen, e.(1986). aluminum in hornblende: an empirical igneous geobarometer. american mineralogist, 11-12, 1297-1313. harrision, c.g.(1994). rates of continental and mountain building. geologische rundschau, 83(2), 431447. hollister, l.s.; grissom, g.c.; peters, e.k.; stowell, h.h. and sisson, v.b.(1987). confirmation of the empirical correlation of al in hornblende with pressure of solidification of calcalkaline plutons. american mineralogist, 9-10, 231-239. johnson, m.c. and rutherford, m.j.( 1989). experimental calibration of the aluminum-in-hornblende geobarometer with application to long valley caldera (california) volcanic rocks. geology, 17(9), 837-841. kesler, s.e. and wilkinson, b.h.( 2006). the role of exhumation in the temporal distribution of ore deposits. economic geology, 101(5), 919-922. kirstein, l.a.(2011). thermal evolution and exhumation of the ladakh batholith, northwest himalaya, india. tectonophysics, 3-4, 222-233. krawczynski, m.j.; grove, t.l.; behren, h.(2012). amphibole stability in primitive arc magmas: effects of temperature, h2o content, and oxygen fugacity. contributions to mineralogy and petrology, 164(2), 317-339. larocque, j.; canil, d.( 2010). the role of amphibole in the evolution of arc magmas and crust: the case from the jurassic bonanza arc section, vancouver island, canada. mineralogy and petrology, 159(4), 475-492. leake, b.e.; woolley, a.r.; arps, c.e.; birch, w.d.; gilbert, m.c.; grice, j.d.; hawthorne, f.c.; kato, a.; kisch, h.j.; krivovichev, v.g.; linthout, k.; laird, j.; mandaring, j.a.; maresch, w.v.; nichel, e.h.; rock, n.m.; schumacher, j.c.; simth, d.c.; stephenson, n.c.; ungaretti, l.; whittaker, e.j. and youzhi, g.(1997). nomenclature of amphiboles:report of the subcommittee on amphiboles of the international mineralogical association, commission on new minerals and mineralnames. american mineralogist, 35,(1), 219-246. li, w.d.(1989). on the yangtze type copper ore deposits and its origin. nangjing institute geology and mineral resource, 10(2), 1-14. li, s.; yang, x.y.; huang, y.; sun, w.d.( 2014). petrogenesis and mineralization of the fenghuangshan skarn cu-au deposit, tongling ore cluster field, lower yangtze metallogenic belt. ore geology reviews, 58, 148-162. liu, m.l.; peng, s.l.(2003). prediction of hidden ore bodies by synthesis of geological, geophysical and geochemical information based on dynamic model in fenghuangshan ore field, tongling district, china. journal of geochemical exploration, 1-3, 81-98. lu, l.n.; fan, h.r.; hu, f.f.; yang, k.f.;lan, t.g.( 2011). emplacement depth of the guojialing granodiorites from the northwestern jiaodong peninsula, eastern china: evidences from hornblende thermobarometry and fluid inclusions. acta petrologica sinica, 27(5), 1521-1532. mao, j.w.; shao, y.j.; xie, g.q.; zhang, j.d.; chen, y.c.( 2009). mineral deposit model for porphyry-skarn polymetallic copper deposits in tongling ore dense district of middle-lower yangtze valley metallogenic belt. mineral deposits, 28(2), 109-119. mao, j.w.; chen, m.h.; yuan, s.d.; guo, c.l.( 2011). geological characteristics of the qinhang(or shihang)metal logenic belt in south china and spatial-temporal distribution regularity of mineral deposits. acta geologica sinica, 85(5), 636-658. mao, z.l.; yang, b.; lai, j.q.(2004). the characteristics of breccias and analysis of origin near ore bodies in fenghuangshan copper deposits, tongling. journal of mineralogy and petrology, 24(1), 29-32. martin, r.f.( 2007). amphiboles in the igneous environment, in: hawthorne, f.c.; oberti, r. amphiboles: crystal chemistry, occurrence and human health: reviews in mineralogy and geochemistry, 67, 323-358. pan, y.m.; dong, p.(1999) the lower changjiang (yangzi/yangtze river) metallogenic belt, east central china: intrusion-and wall rock-hosted cu-fe-au, mo, zn, pb, ag deposits. ore geology reviews, 15(4), 177–242. qu, h.y.; pei, r.f.; li, j.w.; wang, y.l.; deng, y.j.( 2010). geochemical characteristics of the fenghuangshan yanshanian intermediateacid intrusive rocks in tongling, anhui province, and their relationship with the metallogenic process. geology in china, 37(2), 311-323. ridolfi, f.; renzulli, a.; puerini, m.(2010). stability and chemical equilibrium of amphibole in calc-alkaline magmas: an overview, new thermobarometric formulations and application to subduction-related volcanoes. contributions to mineralogy and petrology, 160(1), 45-66. schmidt, m.w.(1992). amphibole composition in tonalite as a function of pressure:an experimental calibration of the al-in-hornblende barometer. contributions to mineralogy and petrology, 2-3, 304-310. shao, y.j.; peng, s.l.; lai, j.q.; wu, g.g.(2003). analysis of the dynamic mechanism of mineral formation of the fenghuangshan copper ore deposit. mineral resources and geology, 17(98), 508-510. smith, j.v.; brown, w. l.(1974). feldspar minerals. springer-verlag, 630pp. stein, e. and dietl, c.(2001). hornblende thermobarometry of granitoids from the central odenwald(germany) and their implications for the geotectonic development of the odenwald. mineralogy and petrology, 1-3, 185-207. tang, y.c.; wu, y.c.; chu, g.z.; xing, f.m.; wang, y.m.; cao, f.y.; chang, y.f.( 1998). geology of copper-gold polymetallic deposits along the yangtz river, anhui province. geological publishing house, beijing, 351pp. 37petrology and mineralogy of the fenghuanshan granodiorite, anhui province, china: implications for petrogenesis and metallogenesis tulloch, a.j. and challis, g.a.(2000). emplacement depths of paleozoic mesozoic plutons from western new zealand estimated by hornblende-al geobarometry. new zealand journal of geology and geophysics, 43(4), 555-567. uchida, e.; endo, s.; makino, m.(2007). relationship between solidification depth of granitic rocks and formation of hydrothermal ore deposits. resource geology, 5(5), 47-56. vyhnal, c.r.; mcsween, h.y.; speer, j.a.(1991). hornblende chemistry in southern appalachian granitoids-implications for aluminum hornblende thermobarometry and magmatic epidote stability. american mineralogist, 1-2, 176-188. wang, j.p.; zhai, y.s.; liu, j.j.; liu, z.j.; liu, j.( 2008). a new approach to post-ore change and preservation of ore deposits: fission track analysis. advances in earth science, 23(4), 421-427. wang, x.j.; peng, s.l.; lai, j.q.; yang, b.; shao, y.j.( 2008). relations between granite and mineral ization in the fenghuangshan copperpolymetallic deposit at tongling, anhui province. geology and prospecting, 44(6), 49-55. wilkinson, b.h. and kesler, s.e.(2007). tectonism and exhumation in convergent margin orogens: insights from ore deposits. the journal of geology, 115(6), 611-627. wu, c.l.; dong, s.w.; paul, t. r.; frost, b.r.; gao, y.l.; lei, m.; chen, q.l.; qin h.p.(2014) petrogenesis of high-k, calcalkaline and shoshonitic intrusive rocks in the tongling area, anhui province(eastern china), and their tectonic implications. geological society of america bulletin, 126, 78-10. xing, f.m.; xu, x.(1996).high-potassium calc-alkaline intrusive rocks in tongling area, anhui province. geochimica, 25(1), 29-38. yasser, a.e.; hassan, m. h.; tomoyuki, s.; masako, y.; shoji, a.; akihiro, t.(2012). mineral chemistry of the neoproterozoic alaskantype akarem intrusion with special emphasis on amphibole: implications for the pluton origin and evolution of subductionrelated magma. lithos, 155, 410-425. zen, e.a.(1989). plumbing the depths of batholiths. american journal, 289(10), 1137-1159. zhai, y.s.; yao, s.z.; lin, x.d.; zhou, x.n.; wan, t.f.; jin, f.q.; zhou, z.g.( 1992). fe-cu (au) metallogeny of the middle-lower changjiang region. geological publishing house, beijing, 235pp. zheng, y.c.; gu, l.x.; tang, x.q.; liu, s.h.; wu, c.z.; li c.h.(2011). oxygen isotope characteristics of the footwall alteration zones in the hongtoushan volcanogenic massive sulfide deposit, liaoning province, china and restoration of their formation temperatures. acta geologica sinica-english edition. 85(3): 683-693. articulo 8c.indd earth sciences research journal geochemistry earth sci. res. sj. vol. 16, no. 2 (december, 2012): 151 164 the geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria jude e. ogala geology department, delta state university, p.m.b.1, abraka, nigeria. e-mail: etunimogala@yahoo.com abstract major and trace element compositions of lignite from the tertiary ogwashi-asaba formation, southern nigeria, have been investigated to determine the prevailing environmental conditions which controlled their formation. seven samples were obtained from outcrops along river valleys, streams and springs in seven localities; these samples were subsequently analysed using fusion combined with inductively-coupled plasma (fus-icp) and total digestion with inductively-coupled plasma (td-icp). geochemical analysis revealed the following concentration range in weight percent (wt. %) for major oxides: sio 2 (0.04-9.78); al2o3 (0.56-6.40), fe2o3 (0.05-1.22), mno (0.001-0.004), mgo (0.02-0.11), cao (0.05-0.15), k2o (0.01 0.04), tio2 (0.016-0.299) and s (0.08-0.39). trace elements indicated the following concentration range in parts per million (ppm): be (2-5), zr (4-63), sr (5-22), y (13-68), ba (9-89), v (4-216), zn (17176), ni (8-28), co (5-13), cr (2-31), cu (1-22) and ga (1-14). the low silica (<10 wt. %) and alumina contents (<7 wt. %) were explained by very limited detrital input during coal formation. magnesium oxide and cao content were relatively low thereby confirming the continental nature of the peat-forming environment. redox-sensitive trace element ratios (ni/co, v/cr and v/v+ni) indicated predominantly oxic environments for coal deposition. the low trace element concentrations determined in the lignite samples did not point to any severe environmental impact from coal use. resumen las composiciones de elementos mayores y trazas de lignita de la formación terciaria ogwashi-asaba, en el sur de nigeria, han sido investigadas para determinar las condiciones ambientales prevalecientes que controlaron su formación. siete muestras fueron tomadas de afloramientos a lo largo de valles de ríos, arroyos y manantiales en siete localidades; estas muestras se analizaron posteriormente mediante fusión combinada (fus-icp) y la digestión total con plasma de acoplamiento inductivo (icp-td). el análisis geoquímico reveló el siguiente intervalo de concentración en porcentaje en peso (% en peso) para los óxidos mayores: sio 2 (0,04 a 9,78); al2o3 (0,56-6,40), fe2o3 (0,05-1,22), mno (0.001-0.004), mgo (0,02 -0,11), cao (0.05-0.15), k2o (0,01 0,04), tio2 (0.016-0,299) y s (0,08 a 0,39). los elementos traza indican el siguiente rango de concentración en partes por millón (ppm): be (2-5), zr (4-63), sr (5-22), y (13-68), ba (9-89), v (4-216), zn (17-176), ni (8-28), co (5-13), cr (2-31), cu (1-22) y ga (1-14). el bajo contenido de sílice (<10 en peso.%) y el contenido de aluminio (<7 wt.%) fueron explicados por la entrada muy limitado de detrítos durante la formación de carbón. el óxido de magnesio y el contenido de cao son relativamente bajos, lo que confirma el carácter continental del entorno de formación de turba. la relación de elementos traza redox-sensibles (ni/co, v/ cr y v/v + ni) indicó ambientes predominantemente óxicos para la deposición de carbón. las bajas concentraciones de trazas de elementos hallados en las muestras de lignito no indican un impacto ambiental severo de la utilización del carbón. palabras claves: geoquímica, lignito, formación ogwashiasaba, ambientte continental, nigeria keywords: geochemistry, lignite, ogwashi-asaba formation, continental environment, nigeria. record manuscript received: 12/01/2012 accepted for publications: 06/05/2012 introduction coal accumulation in nigeria took place mainly during upper cretaceous and tertiary times, resulting in the formation of extensive lignite, sub-bituminous and bituminous coal deposits in the benue trough and anambra basin (simpson, 1954; reyment, 1965; akande et al., 1992; obaje and ligouis, 1996; obaje and hamza, 2000; obaje, 2009). the ogwashi-asaba formation occurs extensively within the niger delta basin jude e. ogala152 in southern nigeria (figure 1); it was first described by wilson (1925) and named the “lignite group.” lithologically, it consists of a sequence of coarse-grained sandstones, light-coloured clays and carbonaceous shales, within which are continental lignite seam intercalations (reyment, 1965; kogbe, 1976; jan du chene et al., 1978). nigeria has the largest known lignite deposits in africa, proven reserves exceeding 300 million tons (orajaka et al., 1990; momsd, 2007). lignite is mined in open pits in nigeria and is used mainly as a substitute for wood as fuel in domestic cooking. the pertinent literature on nigerian lignite deposits’ major and trace element composition is scarce (adedosu et al., 2007; olobaniyi and ogala, 2011). coal is heterogeneous sediment having both organic and inorganic fractions; it contains variable amounts of almost all the elements included in the periodic table (christanis et al., 1998; orem and finkelman, 2003). most elements contained in coal usually show a close association with organic matter or with the coal’s inorganic fraction. changes in the elements’ affinity with either the organic or inorganic fraction may also occur during coalification (christanis et al., 1998). several researchers have attempted to study the geochemical features of trace elements contained in coal to understand and evaluate trace elements’ mode of occurrence as well as their behaviour during combustion (gluskoter et al., 1977; swaine, 1990; finkelman, 1994; meij, 1995; spears and zheng, 1999; davidson, 2000; vassilev et al., 2001). previous studies on trace elements in nigerian coal have revealed their distribution (olajire et al., 2007; ogala et al., 2010a), composition (ndiokwere et al., 1983; olabanji, 1991; ewa et al., 1996; adedosu et al., 2007; ogala et al., 2010b), characteristics (nwadinigwe, 1992; sonibare et al., 2005) and association ( ewa, 2004; ogala et al., 2009). several researchers have studied trace element content in lignite and their behaviour during combustion (foscolos et al., 1989; goodarzi et al., figure 1. map of nigeria showing the location of ogwashi-asaba formation and adjacent units. the geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria 153 1990; georgakopoulos et al., 1994; filippidis et al., 1996; gentzis et al., 1990; gerouki et al., 1996; sakorafa et al., 1996; christanis et al., 1998; chatziapostolou et al., 2006; adamidou et al., 2007). trace elements are of great importance in ascertaining the depositional environment and maturity of geological samples within a particular basin (adedosu et al., 2007). the present study was aimed at investigating the chemical composition (major and trace elements) of lignite samples from the ogwashiasaba formation, southern nigeria, to predict the prevailing geochemical environmental conditions as well as the processes responsible for their distribution. the study area’s geology and stratigraphy the ogwashi-asaba formation (reyment, 1965) occurs extensively within the niger delta basin in southern nigeria, covering a 4,900 km2 area (fig. 1). the formation was originally referred to as a “lignite group” (wilson, 1925; wilson and bain, 1928), “lignite series” (simpson, 1949; 1954), and “lignite formation” (de swardt and casey, 1963). lignite layers have also been encountered in the lowermost strata of the ameki group (fig. 1) and the uppermost strata of the benin formation in drill holes, streams/river banks and road-cut outcrops (okezie and onuogu, 1985). the southern nigerian sedimentary basin’s formation began during the early cretaceous period (albian) following basement subsidence along the benue and niger troughs (nwachukwu, 1972; olade, 1975). folding and uplift occurred during the santonian along a northeast-southwest axis in the abakaliki-benue area. the anambra platform, lying to the west and southwest of the abakaliki folded belts, subsided to form the anambra basin (reyment, 1965; short and stauble, 1967; murat, 1972; benkhelil, 1989). the upper cretaceous stratigraphic succession in the anambra basin began with the deposition of sediment from the marine campanian-maastrichtian enugu/nkporo shales and their lateral equivalent deltaic owelli sandstone (fig. 2). these base units were successively overlain by the lower-middle maastrichtian deltaic coal-bearing mamu formation (lower coal measures) and middle maastrichtian tidal ajali sandstone (false-bedded sandstones) and overlain by the fluvial-deltaic nsukka formation (upper coal measures) . the imo and nsukka formations marine shales were deposited during the paleocene era and overlain by the regressive ameki formation (eocene); the paralic ogwashi-asaba formation (oligocenemiocene) was capped by the continental benin formation (miocene-recent) constituting the tertiary succession (figures 1 and 2). the ogwashiasaba formation consists of a succession of coarse-grained sandstone, clay and carbonaceous shale, containing continental lignite seam intercalations (kogbe, 1976; jan du chene et al., 1978). reyment (1965) suggested an oligocene-miocene age for the ogwashi-asaba formation, but palynological study by jan du chene et al. (1978) proposed a late eocene age for the base part. the lignite seams found within the ogwashi-asaba formation are usually brownish to black, varying in thickness from a few millimetres to a maximum of 6 meters. they are thinly laminated and fissile, having leaf and woody fragments on fresh cleats. the average overburden to figure 2. shows the stratigraphy of three basins in southern nigeriabenue trough, anambra basin and niger delta basin (modified from reyment, 1965). age formation lithology depositional environment basin quaternary benin formation ogwashi asaba formation ameki f ormation/nanka sand imo formation nsukka formation sandstones, clay, shales clay, shales, sandstones, lignite sandstones, cl ay, shales, limestone clay, shales, limestone, sandstone, marl sandstones, clays, shales, coals, limestone continental continental estuarine, shallow marine shallow marine, deltaic fluvio deltaic n ig er d el ta b as in t er ti ar y pliocene miocene oligocene eocene paleocene u p p er c re ta ce o u s maastrichtian ajali formation sandstones, claystones fluvio – deltaic mamu formation enugu/nkporo/owelli formation sandstones, clays, coals shales, sandstones, clay, ironstones, siltstones. shallow marine, deltaic shallow marine, deltaic campanian major unconformity santonian awgu formation sandstones, limestones, clays, coals, siltstones shallow marine, deltaic conician m id d le c re t. turonian eze -aku formation shales, limestones, sandstones shallow marine cenomanian odukpani formation sandstones, limestones, shales shallow marine lo w er c re t. albian asu river group shales, limestones, sandstones shal low marine aptian major unconformity basement complex granites, gneisses, schists, migmatites igneous, metamorphic a n am b ra b as in b en u e t ro u gh jude e. ogala154 and okpuzu rivers in ibusa and at the mgbala and ekete springs in okpanam and nnewi, respectively. lignite seams are also exposed along river valleys at the nkpunkpa and iyiodo springs as well as nnem-agadi river, all occurring in obomkpa. the lignite beds are brownish to black and vary in thickness from a few millimetres to a maximum of 2 meters. they are thinly laminated and fissile with leaf and woody fragments on fresh cleats. the lithological sequence of lignite seam outcropping along the nnem-agadi river was logged (figure 3); seven lignite samples were collected: three samples from obomkpa, two from ibusa and one each from okpanam and nnewi. analytical procedure samples (0.25 g) were powdered in an agate mortar and determined by a combination of methods to investigate lignite chemical composition (major and trace elements), i.e. fusion with inductively-coupled plasma (fus-icp) and total digestion with inductively-coupled plasma (tdicp). loss on ignition (loi) caused by escaping volatiles was measured after heating to about 1,050oc. limits of detectable measurement (ldm) for major elements were 0.01 for sio2, al2o3, fe2o3(t), mgo, cao, na2o, k2o and p2o5, 0.001 for mno and tio2 and 0.01% for s. ldm for trace elements were 2 ppm for ba, sr, zr, bi, te, 1 ppm for y, sc, be, co, cr, cu, ga, hg, ni and zn, 5 ppm for v, sb, tl and w, 0.3 ppm for ag and cd, 10 ppm for u, 3 ppm for as and pb and 4 ppm for sc coal data was calculated on an ash basis (value in ash = element value in coal × 100/ash yield, where ash yield = 100 – loi). major oxides and selected trace elements were analysed by the fusion technique (table 1). samples were mixed with a flux of lithium metaborate and lithium tetraborate and fused in an induction furnace. the molten melt was immediately poured into a 5% nitric acid solution containing an internal standard and mixed continuously until becoming completely dissolved (c. 30 minutes). the samples were run for major oxides and selected trace elements using a combination fusion technique (fus-icp) followed by inductively-coupled plasma-optical emission spectrometry (icp-oes) analysis using simultaneous/sequential thermo jarrell-ash enviro ii icp. a 0.25 g sample was digested with four acids for trace element analysis (table 2), beginning with hydrofluoric acid (hf), followed by a mixture of nitric (hno 3) and perchloric acid (hclo4), heated using precise figure 3. lithologic column of outcrop along nnem-agadi river at obomkpa. lignite ratio is 6:1, thereby appearing to rule out open-cast mining (da swardt and piper, 1957). ogwashi-asaba lignite outcrops along river valleys and streams/ springs in ibusa (okpunzu and mgbiligba river), okpanam (mgbala spring), obomkpa (nnem-agadi river, iyiodo and nkpunkpa springs) and nnewi (ekete spring) (fig. 1). analysis methods field work and sampling field work for the present study took place in outcrops along river valleys, streams and springs in seven localities (fig. 1). lignite seams do not easily outcrop in the field but good exposure was found along the mgbiligba element detection limit analysis method location 1 location 2 location 3 location 4 location 5 location 6 location 7 sio 2 0.01 fus-icp 0.39 0.04 3.94 1.04 9.38 9.78 2.99 al 2 o 3 0.01 fus-icp 0.56 0.65 2.47 1.31 6.4 2.13 2.7 fe 2 o 3 t 0.01 fus-icp 0.05 1.22 0.13 0.21 0.42 0.38 0.17 mno 0.001 fus-icp 0.001 0.004 0.002 0.004 0.002 0.004 0.001 mgo 0.01 fus-icp 0.02 0.02 0.04 0.04 0.04 0.11 0.02 cao 0.01 fus-icp 0.08 0.08 0.09 0.15 0.05 0.43 0.05 na 2 o 0.01 fus-icp < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 < 0.01 k 2 o 0.01 fus-icp < 0.01 < 0.01 0.02 0.01 0.04 0.03 0.02 tio 2 0.001 fus-icp 0.016 0.017 0.095 0.025 0.299 0.07 0.095 p 2 o5 0.01 fus-icp < 0.01 < 0.01 < 0.01 < 0.01 0.02 < 0.01 < 0.01 loi fus-icp 97.45 96.84 91.02 95.92 82.8 87 92.8 total 0.01 fus-icp 98.6 98.88 97.82 98.71 99.45 99.94 98.86 ash yield fus-icp 2.55 3.16 8.98 4.08 17.2 13 7.2 table 1. major element concentration in weight percent (wt. %) for lignites of the ogwashi-asaba formation. the geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria 155 programmer-controlled heating in several ramping and holding cycles thus drying the samples. after dryness was attained, samples were brought back into solution using hydrochloric acid. the sample solution was then analysed for element concentration using varian vista pro inductively-coupled plasma-optical emission spectrometry (icp-oes). usgs and canmet certified reference materials were used for calibration. the activation laboratories in ancaster, canada, performed the chemical analysis. statistical analysis the elements were subjected to univariate (minimum, maximum, mean and standard deviation, correlation coefficient) and multivariate statistical analysis (factor and principal component) using statistical package for social sciences (spss) software (version 16.0). since most parameters measured here were not normally distributed, spearman’s rank correlation was used to examine the correlation between elements (the correlation coefficient matrix measures how well each constituent’s variance can be explained by its relationship to the others). factor analysis (fa)/principal component analysis (pca) was applied to the experimental data standardised through z-scale transformation to avoid misclassification due to differences in the units of measurement. standardisation tends to increase the influence of variables whose variance is small and reduce the influence of variables whose variance is large (liu et al., 2003). factor analysis was applied to the data set to study the association of the trace elements and table 2. trace elements and sulphur concentration for lignites of the ogwashi-asaba formation (values are in parts per million except for s which is in weight percent); 1: low-rank coal world average after seredin and finkelman (2008). elements detection limit analysis method location 1 location 2 location 3 location 4 location 5 location 6 location 7 lrcwa1 ba 2 fus-icp 17 44 54 27 60 93 23 150.0 sr 2 fus-icp 5 6 12 7 22 20 10 110.0 y 1 fus-icp 34 68 57 42 33 18 13 8.6 sc 1 fus-icp < 1 < 1 4 2 7 6 4 4.1 zr 2 fus-icp 4 6 31 10 63 32 24 36.0 be 1 fus-icp 2 4 3 4 5 3 3 1.6 v 5 fus-icp < 5 < 5 47 23 197 45 33 25.0 ag 0.3 td-icp < 0.3 < 0.3 < 0.3 < 0.3 < 0.3 < 0.3 < 0.3 0.1 as 3 td-icp < 3 < 3 < 3 < 3 < 3 < 3 < 3 8.3 bi 2 td-icp < 2 < 2 < 2 < 2 < 2 < 2 < 2 cd 0.3 td-icp 0.4 < 0.3 < 0.3 0.3 < 0.3 < 0.3 < 0.3 0.2 co 1 td-icp 7 5 13 8 9 6 6 5.1 cr 1 td-icp 3 2 12 6 31 16 15 16.0 cu 1 td-icp 1 1 21 3 22 12 5 16.0 ga 1 td-icp 2 1 5 2 14 2 5 5.8 hg 1 td-icp < 1 < 1 < 1 < 1 < 1 < 1 < 1 0.10 mo 1 td-icp < 1 < 1 1 < 1 2 < 1 < 1 2.2 ni 1 td-icp 17 8 28 22 21 17 10 13.0 pb 3 td-icp < 3 < 3 5 < 3 8 3 < 3 7.8 sb 5 td-icp < 5 < 5 < 5 < 5 < 5 < 5 < 5 0.84 s 0.01 td-icp 0.12 0.19 0.14 0.16 0.08 0.34 0.11 te 2 td-icp < 2 < 2 < 2 < 2 < 2 < 2 < 2 tl 5 td-icp < 5 < 5 < 5 < 5 < 5 < 5 < 5 0.68 u 10 td-icp < 10 < 10 < 10 < 10 < 10 < 10 < 10 2.4 w 5 td-icp < 5 < 5 < 5 < 5 < 5 < 5 < 5 1.1 zn 1 td-icp 176 52 105 54 29 36 17 23.0 v/(v+ni) 0.23 0.38 0.63 0.51 0.9 0.73 0.77 ni/co 2.43 1.6 2.15 2.75 2.33 2.83 1.67 v/cr 1.66 2.5 3.92 3.83 6.36 2.81 2.2 jude e. ogala156 extract the principal factors governing trace element distribution (lu et al., 1995). components having > 1 eigenvalue were selected to explain association amongst the measured variables. cluster analysis (ca) established site segregation and desegregation; such analysis is an unsupervised pattern recognition technique which uncovers a data set’s intrinsic structure or underlying pattern without previous knowledge concerning the data being available. this is to enable classification being made based on measured objects’ nearness or similarity, thereby helping to establish relationships among sites in the form of a dendrogram. hierarchical agglomerative ca involved normalising the data set by means of ward’s method (i.e. using euclidean distances as a measurement of similarity). ca was applied to the data set to group similar sampling sites (spatial variation) spread over the region. results and discussion the chemical composition of lignite the chemical composition of the lignite samples is presented in tables 1 and 2. thirty-seven parameters were measured, consisting of ten major elements, loss on ignition (loi) and 26 trace elements. sio 2 (0.04-9.78%), al2o3 (0.56-6.40%) and fe2o3 (0.05-1.22%) were the dominant oxides in most lignite samples; mgo, mno, cao, tio2, na2o, k2o and p2o5 all occurred in traces. loi had high values (82.8097.45%). the low to moderate sio2, al2o3 and fe2o3 concentration was due to quartz, feldspar and pyrite constituting the lignite samples’ main mineral phase (adamidou et al., 2007; kalaitzidis et al., 2010). the low sio 2 (< 10%) and al2o3 content (< 7%) suggested very limited detrital input during coal formation. the relatively low cao (0.05-0.43%) and mgo content (0.02-0.11%) implied a continental origin for the lignite (olobaniyi and ogala, 2011); this agreed with the ogwashi-asaba formation’s continental setting (reyment, 1965; kogbe, 1976; jan du chene et al., 1978). the most abundant trace element was zn (average 67 ppm), followed by v, ba, y and zr (50.71 ppm, 45.43 ppm, 37.86 ppm and 24.29 ppm average values, respectively) (tables 2 and 6). the average mean values for sr, sc, be, cd, co, cr, cu, ga, mo, ni and pb ranged from 0.2 ppm to 18 ppm, ag, as, bi, hg, sb, te, tl, u and w concentration being below the detection limit in all samples. compared to average worldwide coal concentration (seredin and finkelman, 2008), most trace element content was lower than world average values for low-ranking coal due to low ash yield (table 2). y, be, v, co and zn were the only trace elements having relatively higher concentrations in the coal samples used here than the world coal average. redox-sensitive trace element ratios (ni/co, v/cr and v/v+ni) are usually considered powerful geochemical indicators for environmental disv /c r 5 8 6 4 2 0 0 5 10 15 20 ni/co v /( v + n i) 1 0,9 0,8 0,7 0,6 0,5 0 5 10 15 20 ni/co oxic oxicdysoxic dysoxic su bo xi c/ a no xi c suboxic/anoxic suboxic/anoxic d ys ox ic o xi c eu xi ni c a no xi c d ys ox ic (a) (b) figure 4. cross plots of redox-sensitive trace metal ratios (a) v/cr vs. ni/co; (b) v/(v+ni) vs. ni/co. ranges for v/cr and ni/co are from jones and manning (1994); ranges for v/(v+ni) are from hatch and leventhal (1992). element n=7 range minimum maximum mean standard deviation sio 2 9.74 .04 9.78 3.937 4.099 al 2 o 3 5.84 .56 6.40 2.317 1.990 fe 2 o 3 t 1.17 .05 1.22 .369 .398 mno .003 .001 .004 .003 .001 mgo .09 .02 .11 .041 .032 cao .38 .05 .43 .133 .135 na 2 o .000 < 0.01 < 0.01 < 0.01 .000 k 2 o .031 < 0.01 .040 .020 .012 loi 14.65 82.80 97.45 91.976 5.473 table 3. compositional average and standard deviation of major oxides for lignites of the ogwashi-asaba formation. the geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria 157 crimination (lewan, 1984; hatch and leventhal, 1992; jones and manning, 1994; hoffman et al., 1998; rimmer, 2004; rimmer et al., 2004; algeo and maynard, 2004; johnson et al., 2010; saez et al., 2011). jones and manning (1994) suggested that < 5 ni/co ratios inferred oxic conditions, 5-7 dysoxic conditions and > 7 suboxic to anoxic conditions. they also used < 2 v/cr ratios to infer oxic conditions, 2-4.25 for dysoxic conditions and > 4.25 for suboxic to anoxic conditions. lewan (1984) showed that v/v+ni should be greater than 0.5 for organic matter accumulated in euxinic conditions. hatch and leventhal (1992) compared v/v+ni ratios to other geochemical redox indicators, including degree of pyritisation, and suggested that ratios greater than 0.84 showed euxinic conditions, 0.54-0.82 anoxic water and 0.46-0.60 for dysoxic conditions. according to calvert and pedersen (1993), jones and manning (1994), hoffman et al (1998) and algeo and maynard (2004) anoxic environments are characterised by high v/cr ratios and v/v+ni values lying between 0.5 and 0.9 because of the disparate behaviour of v, ni and cr during redox processes in marine environments characterised by fine-grained detritic sedimentation. all the samples analysed and shown on the v/cr cf ni/co and v/ v+ni cf ni/co diagram (figure 4) plotted within the oxic ranges for the four redox-sensitive trace element ratios proposed by jones and manning (1994). figure 5. box plots showing the distributions of major elements. c on ce n tr at io n ( % ) 0,020 0,015 0,010 0,005 0 mno p2o5 ly io do _s pr in g c on ce n tr at io n ( % ) oxides 10 8 6 4 2 0 sio2 al2o3 fe2o3t r iv er _m gb ili gb a ly io do _s pr in g c on ce n tr at io n ( % ) oxides 0,5 0,4 0,3 0,2 0,1 0 mgo cao k2o tio2 r iv er _m ne m _a ga di r iv er _m ne m _a ga di ly io do _s pr in g table 4. correlation coefficients between the major elements in lignite samples from ogwashi-asaba formation sio 2 al 2 o 3 fe 2 o 3 t mno mgo cao k2o tio 2 p 2 o 5 ash_yield sio 2 1.000 al 2 o 3 .744 1.000 fe 2 o 3 (t) -.122 -.127 1.000 mno .075 -.261 .577 1.000 mgo .736 .126 -.077 .503 1.000 cao .513 -.166 -.016 .581 .951 1.000 k 2 o .955 .902 -.111 -.099 .505 .240 1.000 tio 2 .717 .993 -.084 -.288 .077 -.214 .885 1.000 p 2 o 5 .585 .905 .057 -.180 -.020 -.270 .755 .937 1.000 ash_yield .963 .889 -.100 -.030 .547 .276 .994 .870 .739 1.000 jude e. ogala158 table 5. unrotated component matrix from loadings of 9 major elements and loi on three significant factors. elements factor 1 factor 2 factor 3 sio 2 0.929 0.341 -0.105 al 2 o 3 0.935 -0.316 0.080 fe2o3t -0.127 0.193 0.932 mno -0.103 0.764 0.544 mgo 0.451 0.868 -0.203 cao 0.175 0.956 -0.204 k 2 o 0.991 0.063 -0.028 tio 2 0.923 -0.361 0.123 p 2 o 5 0.823 -0.386 0.301 loi -0.988 -0.114 0.011 eigen values 5.485 2.801 1.372 % of variance 54.847 28.008 13.716 cumulative % 54.847 82.855 96.571 major element geochemistry-correlation matrix table 3 shows the composition average and standard deviation for the lignite samples from the ogwashi-asaba formation analysed here. the data set was screened to identify outliers (figure 5). the other elements were not normally distributed, except for sio 2, k2o and mno; a non-parametric spearman test was thus used which applies variables’ ranking order in determining the correlation coefficient. table 4 shows major elements’ correlation coefficients. there was a strong positive relationship between mgo, k 2o, tio2 and p2o5 with sio2, al2o3 and ash yield, suggesting a common detrital source for these elements; the strong positive statistical correlation between al 2o3 and tio2 (r= 0.993) was typical of elements which probably came from clastic rocks. factor analysis of the major elements studied here gained statistical information on the grouping and hence the processes responsible for their formation. the variance/covariance and factor loadings of the variables having eigenvalues were computed; factor analysis results are summarized in a factor matrix (table 5). a 3-factor model was adopted, covering 96.6% of total variance. the first factor expresses 54.9% of the total data variance and showed high positive loading for elements sio 2, al2o3, k2o, tio2, and p2o5 and negative loading for loi (table 6). the strong positive correlations between sio 2, al2o3, and tio2 were typical of clay minerals. the loadings in factor 1 reflected sediments believed to have their source probably in the mafic basement complex (oban massif ) on the north-eastern flank of the niger delta basin (edet et al., 2003; elueze et al., 2009). the presence of tio 2 and p2o5 was indicative of basaltic rocks of oceanic environment (ogala et al., 2009); such interpretation is supported by the strong positive correlation between tio 2 and p2o5 (r= 0.937). the second factor expressed 28% of the total variance and showed high positive loadings with mno, mgo and cao. the grouping in factor 2 suggests a lithogeochemical input representative of weathering and composition of carbonate sediments. the close proximity of the limestone bands/nodules in the ameki formation within the niger delta basin in close proximity may be the possible source. the strong positive correlation between cao and mgo c om p on en t 2 component 1 component 3 1,0 0,5 0,0 -0,5 -1,0 mno cao p2o5 sio2 k2o mgo loi ii i iii fe 2o3t al2o3 -1,0 -1,0-0,5 -0,50,0 0,00,5 0,51,0 1,0 figure 6. principal component analysis for the first 3 components for oxides. figure 7. the results of cluster analyses on the major elements contained in the lignite samples from ogwashi-asaba formation. the geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria 159 trace elements trace elements ba bezr pby cdcr ssr gacu cosc scni zn river_okpuzu lyiodo_spring lyiodo_spring lyiodo_spring mgbala_spring river_nnem_agadiriver_okpuzu figure 8. box plots showing the distribution of trace elements. 200 150 100 50 0 12.5 10.0 7.5 5.0 2.5 0.0 c on ce n tr at io n ( p p m ) c on ce n tr at io n ( p p m ) element n=7 range minimum maximum mean s.d. ba 76.00 17.00 93.00 45.43 26.46 sr 17.00 5.00 22.00 11.71 6.80 y 55.00 13.00 68.00 37.86 19.76 sc 6.01 .99 7.00 3.57 2.37 zr 59.00 4.00 63.00 24.29 20.63 be 3.00 2.00 5.00 3.43 .98 v 192.01 < 5 197.00 50.71 66.72 ag .00 < 0.3 < 0.3 < 0.3 .00 as .00 < 3 < 3 < 3 .00 bi .00 < 2 < 2 < 2 .00 cd .11 .29 .40 .31 .04 co 8.00 5.00 13.00 7.71 2.69 cr 29.00 2.00 31.00 12.14 10.02 cu 21.00 1.00 22.00 9.29 9.14 ga 13.00 1.00 14.00 4.43 4.50 hg .00 < 1 < 1 < 1 .00 mo 1.01 < 1 2.00 1.14 .38 ni 20.00 8.00 28.00 17.57 6.95 pb 5.01 < 3 8.00 3.99 1.92 sb .00 < 5 < 5 < 5 .00 s .26 .08 .34 .16 .09 te .00 < 2 < 2 < 2 .00 tl .00 < 5 < 5 < 5 .00 u .00 < 10 < 10 < 10 .00 w .00 < 5 < 5 < 5 .00 zn 159.00 17.00 176.00 67.00 55.71 table 6. compositional average and standard deviation of trace elements for lignites of the ogwashi-asaba formation. jude e. ogala160 table 7. correlation coefficients between trace elements and ash yield in lignite samples from ogwashi-asaba formation. table 8. unrotated component matrix from loadings of 17 trace elements on four significant factors. table 9. varimax rotated component loadings of 17 trace elements on four significant components explaining 93.80% of the total variance. ba sr y sc zr be v cd co cr cu ga mo ni pb s zn ash_ yield ba 1.000 sr .807 1.000 y -.097 -.412 1.000 sc .705 -973 -.503 1.000 zr .568 .918 -.302 .940 1.000 be .243 .423 .246 .381 .539 1.000 v .387 .803 -.195 .798 .940 .678 1.000 cd -.507 -.468 -.079 -.511 -.467 -.629 -.322 1.000 co .093 .204 .294 .266 .398 .054 .332 -.114 1.000 cr .490 .903 -.453 .943 .981 .521 .932 -.432 .274 1.000 cu .602 .795 -.026 .814 .886 .358 .767 -.432 .729 .791 1.000 ga .188 .679 -.198 .722 .901 .596 .963 -.262 .397 .903 .741 1.000 mo .245 .668 -.104 .639 .830 .709 .968 -.186 .220 .831 .620 .939 1.000 ni .209 .297 .139 .311 .377 .056 .338 -.011 .902 .274 .663 .316 .224 1.000 pb .298 .667 .061 .662 .874 .624 .945 .255 .551 .817 .829 .947 .925 .477 1.000 s .293 .242 -.125 .130 -.147 -.196 -.330 -.224 -.350 -.173 -.099 -.539 -.428 -.132 -.465 1.000 zn -.386 -.500 .279 -.546 -.463 -.653 -.384 .863 .278 -.522 -.198 -.317 -.298 .295 -.178 -.196 1.000 ash_ yield .728 .987 -.380 .981 .969 .469 .870 -.475 .291 .952 .851 .780 .741 .338 .761 .091 -.491 1.000 elements factor 1 factor 2 factor 3 factor 4 ba .557 -.536 .536 .114 sr .897 -.323 .218 -.172 y -.216 .395 .060 .817 sc .909 -.288 .182 -.219 zr .990 -.026 .028 -.105 be .636 -.089 -.453 .548 v .957 .130 -.197 -.075 cd -.508 .498 .007 -.596 co .421 .660 .511 .217 cr .963 -.091 -.087 -.224 cu .882 .185 .390 .073 ga .907 .277 -.286 -.123 mo .858 .209 -.363 -.068 ni .412 .547 .618 .094 pb .900 .384 -.106 .063 s -.165 -.763 .557 .057 zn -.487 .686 .342 -.276 eigen values 9.230 3.002 2.071 1.644 % of variance 54.295 17.660 12.183 9.669 cumulative % 54.295 71.954 84.138 93.806 elements component 1 component 2 component 3 component 4 ba 0.276 0.872 0.164 0.185 sr 0.779 0.606 0.084 0.070 y -0.437 -0.316 0.500 0.577 sc 0.818 0.560 0.072 0.035 zr 0.920 0.292 0.195 0.152 be 0.507 -0.070 -0.040 0.809 v 0.954 0.025 0.151 0.209 cd -0.220 -0.478 0.028 -0.764 co 0.242 -0.048 0.928 0.008 cr 0.958 0.260 0.034 0.088 cu 0.691 0.340 0.600 0.129 ga 0.961 -0.151 0.162 0.152 mo 0.913 -0.164 0.074 0.226 ni 0.235 0.102 0.883 -0.114 pb 0.859 -0.117 0.406 0.239 s -0.386 0.866 -0.152 -0.009 zn -0.374 -0.399 0.474 -0.614 eigen values 7.898 3.028 2.798 2.223 % of variance 46.46 17.81 16.45 13.07 cumulative % 46.46 64.27 80.72 93.80 the geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria 161 (r= 0.951) indicates the presence of dolomite in the lignite samples (yazdi and shiravani, 2004). finally, the third factor expresses 13.7% of the total variance and shows high positive loadings with fe 2o3. the high positive loading of fe2o3 in factor 3 confirms the presence of iron bearing minerals such as pyrite. also the moderate positive correlation between fe2o3 and mno (r= 0.577) indicates that the element is present in carbonates, probably siderite. principal component analysis extraction method was further applied to characterize the geochemistry of major elements. the elements were plotted in rotated space yielding three components (fig. 6) which explained 97% of the total data variance. the chemical compositions were categorized to three groups (fig. 7) by cluster analysis. according to horner and krissek (1992), cluster analysis is a powerful tool that helps in the identification of groups with similar samples, while principal component analysis aids in the identification of elements (variables) that are responsible for the similarities or differences between groups of samples. the three components derived from factor analysis (fig. 6) correspond to the three cluster groups of major elements (fig. 7). cluster i comprises of al 2o3, tio2, p2o5, sio2, and k2o, indicating that the majority of these elements were mainly of terrigenous origin. cluster ii comprises of mgo and cao, indicating sediments of carbonate origin, while cluster iii comprises of fe 2o3, mno and loi indicating affiliation to epigenetic minerals such as pyrite and siderite. trace element geochemistry-correlation matrix table 6 shows the average composition and standard deviation of the trace elements in the lignite sample. the data set was screened to identify outliers (figure 8). among the trace elements analysed, zn, cd, ga, mo, pb, s and co were not normally distributed. spearman’s rank correlation determined the correlation coefficient between trace elements and ash yield (table 8). most trace elements had strong positive correlation coefficients with ash yield and with each other (table 7); this indicated a common detrital source for most of these elements. y, be, cd, co, ni, s and zn poorly correlated with ash yield (0.380 < r < 0.469) which could be explained by organic affinity and these elements’ association with sulphide figure 9. principal component analysis for the first 3 components for trace elements. figure 10. the results of cluster analyses on the trace elements contained in the lignite samples from ogwashi-asaba formation. c om p on en t 2 component 1 component 3 1,0 0,5 0,0 -0,5 -1,0 ni cd co y zn ii i iii -1,0 -1,0-0,5 -0,50,0 0,00,5 0,51,0 1,0 s ba sr cr sc cu zr be mo pb ga phases. the poor correlation of s with cu (r = 0.099), zn (r = 0.196), cd (r = 0.224), ga (r = 0.539), pb (r = 0.465), sr (r = 0.242), sc (r = 0.130), y (r = -0.125), zr (r = 0.147), be (r = 0.196), v (r = 0.330), co (r = 0.350), cr (r = 0.173), ni (r = 0.132) and mo (r = 0.428) indicated that sulphur was not only present in its sulphide form but also as an organic form (finkelman, 1995). there were highly significant correlations (r ≥ 0.90) between the elements in the following couples: sr and sc, sr and zr, sr and cr, sc and zr, sc and cr, zr and v, zr and cr, zr and ga, v and cr, v and ga, v and mo, v and pb, co and ni, cr and ga, ga and mo, ga and pb and mo and pb (table 7). this result was consistent with that recorded by spears and zheng (1999) and was interpreted as being essentially associated with detrital minerals and clay minerals. fa applied to the correlation matrix initially produced overlaps in ba, be, co, ni and cd (table 8), thereby necessitating further analysis. jude e. ogala162 this was achieved by rotating the axis, thereby producing a new set of factors; this was attained after 11 iterations (table 9). each factor involved a sub-set of the original variables with as much minimal overlap as possible (table 9). since the differences between the unrotated and rotated components were significantly different, the rotated option was used for further analysis. a 4-factor model having 93.8% total cumulative variance was obtained (table 9). principal component 1 (pc1) expressed 46.5% of total data variance and showed high positive loadings with sr, sc, zr, v, cr, cu, ga, mo and pb (table 9). the inferred host phases for pc1 were a mixture of lithophile (sr, sc, zr, v and cr; r = 0.803-0.973) and chalcophile (cu, ga, mo and pb; r = 0.620-0.829). the high positive loading of pb in pc1 may have resulted from processes which led to the emplacement of pb-zn mineralisation in the lower benue trough (akande and erdtmann, 1998; ogala et al., 2010b). the positive loading of cu in both pc1 and pc3 indicated dual sources contributing to the presence of cu. further analysis using ca (figure 10) confirmed that cu belonged to pc1. pc1 had a high positive loading with zr (table 9), indicating a concentration of resistant heavy minerals, such as zircon. principal component 2 (pc2) expressed 17.8% of total data variance and had the highest loadings with ba, sr and s. according to finkelman (1995) and christanis et al., (1998), ba and s have a strong organic affiliation, while sr indicates a carbonate origin. principal component 3 (pc3) expressed 16.5% of total data variance and showed positive loading with co, cu and ni. it could be inferred from the data obtained that this group of elements was associated with fe-sulphide (pyrite). the reported occurrence of chalcopyrite (cufes 2) in the abakaliki area of the lower benue trough cannot be ruled out due to the presence of cu (elueze et al., 2009). principal component 4 (pc4) expressed 13% of total variance and showed positive loading with y and be and negative loading for cd and zn. pc4 was made up of a mixture of lithophile (be and y) and chalcophile (cd and zn), suggesting that the elements were associated with silicates and sulphides, respectively. the trace elements were plotted in rotated space yielding three components (fig. 9). ca established trace element association and origin (fig. 10); this clustering (fig. 10) coincided with the results obtained from studying the correlation coefficient matrix (table 7) and pca (fig. 9). cluster 1 consisted of elements associated with detrital input (i.e. zr, cr, sr, sc, cu, v, mo, ga, pb and be) including quartz, clay and accessory minerals such as zircon and rutile commonly representing inorganic constituents (finkelman, 1995). these elements’ strong positive correlation with each other (table 7) and with ash yield indicated a common terrigenous origin. cluster 2 was made up of ba and s, suggesting a strong organic association. the positive statistical correlation between ba and s (r = 0.693) also supported an organic association for these elements. cluster 3 consisting of cd, zn, co, ni and y, indicated a strong association with sulphides such as pyrite (fes 2), sphalerite (zns), chalcopyrite (cufes2) and galena (pbs). the occurrence of polymetallic sulphide lodes of pbs, zns and cufes2 within the lower benue trough cannot be ruled out for these elements (elueze et al., 2009). conclusion the geochemical characteristics of the ignite seams within the ogwashi-asaba formation were investigated; the main mineral phases contained in the lignite were quartz, feldspar and pyrite. the relatively low silica (sio 2) and alumina (al2o3) content suggested very limited detrital input during coal formation. the elements’ geochemical association was examined through statistical correlation and principal component analysis (pca) and cluster analysis (ca) of major and trace elements. the three components derived from fa corresponded to the three major elements’ cluster groups. the first group consisted of clastic rock-derived elements, whilst the second and third groups consisted of elements having carbonate and sulphide affinity, respectively. the concentrations for most trace elements were lower than world average values for low-rank coals. this study has also shown that silicate and sulphide minerals were responsible for the occurrence and distribution of most trace elements in the lignite samples studied here. acknowledgement the author is grateful to activation laboratories, ancaster, canada, for performing the geochemical analysis. references adamidou, k., kassoli-fournaraki, a., filippidis, a., christanis, k., amanatidou, e., tsikritzis, l., and patrikaki, o., 2007. chemical investigation of lignite samples and their ashing products from kardia lignite field of ptolemais, northern greece. fuel, 86, 2502-2508. adedosu, t.a., adedosu, h.o., and adebiyi, f.m., 2007. geochemical and mineralogical significance of trace metals in benue trough coals, nigeria. journal of applied sciences, volume 7, number 20, pp. 3101-3105. akande, s.o., and erdtmann, b.d., 1998. burial metamorphism (thermal maturation) in cretaceous sediments of the southern benue trough and anambra basin, nigeria. american association of petroleum geologists bulletin, volume 82, number 6, pp. 1191-1206. akande, s.o., hoffknecht, a., erdtmann, b.d., 1992. rank and petrographic composition of selected upper cretaceous and tertiary coals of southern nigeria. international journal of coal geology 71, 209224. algeo, t.j., and maynard, j.b., 2004. trace-element behaviour and redox facies in core shales of upper pennsylvanian kansas-type cyclothems. chemical geology 206, 289-318. benkhelil, j., 1989. the evolution of the cretaceous benue trough, nigeria. journal of african earth sciences, 8, 251-282. calvert, s.e., and pedersen, t.f., 1993. geochemistry of recent oxic and anoxic marine sediments: implications for the geological record. marine geology, 113, 67-88. chatziapostolou, a., kalaitzidis, s, papazisimou, s, christanis, k., and vagias, d., 2006. mode of occurrence of trace elements in the pellana lignite (se peloponnese, greece). international of coal geology, 65, 3-16. christanis, k., georgakopoulos, a., fernandez-turiel, j.l., and bouzinos, a., 1998. geological factors influencing the concentration of trace elements in the philippi peatland, eastern macedonian, greece. international journal of coal geology, 36, 295-313. davidson, r.m., 2000. mode of occurrence of trace elements in coal; results from an international collaborative program. iea coal research, london. de swardt, a.m.j., and piper, h., 1957. the lignites of asaba division, benin province. records of the geological survey of nigeria, pp. 5-23. de swardt, a.m.j., and casey, o.p., 1963. the coal resources of nigeria. bulletin of geological survey of nigeria, number 28. du preez, j.w., 1945. geology of oba and nnewi lignite deoposits, onitsha province. annual report of geological survey of nigeria, 26-29. du preez, j.w., 1946. further observation on the lignites of onitsha province. annual report of geological survey of nigeria, 25-26. edet, a.e., merkel, b.j., and offiong, o.e., 2003. trace element hydrogeochemical assessment of the calabar coastal plain acquifer, souththe geochemistry of lignite from the neogene ogwashi-asaba formation, niger delta basin, southern nigeria 163 eastern nigeria, using statistical methods. environmental geology, volume 44, pp. 137-149. elueze, a.a., ekwere, a.s., and nton, m.e., 2009. geoenvironmental assessment of the environs of the aluminium smelting company in ikot abasi, southeastern nigeria. journal of mining and geology, volume 45, number 2, pp. 115-128. ewa, i.o.b., 2004. data evaluation of trace elements determined in nigerian coal using cluster procedures. applied radiation and isotopes, volume 60, pp. 751-758. ewa, i.o.b., adetunji, j., and elegba, s.b., 1996. determination of trace elements in nigerian coal ash by instrumental neutron activation analysis. journal of environmental science and health, part a, volume 31, pp. 1089-1100. filippidis, a., georgakopoulos, a., kassoli-fournaraki, a., misaelides, p., yiakkoupis, p., broussoulis, j., 1996. trace element contents of three lignite seams from the central part of the drama lignite deposit, macedonia, greece. international journal of coal geology 29, 219-234. finkelman, r.b., 1994b. modes of occurrence of potentially hazardous elements in coals: levels of confidence. fuel processing technology, volume 29, pp. 21-34. finkelman, r.b., 1995. modes of occurrence of environmentally sensitive trace elements of coal. environmental aspects of trace elements of coal. kluwer academic publishers, the netherlands, 24-50 pp. foscolos, a.e., goodarzi, f., koukouzas, c.n., and hatziyiannis, g., 1989. reconnaissance study of mineral matter and trace elements in greek lignites. chemical geology 76, 76-130. gentzis, th., goodarzi, f., and broussoulis, y., 1990. petrographic characteristics and depositional environment of greek lignites: ii. ioannina basin, western greece. international journal of coal quality 9(2), 5361. georgakopoulos, a., filippidis, a., and kassoli-fournaraki, a., 1994. morphology and trace element contents of the fly ash from main and northern lignite fields, ptolemais, greece. fuel 73, 1802-1804. gerouki, f., foscolos, a.e., and dimitroula, m., 1996. environmental impact of trace elements encountered in fly ashes from power stations located in the wider area of ptolemaes basin. in: anagnostopoulos, a. (editor), proceedings of 3rd international conference of environmental pollution, thessaloniki, pp. 214-218. gluskoter, h.j., ruch, r.r., miller, w.g., cahill, r.a., dreher, g.b., and kuhn, j.k., (1977): trace elements in coal: occurrence and distribution. illinois state geological survey circular, volume 499, 154p. goodarzi, f., gentzis, th., and yiakkoupis, p., 1990. petrographic characteristics and depositional environment of greek lignite: i. drama basin, northern greece. journal of coal quality 9(1), 26-37. hatch, j.r., and leventhal, j.s., 1992. relationship between inferred redox potential of the depositional environment and geochemistry of the upper pennsylvanian (missourian) stark shale member of the dennis limestone, wabaunsee county, kansas, usa. chemical geology, volume 99, pp. 65-82. hoffman, d.l., algeo, t.j., maynard, j.b., joachimski, m.m., hower, j.c., and jaminski, j., 1998. regional stratigraphic variation in bottom water anoxia in offshore core shales of upper pennsylvanian cyclothems from eastern midcontinent shelf (kansas), u.s.a. in: schieber, j., zimmerle, w., and sethi, p., (editors), shales and mudstones, i.e. schweizerbart’sche verlagsbuchhandlung, stuttgart, pp. 243-269. horner, t.c., and krissek, l.a., 1992. statistical analysis of geochemical patterns in fine-grained permian clastics from the beardmore glacier region, antarctica. recent progress in antarctic earth science. terrapub, tokyo, japan, pp. 241-248. jan du chêne, r., onyike, m.s., sowumi, m.a., 1978. some new eocene pollen of the ogwashi-asaba formation, southeastern nigeria. revista española de micropaleontología 10, 285–322. johnson, c.l., hudson, s.m., rowe, h.d., efendiyeva, m.a., 2010. geochemical constraints on the palaeocene-miocene evolution of eastern azerbaijan, with implications for the south caspian basin and eastern paratethys. basin research 22, 733-750. jones, b., and manning, d.a.c., 1994. comparison of geochemical indices used for the interpretation of palaeoredox conditions in ancient mudstones. chemical geology, volume 111, pp. 111-129. kogbe, c.a., 1976. paleogeographic history of nigeria from albian times. in: kogbe, c.a. (ed.), geology of nigeria. elizabethan publishers, lagos, pp. 237–252. lewan, m.d., 1984. factors controlling the proportionality of vanadium to nickel in crude oils. geochimica cosmochimica acta, volume 48, pp. 2231-2238. liu, c.w., lin, k.h., and kuo, y.m., 2003. application of factor analysis in the assessment of groundwater quality in a blackfoot disease area in taiwan. science of the total environment, v o l u m e 313, numbers 1-3, pp. 77-89. lu, x., zeng, h., xu, t. and yan, r., 1995. chemometric studies of distribution of trace elements in seven chinese coals. fuel, volume 74, number 9, pp. 1382-1386. meij, r., 1995. the distribution of trace elements during the combustion of coal. in: swaine, d.j., goodarzi, f., (editors), environmental aspects of trace elements in coal. kluwer academic publishing, the netherlands, pp. 111-127. ministry of mines and steel development of nigeria (momsd), 2007. http://msmd.gov.ng/aboutmsmd/profileministry.asp. murat, r.c., 1972. stratigraphy and paleogeography of the cretaceous and lower tertiary in southern nigeria. in: african geology (ed. by t.j.f. dessauvagie and a.j. whiteman), university of ibadan press, pp. 251-266. ndiokwere, ch.l., guinn, v.p., and burtner, d., 1983. trace elemental composition of nigerian coal measured by neutron activation analysis. journal of radioanalytical and nuclear chemistry, 79(1), pp. 123-128. nwachukwu, s.o., 1972. the tectonic evolution of the southern portion of the benue trough, nigeria. geological magazine, 109, 411-419. nwadinigwe, c.a., 1992. wax and resin characteristics of nigeria’s lignites and sub-bituminous coals. journal of mining and geology 28(1), 75-80. obaje, n.g., 2009. geology and mineral resources of nigeria. springer verlag, berlin, 221 pp. obaje, n.g., hamza, h., 2000. liquid hydrocarbon source-rock potential of mid-cretaceous coals and coal measures in the middle benue trough of nigeria. journal of african earth sciences 89, 130-139. obaje, n.g., ligouis, b., 1996. petrographic evaluation of the depositional environments of the cretaceous obi/lafia coal deposits in the benue trough of nigeria. journal of african earth sciences 22, 159171. ogala, j.e., akaegbobi, i.m., omo-irabor, o.o., and finkelman, r.b., 2009. statistical analysis of geochemical distribution of major and trace elements of the maastrichtian coal measures in the anambra basin, nigeria. petroleum and coal, volume 54, number 4, pp. 261270. ogala, j.e., omo-irabor, o.o., finkelman, r.b., and akaegbobi, i.m., 2010a. major and trace element distributions in coal and coaly shale seams in the enugu escarpment of southeastern nigeria. global journal of geological sciences, volume 8, number 2, pp. 175-186. ogala, j.e., finkelman, r.b., akaegbobi, i.m., and henry, f., 2010b. evaluation of the trace element contents of some nigerian coal samples. journal of mining and geology, volume 46, number 2, pp. 151-161. jude e. ogala164 okezie, c.n., onuogu, s.a., 1985. the lignites of southeastern nigeria. a summary of available information. geological survey of nigeria occasional paper no 10, 28 pp. olade, m.a., 1975. evolution of the nigeria’s benue trough (aulacogen): a tectonic model. geological magazine, 112, 575-583. olabanji, s.o., (1991): nigerian coal analysis by pixe and rbs techniques. journal of radioanalytical and nuclear chemistry ,149(1), pp. 41-49. olajire, a.a., ameen, a.b., abdul hammed, m., and adekola, f.a., 2007. occurrence and distribution of metals and porphyrins in nigerian coal minerals. journal of fuel chemistry and technology, volume 35, pp. 641-647. olobaniyi, s.b., ogala, j.e., 2011. major and trace element characteristics of the tertiary lignite series within the ogwashi-asaba formation, southern nigeria. proc. 23rd colloquium of african geology/14th conference of the geological society of africa (johannesburg, south africa, january 8-14, 2011), abstract book p. 323. orajaka, i.p., onwemesi, g., egboka, b.c.e., nwankor, g.i., 1990. nigerian coal. mining magazine 162, 446-451. orem, w.h. and finkelman, r.b., 2003. coal formation and geochemistry. in: f.t. mackenzie (editor), sediments, diagenesis, and sedimentary rocks (volume 7), treatise on geochemistry, elsevierpergamon, oxford, pp. 191-222. reyment, r.a., 1965. aspects of geology of nigeria. university of ibadan press. 145 pp. rimmer, s.m., 2004. geochemical paleoredox indicators in devonianmississippian blacks shales, central appalachian basin (u.s.a.). chemical geology 206, 373-391. rimmer, s.m., thompson, j.a., goodnight, s.a., and robl, t.l., 2004. multiple controls on the preservation of organic matter in devonian-mississippian marine blacks shales: geochemical and petrographic evidence. palaeogeography, palaeoclimatology, palaeoecology 215, 125-154. saez, r., moreno, c., gonzalez, f., and almodovar, g.r., 2011. black shales and massive sulfide deposits: causal casual relationships? insights from rammelsberg, tharsis, and draa sfar. mineralium deposita, 46, 585-614. sakorafa, v., michailidis, k., and burragato, f., 1996. mineralogy, geochemistry and physical properties of fly ash from the megalopolis lignite fields, peloponnese, southern greece. fuel 75, 419-423. seredin, v.v., finkelman, r.b., 2008. metalliferous coals: a review of the main genetic and geochemical types. international journal of coal geology 76, 253-289. short, k.c., and stauble, a.j., 1967. outline of geology of niger delta. american association of petroleum geologists bulletin, 51, 761779. simpson, a., 1949. the lignite seams of asaba division, benin province. annual report of geological survey of nigeria, 6-14. simpson, a., 1954. the nigerian coal field: the geology of parts of onitsha, owerri, and benue provinces. geological survey of nigeria bulletin, no 24, 85 pp. sonibare, o.o., ehinola, o.a., egashira, r. and lim, k., 2005. an investigation into the thermal decomposition of nigerian coal. journal of applied science, 5(1): 104-107. spears, d.a., and zheng, y., 1999. geochemistry and origin of elements in some uk coals. international jour. coal geol., 38, pp. 161-179. swaine, d.j., 1990. trace elements in coal. butterworth and co. (publishers) ltd. 278p. vassilev, s.v., eskenazy, g.m., and vassileva, c.g., 2001. behaviour of elements and minerals during preparation and combustion of the pemik coal, bulgaria. fuel processing technology 72, 103-129. wilson, r.c., 1924. brown coal in nigeria. occassional paper number 1 of geological survey of nigeria, 1-15. wilson, r.c., 1925. the geology of the eastern railway: section i, portharcourt to enugu. brown coal in nigeria. bulletin of geological survey of nigeria, number 8, pp. 13-86. wilson, r.c., and bain, a.d., 1928. the nigerian coalfield section ii, parts of onitsha and owerri provinces. bulletin of geological survey of nigeria, number 12, pp. 12-50. yazdi, m., and shiravani, a.e., 2004. geochemical properties of coals in the lushan coalfield of iran. international journal of coal geology 60, 73-79. search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 geofls. colomb. santaf' de sogota d.c ls5n • 0121-2974 los potenciales espontaneos: aspectos basicos y sus aplicaciones luis a. briceno g. profesor asociado departamento de geociencias-facultad de ciencias-universidad nacional de colombia robinson leon departamento de geociencias carrera de geologia facultad de ciencias universidad nacional de colombia briceno, l.a & le6n, r.: los potenciales espontaneos: aspectos baslcos y sus aplicaciones. geofls. colomb. 2:21-24, 1993. issn 0121-2974 resumen los potenciales generados espontaneamente en un medio, sp, se han observado en suelos infrayacidos por arcillas y por areniscas. en cada uno de los casos se miden su dependencia temporal y se establecen rangos de valores: 10 mv para arcillas y 0.6 mv para areniscas, en el caso de que los electrodos de medici6n se coloquen sobre medios homoqsneos. las rapidas y econ6micas medidas ofrecen una gran posibilidad de investigaci6n del subsuelo. abstract the spontaneous potentials, sp, has been observed in two different mediums. in one of them, soil is underlayed by clays and in the other by sandstones. in both cases, time dependency has been observed, and range of values has been established: 10 mv for clays and 0.6 mv for sadstones, if there exists lateral homogenity. because the sp method is really fast and low cost, it provides a wide range of possibilities for doing research. 1. definici6n y causas los potenciales espontaneos "sp",son diferencias de potencial naturales que suelen existir entre dos puntos cualesquiera del terreno. su origen se puede atribuir a diversas causas como: la presencia de capas i6nicas; diferencias de ph, temperatura, salinidad; electrofiltraci6n 6 electro-osmosis. estas causas pueden originar acciones electroqufmicas entre los diferentes minerales y las soluciones con las cuales estas estan en contacto. generalmente. estas soluciones son de composici6n variada 10 que origina presiones de disoluci6n diferenciales en las respectivas superficies de contacto. estas diferencias de disoluci6n generan diferencias de potencial las que a su vez originan un f1ujo de corriente en el subsuelo. 2. utilidad las corrientes naturales de la tierra general mente se miden de manera potenciomstrlca con dos electrodos no polarizables, 0 hechos en plomo 0 grafito y un milivoltlmetro de buena sensibilidad. se puede decir que la profundidad alcanzada en la medida es aproximadamente igual a la separaci6n de los electrodos. esta sencillez operativa caracterizada por un bajo costo del equipo, su simplicidad de rnane]o, la facilidad de transporte y el poco nurnero de operarios requeridos, hacen de este un metodo de alto rendimiento, y sus resultados saran utiles en tanto que la medida de datos sea sistematica y cuidadosa. \22 briceno & le6n: potenciales espontaneos: aspectos basicos y aplicaciones $p (mv) 12 .--.-:..-.....:-------------------------, 10 ;:::j:i;:a::e.e-81b-1;r1:*hh:h3.a:a ..a::...e.:!.-~-~....~....~-.~....~.....~...~...~~.....~...~.....~..~...:...-:..=-....=....~-.:..~....~..-:...-~--.-.............................. _._................................ . -_ __ ._._-._ _._ -_ -._ -_ _ __ _._ __ __ _ _--_ _-_ __ _-8 6 ....................... _ _ .-...-..__ __ _ _ _ _-_ -._ -. 4 ._ -_ _._ _-_ _--.--_ _-_ _.__ _-----_.-_ __ ._ _---_.--_ .._ ..__ .._ __ •.._---_ -_ __ .. 2 ...................................................... _ ••• ,••• _ .. w, , ,_._._._.,_, .. , •••, •••••• _,_ •••• __ ., ••••••••• _._._ -.••,.-.----, •• ,-,,-, ••• -•••••• ,•••• _-••••• _ --.qi9..l.. ...l....i-..... o 4010 20 30 t (min) -t:tserial! d figura 1. sp sustrato arcllloso. ciudad universitaria 07-11-92 ei "sp" depende en gran medida de la composici6n intrfnseca de la roca, por eso sus valores y comporlamiento son reflejo de su iitologia. una aplicaci6n sencilla de la observaci6n superficial de sp, puede ser la de caracterizar rocas sedimentarias con arreglos estructurales sencillos, que sstsn bajo coberleras de poco espesor (hasta 1 m), ya que se espera que exista un cierto rango de sp para una litologia determinada. otra aplicaci6n imporlante es su ulilizaci6n en zonas de actividad votcanica, donde el potencial geoelectrico sufre grandes variaciones inducidas por esta actividad. los cambios en la amplitud, direcci6n y fase de la corriente de la tierra se pueden deber, entre otras, a cambios de resistividad, temperatura, presi6n y condiciones quimicas, que normalmente preceden a los' eventos de actividad rnaqmatica. la debilidad de las corrientes generadas es la principal limitante del metodo. a la que se suman otros factores como la dependencia de las concentraciones quimicas no propias de la roca (aguas subterransas, diagenesis, etc.); la "rigidez" quimica de las rocas cristalinas (igneas y metam6rficas) y su gran irregularidad de distribuci6n geometrica, dificultan la aplicaci6n del rnstodo "sp". por 10 dsrnas es de gran ayuda en la detecci6n de la variaci6n facial de las rocas sedimentarias. 3. objetivos ei presente trabajo se realiza con el objeto de establecer una caracterizaci6n electrica de "tipos faciales" de rocas mediante observaciones de sp. se han tomado dos de las litologias mas representativas: arcillas y arenas; ambas claramente reconocibles, diferenciables, de una amplia presencia en la columna estratiqrafica y ademas de facil caracterizaci6n como patrones por otros rnetodos. se plantea la hip6tesis de que para una iitologia dada existe un rango de potencial generado sspontaneamente en el subsuelo (naturaleza) y observable en la superficie. ai realizar las medidas se observa su comporlamiento con el tiempo y con las condiciones de su medida (calidad y tipo de equipos, condiciones del subsuelo y metodologia de adquisici6n de datos). de modo general, se puede abrir un canal de inquietud sobre el fen6meno inherente a la electricidad sspontanea de la naturaleza y que puede afectar en algun modo la concepci6n sobre la misma, en cuanto a su caracter permanentemente activo. geofisica colombiana n° 2, abril de 1993 sp (mv) 23 0,6 .. _ • ..co>. •• •..... , .. ... .. ..__ ..... ..__ .._. .._--.--.__ .----_ ...if 1.5 ~-.................-................--.---.-----...-.-..---.....-...........-1, 0 ..~-----_--_._ _ ..--.-...._ _ ....-._ -.-.-.-..--,-.,..--5 0.0 l..-. l-' .....l ---ll....j o 2010 t (min) -.serie. c figura 2. sp sustrato arenoso. la punta e1 06-02-93. 4. metodologia en la adquisici6n de datos para nuestro experimento se tuvieron en cuenta cuatro aspectos relacionados con: . a) equipo: utilizaci6n de un milivoltimetro anal6gico marca phoenix y electrodos no polarizables de cobre en soluci6n saturada de sulfato de cobre. b) dispositivo: los electrodos se enterraron aproximadamente 15 cm y la distancia entre ellos se mantuvo constante e igual a 2 m. en cada localidad y ocasi6n se hicieron por 10 menos 4 lecturas en sitios diferentes. c) ei medio: las lecturas se realizaron generalmente sobre suelos 0 'zonas de meteorizaci6n que varian entre 0.2 y 1.0 m de espesor, infrayacidas en unos casos por formaciones arcillosas recientes y en otro por areniscas. d) control temporal: para asegurar la estabilidad y calidad de las medidas, se hicieron en cada ocasi6n medidas multiples en el tiempo, que generaron series de largos perfodos, para observaciones de mas de 30 minutos. 5. resultados la fig.1 corresponde a variaciones de sp, contra el tiempo, sobre un suelo orqanico, de aproximadamente 0.6 m de espesor, que 15 suprayace una unidad arcillosa cuaternaria de varios metros de espesor. las medidas se realizaron en predios de la universidad nacional de colombia en santate de bogota. la curva obtenida es de tipo exponencial creciente, con una amplitud del orden de 10 mv. las figs.2 y 3 corresponden a observaciones realizadas en la vereda la punta (sabana de bogota), donde afloran unas areniscas. elias se tomaron sobre unas lodolitas arenosas, con espesores entre 0.2 y 1.0 m, que suprayacen los bancos de areniscas. en el caso de la fig.2, los electrodos de medici6n se ubicaron sobre las lodolitas, mientras que en la fig.3, uno de los electrodos se ubic6 sobre las areniscas. se observa que en el caso de colocar ambos electrodos sobre las lodolitas, los valores de sp son bastante estables en el tiempo (excepci6n del primer punto), tendiendo hacia un valor de 0,6 mv. cuando se colocan los electrodos en diferentes medios, hay f1uctuaciones en los valores del sp auncuando, en este caso, las amplitudes observadas fueron menores que en los casos anteriores. 6. analisis los rnetodos geoelectricos de prospecci6n que incluyen medidas galvanicas de potencial, deberan siempre tener en cuenta la presencia de potenciales espontaneos. sp, los cuales son 24 briceno & le6n: potenciales espontaneos: aspectos basicos y aplicaciones sp (mv) 0.4 ;=-..:..:::.....:------------------------, 0.3 ~----------_._.0.2 0.1 0.0 l-l l......l.. ---' o 206 10 16 t (min) ~ seriell e figura 3. sp sustrato arenoso. la punta e2 06-02-93. funci6n, tanto de litologias como del tiempo. se puede afirmar que si el medio es aparentemente hornoqeneo, conviene esperar unos pocos minutos, para acercarse a un valor mas estable. fluctuaciones del sp, pueden estar asociadas a inhomogeneidades laterales del medio en el cual se realizan las observaciones. la unidad infrayacente a profundidades del orden de separaci6n de los electrodos, define la amplitud del sp observado. se observ6, en el presente caso, que los sp sobre formaciones arcillosas, son por 10 menos un orden de magnitud mayor que sobre areniscas. este tipo de resultados coincide con estudios realizados en pozos (ver p.ej. shelton, 1973), en donde se muestra la clara diferenciaci6n entre iitologfas, a partir del simple valor del sp. 7. conclusiones los experimentos realizados permiten concluir, con medici ones seneillas pero cuidadosas del sp en superficie: a) si los valores de sp son relativamente estables, existe homogeneidad lateral del medio; b) la amplitud del sp, depende de la litologia a profundidades del orden de separaci6n de los electrodos. para arcillas, el valor medido fue del orden de 10 mv, sobre suelos orqanlcos: mientras que para areniscas sobre lodolitas arenosas, su valor fue del orden de 0.6 mv. se recornienda, observar con detalle los sp, sobre litologias especificas para establecer un rango mas confiable de valores, que permitan su facil uso como identificador facial. observar los sp, en otro tipo de condiciones hidrodinamicas, eventos volcanicos p.ej., pues parece abrirse un gran campo de acci6n y aplicaci6n. bibliografia dobrin. m.b. (1961): introducci6n a la prospecci6n geofisica. omega, barcelona. parasnis. 0.5. (1970): principios de geofisica aplicada. barcelona. metodos electricos cap. 4. davidson. m.j. (1987): unconventional methods in exploration for petroleum an natural gas. geophysics 9. shelton, j. w. (1973): models of sand and sandstone deposits: a methodology for determining sand genesis and trend. oklahoma geological survey, bulletin 118. unesco. (1971): the servillance and prediction of volcanic activity. paris. meto os geofisicos. geociencias-vol 14-1 2010.vp earth sciences research journal earth sci. res. j. vol. 14, no. 1 (june 2010): 111-124 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia john makario londoño b. universidad católica de manizales. facultad de ingenierías y arquitectura email: jmakario@ucm.edu.co abstract an analysis of the seismic activity for volcano-tectonic earthquake (vt) swarms zones at nevado del ruiz volcano (nrv) was carried out for the interval 19852002, which is the most seismic active period at nrv until now (2010). the swarm-like seismicity of nrv was frequently concentrated in very well defined clusters around the volcano. the seismic swarm zone located at the active crater was the most active during the entire time. the seismic swarm zone located to the west of the volcano suggested some relationship with the volcanic crises. it was active before and after the two eruptions occurred in november 1985 and september 1989. it is believed that this seismic activity may be used as a monitoring tool of volcanic activity. for each seismic swarm zone the vp/vs ratio was also calculated by grouping of earthquakes and stations. it was found that each seismic swarm zone had a distinct vp/vs ratio with respect to the others, except for the crater and west swarm zones, which had the same value. the average vp/vs ratios for the seismic swarm zones located at the active crater and to the west of the volcano are about 6-7% lower than that for the north swarm zone, and about 3% lower than that for the south swarm zone. we suggest that the reduction of the vp/vs ratio is due to degassing phenomena inside the central and western earthquake swarm zones, or due to the presence of microcracks inside the volcano. this supposition is in agreement with other studies of geophysics, geochemistry and drilling surveys carried out at nrv. key words: earthquake swarm, volcano-tectonic earthquakes, volcanic activity, vp/vs ratio, volcanic eruption. resumen se llevó a cabo un análisis de a actividad para zonas de enjambres sísmicos tipo volcano-tectónico (vt) en el volcán nevado del ruiz (vnr), para el período 1985-2002, que fue el período sísmicamente más activo hasta el 2010. los enjambres sísmicos se localizaron en fuentes muy bien definidas alrededor del cráter. la fuente sísmica localizada en el cráter, fue la más activa durante este tiempo. la zona ubicada al w del cráter, parece tener una relación con crisis volcánicas posteriores, permaneciendo activa antes de las erupciones de noviembre de 1985 y septiembre de 1989. se cree que la ocurrencia de sismicidad en esta zona sísmica, puede ser usada como premonitorio de actividad volcánica en el vnr. para cada zona sísmica se calculó el vp/vs agrupando sismos y estaciones para cada una de ellas. se encontró que cada zona sísmica tiene un valor vp/vs diferente de las otras. las zonas cráter y oeste (w) presentaron el mismo valor de vp/vs. el valor de vp/vs para la zona cráter y w, fue un 6-7% menor el de la zona norte, y un 3% menor que el de la zona sur. se interpreta que la disminución del vp/vs es debido a desgasificación producida en las zonas sísmicas del cráter y al occidente del volcán, o a la presencia de 111 manuscript received: 10/03/2010 accepted for publication: 25/06/2010 microgrietas en el interior del volcán. esta suposición esta soportada en otros estudios geofísicos, geoquímicos y de datos de pozos geotérmicos, llevados a cabo en el vnr. palabras clave: enjambre de terremotos, terremotos volcán tectónico, actividad volcánica, razón vp/vs, erupción volcánica 1. introduction in many volcanoes around the world, volcano-tectonic (vt) earthquakes are directly or indirectly associated with volcanic activity. in some cases they are associated with magmatic intrusions and rock fracturing (karpin and thurber, 1987). this earthquakes are also known as high-frequency events, and they are thought to be produced by shear failure or slip on faults in the volcano edifice. the only difference with tectonic earthquakes, is their pattern of occurrence. in volcanoes, they occur in swarms, that is a group of many earthquakes of about same size and location with no dominant shock (mcnutt, 2000). the nevado del ruiz volcano (nrv) (fig. 1) began the last eruptive period in december 1984; after this event the most important eruptive processes have been those of november 13, 1985 and september 1, 1989 (bohorquez, 1993). at nrv the vt seismic activity sometimes increased several days before the volcanic crises (eruptions, ash emissions, etc.). vt earthquakes occurred both in a swarm-like pattern and as isolated earthquakes, but mainly in swarm-like pattern (muñoz et al., 1990). several studies of the vt activity at nrv have been carried out with the aim of finding relationships with volcanic activity. all these studies have been done for vt seismicity in nrv but only over short periods of time (e.g., nieto and muñoz 1988; nieto et al., 1990; bohorquez 1993; londono and kobayashi, 1994; londoño 1996, londono and sudo, 2001). nieto et al. (1990) found clusters of vt earthquakes based on the hypocenters obtained for the period july 1985 december 1986. they associated such clusters with a possible ascent of magma. muñoz et al. (1990) identified some geologic features in the area around nrv analyzing hypocenters and composite focal mechanism of vt swarms 112 john makario londoño b. l a ti tu d e (n ). d e g 5.00° 4.95° 4.90 4.85 ° ° 4.80° 75.41° 75.38° 75.35° 75.31° 75.28° 75.25° longitude (w). deg nevado de santa isabel pico del cisne nevado del ruiz volcano 52 00 olleta crater arenas crater pirana crater. n figure 1. location map of the nevado del ruiz volcano (nrv). the topographic curves of 5200 and 5000 m are shown as reference. from 1986-1987. they conclude that the magma chamber of nrv has a very complex geometry. zollweg (1990) studied vt seismicity from december 1985 to may 1986, examining spatial patterns and information on the orientation of local tectonic stress. he also concludes that these kind of earthquakes are related with a magma-transport process. muñoz (1992) made a study of lateral variations of p wave velocity by using a tomographic technique. the preliminary results obtained with this 3d model suggested that there are several regions of low velocity beneath the active crater and northwest of it. bohorquez (1993) analyzed arrival times of p and s waves, polarity of the p wave, spectra of frequencies and particle motion of vt earthquake swarms occurred at august october 1990. she found spatial concentrations of vt around nrv similar to those found by muñoz et al. (1990). londono and kobayashi studied the vt swarms occurred during 1990 at nrv, by using a portable seismic network of 3-components. they found different seismic sources with different spectral characteristics. londono and sudo (2001) developed a method to determine spectral characteristics of seismic zones at nrv, by using main part of spectra of p and s waves. the calculation of the vp/vs ratio in volcanic regions is another interesting subject to study. although today the 3d tomographic technique for calculation of vp/vs is a powerful tool to find its wide spatial distributions, the modified wadati diagram (chatterjee et al., 1985) is also a useful and simple technique to find differences in vp/vs between small earthquake swarm regions. in that technique, the study area is divided in several sectors, and for each earthquake swarm region a group of earthquakes and stations are choose carefully to calculate the vp/vs ratio, to avoid mixing propagation effects between regions. it is known that changes in vp/vs ratio may be related to changes in the physical and chemical properties of the medium, such as partial saturation of rocks, changes in temperature, density, and steam content (birch, 1961; ito et al., 1979; toksoz et al., 1976). in this study we made an evolutionary study of the whole seismicity of swarm-like vt earthquakes, covering the most active periods of activity of nrv from 1985 to 1999, in order to find some relationships with volcanic activity through time. on the other hand, we calculated the vp/vs ratio for each vt swarm zone in order to determine if there are differences between the swarm zones. this paper does not pretend to map the distribution of vp/vs ratios at nrv area as a whole. since only few specific and small zones around the volcano were investigated, we consider that for the purpose of this study the modified wadati diagram is suitable for vp/vs calculation. 2. data and processing we used vt earthquakes recorded by the telemetered local network at nrv area (fig. 2); the stations feature vertical component seismometers with a natural frequency of 1 hz, model l-4c. the hypocenters of vt earthquakes for each swarm zone were selected from the database of the volcanological observatory of nrv for the period 1985 2002. since that year the nrv has not shown swarm like seismic activity until now. we relocated the earthquakes by using the earthquake location program hypo71 (lee and lahr 1985). we used a p velocity model proposed by zollweg (1990) for nrv (table 1). the arrival times of p phases were accurate to 0.05s and those of s phases to 0.1s. we selected those earthquakes which were recorded at five or more stations, with clear onset of p and s phases. the parameters for the selection of the earthquake locations in this analysis were: residual times less than 0.10s, 113 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia table 1. velocity model for p waves used for earthquake location (zollweg, 1990) velocity (km/s) depth at bottom of the layer (km) 3.6 0.00 4.6 0.95 5.5 1.95 6.1 4.00 6.6 14.00 7.8 30.00 vertical errors less than 0.5km, and horizontal errors less than 0.3km. a total of 600 high-quality earthquakes were used for the analysis. figure 2 shows the hypocenters of these earthquake swarms as well as the seismic stations used in this study. we calculated the seismic energy for the vt selected and correlated the resulting values with volcanic activity. the seismic energy (e) was calculated from the following expressions: log e = 9.9 + 1.9 ml 0.024 ml 2 (richter, 1958) ml = 2.82 log(duration) – 2.59 (nieto et al., 1990), where ml is the local magnitude for vt earthquakes at nrv. a modified wadati diagram was used for the calculation of vp/vs ratios using selected groups of stations for each earthquake swarm zone in order to see differences among them (chatterjee et al., 1985). instead of plotting the s-p as ordinate versus travel time of p (tp) as the abscissa 114 john makario londoño b. 5.00° 4.95° 4.90 4.85 ° ° 4.80° l a ti tu d e (n ). d e g 75.41° 75.38° 75.35° 75.31° 75.28° 75.25° longitude (w). deg herr told inde rubi piraolle ref3 bis1 fara alf2 cisn reci pico del cisne isa2 piraña crater b arenas crater c a) b) c d d v c) 75.41° 75.38° 75.35° 75.31° 75.28° 75.25° d e p th (k m ) 0 2 4 6 8 10 12 5.00° 4.95° 4.90 4.85 ° ° 4.80° l a ti tu d e (n ). d e g 0 2 4 6 8 10 12 depth (km) 0 5 km 52 00 figure 2. epicentral and hypocentral location of the earthquake swarms and the stations used in this study. each vt seismic zone is identified with a color and symbol (green circles=crater zone; red squares =north zones; blue rhomboids=west zone; gray circles=south zone). filled triangles represent seismic stations. the topographic curves of 5200 and 5000 masl are plotted as reference. active crater (arenas) and extinct craters (olleta and piraña) of nrv are indicated as well. used in the normal wadati diagram (1933), we plotted the travel time of s (ts) as the ordinate. assuming the poisson’s ratio to be constant along the ray path, the tp versus ts relation is a straight line through the origin with slope equal to the vp/vs ratio. a linear least square fit was made using tp versus ts data. the vp/vs ratio for each earthquake swarm was calculated with a confidence interval of 90% (press et al., 1992). figure 3 shows some examples of the typical seismograms used in this study. 3. analysis sources of error several sources of error are probably involved in the analysis of the evolution of vt seismicity in this study. uncertainty about the earthquake location as well as the origin time are perhaps two of the most important sources of error. the readings of the arrival times of p phase and s phase are also possible sources of error for both vp/vs ratio calculation and analysis of the seismicity. as we are using only the arrival time of the s phase on the vertical component, it is possible to have some error, such as p-to-s conversion, leading to a mispicking of s wave but this possibility is low. although we selected only very clear s phases, we assume an error of 0.5 sec in the arrival time of s phase and it is included in the error of earthquake location and vp/vs calculation. the velocity structure considered for hypocenter location is also a possible source of error, but some tests made previously using controlled blasts showed that the crustal model used for the hypocenter determination at nrv is well constrained. moreover, as zollweg (1990) pointed out, assuming a 1 km/s error in the p velocity model for nrv and an average travel time of 1 sec, the additional error in the location is less than 1km. this systematic error is smaller than a typical 95% confidence ellipsoid major axis lengths that is 1-3km. evolution of the vt seismicity and its relationship with volcanic activity the most recent volcanic activity of nrv started in november 1984. one year later, in november of 1985, a phreato-magmatic eruption occurred, melting about 5% of the ice cap of the volcano. this produced a mudflow, which destroyed armero city located in the eastern part of the volcano, killing about 25000 people. the volume of the eruption was calculated to be about 3.8x107 m3 (naranjo et al., 1986). from 1985 to 1989, several ash emissions occurred. on september 1, 1989 a new phreato-magmatic eruption took place. it was smaller in size than the eruption of november 1985, with a volume of about 1.6x106 m3 (mendez and patiño 1993). more than 1100 earthquakes occurred in only one day. 115 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia figure 3. example of typical seismograms of volcano tectonic earthquake at nrv. the seismograms correspond to an earthquake of crater swarm zone. the line shows the vp/vs ratio for the crater swarm zone. ash continued falling for several days and a small lahar formed. londoño et al., (1998) found a temporal change of coda q before and after these two volcanic eruptions. after the last eruption, the volcanic activity started to decrease gradually. since then, few ash emissions have been observed. table 2 shows a summary of the most relevant volcanic crises of nrv from 1985 to 1999. it seems that a change in the volcanic activity started in 1992. the substantial and continuous decreasing in the vt seismicity is remarkable after that year, compared with previous years. the total energy release for long period (lp) events as well as for all vt earthquakes including those not occurring during swarms is depicted in figure 4. it is clear that as a general tendency, the seismic activity of nrv has decreased gradually since 1992 (except on some particular dates associated with minor volcanic crises), without ash emission or volcanic eruptions. there are two seismic swarm zones that displayed some recurrence in time, but both were low in seismic energy release, and were associated with few volcanic crises. a swarm zone nw far away of the crater exhibited activity a few times, not related with volcanic crises, and was the lowest in seismic energy release. therefore, this swarm zone is not included in this study. there is a seismic swarm zone that has been associated with some of the most important crises at nrv, that is, the eruptions of november 1985 and sep116 john makario londoño b. figure 4. daily total energy (in ergs) releasing (vt and long-period earthquakes) and daily number of vt earthquakes at nrv from 1985 to 2009. asterisks represent volcanic crises. tember 1989. this western swarm-like activity occurred before and after both volcanic eruptions. although this seismic activity recurred only on rare instances, the level of energy was low compared with those recorded before and after both the eruptions mentioned above (figure 5). therefore, it is important to note that not all these vt swarms are related with subsequent volcanic activity. sometimes this was low in energy and in number of events but it was related with a posterior phreatic eruption (fig. 4). in february 1998, another swarm in the western part was recorded. although this swam zone was more energetic than previous mentioned, it did not correlate with any surface volcanic activity (fig. 5). the earthquakes located to the north and to the west were the deepest (4-8 km), while those located beneath the crater and to the south were the shallowest (0-3 km) (figure 2). as is noticed in figure 4, most of the volcanic crises occurred between 1985 and 1992. only two volcanic crises have occurred since 1992. seismic swarm-like activity also occurred concurrently. vt seismicity of this type located in the crater area has been recorded quite continuously after 1992, but it is very low in energy compared to previous years. vp/vs ratios for each vt swarm zone at nrv with the aim to obtain well constrained vp/vs ratios for each vt swarm zone and to avoid mixing propagation effects, we calculated the vp/vs ratio for each swarm zone using different groups of stations for each swarm (figures 2, 6). in order to detect temporal changes of vp/vs for each swarm zone, the vp/vs ratio was calculated for each swarm zone. the figure 7 shows the results. from this figure it is possible to observe that there are some temporal changes in vp/vs with time at nrv for some regions, mainly for crater and west zones. the vp/vs ratios showed low values at the crater and west swarm zones, with vp/vs ratios of 1.67±0.01 and 1.67±0.03 respectively. the vp/vs ratio for the earthquake 117 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia table 2. main volcanic crises at nrv from 1985 to 1999. this list includes phreatic eruptions, phreato-magmatic eruptions as well as strong seismic activity. (lp=longperiod earthquakes; vt=volcano-tectonic earthquakes). no activity date 1 small phreatic eruption september 1985 2 phreato-magmatic eruption 13 november 1985 3 small phreatic eruption january 1986 4 small phreatic eruption may 1986 5,6 small phreatic eruption july 1986 7 small phreatic eruption june 1987 8 small phreatic eruption march 1988 9 intense seismic activity (lp) february 1989 10 intense seismic activity (vt) may 1989 11 intense seismic activity (vt + tremor) june 1989 12 intense seismic activity (vt + lp+tremor ) august 1989 13 phreato-magmatic eruption 1 september 1989 14 small phreatic eruption april 1991 15 small phreatic eruption may 1991 16 intense seismic activity (lp + tremor) april 1994 17 intense seismic activity (vt + lp + hybrids ) july-september 1995 118 john makario londoño b. 0.0e+ 0 2.0 e+ 7 4.0 e+ 7 6.0 e+ 7 8.0 e+ 7 1.0 e+ 8 1. 2 e+ 8 1. 4 e+ 8 08 1985 08 1989 08 1993 08 1997 08 2001 crater swarm zone 0 .0 e+ 0 1 .0 e+ 8 2.0e+8 3 .0 e+ 8 4 .0 e+ 8 5 .0 e+ 8 6 .0 e+ 8 08 1985 08 1989 08 1993 08 1997 08 2001 west swarm zone 0.0e+0 5.0 e+7 1.0e+8 1 .5 e+ 8 2 .0 e+ 8 2.5e+8 08 -1985 08 -1989 08 -1993 08 -1997 08 -2001 south swarm zone 0.0e+ 0 5 .0e+ 7 1.0 e+ 8 1.5 e+ 8 2.0e+8 2 .5 e+ 8 08-1985 08-1989 08-1993 08-1997 north swarm zone date d a il y s e is m ic e n e rg y r e l e a s e (e rg s ) 0 .5 figure 5. energy release for each vt swarm (1985-2002). 119 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia 0 2 4 6 8 0 1 2 3 4 crater vp/vs = 1.67 (0.01)stations=ref3, bis1, fara reci,pira 0 2 4 6 8 0 1 2 3 4 west vp/vs = 1.67 (0.03)stations=ref3, bis1, fara told, 0 2 4 6 8 0 1 2 3 4 south vp/vs = 1.73 (0.02)stations=reci,alf2,cisn,isa2 0 2 4 6 8 0 1 2 3 4 north stations=herr, inde, told, rubi, ref3 , olle vp/vs = 1.78 (002) t (s)s figure 6. modified wadati diagrams for each seismic swarm zone. the group of stations used for vp/vs calculation for each swarm zone and the vp/vs ratio and its error are listed. 120 john makario londoño b. v p v s e p -8 5 n o v8 5 m a r8 8 s e p -8 9 a p r9 4 a g o -9 5 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 02-jun-85 27-feb-88 23-nov-90 19-ago-93 15-may-96 09-feb-99 crater area 02-jun-85 27-feb-88 23-nov-90 19-ago-93 15-may-96 09-feb-99 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 sep-85 n o v8 5 m a r8 8 s e p -8 9 a p r9 4 a g o -9 5 west area 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 02-jun-85 27-feb-88 23-nov-90 19-ago-93 15-may-96 09-feb-99 s e p -8 5 n o v8 5 m a r8 8 s e p -8 9 a p r9 4 a g o -9 5 north area 2 1.9 1.8 1.7 1.6 1.5 1.4 1.3 02-jun-85 27-feb-88 23-nov-90 19-ago-93 15-may-96 09-feb-99 south area s e p -8 5 n o v8 5 m a r8 8 s e p -8 9 a p r9 4 a g o -9 5 date s figure 7. temporal variation of vp/vs for each seismic swarm zone at nrv. asterisks represent volcanic crises. vertical lines represent standard error. swarms located to the south was 1.73±0.02, and for those swarms located to the north was 1.78±0.02. the vp/vs ratios were reduced about 6% for the west and crater swarm zones with respect to the north swarm, and 3% with respect to the south swarm zone. the error bars show that the estimated vp/vs ratios are significantly different among swarm regions, except for the west and crater swarms, whose values of vp/vs are indistinguishable (figure 8). some of the stations used to calculate the vp/vs ratio for the north and the west earthquake swarm zones were the same (figure 6). these two seismic zones were the deepest, as mentioned before. 4. discussion and concluding remarks vt seismic activity is closely related to the volcanic activity at nrv from 1985 to 2002. the swarm-like vt seismic activity was one of the most remarkable phenomena associated with the volcanic crises, occurring before, during or after the crises, although other phenomena such as lp earthquakes and tremor were also associated. the continuous decreasing of seismic activity related to the decrease in volcanic activity for more than 17 years, can suggest that a period of quiescence or calm has occurred since 1992, with some sporadic increment of activity that has not affected the stability of the volcanic system. vt seismic activities located beneath the crater zone and to the west of the volcano were related with the two phreato-magmatic eruptions that had occurred to date at nrv (londoño and sudo, 2001). it is possible that such seismic pattern can be used as a premonitory tool for the major volcanic crises. another volcanic eruption has not occurred yet to confirm this supposition, but we argued, that the level of energy and the conditions of the volcano must be taken into account. it seems that after a certain level of seismic energy release is reached in the western zone, depending on the conditions of the volcano, monitoring of vt becomes an important tool for the surveillance of nrv activity, as is seen on figure 5, where only those dates with high seismic energy release at western zone, are related with phreato-magmatic activity. we suggest that the reduction of 6% of the vp/vs ratio between the north and the west swarm zones is due to the differences in travel path between them. moreover, although it is known that there is a strong relationship between the vp/vs ratios and the depth of the earthquakes, it is interesting to notice that the lowest vp/vs ratios are those for the crater and west zones, which are the shallowest and the deepest seismic zones, respectively. on the other hand, surface geology of nrv around the active crater is fairly homogeneous, consisting mainly of lava flows. based in this, we think that the surface geology did not affect too much the 121 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia 1.90 1.85 1.80 1.75 1.70 1.65 1.60 swarm zone figure 8. comparison of vp/vs for each vt swarm zone at nrv. vertical lines represent standard error. vp/vs ratios. from these results, we suggest that the low vp/vs ratios found in the crater and west zones are the manifestations of different material beneath the volcano. probably these zones have different physical-chemical characteristics from those of the surrounding zones. previous studies carried out in different materials show that factors such as porosity, density, temperature and composition can affect the vp/vs ratio (birch 1961, ito et al., 1979, toksoz et al., 1976). it is possible that such factors affected the vp/vs ratios inside nrv. material such as gas or steam changes the properties of the medium, reducing the vp/vs ratio as was pointed out by chatterjee et al. (1985) who suggested that the low vp/vs ratios in the yellowstone region are due to the presence of steam or gas. we suggest that in the crater and west zones of nrv probably there are more gas available due to continuous degassing of rocks, that affect the vp/vs ratio as well as the volcanic activity. moreover, the temporal changes of vp/vs ratio could be related to temporal changes in gas content inside the volcano. south and north seismic zones did not show any remarkable temporal variation of vp/vs ratio, while west and crater seismic zones did it. on the other hand, the last two seismic zones showed more activity through time (figure 4). the vp/vs ratio found in this study for almost all the vt swarm zones at nrv are rather low. the release of gas or steam from 1985 to 2002, was continuous at nrv, making it possible to observe many fumaroles inside the active crater and huge vapor columns almost daily. on the other hand, a geothermal exploration well, almost 1500m deep was drilled in the nrv area in 1998. the results suggested that: boiling processes are occurring in some places inside the volcano, there is a big hydrothermal system, and the temperatures are higher than 200 °c (monsalve et al., 1998). giggenbach et al. (1990) suggested that there is a huge hydrothermal system in nrv based on petrological and geochemical data. these results support the hypothesis that there is a huge amount of gas or steam inside nrv that affects the vp/vs ratios. this supposition is also in agreement with the idea that there are degassing rocks benath the crater and to the west sectors of the volcano (giggenbach et al., 1990). studies elsewhere suggest that presence of hydrothermal systems at the water-steam transition can explain the reduction of the mean value of the vp/vs ratio (chatterjee et al. 1985; zimmer et al, 2002; maheswar and kalachand, 2007). we think that this is the most likely process occurring at nrv as well. another possible explanation of the low values of vp/vs found in the crater and west swarm zones of nrv compared with the higher values of the north and south zones, is the presence of melting material inside the volcano, or different mechanical properties of the rocks, due probably to the presence of microcracks inside the volcano. a detailed local tomography done at nrv area by londoño and sudo (2002a) seems to support this supposition. they found two low velocity zones to the west and beneath the active crater at nrv. they associated these zones with the presence of degassing magma bodies. no low velocity zones were found at norrh of the volcano. on the other hand, londono and sudo (2002b), modeled the seismic waves attenuation at nrv due to gas content changes through time. they concluded that the gas content is the responsible for phreato-magmatic activity at nrv. therefore, the supposition that the gas content in space and time, is probably one the main causes of spatial and temporal variation of vp/vs ratio at nrv, seems to be reasonable. the relationship between both swarm-like vt seismic zones, the crater and the west zones at nrv, not only with respect to the volcanic activity, but also with respect to the similarity of vp/vs ratios, suggests that these two zones are important for the surveillance and monitoring of the volcano. we think that attention should be raised when vt seismic activity occurs in these places, and when the level of seismic energy release reaches certain levels. of course, it is necessary to correlate such vt seismic activity with other seismic phenomena such as the occurrence of lp earthquakes and tremors, and the conditions of the volcano in order to be more confident about a prediction of volcanic crisis at nrv. acknowledgments the author would like to thank the colleagues at ingeominas, the volcanological and seismological observatory of nevado del ruiz volcano, manizales, colombia, especially álvaro p. acevedo for his help and data processing. this paper was benefited by the discussions with m. yoshikawa, t. mori, and t. hashimoto of kyoto university. special thanks to william hobden for reviewing the english and their comments and suggestions. john jairo sánchez and an anonymous reviewer gave important suggestions and comments that improved the final manuscript. references bohórquez, o.p. (1993). análisis de registros sísmicos digitales en el volcán nevado del ruiz. agosto octubre de 1990. proyecto : “pronóstico de erupciones en los volcanes del p.n.n.”. (in spanish). 122 john makario londoño b. birch, f. (1961). the velocity of compressional waves in rock to 10 kilobars, part 2. elastic properties and equations of state. american geophysical union. 2199-2224. chatterjee, s.n., pitt, a.m., iyer, h.m. (1985). vp/vs ratios in the yellowstone national park region. i. j. volcanol. geotherm. res., 26, 213-230. giggenbach, w.f., garcia, p.n., londoño, c.a., rodriguez, v.l., rojas, g.n., calvache, m.l. (1990). the chemistry of fumarolic vapor and thermal spring discharges from the nevado del ruiz volcanic magmatic hydrothermal system, colombia. j. volcanol. geotherm. res., 42, 13-39. ito, h., de vilbiss, j.,nur, a. (1979). compressional and shear waves in saturated rock during water-steam transition. j. geophys. res. 84, 4731-4735. karpin, t.l., thurber, c.h. (1987). the relation between earthquake swarms and magma transport: kilawea volcano, hawaii. pagepoph, 125,4731-4735. lee, w.h., lahr, j.c. (1985). hypo71pc: a computer program for determining hypocenter, magnitude and first motion patter of local earthquakes. u.s. geol. surv. open-file rep, 75-311. londoño jm, kobayashi y (1994). spectral analysis of volcanic earthquake swarms in nevado del ruiz volcano (colombia), during august-october 1990. bull. volcanol soc. japan 39, 31-47 londoño, j.m., (1996). temporal change in coda q at nevado del ruiz volcano, colombia. j. volcanol. geotherm. res., 73, 129-139 londoño jm, sanchez jj, toro el, cruz gf, bohorquez op (1998). coda q before and after the eruptions of 13 november 1985, and 1 september 1989, at nevado del ruiz volcano, colombia. bull volcanol 59, 556-561 londoño jm, sudo y (2001). spectral characteristics of volcano-tectonic earthquake swarms in nevado del ruiz volcano, colombia. j volcanol geotherm res 112, 39-54 londoño jm, and sudo y, (2002a). velocity structure and a seismic model for nevado del ruiz volcano (colombia). journal of volcanology and geothermal research. vol. 119, no 1-4. p. 61-87. londoño jm, sudo y, (2002b) a warning model based on temporal changes of coda q for volcanic activity at nevado del ruiz volcano, colombia bulletin of volcanology, vol 64. 303-315. maheswar o. kalachand s (2007). maheswar o. kalachand s (2007). maheswar o. kalachand s (2007). marine and petroleum geology volume 25, no 7, 2008, 637-644. mcnutt s. (2000). volcano seismicity. in: encyclopedia of volcanoes. academic press. 1015-1033. méndez, r,a,, patiño, c.j. (1993). estudio estratigrafico de la zona proximal al volcán nevado del ruiz. internal report. ingeominas. (in spanish). monsalve, m.l., rodríguez, r.a., méndez, r.a., bernal, n.f. (1998). geology of the well nereidas i,nevado de ruiz volcano, colombia. geothermal resources council transactions, 22, 20-23. muñoz, c.f., nieto, e.a., meyer, h.j. (1990). analysis of swarms of high-frequency seismic events at nevado del ruiz volcano, colombia (january 1986 august 1987) : development of a procedure. j. volcanol. geotherm. res., 41, 327 354. muñoz, f. (1992). local earthquake tomography for lateral velocity variations and hypocenters at nevado del ruiz volcano-colombia. master thesis, arizona state university. naranjo, j.l., sigurdson, h., carey, j.n., fritz, w.j. (1986). eruption of nevado del ruiz volcano (colombia), 13 nov 1985: tephra fall and lahars. science, 233, 961-963. nieto, a.h. muñoz, f. (1988). energía sísmica liberada por enjambres de sismos de alta frecuencia en el volcán nevado del ruiz, colombia enero 1986 agosto 1987. bol. geol. ingeominas. vol. 29. 3, 3 – 20 (in spanish). nieto, a.h., brandsdottir, b., muñoz, f. (1990). seismicity associated with the reactivation of nevado del ruiz, colombia, july 1985 december 1986. i. j. volcanol. geotherm. res., 41, 315 326. press, w.h., teukolsky, s.a., vetterling, w.t., flannery, b.p. (1992). numerical recipies. the art of scientific computing, 686-693. raigosa, j. (1993). delimitación de zonas de baja velocidad en el volcán nevado del ruiz a partir de retardos en tiempos de viaje de la ondas p en sismos tectónicos regionales. bachelor thesis. universidad de caldas. (in spanish). richter, c.f., (1958). elementary seismology. freeman, san francisco, ca, 201-264. 123 activity and vp/vs ratio of volcano-tectonic seismic swarm zones at nevado del ruiz volcano, colombia toksoz, m.n., cheng, c.h., timur, a. (1976). velocities of seismic waves in porous rocks. geophys. 41,4, 621-645. wadati, k. (1933). on the travel time of earthquake waves, ii, geophys mag., 7, 101-111 zimmer m, prasad m, mavko g. (2002). pressure and porosity influences on vp-vs ratio in unconsolidated sand. the leading edge; soc. explor. gephys.vol. 21 no. 2. 178-183; doi: 10.1190/1.1452609 zollweg, j.e. (1990). seismicity following the 1985 eruption of nevado del ruiz, colombia. j. volcanol. geotherm. res., 41, 355-367. 124 john makario londoño b. determination of effective elastic thickness of the colombian andes using satellite-derived gravity data remy a. galán, iván f. casallas 7 petrography and application of the rietveld method to the quantitative analysis of phases of natural clinker generated by coal spontaneous combustion josé antonio henao m., angélica maría carreño p., john alexander quintero d., sol ángel candela h., carl enso modulations on streamflow characteristics ali ihsan marti, cahit yerdelen and ercan kahya 31 geophysical contribution to evaluate the hydrothermal potentiality in egypt: case study: hammam faraun and abu swiera, sinai, egypt. magdy a. atya, olga a. khachay, aiman abdel latif, oleg y. khachay, gad m. el-qady and ayman i. taha 44 groundwater investigation in awlad salameh, southern sohag, upper egypt aiman abdel-latif and mohamed el kashouty 63 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15th of may city, helwan, egypt magdy a. atya, olga a. khachay, mamdouh m. soliman, oleg yu khachay, ahmed b. khalil, mahmoud gaballah, fathy f. shaaban, i seismic geohistory and differential interformational velocity analysis in the anambra basin, nigeria ekine, a.s. and onuoha, k. m. 88 integrated assessments of possible effects of hydrocarbon and salt water intrusion on the groundwater of iganmu area of lagos metropolis, southwestern nigeria ayolabi, elijah a., folorunso, adetayo f. and obende phillips w. 100 geociencias-vol 12-2 dic.vp earth sciences research journal earth sci. res. j. vol. 12, no. 2 (december 2008): 265-268 discussion of “clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano” by mauricio orozco-alzate, and césar germán castellanos-domínguez mehmet c. demirel1, ercan kahya2 and diego rivera3 1 ph.d. student, department of water engineering and management, university of twente, po box 217, 7500 ae enschede, the netherlands. e-mail: m.c.demirel@utwente.nl 2 civil engineering department, american university of sharjah, po box 26666, sharjah, united arab emirates (corresponding author). e-mail: ekahya@aus.edu 3 professor, department of water resources, university of concepcion, vicente méndez 595, chillán, chile. e-mail: dirivera@udec.cl the authors are to be congratulated for a systematic investigation of the accurate and non subjective classifying approach in seismic research. the authors have conducted several clustering algorithms to the seismic event records from volcanological and seismological observatory at manizales. their objective was to improve the grouping of seismic data (i.e., volcano-tectonic earthquakes, long-period earthquakes and icequakes) digitized at 100.16 hz sampling frequency. their study seems adding new approach to their previous work of langer et al. (2006) who applied different classification techniques to seismic data. the discussers have the following suggestions to improve the author’s investigation on ruiz volcano data and to be a guide for similar future studies. 1. there are five empirical steps that should be followed in the application of cluster analysis which are (i) selection of variables, (ii) selection of standardization technique (if necessary), (iii) dissimilarity metric, (iv) selection of an appropriate method, and (v) test of stability or validation (demirel 2004; everitt 1993; green et al. 1990). these steps are difficult to distinguish in the present study by orozco-alzate and castellanos-domínguez (2007). the users` experience and preferences in these steps may radically affect the resultant cluster structures. for instance, the selected distance metric was not clearly mentioned in the text. did the notation �kl refer to the correlation coefficient between entities k and l? how many stations were selected near the olleta crater and the glacier at nevado del ruiz volcanic complex? were there any scale issues in the dataset which may perturb the dissimilarity matrices? henceforth 265 nov 10-discussion.prn d:\geociencias v-12-2-dic 2008\geociencias-vol 12-2 dic.vp martes, 10 de noviembre de 2009 20:50:03 p p p composite 133 lpi at 45 degrees the span of the data and some statistical information on the data structure should be explained for easy follow of readers and to avoid the aforementioned questions. standardization priori to analysis phase is necessary when the scale differences emerged in a dataset (demirel et al. 2008; everitt 1993; gnanadesikan et al. 1995; milligan and cooper 1988). 2. the authors applied several algorithms on their data and reasoned the following statement: “the lack of a single appropriate clustering algorithm”. however most of the algorithms were already tested in the literature and the relevant shortcomings are given in many text books (bacher 2002; everitt 1993). single linkage produces chain type cluster which is not be desirable for many applications, and complete linkage may create small and compact clusters (demirel 2004; everitt 1993). on the other hand the wards method emerged to make more successive clusters with small inner variance. hence it is herein suggested to use the wards method with the squared euclidean metric to get more distinct clusters in future investigations. 3. in the context of text indications in notation wise, at page 133: the notation “dc” was not explained in the text. at page 133: d(t,t) designates to distance/dissimilarity measure; however, the notation “d” was used for the same purpose in table 1. it is important to maintain consistent use of notations for the dissimilarity measure throughout the text. 4. the mismatches in labeling were counted for the performance comparison and number of runs was given as 10. the author also mentioned that “hierarchical methods report the same number of mismatches over the runs”. it should be noted that cluster structure in the hierarchical methods do not differ in any run as the steps in dissimilarity calculations and cluster delineation has concrete algorithm; thus, it is herein encouraged that issues similar to these unclear points should be justified in the manuscript. 5. the clustering results were not given in the text. the labels of clusters and statistics (i.e., variance, mean) of each cluster should be summarized in the result section. only the averaged numbers of mismatches between class labels were presented but this was not adequate for the readers to have appropriate insights regarding the main objective of the study. since the article is about signal clustering, it would have been very illustrative to put 2 figures: one graph including 3 representative signals (e.g. langer et al., 200) and one figure representing the topological structure of the clusters, e.g. dendrogram. both figures allow analyzing in an intuitive way dissimilarities among signals. 6. at page 135: the authors mentioned that “even though the number of cluster is fixed, single linkage and average linkage find second and third clusters of a few objects only”. the single linkage and average linkage methods are in the group of unsupervised clustering techniques which has no priori knowledge on number of clusters as partitioning methods; therefore, a justification should have been indicated for that matter (demirel and kahya 2007; kahya et al. 2007). 7. as was noted by morlet et al. (1982), seismic signal does vary in amplitude, shape, frequency and phase, versus propagation time. therefore, for clustering it is necessary to analyze signal’s frequency content, as well as to localize in time changes in both, frequency and amplitude. for this task, wavelet transform is a joint time-frequency signal representation that can give the frequency content of the signal at a particular instant of time by filtering (sheikholeslami et al., 1998). it is well suited for signal whose frequencies change with time, but also for signal containing noise and transients (rouyer et al., 2008). also, its multi-resolution property can help detecting the clusters at different levels of accuracy (sheikholeslami et al., 1998). we 266 mehmet c. demirel, ercan kahya and diego rivera nov 10-discussion.prn d:\geociencias v-12-2-dic 2008\geociencias-vol 12-2 dic.vp martes, 10 de noviembre de 2009 20:50:04 p p p composite 133 lpi at 45 degrees propose for further research to apply this technique to ruiz volcano data. a good reference are kumar and foufoula-georgiou (1997) and torrence and compo (1998) for methods. indeed, arciniega-ceballos et al. (2008) applied bandpass filters before clustering in seismic data and rouyer et al., (2008) applied a wavelet-based clustering technique. references arciniega-ceballos a, chouet b, dawson ph. and g asch (2008) broadband seismic measurements of degassing activity asociated with lava effusion at popocatépetl volcano, mexico. journal of volcanology and geothermal research, 170: 12-23. bacher, j. (2002). “cluster analysis.” lecture notes, nuremberg. demirel, m. c. (2004). “cluster analysis of streamflow data over turkey,” istanbul technical university, istanbul. demirel, m. c., and kahya, e. “hydrological determination of hierarchical clustering scheme by using small experimental matrix.” 27th agu hydrology days, fort collins, colorado, 161-168. demirel, m. c., kahya, e., and rivera, d. (2008). discussion of “hydrologic regionalization of watersheds in turkey” by sabahattin isik; and vijay p. singh in asce journal of hydrologic engineering, sep 2008, vol. 13, no. 9, pp. 824-834. doi: 10.1061/(asce) 1084-0699 (2008) 13:9(824) (accepted for publication). everitt, b. (1993). cluster analysis. 3rd edn. , halsted press, division of wiley, new york. gnanadesikan, r., kettenring, j. r., and tsao, s. l. (1995). “weighting and selection of variables for cluster analysis.” journal of classification, 12(1), 113-136. green, p. e., kim, j., and carmone, f. j. (1990). “a preliminary study of optimal variable weighting in k-means clustering”. journal of classification, 7(2), 271-285. kahya, e., demirel, m. c., and piechota, t. c. “spatial grouping of annual streamflow patterns in turkey ” 27th agu hydrology days, fort collins, colorado, 169-176. kumar, p. and foufoula-georgiou, e. (1997). wavelet analysis for geophysical applications. reviews of geophysics, 35:385–412. langer, h., falsaperla, s., powell, t., and thompson, g. (2006). “automatic classification and a-posteriori analysis of seismic event identification at soufri?re hills volcano, montserrat.” journal of volcanology and geothermal research, 153(1-2), 1-10. milligan, g. w., and cooper, m. c. (1988). “a study of standardization of variables in cluster analysis.” journal of classification, 5(2), 181-204. morlet, j., arens, g., fourgeau, e. and glard, d. (1982) wave propagation and sampling theory-part i: complex signal and scattering in multilayered media. geophysics 47, 203-221. orozco-alzate, m., and castellanos-domínguez, c. g. (2007). “clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano.” earth sci. res. j., 11( 2), 131-138 rouyer t, fromentin j-m, stenseth n and b cazelles (2008) analysing multiple tiem series and extending significance testing in wavelet analysis. marine ecology progress series, 359:11-23. sheikholeslami, gh., chatterjee, s. and zhang, a. (1998) wavecluster: a multi-resolution clustering approach for very large spatial databases. in proceedings of the 24th vldb conference. torrence, c. and compo, g. (1998). a practical guide to wavelet analysis. the bulletin of the american meteorological society, 79:61–78. 267 discussion of “clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano” by mauricio orozco-alzate, and césar germán castellanos-domínguez nov 10-discussion.prn d:\geociencias v-12-2-dic 2008\geociencias-vol 12-2 dic.vp martes, 10 de noviembre de 2009 20:50:05 p p p composite 133 lpi at 45 degrees nov 10-discussion.prn d:\geociencias v-12-2-dic 2008\geociencias-vol 12-2 dic.vp martes, 10 de noviembre de 2009 20:50:05 p p p composite 133 lpi at 45 degrees a new approach on the tectonometamorphic mechanisms associated with p–t paths of the barrovian-type silgará formation at the central santander massif, colombian andes castellanos a. óscar m., ríos r. carlos a. and takasu akira 125 deep crustal structure and estimation of moho in the lhasa terrane of central tibet syed mustafizur rahman, m. rezaul islam, mumnunul keramat and sultan-ul-islam 156 geophysical prospecting of the transition zone between the congo craton and the panafrican belt in cameroon tadjou j. m., njingti-nfor, kamguia j. and manguelle-dicoum e. 169 hydrologic homogeneous regions using monthly streamflow in turkey ercan kahya, mehmet c. demirel and osman a. bég 181 intraseasonal variability of rainfall over northern south america and caribbean region j.d. pabón and j. dorado 194 mineralization potentials of pegmatites in the nasarawa area of central nigeria akintola, o. f. and adekeye, j. i. d. 213 multidisciplinary approach to study migmatites: origin and tectonic history of the nason ridge migmatitic gneiss, wenatchee block, cascades crystalline core, wa, usa carlos a. zuluaga c. and harold h. stowell 235 discussion of “clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano” by mauricio orozco-alzate and césar germán castellanos-domínguez mehmet c. demirel, ercan kahya and diego rivera 265 earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014) letter from editor during the outgoing year, successful events took place, as latin american and caribbean seismological commission assembly, on july 23, 24, 25, in bogotá, colombia. in that academic event, the relevant results associated with the seismotectonic behavior of the latin america and caribbean region were addressed, as well as the implications of active exploration of the tectonic conditions, the socio-economic impacts of exploration and extraction of economic resources, and the implementation of nonconventional techniques for exploring and mining that can contribute to sustainable development. another event was the shale colombia 2014, unconventional oil and gas development congress, aimed to determine the next steps towards the commercialization of unconventional oil and gas development for the country. the event hosted case studies from some of the pioneers in colombian development, using the latest reservoir data fiscal updates and global scale technology case studies, to evaluate real potential colombia holds for unconventional development. in the outlook of 2015, one of the main events is the 26th international union of geodesy and geophysics general assembly, to be realized between june 22 and july 2, in prague, czech republic. you are on time to participate in this event yet. in colombia will take place the xv colombian congress of geology on 31 august to 5 september. this event is a bi-annual meeting gathering scientists, academicians, and researchers from the region, with an extensive participation of international guests. the congress wants to provide a scenery to discuss the perspectives of mineral resources, new exploration techniques related to energy supply, proposals for optimized management of mining and mineral resources, between other topics. finally, we want to introduce you the manuscripts selected for this edition. there is a broad range of information about seismology with four papers coming from different countries. the first one is an analysis of the elastic structure beneath australia shown by means of s-velocity maps and determined by the regionalization and inversion of rayleigh wave dispersion. the obtained s-velocity models suggest the existence of lateral and vertical heterogeneity in the structure. from turkey, the second manuscript presents a site response analysis and estimation of s-wave velocity depending on acceleration data. the other two articles in seismology for this edition are based on improving accuracy and reliability of issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.49641 http://dx.doi.org/10.15446/esrj.v18n2.49641 the first arrival times through cross correlation technique, and in the characterization of seismic events using only one seismological station. in other knowledge field, it is presented a mineralization potential study of the intrusive bodies in valis, northwest of iran. a very graphical manuscript coming from the gaza strip shows a study in sand dunes, from morphology, textural characteristics, and associated environmental impacts. from morocco, a study aimed to evaluate geostatistical methods for delimitation of anomalies detected in the phosphatic series of sidi chenane found that the exponential model is the best one to represent the spatial variability of their geoelectric data; models of the geology were successfully obtained from geostatistical method, which help mapping the phosphate deposit inclusions and the estimations of phosphate reserves were improved and better constrained. in geotechnics, three important works are displayed. one is about the experimental triaxial tests developed to estimate undrained critical shear strength under monolithic loadings; the second one reads about classification of caving depth by feature extraction in cutting images, and the other is a simulation using artificial neural network and adaptative neuro-fuzzy system of compression ratio in a municipal solid waste. last two works talk about gis and kriging techniques applied to map a spatial distribution of sediments pollution in the khajeh kory river and an assessment of the land ecological security for yancheng city, in china. we hope you may find interesting information in our esrj. carlos alberto vargas earth sciences research journal editor in chief earthquake light (eql) mainly blue-white flashes from the ground, was observed coseismically during a new zealand m7.1 earthquake, (4.47 am 4-sep-2010 local time). a local production mechanism is most probable. the blue-white flash observations support the freund et al. mechanism, i.e. shockwave disturbance creating electronic “holes” from ruptured peroxy bonds in quartz, and subsequent surface positive charge from the ground, followed by light emission during neutralization. examination of video records shows the flash median length is about half a second and needs to be differentiated from the light during electricity supply short circuits. observed ground-level white colors appear to result from very intense non-specific air ionization. blue colors seem to be nitrogen emission with a short lifetime, succeeded by green oxygen emissions with longer lifetime followed by much lower intensity red. these were created by transient low-to-moderate voltages and probably include significant uv production. the maximum likely radiation dose is small, restricted to the skin, and equivalent at most to a few months natural background radiation. calculations confirm the release of 222rn is not the major mechanism for creating earthquake light, and would contribute minimal radiation dose. other unique observations are: streamers of light changing from blue to green as they passed from west to east, vertical sheets of blue-green light from cracks in an asphalt road surface, created by local shockwaves and shearing forces, daylight observation by fishermen of rapid linear undersea travel of blue light seconds before a 2011 m6.0 aftershock, and a mid-day green-blue glow over nearby hills containing a fault zone, shortly before the further destructive m6.3 earthquake, (12.51 pm 22-feb-2011). las luces de terremoto o eql (del inglés "earth quake light") son principalmente azul blancas y fueron cosísmicamente observadas durante un terremoto de magnitud 7.1 en nueva zelanda (4:47 a.m., 4 de septiembre de 2010, hora local). es muy probable que el fenómeno haya sido producto de un mecanismo local. las observaciones del destello azul blanco coinciden con el mecanismo de freund y otros, esto es, la alteración de un movimiento sísmico que crea "agujeros" electrónicos a partir de lazos de ácido peroxi rotos en cuarzos, con carga positiva superficial consecuente del suelo, y seguida por la emisión de luz durante la neutralización. la revisión de grabaciones de video muestra que la duración del destello es de cerca de medio segundo y debe ser diferenciado de la luz generada por los cortos circuitos del fluido eléctrico. los colores blancos observados a nivel del terreno parecen resultar de una intensa ionización aérea no específica. los colores azules serían emisiones de nitrógeno con un corto período de duración, sucedidas por emisiones de oxígeno verde de mayor duración y seguidas por un rojo mucho menos intenso. estos colores fueron creados por voltajes transitorios de bajo a moderados y probablemente incluyen una producción significativa de radiación ultravioleta. la cantidad de radiación máxima probable es pequeña, restringida a la piel, y equivalente a unos pocos meses de radiación natural regular. los cálculos confirman que la liberación de 222rn (radón) no es el principal mecanismo para crear las luces de terremoto y que este contribuye con una mínima cantidad de radiación. otras observaciones únicas hablan de rayos de luz que cambian de azul a verde mientras van de oeste a este, de cortinas verticales de luz azul verdosa que emergen de grietas en la carretera, creadas por movimientos sísmicos y fuerzas de corte; de observaciones de pescadores a luz del día del trayecto subacuático, rápido y lineal, de una luz azul antes del movimiento sísmico de magnitud 6.0 de 2011, y de un resplandor verde azuloso a mediodía sobre las colinas cercanas a una zona de fallas momentos antes del destructivo terremoto de magnitud 6.3 (12:51 p.m., 22 de febrero de 2011). abstract resumen keywords: : earthquake light, earthquake precursors. palabras clave: luces de terremoto, precursores de terremoto. origin of earthquake light associated with earthquakes in christchurch, new zealand, 2010-2011 record manuscript received: 02/11/2014 accepted for publication: 06/08/2015 how to cite item whitehead, n. e., and ulusoy, u. (2015). origin of earthquake light associated with earthquakes in christchurch, new zealand, 2010-2011. earth sciences research journal, 19(2), 113-120. http://dx.doi.org/10.15446/esrj.v19n2.47000 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v19n2.47000 n.e.whitehead1 and ű.ulusoy2 1corresponding author: whitehead associates, 54 redvers drive, lower hutt, new zealand 5010 64-4-5650263, whiteh@paradise.net.nz 2department of physics, engineering faculty of engineering, hacettepe university, 06800 beytepe, ankara, turkey origen de luces de terremoto asociadas con los terremotos de christchurch, nueva zelanda, 2010-2011 earth sciences research journal earth sci. res. j. vol. 19, no. 2 (december, 2015): 113 120 se is m o l o g y 114 n.e.whitehead and ű.ulusoy introduction eql is mainly flashes of light or glows associated with earthquakes, but because of its transient nature has been treated sceptically by scientists. following capture on film and webcam/security cameras, it is now accepted as real by most seismologists, as judged by the published literature referenced in this paper. there are, by our count, about half a dozen videos available on the internet, (e.g. an aftershock reported by the bbc in sendai, japan: http://www.youtube.com/watch?v=5hrgqlu3y_y, and in lima, peru: http://www.pucp.edu.pe/inras/peru-magneto/mod-3cam-3.exe). most show a white spherical ground-based core surrounded by radiating blue-white light, fading the further it is from the core. there are now thought to be numerous valid records throughout history (derr, 1973), (stothers, 2004), (thériault, st-laurent, freund, and derr, 2014), though the origin is still argued, (straser, 2007), (fidani, 2010), (fidani, 2012), (freund, 2011), and may be multiple. eql has been observed before in new zealand ((aimuri, hutton, 1888); (reefton, anonymous, 1888a; anonymous, 1888b); wellington, 1943, green aerial explosions over the sea (personal communication edith costello, 2012), (grayland, 1957); inangahua 1968, coseismic orange glow, less than 15 min (personal communication w. inwood, 2012)) the references give only superficial descriptions; locations are found in figures 1-3. however, we are still at an early stage, where many more observations should be collected and analysed. this paper attempts to add to the collection and supply some interpretation. two large earthquakes in christchurch, new zealand contributed four unique observations that support some explanations of eql origins in the literature, particularly work by freund et al. (2009), thériault et al. (2014). new zealand is earthquake prone because of its tectonic setting on the “ring of fire” (stirling, 2002). christchurch is a city of 377,000 people on the east coast of the south island of new zealand at 43º s (figure 1). an m7.0 event (4-sep-2010 at 4:35:47 am local time, 16:35:47 ut), had an epicenter 12 km deep, near darfield, 42 km west of christchurch, on the greendale fault (quigley et al., 2010). this main event produced most of the reported eql. there were many thousand aftershocks (bannister and gledhill, 2012). a later event triggered by the first earthquake (stramondo et al., 2011), was on 22-feb-2011, at 12:51:42 pm, local time, 23:51:42 ut, m6.1, 6 km deep (usgs, 2011) and very close to christchurch, on the lyttelton or port hills fault (increasing detail in figures 1, 2, 3). figure 1. location map of new zealand, with christchurch and faults. asterisks mark epicentres. underlined named locations are of eql, some historic. more detail in figures 2,3. figure 2. christchurch environs. black dots mark greendale fault (buried and detected mainly through seismology, with no published defined line at the time of writing). figure 3. christchurch localities. port hills fault zone in black dots (from seismology). black-type localities are those mentioned in text. observer positions asterisked. observations:. observer #1 (typical of many) blue flashes near airport; observer #2; blue/green curtain emissions from cracks in road; observer #3 submarine blue beams; observer #4 blue-green glow above port hills. the city is mostly on a broad river plain of sedimentary greywacke gravel and silt (suggate, 1958). the only igneous rocks are the 500m high port hills eroded remnants of a 10-12 ma-old basalt volcano (stipp and mcdougall, 1968), outside the city to the southeast, within which is a 15km long inlet from the northeast, parallel to the lyttelton/port hills fault (figure 3). the basalt is underlain by torlesse terrane, which is a new zealand mixed sedimentary and metamorphic rock (forsyth et al. 2008). method an appeal was made for earthquake experiences via a then-existing current affairs program (“close-up”) on national tv. the wording was “many people have reported events they think are associated with a later earthquake. for example from animals, birds, plants or in the ground, water, or the air, sky, clouds, electrical appliances or from other events. if you experienced something like this, please visit the website whose address is on the screen”. results were mostly collected on a website specializing in surveys called surveymonkey. for other details including data on animal response, see the parallel paper (whitehead and ulusoy, 2013). for some of the observations, obvious questions arise, but most participants did not 115origin of earthquake light associated with earthquakes in christchurch, new zealand, 2010-2011 supply email addresses for follow-up questions. the sample was obviously biased towards the mentioned categories, but many coseismic accounts were supplied most of which mentioned eql and were used in this paper. the sample will be biased towards the young and computer-literate, and is clearly not representative, but this will not matter for the phenomena described. consideration of reported daylight rainbow segments in clouds was deferred and is not included. results there were nearly 400 usable replies given to the survey, of which 42 mentioned eql for the september 2010 event. eql descriptions were also collected from an unrelated website specializing in meteorology. most people in the city saw brief blue/white coseismic flashes of light surrounding a white core, somewhere near the airport, and a minority, further away, reported a few red colors. the most intense light was a large white flash after tens of seconds of shaking, near the city, rather than at the epicenter 40 km distant. many similar reports were given for aftershocks. the light was captured on security cameras at the christchurch/lyttelton road tunnel (figure 3), princess margaret hospital, (central christchurch), and one less official video, incorporated in a subsequent film documentary, when a city falls, distributed by frank films (new zealand film commission, 2011) the latter available for time analysis, frame by frame. for four flashes, the median length was 0.12 sec (see discussion section). in the following eyewitness summaries, unique observations are marked by asterisk: 1. blue-white lightning from darfield, 5-10 seconds before p-waves at christchurch 2. co-seismic ground lightning near the west of christchurch, mostly blue-white flashes with final intense white flash 3. *mainly coseismic: rivulets of light on the ground, changing from blue to green 4. *mainly coseismic: vertical sheets of blue/green light from road surface (same event as 3). 5. *preseismic: submarine blue streaks of light 6. *preseismic: blue-green haze on hills in daylight (feb-2011 only). selected eyewitness accounts det ai l ed accoun ts are available from the first author or at: www.chchquake.co.nz in the following, similarly in appendix 1 with representative accounts, some spelling has been regularized. editorial comments are in square brackets. a typical description is first, then the four unusual observations. typical observation #1 the colour of the flashes was a blue-white, definitely white like lightning, but more like the sheet lightning we occasionally see in the mountains. there did not appear to be a specific source like a concentrated point or jagged lightning flash, more of a bright round glow in the sky. the position of the glow changed with each flash. our arc of view was south around to west and the variation was in this range. also the elevation would have been below 55 degrees…….. the duration of the flashes was confined to the period of the earthquake and the number would have been at least six and each lit up a large part of the sky. in other words each flash was not confined to a small circle……. unique observations a. sheets of light from cracks in the road/colour changes [this account by observer #2 (figure 3) is significant for both reported colour change and the sheets of light] i was driving into christchurch on the morning of the quake. [southwards] …..4.35am. the sky lit up with an incredible lighting display. i knew it was an aurora. to my left, east, …[ see figures 2 and 3] flashing dark green to light green as it moved away from me. to my right, west, light blues to dark blue. the display was ‘sheeting’. i have seen auroras on tv, and that is what it was. this was not lightning. my car clock clicked over 4.36am, just seconds into the aurora, and the car lurched and tipped and shuddered. the force was enough to have tipped my car over, but due to experience i managed to steer into the shudders. the lights of the motorway were still on. [i.e. this was not yet power-line arcing] ‘rivulets’, or wriggles, were flashing towards me from the south west, like a giant spider web, or ‘radials’, … i would guess that the centre of the spiderweb would have been …. […darfield direction, figure 2, but about 20 km further away]. two ‘rivulets’ flashed passed me on either side, straight into kaiapoi which was on my immediate left, east. the lighting display suddenly became incredibly intense, and was not ‘in the distance’, it was all around me. then with the violent shuddering of the car, there was a huge flash of brilliant white light, and the display vanished, and the whole city sank into blackness [the car functioning was not affected. kaiapoi (figure 3) is 17 km n. of christchurch.]. and further (recapitulation): the light display appeared first …. as the colours changed they moved away, or, as they moved away the colours changed. greens on my left, east, blues on my right, west. the brilliant white light was directly over the city. the range of the light extended right out to sea, east, into the whole distance i could see. i was ‘inside’ the light [and it was hugging the ground, about car height]. …if you could imagine a flat concertina. the radials [= rivulets, earlier] advanced across the road in front of me, from southwest to northeast, and they moved as fast as i could take in the spectacle. the radials were not uniform in that i saw some which appeared to be behind/ ahead of others. each radial, … was each, ker-bump, and a ker-bump, with me travelling south at 100ks per hour. on each side of each radial, the road moved in opposite directions, like a push-pull effect [shearing forces in the asphalt]. as i passed over each radial, the car lurched the other way. each radial looked strangely like the pictures of gophers burrowing along, where they leave a little upraised trail. but there were no open cracks as appeared later. i would guess that the motorway was swaying inside a channel of about one meter overall in the distance i could see. the lights of the motorway were still on until the last tremor. the lights of the city vanished basically simultaneously with the aurora. …., in this instance, the radials shot through like the main radials of a web, and they did not cut through at the same rate. some appeared in front of me, and i ran over them, and some appeared behind me. the light appeared above the ground in sheets. the lights were high in the sky, and also appeared above the cracks. i think one of your correspondents said the light came from multiple sources. i think the light came from the thousands of points where the ground cracked. the impression i have is that gas was forced from the ground, and ‘something’ caused it to appear in colour, …. the lowest point of the appearance of the light, would have been about level with my windsceen. it was fleeting, like a rainbow which disappears when one gets closer…. this correspondent also reported feeling for a few days as though he had been sunburned (discussed below). b. lyttelton submarine observation [from fisherman, observer #3, figure 3, 4 km from the port of lyttelton from the open sea, early afternoon, 11-july-20 11.] “i was looking out the front of the boat and i saw a couple of blue flashes down in the water, was definitely a flash, a dim blue flash. “the blue flashes were about 100 meters to the side of the vessel and shot from left to right in a straight line across us. then the next minute it was like the boat hit something.”[earthquake shock wave from a m6 116 n.e.whitehead and ű.ulusoy116 aftershock (gorman, 2011)]. the flashes were also described as rather milky blue, like the subaerial september events, but visible in daylight. there were two “rays” about 10 m apart which vanished into the distance. the depth was less than 12.4 m, (port company, 2013) and the surface is soft sediment. the time between the observation and the shock was a few seconds. the earthquake was m6.0 and 10 km under lyttelton, on the port hills fault zone . c. blue-green glow on hills observer #4 there was a blueish green glow in the sky just before the quake started [over the port hills to the southeast and above the fault zone. about 30 minutes before the m6.1 quake (12:51, 22-feb-2011), glow visible even though it was midday; no duration noted. weather cloudy]. discussion credibility of reports in the present paper, we have accounts of at least four unique and remarkable observations. however, there is no reason to doubt their content; observers gained little or nothing by reporting them, and in some cases were reluctant to report them because of their strangeness, until they realized that there was historical precedent. related events in the literature were unknown to the observers. most colors at christchurch had a white core (200m?) fading to blue-white further out, as described in the literature (briggs, 1999), (freund, 2003), (ikeya, 2004), with a few reports of green and red (only in the sky). other colors figured much less prominently. in general for the m7.0 quake the bluer colors were reported as being towards darfield, the epicenter, any green tending to be well away, and possibly over the sea to the east. from eyewitness reports, there were perhaps as many as six blue flashes, culminating in an intense white flash of large but uncertain extent (probably a larger version of the blue-white flashes), then a general electricity shutdown. another christchurch report was of a coseismic flash color like tungsten (incandescent) lighting. there were also reported preseismic glows, perhaps hours long, mostly on hills at night, as reported in the previous literature for other earthquakes (tsukuda, 1997). as for many new zealand earthquakes, there was a train-like rumble heard well before the first p-wave, and even before the earthquake origin (figure 4) which is puzzling. this also happened before aftershocks. this paper endorses a previous explanation (gordienko et al., 2008); pre-seismic rumble may originate from very numerous microcracks produced by pre-earthquake strain over a large area. video evidence will be biased in favor of recording the blue-white flashes because they have the highest intensity. glows are much less intense and will require long exposures to register (about 30 seconds, see long (2014) for similar difficulties with aurora detection). neither video facilities nor security cameras will record them, so eyewitness accounts will continue to be important. difficulties in differentiation from transformer arcing the christchurch median flash length was 0.12 s from a frameby-frame analysis, using the 7 movie maker program, compared with similarly derived medians of 0.56 s for an event in lima, peru (hyperlink in heraud and lira (2011)) and 0.22 s for sendai (bbc, 2011). chile data (taylerburdeen, 2010) appears similar. all are much longer than conventional lightning (10-50 µs) and a good rule of thumb is about 0.5 s. since some earthquake lights are several seconds in length, they are perhaps best described as short glows. for luenser (2013), and anonymous (2013) which were superficially similar, but were known to be electrical flashes rather than earthquake light, the analysis gave respective medians of 0.43 s (115 points) and 0.89 s (12 points). using montecarlo statistical methods (e.g. robert and casella, 2005) it was shown the electrical and genuinely earthquake data could not be distinguished. the possibility of transformer arcing must, therefore, be checked for all assertions of earthquake light observations. for the september quake in christchurch (see figure 4), the electrical distribution network was operating a logger, monitoring line voltage. collating eye-witness accounts, the eql appeared at least 10 seconds before the slightest preliminary line voltage fluctuations and were therefore not power supply arcing. no destroyed transformers were reported, which implies no destructive arcing during the whole episode. according to two good independent eyewitness accounts, flashes of light appeared 5-10 seconds in darfield before the first main tremors (p-waves) in christchurch. these flashes continued, but much nearer the city, until the power failure, and strong white flash (again much nearer christchurch than darfield). this confirms that eql may be at least tens of kilometers distant from an epicenter. figure 4. the timing of reported events at christchurch city, september 4, 2010. the epicenter was near darfield, depth 12 km, and 41.9 km distant. the accelerometer trace is from an instrument in the basement of the christchurch westpac trust building (gns, 2011), voltage changes and power failure are from logger data. the eql were coseismic, or a few seconds later than the first p-waves both for darfield and christchurch. for the strongest white flash, it could be a coincidence that this immediately preceded the power shut-down, but the simplest hypothesis is that an intense electromagnetic pulse contributed. more importantly this figure confirms some eql need not be associated with voltage fluctuations in power lines, or arcing. ionization involved. this paper suggests blue, green and red colors arise from ionization levels in nitrogen and oxygen which are relatively long-lived and remain, though only for a few seconds, after the less specific ionizations have disappeared. the white core of the “lightning” looks identical to white air discharges caused by arcing, or malfunctioning in high voltage electrical sources (usually accompanied by some ultraviolet). this is non-specific ionization of air molecules. individual colors are harder to explain, but this paper suggests they are from ionization events visible only because they are relatively long-lived. the blue colors at atmospheric pressure are known to arise naturally through de-excitation of nitrogen molecules, n2 +1, to neutral nitrogen molecules (wiberg, wiberg, and holleman, 2001) but have a lifetime of tens of nanoseconds only. the rarer green colors are due to ionization of oxygen and have a life of about 0.75 s, following which they are in a low but still excited energy level (wiberg et al., 2001). see also gartlein, kimball, and sprague, (1965). as oxygen passes to its lowest energy level with a very long 110 second lifetime, it emits red light. these data may be found in various standard compilations (e.g. wiese and martin, 1986). the colors are also seen in auroras (sorokin and chmyrev, 2014) but at ground level one would need a very intense primary ionization event of perhaps many tens of kilowatts per square meter. could this give a significant radiation dose? radiation dose? the eql themselves must correspond in energy terms only to ev, in the visible to ultraviolet region, and would deliver a dose only to the skin. earth movement could not create higher energy radiation such as gamma rays. observer #2 entirely surrounded by the white flash reported feeling sunburnt, possible if the blue-white flashes had strong uv output. ionized air has a 117origin of earthquake light associated with earthquakes in christchurch, new zealand, 2010-2011 strong uv emission component from atomic transitions in ionized nitrogen (anonymous, 2012), and electric arc welders must use uv protection. conventionally (unscear, 1993), the penetrating radiation dose from inhalation of radon is 1 msv y-1. if we assume that excess radon is released by pre-quake stresses and reaches 10x normal for 10 days before the quake (e.g. igarashi et al., 1995), the dose might be 10/365*1 or 0.027 msv. this is slight compared with the normal yearly radon exposure, and both these doses are negligible in exposure radiation terms. in contrast, normal levels of radon in cities caused through common atmospheric meteorological inversions are about 10 times higher than those likely from earthquakes (ikeya and whitehead, 2013). different analytical treatment, (freund, 2011) also concludes radon is not a factor in eql. thus, although ionizing events are occurring, the radiation dose is not critical. associated electrical phenomena during the week before 4-sep, static electricity was frequently reported, with local tv interference and other electronic glitches (whitehead and ulusoy, 2013). before and after the 22-feb earthquake, one respondent reported 5 minutes of local radio/tv static, though the broadcast signal was unchanged and continuous (personal communication nztv1). another respondent reported this static for 4-sep-2010, 22feb-2011, and 24-mar-2012 quakes. this has been reported for other earthquakes (wadatsumi, 1995), (st-laurent, 2000), (nagao et al., 2002), implying electromagnetic disturbance. however most importantly no thunder ever accompanied the eql, implying quite low voltages compared with conventional lightning, i.e. much less than 30kv cm-1 (lowke, 1992). coseismic dimming and brightening of domestic lights was reported, and noted in available video material, which must be very local changes of electric transmission line voltages, and it seems at least some of this was before the line-voltage fluctuations recorded with the logger. electric appliances not blown (observer #6), suggests a maximum of several thousand kilowatts power variation associated with the earthquake light. otherwise, the common 50 amp domestic electric circuit breakers would be activated. this means electromagnetic effects were involved, whether primary or secondary and were possible at christchurch. this erratic dimming has been previously recorded for incandescent light (terada, 1931). he also reported voltage fluctuations in a submarine cable during the idu earthquake. mechanisms unusual auroral activity was absent. the account from observer #2 described light that occurred on asphalt and seems to have been generated by shearing forces in the asphalt itself or more probably from directly below the cracks. notable is the description of physical forces on the car, coincident with the light, which does not seem to have been recorded elsewhere. this suggests very local seismic forces are strongly associated with these eql. although the previous literature suggests eql are much more frequent near the igneous rock, the present observations show it can occur in asphalt, sedimentary rocks and unconsolidated gravels. however like igneous rocks, the gravels contain small crystals of quartz (e.g. beggs, 1980) which is important for the most likely mechanism suggested below. the furthest observation (coseismic) was 65km north in the hills and above the hills of scargill valley (figure 2), which again means local shockwaves or strain can generate this light. most proposed mechanisms for the creation of high voltage potentials, and emission of ions apply only at the fault; (terada, 1931; lockner, johnston, and byerlee 1983; hedervari and noszticzius 1985). a piezoelectric origin for more distant influence has been considered (derr, 1973; takaki and ikeya, 1998; ikeya, 2004). the theory proposed slightly aligned quartz grains subjected to slowly varying pressure give rise to ulf waves that are one of the very few physical influences that can reach the surface from a deeply buried epicenter. at the surface, electrical fields were produced by these waves, and caused acceleration of local free electrons, creating the light, because ulf itself does not cause ionization. however (st-laurent, derr, and freund, 2006), there are major difficulties with the model, i.e. two mirrorimage forms of quartz will always occur in a 1:1 ratio and their piezoelectric effects cancel when there are many small crystals. laboratory experiments (bishop, 1981; sasaoka et al., 1998), showed that rocks containing quartz gave a weak piezoelectric-like effect proportional to stress. since under their test conditions, other rocks did not, their conclusion was that quartz was somehow involved, but the following work suggests that they were observing a unique mechanism in quartz that is called here “freundtype”. this model (e.g. freund et al., 2009; st-laurent et al., 2006) is based on extensive laboratory experiments in which high pressure on gabbro and diorite produced a flow of current to the surface due to rapid movement of positively charged “holes” in a quartz matrix from breaking peroxy-bonds that are a natural trace impurity (the balancing excess electrons produced do not move). the speed of movement of the holes may be 200 m sec-1. currents were calculated to be in the range 5-100ka per km-2, and many thousand cubic kilometers of rock are probably involved in an earthquake. it is calculated that this creates an intense positive electric field at the surface, air breakdown/positive ionization, then dissipation of the charge (e.g. freund (2011) and references therein). the recombination could give rise to the blue/white flashes observed, and uv. this model has experimental evidence from freund for the creation of these voltages even in sand and soil, so this mechanism could, therefore, apply to the christchurch gravels. it requires a return path for the positive charge to the trapped electrons under the ground, and the flashes in the air are part of this. why was the largest white flash in the city itself, not at the fault, and 30 seconds into the decreasing shaking? this could be because of the severe soil liquefaction in many prone areas in the city, providing a sudden, aqueous, highly conducting return path for electrical charge. individual light sources at christchurch may have been 200 m in diameter, but if they were apparently generated 40 km away in darfield, the implied very narrow beam at the epicenter seems quite improbable. most physical mechanisms should have created wide-beam light spreading from the epicenter, particularly if it is ulf. much more plausible is local stress in christchurch creating local light sources. ulf with its widespread nature may be a better though indirect explanation for long-lived diffuse glows in the sky. ground forces in the “ground aurora” accounts like the above “ground aurora” of observer #2, exist (terada, 1931), (derr, 1973), (tsukuda, 1997), (st-laurent et al., 2006), (fidani, 2010) but mostly with scanty detail or with “streamers”, rather than curtains. a closer parallel is a quebec observation (st-laurent, 2000). the colors produced at christchurch had perhaps a few seconds lifetime and possibly a height of tens of meters, which could be consistent with other literature (st-laurent et al., 2006). the sheets of light emerged from slight ground cracks, like a curtain. this sounds very much like the freund mechanism and may be an example of it. the observer was in the middle of the white flash and not burnt or shocked electrically, but the faraday cage effect from the car body makes magnitude estimation difficult. the lyttelton sub-sea observation at lyttelton, there was an appearance of light before the main shock. in this case, initiation was 10 km deep, but the resulting blue streaks were only 10 m apart, and at most several meters thick–perhaps a few percent of subaerial heights. the effect producing light has apparently penetrated upward through several meters of soft sediment. the milky-blue color again suggests ionization of nitrogen; the solubility of about 20 cm3 l-1 is not impossibly low compared with the 800 cm3 l-1 in the air. however a liquid medium would lead to very brief light compared to air, and since the light was a few seconds before the shockwave, unlike for the blue-white flashes, another mechanism may be responsible. there are other descriptions of marine light: (derr, 1973; kilbourne 118 n.e.whitehead and ű.ulusoy and saucedo, 1981; tsukuda, 1997; soter, 1999; stothers, 2004; ulusoy and ikeya, 2008; granin et al., 2012) differing from each other and with little detail. (heraud and lira, 2011) described surface bluish-white vertical pillars of light from low-lying rocky islets but generation under the sea seems less commonly reported. blue-green haze on hills. the reported color corresponds with ionization of oxygen and nitrogen giving blue-green light. longer-lasting pre-seismic night glows on or above hills have been reported for various earthquakes (e.g. heraud & lira, 2011). if these are coronal lasting glows they would depend on an unprecedentedly high dielectric constant in the earth at these extremely low frequencies (i.e. ulf), and this is uncertain, although the dielectric constant does increase very markedly as frequency decreases (lockner and byerlee, 1985; ikeya, 2004). glow characteristics are so different from the flash-type eql that the mechanism is probably different. the hills supporting the glow were basalt (low quartz), so the underlying torlesse terrane sedimentary/metamorphic rock could be more likely as a source. utility for pre-quake warning? the most possibly useful premonitory earthquake light was diffuse glows on hills the night before the september event giving perhaps several hours warning. unfortunately, this is a night phenomenon and relies on informed observers who are rare at present. however following such an apparent precursor it could be prudent to avoid old and glass-fronted buildings for a day or so. this needs many confirming instances before it is conclusively known to be useful. conclusions the present observations show four types of observation, production of light in a flash on the surface of the ground, change of rivulet/ray color from blue to green, production of light by earthquake impulses, leading to vertical curtains of light above shearing zones, and production without the impulses but due to moderate voltage phenomena which precede p-waves, and may be attributable to local preseismic strain. eql does seem to involve ionization of nitrogen/oxygen which accounts for the colors seen. the actual dose of ionizing radiation is calculated to be very low. there may be some uv emission associated with the flashes, but electronic interference appears to be mainly radio/tv static. possible mechanisms supported include the freund model, particularly for ground-origin light, and ulf for glows in the air. general non-specific rather low voltage ionization gives a white color in the air and corresponding to atomic excited lifetimes, blue, green and red colors follow. preseismic rumbles are consistent with the generation of preseismic strain. acknowledgements sincere thanks to respondents to the survey for their accounts of eql, and orion network personnel for the voltage-logger data and discussion. gerard smythe, for video analysis permission. alan wright for a detailed account of driving through the eql, numerous correspondents for other historical accounts within new zealand, particularly edith costello, dr. martin reyners, warren inwood. thanks to gary vallance for his account of the undersea light, and gps coordinates. thanks to dr. kassym zhumadilov for some tests and discussion, mrs. b.k. whitehead for graphics assistance, dr. y. ikeya for assistance with the japanese literature, mr. d. chord for assistance with copying a christchurch video eql segment. references anonymous. (1888a). telegrams. inangahua. times 13 (31 december): 2. http:// paperspast.natlib.govt.nz/cgi-bin/paperspast?a=d&d=it18881231.2.9 (last accessed october 2014) anonymous. (1888b). the supposed lights of hanmer plain. press (christchurch) 45: (september 7) 5. http://paperspast.natlib.govt. nz/cgi-bin/paperspast?a=d&d=chp18880907.2.32 (last accessed october 2014) anonymous. (2012, june 18). ionized-air glow. retrieved from http:// en.wikipedia.org/wiki/ionized-air_glow (last accessed march 2015) anonymous. (2013). mexico: mysterious lights appear in the sky during the june 16 earthquake. youtube. http://www.desafio2020fpf.com/actualidad/ otras-noticias/mexico-aparecen-misteriosas-luces-cielo-durantetemblor-16-junio-video-noticia-146118 (last accessed march 2015) bannister, s., and gledhill, k. (2012). evolution of the 2010–2012 canterbury earthquake sequence. new zealand journal of geology and geophysics, 55, no. 3, 295–304. doi:10.1080/00288306.2012.680475 beggs, j. m. (1980). sedimentology and paleogeography of some kaihikuan torlesse rocks in mid canterbury. new zealand journal of geology and geophysics, 23, no. 4, 439–445. doi:10.1080/00288306.1980.10424114 briggs, j. w. (1999). earthquake light above mt wilson observatory. astronomy and geophysics, 40, 2.9. derr, j. s. (1973). earthquake lights: a review of observations and present theories. bulletin of the seismological society of america, 63, no.6, 2177–2187. fidani, c. (2010). the earthquake lights (eql) of the 6 april 2009 aquila earthquake, in central italy. natural hazards and earth system sciences, 10, no. 5, 967–978. doi:10.5194/nhess-10-967-2010 fidani, c. (2012). statistical and spectral properties of the l’aquila eql in 2009. bollettino di geofisica teorica et applicata, 1, 135–146. doi:10.4430/bgta0034 forsyth, p. j., barrell, d. j., and jongens, r. (2008). geology of the christchurch area, (76pp). lower hutt, new zealand: gns science. freund, f. (2011). pre-earthquake signals: underlying physical processes. journal of asian earth sciences, 41, no.4-5, 383–400. doi:10.1016/j. jseaes.2010.03.009 freund, f. t. (2003). rocks that crackle and sparkle and glow: strange pre-earthquake phenomena. journal of scientific exploration, 17, no. 1, 37–71. freund, f. t., kulahci, i. g., cyr, g., ling, j., winnicke, m., tregloan-reed, j., and freund, m. m. (2009). air ionization at rock surfaces and pre-earthquake signals. journal of atmospheric and solar-terrestrial physics, 71, no. 17-18, 1824–1834. doi:10.1016/j.jastp.2009.07.013 gartlein, c. w., kimball, d. s., and sprague, g. c. (1965). the origin and morphology of the aurora (134pp). cornell: physics department, cornell university. gordienko, v. a., gordienko, t. v., krasnopistsev, n. v., kuptsov, a. v., larionov, i. a., marapulets, y. v., ruteno, a.n., and shevtsov, b. m. (2008). anomaly in high-frequency geoacoustic emission as a close earthquake precursor. acoustical physics, 54, no. 1, 82–93. doi:10.1134/s1063771008010120 gorman, p. (2011). fishermen see flashes of light as quake hits port. the press (christchurch), 11-july. pa14. grayland, e. c. (1957). new zealand disasters (208pp). wellington: a.h. & a.w. reed. heraud, j. a., and lira, j. a. (2011). co-seismic luminescence in lima, 150 km from the pisco, peru earthquake of 15 august 2007. natural hazards and earth system sciences, 11, 1025–1036. doi:http://dx.doi. org/10.5194/nhess-11-1025-2011 hutton, f. w. (1888). art. xxxii. the earthquake in the aimuri. transaction and proceedings of the royal society of new zealand, 21, 269–293. igarashi, g., saeki, s., takahata, n., sumikawa, k., tasaka, s., sasaki, y., takahashi, m., and sano, y. (1995). ground-water radon anomaly before the kobe earthquake in japan. science, 269, no. 5520, 60–61. doi:10.1126/science.269.5220.60 119origin of earthquake light associated with earthquakes in christchurch, new zealand, 2010-2011 ikeya, m. (2004). earthquakes and animals (295pp). singapore: world scientific. doi:http://dx.doi.org/10.1142/9789812832696 ikeya, m., and whitehead, n. e. (2013). unusual childhood waking as a possible precursor of the 1995 kobe earthquake. animals, 3, no. 1, 228–237. doi:10.3390/ani3010228 long, a. (2014). photographing the aurora borealis (140pp). lakewood, co: first light photography llc. lowke, j. j. (1992). theory of electrical breakdown in air. journal of physics d: applied physics, 25, no. 2, 202–210. doi:10.1088/0022-3727 luenser, b. (2013). fortworth lights. youtube. retrieved from http://www. youtube.com/watch?v=nychbi66izs (last accessed march 2015 ) lyttelton port company. (2013). continuing maintenance dredging operations at lyttelton port of christchurch. application for resource consent and assessment of environmental effects. volume 2 (84pp). christchurch, new zealand. retrieved from http://www.lpc.co.nz/ tempfiles/tempdocuments/other/maintenance dredging/2 aee. pdf, (last accessed march 2015 ) nagao, t., enomoto, y., fujinawa, y., hata, m., hayakawa, m., huang, q., izutsu, j., kushida, y., maeda, k., oike, k., and yoshino, t. (2002). electromagnetic anomalies associated with the 1995 kobe earthquake. journal of geodynamics, 33, 401–411. new zealand film commission (2011) when a city falls the people's story. http:// www.nzfilm.co.nz/film/when-a-city-falls (last accessed march 2015) quigley, m., villamor, p., furlong, k., beavan, j., van dissen, r., litchfield, n., stahl, t., duffy, b., bilderback, e., noble, d., barrell, d., jongens, r., and cox, s. (2010). previously unknown fault shakes new zealand’s south island. eos, transactions american geophysical union, 91, no. 49, 469. doi:10.1029/2010eo490001 robert, c., and casella, g. (2005). monte carlo statistical methods. (649pp) new york: springer. sorokin, v. m., and chmyrev, v. m. (2014). atmosphere-ionosphere electrodynamic coupling. in v. l. bychkov, g. v. golubkov, and a. i. nikitin (eds.), the atmosphere and ionosphere (pp. 97–146). heidelberg: springer. stipp, j. j., and mcdougall, i. (1968). geochronology of the banks peninsula volcanoes, new zealand. new zealand journal of geology and geophysics, 11, no. 5, 1239–1258. doi:10.1080/00288306.1968.10420260 stirling, m. w. (2002). a new seismic hazard model for new zealand. bulletin of the seismological society of america, 92, no. 5, 1878– 1903. doi:10.1785/0120010156 st-laurent, f. (2000). the saguenay, quebec, earthquake lights of november 1988-january 1989: a comparative study with reference to the geoatmospheric lights classification proposed by montandon in 1948, and a description put forward by yasui in 1968. seismological research letters, 71, no. 2, 160–174. doi:10.1785/gssrl.71.2.160 st-laurent, f., derr, j. s., and freund, f. t. (2006). earthquake lights and the stress-activation of positive hole charge carriers in rocks. physics and chemistry of the earth, a/b/c, 31, no. 4-9, 305–312. doi:10.1016/j.pce.2006.02.003 stothers, r. b. (2004). ancient and modern earthquake lights in northwestern turkey. seismological research letters, 75, no. 2, 199–204. doi:10.1785/gssrl.75.2.199 stramondo, s., kyriakopoulos, c., bignami, c., chini, m., melini, d., moro, m., picchiani, m., saroli, m. and boschi, e. (2011). did the september 2010 (darfield) earthquake trigger the february 2011 (christchurch) event? scientific reports, 1, 98. doi:10.1038/srep00098 straser, v. (2007). precursory luminous phenomena used for earthquake prediction the taro valley, northwest appenines, italy. new concepts in global tectonics, no. 44, 18–32. suggate, r. p. (1958). late quaternary deposits of the christchurch metropolitan area. new zealand journal of geology and geophysics, 1, no. 1, 103–122. doi:10.1080/00288306.1958.10422799 taylerburdeen. (2010). terremoto santiago chile 2010 sobre el mirador explosiones de luz. youtube. retrieved from https://www.youtube.com/watch?v=inm8qy0cewu (last accessed march 2015 ) terada, t. (1931). on luminous phenomena accompanying earthquakes. bulletin of the earthquake research institute, tokyo university., 9, no. 3, 225–255. thériault, r., st-laurent, f., freund, f. t., and derr, j. s. (2014). prevalence of earthquake lights associated with rift environments. seismological research letters, 85, no. 1, 159–178. doi:http:// dx.doi.org/10.1785/0220130059 tsukuda, t. (1997). sizes and some features of luminous sources associated with the 1995 hyogo-ken nambu earthquake. journal of physics of the earth, 45, no. 2, 73–82. doi:10.4294/jpe1952.45.73 usgs. (2011). m6.1 south island of new zealand. http://comcat.cr.usgs. gov/earthquakes/eventpage/pde20110221235142350_5#summary (last accessed february 2015 ) wadatsumi, k. (1995). 1519 statements on precursors. (265pp). tokyo: tokyo publishing. whitehead, n. e., & ulusoy, u. (2013). macroscopic earthquake precursors before the september 2010 m7.1 event in christchurch, new zealand. natural hazards and earth system sciences, 13, 167–176. doi:http://dx.doi.org/10.5194/nhess-13-167-2013 wiberg, e., wiberg, n., & holleman, a. f. (2001). inorganic chemistry (1656pp). berlin: associated press. appendix 1 contemporary new zealand eql reports #1. driving to chch [christchurch] from darfield at 9.30pm on friday 03/09/10 [3september-2010, the night of the quake, but about 7 hours before]. saw a strange halo/arc of light over the city of chch. it was blueish i guess. don’t know how long it went on as i was concentrating on driving… [the location of this halo is ambiguous. although, over the city, this could include the port hills.] #2. [rolleston 6-9pm 3september-2010] blue light [seen] towards darfield [this was for an extended time, at least many minutes, not a flash]. #3. i have seen exactly that, [blue light flashes during the september quake] at least 3 all in about 2 kilometers of [from?] each other and maybe 5 kilometers west of parklands [figure 3, tens of km from darfield]. i was deciding whether i should jump out the window when i looked up from the first bright flash and observed the next two, one of which i swore had an arc of lightning leading to it. [this must be forked or arc lighting, rather than sheet lightning] … all the light came from the ground. the one i got a good look at made an amazing amount of light in the form of a sphere on the ground. #4. …there were no clouds, and the light seemed to originate from the ground. we live on dyers pass [eastern edge of the city] and could see flash after flash in the direction of the airport [west of the city and several kilometers away]. #5 i definitely saw the lights (looked a lot like lightning) … we live in broadoaks, cashmere [eastern edge of the city] and what i saw was a blue-red color flash backwards & forwards across the sky. #6 we live in redcliffs [eastern edge of city] and also saw the bluish-white flashes in the sky … what we couldn't understand though is that our lights also flashed like crazy with blue/white flashes as if it was a power surge but all our appliances are ok, which would indicate that wasn't the case at all? #7 the other thing we observed on the cctv footage was what appeared to be a type of lightning (earthquake lightning)? only it appeared to be coming out of the ground. now that sounds weird i know, but members of the police on duty that night also reported it. [in the new zealand context this would more usually mean sheet lightning rather than forked lightning.] #8 [church bay, 50 km south from darfield]. as the earthquake began the sky was lit up with bright blue flashing light... similar to sheet lightning. it appeared 'streaky' like the sun shining through clouds #9 the other scary occurrence i would like to mention is, i awoke before the shaking, to the sound of what i can only describe as 'amplified static 120 n.e.whitehead and ű.ulusoy electricity' coming from my roof (it is iron), which at the time i retorted to my wife 'jeez, what's that, hail?'. we then jumped out of bed and heard the 'train like rumble' people often now compare the eq noise to. my son, downstairs, had his curtains drawn and was awake prior to the quake, he reported seeing blue flashes of light in the sky prior to the quake. the earthquake light was also seen associated with the aftershocks: #10 on saturday night of the big quake my wife and i spoke with a work friend who said he was out driving at 4:35 am when he saw numerous blue flashes in the sky. i was skeptical about what he had seen until the following night. my wife and i were driving back from blenheim home to christchurch [southwards] when just south of amberley [60km n of christchurch] we both saw two blue flashes in the sky ahead of us. one slightly to the left and the later one to the right. a definitive earthquake phenomenon in our opinion. #11 purple light glow the night before 4-september. another witness reported coseismic near-indigo color towards the east, and another, coseismic purple clouds. an inverse mechanics model has been built under work face mining conditions, according to the law of working surface pressure distribution, to explore the mechanical characteristics and stability of the reverse fault under the influence of mining. as a result, a theoretical calculation equation of the normal and shear stresses in the fault zone has been deduced to obtain the stress variation rule between the working surface and the fault layer, under distance conditions of 10, 30, 50, and 70m. with distance conditions of 10 and 30m, the working surface mining stress had a noticeable effect on the reverse fault, resulting in a changing trend of firstly increasing, then decreasing, and increasing again in the normal and shear stresses of the fault zone as a whole. with distance conditions of 50 and 70m, the working face mining stress had little effect on the reverse fault; furthermore, the normal and shear stresses exhibited a changing trend of gradually increasing. at a later stage, a simulation of the above distance plans was conducted using the flac3d numerical simulation software. the results demonstrated that the influence range of the mining stress on the working face under the spacing distances of 10 and 30m included the fault zone, while under the distance conditions of 50 and 70m, the fault zone was excluded. on this basis, the fault zone stability was analyzed under four types of spacing conditions using the mohr-coulomb theory rule and fault activation determination. it is concluded that the fault zone stability was high, with increasing distances between the working face and fault zone. the least sufficient stability was located near the working face, where the fault zone stability was so weak that it is likely to result in impact fracture. abstract keywords: reverse fault; normal stress; shear stress; fault instability; rock burst; analysis of mechanical characteristics and instability law of fault under the influence of mining issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n2.72249 yuanhui li1,rui zhou1,2* 1key laboratory of ministry of education on safe mining of deep metal mines, northeastern university, shenyang 110819, p.r. china. 2china coal technology engineering group, shenyang research institute, shenyang 110016, p.r. china. * email of corresponding author: 3416197979@qq.com earth sciences research journal earth sci. res. j. vol. 22, no. 2 (june, 2018): 139-144 se construyó un modelo mecánico inverso bajo las condiciones de un frente de trabajo minero, de acuerdo con las condiciones de distribución de presión en la superficie de operación, para explorar las características mecánicas y la estabilidad de la falla inversa en estas áreas de explotación minera. como resultado, se dedujo una ecuación teórica para calcular el esfuerzo normal y el esfuerzo cortante en la zona de falla y con el fin de obtener el régimen de variación de esfuerzos entre la superficie de trabajo y la capa de la falla a distancias de 10, 30, 50 y 70 metros. a 10 y 30 metros de distancia, el esfuerzo en la zona de trabajo tiene un efecto notorio en la falla inversa, lo que significa un cambio en la tendencia de los esfuerzos normal y cortante, que primero se incrementan, decrecen y nuevamente aumentan en toda la zona de la falla. a 50 y 70 metros, el esfuerzo en el frente de trabajo tiene poco efecto en la falla inversa; además, los esfuerzos normal y cortante mostraron una tendencia cambiante de crecimiento gradual. por último, se realizó una simulación en cada una de las distancias con el software de simulación numérica flac3d. los resultados demuestran que el rango de influencia del esfuerzo en el frente de trabajo a distancia de 10 y 30 metros incluyen la zona de falla, mientras que a 50 y 70 metros la zona de falla está excluida. de acuerdo con estos resultados, se analizó la estabilidad de la zona de falla en cuatro condiciones de espacio con la teoría de mohr-coulomb y la determinación de activación de falla. se concluye que la estabilidad en la zona de falla es mayor con el incremento de las distancias entre el frente de trabajo y la zona de falla. el menor punto de estabilidad suficiente se ubicó cerca del frente de trabajo, donde la estabilidad de la zona de falla es tan baja que podría derivar en una fractura de impacto. resumen palabras clave: falla inversa; esfuerzo normal; esfuerzo cortante; inestabilidad de falla; explosión rocosa. record manuscript received: 14/11/2017 accepted for publication: 11/06/2018 how to cite item li, y., & zhou, r. (2018). analysis of mechanical characteristics and instability law of fault under the influence of mining. earth sciences research journal, 22(2), 139-144. doi: http://dx.doi.org/10.15446/esrj.v22n2.72249 análisis de las características mecánicas y las condiciones de inestabilidad de fallas en áreas de explotación minera m in in g e n g in e e r in g 140 yuanhui li, rui zhou 1 introduction coal reserves of 1.48 trillion tons have been identified in china, ranking third in the world. however, coal seams are buried deeply, with many folds, faults, and other geological structures, particularly in the tectonic region of fault. the damaged continuity of coal and rock seams poses a challenge to coal mining (tan et al., 2011; xiao et al., 2016; sufiyan et al., 2018). moreover, fault instability generated by the coal mining influence in areas affected by the fault structure may result in disasters such as rock bursts (li et al., 2014; rahim et al., 2018; mahmood et al., 2018), water inrush through the fault (lamoreaux, wu, & zhou, 2014; khanchoul et al., 2018), and coal and gas outbursts (díaz-aguado, & gonzález-nicieza, 2007). the changing law of mining mechanics under the influence of fault has always been the focus of rock burst preve, particularly in the prevention and control of rock burst. gibowicz (1984) and mcgarr (1984) determined that an ore quake is similar to a natural earthquake source mechanism, and shear slip in the fault zone mainly causes large-energy mine earthquakes. the majority of theories regarding the seismic source mechanism can be used for mine earthquakes. joughin and jager (1984), potgieter and roering (1984) discovered that large-energy mine earthquakes are closely related to shear slip in their study of a mine earthquake that occurred in a particular south african gold mine. chinese scholars jiang and liu (2010) proposed a correlation between different fault types and rock burst occurrences. these are more likely to occur in the pressure bump when the working surface is overtaken by the reverse fault, according to data prepared by jiang et al. (2014). it is concluded that a rock burst induced by the reverse fault is more likely to accumulate a significant amount of elastic energy with strong destruction, upon conducting a statistical analysis of a rock burst disaster induced by the reverse fault. through micro-seismic, electromagnetic radiation and other equipment, it is possible to monitor the fact that a mechanical indication in some form occurs before the rock burst induced by the reverse fault, and such that evidence serves as identification of fault instability (veeraragavan et al., 2018; nwankwoala & ememu, 2018). although various scholars have studied the phenomenon, the majority of studies have focused mainly on field monitoring data analysis. less analysis has carried out on the theoretical causes of this mechanical indication, and the changing law of the standard and shear stresses causing the evidence in the fault zone. therefore, according to the actual geological characteristics of the mining face under the influence of a mining area’s reverse fault, the distribution law of the supporting pressure in the coal seam direction is analyzed in this paper (sudhakaran et al., 2018; guan leong et al., 2018). moreover, a mechanics model of the stope has been established under the reverse fault to explore the pressure changing laws and instability of the reverse fault under the mining influence, providing a reference for the prevention and control of the pressure bump induced by the reverse mistake (wali et al., 2018). 2. materials and methods 2.1. mechanics model of the working surface tending to the reverse fault the tendency of the mining working face of the coal seam in a mining area to be affected by the f16 reverse fault was evaluated. the thickness of the coal seam is 5m; the upper roof of the coal seam is a conglomerate of 190m in diameter, and the fault angle is 60°. after the mining face, the overlying strata are collapsed. assuming that the stratigraphic coal and rock mass are an isotropic elastic body, the top plate plane would be taken above the working surface. the pressure distribution of the overlying strata trending along the working surface is shown in fig. 1 (qian, 2003; thiruchelvam et al., 2018; tair & dell, 2018). the pressure of the -∞ to a and i to + ∞ zones indicate the stress of the original rock γh. ai is the stress affecting the working face zone, where ab, bc, fg, and gi are the stress increasing zones (simplified as a linear increase). furthermore, cd, de, and ef are the stress reduction zones (simplified as a linear decrease). the intersection of the fault and the roof plane is the origin o, while the angle with the y-axis is the dip angle θ of the fault (asghar et al., 2018). the working surface is at point c, the working face stress concentration coefficient is k1, and the working face stress reduction coefficient is k2. according to elastic mechanics theory, the stress of a point m (y, x) on the fault can be obtained as follows: figure 1. pressure distribution along working face where σx is the tension of the x-direction in the fault zone, mpa; σy is the stress of the y-direction in the fault zone; τxy is the shear stress of the fault zone, mpa; γ is the rock mass density, kg/m3; h is the buried depth of the working face roof, m; ξ is a variable; g is the gravity acceleration, n/kg; and k1 and k2 are the stress concentration and reduction coefficient, respectively. the specific meanings of the length units a, b, c, d, e, f, g, and i are shown in fig. 1. the usual and shear stresses of the fault zone can be calculated as follows: (4) (5) 141analysis of mechanical characteristics and instability law of fault under the influence of mining table 1. parameter selection figure 2. normal stress distribution of the fault zone under different spacing distances figure 4. numerical model table 2. rock mechanical parameters figure 3. shear stress distribution of the fault zone under different spacing distances where σn is the general stress of the fault zone; τn is the shear stress of the fault zone, mpa; and θ is the dip angle, °. because it is difficult to find an analytical solution with the above calculation equations, relevant parameters are assigned according to the actual coal mine site conditions, to analyze the stress distribution in the fault zone. the specific parameters selected are shown in table 1. the parameters in table 1 are substituted into equations (4) and (5) to obtain the stress distribution of the normal and shear stresses of the fault zone under spacing distances of 10, 30, 50, and 70m between the working surface and fault, as shown in figure 2 and figure 3. it can be observed from figure 2 that the maximum normal stress value in the fault zone appears at a distance of 100m between the fault zone and origin o; that is, the closest position to the working face. here, the normal stress of the fault zone is relatively large at most places under the condition that the working surface is 10m away from the fault; however, the stress is relatively small when the working surface is 70m away from the fault. at a distance of 30 to 40m, the normal stress is relatively large under the condition that the working surface is 70m away from the fault, while it is relatively small under the distance of 30m. it happens because the mining working surface is causing additional normal stress in the fault zone, and the closer the distance to the fault, the more significant the influence will be. it can be seen from figure 3 that the maximum shear stress value in the fault zone is at 100m from the origin o under different distances between the working surface and fault; that is, the closest position to the working face. here, the shear stress of the fault zone is relatively large at most places under the condition that the working surface is 10m away from the fault; however, the stress is relatively small where the working surface is 70m away from the fault. likewise, the shear stress is relatively large under the condition that the working surface is 70m away from the fault, while it is relatively small at 30m, as a result of additional working surface shear stress. combining figure 2 with figure 3, it can be seen that the working face mining stress has a significant effect on the fault zone, causing the changing trend of the normal and shear stresses of the fault zone first increasing and then decreasing, under the conditions that the working surface is 10 and 30m from the fault. the working face mining stress has a small effect on the fault zone, resulting in the changing trend of the normal and shear stresses of the fault zone increasing gradually, under distance conditions of 50 and 70m. 2.2. numerical model analysis a working face-mining-reverse-fault-simulation model was established by utilizing the numerical simulation software flac3d, to better analyze the variation law of the fault zone under the influence of mining, as well as to verify the fault mechanics model analysis. the mining range of the model is 100m in length, 540m in trend length, and 100m in height, as illustrated in figure 4. the working surface excavation process is simulated using dummy cells, using the mohr-coulomb model for simulating the mechanical changes in the coal and rock mass. based on the geological survey and related rock mechanics test results, the rock mass size effect is taken into account. the parameters of each rock formation adopted in the simulating calculations are displayed in table 2. based on the theoretical calculation scheme, a simulation scheme with distances of 10, 30, 50, and 70m between the working face and fault is established when the fault dip angle is 60°. the cross-section of the longitudinal section of the working face mining position and the roof plate horizontal position are intercepted to analyze the effects of the distance between the mining working surface and fault. the simulation results are displayed in figure 5. 142 yuanhui li, rui zhou (a) stress distribution with a distance of 10m between working face and fault (a) variation law of fault instability coefficient (b) stress distribution with a distance of 30m between working face and fault (b) variation law of shear stress with normal stress (c) stress distribution with a distance of 50m between working face and fault (d) stress distribution with a distance of 70m between working face and fault figure 5. stress distribution with different distances between working face and fault figure 6. variation law of fault with different distances it can be seen from the longitudinal section in figure 5 that the stress concentration lies in the roof plate position on the front face of the working surface under four distance conditions between the working surface and fault. the stress in this position should be the maximum; the closer the distance to the fault, the larger the additional stress generated in the fault will be. this point is the same as the maximum position of the fault zone in the theoretical analysis. it can be seen from the horizontal cross-section in figure 5 that when the working surface is 10 and 30m, respectively, from the fault, the roof plate stress range should cover the position of the fault zone, apparently affecting the fault zone stress. under the distance conditions of 50 and 70m, the working surface stress range excludes the fault zone position. according to the numerical simulation results, the influence of the mining stress on the fault zone stress is gradually reduced with an increase in the working distance between the working face and fault, which is also the case in the theoretical calculation. therefore, when combining the elastic mechanic’s theory and numerical simulation results, it can be concluded that the working surface trending mechanics model of the working fault in the working face can be effectively reflected by the variation law of the fault zone stress. 2.3 stability analysis of reverse fault zone two methods for the mohr-coulomb strength criterion and fault activation determination were selected to analyze the fault stability. in determining the strength criterion, the fault friction angle is selected as 10°, and the cohesion is 0.1mpa. since the friction angle and cohesive force of the fault are determined according to the empirical value, and the characteristics of each fault in the fault zone differ, the qualitative analysis could only be conducted on the possibility of fault instability, rather than determining whether the fault was stabilized. the fault instability coefficient is set as α, according to the mohr-coulomb theory strength criterion equation, to facilitate analysis, as is showed in formula 6. the larger the value of α, the higher the probability of fault instability; the smaller the value of α, the more stable the fault will be. the fault activation determination method is conducted according to the ratio between the shear and normal stress of the fault zone; the more significant the ratio, the higher the possibility of instability. (6) where cf is the fault cohesion, mpa, and ɸ is the internal friction angle, °. employing this calculation, when the distances between the working surface and fault are 10, 30, 50, and 70m, respectively, the ratio variation law among the instability coefficient of the reverse fault, the shear stress, and the normal stress can be obtained, as indicated in fig. 6. it can be observed from figure 6 that the fault stability results obtained by the two methods are almost the same. the position with poor stability in the fault zone is located at a distance of 100m between the fault zone and origin o; that is, the closest position to the working face. here, the poorest fault stability is under the distance condition of 10m between the working surface and fault, 143analysis of mechanical characteristics and instability law of fault under the influence of mining figure 7. layout of 25011 working face while the highest fault stability can be obtained at a distance of 70m. under distance conditions of 50 and 70m between the working face and fault, the working face mining stress has no apparent effect on the fault zone; thus, the fault zone stability is gradually increased. under distance conditions of 10 and 30m between the working face and fault, the working face mining stress results in apparent additional pressure on the fault zone, causing its overall stability to exhibit a trend of first increasing, then decreasing and growing again. moreover, negative values exist in the calculation of the ratio of the instability coefficient, shear stress, and normal stress in the fault zone. the physical meaning of the fault zone is that the stress leading position of the fault is different from other areas. in this case, the possibility of fault fracture is also intensified, in opposition to fault stability. therefore, such a stress state should be avoided in actual production work. in summary, it is clear that the working face mining pressure significantly influences the fault stability. moreover, the closer the fault, the higher this influence will be. therefore, the minimum spacing between the working face and fault should firstly be determined in field work, to avoid the risk of fault instability caused by different stress directions in the fault zone. then, comprehensive consideration should be given to actual situations, such as roof plate support on site and economic benefits, to reasonably determine the width of the coal pillar between the working face and reverse fault. 2.4 analysis of actual situations on site the 25110 working face of a particular mine consists of fully mechanized caving faces, located at the level of -445m of the mine 25 mining area. the 2-1 coal seam was evacuated on the working surface. the coal ranges from 7.4 to 13.8m in thickness, moving towards 112 to 27° with a tendency of 202 to 217° and a dip angle of 12°, and belongs to the slowly inclined thick coal seam. the lower lane of the working surface of 25011 is near the f16 reverse fault, as indicated in figure 7. according to the actual situation on site, a total of eight rock burst events occurred in the 25110 working face during the mining process, all within the lower lane of 25011. the first to seventh events occurred during the tunnelling process, while the eighth occurred during the mining process. therefore, we can conclude that the 25011 lower lane approaching the reverse fault is a major area for rockburst occurrence. furthermore, as the mining distance during the early stage approached the reverse fault, one rockburst event was also caused during the mining process; thereafter, no such event occurred. thus, it can be determined that the closer the dip distance of the working face to the fault, the more easily the rock burst will occur, which is consistent with previous analysis results. 3 conclusions (1) based on the law of the working face dip pressure distribution, a reverse fault mechanics analysis model was established to deduce the calculation equation of the normal and shear stresses of the fault zone under the influence of the mining working face. it could be observed that the normal and shear stresses of the fault zone exhibited a changing trend of first increasing and then decreasing under the conditions in which the working surface was 10 and 30m from the fault. the normal and shear stresses of the fault zone were exhibited a changing trend of increasing gradually under distance conditions of 50 and 70m between the working surface and fault. the stress distribution of the longitudinal and transverse horizontal sections under spacing conditions of 10, 30, 50, and 70m between the working and fault distance were obtained by means of the flac3d numerical simulation method. the results indicate that the influence range of the mining stress on the working face at spacing distances of 10 and 30m included the fault zone, while at spacing distances of 50 and 70m, the fault zone was excluded. consequently, it can be concluded that the mining stress not only affects the fault zone, but also changes the variation trend of its normal and shear stresses under spacing conditions of 10 and 30m; however, there was less influence under spacing conditions of 50 and 70m. (2) two mohr-coulomb theory rule methods and fault activation determination were applied to analyse the fault zone stability. when the working surface was 10 and 30m away from the fault, negative values existed in the calculation of the instability coefficient, shear stress, and normal stress ratio in the fault zone. the physical meaning of the fault zone is that the stress leading position of the fault is different from other positions. in such a case, the possibility of fault fracture is also intensified, resulting in a possibility of rock burst. when the spacing was 50 and 70m, the fault zone stability exhibited an overall changing trend of gradually increasing. moreover, the most dangerous fault zone positions under four spacing conditions were located at the closest position to the fault. (3) based on actual site conditions, rock burst was more likely to occur in the lanes approaching the fault zone, indicating that the spacing between the working surface and fault zone was an important factor affecting the occurrence of rock burst. therefore, determining a reasonable distance between the mining working face and inverse fault zone was a foundation for preventing rock burst. acknowledgments this research was financially supported by the yong foundation of china coal technology engineering group (no. 2014qn003); the state key research development program of china (no. 2016yfc0801605); the national natural science foundation of china (no. 51674061); and the state key research development program of china (no. 2016yfc0600706). references asghar, z., ali, w., nasir, a., & arshad, a. (2018). atmospheric monitoring for ambient air quality parameters and source apportionment of city faisalabad, pakistan. earth sciences pakistan, 2(1), 01-04. díaz-aguado, m. b., & gonzález-nicieza, c. (2007). control and prevention of gas outbursts in coal mines, riosa–olloniego coalfield, spain. international journal of coal geology, 69, 253-266. gibowicz, s. j. (1984). the mechanism of large mining tremors in poland(1984). rockbursts and seismicity in mines, 22, :17-28. guan leong, t., tahir, s.h. & asis, j. (2018). stratigraphy of paleogene sequences in wes ton – sipitang, sabah. geological behavior, 2(1), 01-04. jiang, f. x., & liu, w. j. (2010). coupling study of microseismic monitoring and numerical simulation for tectonic activation. journal of china coal society, 35, 900-904. jiang, y. d., pan, y. s., jiang, f. x., dou, l. m., & ju, y. (2014). state of the art review on mechanism and prevention of coal bumps in china. journal of china coal society, 39, 205-213. jiang fuxing, liu weijian. coupling study of microseismic monitoring and numerical simulation for tectonic activation.(2010). journal of china coal society, 35: 900-904. jiang yaodong, pan yishan, jiang fuxing, et al. (2014) state of the art review on mechanism and prevention of coal bumps in china. journal of china coal society, 39: 205—213. joughin, n. c., & jager, a. j. (1984). fracture of rock at stope faces in south african gold mines. proceedings of symposium of rockbursts: 144 yuanhui li, rui zhou prediction and control, 27, 53 -66. khanchoul, k., saaidia, b., & altschul, r. (2018). variation in sediment concentration and water disharge during storm events in two catchments, northeast of algeria. earth sciences malaysia, 2(2), 01-09. james w, lamoreaux, j. w., wu, q., & zhou, w. (2014) new development in theory and practice in mine water control in china. carbonates and evaporites 29, :141–145. li, z. l., dou, l. m., cai, w., wang, g. f., he, j., gong, s. y., and & ding, y. l., (2014)., investigation and analysis of the rock burst mechanism induced within fault–pillars. international journal of rock mechanics and mining sciences, 70, 192–200. mahmood, s., kazmi, s.t., & ali, s.s. (2018). comparison of drinking water bottles of different countries along with zamzam water, pakistan. earth sciences pakistan, 2(1), 05-14. maría b, díaz a, gonzález nc (2007) control and prevention of gas outbursts in coal mines, riosa–olloniego coalfield, spain. int j coal geol 69:253–266. mcgarr, a. (1984). some applications of seismic source mechanism studies to assessing underground hazard(1984). rockbursts and seismicity in mines, 31, : 199-208. nwankwoala, h.o., & ememu, a.j. (2018). hydrogeochemical signatures and quality assessment of groundwater in okpoko and environs, southeastern nigeria. pakistan journal of geology, 2(1), 06-11. potgieter, g. j., & roering, c. (1984). the influence of geology on the mechanisms of mining-associated seismicity in the klerksdrop goldfield.(1984) rockbursts and seismicity in mines, 11, 1984 : 45 -50. qian, m. g. (2003). minggao, shi pingwu, ground pressure and strata control. (xuzhou: china university of mining and technology press.,2003) rahim, i.a., asis, j., & mohd husin, m.a.y. (2018). comparison of different type of friction angle in kinematic analysis. malaysian journal of geosciences, 2(1), 35-38. sudhakaran, m., ramamoorthy, d., savitha, v., & balamurugan, s. (2018). assessment of trace elements and its influence on physio-chemical and biological properties in coastal agroecosystem soil, puducherry region. geology, ecology, and landscapes, 2(3), 169-176. sufiyan, i., zakariya, r., yacoob, r., idris, m.s., & idris, n.m. (2018). swat subbasins parameters and flood risk simulations using 3d in terengganu watershed. earth sciences malaysia, 2(2), 10-15. tair, r. & dell, l. (2018). spring-water as an alternative resource after earthquake for villagers, kota belud sabah. geological behavior, 2(1), 05-11. tan, y. l., ning, j. g., & zhao, t. b., (2011). deformation and control of deep roadways. (beijing: coal industry press., 2011), thiruchelvam, s., ismail, m.f., ghazali, a., mustapha, k.n., norkhairi, f.f., yahya, n., isa, a.m.m., & muda, z.c. (2018). development of humanitraian supply chain performance conceptual framework in creating resilient logistics network. malaysian journal of geosciences, 2(1), 27-30. veeraragavan, s., duraisamy, r., & mani, s. (2018). seasonal variation of soil enzyme activities in relation to nutrient and carbon cycling in senna alata (l.) roxb invaded sites of puducherry region, india. geology, ecology, and landscapes, 2(3), 155-168. wali, e., phil-eze, p.o., & nwankwoala, h.o. (2018). forecasting the future pattern of land use and land cover change in the wetland ecosystem of the port harcourt metropolis. pakistan journal of geology, 2(1), 01-05. xiao, y. x., feng, x. t., li, s. j., feng, g. l., & yu, y. (2016). rock mass failure mechanisms during the evolution process of rockbursts in tunnels. international journal of rock mechanics and mining sciences, 83,174-181. the dynamic properties of subgrade materials are critical factors affecting stability within the traffic engineering discipline. remolded red clays are frequently used as subgrade filling materials, however, to date, a paucity of data exist on to the dynamic properties of this material. accordingly, a large number of dynamic triaxial tests under cyclic loads were carried out to quantify the suitability of remolded red clay as subgrade filling. several potentially influencing dynamic factors were considered, including dynamic stress, vibration frequency, consolidation confining pressure, consolidation ratio and compactness. plastic strain and dynamic strength curves of remolded red clays under varying dynamic loads and load histories have been developed, in addition to the inclusion of those influencing factors. test results show that within the range not exceeding the inherent strength of the test samples, increases soil compactness, confining pressure, and vibration frequency serve to enhance overall dynamic power in concurrence with retarding the development of accumulated plastic strain. conversely, an improvement in the amplitude of the dynamic stress and consolidation ratio was shown to cause a decrease in dynamic strength and acceleration in the development of accumulated plastic strain. an empirical equation relating critical dynamic strength and load histories of remolded red clay has been developed for the provision of fundamental reference data for future studies. las propiedades dinámicas de los materiales de base son factores cruciales que afectan la estabilidad en la ingeniería de tráfico. la arcilla roja reestructurada se utiliza frecuentemente como material de relleno de bases, sin embargo, hasta la fecha, existe una escasez de información sobre las propiedades dinámicas de este material. de acuerdo con esto, se realizó un gran número de pruebas triaxiales dinámicas bajo cargas cíclicas para cuantificar la pertinencia de la arcilla roja reestructurada como material de relleno en bases. se consideraron varios factores dinámicos que podrían ser determinantes, como la fuerza dinámica, la frecuencia de vibración, la presión de confinamiento, el índice de consolidación y la compactibilidad. se desarrollaron las curvas de fuerza plástica y dinámica de arcilla roja reestructurada con varias cargas dinámicas e historia de cargas, además de la inclusión de los factores determinantes. el resultado de las pruebas muestra que dentro del rango de la fuerza inherente a las muestras de estudio, el incremento de la compactibilidad del suelo, la presión de confinamiento y la frecuencia de vibración sirven para mejorar, en general, el poder dinámico al tiempo que retrasa el desarrollo de la fuerza plástica. al contrario, el mejoramiento de la amplitud de la fuerza dinámica y el índice de consolidación muestra una reducción en la fuerza dinámica y una aceleración en el desarrollo de la fuerza plástica acumulada. finalmente, se desarrolló una ecuación empírica que relaciona la fuerza dinámica crítica y las cargas históricas de arcilla roja reestructurada con el fin de proveer información de referencia para estudios futuros. abstract resumen keywords: accumulated plastic strain, dynamic triaxial tests, dynamic strength, remolded red clay. palabras clave: fuerza plástica acumulada, pruebas triaxiales dinámicas, fuerza dinámica, arcilla roja reestructurada. test study on the influences of dynamic stress and load history to the dynamic properties of the remolded red clay issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n4.54400 pruebas sobre la influencia de la carga dinámica y la historia de carga en las propiedades dinámicas de arcilla roja reestructurada earth sciences research journal earth sci. res. j. vol. 20, no. 4 (december, 2016): g1 g8. jian li, shang-xiong chen, and ling-fa jiang state key laboratory of geomechanics and geotechnical engineering, institute of rock and soil mechanics, chinese academy of sciences, china xiaohongshan, wuchang, wuhan 430071, p. r. china author for correspondence: jian li (e-mail: 710171305@qq.com). g e o t e c h n ic s how to cite item li, j., chen, s. x., & jiang, l. f. (2016). test study on the influences of dynamic stress and load history to the dynamic properties of the remolded red clay. earth sciences research journal, 20(4),g1-g8 http://dx.doi.org/10.15446/esrj.v20n4.54400 record manuscript received: 26/11/2015 accepted for publication: 06/01/2017 h2 jian li, shang-xiong chen, and ling-fa jiang i. introduction red clays are widely distributed in southwest and south central regions of china (editorial of engineering geology manual, 2007); accordingly, red clay is frequently utilized as a subgrade filling for fiscal and environmental reasons. thus, studies of the mechanical properties of this subgrade encompass a high practical significance within traffic engineering. several previous studies have undertaken in-situ mechanical testing on red clay foundations, with results indicating significant variation with respect to both the bearing capacity and deformation modulus of red clay bases in concurrence with varying mechanical properties, water content, particle composition, occurrence conditions, soil structure and interior fracture, among other factors (long, chen, xiao, & peng, 2009; kong, luo, & yuan, 1995; kong & luo, 1993; bai, kong, guo & wan, 2010; bai and wan, 2010; peng, lin, wang, & wu, 2015). moreover, studies have shown that the aforementioned red clay characteristics may exert considerable effects on both the safety and suitability of the subgrade when used as a subgrade filling (editorial of engineering geology manual, 2007; kang et al., 2010). the remolding procedure is typically employed to overcome the inherently weak engineering characteristics of red clays. numerous previous studies have examined the static properties of remolded red clays. results indicate that the remolding procedure can effectively eliminate adverse design factors including cracking and uneven distribution, thus enhancing friction, bite force and bearing capacity (yang & lu, 2011; liang & mo, 2010; zhang, jiang, & yang, 2011; kaynia, madshus & zackrisson, 2000). however, within traffic engineering, the dynamic properties of subgrade materials are critical; to date, a paucity of data exist on the dynamic characteristics of remolded red clay, thus representing a significant knowledge gap in light of its frequent use as a subgrade material. the dynamic properties of geomaterials are directly related to dynamic loading which typically encompasses three elements, namely, waveform, amplitude, and frequency. previous investigations have shown that the dynamic loading borne by the subgrade is analogous to a sine wave pattern. thus, the sine wave has frequently been employed to simulate traffic loading within traffic engineering applications (kaynia et al., 2000; liu, 2010). the amplitude of dynamic loading exerts significant influence on the inherent dynamic properties of geomaterials, particularly those of saturated soils. an improvement in the magnitude of dynamic loading may result in significantly decreased dynamic subgrade strength, in concurrence with the accelerated development of accumulated plastic strain (li, 2013; chen, 2015). the influence of vibration frequency on the dynamic properties of geomaterials may be approximately divided into three types. firstly, the dynamic strength of the geomaterial may be improved, in concurrence with a retardation of the accumulated plastic strain due to an increased vibration frequency (duncan and chang, 1970; lin, 2011). secondly, the effect of vibration frequency on the dynamic properties of a geomaterial may not be visible (li, 2010; zhou, 1998); lastly, the dynamic strength and ability to resist deformation may be reduced due to an increase in associated vibration frequency (zhang, 2008). thus, an investigation of the dynamic characteristics of geomaterials under cyclic loading is of critical importance. in the current study, several tests have been designed and undertaken to examine the dynamic properties of remolded red clays. furthermore, previous investigations have shown that loading history exerts significant effects on the dynamic characteristics of geomaterials. an increased consolidation confining pressure has been demonstrated to improve the ability of geomaterials to resist external dynamic loading, in concurrence with reduced dynamic strain (lin et al., 2013; gu, wang, cai, yang & gao, 2012; zhou and gong, 2001). moreover, consolidation ratio also exerts a marked influence upon dynamic soil properties, albeit, with different types and magnitudes of influence associated with different geomaterials. for example, upon increasing the consolidation ratio within a sand or categorically coarse-grained soil, dynamic strength is observed to exhibit an initial increase, followed by a subsequent decrease. however, the associated turning points of various soils have been shown to vary (zhang, 1991; he, 1998; zhang, he, fei & gao, 2006; lunne, berre, andersen, strandvik, & sjursen, 2006; li, chen & jiang, 2015). conversely, an increase of the consolidation ratio of saturated soils reduces the associated dynamic strength (lunne et al., 2006); a significant consolidation ratio has been shown to promote the formation of a shear band due to the inherently low strength of saturated clays. similarly, the effect of compactness upon the dynamic strength of geomaterials is clearly evident; upon compaction, bulk density is increased, thus increasing dynamic strength and retarding the development of accumulated plastic strain (li, 2013; li, chen, & jiang, 2013). in the current study, a series of cyclic dynamic triaxial tests have been undertaken under varying load histories and dynamic loadings to investigate the dynamic properties of remolded red clay and the associated influencing factors. accumulated plastic strain and dynamic strength of samples have been evaluated and compared, leading to the development of predictive mathematical models about both dynamic characteristics. accordingly, the critical dynamic strength of remolded red clay is effectively examined, with empirical results presented for use in future engineering applications, thus representing significant utilities from both an academic and applied perspective. ii. dynamic triaxial tests of remolded red clay samples 1. dynamic triaxial apparatus the employed dynamic triaxial test used a computer-controlled electro-hydraulic servo dynamic triaxial test system (sdt-10), as shown in figure 1. summarily, the system comprises a microcomputer-controlled electro-hydraulic sdt test system host, a standard triaxial pressure cell, a set of electro-hydraulic servo exciters for effective application of confining pressure, a set of digital multi-channel electro-hydraulic servocontrolled systems and a hydraulic source. the test system is capable of providing a maximal axial dynamic load of 10 kn, while the triaxial pressure cell is associated with a maximum confining pressure of 1 mpa. deformation sensor precision is > 0.5%; the precision of measured strain is approximately 10-4. the provided axial and exciting lateral frequency is 0-5 hz, while control type may be stress-controlled. fi figure 1. sdt-10 dynamic triaxial test system employed in the current study. 2. test sample parameters the remolded soil sample used was obtained from an area adjacent to the interurban railway connecting wuhan and xianning. subsequently, the engineering properties and test parameters of the sample were confirmed with several laboratory analyses including grain size, liquid limit, plastic limit and compaction. grain size analyses indicate that grain sizes associated with the test samples were all < 0.075 and thus categorized as being finegrained; measured soil proportion was 2.7. determined liquid limit, plastic limit, and plastic index were 45%, 20.9% and 23.1%, respectively (table h3test study on the inf luences of dynamic stress and load histor y to the dynamic properties of the remolded red clay 1). accordingly, the study sample was shown to exhibit a relatively high plastic index, id est fine grain size in concurrence with high clay content. according to the ‘railway engineering geotechnical test regulations of china,’ the study sample is categorized as cohesive clay with a high liquid limit. compaction test results indicate an optimum moisture content of 20% and a maximum dry density of 1.616 g/cm3 (fig. 2.). 3. test program applied dynamic load amplitude must remain static to comprehensively analyze the developmental accumulated plastic strain and dynamic strength of remolded red clay under varying long term cyclic loads. firstly, varying load histories were applied to the study sample, after which different cyclic loads were applied; the dynamic mechanical properties of the test sample in the presence of varying load histories were continuously monitored; all undertaken analyses were stress-controlled. four influencing factors were considered during test procedures, namely compactness, confining pressure, consolidation ratio and vibration frequency. specific test programs are presented in table 2. table 1. the test result of particle analysis. note: < 0.05 mm, content: 90.7%; < 0.01 mm, content: 52.7%; < 0.002 mm content: 29.6%. figure 2. moisture content/dry density relation curve. 4. soil sample preparation and loading all analyses were undertaken by the ’geotechnical test regulations of china’ to ensure test data reliability. chiefly: (1). all soil samples were prepared using the same method. before dynamic loading, varying confining pressures and consolidation partial pressures were applied to soil samples until deformation conditions were reached (axial deformation <0.005 mm in 5 minutes). during the test procedure, varying cyclic loads were applied to soil samples in concurrence with continuous monitoring of dynamic σ–ε curves (fig. 3.). test termination occurred when vibration time reached 2000 times or accumulated plastic strain reached 15%. grain size/mm 0.5~0.25 0.25~0.075 0.075~0.005 < 0.005 content/% 57.4 42.6 table 2. short-time dynamic triaxial test program. figure 3. load waveform diagram of the grading short-time dynamic triaxial test. iii. test results and analysis 1. test curves of long-term cyclic loading tests 1.1. accumulated plastic strain test curves several accumulated plastic strain curves under different load histories were obtained (figs. 4-7). as shown, under relatively small dynamic loads, the accumulated plastic strain of the test samples developed rapidly during initial loading, followed by gradual stabilization due to soil compaction. under relatively high dynamic loading, accumulated plastic strain grows rapidly, exhibiting a more constant gradient during initial loading, with an elevated rate of development during later loading stages. upon exertion of larger dynamic loading values, developing standards during all loading stages were observed as being relatively similar, with soil deformation occurring fairly rapidly. 1.2. dynamic strength test curves dynamic strength of soils varies with loading rate and the cyclic impacting effects of dynamic stress. typically, dynamic stress is associated with a particular deformation criterion in concurrence with the impacting vibration times of dynamic loading (xie, 1986). in the current study, the definition of dynamic strength resulted in marked variation in parallel with diverse conditions. furthermore, dynamic strength is closely related to the cyclic vibration time and the deformation criteria. accordingly, accurate determination of deformation strain is necessary to an increased understanding of soil-related dynamic stress. the selected approach for determination of dynamic strength was when the accumulated plastic strain reached 5%. dynamic strength variation on remolded red clay samples under diverse conditions are presented in figures 8-11. as shown, under test id influencing factors compactness confining pressure consolidation ratio vibration frequency dt1-1 dt1-2 dt1-3 0.80 0.85 0.95 50 4.0 1 hz dt2-1 dt2-2 dt2-3 0.8 25 50 100 3.0/4.0 1 hz dt3-1 dt3-2 dt3-3 0.8 50 1.0 2.0 3.0 1 hz dt4-1 dt4-2 dt4-3 0.8 50 3.0 1 hz 2 hz 5 hz relation curve between water content and dry density 1.44 1.46 1.48 1.5 1.52 1.54 1.56 1.58 1.6 1.62 1.64 15.0% 16.0% 17.0% 18.0% 19.0% 20.0% 21.0% 22.0% 23.0% 24.0% water content d r y d e n s i t y g / c m 3 h4 jian li, shang-xiong chen, and ling-fa jiang variable test conditions, dynamic strength (σd) was observed to decrease non-parametrically with vibration time. under the effects of relatively large dynamic stress, soil deformation occurred rapidly in concurrence with reduced dynamic loading, gradually increasing deformation vibration time and a stabilizing relational curve. effectively, these conditions indicate that study samples exhibited a critical dynamic strength σcr. thus, soil deformation does not occur when the measured dynamic stress was <σcr. (a) compactness 0.8 (b) compactness 0.85 (c) compactness 0.9 figure 4. accumulated strain curves under varying values of compactness (kc=4.0, σ3=50 kpa, f=1 hz). (a) consolation ratio 2.0 (b) consolation ratio 3.0 (c) consolation ratio 4.0 figure 5. accumulated strain curves under varying consolation ratios (λc=0.80, σ3=50 kpa, f=1 hz). (a) confining pressure 50 kpa (consolation ratio 3.0) (b) confining pressure 100 kpa (consolation ratio 3.0) h5test study on the inf luences of dynamic stress and load histor y to the dynamic properties of the remolded red clay (c) confining pressure 25 kpa (consolation ratio 4.0) (d) confining pressure 100 kpa (consolation ratio 4.0) figure 6. accumulated strain curves under varying values of confining pressure (λc=0.80, f=1 hz). (a) vibration frequency 1 hz (b) vibration frequency 2 hz (c) vibration frequency 5 hz figure 7. accumulated strain curves under varying vibration frequencies (λc=0.80, σ3=50 kpa, kc=3.0). figure 8. dynamic strength results of remolded red clay samples (influenced by compactness). figure 9. dynamic strength results of remolded red clay samples (influenced by confining pressure). h6 jian li, shang-xiong chen, and ling-fa jiang compactness consolidation ratio confining pressure/kpa vibration frequency/ hz σcr /kpa 0.80 4.0 50 1.0 59.5 0.85 134.5 0.90 209.2 0.8 2.0 50 1.0 139.1 3.0 118.5 4.0 59.5 0.8 4.0 25 1.0 49.8 100 78.2 0.8 3.0 50 2.0 123.1 5.0 140.6 figure 10. dynamic strength results of remolded red clay samples (influenced by consolidation ratio). figure 11. dynamic strength results of remolded red clay samples (influenced by vibration frequency). additionally, results suggest soil dynamic resistance is improved upon enhancement of soil compactness, soil confining pressure and vibration frequency of dynamic stress. the association between measured consolidation ratio and dynamic strength was dissimilar to other influencing factors. varying factors were shown to exert different effects on the dynamic strength of study samples. for example, measured soil compactness had a highly significant effect on the measured dynamic strength of study samples, while other potentially influencing factors (e.g. confining pressure and vibration frequency) were shown to produce more subdued effects. overall, varying load histories demonstrated a marked impact on the dynamic strength of remolded red clays. 2. critical dynamic strength and equation development dynamic strength is a first soil parameter regarding dynamic loading to reflect and relate the development of soil deformation in concurrence with fluctuating strength. moreover, as previously outlined, a decreased dynamic stress is shown to be associated with and increased vibration time. consequently, below a specified level of dynamic stress, soil deformation does not occur, with this specified level of stress defined as the critical strength (σcr) (heath, waters, shenton & sparrow, 1972), as previously demonstrated (liu, 2010; li, 2010; zhou, 1998). based upon results and figures about dynamic stress obtained within the current study, characteristic σcr for test samples are presented in table 3. table 3. critical dynamic strength under varying conditions. while cyclic triaxial testing is the most frequently employed method for the determination of critical dynamic stress within a specified soil type or sample, its use on-site is problematic due to the necessarily high number of tests required for accurate approximation, the relatively high level of proof complexity and the potential presence of variable influencing factors. accordingly, in terms of practical application, a generalized empirical equation comprising compactness, confining pressure, consolidation ratio and vibration frequency has been developed via fitting analysis of calculated critical dynamic strengths. the suitable effect of the developed empirical equation was appropriately high, with an rsq of 0.98. due to influencing factors on the critical dynamic strength of test samples, in concurrence with inherent limitations associated with test methods employed, it is recommended that use of the presented equation be confined to conditions analogous to those maintained in its development. the developed empirical equation is as follows: (1) 3. influencing factor analyses of accumulated plastic strain and critical dynamic strength 3.1. accumulated plastic strain analysis εp is shown to be influenced to varying degrees by dynamic stress, compactness, confining pressure, consolidation ratio and vibration frequency based upon developmental curves of accumulated plastic strain (figs. 4-7); these variations are primarily due to interior soil conditions being differentially affected by dynamic stress and varying load histories. factors accelerating εp development: increased dynamic stress may accelerate intra-particle friction and promote the formation of sliding surfaces, thus accelerating εp development. εp may be accelerated via enhancement of the consolidation ratio due to increasing partial stress, in concurrence with an increased consolidation ratio and relatively large partial stress thus causing soil structure deformation, an inherently decreased dynamic strength and an acceleration of the development of εp. (ii) factors delaying εp development soil εp development may be retarded via increased soil compactness and confining pressure; the effect of varying soil compactness on observed εp is particularly significant. upon improvement of soil compactness and confining pressure, the void ratio is decreased, and friction between soil particles is increased, thus enhancing to soils ability to resist external dynamic loading. h7test study on the inf luences of dynamic stress and load histor y to the dynamic properties of the remolded red clay in the case of stiff remolded red clays, the development pattern of εp may also be effectively retarded, due to the combination of compaction over a short time and a significant vibration frequency. conversely, friction between soil particles will be increased in the presence of a high vibration frequency, resulting in increased resistance to external dynamic loading. (ii) influencing factor analysis of σcr as shown (figs. 8-11; table 3.), σcr is influenced to varying degrees by diverse loading histories; parametric associations exist between measured compactness, confining pressure, consolidation ratio, vibration frequency and critical dynamic strength of the remolded red clay samples tested in the current study. accordingly, an improvement in soil compactness, confining pressure and vibration frequency will result in an increased critical dynamic strength increase. conversely, an increased consolidation ratio, results in a decreased dynamic strength. it happens primarily due to the inherently increased soil density and a higher level of soil particle contact associated with increased soil compactness, thus leading to increased critical dynamic stress. increased confining pressure, also associated with higher levels of soil compaction, resulted in a lower level of dynamic strength increase. it is theorized that upon increasing vibration frequency, the contact area between soil particles is significantly increased in concurrence with increased intraparticle frictional forces, thus improving the soils’ ability to resist external loading. when consolidation ratio was increased, measured critical dynamic strength was decreased; the exertion of high large partial stress results in localized soil deformation, leading to a reduced level of soil dynamic strength. several large consolidation ratios were used to study the effect of the consolidation ratio on critical dynamic strength; an inversely proportional relation was found to exist between consolidation ratio and critical dynamic strength. in the case of consolidation ratios < 2.0, the authors recommend further analyses. iv. conclusions several dynamic triaxial tests were carried out on remolded red clays under varying load histories, thus permitting examination of the effects of dynamic stress and different load histories on the dynamic properties of these subgrade materials. based on test results, the following conclusions have been developed: (1) load records were shown to directly influence the dynamic properties of remolded red clays, with diverse load histories exhibiting equally different levels of influence. increased levels of compactness, confining pressure and vibration frequency, were shown to increase the test samples ability to resist external loading, thus retarding accumulated plastic strain development and improving the critical dynamic strength of the soil. due to inherent soil properties, the effects of compactness and confining pressure are especially significant. conversely, accumulated plastic strain development was accelerated in concurrence with decreased critical dynamic strength via improvement of the consolidation ratio, primarily due to the development of localized deformation and increased partial stress. (2) an increased vibrational frequency was shown to result in significantly improved dynamic strength and an increased ability to resist external loading within test samples. within categorically stiff unsaturated clays, the effects of increased vibrational frequency pore water pressure in soil are not significant. therefore, soil dynamic strength and modulus remain under the influence of friction between soil particles. based on analyses of the effects of varying load histories on the dynamic characteristics of remolded red clay, an empirical equation for relating critical dynamic strength to these load records has been developed and presented; this may be utilized for future engineering applications. acknowledgements this study was aided by financial support from the national natural science foundation of china (grant nos. 51027002), the natural science foundation of hubei province (grant nos. 2015cfb363) and the national science and technology support plan of china (grant nos. 2011bab10b02). references bai, w., kong, l., guo, a. g., & wan, z. (2010). research on distribution characteristics of bearing capacities and deformation parameters of laterite ground. rock and soil mechanics 31(s2), 164-169. bai, w. & wan, z. (2010). extension comprehensive prediction method for evaluation of subgrade bearing capacity of red clay area. highway, 7, 85-90. chen, g. h., sun, x. z., hu, h. y. & hu y. w. (2015). research on modeling and algorithm of supply chain’s reliability based on ccfsm. journal of coastal research, special issue no.73, 99-103. doi: http://dx.doi. org/10.2112/si73-018.1 duncan, j. m. & chang, c. y. (1970). nonlinear analysis of stress and strain in soils. journal of the soil mechanics and foundations division, 103(6), 517-533. editorial of engineering geology manual. (2007). engineering geology manual. architecture and building press, beijing, china. gu, c., wang, j., cai, y. q., yang, z., & gao, y. (2012). undrained cyclic triaxial behavior of saturated clays under variable confining pressure. soil dynamics and earthquake engineering, 40, 118-128. doi: http://dx.doi.org/10.1016/j.soildyn.2012.03.011 he, c. r. (1998). dynamic triaxial test on elongation and tensile characteristics of soil. journal of sichuan union university (engineering science edition), 2(4), 49-52. heath, d. l., waters, j. m., shenton, m. j., & sparrow, r. w. (1972). design of conventional rail track foundations. proceedings of the institution of civil engineers, 51(2), 251-267. doi: http://dx.doi. org/10.1680/iicep.1972.5952 kang, j. w., gan, y., wang, h. l., huang, l. h., wu, x. g., & peng, s. m. (2010). original stamping foundation treatment of red clay of new airport in kunming. chinese journal of geotechnical engineering, 32(s2), 469-500. kaynia, a. m., madshus, c. & zackrisson, p. (2000). ground vibration from high-speed trains: prediction and countermeasure. journal of geotechnical geoenvironmental engineering, 126(6), 531-537. doi: http://dx.doi.org/10.1061/(asce)10900241(2000)126:6(531)#sthash.sn1mdja4.dpuf kong, l. w. & luo, h. (1993). effect of the conversion in form of free iron oxide on the engineering property of the red clay. rock and soil mechanics 14(4), 25-39. kong, l. w., luo, h. x., & yuan, j. x. (1995). preliminary study on the effective cementation characteristics of the red clay. chinese journal of geotechnical engineering, 17(5), 42-47. li, j. (2013). on dynamic response of the red clay foundation under highspeed train loading. institute of rock and soil mechanics, chinese academy of sciences, wuhan, china, 36-39. li, y. (2010). study on dynamic behaviors of unsaturated loess in different vibration frequencies. northwest a&f university, yangling, china, 45-49. li, j., chen, s. x., & jiang, l. f. (2013). resonant column test on dynamic shear modulus and damping ratio of the remolded red clay and its law analysis. in: xie, l. (eds.) modeling and computation in engineering ii. 129-140. doi: 10.1201/b14896-21 li, j., chen, s., & jiang, l. (2015). dynamic strength and accumulated plastic strain development laws and models of the remolded red clay under long-term cyclic loads: laboratory test results. polish maritime research, 22, special issue s1 (86), 89-94. liang, b. & mo, k. (2010). research on remodelling red clay’s shear strength with different moisture ratio. shanxi architecture, 36(04), 101-102. lin, q. (2011). experimental investigation on dynamic behavior of soils affected by vibration frequencies. shanghai jiao tong university, shanghai, china, 59-61. h8 jian li, shang-xiong chen, and ling-fa jiang liu, x. h. (2010). research on dynamic stability of red clay subgrade under ballastless track of high-speed railway. unpublished phd thesis, central south university, changsha, china, 25-27. long, w. x., chen, k. s., xiao, t., & peng, x. p. (2009). research of general triaxial test for unsaturated red clay. rock and soil mechanics, 30(s2), 28-33. lunne, t., berre, t., andersen, k. h., strandvik, s., & sjursen, m. (2006). effects of sample disturbance and consolidation procedures on measured shear strength of soft marine norwegian clays. canadian geotechnical journal, 43, 726-750. peng, w., lin, z., wang, l. & wu, j. (2015). effect of weakly alkaline salt pretreatment on bio-boards for medicine safety. journal of pure and applied microbiology, 9(3), 1913-1917. xie, d. y. (1986). soil dynamics. xi’an jiaotong university press, xi’an, china, 69-73. yang, s. & lu, t. h. (2011). research on unventilated-undrained triaxial shear test of red cohesive soil. rock and soil mechanics, 32(s1), 356-359. zhang, j. h. (1991). research on pore water pressure strength and liquefaction properties of unsaturated sandy soil. chengdu university of science and technology, chengdu, china, 36-38. zhang, j. l., jiang, z. g., & yang, g. (2011). experimental study on mechanical behaviors of polypropylene fiber reinforced clay. chinese journal of geotechnical engineering, 33(1), 420-425. zhang, r., he, c. r., fei, w. p., & gao, m. z. (2006). effect of consolidation stress ratio on dynamic strength and dynamic pore water pressure of soil. chinese journal of geotechnical engineering, 28(1), 101-105. zhou, j. (1998). research on properties of the saturated soft clay under cyclic loads. zhejiang university, hangzhou, china, 26-29. zhou, j., & gong, x. n. (2001). strain degradation of saturated clay under cyclic loading. canadian geotechnical journal, 38(1), 208-212. doi: 10.1139/cgj-38-1-208 earth sci. res. j. vol. 11, no. 2 (december 2007): 131-138 clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano mauricio orozco-alzate, and césar germán castellanos-domínguez universidad nacional de colombia sede manizales, grupo de control y procesamiento digital de señales, campus la nubia, km 7 vía al magdalena, manizales, colombia. corresponding author: mauricio orozco-alzate, email: morozcoa@unal.edu.co abstract classification of seismic signals at colombian volcanoes has been carried out manually by visual inspection. in order to reduce the workload for the seismic analysts and to turn classification reliable and objective, the use of supervised learning algorithms has been explored; particularly classifiers built in dissimilarity spaces. nonetheless, the performance of such learning methods is subject to the availability of a representative and a priori well classified training sets. to detect mislabeled events, the use of clustering techniques on the dissimilarity representations is proposed. our experiments, performed on re-analyzed seismic signals, show a significant improvement respect to recognition accuracies for the original data sets. key words: clustering, dissimilarity, mislabeling, seismic signals. resumen la clasificación de las señales sísmicas en los volcanes de colombia ha sido llevada a cabo manualmente mediante inspección visual. con el fin de reducir la carga de trabajo de los analistas y para tornar la clasificación confiable y objetiva, se ha explorado el uso de algoritmos de aprendizaje supervisado; particularmente, clasificadores construidos en espacios de disimilitud. no obstante, el desempeño de dichos métodos de aprendizaje está sujeto a la disponibilidad de un conjunto de entrenamiento representativo y, a priori, bien clasificado. para detectar eventos mal clasificados, se propone el uso de técnicas de agrupamiento sobre las representaciones de disimilitud. los experimentos, realizados sobre las señales sísmicas verificadas, muestran una mejora significativa respecto a las tasas de reconocimiento para los datos originales. palabras claves: agrupamiento, disimilitud, etiquetado incorrecto, señales sísmicas. 131 earth sciences research journal manuscript received september 9 2007. accepted for publication november 30 2007. introduction in many applications of pattern recognition, it is extremely difficult or expensive, or even impossible, to reliably label a training sample with its true category (jain et al., 2000). particularly, in automatic classification of seismic-volcanic signals, night and rotating shift work schedules, tedious evaluations, and changes of personnel turn the task of recognition by visual inspection susceptible to human errors. besides, analysts often engage in differences of opinion about interpretations of dubitable signals. in order to reduce the workload for the seismic analyst and the risks associated to subjective judgments, a number of supervised classification methods have been used (scarpetta et al., 2005; langer et al., 2006; orozco-alzate et al., 2006a). it is supposed for those supervised classification techniques that a well-labeled data set is available. however, due to the same reasons cited above, it is highly likely that training sets include mislabeled events. in langer et al. (2006), an automatic classification of seismic events at soufrière hill volcano was carried out. in addition, a careful manual revision of the original a-priori classification was achieved by an expert not involved in the previous labeling of the data set. it was found that a considerable number of the events were erroneously attributed to other classes. as a result, a remarkable improvement in classification accuracy was obtained when the revised data set was used. the nevado del ruiz volcano is monitored by the volcanological and seismological observatory at manizales (vsom). because of the considerable amount of data, the labelling task of the recorded seismic signals is distributed among several analysts (e.g. one trainee per volcanic station). a second or third opinion is requested just in case of serious doubt. as a result, classifications performed by different experts are not available and an analysis of concordance for such a-priori labels was not conducted. in this study, a revision of the original labelled nevado del ruiz volcano (ruiz) data set is conducted. in contrast to the approach followed by langer et al. (2006), the revision by using clustering techniques was automated. several clustering algorithms on a given data set were used due to the lack of a single appropriate clustering algorithm (jain et al., 2000). therefore, experiments were conducted by using the most popular clustering approaches, which belong to two basic strategies: hierarchical and partitioning methods. in addition, the ruiz data set was arranged to consider two separated problems: the ruiz-vt,lp (two classes) and the ruizall (three classes) data sets. revised data sets were used according to our previous dissimilarity-based classification approach (orozco et al., 2006a, orozco et al., 2006b) and compared against the performances obtained with the original data sets. dissimilarity representation and classifier differences in spectral content allowed a visual discrimination of different types of volcanic earthquakes. therefore, spectra of seismic records are commonly used for classification and monitoring of seismic activity (zobin, 2003). in addition, recent studies have claimed that the dissimilaritybased classification approach is a feasible and sometimes advantageous alternative to the feature-based method (duin et al., 1998, pękalska et al., 2001, pękalska and duin, 2002, paclík and duin, 2003b, pękalska and duin, 2005). according to those facts, a dissimilarity representation for the ruiz data set can be derived as follows: (i) the power spectral density (psd) for each clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano 132 orozco and castellanos. esrj vol. 11, no. 2. december 2007 133 record is estimated via the yule-walker autoregressive method: dc bias must be removed before computing the spectra, (ii) a dissimilarity measure between normalized spectra is calculated as the area difference of the non-overlapping parts (l1-norm) between spectra, see fig. 1. figure 1. dissimilarity measure as the difference between normalized spectra. a dissimilarity matrix d(t,t) was constructed by having those pairwise measures. each entry dij of d corresponds to the dissimilarity between a pair of seismic records from the training set t. then, a proper classifier can be defined on such a dissimilarity representation, either by using the entire training set t or a representation set r⊆t. linear normal density based classifier a number of studies have showed that normal density based classifiers perform well in dissimilarity spaces (pękalska et al., 2001, pękalska and duin, 2002, paclík and duin, 2003b, paclík and duin, 2003a, pękalska et al., 2004, orozco et al., 2006a). particularly, in our previous study with the nevado del ruiz volcano data set (orozco et al., 2006b), the linear normal density based classifier (bayesnl) outperformed the nearest neighbor rule 1-nn and the quadratic normal density based classifier (bayesnq). for a two-class problem, the bayesnl classifier is given by              2 1 21 1 21 log2 1 ),()),(( p p crxdrxdf t       mmmm (1) where c is the sample covariance matrix; m(1), m(2) are the mean vectors and p(1), p(2) are the class prior probabilities. if c is singular, a regularized version must be used. the following regularization is typically used with λ equals 0.01 or less (pękalska et al., 2006):    cdiagccreg   1 . (2) clustering techniques unsupervised classification refers to figure 1. dissimilarity measure as the difference between normalized spectra. situations where the objective is to construct decision boundaries based on unlabeled training data (jain et al., 2000). hierarchical and partitioning methods are the two basic strategies to find clusters. in this study, the following clustering techniques are used: single linkage (sl), average linkage (al), complete linkage (cl), k-means and k-centres. sl, al, and cl are hierarchical, whereas the latter are partitioning methods. a brief description of these approaches is given below. hierarchical clustering the most popular hierarchical techniques for clustering are the agglomerative methods. at the beginning, each object is considered as a single cluster; then, the closest two clusters are merged iteratively until a specified number of clusters is reached (pękalska and duin, 2005). let ck and cl be two clusters of the cardinalities nk and nl respectively, and let ρkl be a dissimilarity measure between them. three basic criteria for the agglomerative methods are summarized in table 1. partitioning clustering partitioning methods group the objects into k clusters, usually by using representatives or by assuming a specific geometrical structure. objects are assigned to the clusters, new representatives are estimated and the process is repeated until a stable solution is reached. two typical partitioning methods are k-means and k-centres; see table 2 for a brief description, a detailed one can be found in pękalska and duin (2005). table 2. clustering methods. experimental results volcano-tectonic (vt) earthquakes, longperiod (lp) earthquakes and icequakes (ic) are the seismic signals classes considered in this study. they are contained in the ruiz-all data set. the ruiz-vt,lp data set includes only the first two classes. signals were digitized at 100.16 hz sampling frequency by using a 12 bits analog to digital converter. 134 method � kl emphasis/comment sl  jicpcp ppdljki ,minmin  connectedness. resulting clusters are elongated and chain-like. cl  jicpcp ppdljki ,maxmax  compactness. it performs well when the objects form naturally distinct clouds. al     ki ljcp cp ji lk ppd nn ,1 connectedness and compactness. it performs well for naturally distinct clouds and elongated clusters. table 1. hierarchical clustering methods. clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano 135 recording stations are located near to the olleta crater and the glacier at nevado del ruiz volcanic complex. in order to explore the level of agreement/ disagreement between the labels given by the experts and the ones produced by the clustering algorithms, the number of mismatches for the entire data sets is considered. the averaged number of mismatches over 10 runs is reported in table 3. hierarchical methods report the same number of mismatches over the runs, therefore their standard deviations are zero. sl hierarchical criterion for both the ruizvt,lp and the ruiz-all problems presents a rate of disagreement considerable high; similarly, mismatches of al results for the ruiz-all problem reach 45%. in fact, even though the number of cluster is fixed, sl and al find second and third clusters of a few objects only. as a result, valid data subsets, i.e. randomly generated and including enough objects per class, are not always warranted. in consequence, al for the ruiz-all problem and sl in both cases are not considered in the subsequent classification experiments. orozco et al. (2006b) observed an asymptotic behaviour for training set sizes greater than 60 examples per class. in addition, the bayesnl provided the best overall performance, outperforming the 1-nn and the bayesnq. according to that, the experiments were conducted with the bayesnl using training sets of a fixed size of 60 objects per class. clustering was performed on the entire data sets. then, training and test sets are randomly extracted for each run. the results are shown in table 4. for comparison, the results using the original data are also presented. it is clear that performances for re-labelled data sets are much better than those for the original data. discussion and conclusion a revision of the original labelled seismic events recorded by the vsom staff provides a significant improvement in the performance of supervised dissimilarity-based classifiers such as the observed for the bayesnl classifier. the use of events labelled by clustering confirmed that labelling errors are frequent and recurrent. clustering uses a notion of proximity, judged in a numerical way. in contrast, labels assigned by experts obey to the method description k-means representatives are estimated by cluster mean vectors. the dissimilarity is the euclidean distance of an object to the cluster means. k-centres centre objects are chosen such that the maximum of the distances over all objects to the nearest centre is minimized. results depend on random initialization. table 2. clustering methods. orozco and castellanos. esrj vol. 11, no. 2. december 2007 136 clustering method ruiz-vt,lp ruiz-all sl 482 1108 cl 367 495 al 164 861 k-centres 158.2 (25.2049) 507.5 (52.2797) k-means 135.6 (0.5164) 506.6 (0.5164) total 1063 1891 table 3. averaged number of mismatches between the class labels assigned by the vsom staff and labels assigned by the clustering method over the entire data sets. clustering method ruiz-vt,lp ruiz-all sl — — cl 2.3494 (0.4045) — al 4.9646 (1.1176) 3.77 (0.76) k-means 2.7524 (0.7405) 5.6722 (0.6109) k-centres 2.8810 (0.7441) 4.6792 (0.8413) total 13.0075 (1.0354) 20.02 (0.81) table 4. classification error (in % and averaged over 25 runs) with its standard deviation (in %) for the rnlc applied to the revised data sets. clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano 137 visual resemblance between the event and a canonical waveform which analysts have learnt by reference or experience. obviously, such a method is highly subjective and supposes that differences are easily detected by visual inspection but in many cases this is not true. since the final rule used (stand-alone) was calculated from all the data, clustering methods were used on the entire data sets instead of applying them to the training sets only. al and cl offer the smallest errors for the ruiz-vt,lp and ruiz-all problems respectively. even tough, the best clustering is hierarchical in both cases; differences are not enough to claim that hierarchical methods should be preferred over the partitioning ones. nonetheless, a general conclusion can be drawn from our study: the use of a clustering method to confirm labels assigned by experts is highly beneficial for constructing reliable and accurate supervised classifiers of seismic events. acknowledgements we thank the vsom staff for providing the raw data set. references duin, r. p. w., de ridder, d., and tax, d. m. j. (1998). featureless pattern classification. kybernetika, 34, no. 4, 399–404. jain, a. k., duin, r. p. w., and mao, j. (2000). statistical pattern recognition: a review. ieee trans. pattern anal. machine intell., 22, no 1, 4–37. langer, h., falsaperla, s., powell, t., and thompson, g. (2006). automatic classification and a-posteriori analysis of seismic event identification at soufrière hills volcano, montserrat. journal of volcanology and geothermal research, 153, 1–10. orozco, m., garcía, m. e., duin, r. p. w., and castellanos, c. g. (2006a). dissimilaritybased classification of seismic volcanic signals at nevado del ruiz volcano. 2nd latin-american congress of seismology, bogotá, colombia, august, cd-rom. orozco, m., garcía, m. e., duin, r. p. w., and castellanos, c. g. (2006b). dissimilaritybased classification of seismic volcanic signals at nevado del ruiz volcano. earth sciences research journal, 10, no. 2, 57– 65. paclík, p. and duin, r. p. w. (2003a). classifying spectral data using relational representation. proceedings of the spectral imaging workshop, graz, austria, april, 3134. paclík, p. and duin, r. p. w. (2003b). dissimilarity-based classification of spectra: computational issues. real time imaging, 9, no. 4, 237–244. pękalska, e. and duin, r. p. w. (2002). dissimilarity representations allow for building good classifiers. pattern recognition lett., 23, no. 8, 943–956. pękalska, e. and duin, r. p. w. (2005). the dissimilarity representation for pattern recognition: foundations and applications. world scientific, singapore, 636pp. pękalska, e., duin, r. p. w., günter, s., and bunke, h. (2004). on not making dissimilarities euclidean. proceedings of structural and statistical pattern recognition, lisbon, portugal, august, 1143–1151. pękalska, e., duin, r. p. w., and paclík, p. (2006). prototype selection for dissimilaritybased classifiers. pattern recognition, 39, no. 2, 189–208. orozco and castellanos. esrj vol. 11, no. 2. december 2007 138 pękalska, e., paclík, p., and duin, r. p. w. (2001). a generalized kernel approach to dissimilarity based classification. j. mach. learn. res., 2, no. 2, 175–211. scarpetta, s., giudicepietro, f., ezin, e. c., petrosino, s., pezzo, e. d., martini, m., and marinaro, m. (2005). automatic classification of seismic signals at mt. vesuvius volcano, italy, using neural networks. bulletin of the seismological society of america, 95, no. 1, 185–196. zobin, v. (2003). introduction to volcanic seismology. elsevier, amsterdam, the netherlands, 302pp. clustering on dissimilarity representations for detecting mislabelled seismic signals at nevado del ruiz volcano geociencias-vol 12-1 sep.vp earth sciences research journal earth sci. res. j. vol. 12, no. 1 (june 2008): 62-71 denoising resistivity phosphate “disturbances” using haar mother wavelet transform (sidi chennane, morocco) saad bakkali1 and mahacine amrani2 1earth sciences department geosciences & environment group faculty of sciences and techniques, abdelmalek essaadi university, tangier, morocco saad.bakkali@menara.ma 2engineering process department faculty of sciences and techniques, abdelmalek essaadi university, tangier, morocco amrani.mahacine@menara.ma abstract wavelet transforms originated in geophysics in the early 1980s for the analysis of seismic signals. since then, significant mathematical advances in wavelet theory have enabled a suite of applications in diverse fields. in geophysics, the power of wavelets for analysis of non stationary processes that contain multiscale features, detection of singularities, analysis of transient phenomena, fractal and multifractal processes, and signal compression is now being exploited for the study of several processes including resistivity surveys. the present paper deals with denoising moroccan phosphate “disturbances” resistivity data? map using the haar wavelet mother transform method. the results show a significant suppression of noise and a very good smoothing and recovery of resistivity anomalies. key words: resistivity, phosphate, disturbance, haar, wavelet, sidi chennane, morocco. resumen la transformada wavelet tuvo sus orígenes a inicios de los 80’s en el análisis de señales sísmicas, que debido a avances matemáticos significativos han permitido su aplicación a diversos campos. la energía de la ondícula; usada en el análisis de procesos no estacionarios con rasgos de múltiples escalas, detecciones de singularidades, análisis de fenómenos transientes, procesos fractales y multifractales, y compresión de señales, es aplicada a diferentes procesos incluyendo sondeos de resistividad. este artículo muestra la atenuación del ruido en el mapa 62 manuscript received april 7, 2008. accepted for publication june 2, 2008. de perturbaciones de resistividad en los fosfatos marroquíes mediante el uso de la ondícula haar en la transformada wavelet. los resultados indican una atenuación significativa del ruido, un buen suavizado y recuperación de las anomalías de resistividad. palabras clave: resistividad, fosfato, perturbaciones, haar, ondícula, sidi chennane, marrueco. introduction geophysical data are often contaminated with noise and artifacts coming from various sources. the presence of noise in data distorts the characteristics of the geophysical signal resulting in poor quality of any subsequent processing. consequently the first step in any processing of such geophysical data is the “cleaning up” of the noise in a way that preserves the signal sharp variations. wavelet transforms are relatively recent developments that have fascinated the scientific, engineering, and mathematics community with their versatile applicability. for geophysical processes, in particular, tools that offer the ability to examine the variability of a process at different scales are especially important. wavelet analysis is an emerging field of applied mathematics that has provided new tools and algorithms for solving such problems as are encountered in fault diagnosis, modelling, identification, and control and optimization (kumar et al., 1994). the theory has acquired the status of a unifying theory underlying many of the methods used in physics and signal processing. the decision as to which representation (expansion) to use for a signal, for example wavelet expansion versus fourier or spline expansion depends on the purpose of the analysis. wavelets have become increasingly popular for analyzing data in the geosciences. wavelets re-express data collected over a time span or spatial region such that variations over temporal/spatial scales are summarized in wavelet coefficients. individual coefficients depend upon both a scale and a temporal/spatial location, so wavelets are ideal for analyzing geo-systems with interacting scales (riedi, 1998). so, the wavelet transform filtering method has become a powerful signal and image processing tool which has found applications in many scientific areas. this method is a widely used technique that is applicable to the filtering geophysical data (kumar et al., 1997). the present paper deals with denoising moroccan phosphate “disturbances” resistivity data map using the haar mother wavelet transform method. (include reference) the results show a high significant suppression of the noise and a very good smoothing and recovery of the resistivity anomalies signal. so the haar wavelet mother transform processing is thought to be a good method to geophysical anomaly filtering and optimizing estimation of phosphate reserves. the geophysical context resistivity is an excellent parameter and marker for distinguishing between different types and degree of alteration of rocks. resistivity surveys have long been successfully used by geophysicists and engineering geologists and the procedures are well established. (include reference) the study area is the oulad abdoun phosphate basin which contain the sidi chennane deposit. the sidi chennane deposit is sedimentary and contains several distinct phosphate-bearing layers. these layers are found in contact with alternating layers of calcareous and argillaceous hardpan. however, the new deposit contains many inclusions or lenses of extremely tough hardpan locally known as “derangements” or “disturbances” (figure 1), found throughout the phosphate-bearing sequence (kchikach et al., 2002). the hardpan pockets are normally detected only at the time of drilling. direct exploration methods such as well logging or surface geology are not particularly effective they interfere with field operations and introduce a severe bias in the estimates of phosphate reserves (figure 2). 63 denoising resistivity phosphate “disturbances” using haar mother wavelet transform (sidi chennane, morocco) the study area was selected for its representativity and the resistivity profiles were designed to contain both disturbed and enriched areas. the sections were calibrated by using vertical electrical soundings. high values of apparent resistivity were encountered due to the presence of near-vertical faulting between areas of contrasting resistivity, and fault zones which may contain more or less highly conducting fault gouge. the gouge may contain gravel pockets or alluvial material in a clay matrix. such anomalous sections are also classified as disturbances. apparent resistivity values in these profiles locally exceeded 200 �?m. (bakkali, 2005; bakkali et al., 2006). the apparent resistivity map (figure 3) obtained from a further survey was considered in fact a map of discrete potentials on the free surface, and any major singularity in the apparent resistivities due to the presence of a perturbation will be due to the crossing from a “normal” into a “perturbed” area or vice versa. in other words, the apparent resistivity map may be considered a map of scalar potential differences assumed to be harmonic everywhere except over the perturbed areas. interpretation of resistivity anomalies is the process of extracting information on the position and composition of a target mineral body in the ground. in the present case the targets were the inclusions called perturbations. the amplitude of an anomaly may be assumed to be proportional to the volume of a target body and to the resistivity contrast with the mother lode. if the body has the same resistivity as the mother lode no anomaly will be detected. thus assumed in fact and in first approach that the resistivity anomalies would be representative of the local density contrast between the disturbances and the mother lode. level disturbance of the anomalous zones is proportionnal to resistivity intensity (figure 4). (bakkali, 2005; bakkali, 2006). 2006 (1)? 2006 (2)? the wavelet analysis approach the wavelet transform is a time-frequency decomposition which links a time (or space) domain function to its time-scale wavelet domain representation. the concept of scale is broadly related to frequency. small scales relate to short duration, high frequency features and correspondingly, large scales relate to long duration, low frequency features (daubechies, 1990). wavelets are functions that satisfy certain mathematical requirements and are used in representing data or other function. in the signal analysis framework, the wavelet transform of the time (or 64 saad bakkali and mahacine amrani * phosphate “disturbance”. space) varying signal depends on the scale that is related to frequency and time (or space) (daubechies et al., 1992). the 2d wavelet method provides information on many more resolution than the former method. it is a powerful tool particularly suitable in denoising, filtering and analyzing problems and potential singularities in geophysical context (foufoula-geogiou et al., 1994) (grossmann et al., 1989). moreover this property is crucial for performing an efficient denoising resistivity anomaly map of the moroccan phosphate deposit “disturbances”. theorical review traditionally, fourier transform has been used to process stationary signals acquired by computers. in this way, the representative spectrum of frequencies is obtained from the time series produced during ac65 denoising resistivity phosphate “disturbances” using haar mother wavelet transform (sidi chennane, morocco) figure 2: (a) location of the studied area in the sedimentary basin of ouled abdoun. (b) section showing the disruption of the exploitation caused by disturbances. (c) stratigraphical log of the phosphatic series of sidi chennane: (1) hercynian massif; (2) phosphatic areas; (3) marls; (4) phosphatic; marls; (5) phosphatic layer; (6) limestones; (7) phosphatic limestone; (8) discontinuous silex bed; (9) silex nodule; (10) dérangement formed exclusively of silicified limestone; (11) dérangement constituted of a blend of limestone blocks, marls and clays; (12) dérangement limit; (13) roads. quisition of the signal by the computer. for non stationary signals, typical of engineering processes, the existing methodologies have not been fully developed. windowed fourier transform, also called short-tine? fourier transform, was first applied using a gaussian type window (walker, 1997). for a given signal f(t), a conventionaly defined signal g(t – t0) is applied to a window of time that moves along with the original signal, forming a new family of functions: fg(t0,t) = f(t)g(t – t0). functions formed this way are centred on and have a duration defined by the characteristic time window of the function g(t). windowed fourier transform is thus defined as: � �f w t f t g t t e dtg fwt, ( ) ( )0 0� � � (1) this transform is calculated for all t0 values and it gives a representation of the signal f(t) in the time frequency domain. if a space function f(x) instead of a time signal, is considered, a representation is given in the spacefrequency domain (meyer, 1993). however as a windowed fourier transform represents a signal by the sum of it sine and cosine functions, it restricts the flexibility of the function g(t – t0) or g(x – x0) making a characterization of a signal and simultaneous location of its high frequency and low frequency components difficult in the time-frequency domain or the space-frequency domain. wavelets transform were developed to overcome this deficiency of windowed fourier transform in representing non-stationary signals. wavelets transform is obtained from a signal by dilatation-contraction and by the translation of a special wavelet within the time or space domain. the expansion of this signal into wavelets thus permits the signal’s local transient behaviour to be captured, while the sine and the cosines can only capture the overall behaviour of the signal as they always oscillate indefinitely (walker, 1997). 66 saad bakkali and mahacine amrani figure 3: a map of resistivity anomalies for ab=120 m using shepard’s method for resistivity data excentrility? signal analysis and the haar wavelet mother in the fourier analysis, every periodic function having a period of 2 and an integrable square is generated by an overlay of exponential complexes, wn(x) = ejnx, n = 9, � 1, � 2, �obtained by dilations of the function w(x) = ejx : wn(x) = w(nx). extending the idea to space for � integrable square functions, the following is defined : � �a b a t b a a b r a, , , ,� � � �� � � �� � � 1 0 (2) the function � is called a mother wavelet, where a is the scale factor and b is the translation parameter. the family of simpler wavelets, which will be adopted in the present work, is that the haar wavelet : � � � ( ) , ( ) , ( ) [ , ] x if x x if x x if x � � � � � � � � � 1 0 1 2 1 1 2 1 0 0 1 for the one-dimensional no stationary function f(x) that decrease to zero when x � , the following assumption is normally adopted : � �p q px x q p , ( ) ( )� 2 22 (3) the scale factor of 2-pq is called the localization or dyatic translation and k is the translation index associated with the localization, where p and q � � (meyer, 1982) proved that wavelet thus defined are orthogonal, i.e., � � � �p q l m p l q m, , , ,� � for p q l m z, , � �where is equal to the scalar product and � refer to the delta function of dirac. thus the function f(x) can be rewritten as follows : f x c xp q p q qp ( ) ( ), ,� � � � � �� � (4) the values of the constant cp,q are obtained by wavelet transform in its discrete form. then is expanded into a series of wavelets with their coefficients obtained from c f f x x dxp q p q p q, , ,( ) ( )� � � � �� � (5) 67 denoising resistivity phosphate “disturbances” using haar mother wavelet transform (sidi chennane, morocco) figure 4: a map of the disturbed noisy phosphate zones corresponding to figure 3. the wavelet transform can also be calculated using special filters called quadrature mirror filters (mallat, 1989). they are defined as a low-pass filter, associated with the coarser scale, and a high-pass filter to characterize the details of the signal. the signal f(x) then is described as: f x c x d xp q p q p q o q qp pq ( ) ( ) ( ), , , ,� � ��� � 0 0 0 � � (6) where c f x x dxp q p q0 0, ,( ) ( )� � � � (7) and d f x x dxp q p q, ,( ) ( )� � � � (8) in the expansion of f(x) by equation 6, the first term represents the approximation of the signal and the second the signal details, filtered by the approximation. the function �p0,q(x) is denominated a scale function or father wavelet, and it is responsible for obtaining the approximation of the signal, while the mother wavelets, �p,q are responsible for the generation of the details filtered by the approximation (polikar, 1999) (wickerhauser, 1994). for the family of haar wavelets, the scale function is �p q x if x0 1 0 1, ( ) [ , ]� � and �p q x if x0 0 0 1, ( ) [ , ]� � . the mother wavelets, responsible for the details in the haar family, are expressed as: � p q p px if q x q p , ( ) ,� � � � � � � � � � 2 2 2 1 2 2 (9) � p q p px if q x q p , ( ) , ( )� � � � � � � � � � � 2 2 1 2 2 12 (10) and �p,q(x) = 0, otherwise. the processing data the resistivity data base is a compilation of 51 traverses at a spacing of 20 m. there were 101 stations at 5 m distance for every traverse, which makes 5151 stations all together in the resistivity survey. we choose the haar wavelet basis for its smoothness and compact support (torrence et al., 1998). we calculated the magnitude square of the haar wavelet transform coefficients (rioul et al., 1991) using origin pro 8 routine (origin pro, 2007) for each resistivity traverse (figure 5). then we deferred all the results to built a 2d wavelet spectrum regular maps which represent in fact filtering and denoising map of the phosphate deposit “disturbances”. since a major potential application of wavelets is in image processing, the 2d wavelet transform is a necessity to be applied as a detector and analyser of singularities like edges, contours or corners (ucan et al., 2000). (tsivourakipapafotiou et al., 2005). results & conclusions figure 6 represents an indicator of the level of variation of the contrast of density between the disturbances and the normal phosphate-bearing rock. the surface modeling of resistivity anomalies is obtained by autosignal routine from our apparent resistivity survey transformed data obtained using the haar wavelet mother response. these procedure enables us to define the surface phosphate disturbed zones. the haar wavelet analysis surface of phosphate deposit disturbance zones modeling as obtained by the above procedure in the study area provided a direct image for an interpretation of the resistivity survey. these method enable us to identify the anomalies area which turned out to be strongly correlated with the disturbances. the use of magnitude square of the haar mother wavelet transform represent an effective filtering method which makes it possible to attenuate considerably the noise represented by the minor dispersed and random disturbances. the overall effect is that of scanning and denoising the anomalous bodies. comparatively to classical approaches used in filtering and denoising the same geophysical data map (bakkali, 2007), the advantage of the haar wavelet transform method is doesn’t introduce significant distorsion to the shape of the original resistivity signal . the haar wavelet output of the apparent resistivity which correspond to the wavelet output of the anomalous phoshate deposit map obtained from such a technical tool represent the crossing dominate area from a “normal” into a “perturbed” area or vice versa. moreover the level of disturbance is very clearly 68 saad bakkali and mahacine amrani shown. the proposed filtering and denoising method using haar wavelet transform tends to give a real estimation of the surface of the phosphate deposit “disturbances” zones with a significant suppression of the noise. the level disturbance resulting from such method is also more defined in all the disturbed zones. we have described an analytical procedure to analyze the anomalies of a specific problem in the phosphate mining industry. the results proved satisfying. data processing procedures as the haar wavelet mother response transform of resistivity data map was found to be consistently useful and the corresponding map may be used as auxiliary tools for decision making under field conditions. 69 denoising resistivity phosphate “disturbances” using haar mother wavelet transform (sidi chennane, morocco) figure 6: wavelet output of the phosphate deposit “disturbances” map given in figure 4 . 350 300 200 150 250 100 50 350 300 200 150 250 100 50 0 100 200 300 400 500 distance in meters r e s is ti v it y in o h m .m r e s is ti v it y in o h m .m disturbed zone phosphate zone phosphate zone b. 350 300 200 150 250 100 50 0 100 200 300 400 500 350 300 200 150 250 100 50 distance in meters r e s is ti v it y in o h m .m r e s is ti v it y in o h m .m disturbed zone phosphate zone disturbed zone n o is e n o is e a. figure 5: example of real resistivity traverse data of the survey (a) and the corresponding haar wavelet transform denoised output (b). references bakkali, s., (2005). analysis of phosphate deposit “disturbances” using the horizontal-gradient responses of resistivity data (oulad abdoun, morocco), earth sci. res. j. vol.9, n°2, p123-131. bakkali, s., (2006). a resistivity survey of phosphate deposits containing hardpan pockets in oulad abdoun, morocco, geofisica internacional, 45 (1), p73-82. bakkali, s. and l. bahi., (2006). cartographie des «dérangements» de séries phosphatées par mesures de résistivités électriques, journal des sciences pour l’ingénieur, j.s.p.i., 6, p1-10. bakkali, s., (2007). enhancement of edges of sidi chennane phosphate “disturbances” using sunshading responses of resistivity data, russian geology and geophysics, rgg, 48 (9), 775–781. daubechies, i., (1992). ten lectures on wavelets, 357 pp., soc. for ind. and appl. math., philadelphia, pa. daubechies, i., mallat, s. and a. s. willsky., (1992). introduction to the special issue on wavelet transforms and multiresolution signal analysis, ieee trans. inf. theory, 38(2), 529–531. daubechies, i., (1990). the wavelet transform, time-frequency localization and signal analysis, ieee trans. inf. theory, 36(5), 961–1005. foufoula-georgiou, e., and p. kumar., (1994). wavelets in geophysics, academic, san diego, calif., 384 pp. grossmann, a., kronland-martinet, r. and j. morlet., (1989). reading and understanding continuous wavelet transforms, in wavelets:time-frequency methods and phase space, edited by j. combes, a. grossmann and p. tchamitchian, 2–20, springer-verlag, new york. kchikach, a., jaffal, m., aifa, t. and l. bahi., (2002): cartographie de corps stériles sous couverture quaternaire par méthode de résistivités électriques dans le gisement phosphaté de sidi chennane (maroc). comptes rendus. geosciences, 334, 379-386. kumar, p., and e. foufoula-georgiou., (1994). wavelet analysis in geophysics: an introduction, in wavelets in geophysics, edited by e. foufoula-georgiou and p. kumar, academic, san diego, calif. 1–43. kumar, p. and e. foufoula-georgiou., (1997). wavelet analysis for geophysical applications, agu, reviews of geophysics, 35, 4, 385–412. mallat, m., (1989). a theory for multiresolution signal decomposition; the wavelet representation, ieee trans. on pami, 11(7), 674-693. meyer, y., (1992). wavelets and operators, cambridge univ. press, new york. meyer, y., (1993). review of an introduction to wavelets by c. k. chui and ten lectures on wavelets by i. daubechies, bull. am. math. soc., 28(2), 350–360. origin pro 8 (2007). about origin pro 8 sro,v8.0724 software, copyright 1991-2007, origin lab corporation, www.originlab.com. polikar, p., (1999). the engineer’s ultimate guide to wavelet analysis , the wavelet tutorial, http:/engineering. rowan.edu/~polikar. riedi, r.h., (1998). multifractals and wavelets: a potential tool in geophysics, seg meeting 1998, technical session on novel transforms, new orleans, la, september 1998. rioul, o., and m. vetterli., (1991). wavelets and signal processing, ieee signal process. mag, 8(4), 14–38. systat (2002). about autosignal v1.6 software, copyrignt 2002 aisn software inc. torrence, ch. and g.p. compo., (1998). a practical guide to wavelet analysis, bull. am. meteorol. soc., 79, 61-78. tsivouraki-papafotiou, b., tsokas, g.n. and p. tsurlos. (2005). wavelet denoising of magnetic 70 saad bakkali and mahacine amrani prospecting data, journal. balkan geophysical society, vol 8, 2, 28-36. ucan, o.n., seker, s., abora a.m and a. ozmen, a., (2000). separation of magnetic fields in geophysical studies using a 2-d multiresolution wavelet analysis approach. journal. balkan geophysical society, vol 3, 3, 53-58. walker, j.s., (1997). fourier analysis and wavelet analysis, notices of the ams, vol 44, 6, 658-670. wickerhauser, m. v., (1994). adapted wavelet analysis from theory to software, 486 pp. 71 denoising resistivity phosphate “disturbances” using haar mother wavelet transform (sidi chennane, morocco) letter from editor a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombiacarlos alberto ríos reyes, oscar mauricio castellanos alarcón and akira takasu 7 material assimilation in a shallow diapiric forceful intrusion: evidence from microstructures and csd analysis in a porphyritic intrusive body, “la línea” tunnel, central cordillera, colombialorena rayo and carlos a. zuluaga 31 mineralization controls and petrogenesis of the rare metal pegmatites of nasarawa area, central nigeriaakintola, o.f. and adekeye, j.i.d. 44 denoising resistivity phosphate “disturbances” using haar mother wavelettransform (sidi chennane, morocco)saad bakkali and mahacine amrani 62 petrogenesis of the metacarbonate and related rocks of the silgará formation, central santander massif, colombian andes: an overview of a “reaction calcic exoscarn”ríos, c.a., castellanos, o.m., gómez s.i. and ávila, g.a. 72 seismic velocity determination in gravel and sands using piezocrystalsarmando luis imhof and juan carlos santamarina 107 earth sciences research journal earth sci. res. sj. vol. 16, no. 2 (dcember, 2012): 3 letter from the editor, greatly appreciated readers of esrj, i am pleased to bring you vol. 16(2) 2012 which contains interesting articles related to applying methods regarding seismic prospecting, geotechnics, geomorphology, seismology and geochemistry in such varied countries as japan, turkey, iran, nigeria and cameroon. procedures for suitably treating seismic signals by using discrete wavelet transform (dwt), or treating clays aimed at modifying their physical attributes form part of the basic contributions regarding seismic engineering and geotechnics’ immense usefulness. this issue also contains the results of geomorphological analysis carried out on the nw coast of kyushu, japan, inferring paleoclimatic conditions and the configuration of the paleoshoreline during the holocene age. such information contributes towards knowledge concerning coastal dynamics and its relationship with ancient populations’ evolution in the japanese archipelago. a systematic analysis concerning instrumental seismicity led to inferring conditions regarding the turkish peninsula’s fracture regime. such work illustrates the concentration of events and their fractal characterisation, possibly being related to this active region of the world’s seismotectonic conditions. four pieces of geochemical work, one of them related to the effects of runoff from a coal mine in iran and the other three related to the environmental characterisation of several mesozoic and palaeozoic formations in nigeria and cameroon, emphasise new geological knowledge emerging from asia and africa. we would like to thank all our authors who submitted work for making their contributions and their willingness to support esrj, and invite the whole community to continue sending us new research work coming within the scope of our geoscientific journal. on a different matter, i would like to take this opportunity to invite the geoscientific community to participate in two interesting events to be held in colombia during 2013: the 14th colombian congress of geology to be held in bogota (july 31st to august 2nd 2013) will offer an appropriate scenario for presenting and discussing recent results regarding knowledge of colombian geology, associated with this traditional meeting, will be held the simposio de exploradores, that will allow to identify new exploration horizons in o&g and mining in latin america and caribbean. and the aapg ice will be held in cartagena (8-11th september) where topics related to latinamerican basins and petroleum systems, unconventional resources, challenges concerning heavy oil, mature fields, deep water exploration and production, and environmental geology will be dealt with within the framework of the o&g industry, as well new academic foci and government interests. the aforementioned events will surely enable new advances to be made which will be of interest to all. we hope to see you in colombia during these events and the scientific workshop and commemorating of the first 100 years of san calixto observatory in la paz, bolivia. other important events in latin america will be: the meeting of the americas, a joint assembly that covers topics in all areas of the geophysical sciences. the meeting will be held at the cancun center in cancun, mexico, 14–17 may 2013. this important meeting is leadership by the agu and also sponsored by institutions like: asociación argentina de geofísicos y geodestas (aagg), asociación latinoamericana de geofísica espacial (alage), asociación latinoamericana de paleomagnetismo y geomagnetismo (latinmag), chilean geological society (cgs), centro regional de sismología para américa del sur (ceresis), geochemical society (gs), mexican association of exploration geophysicists (amge), mexican geological society (sgm), mexican union for quaternary studies (umec), mineralogical society (ms), sociedade brasileira de geofísica (sbgf ), society of exploration geophysicists (seg) and unión geofísica mexicana (ugm). finally i shall mention the official opening of the national laboratories network of geosciences of colombia, an initiative sponsored by the colombian state through the national hydrocarbons agency, colciencias (national science foundation of colombia), five colombian universities (national university of colombia, caldas university, universidad industrial de santander, uptc, eafit) as well other local institutions associated to the gesociences (the colombian geologic survey, the cartographic institute and the international center of physics). a large, complete networking, of last generation, distributed countrywide, that support specialized analyses for the o&g and mining industries, academy and research centers. many of our lectors may find in this networking an interesting source of mediums for new studies along the worldwide. carlos a. vargas editor untitled earth sciences research journal earth sci. res. j. vol. 13, no. 1 (june 2009): 64-73 radiogenic components of the nigerian tar sand deposits akinmosin, a.1, osinowo, o.o.2 and oladunjoye, m. a.2 1department of earth sciences, olabisi onabanjo university, ago-iwoye. nigeria 2department of geology, university of lbadan, lbadan. nigeria corresponding author email: wale.osinowo@mail.ui.edu.ng abstract a combination of factors has prevented the exploitation of the nigerian tar sand deposits to date, among which is the environmental factor which may pose some dangers to both physical and biological components of the area of occurrence. gamma ray spectrometric analysis was carried out on samples of bituminous sand deposits in parts of southwestern nigeria. the aims were to determine the presence and level of radioactivity of selected radionuclides and to assess the possible impact on the environment, and provide geochemical baseline that could be useful in planning appropriate environmental management programs that will reduce potential negative effect of exploiting the resources on the environment. twenty air-dried samples collected for this study were weighed and sealed for 28 days to enable them attain a state of secular equilibrium. they were subsequently analyzed for gamma-emitting radionuclides using gamma-ray spectrometer fitted with a calibrated canberra vertical coaxial high purity germanium detector (hpge) system. the radio nuclides identified with reliable regularity belong to the decay series of naturally occurring radio nuclides headed by 238u, 232th and naturally occurring 40k. result showed that the radiogenic composition of the clay overburden (0.631msvy-1), shale-(0.193msvyr-1), and bituminous sand (0.446msvyr-1), are lower than the normal background value considered harmful to man. key words: spectrometric, gamma ray, radionuclides, nigerian. 64 manuscript receiver: january 07th, 2009. accepted for publication: june 15th, 2009. resumen una combinación de factores ha evitado la explotación de los depósitos de arena bituminosa de nigeria hasta la fecha, en ellos el factor ambiental plantea algunos peligros para los componentes físicos y biológicos de la zona. se realizó un análisis de espectrometría de rayos gamma a las muestras de los depósitos de arena bituminosa en partes del suroeste de nigeria. los objetivos fueron determinar la presencia y el nivel de radiactividad de determinados radio nucleídos para evaluar el posible impacto sobre el medio ambiente, y proporcionar un referente geoquímico que sería útil en la planeación de programas de manejo ambiental que redujeran los posibles efectos negativos de la explotación de los recursos al medio ambiente. para el desarrollo de este estudio se tomaron veinte muestras las cuales fueron pesadas y selladas durante 28 días a fin de que puedan alcanzar un estado de equilibrio secular. un análisis de emisión de rayos gamma por radio nucleídos fue realizado posteriormente con un espectrómetro de rayos gamma equipado con un detector coaxial vertical de germanio de alta pureza (hpge). la radio nucleídos, identificados con una regularidad confiable, pertenecen a la serie de radio nucleídos de desintegración natural, encabezada por 238u, 232th y 40k. los resultados mostraron que la composición radiogénica de la arcilla (0.631msvy-1), de las lutitas-(0.193msvyr-1), y de la arena bituminosa (0.446msvyr-1), son inferiores al valor mínimo, considerado dañino para el hombre. palabras clave: espectrometría, rayos gama, radio nucleídos, nigeria. 1. introduction tar sand is composed of a mixture of bitumen, which makes up about 10-20% and about 80-85% mineral matter including sands, clays and 4-6% water. tar sand has similar composition as the light crude (i.e. h, c and minor amount of s and o). they are believed to have formed from biodegradation and water-washing of light crude due to lack of cap rock. the nigerian tar sand is believed to have formed in a similar process. trace elements occur naturally in rock forming minerals and ore minerals; hence they can reach the environment from natural processes. weathering is a physical and chemical processes that breakdown rocks and then release these trace elements naturally into the environment. the dahomey basin (fig. 1) is a marginal pullapart basin (klemme, 1975) or margin sag basin (kingston et al., 1983), which was initiated during the early cretaceous separation of african and south american lithospheric plates. occurrence of seepage and tar sand deposits over the okitipupa ridge in the dahomey basin provided the initial impetus for oil exploration in nigeria. from the turn of 19th century up till date, no less than over twenty groups comprising public and private ventures have shown degrees of interest. the occurrence of these deposits has been known since early last century, however, intense investigations commenced from mid 70’s till now. the pioneering efforts were initiated by the geological consultancy unit of the university of ife (now obafemi awolowo university). the geology of these deposits, oil saturation and reserve estimates as well as textural characteristics of the associated sands has been described (adegoke et al., 1980, and 1981; enu, 1985). the physiochemical properties of the bitumen in relation to production and processing have been studied (adegoke et al., 1980; oshinowo et al., 1982; ekweozor, 1985; oluwole et al., 1985). the origin of the bitumen has been considered (coker, 1982; ekweozor, 1985). other radiogenic components of the nigerian tar sand deposits 65 relevant studies on the deposit include works done by ako et. al (1983); ekweozor (1986 and 1990); ekweozor and nwachukwu (1989); enu (1987, 1990); enu and adegoke (1984). these works have highlighted relevant aspects of the geochemical and sedimentological characteristic of the deposit. a combination of factors has prevented the exploitation of this resource to date; the most important is the environmental effects that may pose treat to both physical and biological components in the area of occurrence. it is therefore, of utmost importance to know the average background amounts of each metal in the natural uncontaminated geological materials, soils, and waters before assessing the contribution of anthropogenic sources to environmental contamination that may be associated with mining project when it eventually commences. stratigraphy of the dahomey basin the study area lies within latitude 0060381n0060401n and longitude 0040341e0040371e (fig. 2), and falls within the eastern dahomey basin. the reviewed work of omatsola and adegoke (1981) on the cretaceous stratigraphy of the dahomey basin has recognized three formations belonging to the abeokuta group. these are: the ise formation, consisting essentially of continental sands, grits and siltstones, overlying the basement complex uncomformably. neocomian to albian age has been assigned to this formation. overlying the ise formation is the afowo formation, which consists of coarse to medium-grained sandstones with variable interbeds of shales, siltstones and clay. the sediments of this formation were deposited in a transitional to marginal marine environment. turonian to maestritchtian age has been assigned to this formation. the araromi formation consists essentially of sand, overlain by dark-grey shales and interbedded limestone and marls with occasional lignite bands. the formation conformably overlies the afowo formation and maestrichtian to paleocene age has been assigned (omatsola and adegoke, 1981). overlying the abeokuta group conformably is the imo group, which comprises of shale, limestone 66 akinmosin, a., osinowo, o.o. and oladunjoye, m. a. a afowo union volta delta lagos ojo orimedu ise bede gbekebo gilli-gilli benin w. benin flank awaizombe ihuo recent miocene cretaceous miocene a’ oban hills okitipupa highbasement complex dahomey miogeocline eo cen e pa leo cen e cretaceous aeromgnet t c ba se men ba se m en t c om pl ex 0 10000 20000 30000 feet continental crust transitional c-o crust oceanic crust part of triple junction and abakaliki arm transitional c-o crust continental crust afowo bede basement ba se m en t volta delta prolongation of chain and charcot f.z.o. according to emery et al 1978 gilli-gilli oban hills ihuo a’ ch ar co t f .z . warri delta nose ch ain f.z . 0 0 250 150 r . n ig e t port harcourt figure 1: east-west geological section showing the dahomey basin and upper part of the niger delta (after whiteman, 1982) and marls. the two-lithosratigraphic units under this group are: ewekoro formation that consists of thick fossiliferous limestone. adegoke (1977) described the formation as consisting of shaley limestone 12.5m thick which tends to be sandy and divided it into three microfacies. ogbe (1972) further modified this and proposed a fourth unit, which he said is paleocene in age and associated with shallow marine environment due to abundance of coralline algae, gastropods, pelecypods, echinoid fragments and other skeletal debris. akinbo formation lies on the ewekoro formation and it comprises of shale, glauconitic rock bank, and gritty sand to pure grey sand with little clay. lenses of limestone from ewekoro formation grades laterally into the akinbo shale very close to the base. the base is characterized by the presence of a glauconitic rock. the age of the formation is paleocene to eocene. overlying the imo group is the oshoshun formation; it is a sequence of mostly pale greenish-grey laminated phosphatic marls, light grey white-purple clay with interbeds of sandstones. it also consists of claystone underlain by argillaceous limestones with light grey shale at the bottom. there are inclusions of phosphatic and glauconitic materials in the lower part of the formation and the upper part is made up of medium to coarse-grained silty sandstone (adegoke, 1969). the formation is eocene in age (agagu, 1985). the sedimentation of the oshoshun formation was followed by a regression, which deposited the sandstone unit of ilaro formation (kogbe, 1976). the sequence represents mainly coarse sandy estuarine deltaic and continental beds, which show rapid lateral facies change. the coastal plain sands are the youngest sedimentary unit in the eastern dahomey basin, it probably overlay the ilaro formation unconformably, but convincing evidence as to this is lacking (jones and hockey, 1964). it consists of soft, poorly sorted clayey and pebbly sands, of oligocene to recent age. method of study gamma ray spectrometric analysis was carried out on twenty samples. these samples were transferred from the bags into clean air-tight containers and sealed for 28 days in order for the samples to attain secular equilibrium and also to prevent escape of ra-226 gas. moreover, an empty clean container was radiogenic components of the nigerian tar sand deposits 67 a legend river n footpath afowo fm. road settlement 0 1km 4 37e0 6 40n06 40n0 4 37e 0 6 38n 0 6 38n 0 4 34e 0 4 34e 0 idiobilayo idiopopo tar sand outcrop figure 2: location map of the study area showing tar sands outcrop points. equally weighed and sealed for minimum number of days to serve as the background count. various detectors have been used to measure radiation energy deposited in a medium. the choice of a detector for low counting revolves around many factors, but energy resolution and counting efficiency are the most important. two types of detectors used in this work are: inorganic scintillation counter sodium iodide (nai) activated with thalium (ti), and the other is high purity germanium detector (hpge). after a minimum of 28 days for the sealed samples to attain secular equilibrium, each sample was placed on hpge detector enclosed in a lead shield and counted for 10 hours (36 000 s). an identical but empty container that has been sealed for same 28 days was also counted for 10 hours. a standard soil sample that contains certified radioactivity concentration due to 238u, 232th and 40k by weight was also counted for 10 hours. the high resolution of hpge detector made identification of a wide spectrum of ã-ray energy in the samples possible, and the photo peaks observed with regularity in the samples were identified. the area under each peak from the spectra produced was analyzed using a spectra analysis program, sampo-90, which matched ã-energies at various energy levels to a library of possible radioisotopes. the background peak height area was subtracted from the standard peak area in order to get net-peak area values for both the standard and the sample respectively. results and discussion the measured radioactivity concentration (bqkg-1) of gamma emitting radionuclides in the samples is presented in appendix 1. it was observed that the radionuclides identified and quantified from the gamma ray spectral (apendix2) are decay daughter products of naturally occurring radioactive element 238u and 232th. the radioactive daughter product of 238u, are 214bi, 214pb and 226ra, likewise the daughter products of 232th are 208ti and 228ac. 40k that is also naturally occurring but a non series radioactive isotope was also detected and selected for analysis at 1460.75kev. radionuclides were detected in all samples analyzed for, but in varying proportions. for instance, naturally but non-series 40k was detected at 1460.75kev in all the samples except in tar l2, tar l3, tar sand 13, tar sand 17 and tar sand 18; there activity values range between 7.04 ± 0.45bqkg-1 and 1003.91 ± 48.57bqkg-1 the 238u series are more prominent in the samples analyzed than the 232th series.214bi activity values range between 5.02 ± 0.33bqkg-1 and 193.01 ± 19.94bqkg-1; 226ra activity values range between 6.10 ± 0.84bgkg-1 and 148.07 ± 9.81bgkg-1 ; 228ac activity values range between 2.14 ± 0.60bqkg-1 and 53.66 ± 4.44bqkg-1 ; 208ti values range between 1.09 ± 0.38bqkg-1 and 86.81 ± 2.6bqkg-1. the absorbed dose rates in air calculated from the radioactivity concentration of the various samples are given in tables 1 and 2 for the overburden, shale, and tar sand respectively. the corresponding absorbed dose rates have been calculated using the relationship derived by beck et al (1972) which is given as: d = 0.042ac (k) + 0.429ac (u) +0.666 ac (th) where d is in gyhr-1 and represents the absorbed dose rate in air due to the specific activity concentrations of ac (k). ac (u) and ac (th) in (bqkg -1), respectively. the result may be compared with those units set by the international commission on radiological protection (icrp) as shown in table 3. it could be established that the mean dose equivalent calculated for clay overburden 0.631m svyr-1, shale 0.193m svyr-1 and bituminous sand 0.446m svyr-1 are still lower than the normal background value to all humans on earth considered harmful or hazardous to the environment. conclusion the presence of radioactive elements in an environment has been a source of concern, under certain conditions; they can generate a significant radiation that can be injurious to health. the result of the study using gamma spectrometric method of analysis indi68 akinmosin, a., osinowo, o.o. and oladunjoye, m. a. radiogenic components of the nigerian tar sand deposits 69 table 1: absorbed dose rate in tar sand sample (ngyhr-1). sample no. sample description ac (u) ac226 (bqkg-1) ac (th) ac228 (bqkg-1) absorbed dose rate (ngyhr-1) dose equivalents (msvyr-1) s1 tar sand 35.80 ± 6.92 14.22 ± 0.7 24.69 0.216 s6 tar sand 20.48 ± 1.30 11.13 ±0.56 16.11 0.141 s8 tar sand 88.4. ± 3.27 25.11 ± 0.9 54.64 0.473 s9 slightly impregnated bituminous sand nd 2.23 ±0.28 1.48 0.013 s4 tar sand 148.07±9.81 86.81 ± 2.6 121.33 1.06 s10 tar sand 30.8 ± 1.16 17.76 ±0.52 24.63 0.20 s11 tar sand nd 15.19 ±6.08 65.07 0.57 s12 tar sand 48.80 ± 3.53 20.93 ±7.53 90.01 0.80 s13 tar sand 6.10 ± 0.84 nd 25.86 0.23 s14 tar sand 59.61 ± 4.97 32.40 ±9.54 130.33 1.14 s15 tar sand 66.41 ± 5.00 53.66 ±4.44 92.37 0.81 s16 tar sand 39.19 ± 3.11 26.52 ±5.22 78.36 0.68 s17 tar sand 21.42 ± 1.58 16.85 ±1.56 36.30 0.32 s18 tar sand 8.43 ± 0.76 5.97 ± 1.79 11.89 0.10 s19 tar sand 10.73 ± 0.89 7.92 ± 1.61 14.78 0.13 s20 tar sand 16.95 ± 1.26 12.61 ±1.37 30.01 0.26 mean 37.20 ± 2.78 21.83 ±2.63 51.12 0.446 table 2: absorbed dose rate in clay overburden and shale (ngyhr-1) sample no. sample description ac (u) ra226 ac (th) ac228 absorbed dose rate (ngyhr-1) dose equivalents (msvyr-1) s7 clay overburden 107.20±9.61 39.21±1.26 72.09 0.631 s6 shale 31.49 ± 1.24 12.83±0.53 22.04 0.19 s5 shale 10.50 ± 0.60 2.14 ± 0.60 5.92 0.05 s1 shale 57.70 ± 1.91 21.30 ± 0.7 38.91 0.34 mean 51.72 ± 3.43 18.87±0.77 34.74 0.303 mean absorbed dose rate (ngyhr-1), calculated mean equivalent dose is in svyr-1. cated that the radionuclides identified and quantified belong to the naturally occurring decay series of 238u and 232th. the naturally occurring non-decay isotope 40k was also identified and quantified. although the radioactivity levels of the radionuclides were found to be low. the natural gas 226ra, that is generally associated with petroleum from decay of uranium and thorium is noted in the bituminous sands in relatively low concentration. the specific activity of these radionuclides in the bituminous sands is not expected to cause any health hazards now, but the mode of exploitation and extraction can still raise the level if not cautiously chosen and adopted in future. open cast mining method might concentrate and lead to radiation of some of these elements in the overburden during excavation and thus could have severe health implication. references adegoke, o.s. 1969. eocene stratigraphy of southern nigeria, bulleting bureau de research geologic et miners memoirs. vol. 69 pp 22-243. adegoke, o.s. 1977. stratigraphy and paleontology of the ewekoro formation (paleocene) of sw nig. bull. am. paleontol; vol.71, no 295, 375 pp. adegoke o.s., ako, b.d., enu, e.i. 1980. geotechnical investigations of the ondo state bituminous sands. vol. 1. geology and reserve estimate. rept. geological consulting unit, dept. of geology, university of ife. 257pp. agagu, o.k., 1985. a geological guide to bituminous sediments in southwestern nigeria. unpubl. rept. dept. of geology, university of ibadan. ako, b.d., alabi, a.o., adegoke, o.s., and enu, e.i., 1983. application of resistivity sounding in exploration for nigerian tar sand, energy exploration and exploitation vol. 2, no 2. pp.155-164. beck, h. l, decampo, j. and gogolak, c., 1972. in-situ ge(li) and nai(tl) gamma ray spectrometry. rept. hasl-258, u.s atomic energy comm. coker s.j.l. 1982. some aspects of the geology of the bituminous sands of parts of the benin basin. nig. min. and geosc. soc. 19, pp121. ekweozor, c.m. 1985. nigerian tar sands. bitumen chemical properties and their relation to origin, 70 akinmosin, a., osinowo, o.o. and oladunjoye, m. a. table 3: dose limits and their biological effects radiation dose rate duration of exposure likely effects/implication 10,000msv/yr short-term dose immediate illness and subsequent death within few weeks. 1,000msv/yr short-term dose nausea and decreased white blood cell, but not death. 50msv/yr over 5 year the lowest dose rates where there is evidence of cancer being caused. above this, the probability of cancer occurrence increases with dose. 2msv/yr normal background to all humans on earth 0.3-0.6msv/yr artificial sources of radiation, mostly medical equipment. source: uranium information centre, icrp (2002). production and processing. book of abstracts: 21st ann. conf. nig. min. & geosc. soc., jos nigeria, 11th-15th march 1985. ekweozor, c.m. 1986. characteristics of the non-asphaltene products of mild chemical degradation of asphaltene. org. geochem. 10. pp. 1053-1058. ekweozor, c.m., 1990. geochemistry of oil sands of southwestern nigeria in: b.d. ako, e.i. enu (editors), occurrence, utilization and economic of nigerian tar sands:a workshop held in ogun state university ago-iwoye, nigeria on 29-31 may, 1990.published by the nigerian mining and geosciences society, ibadan chapter, 50-62. ekwezor, c. m., and nwachukwu, j. i., 1989. the origin of tar sands of southwestern nigeria. nape bull. 4, 80-94. enu, e.i., 1985. textural characteristic of the nigerian tar sands. sedimentary geology. v. 44, pp 65-81. enu, e.i., 1987. the paleoenvironment of deposition of late maastrichtian to paleocene black shales in the eastern dahomey basin, nigeria geologie en mijinbouww. 66 pp, 15-20. enu, e.i., 1990. textural and occurrence of tar sands in nigeria in: b.d. ako, e.i. enu (editors), occurrence, utilization and economic of nigerian tar sands: a workshop held in ogun state university ago-iwoye, nigeria on 29-31 may, 1990.published by the nigerian mining and geosciences society, ibadan chapter, 11-16. enu, e.i., and adegoke, o.s., 1984. potential industrial mineral resources associated with the nigeria tar sands. 27th international geological congress mosco. book of abstract. vol. vii, pp 345. international commission on radiological protection (icrp), 2002 jones m.a and hockey r.o., 1964. the geology of part of south western nigeria.nig.geol.surv. bull.no 31, 1-101. klemme, h.d., 1975. geothermal gradients, heatflow and hydrocarbon recovery. in : a.g. fischer and s. judson (eds), petroleum and global tectonics. princeton, new jersey, princeton univ. press, pp. 251-304. kingston, d.r., dishroon, c.p. and williams, p.a. 1983. global basin classification system. american association of petroleum geologists. bulletin 31, 87p. kogbe, c.a., 1976. geology of nigeria. (second revised edition). published by rock view (nig.) 455pp.c.a kogbe (editor). ogbe f.g.a. (1972): stratigraphy of strata exposed in ewekoro quarry, southwestern nigeria. in: africa geology pp305-322. oluwole, f.f., adegoke, o.s., kehinde, l.o., borishade, a.b., ako, b.d., ajayi, t.r., enu, e.i., nwachukwu, j.i., coker, s.j.l., wallace, d., asubojo, o.i and ogunsola, o., 1985. chemical composition of bitumen extracts from nigerian tar sands. in: proceedings of 3rd intern. confer. on heavy crude and tar sands; long beach california; chap. 33, pp 373-379. omatsola m.e and adegoke o.s. (1981): tectonic and cretaceous stratigraphy of dahomey basin jour. of min. geol. vol.54, pp 65-87. oshinowo, t., ademodi, b. and adediran, a.s., 1982. bituminous tar sand of nigeria. analysis of oils-part 1. jour. nig. soc. chem. engrs. 1, 44-48. taylor, s.r., 1964. abundance of chemical elements in the continental crust; a new table. geochim cosmochim acta, 28. 1273-1285. whiteman, a.j., 1982. nigeria: its petroleum geology, resources and potential, vol. 2, graham and trotman, london. radiogenic components of the nigerian tar sand deposits 71 72 akinmosin, a., osinowo, o.o. and oladunjoye, m. a. appendix 1: activity concentration in bgkg-1 (hp ge detector) sample description k40 (1460.75kev) u238 th 232 absorbed dose rate (ngyhr-1) equivalent dose rate (msvyr-1) bi214 (609.31 kev) pb214 (351.91 kev) ra226 (911.9 kev) ac228 (911.9 kev) ti208 (583 kev) shale l1 10.32 ± 0.78 51.69 ± 1.9 73.57 ± 7.7 57.7 ± 1.91 21.3 ± 0.7 21.3±0.76 104.71 0.917 tar l2 nd 48.32 ± 2.0 75.52 ± 8.1 35.8 ± 6.92 14.02 ± 0.7 14.00±0.7 87.14 0.76 tar l3 nd 20.48 ± 2.0 82.29 ± 1.3 20.48 ± 1.3 1.13 ± 0.56 11.12±0.59 67.7 0.59 shale s4 70.19 ± 4.04 180.65 ± 5.8 193.01 ± 19.9 148.07±9.81 86.81 ± 2.6 86.81 ± 2.6 242.41 3.009 shale s5 18.16 ± 1.14 10.50 ± 0.6 10.46 10.50 ± 0.95 2.14 ± 0.6 2.137± 1.6 17.09 0.14 shale s6 7.04 ± 0.45 31.49 ± 1.29 35.1± 3.79 31.49 ± 1.24 12.83±0.53 12.82±0.53 59.38 0.52 clay s7 59.903± 0.5 150.36 ± 4.9 165.88 ± 17.2 107.2 ± 9.69 39.21±1.26 19.11± 1.2 236.39 2.07 tar sand 6 92.69 ± 57.5 88.4 ± 3.2 132.46 ± 14.4 88.4 ± 3.27 25.11 ± 0.9 25.1 ± 0.9 170 1.49 tar sand s7 13.13 ± 0.78 nd nd nd 2.23 ± 0.28 2.23 ± 0.27 16.73 0.19 tar sand s8 376.69±18.55 30.79 ± 1.16 42.44 ± 4.47 30.8 ± 1.16 17.16±0.59 17.16±0.58 83.3 0.73 tar sand s9 658.72±29.77 28.66 ± 3.39 15.7 ± 2.4 nd 15.19±6.08 12.39±1.76 65.07 0.57 tar sand10 847.22 ± 38.4 25.79 ± 3.54 12.52 ± 2.72 42.8 ± 3.53 20.93±7.53 8.54 ±1.89 90.01 0.80 tarsand11 518.58±22.75 3.5 ±1.73 nd 6.1 ± 0.84 nd nd 25.86 0.23 tar sand12 1003.91±48.57 39.51 ± 4.68 37.74 ± 3.96 59.61 ± 4.97 32.4 ± 9.54 11.84±2.44 130.33 1.14 tar sand13 nd 16.75 ± 1.98 26.8 ± 0.81 66.41 ± 5.0 53.66±4.44 14.23 ± 1.0 92.37 0.81 tar sand14 458.96 ± 20.58 23.83 ± 2.73 20.36 ± 1.99 39.19 ± 3.11 26.52±5.22 8.48 ± 1.29 78.36 0.68 tar sand15 50.93 ± 3.28 9.68 ± 1.14 15.58 ± 0.4 21.42 ± 1.58 16.85±1.56 4.36 ± 0.36 36.30 0.32 tar sand17 nd 2.20 ± 0.74 5.02 ± 0.33 8.43 ± 0.76 5.97 ± 1.79 1.81 ± 0.55 11.89 0.10 tar sand18 nd 4.11 ± 0.67 5.63 ± 0.28 10.73 ± 0.89 7.92 ± 1.61 1.09 ± 0.38 14.78 0.13 tar sand19 92.45 ± 4.69 8.28 ± 0.97 8.28 ± 0.25 16.95 ± 1.26 12.61±1.37 4.68± 0.36 30.01 0.26 mean 213.944±12.589 34.749±2.21 47.918±2.807 40.104±2.91 21.20±2.393 13.960±1.007 82.992 0.784 appendix 2 radiogenic components of the nigerian tar sand deposits 73 1600 1400 1200 1000 800 600 400 200 0 -200 0 1000 2000 3000 4000 5000 channels spectrum 1 c o u n ts /c h a n n e l 1600 1400 1200 1000 800 600 400 200 0 -200 0 1000 2000 3000 4000 5000 channels spectrum 7 c o u n ts /c h a n n e l this study applies the ecological green equivalent approach to evaluating the land use structure of mt. yuelu scenic area in hunan province, china. a mathematical model is established based on land use and land cover data, and then arcgis used to extract the spatial extent of the ecological green equivalent within each of the relevant elements. results show that the area has a relatively reasonable land use structure even though the forest coverage rate is slight below the optimum. the overall green equivalent (1.13) was higher than the optimum forest coverage ecological green equivalent (1.00). the distribution of forest was uneven, with most of the forest within the site; the area’s land use structure could thus be improved by extending the green area outside of mt. yuelu. we conclude by reiterating that landscape and infrastructure development should consider ecological system conditions. earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): n1 -n5 abstract keywords: ecological green equivalent, land use structure, landscape, scenic area, forest, china. issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.54138 e n v i r o n m e n t a l g e o l o g y xi chen 1,2,3, ke-lin wang 2, wei-jun zhou 4, hong-bin li 4, muhammad aqeel ashraf 5 1. college of bioscience and biotechnology, hunan agricultural university, changsha, hunan 410128, china 2. key laboratory of agro-ecological processes in subtropical region, institute of subtropical agriculture, chinese academy of sciences, changsha, hunan 410125, china; kelin@isa.ac.cn; (for correspondence) 3. biological research institute of chenzhou, chenzhou, hunan 423000, china 4. college of resources and environment, hunan agricultural university, changsha, hunan 410128, china 5. faculty of science and natural resources, university malaysia sabah 88400 kota kinabalu sabah, malaysia optimization of land use at mt. yuelu scenic area: an analysis using the ecological green equivalent record manuscript received: 12/11/2015 accepted for publication: 27/02/2016 how to cite item chen, x., wang, k., zhou, w., li, h., & ashraf, m. a. (2016). optimization of land use at mt. yuelu scenic area: an analysis using the ecological green equivalent. earth sciences research journal, 20(1), n1-n5. doi: http://dx.doi.org/10.15446/esrj.v20n1.54138 introduction many different methods presently exist to evaluate the utilization of ecotourism resources. focused on the environmental impact of tourism, the ecological footprint of tourism and other concepts assess the environmental carrying capacity related to tourism (wu, 2013). such information provides an essential basis for the sustainable development of ecotourism, especially in china, but these methods have limitations and shortcomings (gu and he, 2014). for example, the ecological footprint of tourism is hard to calculate. given the inherent mobility involved in tourism, particularly regional tourism, it is hard to define the ecological footprint accurately. moreover, tourism ecological footprint calculations often focus on analyzing the current multi-part footprint of energy, shopping, construction land, food, and other items, but the consumption footprint related to contaminated waste in environmental systems is not considered (hu and zheng, 2008; yang et al, 2012). in this study, we try to develop a new evaluation method by using ecological green equivalent analysis to analyze and evaluate land use at hunan province’s mt. yuelu scenic ecotourism area. this work should provide a scientific basis for the further development and utilization of ecotourism zones. the rest of the paper proceeds as follows. section 1 offers an overview of the green equivalent concept while section 2 describes the research methods. section 3 then presents results, which are used to inform the conclusions and recommendations of the final section. 1. the green equivalent concept and land type classification forests, the largest ecosystem on land, carry out key ecosystem services including climate regulation, soil and water conservation, and air purification. in the process of developing and using tourism resources, it is often difficult to avoid damaging forest ecosystems; such damage, in turn, affects the sustainable development of a region’s ecotourism industry. some scholars have thus suggested providing compensation for the destruction of or harm to regional forest ecosystems and their accompanying services (hu and zheng, 2008; hardin and jensen, 2007). developing efficient and practical compensation measures, however, requires a means of estimating comparable ecological function “equivalents.” the concept of “green equivalent” has been put forward to meet this need (li and bian, 2012). n2 xi chen, ke-lin wang, wei-jun zhou, hong-bin li and muhammad aqeel ashraf the “green equivalent” is defined as the minimum amount of green space required to ensure sufficient green biomass to deliver the same ecosystem services essentially—that is, the ecological function “equivalent” (mao, 1998). this biomass may occur in the form of a variety of green vegetation types, such as lawns, gardens, and farmland. the green equivalent is thus the ratio of the amount of other green vegetation to the green forest area that can deliver the same services (zhang et al., 2012). currently, ecological green equivalent analysis is mainly used to optimize land use structure when constructing regional development resources because land is typically the basis of the ecosystems on which tourism relies. planners can thus analyze the ecological green equivalent to optimizing the use of tourism land resources or use it to evaluate the green resources available within the land use structure of the tourism ecosystem after changes are made (sun, 2013). under the ecological green equivalent concept, traditional land use patterns can be divided into three categories (zhao et al., 2011; liu et al., 2002). first, there is ecological green equivalent land, characterized by its ability to be quantified regarding ecological mechanisms carried out by forest ecosystems. this includes farmland, woodland, orchards, and certain types of unused land. second is the implicit green equivalent land. quantification of this is difficult and must be done via qualitative research. the third is the non-green equivalent land (table 1). table 1. land classification under the ecological green equivalent approach. 2 research methods 2.1 description of the study area mt. yuelu scenic area is located in the changsha city of hunan province (longitude 112º44’-112º48’, latitude 28º20’-28º27’). it is about 4 km long in the north-south direction and about 1.5-2 km wide in the eastwest direction; the main mountain area is 8 km2, and the main mountain peak, bi xu, has an altitude of 300.8 m above sea level. the zone includes several scenic areas, including yuelu mountain, tianma mountain, orange isle, peach blossom mountain, shi jialing, zhai zi mountain, and xianjia lake. these combine with the surrounding protection zone for a total area of 38.4 km2. the area enjoys a humid subtropical monsoon climate, with an average annual temperature of 17ºc, rainfall of 1200-1400 mm, and relative humidity of 80%. it is one of china’s rainier zones (yang, 2004). the weather is damp or rainy and hot in spring and summer: july average and higher temperatures are 28.6ºc and 40.6ºc, respectively. in late autumn and early winter there is less rainfall, and january temperatures average 4.6ºc, plummeting to extreme lows of -11.3ºc. vegetation types in mt. yuelu scenic area are dominated by typical subtropical evergreen broadleaf forest and subtropical warm coniferous forest. according to species lists, the mountain hosts 285 kinds of forest plants (excluding foreign species) from 68 families and 159 genera; this accounts for 59.7%, 28.9%, and 14.3% of the respective families, genera, and species of hunan province’s native forest plants (yang, 2004). 2.2 methods of ecological green equivalent calculation and ecosystem analysis 2.2.1 spatial analysis of the study area in this study, we use arcgis 9.3 to extract the relevant factors under the spatial extent of ecological green equivalent given by the urban land use map (scale 1:10000, 2005); we combine this with the national land classification (trial) to classify the area of each land use type (zhao et al., 2011; xiao, 2013). using forest ecosystems as a standard, we then calculate the total ecological service function scores for all kinds of green land cover to obtain the green equivalent for each type of ecological system. by estimating the actual total area of green equivalent, we identify the lowest ecological optimization criterion to meet the minimum needs of the research area. finally, the structure of the ecological land can be adjusted within the research area to achieve ultimately a reasonable regional land resource utilization structure and promote the sound development of ecotourism resources (luo and zhang, 2007). 2.2.2 calculate total ecological service function score land use patterns change, as do ecosystem processes and functions. changes in land use patterns can have profound impacts on regional climate, soil, water, and biological processes. therefore, considering the regulation and protection of natural, economic, and social subsystems, criteria must be used to select typical, representative, and systematic indicators. table 2 lists 17 kinds of ecological functions of ecosystems, encompassing five aspects: the atmosphere, water, soil, biological processes, and natural disasters. we used the evaluations of japanese experts, obtained via surveys (mao, 1998), combined with the actual mt. yuelu context, to determine assessment scores of the various environmental features in the ecosystem (mao, 1998). the scores in table 2 take into consideration the environmental sub-functions with appropriate weights to determine the cumulative summation of the total value of ecosystem services that derive from each sub-function (or index value): in the formula, the total value of ecosystem services is p, f is an index value, and the number of indicators in the index system is i. based on this, the ecological services value of forestland is estimated as 169.1, considering the full year and area. under the same circumstances, the ecological services value of natural grasslands is 132.26, that of grasslands is 121.84, that of wetlands is 127.56, that of dry land is 113.55, and the garden plots is 124.53 (zhao, 2006). n3optimization of land use at mt. yuelu scenic area: an analysis using the ecological green equivalent table 2. the scores of different ecological system by environmental protection features notes: evaluation significance: optimal is 10, acceptable is 7.5, and minimal is 5. “atmospheric composition improvement -1” indicates co2 sequestration functions. “atmospheric composition improvement -2” indicates o2 sequestration functions. “atmosphere purification -1” and “atmosphere purification -2” indicate dust capture and toxic gas cleansing ecological service functions, respectively. 2.2.3 determining average ecological green equivalents in the ecosystem according to the different estimated land ecosystem service values, we assume that optimal woodland forest coverage corresponds to green equivalent 1, under the premise of year-round, wholly green coverage. it follows that xi= fi /fl the ecological services value score for woodland is fl and the total score for all ecological services of the green surface cover of class i is fi. thus, the surface green cover of class i regarding ecological green equivalent in the ecosystem is fi. based on zhao et al. (2006), the green equivalent of garden plots is 0.73, that of the natural meadow is 0.76, grassland is 0.73, paddy land is 0.77, and ordinary upland is 0.68. however, due to climatic differences between regions, there are both interregional and intertemporal differences in green equivalents. given the actual circumstances in each region, the above green equivalent results must be multiplied by a coefficient representing the growing period about a one-year-period. according to expert-recommended values (table 3), when three crops are grown in a year in subtropical regions, the growing period parameter is 0.83 (liu et al., 2002). table 3. average green equivalent of cultivated land and grassland notes: relative growing period parameter = (length of growing period (months) / 12). the harvest frequency and growing period are based on expert advice. 2.3 model for optimizing the land utilization structure based on the ecological green equivalent we use the ecological green equivalent of optimal forest cover as the basis for ecosystem optimization. the total area of the research is st, the optimal forest coverage is rmax, regional forest area under the optimal forest coverage requirements is sf, actual forest area is srf, si is the area of land used for land use type i (i is an integer, i ϵ [1, 4]), and ecological green equivalent is gi. we can then use soil and precipitation data from the meteorological department and the agricultural sector to consider the benefits arising from ecological protection and apply the method of yuanbo (zuo and ruan, 2009; gong et al., 1996) to determine the optimal forest cover rate (rmax) (liu eat al., 2009): rmax% =(p × s1) / (w × st) × 100% in this formula, the region’s total land area is st, the daily maximum precipitation (t/hm2) within a year is p. s1 = st -(surface area of urban, mining, transport, and paddy field land) (hm2), and the forest soil’s saturated water storage capacity per unit area is w (t/hm2) (zuo and ruan, 2009). sf = st × rmax, and the corresponding ecological green equivalent is =1. the ecological green equivalent of the actual woodland area is given by: the region’s green equivalent x is calculated as follows: comparing the sizes of x and a determines whether the measure is on target. if ≥ , then the standard is met. if < , the areas of various types of green space must be adjusted, with the index recalculated for verification (liu et al., 2002). 3. results the tourist ecosystem is shaped by a combination of factors involving land, the environment, technology, policy, human needs and behaviors, and ecology. changes to any one factor will lead to changes in the existing ecosystem, altering its balance. although the tourism ecosystem itself has a certain ability to recover, this is limited. therefore, if the tourism industry engages in the excessive pursuit of economic benefits, the extent of land use for economic purposes will be overly emphasized. such a process ignores environmental protection, eventually harming tourism itself. through calculating the green equivalent of the tourism ecosystem (wu et al., 2011; n4 xi chen, ke-lin wang, wei-jun zhou, hong-bin li and muhammad aqeel ashraf xiang and meng, 2013), we will be able to address the problem of properly using tourism ecosystem land resources (via structural optimization) through a combination of qualitative and quantitative analysis. according to the region’s endowment of natural resources and its environment, we determine the ecological minimum standards; we then use ecological green equivalent calculations for the area to protect the lowest stock of land required to maintain ecosystem functions. this procedure ensures that the degree of regulated land use does not exceed the capacity of the ecosystem, thereby achieving the coordination between ecological, social, and economic motivations that is the goal of sustainable development (lu et al., 2013). 3.1 ecological green equivalent by drawing on previous research defining forest ecosystem services values, we quantitatively evaluate the ecological service value of various types of green spaces in the ecosystem (hardin and jensen, 2007). to investigate based on a “considerable green amount” of land area regarding the relationship between forest and farmland and gardens and lawns, we measure the ecological green equivalent of cultivated land, grasslands, and garden plots. this quantifies, compared to forests, the ecological functions of a given area of agricultural land (including cropping land, grassland, and garden plots) (sun, 2013). calculating the green coverage within the scenic area, we can thus explore how to optimize the structure of ecological standards within the scenic area. this insight can strengthen the construction of scenic environments (zhang et al., 2011; zhang et al., 2012; li et al., 2014). based on the above approach, we determine that mt. yuelu scenic area has an ecological green equivalent of x = 1.13; as this is greater than 1.00, it implies a proper structure of land use. among the contributors, the ecological green equivalent of cultivated land is 0.20, that of garden plots is 0.10, and that of forest land is 0.84. table 4. green equivalents in mt. yuelu scenic area despite this positive indication, to accelerate the development of the area in recent years, a large number of human landscapes have been built, causing massive destruction to forest vegetation. the woodland area and the total ecological green equivalent have both fallen. as a result, habitats have been destroyed for key species in the biological community and biodiversity has been reduced. community characteristics and ornamental values have also declined, with inadequate connections within the area’s greenbelt: some roads lack roadside greenery and proper green infrastructure or rely on solely a single species. 3.2 land use structure land resources support various types of ecotourism resources. therefore, this study seeks to apply the concept of ecological green equivalent to optimizing the allocation of ecotourism land use structures. this represents an application of systems theory thinking and the principles of ecosystem ecology. we thus ecologically optimize land use for ecotourism in the mt. yuelu scenic area to safeguard the ecological environment while balancing economic, social, and ecological benefits. according to a mapping survey of land use within the scenic area, which covers an area of 38.49 km2, the land use structure is as given in table 5. table 5. land use in the mt. yuelu scenic area using this current land use structure for mt. yuelu scenic area and basic data on the area (e.g., p = 1165 t/hm2, w = 2500 t/hm2, and s1 = 2500 hm2), we calculate that optimal ecological protection benefits are achieved when there is 30.3% forest cover. although mt. yuelu scenic area’s forest coverage rate was 25.4%, lower than optimal, results show that its overall green equivalent (1.13) is higher than the green equivalent of the optimal forest cover rate (1.00), indicating that the scenic area currently has an extremely reasonable land resource utilization structure. however, mt. yuelu’s woodlands are unevenly distributed; on the mountain itself, for example, total forest area is 5.33 km2 of an area of only 8.00 km2; 20.8% of the total scenic land area thus accounted for 54.5% of the woodland area. 4 conclusions and recommendations 4.1 conclusions the essence of landscape pattern optimization is using landscape ecology principles to achieve reasonable land use by jointly considering natural and social data. by analyzing the appropriate and reasonable distribution pattern of landscape types, one can adjust the landscape distribution (in both location and scope) to maximize the total ecological value of the landscape. the results of this research indicate that the mt. yuelu scenic area has a proper land resource utilization structure but with less-than-optimal forest cover. this suggests a need for the area’s managers to optimize land use through structural adjustment. this study investigated ecological green equivalent to considering ecotourism resource allocation. further research could consider combined optimization of both the land quantity resources for ecotourism and the spatial structure based on a comprehensive model of quantity optimization and spatial pattern analysis. to facilitate realtime monitoring of mt. yuelu’s ecotourism resources and environment and provide dynamic warnings on ecological security, research should also take advantage of “3s” technology for analysis and monitoring. 4.2 recommendations to promote the ecological health of the mt. yuelu scenic area, we suggest four areas for further optimization based on the problems found in this research. first, this study is the first to apply the green equivalent analytical model to ecotourism resources for the purpose of determining scientific and operable land use planning guidance (jim, 2001; li et al., 2012). results indicated that the area’s land resource utilization structure was reasonable, but the area’s woodlands are unevenly distributed. their scope should thus be increased outside of the core scenic area to support the forest coverage rate to reach the desired level. moreover, the use of the forest ecological environment quality as the overall situational indicator reflects regional limitations; there is thus a need for continued research to improve the computational model and select better indicators (hu and zheng, 2008; yang et al, 2012). it should also be noted that the quality of woodland is not entirely reflected in the regional green equivalent, highlighting a further need to improve the index method as part of future research efforts (xiao, 2013). n5development of tropical lowland peat forest phasic community zonations in the kota samarahan-asajaya area, west sarawak, malaysia second, optimization of land use structures helps improve ecosystem functions and services, thereby realizing regional ecological security objectives. as the scenic area’s woodlands are unevenly distributed, woodland resources should be reallocated, and the green area outside of the mt. yuelu scenic area should be increased. by optimizing the structure of construction land and safeguarding the ecological health and regional ecological security through the land use pattern (yang et al, 2012), the scenic area’s ecological green equivalent can be targeted to meet requirements while maximizing eco-efficiency in line with the needs of economic and social development. third, we note that, in recent years, the mt. yuelu area’s environment has been damaged, impinging upon the sustainable social, economic, and environmental development. the development of the area’s cultural landscape through the construction of infrastructure such as roads must thus be done in compliance with ecological priorities (jim, 2001). this should be based on existing roads and should aim to widen green spaces, optimize species distributions, and increase the scenic area’s green space. finally, these results have also revealed that there is scope for improving the consistency of land use and ecological landscape functions in the mt. yuelu area. the landscape spatial pattern should be aligned to local conditions, with space allocation made in a manner that is reasonable for a given project, fully aligning the project’s properties and functions with the functions and spatial characteristics of the land, thus achieving ecological coordination, system harmony, and scenario blending (hardin and jensen, 2007). future work in the area should also consider the renewal and transformation of the area’s tree species distributions to include some ornamental (e.g., flower, foliage, and concept tree) species, such as magnolia, ginkgo, and elaeocarpus decipiens. the goal of this work should be to enhance the natural colors of the forest and thus achieve coordination between ecological functions of the forest and its landscaping and greening structure, thereby improving the area’s landscape efficiency. simultaneously, it may be appropriate to develop additional tourism options, such as agriculture and horticulture tours (e.g., tourist farms), and to adjust the industrial structure actively to promote the balanced development of the ecosystem (ye et al., 2012; zhu, 2011; chi et al., 2014). acknowledgements this work was supported by the national natural science foundation of china (grant no. 41071340). our special thanks are owed to professor ke-lin wang and professor wei-jun zhou, for their helpful discussions while preparing the manuscript. references chi, w., shi, w. and kuang, w. (2015). spatio-temporal characteristics of intraurban land cover in the cities of china and usa from 1978 to 2010. journal of geographical sciences, 25(1), 3-18. doi: 10.1007/s11442-015-1149-z gong, y.b, zhang, j. and chen, l.w. (1996). sichuan basin hilly district level best protection benefits of forest cover quantitative research. journal of sichuan agricultural university, (2), 231-236. gu, w.p. and he, q.y. (2014). advantages, opportunities, and countermeasures dongting lake ecological economic zone development of tertiary industry. research of agricultural modernization, 35(3), 281-285. hardin, p.j. and jensen, r.r. (2007). the effect of urban leaf area on summertime urban surface kinetic temperatures: a terre haute case study. urban forestry & urban greening, 6(2), 63-72. hu, h.k. and zheng, c.y. (2008). environmental impact assesment of land use change based on ecological green equivalent--a case study of meizhou. research of agricultural modernization, 29(6), 743-746. jim, c.y. (2001). managing urban trees and their soil envelopes in a contiguously developed city environment. environmental management, 28(6), 819-832. doi: 10.1007/s002670010264 li, b. and bian, j. (2012,). optimization of land use structure based on ecological green equivalence in fuling district of chongqing. journal of hainan normal university: (natural science edition), 25 (2), 212-220. li, q.q., zhang, x. and gao, x.s. (2014). counties in the hilly land use degree coupled with the benefits of coordination pattern analysis. research of agricultural modernization, 35(1), 97-102. li, z., wang, j. and bai, z.k. (2012). guizhou province land use and ecosystem service value and gray prediction. advances in geographic, 31(5), 577-583. liu, y.f., ming, d.p. and yang, j.y. (2002). the land use structure based on ecological green equivalent optimization. wuhan university journal (information sciences), 27(5), 493-500. liu, y.j., ma, h.l. and ma, y.g. (2009). changes in environmental impact assessment based on ecological green equivalent of meizhou city land use. southern gardening, 20(2), 34-36. lu, m., fei, w. and li, d.h. (2013). based on the green equivalent of jinan city green building programs of ecological compensation. journal of shandong architecture university, 28(2), 95-100. luo, z.j. and zhang, j. (2007). ecological green equivalent and its application to land use structure optimization a case study of xinjian county. journal of jiangxi agricultural university, 29(5), 851-856. mao w.y. (1998). an introduction to ecological impact assessment. beijing: china environmental science press. sun, y. a. (2013). the land use structure based on ecological green equivalent taking yongchuan district of chongqing city chongqing. southwestern university. wu, d. h. (2013). land use master plan environmental impact assessment. nanning: guangxi university. wu, z.j., ni, j.p. and wei, c.f. (2011). cultivated land use change and its mechanism chongqing three gorges project reservoir under stress. journal of southwest university (natural science edition), 33(3), 50-57. xiang, y.y. and meng, j.j. (2013). the land use structure-based ecological benefits of wuhan city circle optimization. resources and environment in the yangtze, 22(10), 1297-1304. xiao, q. (2013). on optimizing land use structure based on ecological green equivalent in yongchuan district, chongqing. journal of southwest china normal university (natural science edition), 38(8), 46-49. yang, d., zheng, f.j., pan j h., zhou, w. and liu, h. (2012). improved topsis method in the evaluation of rationality of land use structure--a case study of lanzhou city. chinese journal of soil science, 43(1), 120-124. yang, d.h. (2004). the landscape analysis and evaluation research of mt. yuelu scenic spot. changsha: central south forestry university. ye, t.p., liao, w. and li l. (2012). the land-use structure and pattern of danjiangkou reservoir optimization. research soil and water conservation, 19(1), 11-16. zhang, b.l., yang, q.y. and feng, y.b. (2011). t effects of land use change on regional economic development in chongqing. journal of southwest university (natural science edition), 33(4), 151-155. zhang, g., he, d. and yang, q. (2012). optimization of the sanxia reservoir area land use structure based on ecological green equivalent a case study of yunyang county in chongqing city. journal of southwest agricultural university (social sciences), 10(8), 5-9. zhang, x.h., sun, q. and zhang, h.b. (2012). typical urban land use change and its ecological effects a case study in daqing. research of agricultural modernization, 34(4), 465-469. zhao, d., li, f. and wang, r.s. (2011). urban land use structure optimization based on ecological green equivalent a case study in ningguo. journal of ecology, 31(20), 6241-6250. zhao, y.q. (2006). regional landscape ecological planning. chongqing, southwestern university. zhao, y.q., yang, q.y., yan, l. and yang, y. (2006). the application of ecological green equivalent on land use structure optimization -a case study of jiangbei district, chongqing. journal of southwest china normal university (natural science edition), 31(1), 170-174. zhu, g.f. (2011). ecological landscape pattern changes nanjing xianlin university town planning and optimization research. nanjing: nanjing agricultural college. zuo, z. and ruan, x.l. (2009). the application of ecological green equivalent on land use structure optimization and example. heilongjiang environmental journal, 33(3), 11-13. structured soils, such as residual laterites or volcanic ashes, abound in tropical regions. the hydromechanical behavior of these soils is clearly related to their structure. new techniques based on microstructural studies constitute a powerful tool for investigating these soils. this work presents an experimental investigation that documents the relationships between the microstructure and mechanical behavior of these soils. this study focuses on the microstructural behavior of brasília clay, a structured, highly porous tropical soil (n>50%) with a bimodal pore size distribution (psd). the microstructure was investigated using psd analyses of different structural states and void ratios. the mercury intrusion porosimetry (mip) technique was used to explore the psd of samples under various conditions, i.e., natural, compacted, slurry or consolidated states. modeling of the psd curves was performed using the bimodal van genuchten curve, which permits the linking of psd curves to the water retention curve. we observed that loading and compacting did not affect the micropores of this soil and concluded that the changes occurred entirely within the macropores. abstract keywords: hydromechanical behavior, microstructural behavior, pore size distribution. study of the relationship between the hydromechanical soil behavior and microstructure of a structured soil issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n2.65640 manoel porfirio cordão neto1, oisy hernández2, raydel lorenzo reinaldo3, camila borges4, bernardo caicedo5 1. department of civil and environmental engineering, university of brasilia, brazil 2. department of civil engineering, federal university of west bahia, brazil 3. department of civil engineering, federal university of tocantins, brazil 4. department of civil engineering, federal institute of goias, brazil 5. geotechnical post-graduation program, university of andes, colombia earth sciences research journal earth sci. res. j. vol. 22, no. 2 (june, 2018): 91-101 los suelos estructurados, tales como las lateritas residuales o las cenizas volcánicas, abundan en las regiones tropicales. el comportamiento hidromecánico de estos suelos está claramente relacionado con su estructura. las nuevas técnicas basadas en estudios microestructurales constituyen una poderosa herramienta para investigar estos suelos. este trabajo presenta una investigación experimental que documenta las relaciones entre la microestructura y el comportamiento mecánico de estos suelos. este estudio se centra en el comportamiento microestructural de la arcilla de brasília, un suelo tropical estructurado y altamente poroso (n> 50%) con una distribución de tamaño de poros (psd) bimodal. fueron utilizados los análisis de psd para investigación de la microestrutura del suelo en diferentes estados y con diferentes índices de vacíos. se utilizó la técnica de porosimetría por intrusión de mercurio (mip) para explorar la psd de muestras en diferentes condiciones, es decir, natural, compactado, lodo o estados consolidados. se realizó la modelación de las curvas psd utilizando la curva bimodal de van genuchten, que permite el enlace de curvas psd a la curva de retención de agua. observamos que la aplicación de carga sobre las muestras de suelo y la compactación de las mimas no afectaron los microporos y se concluyó que los cambios que ocurrieron en este suelo fueron completamente dentro de los macroporos. resumen palabras clave: comportamiento hidromecánico; comportamiento microestructural; distribución del tamaño de los poros. record manuscript received: 13/06/2017 accepted for publication: 16/03/2018 how to cite item cordão neto, m., p., hernandez, o., lorenzo, r., borges, c., & caicedo, b. (2018). study of the relationship between hydromechanical soil behavior and microstructure of a structured soil. earth sciences research journal, 22(2), 91-101. doi: http://dx.doi.org/10.15446/esrj.v22n2.65640 estudio de la relación entre el comportamiento hidromecánico y la microestructura de un suelo estructurado g e o t e c h n ic s 92 manoel porfirio cordão neto, oisy hernández, raydel lorenzo reinaldo, camila borges, bernardo caicedo 1. introduction since the initial description of effective stress for unsaturated soils, significant progress has been made in defining constitutive models. most of these models are overviews of cam clay or modified cam clay (alonso et al., 1990, 1999, 2013; wheeler et al., 2003; sheng et al., 2008). currently, most studies attempt to link the soil microstructure to the overall soil behavior (gens and alonso, 1992; alonso et al., 1999, 2010, 2013; sanchez et al., 2005; koliji et al., 2008; pinyol et al., 2012; della vecchia et al., 2013). this trend is justified by the fact that according to mitchell and soga (2005), the mechanical properties of soil, including stressdeformation behavior, strength, compressibility and permeability, depend on its structure. for this reason, microstructural experimental studies have been an important tool for improving the research community’s understanding of the relationship between the microand macroscopic behaviors of soils (romero, 2013). several studies (e.g., alonso et al., 1999, 2010, 2013; loret and khalili, 2000; della vecchia et al., 2013; romero, 2013) have followed this line of research. based on constitutive laws and modeling, microstructural analysis distinguishes between variables that depend on the state of the soil, e.g., slurry, undisturbed or compacted. an example of the methodology used to distinguish between state-dependent and intrinsic soil properties is the concept of the intrinsic compression line, first presented by burland (1990). figure 1 shows the same soil in three different states: intact, slurry and remolded. the three curves, which represent the different soil states, tend to converge. a simple conclusion is that the initial structural state is destroyed in the presence of plastic strain and that at a certain point, no memory of the initial state remains, and the sample response is either equal or very similar. figure 1. structure states (vaughan and leroueil, 1990). figure 2. grain size distribution. (a) laser granulometric test; (b) hydrometer test. microstructural studies have been found to be useful in studying soils with limited bonding between particles. these studies are very useful for understanding the behavior of structured tropical soils, such as brasília porous clay (camapum de carvalho and mortari, 1994), colombian volcanic ash (lizcano et al., 2006) and san juan alluvium (zhang and rothfuchs, 2004). these soils have a very complex structure; therefore, defining a constitutive model that extracts all the features of these materials is difficult. in this work, we investigate the relationship between the mechanical behavior and microstructural features of brasília clay via mip tests. the mip samples were in different states (slurry, undisturbed (natural) and compacted) with different void ratios following the consolidation curve. all mip tests were modeled using the modified van genuchten equation (van genutchen, 1980). modeling analyses were conducted to better understand how mechanical loading affects the microstructural behavior. the results show that the mechanical loading erases the initial structure and leads to a convergent behavior. materials and methods the characteristics of brasília porous clay are highlighted here. this clay contains aluminum and iron oxides, as well as bonding particles, which form clumps of particles (figures 2 and 3), i.e., soil aggregates. these aggregates produce a bimodal soil structure that is evident in the pore size distribution (psd) curve (figure 4) and the soil-water retention curve (figure 5). figure 2 also shows how the aggregates (package) are subsequently destroyed by the dispersant and the ultrasound. furthermore, this clay contains two dominant pore sizes (microand macropores), and the distance between pores is very large (approximately 3 orders of magnitude) compared to other soils. finally, the micropores are not affected by loading or compaction processes (figures 4 and 5). (a) (b) all tests presented in this study were performed on soil samples obtained from the experimental field of the university of brasília (unb) (from a depth of 2 m) in brasília city, df, brazil. the characteristics of this site have been investigated in detail by many researchers of the unb geotechnical group. the first ~8 m of this site are characterized by a lateritic porous clay known as “brasília porous clay” (camapum de carvalho and mortari, 1994). this soil’s high porosity and weak cementation between particles originate from two specific physicochemical geological processes: lixiviation and laterization (camapum de carvalho and mortari, 1994). in general, the macrostructure of lateritic soils depends on the laterization and weathering processes (miguel and bonder, 2012). this soil is characterized by a high percentage of clay aggregates, which are often described as “packages” of clay (figure 3). despite the clayey matrix, the soil’s hydraulic behavior is similar to that of granular soil (kw=10-5 m/s), and the air entrance value (aev) of the water retention curve is approximately 10 kpa. evidence of a granular nature can be seen 93study of the relationship between the hydromechanical soil behavior and microstructure of a structured soil in the grain size distribution curves measured using a laser granulometer, with and without ultrasound (figure 2a), and in the hydrometer test results, with and without dispersant (figure 2b). furthermore, two other important structural components of the brasília soil are its weak cementation (aluminum and iron oxides) and high porosity values (greater than 50% in the natural state), which give the soil a very high potential for volumetric collapse (figure 6). figure 3. photomicrograph of an intact sample of brasília porous clay (guimarães, 2002). figure 6. consolidation curve for natural water content (wnat) and saturated (sr=100%) conditions. depth=2.0 m (guimarães, 2002). figure 4. pore size distribution and its evolution with loading for the brasília porous clay (silva, 2009). figure 5. water retention curve of brasília soil in different states. another important characteristic of this tropical soil is the bimodal distribution of the psd curve. figure 4 illustrates a typical psd curve for compacted brasília porous clay and shows how it changes along a loading path. the same feature (i.e., bimodal distribution) is observed in the water retention curve presented in figure 5. despite the different states and void ratios, the micropores do not change (suction greater than 10.000 kpa), as observed in the psd curves. the soil samples present a liquid limit wl=36%, plastic limit wp=26% and specific solid weight gs=2.74. table 1 summarizes other geotechnical properties of the samples. furthermore, the samples presented in table 1 had different fabric consolidation states. the following nomenclature is used to describe the initial structure and consolidated states: natural soil (n) denotes an intact sample, compacted soil (c) represents soil compacted using the standard proctor procedure on the dry side (figure 7), and slurry (s) denotes samples with water contents at the liquid limit (wl). the procedure used to produce slurry samples (s) is similar to the one used to obtain the consistency index of soils, i.e., the liquid limit and plastic limit. table 1 also lists the effective vertical stresses applied during the consolidation tests; for instance, n-50 indicates a natural sample consolidated using a vertical effective stress equal to 50 kpa. 94 manoel porfirio cordão neto, oisy hernández, raydel lorenzo reinaldo, camila borges, bernardo caicedo table 1. geotechnical properties of the samples. figure 7. compacted curve – standard proctor. field data from (silva, 2009). figure 8. comparison between mip samples and the respective consolidation load points. (a) natural sample; (b) slurry and compacted samples. results the mercury intrusion porosimetry (mip) technique has been used to easily and rapidly study structural psds in soils. although mip has certain limitations regarding the minimum and maximum measurable pore radii, it is one of the most commonly used techniques for describing the structure of soils and can provide insights into the link between the microstructure and the geotechnical properties of a soil (romero, 2013). prior to mip testing, all samples were freeze-dried via the lyophilization technique with liquid nitrogen. the drying process helps avoid strains on the samples. next, to investigate how the consolidation process affects the psd, the samples were subjected to consolidation. a sample was placed in the equipment and saturated, and then a target load was applied and maintained for 24 hours. then, the sample was removed, and an mip sample was prepared. a different sample was used for each vertical effective stress. figure 8 presents a comparison between the mip results obtained by the above procedure, as well as their consolidation curves. in general, the results obtained from this procedure follow the consolidation curve. the natural sample was very heterogeneous; consequently, differences exist between the mip samples and the consolidation curves. (a) 95study of the relationship between the hydromechanical soil behavior and microstructure of a structured soil figure 9. brasília porous clay in different states. (a) water relation; (b) pore size distribution. figure 10. brasília porous clay in the natural state. (a) water relation; (b) pore size distribution. (a) (a) (1) (b) (b) in general, the results verify the influences of the initial structure and changes during loading on the microscopic behavior of brasília clay. starting with the initial structural state, figure 9 shows the psds for three different structures (natural, compacted and slurry). the vertical axis, enw, represents the ratio between the volume of the voids filled with mercury and the volume of the solids. the psd is defined as follows: where d is the apparent pore diameter. the range of diameters shown in the psd is approximately 0.01 μm to 200 μm. bimodal behavior is observed in this range, and the distance between dominantly microand macropores is nearly 3 orders of magnitude. this type of behavior has been observed in other studies, such as almeida and futai (2005) and miguel and bonder (2012), which present psd curves for similar tropical soils. due to the limitations of mip associated with pores smaller than 0.1 μm, otálvaro (2013) performed both mip and n2 adsorption tests on the same soil and concluded that the differences between the micropores determined via mip and the n2 tests were not significant. the micropores are probably associated with aggregates (packages), as shown and discussed above (figure 2). interestingly, the samples with different initial structures (natural, compacted and slurry) featured similar pore distributions for diameters less than 1 μm. furthermore, the extrusion curves were also very similar. therefore, we conclude that the compaction and slurry processes are not able to affect the micropores inside the aggregates. consequently, these micropores inside the aggregates influence the liquid and plastic limits. the natural and compacted samples have very similar psds, and their main distinction is the final void ratio. according to their curves, these samples are similar for pore diameters smaller than 1 μm, and their distributions of micro and macropores are similar. for >1μm pore diameters, certain differences exist, with the main difference being that between the inflection points in the slopes of the curves. based on the comparison of these two specific curves with that of the slurry sample, the dominant macropore diameter in the slurry is obviously smaller than that in the other two. however, verifying the macropore range uniformity is not possible. figures 10, 11 and 12 illustrate the changes in the psd during loading for different initial structures. the consolidation process similarly affects the psd of natural, compacted and slurry samples. overall, no noteworthy variations are present in the curves for pore diameters smaller than 0.5μm (micropores). additionally, the extrusion curves are similar. the majority of differences are observed among the macropores. 96 manoel porfirio cordão neto, oisy hernández, raydel lorenzo reinaldo, camila borges, bernardo caicedo for the natural sample, the macropores are quite uniform, and the dominant diameter is approximately 50 μm. the macropore distribution was less uniform during loading, indicated by the change in slope of the curves (figure 10a) and the pore size density (figure 10b). in addition, the dominant pore size peak is also reduced. importantly, if the effective vertical stress was increased, the dominant diameter peak for macropores would disappear, i.e., the psd would assume a unimodal behavior. the mip results for the slurry sample are presented in figure 11. most of the above observations also apply to the slurry sample. however, the distribution of >1 μm pore diameters is less uniform than that of the natural and compacted samples, and the distribution is even less uniform after the consolidation process (figure 11b). the modeling results presented in the next section enable the evaluation of the uniformity of the curves using a model fitting parameter. figure 11b shows that the pore size density of macropores decreases more slowly in the slurry than in the natural samples. (a) (a) (b) (b) figure 11. brasília porous clay in the slurry state. (a) water relation; (b) pore size distribution. figure 12. brasília porous clay in the saturated compacted state. (a) water relation; (b) pore size distribution. the behavior of the compacted sample (figure 12) is similar to that of the other two states. however, the compacting process slowed the pore size changes, i.e., curves with loads less than 800 kpa are similar. this behavior agrees with the fact that a compacted sample is less deformable; consequently, it will experience fewer changes in structure. however, after the sample point c-800, the changes are noteworthy. importantly, the position of c-400 is unexpected because it is located near sample c-200 and not directly in between c-800 and c-200 (figure 12a). finally, as in other samples, unimodal behavior is expected if larger loads are applied. the results presented in figures 11 and 12 agree with the water retention curves presented in figure 5. the water retention curves for natural (guimarães, 2002; silva, 2007), compacted ((silva, 2009) e=1.16), and compacted and loaded states ((silva, 2009), e=0.98 and e=0.77, respectively) coincide at suction values larger than 1000 kpa because the water content links the micropores. compaction and loading also affect the curves at values less than 100 kpa, a range associated with macropores. figure 13 presents the water retention curves obtained via conventional methods (prapaharan et al., 1991) and the mip results. the water retention curve obtained via the mip test was calculated using the procedure presented by otálvaro et al. (2016). although the curves do not coincide, the mip results reflect the aevs of the macroand micropores. 97study of the relationship between the hydromechanical soil behavior and microstructure of a structured soil figure 13. comparison between the water retention curves obtained via conventional techniques and mip. table 2. pore size distribution parameters for full-fit procedure. modeling and discussion because a visual analysis of the effects of the consolidation process would be difficult, the results were modeled, and an analysis of fitting parameters was performed to better understand the effects of structural and consolidation state changes on the psds. the cumulative pore size curves were fitted using the equation proposed by durner (1994). this equation is a modification of the van genutchen (1980) equation and is able to separate and identify the microand macrocomponents of the soil structure. thus, the cumulative pore size curve was adjusted using equation 2: (2) where the parameters with the subscript m are associated with micropores and the parameters with the subscript m are associated with macropores. the parameters em and em are the void ratios; αm and αm represent the dominant pore sizes; and nm and nm are the fitting parameters associated with pore size uniformity. the fitting was done in two steps. first, all six parameters of equation 2 were calculated using a standard procedure, which means that all possible variables were varied until the best fit was obtained. this step is called full-fit; the results are presented in table 2. in the next step, only the fit parameters related to the macropores (em, αm and nm) were adjusted; the micropore parameters (em, αm and nm) remained constant and equal to the average of values obtained in the first step (full-fit) considering all the states and consolidation loads. additionally, for this procedure, em was computed by imposing the total void ratio. this step is referred to as macro-fit; the results are presented in table 3. according to tables 2 and 3, the r2 coefficient is greater than 97%, which means that the modeling process was able to mimic the main features of the data. the micropore parameter αm, which is associated with the dominant pore size, exhibits little variation. this agrees with the observations in figures 10, 11 and 12. additionally, the experimental results lead to the conclusion that all micropore ranges are small. consequently, the macro-fit results can be used to identify the main features of the data (r2>97%), demonstrating that the micropore parameters had little influence on the final results. figure 14 presents the fit and experimental data for the smaller r2 value. even this result exhibits a good agreement and allows us to conclude that the fit parameters are representative of the data. the macropore parameters evolved as expected. for example, the macropore void ratio (em) decreased during loading. however, the fit parameters highlight another point. table 4 presents the void ratio of the micropore curve (em). the suggestions of alonso et al. (2013) and delage and lefebvre (1984) were used to determine the values of em. in general, the values obtained by modeling were similar to but less than the extrusion curve values. the fit parameter em is the value around the inflection point of the micropores, and the data show that the fit parameter is an alternative to determining the microand macropore void ratios. 98 manoel porfirio cordão neto, oisy hernández, raydel lorenzo reinaldo, camila borges, bernardo caicedo table 3. pore size distribution parameters for macro-fit procedure. (b) (a) (c) (d) figure 14. water relation curves and adjustment with model validation. (a) n, (b) c, (c) c-50, and (d) s. 99study of the relationship between the hydromechanical soil behavior and microstructure of a structured soil (b) (b) (a) (b) (a) (a) figure 15. variations in the fit parameters of the natural sample. (a) dominant pore size (αm); (b) uniformity coefficient (nm). figure 17. variations in the fit parameters of the compacted sample. (a) dominant pore size (αm); (b) uniformity coefficient (nm). figure 16. variations in the fit parameters of the slurry sample. (a) dominant pore size (αm); (b) uniformity coefficient (nm). table 4. microstructural void ratio values from the extrusion curve. the behavior of the psd curves can be better analyzed by using the parameters presented in table 2 and table 3. figures 15, 16 and 17 present variations in the parameters related to the dominant macropores (αm) and macropore uniformity (nm). 100 manoel porfirio cordão neto, oisy hernández, raydel lorenzo reinaldo, camila borges, bernardo caicedo for all samples, we observed that αm decreases during loading in the full-fit and macro-fit results (figures 15a, 16a, and 17a). the slope of the αm decrease is steeper for the natural samples than for the compacted and slurry samples. this behavior can be attributed to the presence of weak cement bonding between aggregates that is not present in the slurry and compacted samples. after the bonds are destroyed, the natural samples behave similar to the other samples. furthermore, despite the differences in initial structures, the values of αm approach zero, meaning that all samples exhibit unimodal behavior. the macropore uniformity parameter (nm) did not change after the consolidation process. for the macro-fit procedure, the curves presented in figures 15b, 16b, and 17b can be considered constant at approximately 1.5 (except for the highest void ratio). the compacted sample results of nm indicate that the samples become more uniform during the full-fit procedure. however, according to these results, the variation is small and can be regarded as a statistical phenomenon. the natural samples from n (e=1.69) to n-50 (e=1.55) change significantly (see figure 18), indicating that the consolidation process does not affect the distribution of pores beyond a certain point. this result agrees with the above αm results, in which the natural sample begins to behave like the other samples when the weak bonds are destroyed. across all the samples with different structures and consolidation states, the value of nm was approximately 1.5. therefore, this parameter is not related to the structure and is instead an intrinsic soil property associated with grain size, grain shape, interactions between mineral-clays and water, etc. the number of tests in this study is not sufficient to make our results conclusive, but our results can be considered the first insight into how mip results can be used to identify non-structural properties. final discussion and conclusions here, we present the test results and discuss our conclusions: 1. for brasília clay, the micropores (<0.1 μm in diameter) are not affected by consolidation, compaction, or processes used to produce a slurry. 2. the macropores are destroyed by the consolidation process, which can shift the sample from bimodal to unimodal distributions. 3. the initial structure is destroyed by the consolidation process, and after a certain degree of consolidation, the natural, compacted and slurry samples achieve very similar structures. the third point can be linked with the macroscopic behavior of the soil. as discussed at the beginning of this paper, the mechanical properties of soils depend on the soil structure. however, if the consolidation process destroys the initial structure, samples with different initial structures tend to exhibit similar behaviors after the initial structure is destroyed. this conclusion agrees with the concept of the intrinsic consolidation line presented by burland (1990) and can be observed in figures 1 and 18 for two different soils. figure 18 presents the consolidation curve for brasília clay with different initial states (natural, compacted and slurry). despite the different initial behaviors (which are related to the formation processes), all consolidated samples tend to exhibit similar behaviors, and the consolidation curves converge. the conclusions presented in this study need to be verified for other soils because brasília clay has very specific features, such as aggregations, a bimodal structure and the conservation of the micropore structure. however, the methodology used to investigate the effects of the consolidation process on psd proved to be efficient. additionally, the modeling process aided our understanding of the microstructural behavior and proved to be a very efficient tool for analyzing the psd. figure 18. results of oedometric compression tests on brasília clay in different initial states. references almeida, m. s. s. & futai, m. m. (2005). an experimental investigation of the mechanical behaviour of an unsaturated gneiss residual soil. géotechnique, 55(3), 201-213. alonso, e. e., josa, a. & gens, a. (1990). a constitutive model for partially saturated soils. géotechnique, 40(3), 405–430. alonso, e. e., pereira, j., vaunat, j. & olivella, s. (2010). a microstructurally based effective stress for unsaturated soils. géotechnique, 60(12), 913–925. alonso, e. e., pinyol, n.m. & gens, a. (2013). compacted soil behaviour: initial state, structure and constitutive modeling. géotechnique, 63(6), 463–478. alonso, e. e., vaunat, j. & gens, a. (1999). modelling the mechanical behaviour of expansive clay. engineering geology, 54, 173–183. burland, j. b. (1990). on the compressibility and shear strength of natural clays. géotechnique, 40(3), 329–378. camapum de carvalho, j. & mortari, d. (1994). caracterização geotécnica de solos porosos do distrito federal. iii simpósio brasileiro de escavações subterrâneas, brasilia, 109–122. delage, p. & lefebvre, g. (1984). study of the structure of a sensitive champlain clay and its evolution during consolidation. canadian geotechnical journal, 21, 21–35. durner, w. (1994). hydraulic conductivity estimation for soils with heterogeneous pore structure. water resources research, 30(114), 211–223. gens, a. & alonso, e. e. (1992). a framework for the behaviour of unsaturated expansive soils. canadian geotechnical journal, 29, 1013–1032. grau, e. d. a. (2014). efeito da variação de umidade no empuxo em solos tropicais. msc. thesis, g.dm-240/2014, civil and environmental engineering department, university of brasilia, brasilia, 105p. van genutchen, m. (1980). a closed form equation for predicting the hydraulic conductivity of unsaturated. soil science society of america journal, 44, 892–898. guimarães, r. c. (2002). análise das propiedades e comportamento de un perfil de solo laterítico aplicada ao estudo do desempenho de estacas escavadas. msc. thesis, civil and environmental engineering department, university of brasília, brasilia 207p. koliji, a., vulliet, l. & laloui, l. (2008). advanced constitutive model for unsaturated structured soil with double porosity. 2nd international conference of international association for computer methods and advances in geomechanics, goa, 709– 715. lizcano, a., herrera, m.c. & santamarína, j.c. (2006). suelos derivados de 101study of the relationship between the hydromechanical soil behavior and microstructure of a structured soil cenizas volcánicas en colombia. revista internacional de desastres naturales, accidentes e infraestructura civil, 6, 167–198. loret, b. & khalili, n. (2000). a three phase model for unsaturated soils. international journal for numerical and analytical methods in geomechanics, 24, 983–027. miguel, m. g. & bonder, b. h. (2012). soil–water characteristic curves obtained for a colluvial and lateritic soil profile considering the macro and micro porosity. geotechnical and geological engineering, 30, 1405–1420. mitchell, j. k. & soga, k. (2005). fundamentals of soil behaviour. john wiley & sons, new jersey, usa. otálvaro, i. f. (2013). comportamiento hidromecánico de un suelo tropical compactado. dsc. thesis, civil and environmental engineering department, university of brasília, 148p. otalvaro, i. f., cordão neto, m. p., delage, p., & caicedo, b. (2016). relationship between soil structure and water retention properties in a residual compacted soil. engineering geology, 205, 73-80. pinyol, n. m., alonso, e. e. & gens, a. (2012). modelling compacted soil behaviour including microstructural features. unsaturated soils: research and applications, springer, berlin, germany, 119-127. prapaharan, s., white, d.m. & altsschaeffl, a.g. (1991). fabric of fieldand laboratory compacted clay. journal of geotechnical engineering, 117(12), 1934–1940. romero, e. (2013). a microstructural insight into compacted clayey soils and their hydraulic properties. engineering geology, 165, 3–19. sanchez, m., gens, a., guimaraes, l. & olivella, s. (2005). a double structure generalized plasticity model for expansive materials. international journal for numerical and analytical methods in geomechanics, 29, 751-787. sheng, d., fredlund, d. g. & gens, a. (2008). a new modelling approach for unsaturated soils using independent stress variables. canadian geotechnical journal, 45(4), 511–534. silva, j. (2007). estudos preliminares para implantação de trincheiras de infiltração. msc. thesis, civil and environmental engineering department, university of brasília, 109p. silva, m. t. m. g. (2009). metodologia para determinação de parâmetros para solos não saturados utilizando ensaios com umidade conhecida. msc. thesis, civil and environmental engineering department, university of brasília, 113p. vaughan, p. r. & leroueil, s. (1990). the general and congruent effects of structure in natural soils and weak rocks. géotechnique, 40(3), 467–488. della vecchia, g., jommi, c. & romero, e. (2013). a fully coupled elastic–plastic hydromechanical model for compacted soils accounting for clay activity. international journal for numerical and analytical methods in geomechanics, 37(5), 503–535. wheeler, s. j., sharma, r. s. & buisson, m. s. (2003). coupling of hydraulic hysteresis and stressstrain behaviour in unsaturated soils. geotechnique, 53(1), 41–54. zhang, c. & rothfuchs, t. (2004). experimental study of the hydromechanical behaviour of the callovo-oxfordian argillite. applied clay science, 26(october), 325–336. untitled earth sciences research journal earth sci. res. j. vol. 13, no. 1 (june 2009): 30-39 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india srinivasamoorthy k1, sarma vs2, vasantavigar m1, vijayaraghavan k1, chidambaram s1 and rajivganthi r1 1department of earth sciences, annamalai university, annamalainagar – 608 002 2national geophysical research institute, hyderabad -500606 (council of scientific & industrial research, new delhi) corresponding author email: moorthy_ks@yahoo.com abstract an attempt was made to identify the extent of pollution in the aquifer matrix of tirupur, a highly industrialized zone of tamilnadu state, south india. electrical imaging techniques were adopted with a syscal pro-96 system, for measuring apparent resistivity values using different electrodes separation. the first profile conducted at valipalayam recorded a resistivity range of <10 ù m at a depth of 8 m, which indicates contamination of top soil due to discharge of effluents. an increase in resistivity >45.5 ùm was observed at a depth of 27 to 47 m indicating the possibility of contamination. the second profile conducted at pethichettipuram indicates source of contamination at left end corner with a drop in resistivity <46.5 ù m at a depth of 7.91 m. a drop in resistivity <21.6 ù m was also observed at a depth of 11.5 m indicating a contaminated zone in deeper regolith. the third survey conducted in palayakadu indicates contamination of regolith at a depth of 0 to 20 m with a resistivity less than 40 ùm. the fourth survey at chellapuram indicates contamination of overburden with resistivity >11.5 ùm, to a depth of about 10 m. five imaging surveys conducted across the contaminated sites reveals that shallower regoliths are highly contaminated and deeper aquifers are free from contamination except a few locations. key words: industrial zone, resistivity, tomography, 2d model, regolith, contamination. 30 manuscript receiver: january 05th, 2009. accepted for publication: may 05th, 2009. resumen se intentó determinar el alcance de la contaminación en la matriz acuífera de tirupur, una zona altamente industrializada del estado de tamilnadu, al sur de la india. se usaron sondeos eléctricos para obtención de imágenes con un sistema syscal pro-96, para medir los valores de resistividad aparente con diferentes separaciones de electrodos. el primer perfil realizado en valipalayam registró un rango de resistividad <10 ù m, a una profundidad de 8 metros, lo cual indica la contaminación de la parte superior del suelo debido a la descarga de efluentes. un aumento de la resistividad > 45,5 ùm se observó a una profundidad de 27 a 47 m, indicando la posibilidad de contaminación. el segundo perfil realizado en pethichettipuram indica la fuente de contaminación en el extremo izquierdo con una caída de la resistividad <46,5 ù m, a una profundidad de 7,91 m. un descenso en la resistividad <21,6 ù m también se observó a una profundidad de 11,5 m, indicando una zona contaminada en lo más profundo. el tercer sondeo realizado en palayakadu revela una contaminación a una profundidad desde 0 a 20 m con una resistividad inferior a 40 ùm. el cuarto perfil en chellapuram indica una contaminación de los terrenos de resistividad >11,5ùm, a una profundidad de unos 10 m. cinco estudios de imágenes realizados a través de los sitios contaminados revelan que los acuíferos superficiales están altamente contaminados y los acuíferos más profundos están libres de contaminación, a excepción de unos pocos lugares. palabras clave: zona industrial, resistividad, tomografía, 2d modelo, contaminación. introduction the purpose of electrical surveys is to determine the subsurface resistivity distribution by making measurements on the ground surface as these measurements help to estimate the true resistivity of the subsurface. the true resistivity is related to various geological parameters such as the mineral and fluid content, porosity and degree of water saturation in the rock. in many engineering and environmental studies, the subsurface geology is very complex and the resistivity can change rapidly over short distances (keller and frischknecht 1966; daniels and alberty 1966). resistivity values have a much larger range compared to other physical quantities mapped by other geophysical methods. 2-d and 3-d electrical surveys are the practically commercial techniques used with the recent development of multi-electrode resistivity surveying instruments (griffiths et al. 1990) with aid of computer inversion softwares (loke, 2004). igneous and metamorphic rocks typically have high resistivity values but vary on the degree of fracturing, and the percentage of the fractures filled with ground water. the resistivity of ground water approximately varies from 10 to 100 ù m depending on the concentration of dissolved salts present. the low resistivity (about 0.2 ù m) in groundwater is mainly due to the presence of industrial contaminant metals such as fe, cu, pb and zn along with leaching of cations and anions like potassium, chloride, sodium, bicarbonate, silicates etc., from rock sources and manmade influences greatly reduce the resistivity of ground water to less than 1 ù m even at fairly low concentrations (pathak and venkatesshwar, 2001) electrical imaging techniques new developments in recent years is the use of electrical imaging surveys, where the resistivity changes in vertical direction, as well as in horizontal direction along the survey line, to map areas with complex geology (griffiths and barker, 1993). these surveys are usually carried out using 25 or more electrodes connected to a multi-core cable which is attached to an electronic switching unit, connected to a lap top computer with an electronic switching unit to automatically select the relevant four electrodes along a 31 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india straight line. electrical imaging will provide information about distinct subsurface boundaries and conditions, which can indicate soil or bedrock lithology variations (edwards, 1977). from the measured field data, simulated depth sections are constructed (apparao and sarma, 1981 1983) with over lapping data levels. to plot the data from a 2-d imaging survey, the simulated section contouring method is normally used due to its ease in pictorial representation with different arrays for mapping the same region, which gives rise to very different contour shapes in the simulated section plot. the syscal pro-96 system is capable of measuring apparent resistivity values with different electrode configurations for wenner electrode arrangement. resistivity simulated sections was prepared using the apparent resistivity values and interpretation were carried out using res2dinv software (loke, 1997). the output is presented in the form of subsurface images which is a useful system for electrical resistivity tomography (ert). concept the concept consists of using multi core cables, which contain arrangement of cables and electrodes one takeout every 5m with 64 electrodes. the measuring unit includes relays, which automatically carries out the sequence of readings introduced in its internal memory. the system takes readings for many combinations of transmission and reception pairs so as to achieve a mixed profiling and sounding pairs (abem, 2004). the total length of the cable is equal to the spacing of electrodes which determines the depth of investigations (ron barker et.al, 2001). the final depth of the investigations of a lund imagine survey depends on, geometry of cables (type of array, number of electrodes, spacing between electrodes and number of segments) and the measurement of signal by the equipment, namely the amplitudes of the signal, existing noise, power specifications of the equipment and its ability of filtering the noise through the stacking process (loke, 2001). methodology a known amount of current (i) is pumped to energize the subsurface using current electrodes and the response is measured on the ground surface in the form of voltage (v) through potential electrodes. resistance (v/i) is calculated and further, the apparent resistivity is computed by the formula ña = k× (v/i), where k is the geometrical factor which depends upon the type of configuration that is used. in the case of profiling, the apparent resistivity is obtained along a line using an appropriate electrode arrangement. from the resistivity plot the anomalous zones are identified without any depth estimation. in the case of sounding, apparent resistivity is obtained at a particular place (identified through profiling) by increasing electrode separations and the data is interpreted in terms of layer parameters quantitatively. the present proposed technique is the wenner array which has the strongest signal strength for a comprehensive subsurface picture. forward modeling program the free program, res2dmod.exe, is a 2-d forward modeling program which calculates the apparent resistivity pseudo section for a user defined 2-d subsurface model. the program helps to choose the finite-difference (dey and morrison, 1979a) or finite-element (silvester and ferrari, 1990) method to calculate the apparent resistivity values. the res2dmod.exe program indicates the contours in the simulated section produced by the different arrays over the same structure. study area tirupur is located 50 km east in coimbatore district of tamilnadu state in south india at latitudes and longitudes 11.18° n 77.25° e with a total extent of 27 sq.km (figure 1). tirupur an important trade center of india gained recognition as the leading source of knitted garments, casual wear and sportswear. geology of the study area is entirely composed of precambrian shield area called as the indian peninsu32 srinivasamoorthy k, sarma vs, vasantavigar m, vijayaraghavan k, chidambaram s and rajivganthi r lar complex with wide range of igneous and metamorphic rocks. the most common rock type in the study area is gneiss. gneiss is a generic term for a large variety of metamorphic rocks, and can have both sedimentary and igneous origin. the gneiss found in the tirupur region is of high metamorphic grade and is mainly of the biotite type, but quartzo-feldspatic gneiss is also found. these rocks are thought to have been formed during the archaean time period, approximately 3.8 to 2.5 billion years ago, which means that they are among the world’s oldest rocks (gustavsson et al., 1970), (sankararaaj et al., 2002). thestudy area is an undulating plain with gentle slopes. the elevation ranges between 225 33 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india figure 1. study area with imaging locations. and about 300 m (kristina furn, 1998). the climatic conditions in the study area are “semiarid” with a mean annual temperature of 29.4°c. (ssluo, 1998). the annual average rainfall in the study area is 527.2 mm. hydrogeology of the study area groundwater occurs in two different aquifers, one shallow aquifer formed by the weathered zone and the other deeper aquifers connected to the fracture system in the rock with vertical and horizontal fractures. these fractured zones extend down as 200 m or more. the transmissivity of the shallow aquifers varies from 11.4 to 51.0 m2/day, with an average of about 30.7 m2/day. the hydraulic conductivities vary from 2.9 to 20.0 m/d with an average of 9.2 m/d. the average thickness of the shallow aquifer is 3.7 m. the average specific yield is 4.0×10–2 (cgwb, 1993). the vertical fracture zones constitute an unconfined aquifer, demarcated as the most potential aquifer in the area, in which groundwater development is focused. transmissivity in the vertically fractured aquifer vary from 52.1 to 497.0 m2/day. the hydraulic conductivity is 1.8 m/day with a range of 0.9 to 3.4 m/day. the aquifer ranges from approximately 45.0 to 142.3 m below ground level with an average thickness of 97.3 m (table 1). in horizontal fracture aquifer the transmissivity is 52.8 m2/day with a hydraulic conductivity of 0.5 m/day. the horizontally fractured layer depth ranges from 28.8 to 129.5 m below ground level (cgwb, 1993). the groundwater table more or less follows the topography, but with a smaller slope than the surface. the hydraulic gradient is approximately 2.8 m/km with a steeper slope of about 6.5 m/km towards the river noyil indicating the groundwater flow along east west direction. groundwater is the main source of water supply in the study area although presence of river noyyal which dries during summer season. the study area has been demarcated for heavy exploitation of groundwater used mainly for industrial purposes (senthilnathan and azeez, 1999). at present 9000 knitting, processing and manufacturing units consume nearly 86 mld (million liter per day) of water, but 90 mld of water used are discharged as effluents containing a variety of dyes and chemicals like acids, salts, wetting agents, soaps and oil, leading to contamination of the ground and surface water and soil in and around the study area (rajaguru and subbram, 2000). the pollution load calculated from pollution control board, 2008 was as follows: total dissolved solids 23.54 lakh tonnes, chloride 13.11 lakh tonnes, sulphate 1.25 lakh tonnes, total suspended solids (tss) 0.97 lakh tones, chemical oxygen demand 0.90 lakh tonnes, biological oxygen demand 0.29 lakh tonnes and oil & grease 0.01 lakh tones (jacob thomson, 1998). hence an attempt was made to get further information on the extent of groundwater contamination by using ert studies. results and discussion the present study was made to delineate the pollutant zone through ert method. in tirupur surveys, the imaging system was used with single cable with 25 takeouts at 5 m interval by syscol pro 96 instrument by adopting wenner array. electrodes were 0.5 m long and made of stainless steel; they were planted to 34 srinivasamoorthy k, sarma vs, vasantavigar m, vijayaraghavan k, chidambaram s and rajivganthi r table 1. aquifer parameters of the study area physical parameter (unit) hydraulic conductivity (m/day) thickness (m) storativity porosity (%) shallow 9.2 3.7 4.0* 10-2 4.5 vertical 1.8 97.3 5.2* 10-3 8.0 a depth of 0.4 m. each electrode was watered to ensure good contact with the ground. a total of four image lines were measured with the images varying in length from 0 to 50 m parallel to the river noyyal. this effectively gave a maximum depth of imaging of 20 m. the first survey was conducted at valipalayam 1km from river noyyal (figure 2). the profile shows a low resistivity zone ranging from 10 to 100 ù m, indicating presence of highly weathered rock materials. the basic concept of electrical resistivity method is to demarcate higher resistivity zones within the low electrical resistivity rocks at the sub surface. this is because; the very low resistivity is an indicator of highly weathered rock material. the regolith with a resistivity range of <10 ù m is found at a depth of 8 m indicating the contamination of top soil due to the discharge of effluents. the weathered and fractured zones were identified at a depth of 27 to 47 m with increase in resistivity from 46 to 95.5 m indicating that the deeper layers are exposed to groundwater contamination in the absence of clay materials (ron barker and others, 2001). the second survey was conducted near pethichettipuram just 1km away from the noyyal river. the survey indicates source of contamination at left end corner with a drop in resistivity by <46.5 ù m at a depth of 7.91 m (figure 3). the same trend was also noted at a depth of 11.5 m with a drop in resistivity by <21.6 ù m indicating the contaminated zone at deeper regoliths. this is supported by a groundwater sample collected in a dug well to a depth of 15m showing higher tds value >3,500 ppm. the fractured and massive rocks revealed higher resistivity varying from 46.7 to 2200 ù m indicating the non polluted nature of deeper formations. a thin subsurface layer with low resistivity extends to deeper area in the middle of the profile, followed by a thick layer with high resistivity (resistivity greater than 1000 ù m) acting as a basin at deeper depth (apparao and sarma, 1993). this basin is more dominant at the last quarter of the traverse where it is closer to the surface. this high resistivity layer could represent the bedrock, which is competent without any structures. from the survey result a geological division is observed between the lower and high resistivity materials. in the central part of the profile a very low resistivity zone exist, indicating existence of an aquifer within the profile, based on its lower resistivity in relation to the background resistivity. the third survey was conducted in palayakadu (figure 4), indicating extensive contamination of regolith from ground surface to 20 m as the resistivity of the regolith drops less than 40 ùm. bed rock resistivity was higher >1058 ù m indicating their massiveness at shallow depth. the profile produced three different resistivity of layered rock which is clearly defined by their different resistivity layers at various depths as follows; a thin subsurface layer with comparatively low resistivity values ranging from 10 to 46.7 ùm at an depth of about 15 m. this layer of low resistivity is typical of weathered rock materials of the underlying rocks in the area. an intermediate resistivity layer ranging from 101 to 218 ù m at the intermediate depth zone, which could be represented as weathered to moderately weathered rock material. this depth layer between the depth ranges (15-40 m) has a comparatively moderate resistivity from 150 to 1018 ù m. a thick layer with a comparatively high resistivity (>218 ùm) is also observed below the layers of low to intermediate resistivity rocks. this relatively high resistivity rock layer represents the presence of fresh rock material with no structural patterns like fractures and joints, as good indications for aquifers (kelley, 1976) due to their sheared nature represented by the wavy pattern. the shallow depth layers could also be interpreted to be layers of different rock materials. the resistivity layers of the different rock materials in the range (180 to 1080 ù m) and its contact with the fresh rock observed at a depth range of about 55 m was also accounted. the fourth survey conducted at chellapuram is shown in (figure 5). the profile indicates that the area is underlined by varying high resistivity rock materials by sharp changes in their electrical resistivity values. the high resistivities observed are typical of fresh granite rocks. the general structural trend observed is as follows; an overburden with low resistivity to a 35 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india 36 srinivasamoorthy k, sarma vs, vasantavigar m, vijayaraghavan k, chidambaram s and rajivganthi r 2 d valipalayam .bin ps.2 0.0 80.0 160.0 240.0 n. 2.60 8.53 14.3 20.1 25.8 31.6 37.0 42.5 47.9 regolith weathered and fractured zone mesured apparent resistivity pseudosection 0.0 80.0 160.0 240.0 n.ps.2 2.60 8.53 14.3 20.1 25.8 31.6 37.0 42.5 47.9 0.0 80.0 160.0 240.0 n. depth 1.25 6.38 12.4 19.8 28.7 33.8 39.4 45.6 calculated apparent resistivity pseudosection iteration 3 rhs error 3.7% inverse model resistivity section 0.197 0.325 0.536 0.884 1.46 2.41 3.97 6.55 resistivity in ohn.n unit electrode spacing 5.00 n. figure 2. resistivity imaging at valipalayam. 37 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india 0.0 32.0 64.0 96.0 128.0 m. 0.0 32.0 64.0 96.0 128.0 m. 0.0 32.0 64.0 96.0 128.0 m. pethuchettipuram-l1 pethuchettipuram-l2 pethuchettipuram-l3 depth teration 5 rms error = 25% 0.500 3.70 7.91 11.5 15.8 21.0 inverse model resistivity section 10.0 21.6 46.7 101 218 471 1018 2200 resistivity in ohm.m unit electrode spacing 2.00 m unit electrode spacing 2.00 m depth teration 5 rms error = 1.46% 10.0 21.6 46.7 101 218 471 1018 2200 resistivity in ohm.m depth teration 5 rms error = 2.0% 0.500 3.70 7.91 11.5 15.8 21.0 0.500 3.70 7.91 11.5 15.8 21.0 unit electrode spacing 2.00 m 10.0 21.6 46.7 101 218 471 1018 2200 resistivity in ohm.m inverse model resistivity section inverse model resistivity section contaminated zone fractured and massive rock aquifer zone figure 3. resistivity imaging at pethuchettipuram. 38 srinivasamoorthy k, sarma vs, vasantavigar m, vijayaraghavan k, chidambaram s and rajivganthi r ws-07palayakadu 0.0 80.0 160.0 240.0 320.0 400.0 m. contaminated zone maximize hard rock 2.60 10.5 18.2 25.8 33.5 41.2 48.8 56.5 2.60 10.5 18.2 25.8 33.5 41.2 48.8 56.5 1.25 9.26 19.0 28.7 39.4 52.4 0.0 80.0 160.0 240.0 320.0 400.0 m. 0.0 80.0 160.0 240.0 320.0 400.0 m. depth inverse model resistivity section mesured apparent resistivity pseudosection calculated apparent resistivity pseudosection iteration 5 abs error 35.4% ps.2 ps.2 10.0 21.6 46.7 101 218 471 1018 2200 resistivity in ohn.m unit electrode spacing 5.00 m. figure 4. resistivity imaging at palayakadu. depth of about 29.3m with a resistivity range of (11to 45 ù m). this layer is due to the presence of weathered materials with greater risk of contamination (urish, 1983). an intermediate zone with resistivity range from 663 to 3328 ù m at a depth of 21.1 to 38.7 m is noted. a very high resistivity zone (125418 ù m) is observed at a depth of 55 m indicates fresh rock material without any structural pattern like folds and faults. resistivity distributions observed at greater depth are high when compared with low resistivity values observed at the surface. the profile therefore could be interpreted as the different layering of weathering with low resistivity values on the surface and higher resistivity values are confined to the fresh electrical resistant rock materials. conclusion an attempt was made in tirupur a highly industrialized zone to determine the extent of pollution in aquifers by using ert techniques. a total of 4 profiles were conducted more or less 1 km and parallel to river noyyal. the profile at valipalayam indicates contamination of top soil by a drop in resistivity by < 10 ù m at a depth of 8 m. the weathered and fractured zones identified at depth of 27 to 47 m with greater resistivity range indicate deeper layers without any groundwater contamination. the second profile conducted near pethichettipuram indicates sources of contamination at a depth of 7.91 m with a drop in resistivity <46.5 ù m. another drop in resistivity at a depth of 11.5 m indicates the contaminated nature of the deeper regolith. the fractured and massive rocks follow the same trend like that of the first location with greater resistivity. the third survey conducted in palayakadu indicates greater contamination of regolith at a depth of 0 to 20 m with a resistivity less than 40 ùm. bed rock resistivity was higher >1058 ù m indicating their massiveness present at shallow depth. the fourth survey conducted at chellapuram indicates high resistivity rock materials which are layered and boundaries of the different resistivity materials are clearly defined by the sharp change in electrical resistivity. the high resistivity observed is typical of fresh granite rocks. from a total of four profiles the first, third and fourth showed top 10 to 25 m of regolith has resistivity of less than 10 ù m, with top 5 m having a resistivity of less than 10 ùm indicating soil with greater contamination. the second profile has shown low resistivity at pockets at shallower depth and resistivity of above 100 ùm is not contaminated. none of the five images measured across the contaminated sites show any strong lateral change in resistivity and it must be admitted that similar information could be obtained with resistivity sounding. a few soundings over the 39 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india top soil weathered and fractured zone massive hard rock unit electrode spacing 5.00 m.resistivity in ohm.m 11.6 43.3 165 622 2344 883 33284 125418 inverse model resistivity section iteration 5 rms error = 4.8%depth 0.0 80.0 160 240 320 m. 13 10.8 21.7 29.3 38.7 50.6 65.3 figure 5. resistivity imaging at chellapuram. area can indicate likely sites for low resistivity regolith and heavy contamination. acknowledgements corresponding author thank the indian national science academy (insa) for the visiting fellowship award and the director, national geophysical research institute (ngri) for permission to work under dr.v.s.sarma. the author acknowledges the reviewer for his help to improve the ms in the present form. references abem, (2004). instruction manual, terrameter sas 4000/sas 1000. apparao, a. and sarma, v.s. (1981). a modified pseudo depth section as a tool in resistivity and ip prospecting, geophysical research bulletin, ngri, india, v. 19, 187-208. apparao, a. sarma, v.s. (1983). the concept of depth of detection and depth of investigation in resistivity and ip, geophysical research bulletin, ngri, india, no. 21, 51-61. apparao, a. subrahmanya sarma, v. sivasarama sastry, r. and srinivas, g.s. (1993). a catalogue of modified pseudodepth sections with different electrode arrays over buried conductive targets, ngri, hyderabad (india), 52 pp. barker, r. d. venkateswara, r.t. and thangarajan, m. (2001). delineation of contaminant zone through electrical resistivity imagine technique. current science journal. 81. no. 3, 277-283. central groundwater board (cgwb) (1993). groundwater resources and development prospects in coimbatore district, tamil nadu, central groundwater board, 30. daniels, f. and alberty, r.a. (1966). occam’s inversion to generate smooth, two dimensional models form magneto telluric data, geophysics. 55, 1613-1624. dey, a. and morrison, h.f. (1979a). resistivity modeling for arbitrary shaped twodimensional structures. geophysical prospecting. 27, 1020-1036. edwards, l.s. (1977). a modified pseudo section for resistivity and induced-polarization. geophysics. 42, 1020-1036. griffiths, d.h. and barker, r.d. (1993). two-dimensional resistivity imaging and modeling in areas of complex geology, applied geophysics. 29, 211-226. gustavsson, y., jacks, g., nilsson, l.y. and taranger, b. (1970). water resources and water supply in coimbatore district, tamil nadu, india, department of land improvement and drainage school of surveying, royal institute of technology, stockholm. jacob thomson (1998). impact of industries on the ground water quality of tiruppur and its ethical implications, ph.d. thesis, dept. of zoology, university of madras, chennai. keller, g.v. and frischknecht, f.c. (1966). electrical methods in geophysical prospecting, pergamon press, oxford. kelly w.e. (1976). geoelectric sounding for delineating groundwater contamination, ground water, 14, 6-10. kristina furn, (2004). effect of dyeing and bleaching industries on the area around the orathapalayam dam in southern india, uptec w04033, issn 1401-5765, tryckt hos institutionen för geovetenskaper, uppsala universitet, uppsala, pp 99. loke, m.h. (1997). electrical imaging surveys for environmental and engineering studies – a practical guide to 2d and 3d survey, penang, malaysia, universiti sains malaysia, unpublished short training course lecture notes. 40 srinivasamoorthy k, sarma vs, vasantavigar m, vijayaraghavan k, chidambaram s and rajivganthi r loke, m.h. (1997). res2dinv ver. 3.3 for windows 3.1, 95, and nt, advanced geosciences, inc, pp. 66. loke, m.h. (2004). rapid2-d resistivity and ip inversion using the least-square method; manual for res2dinv, version 3.54, p53. pathak b. and venkatesshwar rao,(2001). geoelectrical investigation to locate groundwater resources for urban development: a case study in parts of greater guwahati urban area, assam, geophysics, xxii-1:41-46. public works department (pwd) (2003). statement showing the hydraulic particulars of p.w.d. tanks, sub-division irrigation, coimbatore. raja guru, p and subbram v (2000), ground water quality in tiruppur, environmental awareness on quality management of irrigation water, edited by lakshmanaperumalsamy and krishnaraj, department of environmental sciences, bharathiyar university, coimbatore. sankaraaj, l., subramanian, t.p., siddhamalai, a. & farooque ahmed, n. (2002). quality of soil and water for agriculture in noyyal river basin, tamil nadu. joint director of agriculture (research), soil survey and land use organization, tamil nadu department of agriculture, coimbatore. senthilnathan, s and azeez, p.a. (1999). influence of dyeing and bleaching industries on ground water of tirupur, tamil nadu, india. bull environ contam toxicol, 62:330-335. silvestre, p.p. and ferrari, r.l. (1990). finite elements for electrical engineers, 2nd edition, cambridge university press. ssluo (soil survey & land use organisation) (1998). soil atlas coimbatore district, department of agriculture tamil nadu, coimbatore. urish, d.w. (1983).the practical application of surface electrical resistivity to detection of ground water pollution. groundwater, 21:144-152. 41 electrical imaging techniques for groundwater pollution studies: a case study from tamil nadu state, south india sandy pebble stratum is a typical discrete particle unstable stratum, mainly consisting of sand and pebble. however, the effect of coarse-grained content on the stability of stratum is not clear. in this paper, taking the sandy pebble soil of different coarse-grained content in chengdu city, sichuan province, china as the research object, a research on macro-mesomechanical properties of sandy pebble soil of different coarse-grained content was carried out using the method combining the indoor large-scale triaxial test of coarse-grained soil with the discrete element numerical triaxial test. the research results showed that the stress-strain curve of sandy pebble soil exhibited strain softening with the increase of coarse-grained content; when the confining pressure was the same, the stress peak increased and the strain when the peak was reached decreased gradually with the increase of coarse-grained content. it revealed the functional relationship between coarse-grained content and mechanical indexes of sandy pebble soil such as internal friction angle and cohesion, the internal friction angle and cohesion of sandy pebble soil linearly increased with the rise of coarsegrained material; it proposed the particle discrete element micro parameters of sandy pebble soil of different coarsegrained content, including contact modulus, friction coefficient, particle stiffness ratio, contact bond strength. the research results provided the theoretical support for the fine design and construction of sandy pebble stratum project. abstract keywords: sandy pebble soil; coarsegrained content; discrete element; macromicromechanical properties. macro-micromechanical properties of sandy pebble soil of different coarse-grained content issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n1.66105 junfu lu, di li* xiaoqiang xue, shenlin ling state key laboratory of geohazard prevention and geoenvironment protection, chengdu university of technology, chengdu610059, china *e-mail: lidi2014@cdut.edu.cn earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018): 65-71 la capa de arena guijosa es un manto típico de partículas inestables que consisten principalmente de arena y guijarros. sin embargo, el efecto del contenido de grano grueso en la estabilidad de la capa no es claro. el área de investigación en este estudio es la ciudad de chengdu, provincia de sichuan, en china. allí se tomaron muestras de diferente clasificación granulométrica para analizar las propiedades macro y mesomecánicas del suelo de arena guijosa a través del método combinado de la prueba triaxial a gran escala en laboratorio para suelos de grano grueso con la prueba triaxial numérica de elementos discretos. los resultados de la investigación muestran que la curva del ensayo de tracción en suelos de arena guijosa presenta un aplacamiento en la presión con el incremento del contenido de grano grueso. cuando la presión de confinamiento fue la misma, el pico de esfuerzo se incrementó y la presión empezó a caer gradualmente cuando se alcanzó el pico con el incremento del contenido de grano grueso. esto revela la relación funcional entre el contenido de grano grueso y los indicadores mecánicos para suelos de arena guijosa como el ángulo de fricción interna y la cohesión; estos factores se incrementaron con el aumento del material de grano grueso. esto sugiere los microparámetros de particulas de elementos discretos en este tipo de suelos, que incluyen el modulo de contacto, el coeficiente de fricción, la proporción de rigidez de partículas y la fuerza del vínculo de contacto. los resultados de la investigación proporcionan el apoyo teórico para el diseño y construcción de nuevos proyectos sobre la capa arenosa y guijarrosa. resumen palabras clave: suelo arenoso y guijarroso; contenido de grano grueso; elementos discretos; propiedades macro y micromecánicas. record manuscript received: 14/11/2017 accepted for publication: 23/04/2018 how to cite item lu, j., li, d., xue, x., & ling, s. (2018). macromicromechanical properties of sandy pebble soil of different coarse-grained content. earth sciences research journal, 22(1). 65-71. doi: http://dx.doi.org/10.15446/esrj.v22n1.66105 b io g e o c h e m is t r y propiedades macro y micromecánicas de suelos de arena guijosa con diferente clasificación granulométrica 66 junfu lu, di li, xiaoqiang xue, shenlin ling introduction sandy pebble soil is widely distributed in various regions of the world, and its mechanical properties are further complicated by the randomness of geometrical distribution of pebble and discreteness of particle. the sandy pebble stratum widely distributed in chengdu city, sichuan province, china has a great impact on the construction safety of tunnel and underground engineering(wei et al., 2008; omarashahestan and omarashastani, 2017), so the study on strength characteristics and stability of sandy pebble soil has been paid increasing attention (si et al., 1990; hejazi et al., 2017). guo qingguo (1998)proposed 5 mm as the dividing line of fine and coarse particles, and that takes 30% and 70% of coarse-grained contents as the two research points. at the same time, many scholars believed that the physical and mechanical properties of sandy pebble soil of different coarse-grained content such as cohesion, internal friction angle and permeability had a great impact (wei et al., 2008; yasin et al., 2017).most of them conducted studies on macromechanical properties of sandy pebble soil using the triaxial test and direct shear test but lacking micromechanical parameters of the composite structure of sand and pebble particles. discrete element method is a numerical simulation method for solving discontinuous media problems, and has received attention from many scholars and was developed since being proposed in the 1970s by cundall (1997). geng li(2011)simulated the triaxial test process using pfc3d, and analyzed the effect of particle micro parameters on the strength of coarse-grained soil, however, it researched on only a gradation of soil, so the conclusions were not representative. xu xiaofeng (2013) et al established the calculation model of discrete element simulation of four kinds of gravelly soils of different coarse-grained content, and analyzed the effect mechanism of coarse-grained content on the macro-micromechanical properties of gravelly soil, however, the micro parameters obtained were only suitable for a confining pressure, so it was not accurate enough. qi yang (2015)et al simulated the uniaxial loading test of coarse-grained soil in various main stress directions under triaxial state by means of pfc3d, and analyzed the stress-strain characteristics of coarse-grained soil in complex stress state from the macro and micro points of view, believing that the complex stress state was the inducing factor for coarse-grained soil to produce various anisotropies; in addition, many scholars at home and abroad solved the problems of sandy pebble soil encountered in the engineering by using the discrete element theory (qi et al., 2015; ahmad et al., 2017). in this paper, relying on the sandy pebble stratum in chengdu city, sichuan province, china, the influencing law of coarse-grained content on the macromechanical properties of sandy pebble soil was studied using the large-scale triaxial test of coarse-grained soil, on this basis, a particle discrete element numerical triaxial test platform was established, and the micromechanical parameters of sandy pebble soil of different coarsegrained content was studied using the discrete particle element numerical triaxial test (yasin et al., 2017). the research results provided essential reference and guidance for the fine design, construction, and maintenance of sandy pebble stratum project. research method triaxial test of coarse-grained soil test scheme in this test, a stress type large-scale triaxial test apparatus was adopted, and its specifications included a diameter of 300 mm, a height of 600 mm, a maximum axial stress of 21 mpa, a maximum vertical load of 1500 kn, maximum confining pressure of 3.0 mpa, and maximum axial stroke of 300 mm. the shear strength indexes c and φ of coarse-grained soil with the particle size of not less than 60 mm can be measured, as shown in figure 1. figure 1. large-scale triaxial apparatus of coarse-grained soil. specimens in this test, the sandy pebble soil in chengdu city, sichuan province, china was taken as the research object, for the oversized particles, the weighted average substitution was conducted using the sandy pebble soils with the maximum allowed particle size of 60 mm to 5 mm according to the content, namely equivalent substitution method (yasin, 2017; kudus et al., 2017). to obtain the shear strength index of sandy pebble soil of different coarse-grained content, a total of four groups of specimens with p5 (mass percentage of particles with particle size of greater than 5 mm) of 73%, 75%, 76.5% and 80% were adopted in this test, with density of 2.05 g/cm3, moisture content of 1.6%.each group of specimens was exerted with confining pressures of 100 kpa, 200 kpa, and 300 kpa. the particle gradation of sandy pebble soil used in the test is shown in table 1 and figure 2. table 1. composition of particle gradation of sandy pebble soil of different coarse-grained content. the triaxial specimens were divided into three layers for preparation, which allows realizing the uniform mixing of particles. the shearing with an axial strain rate of 1.5 mm/min was carried out using the test method of isotropic consolidation undrained shear until the failure of the specimen occurred (1999), and the specimens are shown in figure 3. 67macro-micromechanical properties of sandy pebble soil of different coarse-grained content figure 2. gradation curve of sandy pebble soil. figure 3. large-scale triaxial specimen of sandy pebble soil. particle discrete element numerical triaxial test establishment of numerical triaxial test platform the test silo used in the numerical triaxial test of coarse-grained soil adopted a cylinder with diameter of 300 mm and height of 600 mm, its upper and lower planes were the loading plates with stiffness of greater than particle stiffness, the wall of the test silo adopted the flexible constraint, and its normal stiffness was smaller than normal stiffness of particles (wun et al., 2017). in the effective area of the test silo, the sphere unit was adopted to replace the soil particle according to the particle gradation of sandy pebble soil tested in the test, the minimum unit diameter in the numerical triaxial test was dmin=1 mm, and the particles of large particle size were composed of small units and classified as a particle group (i.e., clump unit) (ridzuan et al., 2017). the contact bond model among particles was adopted, and the servo loading was adopted in the test to keep constant confining pressure. the model of the numerical triaxial test is shown in figure 4. figure 4. triaxial numerical model. method for determining the micro parameters of discrete element (1) determination of particle contact stiffness ec and ratio kn/ks of normal stiffness to shear stiffness. for the contact bond model, the macro elastic modulus e was mainly determined by the ratio of particle contact stiffness to stiffness. when the particle stiffness ratio was constant, the micro parameter ec and macro elastic modulus presented linear correlation (rahman et al., 2017). in the numerical triaxial test, the particle contact stiffness ec needed to be continuously adjusted until the elastic modulus of numerical model was consistent with the indoor large-scale triaxial test, and the ec at this time was the micro parameter determined. the stiffness ratio kn/ ks was mainly determined by the poisson’s ratio, the greater the stiffness ratio was, the greater the poisson’s ratio was; (2) determination of particle micro parameter friction. the friction coefficient had a great effect on the peak intensity of the model, and little impact on the elastic modulus of the material, by adjusting the friction coefficient until it was similar to the peak intensity of the indoor large-scale triaxial test, the friction coefficient can be determined; (3) determination of particle bond strengths σ and τ. the particle bond strength was mainly related to the peak intensity of the model, when the ratio σ mean/σ dev of bond strength average to unbalanced load value was constant, the peak intensity of the model was proportional to the bond strength average. table 2. microparameters of the discrete element of sandy pebble soil of different gradation. analysis of research results the stress peak of traditional the triaxial test model of particulate matter under different confining pressure was slightly lower than that of the indoor test, and potyondy and hoek et al. (1998) believed that by changing the particle micro parameters to overcome this disadvantage had little effect. in this numerical triaxial test, clump unit was introduced, which efficiently solved this problem. clump unit was a large particle composed of a plurality of balls cemented together simulating the particles of irregular shape, which can realize the irregularity of sandy pebble soil, increase the occlusion and embedding among particles, thus appropriately improving the shear strength of the numerical triaxial test. analysis of macromechanical properties of sandy pebble soil stress-strain characteristics of sandy pebble soil of different coarse-grained content from figure 5, it can be seen that the stress-strain curve of sandy pebble soil exhibited strain softening with the increase of coarse-grained content, this was because, with the increase of coarse-grained content, particles became overhead, formed large pores, and lacked fine particles to fill, resulting in the decrease of stress with the increase of strain (zhang et al., 2013; basarian and 68 junfu lu, di li, xiaoqiang xue, shenlin ling tahir, 2017). from figs. 6 and 7, it can be seen that when the confining pressure was the same, the stress maximum increased and the strain when the peak was reached decreased gradually with the increase of coarse-grained content. (a) stress-strain curve of sandy pebble soil of gradation 1. (b) stress-strain curve of sandy pebble soil of gradation 2. (c) stress-strain curve of sandy pebble soil of gradation 3. (d) stress-strain curve of sandy pebble soil of gradation 4. figure 5. stress-strain curve of sandy pebble soil of different coarsegrained content. figure 6. peak stress of sandy pebble soil of different coarse-grained content. φ =0.84p5-23.39 (1) c =4.13p5-208.8 (2) figure 7. peak strain of sandy pebble soil of different coarse-grained content. analysis of macromechanical parameters of sandy pebble soil from figure 8, it can be seen that the internal friction angle and cohesion of sandy pebble soil linearly increased with the increase of coarse-grained content, but with different increasing magnitude, by fitting, the relationship between mechanical parameters and coarsegrained content of sandy pebble soil was obtained as follows: where: φ—internal friction angle, (°); c— cohesion, (kpa) , 69macro-micromechanical properties of sandy pebble soil of different coarse-grained content (a) relationship between internal friction angle and coarsegrained content. (b) relationship between cohesion and coarse-grained content. figure 8. relationship between internal friction angle and cohesion and coarse-grained content. p5—mass percentage of particles with particle size of greater than 5 mm analysis of micromechanical properties of sandy pebble soil based on the results of the indoor large-scale triaxial test of coarse-grained soil, the optimal micro parameters of sandy pebble soil of different coarse-grained content were obtained by using the approximation method of parameters of particle numerical triaxial test. stress-strain curve of numerical triaxial test of sandy pebble soil of different coarse-grained content as can be seen from figure 9, the stress-strain curve of the discrete element numerical triaxial test was in substantial agreement with that of the indoor test, and the micro parameters of the discrete element determined by this method can well realize the macromechanical properties of sandy pebble soil. when the triaxial confining pressure was the same, the stress peak increased, but with different initial modulus, this was primarily because the basic unit of discrete element was the rigid sphere, and the occlusion of particles was weak, and the particles were arranged in single form, even if the contact bond was selected to improve the cohesion of particles (bilal et al., 2017). however, with the increase of axial load, the bond fracture led to strength characteristics of sandy pebble soil different from the actual strength characteristics of sandy pebble soil, which also needed to be improved and further optimized. (a) stress-strain curve of sandy pebble soil of gradation 1 (b) stress-strain curve of sandy pebble soil of gradation 2 (c) stress-strain curve of sandy pebble soil of gradation 3 (d) stress-strain curve of sandy pebble soil of gradation 4 figure 9. stress-strain curve of sandy pebble soil of different coarse-grained content in the real and numerical triaxial test. 70 junfu lu, di li, xiaoqiang xue, shenlin ling micromechanical parameters of discrete element of sandy pebble soil by checking calculation, the micro parameters of discrete element able to reflect the macromechanical properties of sandy pebble soil of different coarse-grained content were obtained, the determination of the micro parameters of the discrete element provided the basic data for the numerical analysis and design of sandy pebble soil. the specific micro parameters of the discrete element are shown in table 2. table 2 showed that the micromechanical parameters of sandy pebble soil were mainly determined by contact modulus, porosity, friction coefficient, bond stiffness and bond stiffness ratio. the more the coarse-grained content was, the larger the contact modulus and friction coefficient were, and the higher the internal friction angle and cohesion of sandy pebble soil were, but the other micro parameters showed little change (nema et al., 2017). therefore, the fact that the strength of sandy pebble soil increased with the increase of coarsegrained content was caused by particle stiffness and friction coefficient. analysis of displacement of sandy pebble soil particles the displacement of sandy pebble soil particles in the triaxial compression process was analyzed, to reveal the micromovement of sandy pebble soil particles in the triaxial compression process. because the coarse-grained content of sandy pebble soils of four different gradations had little difference, so their particle displacement fields showed no difference (andik and sarang, 2017). when the sandy pebble soil of gradation one was selected, the displacement was analyzed with particle displacement field of 100 kpa of confining pressure as the representative, and the microdisplacement diagram of particles under different strain is shown in figure 11. from the beginning of compression shear, due to movement of upper and lower loading plates, the particles in the upper part of the model moved downward, and the particles in the lower part of the model moved upward. as the loading progressed, the particles in the middle of the model moved outward due to pressure, resulting in the dilatancy of specimens in the middle of the model, which was the same as the failure mode (figure 10) of samples in the indoor test. figure 10. failure appearance of specimens. (a)ε=2.5% (b) ε=8% (c) ε=12% (d) ε=16.5% figure 11. displacement field of particles. conclusion in this paper, the macromechanical parameters of sandy pebble soil of different coarse-grained content were obtained using the indoor large-scale triaxial test of sandy pebble soil, and the discrete element numerical triaxial model of sandy pebble soil was established. by checking calculation, the micromechanical parameters of sandy pebble soils of four different gradations were obtained, and the effect of coarse-grained content on the mechanical properties of sandy pebble soil was analyzed from the macro and micro points of view, and the following conclusions were obtained: the stress-strain curve of sandy pebble soil exhibited strain softening with the rise of coarse-grained content; when the confining pressure was the same, the stress peak increased gradually with the increase of coarse-grained content; the shear strength of sandy pebble soil increased steadily with the increase of coarse-grained content, and the internal friction angle and cohesion of sandy pebble soil linearly increased with the rise of coarse-grained content; based on the micromechanical parameters of sandy pebble soil in the triaxial test, the micromechanical parameters of discrete element of sandy pebble soil of four different gradations were obtained using the comparative data method of numerical triaxial test, providing the theoretical support for the fine research on stability of sandy pebble soil; the increase of the coarse-grained content of sandy pebble soil resulted in increased contact stiffness and friction coefficient of micro parameters of discrete element, thus increasing the strength of sandy pebble soil; the dilatancy of specimens was caused by the particles in the upper and lower parts of the model moving to the middle and the particles in the middle of the model moving outward, and the numerical triaxial test can better reproduce the triaxial compression process. acknowledgements this work was supported by the national natural science foundation of china (51208069) and the state key laboratory of geohazard prevention and geoenvironment protection independent research project (sklgp2012z005, sklgp2016z010). 71macro-micromechanical properties of sandy pebble soil of different coarse-grained content references ahmad, n., hussain, t., awan, a.n., sattar, a., arslan, c., tusief, m.q., mariam, z. (2017). efficient and eco-friendly management of biodegradable municipalsolid waste (msw) using naturally aerated windrow compostingtechnique in district lahore pakistan. earth science pakistan,1(1), 01-04. andik, b., sarang, a. (2017). daylighting buried rivers and streams in tehran. water conservation and management, 1(2), 01-04. basarian, m.s., tahir, s.h. (2017). groundwater prospecting using geoelectrical methodat kg gana, kota marudu, sabah. earth science malaysia, 1(2), 07-09. bilal, a., yasin, m., ali, a. (2017). the geology and structure of neogene rocks in dadyal andadjacent areas, in the sub-himalayas, azad jammu and kashmir,pakistan. earth science malaysia,1(1), 15-20. cundall, p. a., &strack, o. d. l. (1979). a discrete numerical model for granular assemblies. geotechnique, 29(1), 47-65. geng, l., huang, z., & miao, y. (2011). meso-mechanics simulation triaxial test of coarse-grained soil. journal of civil engineering and management,28(4): 24-29. guo, q. (1998). the engineering properties and application of coarse grained soil. zhengzhou: yellow river water conservancy press. hejazi, s.m., lotfi, f., fashandi, h., alirezazadeh, a. (2017). serishm: an eco-friendly and biodegradable flameretardant for fabrics. environment & ecosystem science,1(2), 05-08. hoek, e.,&brown, e. t. (1998). practical estimates of rock mass strength. international journal of rock mechanics and mining sciences, 34(8), 1165-1186. kudus, k., grasian, i.,madasamy, s., john, a. (2017). immunomodulatory effect of alginic acid from brown seaweed sargassumwightiion disease resistance in penaeus monodon. journal cleanwas,1(1), 26-29. nema, a., yadav,k.d., christian, r.a. (2017). effect of retention time on primary media for greywater treatment. water conservation and management, 1(1), 01-03. omarashahestan, m.j., omarashastani, s. (2017). evaluating environmental considerations withchecklist and delphi methods, case study: suran city, iran. environment & ecosystem science,1(2), 01-04. qi, y., tang, x., & li, x. (2015). stress-induced anisotropy of coarsegrained soil by true triaxial tests based on pfc. chinese journal of geotechnical engineering, 37(12),2292-2300. rahman, n.a.,tarmudi, z., rossdy, m., muhiddin, f.a. (2017). flood mitigation measres using intuitionistic fuzzy dematelmethod. malaysian journal geosciences, 1(2), 01-05. ridzuan, a.a., zahar, u.a.u., noor, n.a.m. (2017). association of evacuation dimensions towards risk perception of the malaysian studentswho studied at jakarta, medan, and acheh in indonesia. malaysian journal geosciences, 1(1), 07-12. si, h. (1990). on the mechanical properties of cohesionless sand cobbles and rockfill materials. chinese journal of geotechnical engineering, 12(6), 32-41. wei, h., wang, r., & hu, m. (2008).strength behavior of gravelly soil with different coarse-grained contents in jiangjiagou ravine. rock and soil mechanics, 29(1), 48-51. wun, w.l., chua, g.k., chin, s.y. (2017). effect of palm oil mill effluent (pome) treatment byactivated sludge. journal cleanwas,1(2), 06-09. xu, x., wei, h., & meng, q. (2013). dem simulation on effect of coarse gravel content to direct shear strength and deformation characteristics of coarse-grained soil. journal of engineering geology, 21(2), 311-316. yasin, h., usman, m., rashid, h., nasir,d.r., randhawa, d.i.a. (2017). alternative approaches for solid waste management: a casestudy in faisalabad pakistan. earth sciences pakistan,1 (2), 07-09. yasin, m. (2017). diagenesis of miocene sandstone inthe district sudunhoti and poonch, azad jammu and kashmir, pakistan. pakistan journal of geology, 1(1), 05-07. yasin, m., shahzad, a., abbasi, n., ijaz, u., khattak, z. (2017). the use of stratigraphic section in recording quagmire ofinformation for the fluvial depositional environment – aworked example in district poonch, azad jammu and kashmir,pakistan. pakistan journal of geology, 1(2), 01-02. zhang, c., zhan, x., & yang, c. (2013). mesoscopic simulation of strength and deformation characteristics of coarse grained materials. rock and soil mechanics, 34(7), 2077-2083. lópez et al.indd earth sci. res. j. vol. 10, no. 2 (december 2006): 105-116 geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations jesús m. lópez k1., jesús moreira2, and josé gandarillas3. (1) geophysics and astronomy institute. department of geo-environment studies. calle ave. 212 no. 2906, entre 29 y 31, la lisa, cp 17100, ciudad de la habana, cuba. kramer@iga.cu (2) geology and paleontology institute. vía blanca y línea del ferrocarril, san miguel del padrón, cuba. moreira@igp.minbas.cu, moreira@gms.minbas.cu. (3) geophysics and astronomy institute. department of applied informatics. calle ave. 212 no. 2906, entre 29 y 31, la lisa, cp 17100, ciudad de la habana, cuba. pepegh@iga.cu abstract this paper describes the results from a study of the florencia gold-telluride deposit in central cuba, including mineralogical, petrographical, microprobe and chemical analysis. valuable information is provided for the exploration, mining and processing of gold ores from other nearby deposits with similar characteristics. results highlight changes in the mineralogical composition of the ores between the north and south sectors of the deposit, as reflected in metallurgical concentrates after beneficiation and flotation of samples from these sectors. it is shown that gold deposits of the cretaceous volcanic arc of cuba largely consist of native gold, telluride and pyrite, where arsenopyrite is almost absent. traces of lead, zinc and cadmium are present in the periphery of the main ore zones. key words: volcanic arch, cretaceous, mineralogy, electronic microprobe, petrography, gold, telluride, florencia, cuba. 105 earth sciences research journal manuscript received march 9 2006. accepted for publication december 4 2006. introduction despite the long history of mining of auriferous resources in cuba, the technique-productive and technological information was not conserved for different reasons. during several centuries the spanish settlers, and later on the small companies, developed underground or rudimentary mining in the holguín, guaimaro-jobabo, placetas and isla de la juventud areas, which were interrupted by different economical and historical factors. the florencia deposit is located in the e area of the country, in the guaimaro municipality, camagüey. it is known from the last century as a gold mine district and there are several reports from that period describing its mining and infrastructure. in the nineties, the international economic association macdonald mines ltd and geominera s.a. carried out an intense exploration program in florencia and in the surroundings of jobabo area (fig. 1), including maclama and golden hill located to the south. this exploration program led to a pre-feasibility study. works were stopped and no inversion took place because of low prices of gold. the results exposed in this paper include the underground explorations that were developed during the years 70-80 as a cooperation program between the cuban and u.s.s.r government. this information was taken from field works such as drillings and gallery samples, as well as available literature of the florencia district. it was also obtained important laboratory information such as petrographical, microprobe and chemical analysis. the principle aim of this study was to determine the mineral assemblage of gold ore deposits and the way in which precious gold and silver elements occur. the mining of these deposits may cause potentially adverse environmental impacts and some corrective measures are presented in order to mitigate them. materials and methods data were provided by the oficina nacional de recursos minerales from important areas with manifestations of gold located in the cretaceous volcanic arch and information published by foreign companies that carried out works of geological prospecting in different areas of the country. 15 thin sections, 35 polished sections, and 10 fire-essay analysis were also analysed. the results highlight the mineralogical differences between the products of the technological samples obtained in the north and south areas. the presence of minerals of the telluride group and the arsenopyrite absence in the ores, playing the pyrite an important part, is a characteristic feature of the auriferous sites located in the volcanic cretaceous arch. geological setting the cuban tectonic zonation is very distinctive. a continental scarp extends along the northern margin of the island and the narrow ophiolitic belt belonging to the ancient oceanic crust located to the south. the volcanic – sedimentary deposits complex, which represents the formation of an island arc, is developed along the central and southern parts of the territory. metamorphic rocks from the continental mass are well know in western and south-eastern cuba. during the aptian to campanian, a geodynamic volcanic island arc regime was established in the caribbean. outcrops can be observed along the camagüey province. in the central region of the country, the los pasos formation contains several deposits and prospects of the vms (kuroko) type with appreciable contents of cu, zn, and subordinated ag, au, cd and pb. several high anomaly areas have been described with presence of auriferous mineralization of epithermal type in the rocks of the cretaceous volcanic arch of the eastern part in the country, forming the ciego de avila camagüey las tunas and holguín belt (lópez et al., 1990; primelles et al., 1998; primelles et al., 2003; simon et al. (1999; capote, 2004). the camagüey region, is the most important because of its vast number of deposits, and constitutes a district of epithermal deposits. according to díaz de villalvilla et al. (2003), the nature of the rocks of the volcanic arc in camagüey allows to correlate them with those of the dominican republic (fig. 1). there are several types of mineral deposits in central cuba including high sulphide (golden hill) and low sulphide systems (beatriz, florencia), cu-mo-au porphyry (palo seco y la purísima) and au-ag-zn skarns (la purísima) deposits (primelles et al., 1998; primelles et al., 2003; simon et al., 1999). these authors considered the gold mineralization in the territory as an intrusive-related model with intrusive rocks and other representatives of the rocks of the arch, and state that epithermal deposits of the camagüey district are not related genetically with the plutonic bodies of the cretaceous geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations. 106 lópez et al., esrj vol. 10, no. 2. december 2006 fig. 1. central cuba deposit location. volcanic arch. simon et al. (1999) suggest that there is a relationship between the mineralization and the rhyolites of the sierra formation. this paper underlines the mineralogical characteristics of these deposits by pyrite presence as the main mineral, either disseminated or as sulphur; the absence of arsenopyrite and the occurrence of a variety of minerals such as gold, silver, lead, bismuth and nickel of the telluride’s group: petzite, calaverite, silvanite, hessite, bolinskite, shtiutzite, melonite, and altaite, as well as the presence of gold of high purity and electrum (torres et al., 2005; lópez, 1988; lópez et al., 1998; bortnikov et al., 1988). the geochemistry of halos of cu, ag, pb, zn and mo are characteristic for this type of location. el pilar epithermal prospect is located in the western part of cuba (fig.1), about 24 kms from ciego de avila city. beatriz is the better-studied deposit, located to the north of the guaimaro area (simon et al., 1999). the second in importance is florencia located also in the guaimaro area. golden hill is located to the south of las tunas, and it is representative of sulphate, exposing an area of oxidation. the jobabo municipality area exhibits several groups of small auriferous manifestations such as georgina, iron hill, and maclama. golden hill (little golden hill, three hill and big golden hill sectors), florencia and maclama are located close to the guaimaro village. they are located in a smooth relief and the presence of an oxidation zone makes them feasible for open pit mining using the heap leaching method or using small facilities as vat leaching. regardless of beneficiation-mining process used, it is necessary to be cautious because toxic elements are present, and there are water reservoirs and rivers crossing the region. mineralization the mineralization studied in the florencia deposit has a structural control that extends the body laterally up to 1 km, with thickness of 3040 m. the well-known mineral bodies are four: three in the north area and one in the south. they have forms of long narrow veins, with at least three quartz generations. the quartz types 107 are sterile milk-like quartz, darker quartz -where mineralization is foundedand typical quartz filled open veins with calcite carbonates. gas – liquid inclusions were studied by bortnikov et al. (1988), paded mainly for quartz and carbonates (fig.2). the florencia host rocks are traquibasalts and cretaceous (albian-turonian) andesite basalts. the volcanic rocks are cut by the granodiorite stock and quartz-diorites belonging to the guaimaro stock, and also by a few younger diorites and granodiorites dikes. in the limits of the mineral field, there are two isolated areas denominated north and south, which differ in their mineralogical composition. they are represented mainly by vein – disseminated sulphurous mineralizations. the north area is divided in three bodies with some ramifications, and the dominant mineralization is pyrite-chalcopyrite (fig 3). the south area is lens-shaped with pb-zn-ag-cu poly-metallic mineralization, being better developed the pbzn-ag poly-metallic mineralization. the mineral bodies of the north and south area have been explored using perforation wells up to 170 m depth and 300 m of strike, and underground mining. fig. 2. florencia’s mineral bodies’ district map (shestokanova, 1967). geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations. 108 fig. 3. small vein located in the second level of the south wall. the main ore is also indicated in the right part of the photo. mineralogical characteristics near 30 minerals are recognized in the veins at the florencia deposit. by-product minerals include different quartz types and textures such as crustiform, banded and cockade with different colours, in occasions grey with open cavities, typical of crack fills. starting from the thermodynamic analysis of the mineral associations in the system au-ag-te (bortnikov et al., 1989), the formation temperatures of the deposit varied between 70 and 280 °c. the main economic components of some veins are gold and silver. the native gold is the major metallic mineral in the ores. it is observed in polished sections in grains of different sizes. table i shows the results obtained by the author related to the count of the grains of gold with different sizes observed in 100 polished sections. in table ii, the study of the mineral composition of the ores demonstrated that the finest gold of high purity is present, as well as gold and silver in form of telluride’s calaverite (aute2), silvanite (au ag te4), petzite (agaute2), hessite (agte), shtiutzite and bolinskite (agbite2). the main concentrations of gold are found in native gold and petzite, broadly distributed among tellurides, and in silver concentrates, mixed with native gold, hessite, and petzite. these data show how gold and silver are represented in the ores, being the tellurides important in their composition. comparing north and south ore sectors, it is observed a smaller development of the sphalerite and galena in the north. here the pyrite contains inclusions of native gold sizes mm number of grains 0, 00 n – 0,015 16 0,020 – 0,040 6 0, 040 – 0,080 4 0,080 – 0,160 2 table. i. grains of gold counting lópez et al., esrj vol. 10, no. 2. december 2006 109 fig. 4. hessite and altaite in galenite. increase 200 x. fig. 5. vein of native gold and petsite in pyrite ii. increase 200 x. geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations. 110 table. 2. electron microprobe analyses (mass. %) and segregation of gold tellurides (calaverite and petzite) (fig 3, 4, and 5).the analyses of the pyrite mono-fractions with high contents of gold (64 up to 450 g/t), underline the significant role of the pyrite as concentrator of gold in this sector. in the south sector, the galena and the sphalerite are diffused, the native gold prevails in form of small inclusions in the pyrite; and inside the group of tellurides, the altaite and silver tellurides are diffused. results. there are two mineral forms present in florencia deposit ores: gold and silver. the presence of high purity gold and silver tellurides, evidences that a quantitative balance of the interrelations between these two types of ores is needed. during exploration works in 1990, the camagüey geology company took two technological samples of 4 tons each one: technological sample-3 (mt3) in the north sector and technological sample4 (mt-4) in the south sector. the analyses were carried out by the centre of investigations for the mining metallurgic industry executing the investigations for beneficiation by flotation and gravitation enrichment methods. the authors studied the flotation concentrate and gravitation of the mt-4 sample and flotation of the mt-3 sample. the analysis of the three concentrates of the technological mt-3 and mt-4 samples were carried out by the moscow institute of geology and geochemistry of the minerals in order to know the contents of au, ag, te and bi. portions of the samples were dissolved in concentrated hno3 and heated for one hour, and the remaining product, was solved in regal water. the determination of au, ag, te, and bi was done in the concentrate samples and in the break-up samples (after the concentrated hno3 treatment) and later all samples were tested using atomic absorption spectroscopy methods. the analysis was carried out with the contents of no. mineral au ag pb bi te s total 1 petzite 24.80 41.62 33.43 100.05 2 petzite 25.95 40.36 32.61 98.92 3 petzite 49.36 34.62 102.38 4 petzite 11.07 51.72 33.22 98.01 5 hessite 10.01 54.12 36.66 100.79 6 hessite 1.43 59.10 0.45 39.40 100.38 7 hessite 0.60 59.70 0.44 0.27 39.16 100.17 8 hessite 0.12 59.81 1.41 0.10 38.34 99.78 9 hessite 60.00 0.90 38.58 99.48 10 hessite 1.54 60.54 0.37 36.25 100.70 11 shtiutzite 54.70 0.20 42.47 97.39 12 calaverite 41.18 0.47 57.27 98.92 13 calaverite 43.21 56.86 102.09 14 bolinskite 16.67 0.36 36.02 42.59 0.83 98.71 15 bolinskite 13.39 0.10 37.57 43.57 1.13 98.79 16 bolinskite 18.31 35.16 46.14 99.61 17 tsumoite 61.65 37.31 98.66 18 raklidzite 1.61 14.33 37.53 45.52 0.16 99.45 19 altaite 0.18 0.90 57.65 1.23 39.25 98.21 20 altaite 0.60 58.26 1.15 38.92 98.93 21 altaite 0.24 1.09 58.66 0.32 37.03 97.34 lópez et al., esrj vol. 10, no. 2. december 2006 111 these elements in the levels from 1-10 g/t after their concentration. the results are shown in the table iii highlighting the differences between the contents in the elements analysed in the mt-3 and mt4 samples. the content of gold in mt-3 is four times the content in mt-4, as well as the silver and telluride concentrations, in both concentrates are similar but bigger than gold concentration. the relationship ag/au, in the concentrate mt 4 is 5.1 (gravitation) and 4.4 in flotation, and 1.35 in mt-3. the most important quantitative evaluation of precious metals is obtained starting from the analysis of the minerals of the telluride groups in the concentrate and in the solutions after its treatment with hno3. the quantity of gold that passes to the solution used to find telluride forms in the ores is 10.4 11.5 -% for mt-4 and 7.1%. for mt-3 . the silver, contrary to the gold, is in telluride form in more than 95% .it passes to the breaks-up with the hno3 treatment of the concentrate. it is important to highlight some differences in the composition of the concentrates obtained in mt-3 and mt-4 technological samples shown in the spectral analysis: 1. the biggest pb and zn contents in mt-4 concentrate reflect relative bigger sphalerite and galena quantities in the ores of the south sector. 2. the biggest quantities in sphalerite are related with biggest contents of cadmium (0.15%) in the mt-4 flotation concentrate. 3. it is important to note the cobalt high contents in the concentrate (table iv) and high co/ni relationship values (from 10 up to 20). the content of cobalt increases twice in the mt3 concentrate and it reaches up to 0.11%. this is conditioned by the quantities of pyrite in this concentrate and it is also confirmed by the pyrite mono-fraction results where the cobalt content varies from 0.15% to 0.40%. the data demonstrate that future technological research should take into account the presence of minerals from the telluride group in ores related to precious metals (table v), the silver and the presence of small inclusions of native gold and gold telluride in pyrite. geological factors with potential environmental effects some epithermal gold deposits have been defined and studied in the volcanic cretaceous arch (e.g., little golden hill, big golden hill, three hill, florence, and jacinto). there are other important such as el pilar-san nicolás, la purisima, tres antenas, corral de rojas, and la mina. the presence of several types of mineral deposits genetically distinctive in these formations are characterized by pyrite presence as main mineral, table. iii. chemical analyses results of the concentrate phases, solutions, and break-up remains. concentrates analysis g/t element au ag te bi mt–4 gravitation concentrates 35 179 188 44 break-up 4.0 172 162 42 mt–4 flotation break-up remains 32 7.2 18 5 concentrates 39 172 178 44 break-up 4.1 165 160 42 mt–3 flotation break-up remains 35 5.0 15 5 concentrates 140 190 220 26 break-up 10 182 200 25 break-up remains 128 7.5 21 5 geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations. 112 table. iv. r esults of the spectral quantitative analysis of the concentrate. c oncentrates e l e m e n t s a ug/t a g g/t z n% c d% m n% p b% b i % c u% c o% n i % sb% a s% sn% m t-4 z one south g ravitation 39.8 100.0 1.9 0.055 0.031 1.0 0.007 0.32 0.047 0.0043 0.090 0.056 m t-4 z one south flotation 33.0 84.0 5.0 0.150 0.090 1.0 0.83 0.50 0.0050 0.040 0.050 0.005 m t-3 z one n orth flotation 122.0 90.0 1.2 0.012 0.027 0.058 1.50 0.110 0.0050 0.022 0.050 0.002 lópez et al., esrj vol. 10, no. 2. december 2006 113 either disseminated or accompanied by quartz in veins. the almost absence of arsenopirite in the ores and the presence of a variety of minerals such as gold, silver, lead, bismuth and nickel belonging to the telluride group is an important common feature to keep in mind during future mining operations. the existence of at least three small deposits with advanced studies and the data presented here, as well as the high prices of gold and silver at present time, permitted to start the mining process in golden hill, florencia and beatriz, and defined the corresponding environmental studies in order to take the systematic measures to avoid or decrease damage as possible. • size of the deposit: they are small and located near each other. they stand out as small elevations on the surface. typical oxidation zones reach 10-20 m depth. ores are formed by pyrite as the main mineral, quartz, carbonates, and mineral loamy product of the wall – rock alteration. the reserves are approximately 5 10 tons of gold. the veins usually have high gold contents and erratic distributions could be present. small mining operations are needed to open pit facilities and benefit chord in the studied resources. • the mobility of the metals and the acid drainage product of the water circulation and the pyrite presence during the mining activity (open pit), could be limited directly by the presence of carbonates in the mineral bodies and associated with the wall rocks. • the oxidation of these sulphurs could generate acid drainage, which can potentially affect the superficial drainages of the area of influence in the mine and consequently enter in the food-chain. some detailed studies are proposed about risks in the area (e.g., the use of some types of biotechnological barriers such as bacteria or mushrooms that neutralises the environmental contamination during the exploitation works in the area). du bray (2004) recollected samples of stream sediment using the open-pit method and heap leaching showing results of < 5 to 12 silver ppm, 0.04 to 0.4 ppm of gold, < 5ppm cadmium, 18 to 81 ppm cooper, 1.4 to 4.2% of iron 200 to 2.020 ppm, manganese, 7 to 197 ppm lead, and 70 to 1.131 ppm zinc. the existence of at least 3 small deposits with advanced studies and the data presented here, as well as the high prices of gold and silver at present time, permitted to start the mining process in golden hill, florencia and beatriz, and defined the corresponding environmental studies in order to take the systematic measures to avoid or decrease damage as possible. table. v. te determination in ores by hidruros generation methods (north area). level 0. mineral body # samples content te g/t 1 5 91 1 7 87.5 1 8 36.6 5 14 13.4 1 19 12.2 3 21 24 1 22 40.6 4 28 2.2 geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations. 114 conclusions. 1. the presence of telluride group minerals and the absence of arsenopyrite in the cretaceous volcanic arch ores, being the pyrite the most significant, represent a characteristic feature for the auriferous locations. 2. the florencia ore deposit has important mineralogical differences between the two main areas: north and south. 3. the highest pb and zn contents in the mt-4 concentrate reflect the relative bigger quantity of sphalerite and galena in a small part of the south sector. 4. the biggest quantities of sphalerite are related to the biggest contents of cadmium (0.15%) in the mt-4 flotation concentrate. 5. the environmental impact caused by open pit and the pyrite presence (acid drainage mining), and the beneficiation methods used will give tellurides an inadequate final destination as residuals from the mining or during the mining processes. recommendations. optimal operation conditions should be studied more in detail at a proper scale for the industrial recovery of the gold and silver associated to tellurides, as well as the potentially environmental impacts for that process. references bortnikov, n. s., lópez, j. m., guenkin, a. d., krapiva, l. y. and santa-cruz, m. (1988). paragenesis of gold and silver tellurides in the florencia deposit. cuba. international geology review. 30. 294-306. capote, c. m. (2004). ‘análisis estructural sin-arco: una herramienta imprescindible en la evaluación metalogénica de cuba centrooriental’. in: estudios sobre los arcos volcánicos de cuba. centro nacional de información geológica. i.g.p. cd. cuba. 3342. díaz de villalvilla, l., milia, i., santa-cruz, m. and aguirre, g. (2003). ‘formación los pasos: geología, geoquímica y su comparación con el caribe’. in: estudios sobre los arcos volcánicos de cuba. centro nacional de información geológica. i.g.p. cd. cuba. 54-61. du bray, e. (2004). preliminary compilation of descriptive geoenvironmental mineral deposit models.u.s geological survey. open file report. 95 – 831. escobar, e. (1994). mineralización de metales básicos y preciosos, asociados al arco volcánico cretácico de la región ciego de ávila -camagüeylas tunas. in: congreso cubano de geología y minería. ii congreso de geología y minería. programas y resúmenes. santiago de cuba: [s.n.], p. 96. lópez, j. m., moreira, j., pantaleon, g. j., lavandero, r. m., montano, j. and cruz–martín, j. (1998). tipos mineralógicos de algunos yacimientos auríferos de cuba. iii congreso cubano de geología y minería (geomin 98). geología y minería 98. la habana. 371-374. lópez, j. m. (1990). informe del tema 40109. evaluación pronóstico de la mineralización aurífera de cuba, para el mapa metalogénico pronóstico, a escala 1:500 000 instituto de geología y paleontología. archivo igp. archivo onrm. poznaikin v., morales alberto., cañete carlos. lópez, j. m. (1988). composición sustancial y asociaciones parageneticas de los yacimientos hidrotermales auríferos de cuba. referat. moscú. 1-27. primelles, l., alvarado, b. and torres, m. (2003). ‘campo mineral maclama. breve caracterización geológica y mineralógica de la mineralización oro-telurídica. tipos esperados de depósitos’. in: estudios sobre los arcos volcánicos de cuba. cd. centro nacional de información geológica. i.g.p. cuba. 33-42. primelles, l., barroso, a., lugo, r. and escobar, e. (1998). geología y metalogenia del campo mineral guaimaro. enfoque actual. iii congreso cubano de geología y minería (geomin 98). la habana. 390-393. lópez et al., esrj vol. 10, no. 2. december 2006 115 simon, g., kessler, s. e., russell, n., hall, c. m., bell, d. and piñero, e. (1999). epithermal gold mineralization in an old volcanic arc: the jacinto deposit, camagüey district, cuba. economic geology. 94. 487-506. torres, m., sastroputro, s., ulloa, m., pérez, m., de la nuez, d., santa-cruz, m. and alvarado, b. (2005). petrología, mineralogía y alteración hidrotermal de la manifestación de cu porfídico del sector palo seco. camagüey, cuba. i convención cubana de ciencias de la tierra. geociencias 2005. s.c.g. 141-150. zhestokanova v, (1997). informe de los trabajos de búsqueda y exploración realizados en los años 1964-65 en los yacimientos auríferos de la zona de guaimaro en la provincia de camagüey. c.n.f.g, cuba. geology of the florencia gold – telluride deposit (camagüey, cuba) and some metallurgical considerations. 116 untitled earth sciences research journal earth sci. res. j. vol. 13, no.1 (june 2009): 6-15 vertebrates of the marília formation (late maastrichtian) from the peirópolis paleontological site: toward a better understanding carlos roberto a. candeiro laboratório de geologia, curso de geografia, campus do pontal, universidade federal de uberlândia, avenida josé joão dib, 2545, bairro progresso, zip code 38302-000, ituiutaba, minas gerais state, brazil corresponding author. tel.: ++ (55) 32692389. e-mail address: candeiro@yahoo.com abstract the peirópolis paleontological site (late cretaceous, maastrichtian) in minas gerais state, brazil yielded an important assemblage of fossil vertebrates. the typical occurrence of south american widespread taxa in peirópolis is important for correlation between the brazilian bauru basin and argentinean late cretaceous basins. the fishes, turtles, anuran, crocodilians and dinosaurs known from peirópolis (marília formation) resemble the patagonian latest late cretaceous vertebrate faunas but lacks ornithischian dinosaurs. key words: vertebrates, late cretaceous, marília formation, minas gerais state, brazil. resumen en el sítio paleontológico de peirópolis (cretácico superior, maastrichtiano) ubicado en la província de minas gerais, brasil, hay un contenido muy importante de vertebrados fósiles. la ocurrencia en peirópolis de taxa típicos y de amplia distribución en la américa del sur es importante para la correlación entre la cuenca bauru y cuencas argentinas del cretácico superior. los pesces, tortugas, anuros, cocodrilianos y dinosaurios de peirópolis (formación marília), se equivalen a las faunas de vertebrados del neocretácico de patagonia, sin embargo, sin la presencia de dinosaurios ornitísquios. palabras claves: vertebrados, neocretácico, formación marília, província de minas gerais, brasil. 6 manuscript received: january 10th, 2009. accepted for publication: april 30th, 2009. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:37 p p p composite 133 lpi at 45 degrees introduction the upper cretaceous beds at the peirópolis paleontological site of (uberaba municipality, minas gerais state; fig. 1) consist on an entirely continental sedimentary succession. the upper cretaceous bauru group units at peirópolis are divided into the uberaba formation (coniacian-santonian) and marília formation (late maastrichtian) in ascending order. the uberaba formation occurs as small patches mainly in the peirópolis village. although extensive paleontological studies have been carried out in peirópolis area (e.g., price, 1955; estes and price, 1973; báez and peri, 1989; gayet and brito, 1989; carvalho et al., 2004; frança and langer, 2005; kellner and campos, 2005; kellner et al., 2005; campos et al., 2005; novas et al., 2005; santucci and bertini, 2006; novas et al., 2008; salgado and carvalho, 2008; candeiro et al., 2008), fossil vertebrates are unknown so far from uberaba formation, except for unstudied dinosaur eggs found near peirópolis. on the other hand, abundant vertebrate remains have been reported from the late maastrichtian marília formation (specimens and localities are listed in the tab. 1 and 2). the first discovery of vertebrate fossils was vertebrate fragments from the “ponto 1 do price” from the serra da galga member (marília formation) in the 1940´s (price, 1951; candeiro and bergqvist, 2004). since then, scattered dinosaur bones and other vertebrate fossils including fishes, turtles, and crocodilians have been found in quarries, roads cuts, and stream beds as a result of a centro de pesquisas paleontológicas llewelyn ivor price (peirópolis) team survey. abundant plants (palynomorphs) and mollusks have been located from the peirópolis quarries. they clearly indicate that diversity of vertebrate communities existed during the time of marília formation deposition in the peirópolis area. except for some the crocodilians, however, the vertebrate faunas remain inadequately published. therefore, the purpose of this paper is to review the vertebrate fossils from the marília formation at this locality and report their occurrence to the paleontological community. comments and correlations with other vertebrate-bearing formations of late cretaceous age from argentina are discussed. the potential significance of marília formation vertebrates from peirópolis may be seen especially in the 7 vertebrates of the marília formation (late maastrichtian) from the peirópolis paleontological site: toward a better understanding n 0° 70°w 10°s 20°s 40°w 50°w 60°w 51”00’ 18”00’ 47”30’ 18”00’ 20”30’ 51”00’ brazil 20”30’ goiás state são paulo state 47”30’ 0 10 20 km legend town localization study area road river par ana iba rive r araguari river tejuco river peirópolis grande river uberaba prata ituiutaba araguari uberlãndia figure 1. geographic distribution of the upper cretaceous “marília-points” of peirópolis paleontological site. point 1 (5 quarries); point 2 (1 quarry); point 3 (1 quarry); point 4 (2 quarries); point 5 (1 quarry). julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:39 p p p composite 133 lpi at 45 degrees context of roughly contemporaneous areas from southern south america (candeiro and bergqvist, 2004). methodology the data on vertebrates from peirópolis paleontological site for the present work are mostly based in literature sources as well as in the direct observation of the specimens deposited at museu dos dinossauros (peirópolis district, uberaba town, minas gerais state, brazil). for the stratigraphic units in the bauru group i follow the arrangement proposed by dias brito et al. (2001). results geological setting the peirópolis paleontological site is northeast of bauru basin (upper cretaceous) (fernandes and coimbra, 1996). the bauru group in triângulo mineiro region is represented by sediments belonging to the adamantina, marília and uberaba formations (fig. 2). these layers overlie basalts belonging to serra geral formation (são bento group, paraná basin), sandstone belonging to botucatu formation (são bento group), metamorphic and proterozoic outcrops belonging to araxá and canastra groups (sanfransiscana basin) and mesozoic intrusion of the 8 carlos roberto a. candeiro figure 2. geological map of bauru group in triângulo mineiro region (modified fernandes and coimbra, 1996). julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:39 p p p composite 133 lpi at 45 degrees soerguimento of alto paranaíba (suguio et al., 1979). outcrops of the marília formation are unevenly distributed throughout the peirópolis paleontological site. according to barcelos and suguio (1987), that unit was deposited by coalescing alluvial fans, and later reworked by a braided system in association with calcretes and lacustrine calcareous sediments. barcelos (1984) subdivided the marília formation into the echaporã, ponte alta, and serra da galga members, although only the latter two members are exposed in the peirópolis area. ponte alta and serra da galga members show the following characteristics (garrido et al., 1992): ponte alta member– a basal “calcareous white member”, characterized by thin and medium calcareous beds with pebbles and calcareous nodules which were deposited in lakes; serra da galga member– “member of white sandstone and conglomerate”, composed of bleached conglomerate and sandstone with feldspar matrix deposited by alluvial fans, braided rivers and lakes. those two members, according to suguio (1973, 1980), were formed in a partially dry climate, which would have favored the fossil preservation found in these layers. dias-brito et al. (2001) assigned a late maastrichtian age to the marília formation. this lithostratigraphic unit yielded invertebrates and plants, and abundant vertebrates (fig. 3). vertebrates from the marília formation are represented by anurans (baurubatrachus pricei), lizards (pristiguana brasiliensis), crocodilians (itasuchus jesuinoi, peirosaurus tormini, uberabasuchus terrificus), chelonians (cambaremys langertoni), dinosaurs (abelisauridae, carcharodontosauridae, maniraptora, and eutitanosaurian baurutitan britoi, trigonosaurus pricei, uberabatitan ribeiroi, aeolosaurus). vertebrate paleontology the vertebrate fossils of the triângulo mineiro region have been known since 1920, however only with the pioneer studies of paleontologist llewellyn ivor price, which started in 1947, the fossils found in the peirópolis paleontological site became not only more known, but at the same time better classified. just a small part of collected material in this paleontological site was described to the level of species due to fragmentary nature of most the specimens which made a more refined description impossible (e.g., fig. 4). abundant fish fossils have recently been found in the peirópolis. to date four localities have yielded isolated scales and teeth. they produced characiformes, perciformes and siluriformes. the species which were described formally are represented by amphibian baurubatrachus pricei báez and peri, 1989; lepidosaurian pristiguana brasiliensis estes and price, 1973 and crocodilians itasuchus jesuinoi, peirosaurus tormini price, 1955 9 vertebrates of the marília formation (late maastrichtian) from the peirópolis paleontological site: toward a better understanding bauru group peirópolis site serra da galga mb. (marília fm.) ponte alta mb. (marília fm.) uberaba fm. legend plant vertebrate fish frog lizard turtle crocodilian sauropod theropod figure 3. upper cretaceous bauru group generalized geological section in peirópolis showing occurrence of the biota. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:39 p p p composite 133 lpi at 45 degrees and uberabasuchus terrificus carvalho, ribeiro and avilla, 2004. the dinosaurs represent the most abundant remains and they are represented by theropods which were attributed by candeiro (2002) and candeiro et al. (2004) to abelisauridae and carcharodontosauridae, first abelisauridade bone remains by (novas et al., 2008) and maniraptoran claw recently described by novas et al. (2005). the sauropods are represented by the titanosaurians baurutitan britoi campos and kellner, 2005 and trigonosaurus pricei campos, kellner, bertini and santucci, 2005; uberabatitan ribeiroi salgado and carvalho, 2008 and others different elements e.g., teeth, vertebrae, ribs, hemal arches, phalanges and osteoderms. these specimens were studied by campos and kellner (1999), 10 carlos roberto a. candeiro figure 4. reptilian fossils from the peirópolis paleontological site. a, lower jaw itasuchus jesuinoi (modified from price, 1955); b, pelvis of the “titanosauridae” (modified from campos and kellner, 1999); c, abelisauridae teeth; d, carcharodontosauridae tooth; e, theropoda indet. tooth. scale bar in mm. table 1. vertebrate fauna of the marília formation from peirópolis paleontological site. area 1 area 2 area 3 area 4 area 5 ponte alta member x x x serra da galga member x x vertebrata vertebrata indet x x x x x pisces characiform x x perciform x x siluriform x x anura baurubatrachus pricei x reptilia squamata pristiguana brasiliensis chelonia chelonia indet. podcnemidae crocodylomorpha itasuchus jesuinoi peirosaurus tormini uberabasuchus terrificus dinosauria sauropoda titanosauria x x x x x x julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:40 p p p composite 133 lpi at 45 degrees 11 vertebrates of the marília formation (late maastrichtian) from the peirópolis paleontological site: toward a better understanding area 1 area 2 area 3 area 4 area 5 “titanosauridae” a “titanosauridae” b “titanosauridae” c “titanosaurinae” “titanosaurus” aeolosaurus sp. baurutitan britoi trigonosaurus pricei theropoda theropoda indet. abelisauridae indet. carcharadontosauridae indet. maniraptora x x x x x x x x x x x x x table 2. fossiliferous areas (quarries), in the peirópolis paleontological site, minas gerais state. areas generalities reference selected area 1 (quarry 1) this outcrop has been known as “caieira” campos and kellner (1999) area 1 (quarry 2) this outcrop is inside of “caieira” kellner (1996) area 1 (quarry 3) this quarry has been known as “point 2” price (1955) area 1 (quarry 4) this outcrop has been known as “point 3” campos and kellner (1999), trotta et al. (2002) área 1 (quarry 5) this outcrop is known as “mumbuca” powell (1987, 2003) area 2 the outcrop is situated on the br-262 throughway about 3,5 km east of peirópolis campos and kellner (1999) area 3 the outcrop is situated on the ponte alta district throughway about 18 km east from peirópolis magalhães-ribeiro (2002) area 4 (quarry 1) quarry 1 is known as the “ rio tejuco 1” albuquerque and candeiro (2003) area 4 (quarry 2) quarry 2 is known as the “rio tejuco 2” albuquerque and candeiro (2003) area 5 this area has been known as “partezan quarry” bertini et al. (1993) julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:40 p p p composite 133 lpi at 45 degrees santucci (2002), powell (2003) and santucci and bertino (2006) who attributed part of these remains to titanosauria (see detailed assignments in tab. 1). dinosaur eggs found in sediments of the serra da galga member are related to “megaloolithidae” oofamily, according to magalhães-ribeiro (2002). furthermore there are numerous cranial and post cranial remains belonging to crocodilians and turtles. numerous fragments of disarticulated bones and coprolites are found in marília formation sediments in the peirópolis paleontological site, however they have not been attributed to taxonomic level below vertebrata; they are housed in the centro de pesquisas paleontológicas llewellyn ivor price and museu de ciências da terra (rio de janeiro). fossil sites from marília formation in the peirópolis paleontological site of: the first fossil site was studied by price in the 1940´s. since then, new fossil localities of the marília formation have been found in the uberaba municipality with five mentioned in literature so far (fig. 1; tab. 2). the best known fossil vertebrates of marília formation found in the peirópolis paleontological site of are listed below. discussion the present study and those of the previous works in the marília formation peirópolis paleontological site), it is attempted here to point a comprehensive point of view of this area. the peirópolis paleontological site contains a moderately diverse late cretaceous (maastrichtian) vertebrate assemblage, from which some of the taxa are reported also in argentina (e.g., chubut, neuquén and malargüe groups; candeiro and martinelli, 2003; lamanna et al., 2003; martinelli and forasiepi, 2004; leanza et al., 2004). some of the vertebrate taxa reported at peirópolis occur with other vertebrates known from other areas of argentina, and are important for correlation with argentinean units. supporting this correlation are the mesoeucrocodylian peirosaurus tormini and the titanosaur aeolosaurus, which are found in bajo de la carpa, allen, angostura colorada, los alamitos, and bajo barreal formations in patagonia. abelisaurid and carcharodontosaurid remains are known to occur in the bauru group. in addition to the titanosaurian material from peirópolis area, remains of those sauropods are widely known in upper cretaceous units of argentina. additional support for correlation is given by the presence in peirópolis of the gondwanan “megaloolithidae” eggshells. other support of the correlation between peirópolis area and argentinean areas may be given by other vertebrate groups, but more detailed studies are needed before this can be done. for example, characiform, siluriform and perciform fishes and podocnemid turtles are known in most of the neuquén and malargüe groups, but they cannot be used as correlation tools since their remains are not well known from the marília formation. ornithischian records are remarkable in argentina (late cretaceous), however, not a single ornithischian remain has been identified so far in the marília formation or the bauru group. considering that ornithischians occurred in argentina, this bias may be from lack of extensive or least systematic prospecting in the marília formation rather than from depositional or preservational events. conclusions so far, four fossil sites have been recorded at the peirópolis paleontological site of, the marília formation with a diverse vertebrate fauna of upper maastrichtian age. these fossils are represented mainly by fishes, frogs, turtles, crocodilians, and dinosaurs. the diverse fossiliferous, but poorly known marília formation accumulated in the northwestern bauru basin during the latest cretaceous. deposition occurred on a fluvio-lacustrine environment. the maastrichtian age of the marília formation suggested by previous authors has several important biogeographical and paleontological correlations. first, it indicates that the marília formation assemblage is approximately contemporaneous in age with 12 carlos roberto a. candeiro julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:40 p p p composite 133 lpi at 45 degrees the vertebrate assemblage recovered from the patagonia (e.g., chubut, malargüe, and neuquén groups). this finding is certainly consistent with striking similarities between faunas from these two southern south america areas. second, this maastrichtian marília formation fauna, when considered in the light of the late cretaceous discoveries, was different by the lack of ornithischian records in brazil, but which are found in argentina. acknowledgments this paper forms part of the undergraduate project of the r. candeiro who gratefully acknowledges financial assistance in the form of an undergraduate fellowship (1998-1999) received from the council of scientific from brazil (cnpq/166/98) under dr. ignácio brito advisor (financial supported by fapemig/ufu/801/98 fellow). the manuscript was substantially improved after the revision work of dr. yaneth muñoz-saba (universidad nacional de colombia). i would like to thank prof. luiz carlos borges ribeiro and the centro de pesquisas paleontológicas llewellyn ivor price crew, dr. adriano rodrigues dos santos (universidade federal de uberlândia, ufu), dr. lílian paglarelli bergqvist (universidade federal do rio de janeiro, ufrj) and thiago da silva marinho (ufrj) for reviewing the earlier drafts of the manuscript. the manuscript was substantially improved by of dr. simone magnuco (museo civico di storia naturale di milano, italy), dr. walter hartwig (touro university, united states) and dr. david gillette (museum of northern arizona, united states). the author also would like to thank the friends, lisete, ana and fernando brito, that in some ways helped us to start our paleontological studies, since they are wife, daughter and son of professor ignácio machado brito, who opened the doors of his house and ignited my life into paleontology. references albuquerque, j.a.c. and candeiro, c.r.a., (2003). new potential fossiliferous site uberabauberlândia higway (maastrichtian) – triângulo mineiro, minas gerais state, brazil. paleontologia em destaque. 40. 4-5. báez, a.m. and peri, s., (1989). baurubatrachus pricei, nov. gen. et sp., un anuro del cretácico superior de minas gerais, brazil. anais da academia brasileira de ciências. 61. 447-458. barcelos, j.h., (1984). reconstrução paleogeográfica da sedimentação do grupo bauru baseada na sua redefinição estratigráfica parcial em território paulista e no estudo preliminar fora do estado de são paulo. rio claro. 190 p. dissertation (livre docência). departamento de geociências, universidade estadual paulista. barcelos, j.h. and suguio, k., (1987). correlação e extensão das unidades litoestratigráficas do grupo bauru, definidas em território paulista, nos estados de minas gerais, goiás, mato grosso do sul e paraná. boletim, 6º simpósio regional de geologia, unesp, são pedro, são paulo: 313-321. bertini, r.j., marshall, l.g., gayet, m. and brito, p. m., (1993). the vertebrate fauna of the adamantina and marília formations, upper cretaceous of the paraná basin, southerneast brazil. neus jahrbuch fur mineralogie, geologie und paläeontologie. 188. 71-101. campos, d.a. and kellner, a.w.a., (1999). on some sauropod (titanosauridae) pelves from the continental cretaceous of brazil. national science museum monographs. 15. 143-166. campos, d.a., kellner, a.w.a., bertini, r.j., and santucci, r.s., (2005). on a titanosaurid (dinosauria, sauropoda) vertebral column from the bauru group, late cretaceous of brazil. arquivos do museu nacional. 63 (3). 565-593. candeiro, c.r.a., (2002). dentes de theropoda da formação marília (santoniano-maastrichtiano), bacia bauru, região de peirópolis, uberaba, minas gerais, brasil. rio de janeiro. 136p. msc thesis, programa de pós-graduação em geologia, universidade federal do rio de janeiro. 13 vertebrates of the marília formation (late maastrichtian) from the peirópolis paleontological site: toward a better understanding julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:40 p p p composite 133 lpi at 45 degrees candeiro, c.r.a. and martinelli, a.g., (2003). comparison of dinosaurs and sedimentary environments between the bauru and neuquénmalargüe groups (upper cretaceous): a preliminary approach. paleontologia em destaque. 17. 48-48. candeiro, c.r.a. and bergqvist, l.p., (2004). paleofauna cretácica do centro de pesquisas paleontológicas llewellyn ivor price, peirópolis, uberaba, minas gerais. revista sociedade & natureza. 16 (31). 27-35. candeiro, c.r.a., bergqvist, l.p., novas, f.e. and currie, p.j., (2004). theropod teeth from the marilia formation (upper maastrichtian), minas gerais state, brazil. journal of vertebrate paleontology. 24 (3, suppl.). 43a. candeiro, c.r.a., santos, a.r., bergqvist, l.p., ribeiro, l.c. and apesteguía, s., (2008). the late cretaceous fauna and flora of the uberaba area (minas gerais state, brazil). journal of south american earth sciences. 25. 203-216. campos, d. a.,; kellner, a.w., bertini, r.j., santucci, r.m., (2005) . description of a titanosaurid caudal series from the bauru group, south-east brazil. arquivos do museu nacional. 63. 565-593. carvalho, i.s., ribeiro, l.c.b. and avilla, l.s., (2004). uberabasuchus terrificus sp. nov., a new crocodylomorpha from the bauru basin (upper cretaceous), brazil. gondwana research. 7 (4). 975-1002. dias-brito, d., musacchio, e.a., castro, j.c., maranhão, m.s.a.s., suárez, j.m. and rodrigues, r., (2001). grupo bauru: uma unidade continental do cretáceo no brasil – concepções baseadas em dados micropaleontológicos, isotópicos e estratigráficos. revue paléobiologie. 20 (1). 245-304. estes, r. and price, l.i., (1973). iguanid lizard from the upper cretaceous of brazil. science, 180. 748-751. fernandes, l.a. and coimbra, a.m., (1996). a bacia bauru (cretáceo superior, brasil). anais da academia brasileira de ciências. 68 (2). 195-205. garrido, a.e., ferreira, a.m. and garcia, a.j.v., (1992). estratigrafia e sedimentologia do grupo bauru em peirópolis município de uberaba, minas gerais. revista da escola de minas. 45. 112-114. gayet, m. and brito, p. m., (1989). ichtyofaune nouvelle du crétacé supérieur du groupe bauru (états de são paulo et minas gerais, brésil). geobios. 22. 841-847. kellner, a.w.a., (1996). remarks on brazilian dinosaurs. memoirs of the queensland museum. 39 (3). 611-626. kellner, a.w.a., campos, d.a. and trotta, m.n., (2005). description of a titanosaurid caudal series from the bauru group, late cretaceous of brazil. arquivos do museu nacional. 63 (3). 529-564. frança, m.a.g. and langer, m.c., (2005). a new freshwater turtle (pleurodira, podocnemididae) from the upper cretaceous (maastrichtian) of minas geris, brazil. geodiversitas, 27 (3). 391-411. lamanna, m.c., luna, m.a., casal, g.a., martinez, r.d., ibiricu, l. and sciutto, j.c., (2003). new crocodyliform and dinosaur discoveries from the upper cretaceous (campanian¿maastrichtian) upper member of the bajo barreal formation, southern chubut province, argentina. journal of vertebrate paleontology. 23 (3, suppl.). 70a. leanza, h.a., apesteguía, s., novas, f. and de la fuente, m., (2004). cretaceous terrestrial beds from the neuquén basin (argentina) and their tetrapod assemblages. cretaceous research. 25. 61-87. magalhâes-ribeiro, c.m.m., (2002). ovo e fragmentos de cascas de ovos de dinossauros, provenientes da região de peirópolis, uberaba, 14 carlos roberto a. candeiro julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:41 p p p composite 133 lpi at 45 degrees minas gerais. arquivos do museu nacional. 60 (3). 223-228. martinelli, a.g. and forasiepi, a.m., (2004). late cretaceous vertebrates from bajo de santa rosa (allen formation), río negro province, argentina, with the description of a new sauropod dinosaur (titanosauridae). revista del museo argentino ciencias naturales “bernardino rivadavia”. 6 (2). 257-305. novas, f.,e., carvalho, i.s., ribeiro, l.c.b. and méndez, a.h., (2008). first abelisaurid bone remains from the maastrichtian marília formation, bauru basin, brazil. cretaceous research. 29. 625-635. powell, j.e., (1987). morfologia del esqueleto axial de los dinosaurios titanosaurios (saurischia, sauropoda) del estado de minas gerais, brasil. atas, 10º congresso brasileiro de paleontologia, sbp/sociedade brasileira de paleontologia, rio de janeiro: 155-171. powell, j.e., (2003). revision of south american titanosaurid dinosaurs: paleobiological, paleobiogepgraphical and phylogenetic aspects. records of the queen victoria museum. 111. 1-173. price, l.i., (1951). ovo de dinossauro na formação bauru, do estado de minas gerais. notas preliminares da divisão de geologia de mineralogia. 53. 1-7. price, l.i., (1955). novos crocodilideos dos arenitos da série bauru, cretáceo do estado de minas gerais. anais da academia brasileira de ciências. 27. 487-498. salgado, l. and carvalho, i.s., (2008). uberabatitan ribeiroi, a new titanosaur from the marília formation (bauru group, upper cretaceous), minas gerais, brazil. palaeontology. 51. 881-901. santucci, r.m., (2002). revisão dos titanosauridae (saurischia, sauropoda) do cretáceo superior continental da bacia bauru. rio claro. 179 p. msc. thesis. instituto de geociências e ciências exatas, universidade estadual paulista. santucci, r.m. and bertini, r.j., (2006). a large sauropod titanosaur from peirópolis, bauru group, brazil. neues jahrbuch für geologie und paläontologie. monatshefte. 6. 344-360. suguio, k., (1973). calcretes of the bauru group (cretaceous), brazil: petrology and geological significance. são paulo. 236 p. dissertation (livre docência). instituto de geociências, universidade de são paulo. suguio, k., (1980). fatores paleoambientais e paleoclimáticos e subdivisão estratigráfica do grupo bauru. resumos expandidos, 1ª mesa redonda: a formação bauru no estado de são paulo e regiões adjacentes, sbg/sociedade brasileira de geologia, são paulo: 15-26. suguio, k., sviero, d.p. and felitti filho, w., (1979). conglomerados polimítricos diamantíferos de idade cretácica de romaria (mg): um exemplo de sedimentação de leques aluviais. atas, 2º simpósio regional de geologia, sbg/sociedade brasileira de geologia, rio claro: 217-229. 15 vertebrates of the marília formation (late maastrichtian) from the peirópolis paleontological site: toward a better understanding julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:41 p p p composite 133 lpi at 45 degrees the study investigated the feasibility of applying the depth-invariant index algorithm to delineate the shallow underwater geomorphological features using quickbird imagery. although the technical procedures of the depthinvariant index are already demonstrated in the professional literature, the algorithm needs to be applied and tested in several coastal regions with different seawater conditions and properties. some of the geomorphological features in the coastal zone of the gaza strip are seasonal features such as underwater sand bars. however, since these features are non-permanent and dynamic, it would be useful to utilize satellite data to detect changes in both locations and depth, and also to monitor this feature in shallow coastal water. the importance of identifying the underwater sand bars in the coastal zone can be significant in the case of rip currents, and swimming conditions. additionally, using satellite imagery for detecting dynamic underwater coastal morphologies may enable local authorities to adopt such an algorithm with a combination of satellite imagery for fast, efficient and extensive coverage mapping. while information about seafloor features can be obtained using active satellite data such as radar or lidar imagery, active satellite data are not available and require advanced processing and logistical resources, which are difficult to obtain in the case of the gaza strip. the study instead used passive quickbird satellite imagery to map and delineate the seabed morphologies through the calculating the depth-invariant index for the spectral bands. by using the depth-invariant index, it was possible to distinguish and delineate the major underwater sand bars within the entire coastal zone of the gaza strip. also, the spectral and spatial resolution of quickbird imagery exhibited a substantial capability in mapping and delineating the underwater seabed morphologies. el estudio investigó la factibilidad de aplicar el algoritmo índice de profundidad invariante para delinear las características geomorfológicas del suelo bajo el agua a través de las imágenes quickbird. a pesar de que los procedimientos técnicos del índice de profundidad invariante se han demostrado ampliamente en la literatura profesional, el algoritmo debe ser aplicado y verificado en varias regiones costeras con diferentes condiciones y propiedades del agua marina. algunas de las condiciones geomorfológicas en la zona costera de la franja de gaza dependen de las estaciones, como las barreras de arena bajo el agua. sin embargo, como estas condiciones son dinámicas, sería útil usar la información satelital para detectar cambios en las locaciones y en la profundidad, y para vigilar estas condiciones en el agua costera poco profunda. la importancia de identificar las barreras de arena en la zona costera puede ser determinante en el caso de corrientes violentas o para las condiciones de inmersión. adicionalmente, el uso de imágenes satelitales para determinar las dinámicas morfológicas costeras bajo el agua podría permitir a las autoridades locales adoptar un algoritmo que combinado con dichas imágenes presente un mapeo más rápido, más eficiente y de mayor cobertura. si bien la información sobre las características del suelo oceánico pueden obtenerse con información satelital activa como las imágenes de radar o lidar, la información no está disponible, además de requerir un procesamiento avanzado y recursos logísticos, que son difíciles de obtener en el caso de la franja de gaza. este estudio, por su parte, utiliza imágenes satelitales pasivas de quickbird para mapear y delinear las morfologías del lecho del mar a través de calcular el índice de profundidad invariante para las bandas espectrales. al utilizar el índice de profundidad invariante fue posible determinar y delinear las mayores barreras de arena en toda la costa de la franja de gaza. la resolución espacial y espectral de las imágenes de quickbird mostraron una capacidad sustancial en el mapeo y delineamiento de las morfologías del lecho marino abstract resumen keywords: remote sensing, seabed morphology, depth-invariant index, gaza strip, palestine. palabras clave: teledetección, morfología del suelo marino, índice de profundidad no variable, franja de gaza, palestina. mapping of underwater seabed morphology of the gaza strip coastal zone using remote sensing technique record manuscript received: 19/04/2015 accepted for publication: 26/04/2016 how to cite item abualhin, k. (2016). mapping of underwater seabed morphology of the gaza strip coastal zone using remote sensing technique. earth sciences research journal, 20(2), c1-c7. doi:http://dx.doi.org/10.15446/esrj.v20n2.50256 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n2.50256 khaldoun abualhin al-azhar university-gaza, gaza, gaza strip, p.o. box 1277, palestine, khaldon@alazhar.edu.ps earth sciences research journal earth sci. res. j. vol. 20, no. 2 ( june, 2016 ) : c1 c7, mapeo de la morfología del suelo oceánico en la zona costera de la franja de gaza a través de la técnica de detección remota h y d r o g e o l o g y c2 khaldoun abualhin introduction several techniques and algorithms have been developed to use satellite imagery for environmental monitoring and coastal habitats mapping. the mapping of underwater seabed morphologies includes several features such as massive valleys, canyons, rocky outcrops, geomorphological features, and sand bars [goudie, 2004; pavlopoulos et al., 2009]. seabed morphologies are an important element of coastal protection and planning elements and also in mapping marine habitats and their management. accordingly, it is necessary for the researcher to get detailed information about seabed morphology and habitats in the nearshore coastal zone. recently, modern methods of marine surveying such as echo sounder and active satellite sensors like lidar data are used for scanning the seabed. the advantages of these modern techniques of mapping are on the accuracy and also a full range of depth coverage. on the other hand, these methods of seabed mapping are expensive, time-consuming, and difficult to process. in contrast, passive remotely sensed data have been shown to offer a reasonable alternative in providing reliable and low-cost information about coastal habitats and morphologies [mumby and edwards, 2002; mumby et al., 2004]. different passive satellite imagery has been used in studying coral reef and classification, benthic habitats classification, sand bottoms delineation and in other studies [ahmad and neil, 1994; andréfou t et al., 2003; edwards et al., 1999; green et al., 2000; hochberg and atkinson, 2003; michalek et al., 1993; mishra et al., 2006; mumby et al., 1998; mumby and edwards, 2002; mumby and harborne, 1999; mumby et al., 2004; ong et al., 1998; pasqualini et al., 2005; stumpf et al., 2003; vanderstraete et al., 2003]. mishra et al. (2006) utilized quickbird multispectral data for identifying and classifying tropical-marine benthic habitats. in 1998, ong and hick combined spectral data and digital bathymetry to auto-classify larger benthic habitats. pasqualini et al., (2005) used spot5 satellite imagery to map seagrasses. mumby et al., (1997a,b, 1998) compared the descriptive resolutions of various remote sensing methods for mapping coral reef habitats of the turks and caicos islands. other researchers simulated ikonos data for different seabeds and coral reef habitat mapping [andréfouet et al., 2003; hochberg and atkinson, 2003]. mumby et al., (2004) provided a general review of what can and cannot be mapped using passive remote sensing data and summarized the capacity of various methods to predict and derive bathymetry from different types of optical imagery [jupp, 1988; stumpf et al., 2003]. in terms of satellite spectral and spatial resolutions, multispectral imagery with coarse and medium spatial resolution, landsat tm, landsat etm+, aster, and ikonos satellite data are considered useful tools for monitoring the spatial dynamics of habitats [mumby and edwards, 2002; mumby and harborne, 1999]. however, taking into consideration the similarity in spectral resolution and the width of the spectral band between quickbird and ikonos sensors, the ability of quickbird data to discriminate seabed morphology in the shallow coastal region is comparable to ikonos data. in this study, quickbird satellite imagery was used to delineate the seabed morphology of sand bars mainly in the nearshore of the gaza strip coastal zone. in addition to oceanographic factors, the coastal region of the gaza strip suffers from intensive anthropogenic pollution and other manmade interventions along the entire coastal zone. moreover, as a result of israel’s blockade of the gaza strip and the subsequent limitation on resources, it is not always possible to use the most recent technologies and tools for mapping and monitoring the coastal zone, including the mapping of underwater morphologies. as it is important to get up-todate information about seabed morphologies for monitoring the changing conditions of the coastal zone, a quickbird satellite image was used to extrapolate the underwater morphology of sand bars along the coastal zone of the gaza strip. underwater bars can defined as an aggregation of ridges of sediments whose formation, morphology and behavior were determined by interactions between waves, currents, tides, local slope and grain size. bowman and goldsmith (1983) delineated a wide range of nearshore sand bars and shoal patterns along the southeastern mediterranean coastline. their study identified three major types of sand bars: non-rhythmic parallel-meandering bar; inner single-crescentic bars; and, double-crescentic bars [bowman and goldsmith, 1983]. study area the gaza strip is a part of historical palestine and is located on the south eastern mediterranean coast and bordered by sinai from the south and the negev desert from the east (figure 1). the current area of the gaza strip is 360 km² with an estimated human population density of 4167 per km². the geographic coordinates of the lower left and upper right corners of the gaza strip are 31.31 n, 34.2 e; and 31.53 n, 34.56 e. the shoreline of the gaza strip is a smooth line that extends 42 km. the coastal zone includes sand dunes in the south and north, coastal cliffs, and the wadi gaza mouth area. wadi gaza is the main seasonal river in the gaza strip and influence the biodiversity in the gaza strip than it does today. figure 1. map of the gaza strip, palestine methodology seabed feature and underwater patterns can be mapped from passive remote sensing imagery if the influence of the water column was removed. the implementation of the physical relation between spectral reflectance and light attenuation is the base of extracting information from multispectral satellite imagery over water bodies. the incident radiation on the water column is absorbed and scattered or reflected unequally. the attenuation influence can be estimated over the entire multispectral image by getting water depth at several pixels in the image [lyzenga, 1981]. however, to neutralize the influence of seawater on the reflected signal, an attenuation coefficient must be calculated at each pixel in the image [green et al., 2000]. lyzenga (1978, 1981) proposed a simple image-based technique to combine the information in various spectral wavelengths to produce the depth-invariant index. in this method, the ratio of water attenuation coefficients are required and optical properties considered uniform. depth-invariant index the concept of extracting bottom-type information in shallow water depends on the fact that the reflected seabed-radiance is a linear function of bottom reflectance and an exponential function of water depth [lyzenga, 1978]. the technique to calculate depth-invariant index involves three main steps: (1) generate a linear relationship between depth and radiance; (2) calculate the ratio of attenuation coefficients of band pairs; and, (3) calculate the depth-invariant index of bottom type (figure 2). c3mapping of under water seabed mor phology of the gaza strip coastal zone using remote sensing technique figure 2. flow chart for generating depth-invariant index since the relationship between the depth and reflected radiance over water bodies is an exponential relation, this relationship can be transformed using a natural logarithm (equation 1) to become a linear relationship. for a given bottom type, the transformed radiance in bands (i) and (j) is linearly (figure 3). the transformed radiance will attenuate linearly with increasing water depth. xi = ln(li) (1) where has transformed reflectance in the band (i). li has atmospherically corrected radiance in the band (i). ln is the natural logarithm. accordingly, the transformed radiance in the band (i) is plotted versus the transformed radiance in the band (j) at variable depth and constant bottom type (figure 3. the result falls along a straight line. consequently, different bottom types should provide different parallel lines [lyzenga, 1981]. the slope of the lines represents the ratio of attenuation coefficients for each spectral band (figure 4 and figure 5). figure 3. linear exponential relationship between reflected radiance and depth using natural logarithms. the gradient of the line represents the ratio of attenuation coefficient. (band i has a shorter wavelength, and, therefore, attenuates less rapidly, than band j) figure 4. scatter plot of the natural logarithm (log) of quickbird radiance. quickbird band 4 was plotted versus band 3, 2 and 1. (the dn values were obtained from 257 pixels over sandy bottom-type at varying depths in the coastal zone of the gaza strip.) figure 5. scatter plot of the natural logarithm of quickbird radiance in the band (3) versus radiance in the band (4) (the quickbird image for the gaza coastal zone was acquired on 21 january 2009) the ratios of the attenuation coefficients of the spectral bands were derived from the scattering bi-plot. the slope of the scatter plot trend line represents the ratio of attenuation coefficients (figure 5). the slope of the trend line will depend on the variable chosen to be the dependant variable. accordingly, using equation 2, the ratio of attenuation coefficients (ki/kj) will not depend on the order the variable. where, (2) (3) where σii and σjj are the variances for i and j measurements respectively, and σij is the covariance between i and j bands. once the ratios of attenuation coefficients are calculated, the depthinvariant indexes can be calculated using equation 3 for the entire image. each pair of spectral bands produces a single depth-invariant index for the bottom type. implementation of depth-invariant index algorithm on quickbird image a quickbird satellite image acquired on 21 january 2009 at 08:35 gmt was used to calculate the depth-invariant indices and then to map the seabed morphology along the coastal zone of the gaza strip. the quickbird image was radiometrically and geometrically corrected. the image was registered to the topographic map with rms less than 0.4 pixels (table 1). the atmospheric correction was then performed using the envi flaash module. the homogeneous regions in the satellite image with a uniform bottom reflectance were selected to be used for attenuation coefficient calculation. the ratio of attenuation coefficients was calculated based on the equation 2. table 2 shows the calculated attenuation coefficients ratios (ki/kj) for each pair of spectral bands. table 1. satellite data characteristics c4 khaldoun abualhin table 2. the calculated variance (σii, σjj), covariance (σij) and the ratio of attenuation coefficients (ki/kj) for each pair of logarithmic transformed quickbird bands results and discussion several depth-invariant indices were calculated from quickbird imagery for the coastal zone of the gaza strip, portraying different abilities to recognize seabed morphology. for example the calculated index from band 1 and band 3 demonstrated the ability to map features in shallow depth up to 10 m. the index derived from band 2 and band 3 showed the feasibility of mapping features at shallow water depth. the index of band 2 and band 4 as well as an index of the band 3 and band 4 obtained reasonable results in mapping the very shallow coastal bars at depth varying from 0.0 -1.5 m. these bars welded to the coastline. accordingly, different types of seabed bars were clearly observed and delineated at different locations in the entire coastal zone and nearshore zone of the gaza strip. types of bars are often distinguished based on their shape and orientation about the shoreline. they may be linear, sinuous or crescentic, or they may consist of alternating transverse bars, which are welded to the shoreline and are separated by channels occupied by rip currents, as seen in figure 6 [goudie, 2004]. figure 6. idealized cross-shore barred profile (a) and some examples of bar types [goudie, 2004]. the mapped sand bars in the coastal zone of the gaza strip were similar to those observed by [almagor and hall, 1984; bowman and goldsmith, 1983; goldsmith et al., 1982], in the south-eastern mediterranean coast. some of these sand bars are seasonal occurrences while others occur throughout the year. the delineated underwater sand bars in the gaza strip coastal zone were as follows: single-crescentic bars; anvil-shaped transverse bars, oblique transverse bars, initial crescentic bars, double crescentic bars and triple crescentic bars. figure 7 shows the different types of underwater sand bars delineated in the nearshore of the gaza strip. as mentioned above, depth-invariant indices demonstrated disparities in their sensitivity to mapping seabed features in the shallow coastal region. for instance, double and triple crescentic bar systems can be delineated using the ratio index of band 1 and band 2. the ratio of band 2 and band 3 achieved a respectable result in mapping different sand bar systems at different locations. the indices of band 2 and band 4 and of band 3 and band 4 showed less capability in mapping sand bar features. in addition, the calculated depth-invariant indices provided a dimensionless bottom relief in the nearshore zone of the gaza strip. several bottom relief profiles were generated perpendicular to the coastline (figure 7 and figure 8). figure 8 shows a variation in the seabed relief on both sides of the fishing port. this variation may be the result of sediments being transported from the nile delta to the southeastern mediterranean nearshore shelf by northward longshore currents [rohrlicht and goldsmith, 1984; zviely et al., 2007]. since the sediments are trapped and accumulated at the southern side of the fishing port, the sand supply to the northern side of the fishing port has decreased. two sandbars appeared at the southern side of the fishing port and located at a distance of 100 m up to 400 m from the coastline. figure 9 illustrates the bottom relief of the general morphology and distribution of the underwater sand bars. a. transverse bars (oblique) c. single crescentic bar d. double crescentic bars b. double crescentic bars system figure 7. example of sand bars observed along the coastal zone of the gaza strip. (the bars were extracted from quickbird imagery using depth-invariant index technique [lyzenga, 1981]) c5mapping of under water seabed mor phology of the gaza strip coastal zone using remote sensing technique d. double crescentic bars b. double crescentic bars system figure 8. the bottom relief on the southern side of the seaport and up to 400m seaward. (the figure reveals the bars crest and trough about the distance of 500 m from the shoreline. the profile was obtained from the calculated depth-invariant index for quickbird imagery.) figure 9. bottom relief profiles. (the profiles were extracted from the calculated depth-invariant index for quickbird band 2 and band3. the horizontal component of the patterns was relatively accurate where the vertical component is pseudo-depth.) c6 khaldoun abualhin utilizing the envi topographic modelling tool, the slope, the aspect and the shaded relief image were derived from the seabed indices (figure 10). index from band 1 and band b3 demonstrated a slope gradient of the gaza strip coastal zone ranging from 0.57° to 0.41° at a distance of 50 m to 100 m from the coastline. figure 10. shaded relief surface generated from the depth-invariant index of quickbird band 2 and band 3. (the sand bars appear in three sets: the first one attached to the coastline, the second located within a very shallow water from 2-4 m, and the third located at 200-400 m distance from the coast at a depth of 7-10 m.) conclusion gathering information about seabed morphology and bottom profiles requires advanced instruments and sometimes it is a time-consuming process. data collection may also be complicated by challenging field and logistical circumstances beyond the control of researchers, as has been the case in the gaza strip. nevertheless, remotely sensed imagery has provided an alternative method for generating bathymetric maps with reasonable accuracy. in particular, the utilization of the depth-invariant index to map and delineate underwater seabed morphology has proven effective. the study confirmed the possibility of deriving underwater seabed morphology in shallow water. the calculated indices provided interesting results on mapping seabed morphology such as sand bars and bottom profile. the method enables mapping and delineates the major underwater sand bars along the entire coastal zone of the gaza strip. importantly, various types of seabed morphologies were clearly observed and defined at different regions and depths in the gaza strip coastal zone such as crescentic bars, single-crescentic bars, double-crescentic bars, triple crescentic bars and transverse bars. some bar systems were located onshore and attached to the shoreline at a very shallow depth of up to 2 meters, as was the case for transverse bars, which then moved offshore to form crescent bars. consequently double crescent bars are formed at a deeper depth and distance from 150 up to 400 meters offshore. finally, as the nearshore region of the gaza strip is homogeneous and consists mainly of sandy sediments, the spectral and spatial resolution of quickbird imagery were useful in distinguishing natural variations and contrasts due to optical physical properties of coastal habitats. quickbird imagery exhibited a substantial capacity to delineate underwater seabed morphology by using depth-invariant index technique. the achieved findings were quite satisfying and consistent with published information on the region. acknowledgment special gratitude is extended to the daad (german exchange service) for funding support and the purchase the quickbird image. references ahmad, w., & neil, d. (1994). an evaluation of landsat thematic mapper (tm) digital data for discriminating coral reef zonation: heron reef (gbr). international journal of remote sensing, 15, 2583-2597. almagor, g., gill, d., & perath, i. (2000). marine sand resources offshore israel. marine georesources and geotechnology, 18, 1-42. almagor, g., & hall, j. (1984). morphology and bathymetry of the mediterranean continental margin of israel. geological survey of israel bulletin, 77. andréfouet, s., kramer, p., torres-pulliza, d., joyce, k., hochberg, e., garzapérez, r., mumby, p., riegl, b., yamano, h., & white, w. (2003). multi-site evaluation of ikonos data for classification of tropical coral reef environments. remote sensing of environment, 88, 128-143. bowman, d., & goldsmith, v. (1983). bar morphology of dissipative beaches: an empirical model. marine geology, 51, 15-33. edwards, a., mumby, p., green, e., & clark, c. (1999). applications of satellite and airborne image data to coastal management: seventh computerbased learning module (bilko for windows). paris: unesco. goldsmith, v., bowman, d., & kiley, k. (1982). sequential stage development of crescentic bars: hahoterim beach, southeastern mediterranean. journal of sedimentary petrology, 52, 233-249. goudie, a. (2004). encyclopedia of geomorphology. new york: routledge ltd. green, e., mumby, p., clark, c., & edwards, t. (2000). remote sensing handbook for tropical coastal management. paris: unesco. green, e., mumby, p., edwards, a., & clark, c. (1996). a review of remote sensing for the assessment and management of tropical coastal resources. coastal management, 24, 1-40. hochberg, e., & atkinson, m. (2003). capabilities of remote sensors to classify coral, algae, and sand as pure and mixed spectra. remote sensing of environment, 85, 174-189. huston, m. (1994). biological diversity: the coexistence of species. cambridge: cambridge university press. jupp, d. (1988). background and extension to depth of penetration(dop) mapping in shallow coastal waters. in, proceedings of the symposium on remote sensing of the coastal zone. gold coast, queensland. lyzenga, d. (1978). passive remote sensing techniques for mapping water depth and bottom features. appl. opt, 17, 379-383. lyzenga, d. (1981). remote sensing of bottom reflectance and water attenuation parameters in shallow water using aircraft and landsat data. international journal of remote sensing, 2, 71-82. michalek, j., wagner, t., luczkovich, j., & stoffle, r. (1993). multispectral change vector analysis for monitoring coastal marine environments. photogrammetric engineering and remote sensing, 59, 381-384. mishra, d., narumalani, s., rundquist, d., & lawson, m. (2006). benthic habitat mapping in tropical marine environments using quickbird multispectral data. photogrammetric engineering and remote sensing, 72, 1037. mumby, p. j., green, e. p., edwards, a. j., & clark, c. d. (1997). coral reef habitat mapping: how much detail can remote sensing provide?. marine biology, 130(2), 193-202. mumby, p. j., green, e. p., clark, c. d., & edwards, a. j. (1997). reefal habitat assessment using (casi) airborne remote sensing. in proc 8th int coral reef symp (vol. 2, pp. 1499-1502). mumby, p., clark, c., green, e., & edwards, a. (1998). benefits of water column correction and contextual editing for mapping coral reefs. international journal of remote sensing, 19, 203-210. mumby, p., & edwards, a. (2002). mapping marine environments with ikonos imagery: enhanced spatial resolution can deliver greater thematic accuracy. remote sensing of environment, 82, 248-257. mumby, p., & harborne, a. (1999). development of a systematic classification scheme of marine habitats to facilitate regional management and mapping of caribbean coral reefs. biological conservation, 88, 155-163. mumby, p., skirving, w., strong, a., hardy, j., ledrew, e., hochberg, e., stumpf, r., & david, l. (2004). remote sensing of coral reefs and their physical environment. marine pollution bulletin, 48, 219-228. c7mapping of under water seabed mor phology of the gaza strip coastal zone using remote sensing technique ong, c., hick, p., burt, j., & wyllie, a. (1998). marine habitat mapping using data from the geoscan airborne multi-spectral scanner. geocarto international, 13, 27 34. pasqualini, v., pergent-martini, c., pergent, g., agreil, m., skoufas, g., sourbes, l., & tsirika, a. (2005). use of spot 5 for mapping seagrasses: an application to posidonia oceanica. remote sensing of environment, 94, 39-45. pavlopoulos, k., evelpidou, n., & vassilopoulos, a. (2009). mapping geomorphological environments. berlin: springer verlag. rohrlicht, v., & goldsmith, v. (1984). sediment transport along the southeast mediterranean: a geological perspective. geo-marine letters, 4, 99-103. stumpf, r., holderied, k., & sinclair, m. (2003). determination of water depth with high-resolution satellite imagery over variable bottom types. limnology and oceanography, 48, 547-556. vanderstraete, t., goossens, r., & ghabour, t. (2003). can aster-data be used for bathymetric mapping of coral reefs in the red sea using digital photogrammetry? in, proceedings of the second workshop of the earsel special interest group on remote sensing for developing countries (pp. 212-218). bonn, germany. zviely, d., kit, e., & klein, m. (2007). longshore sand transport estimates along the mediterranean coast of israel in the holocene. marine geology, 238, 61-73. earth sciences research journal eart sci. res. j. vol. 18, no. 2 (december, 2014): 173 179 spatial distribution of sediment pollution in the khajeh kory river using kriging and gis maryam zare khosh eghbal department of mathematics and science, astara branch, islamic azad university, astara, iran e-mail: m.zare@iau-astara.ac.ir gis the aim of this research was to study of sediment pollution conditions in the river bed of khajeh kory, which passes from the southern border of the caspian sea in astara city. to determine the pollution of river bed sediment along the route, from the river source in the heights to estuary in the caspian sea, sampling was conducted at 10 stations. afterward, each sample was divided into two groups based on size (>63 µm and <63 µm). the concentration of heavy metals cu, cd, co, zn, pb, ni, fe, and mn in each group was measured using icp-oes. in the next stage, the enrichment of samples was determined by normalization with aluminum. the data were then interpolated using the kriging method and various models (including spherical, circular, exponential, gaussian) were fitted to the data, and the best method was selected using the cross validation method. using the obtained enrichment outcomes, an information layer was produced for each element in the gis environment utilizing the kriging method. the layers were aggregated and the compiled layer was classified into 4 layers again. using this method, the length of the river from its source to the sea estuary was characterized based on the contamination of heavy metals, and contaminated and unpolluted areas in river sediments could be observed. the source of this pollution, either natural or manmade, was revealed through this method. el objeto de esta investigación fue estudiar las condiciones de contaminación sedimentaria en el lecho del río khajeh kory, que se extiende desde el borde sur del mar caspio en la ciudad de astara. para determinar la contaminación del lecho sedimentario, se recolectaron muestras en 10 estaciones establecidas entre la fuente del río en sus partes altas hasta los estuarios en el mar caspio. después, cada muestra fue dividida en dos grupos de acuerdo con el tamaño (>63 µm and <63 µm). la concentración de metales pesados cu, cd, co, zn, pb, ni, fe y mn en cada grupo se midió con la técnica icp-oes. en la siguiente etapa, se determinó el enriquecimiento de las muestras por la normalización con aluminio. los datos se interpolaron luego con el método kriging y varios modelos (incluidos el esférico, el circular, el exponencial, el gaussiano) se ajustaron a la información y se seleccionó el mejor método a través de la validación cruzada. con los resultados obtenidos del enriquecimiento se produjo una capa de información para cada elemento en el ambiente gis a través del método kriging. las capas fueron agregadas y la compilación de estas se clasificó nuevamente en otras cuatro capas. de esta forma se caracterizó el río a lo largo, desde su nacimiento hasta el estuario marino, con base en la contaminación de metales pesados y se pudo distinguir entre las zonas contaminadas y las áreas limpias en sedimentos. las fuentes de esta contaminación, sean naturales o humanas, también se identificaron con este método. abstract resumen key words: sediment pollution, khajeh kory river, kriging, gis palabras clave: sedimentos contaminados; río khajeh kory; kriging; gis record manuscript received: 19/07/2014 accepted for publication: 16/09/2014 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v18n2.44491 http://dx.doi.org/10.15446/esrj.v18n2.44491 174 maryam zare khosh eghbal, shahriar sobh zahedi introduction: rivers are an essential aquatic ecosystems, and rivers are polluted by various contaminants (bednarova et al., 2013). one class of most important pollutants is heavy metals. heavy metals are ecologically impactful because they experience high resistance in the environment, accumulate in natural resources, and ultimately enter the food cycle (varol & şen, 2012). river sediments are the ultimate destination of these metals (maanan et al, 2013). geochemistry-based methods have been considered appropriate in determining the contaminated condition of these deposits (meybeck, horowitz, & grosbois, 2004). the guilan province is especially notable for contamination as various rivers directly or indirectly pour into the caspian sea. within this province a number of urban areas, urban sewage, domestic wastewater, and industrial sewage play important roles in the contamination of rivers (gheshlagh, ziarati, & bidgoli, 2013). domestic wastewaters are poured into special attraction wells within each residential unit in rural areas. however, various agricultural fertilizers and pesticides are common in this area because of rice cultivation and the use of other agricultural lands. these materials enter the river along with surface drainage and cause pollution. additionally, the land may also contain high levels of heavy metals due to special lithology and geology, naturally causing contamination without human interference (salati & moore, 2010). the khajeh kory river is located in the western part of guilan and passes along forest, agricultural, and residential areas upstream and crosses astara downstream to pour into the caspian sea. the aim of this study is to investigate contamination conditions in this river from different sections from its source (areas without agricultural lands and human sewage) to river estuaries at the sea (areas that receive urban and rural sewage, as well as agricultural land surface drainage). one problem in measuring the contamination of an area or river involves the limited sampling possible within an area or along the river length; to overcome this problem, interpolation methods should be used. interpolation analysis is in line with the spatial characteristics of methods that are common in spatial statistics studies (xie et al., 2011) and is based on the principle that points that are closer in space are also likely to be more similar to each other than to points that are further apart. there are various spatial interpolation algorithms that are based on geostatistical and geometric methods. recently, the use of gis software has been utilized in the spatial modeling arena and a variety of studies have been conducted in interpolation analyses for zoning contamination sites (mcgrath, zhang, &carton, 2004). validation of these methods requires precision and various factors are recommended. in the present study, geographic information system (gis) and standard kriging methods have been used to prepare interpolation maps to examine heavy metal contamination in the khajeh kory river. materials and methods: in the studied area, the sediments of the khajeh kory river covers a watershed area of 10233 hectare, located in the guilan province, which ranges between 38° 20΄ 45˝ and 38° 26΄ 45˝ latitude and 48° 42΄ 22˝ and 48° 52΄ 28˝ longitude. this watershed is covered with forest (63.12%), agricultural and garden lands(12.71%) and 2.47% rural and urban areas (figure 1). geologically, the basin is covered with structures from the mesozoic and cenozoic ages. the main part of basin (44.5%) is composed of pevet units from the paleocene age with alternation of tuff ,tuffereous sandstone with pyroxin.   caspian  sea figure 1: land use map in the khajeh kory watershed area. figure 2: geological map of the khajeh kory watershed area. this special lithology can cause an increase in the measurements of elements in the area and is given special attention. qcs units encompassing the quaternary age cover 27.33% of the total area and constitute the second most prevalent the type of sediment in the caspian sea coastal plain (figure 2). figure 2 is a geological map of the area and was prepared using arc gis 10 software and georeferencing a 1:100000 map of astara geology. sampling stations were selected to cover a range of various sites depending on the land use and type of lithology in the area. the pattern of sampling and the condition of each area are represented in figure 1. after sampling stations were determined (singh, mohan, singh, & malik, 2005), trapped sediments were sampled from each station (horowitz, 2009). sediments were dried at ambient air temperature in laboratory conditions. dried sediments were powdered and divided by size into two groups (>63 µm (muddy part) and <63 µm), and each group was subjected to icp-oes (inductively coupled plasma optical emission spectrometry) to determine the concentration of elements, including cu, cd, co, zn, pb, ni, fe, and mn. icp-oes uses plasma to prepare qualitative and quantitative analyses of elements. in our analysis, the argon gas was ionized using a magnetic field (radio frequency of 27-40 mh) and the temperature was ~10,000 k. the sample was infused into the plasma using a nebulizer, and the high temperature sample was converted to ions and diffused through the sediments. the rate of diffusion was measured by the apparatus. after determining the elemental concentrations, a method must be applied for quantitative measurement of elemental contamination. one such method uses aluminum as a standard in normalizing sediments samples (choi, kim, hong, & kim, 2013). aluminum is present in high amounts as a result of erosion in stones and its presence is sediment is not affected by human activities. conversely, aluminum has variable solubility, can chemically react in soil, and migrates after being deposited in sediment. normalization based on aluminum   175spatial distribution of sediment pollution in the khajeh kory river using kriging and gis content occurs in two stages. first, the concentrations of each element in a sample are divided by the concentration of aluminum, and the sample with the least variation is selected as the reference sample. second, the amount of pollution is obtained by the ratio of a given element to aluminum in the sample and compared to an inner reference sample using the following formula: ef = [x / al] sample /[x / al] ref in the following method, the descriptive statistics of analysis and geostatistics will be used to fit the best data to an enrichment model. finally, an enrichment map is prepared for each element, and after merging the elements, a cumulative map of pollution weight in river foot can be produced. statistical methods: 1. descriptive statistics of data: to investigate the distribution of element enrichment and to summarize the statistical information of contamination, concentration of each element in ppm was evaluated for maximum, minimum, mean, standard deviation and variance using spss software. 2. geostatistical analyses: geostatistics is capable of modeling uncertain temporal and spatial phenomena. basically, geostatistics estimates each unknown value as a random number within a known probable distribution across an understudied space (dao, morrison, kiely, & zhang, 2013). before using the data, the normality of data needs to be determined (kroulík, mimra, kumhála, & prošek, 2006.) and both kolmogorov and smirnov tests are used to test the distribution of data with a reliability of 5% and 1%, respectively. after determining the normality of the data, geostatistics is used to correlate information of a point to the information of that area on a map. the kriging method of geostatistics was used to interpolate, zonate, and forecast the probability of data correlation. 3. analyzing variograms: after checking the normality of the data, kriging measurements were implemented in calculating a variogram. a variogram is a function that measures the variability of spatial data; it is extremely important in spatial data correlation and is the base of geostatistics (kaushik, kansal, kumari, & kaushik, 2009; maanan, et al., 2013). dividing a variogram in two, we can obtain easily understood semivariograms, which is the practical use in geostatistical methods. semivariograms are represented by γ(h) and can be calculated according to the following relation: in this equation, n is the number of sample couples per distance, h; zi (x) is the value of variance in i point; and indicates the value of variance in a point that is h distance from point i. increasing h, the value of semivariogram is increased to a certain distance and then fixed at a threshold level, sill. the distance between samples at which variable values have minimal effect on each other and the value of semivariogram does not significantly differ is called the range of influence. the value of the semivariogram when h=0 is called the nugget effect (figure 3) (liu, wu, & xu, 2006). the nugget effect usually results from errors in data sampling, measurement and, analysis. the less the nugget effect, the less the estimation error will be. variogram modeling is an iterative process. in all geostatistical interpolation methods, including kriging, unknown values are estimated using the following equation: figure 3: schematic image of the variogram and its parameters. γ(h)= 1 2n zi(x)−zi(x+h)[ ] i=1 n ∑ 2 zi(x+h)   sill range of influence nugget effect structured  variance   z*(x)= λiz(xi) i=1 n ∑ in this equation, z(xi) indicates the value of observable z variable at xi, λi indicates the weight of a given ratio to z variable at xi, and n is the number of observations. after drawing an experimental variogram, four models (including circular, spherical, exponential, and gaussian) were fitted for each data set. 4. assessment methods, criterion, and validation models: in the present study, the methods used to control and validate parameters used in the estimation will be assessed as a method of cross-validation. this method requires deleting samples and re-estimating them using the kriging method, followed by using other samplings and fitted models in an experimental variogram (angulo-martínez, lópez-vicente, vicente-serrano, & beguería, 2009). following the subtraction of real and estimated values, the estimations are statistically evaluated (castrignanò, giugliarini, risaliti, & martinelli, 2000). in the present research, validation models and estimations were evaluated by calculating the root mean square (rms) criteria and the standard root mean square (srms) values using the following formulas: in these formula, n is the number of points zi,act indicates the value of i at a known point, zi,est is the estimation of i known point, and s indicates error variation. the best estimation should have a minimal rms, and srms should approximate 1. if srms equals 1, it means that rms is equal to s (angulo-martínez, et al., 2009). generally, the smaller the rms, the more precise the method will be. theoretically, if this criterion equals zero, the method precision is one hundred percent and the estimated value of a quantity is equal to its real value. discussions and results: table 1 details the changes in the elemental concentrations in the sediments from the khajeh kory river. these results demonstrate that the heavy metal concentrations were mainly found in fine-grain deposits (fang, jiang, wang, & xie, 2010; kaushik, et al., 2009). additionally, the highest concentrations of iron were found in station s10; the maximum concentrations of manganese, cobalt, and cadmium were found in s3; the maximum concentration of nickel was found in s9; the maximum concentration of lead was found in s1; and the maximum concentration of zinc was found in s7. in regard to sampling station placements by geological units and land use, station 1 (s1) reveals high values of lead, which result from urban life and petrol use in that area. the abundance of iron in s10 comes from natural, geological sources as there are no contamination factors from agriculture, urban and rural sewage, or industry. conversely, constitutive rocks in this area are mainly composed of pyroxene, which contain iron and magnesium components. three stations, including s9, s8, and s7, have similar compositions. therefore, the noted frequency of zinc in s7 is of non-geological origins and likely originates from the neighboring besat residential suburb, which utilizes gable roofs that are composed of galvanized zinc. this zinc can decompose due to rainfall and enter surface water, resulting in higher concentrations in sediments. abundance of manganese, cobalt, and cadmium in s3 among agricultural lands is likely a result of chemical fertilizers and pesticides used in rice fields, which are common in the area. rms = 1 n (zi,act i=1 n ∑ −zi.est)2 srms = 1 n (zi,act −zi,est) 2 s2i=1 n ∑ = rms s 176 maryam zare khosh eghbal, shahriar sobh zahedi table 1: concentrations of selected elements of sample sediments from sampling stations. f * indicates fine-grain (smaller than 63 micron) and c indicated coarse-grain (greater than 63 μm). the enrichment factor values and use of normality method by al in various stations is described in table 2. enrichment factors are used in this analysis instead of absolute elemental concentrations because of differences in densities between elements. for instance some elements, including iron and manganese, are more common in the crust and are measured in percent, whereas other elements are less abundant and are measured in ppm. using enrichment as an index without dimension is effective and efficient because it allows for different analyses to be conducted without needing to consider the differences between the amounts of individual elements. these results indicate that the pollution is mostly in fine-grain sediment (rodríguez, ruiz, alonsoazcárate, & rincón, 2009; seshan, natesan, & deepthi, 2010). additionally, s3 stations contained the highest amounts of mn, co, and cu. the level of zn enrichment was found to be 30 times higher in this area. after the river crosses astara city, where it begins to pour into the caspian sea, there is an 18.73-enrichment of zinc, which demonstrates that there is a relationship with residential and agricultural areas (bai et al., 2011). table 2: the results of element enrichment factors in sediment samples. sample x y fe mn cd co cu ni pb zn s1 f 313741 4255467 0.61 0.76 0.7 0.67 0.97 0.72 3.79 15.92 s1 c 313741 4255467 0.89 0.9 1.03 0.95 0.85 1.03 0.89 1.01 s2 f 313277 4255522 0.56 1.28 0.69 0.56 0.9 0.77 2.08 3.03 s2 c 313277 4255522 1 1 1 1 1 1 1 1 s3 f 309388 4256292 0.7 2.22 1.03 0.8 1.15 0.76 1.13 0.85 s3 c 309388 4256292 0.9 0.82 0.97 0.82 1.22 0.82 0.85 0.93 s4 f 308955 4255522 0.62 0.58 0.72 0.59 0.9 0.72 1.09 4.68 s4 c 308955 4255522 0.87 0.81 0.99 0.91 0.87 1.03 1.23 1.06 s5 f 308627 4255966 0.6 0.53 0.65 0.6 0.91 0.71 1.01 12.91 s5 c 308627 4255966 1.22 0.99 1.03 1.12 1.04 1.17 1.93 1.15 s6 f 307798 4253708 0.65 0.85 0.67 0.7 0.91 0.82 1.25 5.25 sample fe 7.14 cd co cu ni pb zn s1 f 4.86 3.76 0.27 25 63 50 117 2116 s1 c 4.64 5.37 0.26 23 36 47 18 88 s2 f 4.48 0.79 0.27 21 59 54 65 407 s2 c 5.16 0.63 0.25 24 42 45 20 86 s3 f 5.59 7.14 0.4 30 75 53 35 114 s3 c 5.14 3.76 0.27 22 57 41 19 89 s4 f 5.03 5.37 0.28 22 59 51 34 630 s4 c 4.56 0.79 0.25 22 37 47 25 92 s5 f 5.37 0.63 0.28 25 66 55 35 1920 s5 c 6.35 7.14 0.26 27 44 53 39 100 s6 f 5.36 3.76 0.27 27 61 59 40 723 s6 c 6.03 5.37 0.26 27 51 58 29 104 s7 f 4.95 0.79 0.27 24 57 53 37 4529 s7 c 5.94 0.63 0.24 26 48 57 32 119 s8 f 4.85 7.14 0.8 27 64 60 51 201 s8 c 6.42 3.76 0.25 26 47 61 21 100 s9 f 3.76 5.37 0.28 17 60 44 100 2696 s9 c 5.75 0.79 0.23 23 40 56 19 96 s10 f 5.93 0.63 0.26 26 69 45 28 345 s10 c 7.14 7.14 0.25 30 47 60 48 113 max 7.14 3.76 0.8 30 75 61 117 4529 min 3.76 5.37 0.23 17 36 41 18 86 average 5.37 0.79 0.3 24.7 54.1 52.45 40.6 733.4 standard deviation 0.79 0.63 0.12 3.11 11.1 6 26.27 1182 variance 0.63 7.14 0.02 9.69 123.15 36.05 689.94 1396988.9 177spatial distribution of sediment pollution in the khajeh kory river using kriging and gis table 3: the results of kolmogorov – smirnov in enrichment rate of river sediments. fe mn cd co cu ni pb zn n 10 10 10 10 10 10 10 10 normal parameters a, b mean 0.61 0.99 0.85 0.63 0.94 0.73 1.74 10.26 std. deviation 0.05 0.51 0.42 0.09 0.08 0.09 1.07 11.06 most extreme differences absolute 0.16 0.21 0.36 0.14 0.26 0.22 0.28 0.28 positive 0.15 0.21 0.36 0.14 0.26 0.13 0.28 0.28 negative -0.16 -0.18 -0.23 -0.10 -0.16 -0.22 -0.18 -0.20 kolmogorov-smirnov z 0.52 0.65 1.125 0.45 0.819 0.684 0.873 0.87 asymp. sig. (2-tailed) 0.95 0.79 0.16 0.99 0.51 0.74 0.43 0.44 s6 c 307798 4253708 0.89 0.76 0.79 0.85 0.92 0.98 1.1 0.92 s7 f 308689 4254667 0.63 1.01 0.7 0.65 0.89 0.77 1.21 34.35 s7 c 308689 4254667 0.98 0.77 0.81 0.92 0.97 1.08 1.36 1.17 s8 f 306606 4252861 0.58 1.24 1.99 0.7 0.95 0.83 1.58 1.45 s8 c 306606 4252861 1.1 0.81 0.89 0.96 0.99 1.2 0.93 1.03 s9 f 305939 4252099 0.51 0.77 0.79 0.5 1.01 0.69 3.53 22.1 s9 c 305939 4252099 1.27 0.93 1.05 1.09 1.08 1.42 1.08 1.27 s10 f 305527 4251633 0.6 0.62 0.54 0.57 0.86 0.52 0.73 2.1 s10 c 305527 4251633 1.22 0.88 0.88 1.1 0.99 1.18 2.12 1.16 parameter model nugget effect partial sill nugget/ partial sill rms srms fe circular 0.00151 0.00135 1.1156 0.05047 1.022 spherical 0.00150 0.00137 1.0950 0.05028 1.017 exponential 0.00146 0.00153 0.9517 0.05119 1.023 gaussian 0.00167 0.00127 1.3165 0.05061 1.031 circular 0 0.36795 0 0.5347 0.8304 mn spherical 0 0.36795 0 0.5399 0.8376 exponential 0.01416 0.36795 0.0385 0.554 0.8569 gaussian 0.00500 0.36795 0.0136 0.5352 0.8459 circular 0 0.24437 0 0.4221 0.8868 cd spherical 0 0.24816 0 0.4274 0.8901 exponential 0 0.29234 0 0.4552 0.9356 gaussian 0.00028 0.27828 0 0.4142 0.9015 circular 0 0.00867 0 0.08893 0.9204 co spherical 0 0.00867 0 0.08956 0.9236 exponential 0 0.03115 0 0.09237 0.9404 gaussian 8.98e-06 0.00898 0 0.08999 0.9269 circular 0 0.01064 0 0.09753 0.8729 cu spherical 0.00262 0.00936 0.2794 0.09714 0.9326 exponential 0 0.01237 0 0.09722 0.8957 gaussian 1.09e-05 0.01090 0.0010 0.09552 0.8687 to process geostatistics, the normality of the data must first be evaluated, and the kolmogorov-smirnov test is used to distribute normal data (dao, et al., 2013). the results of this test are illustrated in table 3. this test indicated that all parameters have a normal distribution at a 0.05 level. after calculating and drawing experimental variograms, four models (including spherical, circular, exponential, gaussian) were fitted using experimental semi-variograms for each data set; the parameters for each one are displayed in table 4. table 4: variogram parameters for each of the elements. 178 maryam zare khosh eghbal, shahriar sobh zahedi as indicated in table 4, the nugget effects of zn, co, and cd elements for three models (including spherical, circular, and exponential) equals zero. low ratios of nugget/partial sill represent the spatial coherence of variables and the lesser this ratio, the greater the spatial coherence of the data will be (angulo-martínez, et al., 2009). the smallest ratios of this value were observed for cadmium, cobalt, and zinc in the three models. it is not possible to comment on the best data fit model because of the low nugget/sill for zn, cd, and co in the three different models; more calculations are required. cross-validation was conducted to validate the data using standard root mean square (srms) value and root mean square (rms) values. the calculated srms and rms values fitted from various models on different semi-variograms are given in table 4. the variogram with the smallest rms indicates the best model for the sample. the best models for each element were as follows: the exponential model for zn; the gaussian model for pb, cu, cd, and ni; the spherical model for fe; and the circular model for co and mn. in regards to the srms values, the use of kriging in spatial changes of selected data indicated that optimal estimation had been performed. preparing the map of contamination zoning: as described using the kriging method, the smallest rms establishes the criterion for selecting the best model. with this consideration, the best model was obtained for enrichment values of each elements. high precision estimation of the parameters in this and other studies (angulo-martínez, et al., 2009; bai, et al., 2011; bednarova, et al., 2013; castrignanò, et al., 2000; dao, et al., 2013) indicates that geostatistical methods are appropriate for studying spatial changes in data, using the necessary calculations when there is a lack of data. to obtain the best kriging models in gis, which are used to determine and interpret pollution in river sediments, raster layer were first generated for each element based on the best selected model during the kriging application stage (table 4). then, all raster layers were compiled, the resulting map was grouped into four classes, and the pollution zoning map was used to interpret pollution deposits in the khajeh kory river. figure 4: map of the superficial sediments accumulating in river pollution.   sample  location   to decrease uneven weighing effects by various factors, the raster map for each parameter was divided into four classes. these four class maps were collected, and the cumulative map was divided into four classes. the final map indicates that much of the river length contains high accumulations of contamination and that upstream parts of the river basin have been mainly formed from pyroxene deposits from the besat suburb, which contains the most pollution and largely depends on the igneous lithology of area (figure 4). as heavy metals exclusively demonstrate different contamination levels along the route and because cumulative contamination relies on multiple factors that are unique to each heavy metal, this implies different responses in the soil contamination of an element. conclusion: the conducted studies revealed that geochemical characteristics and river contamination differ at various points. the reason for this difference is based on the concentration of population centers, the way of life within these centers, land uses, and geology. urban population primarily concentrates in the eastern part of the drainage basin and decreases from east to west. the results demonstrate that elemental concentrations, excluding iron, increase as the river approaches the sea. iron concentration are higher upstream, as in station s10, despite the lack of contaminating agricultural land and residential areas; this is due to the ferruginous lithology of the area. additionally, the zn concentrations do not follow a clear pattern as the concentration fluctuates among different basin points and is at high concentrations after crossing all the basins and prior to entering to the sea. data from fine-grain samples from each station (s1f, s2f, s3f, s4f, s5f, s6f, s7f, s8f, s9f, s10f) were normalized, using aluminum, to determine enrichment. pollution can concentrate in fine-grain texture, and the data from this texture indicates area contamination. then, using the kriging method, the enrichment factor of the samples was converted to a map in an gis environment. to validate this map, rms factors were calculated and the most validated model was determined to fit each sample. next the data were converted to a surface map based on the best interpolation model as determined using kriging method. the obtained maps for each element were then converted to raster layer and these layers were collected, which may provide evidence of pollution. the results of classification indicate that much of the river length experiences low values of contamination. in addition, major parts of the river route contain low values of contaminants. the rest of the river route contains medium to high contamination. the highest amounts of contamination were observed between where the rivers exits the city and enters the sea, as well as in the areas crossing rice fields. in addition, in the heights with igneous lithology, high levels of contamination were observed around station s7 next to the besat residential suburb. as the river passes among pyroxene masses in the heights, the pollution is observed to be andesitic. this work highlights that the high rate of contamination in the khajeh kory drainage basin is from either natural or manmade sources. in the heights, the residential areas do not have considerable concentrations of heavy metals. additionally, there are no rice fields that can circular 0.00167 0.00554 0.3016 0.08104 0.9855 ni spherical 0.00179 0.00546 0.3282 0.0809 0.9866 exponential 0 0.00726 0 0.08408 1.005 gaussian 0.00290 0.00439 0.66013 0.0807 0.9896 circular 0.74358 0.59072 1.2588 1.158 1.096 pb spherical 0.73107 0.55505 1.3171 1.155 1.098 exponential 0.82256 0.34810 2.3630 1.154 1.07 gaussian 0.65856 0.94277 0.6985 0.4142 0.9015 circular 0 296.44 0 15.22 1.09 zn spherical 0 267.18 0 14.97 1.046 exponential 0 239.63 0 13.75 0.922 gaussian 15.83 243.24 0.0651 15.17 1.063 179spatial distribution of sediment pollution in the khajeh kory river using kriging and gis cause contamination, and the dominant coverage is natural rainforest. therefore, the existence of contamination in the heights is definitely related to the geology of the area. abundance of iron in river sediment after passing through the mass confirms its role in increasing elemental concentrations. the main reason for manmade contamination in the basin, excepting residential areas, are fertilizers and pesticides used in rice fields and the use of highly enriched lead, petrol and worn tires of transportation vehicles in urban areas. additionally, in urban areas, the main cause of contamination is domestic and urban sewage pouring into the river and the accumulation of waste on its banks. recommendations: contamination is one of the most important factors threatening the khajeh kory river ecosystem. as noted previously, wastes and effluents are considered to be the main sources of contamination. drainage water from agricultural lands contains high percentages of chemical fertilizers and materials that pour into the river. it is recommended that rivers should be managed to prevent solid waste accumulation and promote recycling and that the concentration of heavy metals in sewage feeding into the river should be monitored. additionally, sampling time intervals should be reduced and conducted in different seasons of the year to better and precisely analyze contamination in future research. references: angulo-martínez, m., lópez-vicente, m., vicente-serrano, s. m., & beguería, s. (2009). mapping rainfall erosivity at a regional scale: a comparison of interpolation methods in the ebro basin (ne spai). bai, j., xiao, r., cui, b., zhang, k., wang, q., liu, x., et al. (2011). assessment of heavy metal pollution in wetland soils from the young and old reclaimed regions in the pearl river estuary, south china. environmental pollution, 159(3), 817-824. bednarova, z., kuta, j., kohut, l., machat, j., klanova, j., holoubek, i., et al. (2013). spatial patterns and temporal changes of heavy metal distributions in river sediments in a region with multiple pollution sources. journal of soils and sediments, 1-13. castrignanò, a., giugliarini, l., risaliti, r., & martinelli, n. (2000). study of spatial relationships among some soil physico-chemical properties of a field in central italy using multivariate geostatistics. geoderma, 97(1), 39-60. choi, k. y., kim, s. h., hong, g. h., & kim, c. j. (2013). metal contamination and potential toxicity of sediment from lock gate port in south korea. dao, l., morrison, l., kiely, g., & zhang, c. (2013). spatial distribution of potentially bioavailable metals in surface soils of a contaminated sports ground in galway, ireland. environmental geochemistry and health, 1-12. fang, f., jiang, b., wang, h., & xie, h.-f. (2010). particle size distribution and health risk assessment of heavy metals in surface dust of wuhu urban area. geographical research, 29(7), 1193-1202. gheshlagh, f. s.-n., ziarati, p., & bidgoli, s. a. (2013). seasonal fluctuation of heavy metal and nitrate pollution in ground water of farmlands in talesh, gilan, iran. horowitz, a. j. (2009). monitoring suspended sediments and associated chemical constituents in urban environments: lessons from the city of atlanta, georgia, usa water quality monitoring program. journal of soils and sediments, 9(4), 342-363. kaushik, a., kansal, a., kumari, s., & kaushik, c. (2009). heavy metal contamination of river yamuna, haryana, india: assessment by metal enrichment factor of the sediments. journal of hazardous materials, 164(1), 265-270. kroulík, m., mimra, m., kumhála, f., & prošek, v. (2006). mapping spatial variability of soil properties and yield by using geostatic method. research in agricultural engineering, 1, 17-24. liu, x., wu, j., & xu, j. (2006). characterizing the risk assessment of heavy metals and sampling uncertainty analysis in paddy field by geostatistics and gis. environmental pollution, 141(2), 257-264. maanan, m., landesman, c., maanan, m., zourarah, b., fattal, p., & sahabi, m. (2013). evaluation of the anthropogenic influx of metal and metalloid contaminants into the moulay bousselham lagoon, morocco, using chemometric methods coupled to geographical information systems. environmental science and pollution research, 1-13. mcgrath, d., zhang, c., & carton, o. t. (2004). geostatistical analyses and hazard assessment on soil lead in silvermines area, ireland. environmental pollution, 127(2), 239-248. meybeck, m., horowitz, a. j., & grosbois, c. (2004). the geochemistry of seine river basin particulate matter: distribution of an integrated metal pollution index. science of the total environment, 328(1), 219-236. rodríguez, l., ruiz, e., alonso-azcárate, j., & rincón, j. (2009). heavy metal distribution and chemical speciation in tailings and soils around a pb–zn mine in spain. journal of environmental management, 90(2), 1106-1116. salati, s., & moore, f. (2010). assessment of heavy metal concentration in the khoshk river water and sediment, shiraz, southwest iran. environmental monitoring and assessment, 164(1-4), 677-689. seshan, b., natesan, u., & deepthi, k. (2010). geochemical and statistical approach for evaluation of heavy metal pollution in core sediments in southeast coast of india. international journal of environmental science and technology, 7(2), 291-306. singh, k. p., mohan, d., singh, v. k., & malik, a. (2005). studies on distribution and fractionation of heavy metals in gomti river sediments—a tributary of the ganges, india. journal of hydrology, 312(1), 14-27. varol, m., & şen, b. (2012). assessment of nutrient and heavy metal contamination in surface water and sediments of the upper tigris river, turkey. catena, 92, 1-10. xie, y., chen, t.-b., lei, m., yang, j., guo, q.-j., song, b., et al. (2011). spatial distribution of soil heavy metal pollution estimated by different interpolation methods: accuracy and uncertainty analysis. chemosphere, 82(3), 468-476. keywords: manabí province; git; soil erosion risk. palabras clave: erosión; mdt; ndvi; usle; ndvi; imágenes de satélite. how to cite item pacheco, h. a., mendez, w., & moro, a. (2019). soil erosion risk zoning in the ecuadorian coastal region using geotechnological tools. earth sciences research journal, 23(4), 293-302. doi: https://doi. org/10.15446/esrj.v23n4.71706 the activation of erosive processes on the slopes of the drainage catchments in the ecuadorian coastal region represents a serious environmental problem that results in a loss of soil resource with the consequent reduction in agricultural productivity, increasing lack of areas with vegetation, and decreased water quality. these issues increase, among other things, the probability of flooding. the purpose of this research is to zone soil erosion risk in the manabí province, which is located on the ecuadorian coast. the methodology is supported by the employment of geographic information technologies (git) in order to access and process information of interest, such as: precipitation data with which to calculate the r factor of the universal soil loss equation (usle); topography, which was employed to create a slope map from the digital elevation model (dem) aster; vegetation cover, by applying the normalized difference vegetation index (ndvi); land use, through the interpretation of orthophotos and a field survey; and infiltration capacity, obtained by considering the texture of the soil. the factors were combined by means a weighted relationship (map algebra) and exploiting the potential of software for the design of geographic information systems (gis). the results indicate the presence of soils with little or limited vegetal protection, in areas with steep slopes, moderate rainfall and extensive farming activities. these characteristics determine conditions of high soil erodibility, and propitiate the triggering of erosive processes. indeed about 19% of the surface of the manabí province classifies with soil erosion risk between high and very high. these areas should be prioritized by regional government agencies to implement soil restoration and protection policies. surfaces with medium soil erosion risk represent 67% of the territory, which are also areas that require certain levels of attention. the surfaces with low soil erosion risk represent only 14% of the province, corresponding to areas of low slope. abstract soil erosion risk zoning in the ecuadorian coastal region using geo-technological tools zonificación de riesgo erosivo en la costa ecuatoriana usando herramientas geotecnológicas issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n4.71706 la activación de fuertes procesos erosivos en las vertientes de las cuencas hidrográficas de la provincia de manabí, representa una seria problemática ambiental que se traduce en la pérdida del valiosísimo recurso suelo, con la consecuente disminución en el rendimiento de la productividad agrícola, el incremento de áreas desprovistas de vegetación, la disminución de la calidad del agua, la intensificación de las inundaciones, entre otras cosas. el objetivo de ésta investigación consiste en zonificar el riesgo erosivo en la provincia de manabí; para ello se hace uso de las tecnologías de la información geográfica en el acceso y procesamiento de información de interés como es el caso de las precipitaciones con el cálculo del factor r de la usle, topografía, elaborando un mapa de pendientes a partir del mde aster, cobertura vegetal, usando el índice normalizado de diferencia de vegetación (indv), uso del suelo, a través de interpretación de ortofotos y levantamiento de campo, y capacidad de infiltración, considerando la textura del suelo. la combinación de los factores se realizó a través de una relación ponderada, aprovechando las potencialidades de softwares para trabajo en ambiente gis, específicamente las relacionadas con el álgebra de mapas. los resultados indican la presencia de suelos con escasa o nula protección vegetal, en áreas de pendientes elevadas, pluviosidad considerable y actividades agropecuarias extensivas, lo cual condiciona la fuerte acción de los procesos erosivos, con las consecuencias de bajo rendimiento en la producción agropecuaria, azolve y contaminación de los cuerpos de agua con la consecuente disminución de la calidad de vida en la población, evidenciando la necesidad de implementar políticas, estrategias y acciones, por parte de los gobiernos autónomos descentralizados, orientadas a impulsar el fortalecimiento de capacidades institucionales y comunitarias que promuevan la sostenibilidad ambiental. resumen record manuscript received: 14/04/2018 accepted for publication: 16/08/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 4 (december, 2019): 293-302 henry antonio pacheco1*, williams méndez1, armando moro1 1universidad técnica de manabí * corresponding author: henrypacheco@gmail.com g eo lo g ic a l en g in ee r in g 294 henry antonio pacheco, williams méndez, armando moro introduction one of the main environmental problems treated in recent years has been to solve soil erosion, which has increased worldwide, due to the need to expand the agricultural frontier to produce food and provide habitat for a growing population. situation that has led to the analysis, in many countries, looking to establish as anthropogenic conditionings, interact in the activation of erosive processes, in order to evaluate and zone the risk levels, and to design measures responding to each environmental reality. in the case of ecuadorian coastal region the socio-environmental conflicts, are specifically because more than 60% of the population, is engaged in agricultural activities not productive, in addition to the fact that they employ few mechanisms that are sustainable for the environment, which leads to pollution and the degradation of soils, forests and waters, thus decreasing the population’s quality of life (instituto nacional de estadística y censos [inec], 2010). in the drainage catchments of the manabí province, the sensitivity to the activation of erosive processes, represents a serious environmental problem, that results in the loss of soil resource, with the consequent reduction in agricultural productivity, increase in areas without vegetation, decrease in water quality, and the intensification of floods, among other scenarios (desinventar, 1994). so the qualification and quantification of soil erosion, is important as regards, assessing land units because it has effects such as the loss of the nutrients required for crop growth, the accumulation of sediments in river channels, and a reduction in water storage capacity owing to soil loss, thus reducing the natural flow of catchments (fonseca et al., 2014). it is as these in recent years, there have witnessed a significant increase in the use of geotechnological tools (remote sensing [rs] and gis) to study soil erosive processes. the principle objectives of many of these contributions are: (a) the multi-temporal and spatial modeling of erosive processes (zeng et al., 2013); (b) the analysis of the effects of the changes in vegetal cover and land use (lech-hab et al., 2015); (c) the quantification of soil loss (parveen and kumar, 2012; farhan et al., 2013; amsalu and mengaw, 2014; lahlaoi et al., 2015; ganasri and ramesh, 2016; tahiri et al., 2016); (d) the zoning of the hazard and risk control (prasannakumar et al., 2012; gaatib and larabi, 2014; uddin et al., 2014; chirala et al., 2015); and (e) soil mapping and classification (alves et al., 2015; rizzo et al., 2015; moura-bueno et al., 2016). likewise, in most of the contributions mentioned above, emphasis is placed on the use of the r factor of the revised universal soil loss equation (rusle) as an indicator of the soil erosion risk. however, little has been done about it in ecuador. in this sense, quantitative studies conducted in the ecuadorian sierra by de noni and trujillo (1986) reveal that the land lost due to erosion are considerable. for example, the results fluctuate among 200 to 500 t.ha-1.y-1 for 1982. the experiments clearly demonstrated the close relationships between runoff and rainfall intensity, in particular the maximum rainfall intensity in mm h-1 during 30 min and of medium frequency (im30), which fluctuate in the ecuadorian coastal region between 40 and 70 mm h-1. in this order of ideas and taking into account, the relevance of the manabí province, for the ecuadorian productive sector, in terms of agricultural potential and the propensity of the physical environment, to activate erosive processes, the purpose of this paper, was to use geotechnological tools, to zone the soil erosion risk, in order to spatiality analyse the problem, identify the most critical areas in the province, and propose possible control measures. study area the manabí province is divided into 22 cantons that cover an area of 1,942,764.33 ha located on the north-western pacific coast of ecuador and whose geoastronomical coordinates are latitude of -1.39° and 0.55° s and n, respectively, longitude of 79.38° 80.93° w. the area has an altitudinal range from 0 to 851 m.a.s.l. (figure 1). the rainfall regime is unimodal with more than 9 dry months on the coast with annual amounts of less than 200 mm, and progressive increase in a w-e direction until reaching an annual rainfall of 3,000 mm in the north-eastern region (instituto nacional de meteorología e hidrología [inamhi], 2016). large coastal plains and fluvial valleys, in addition to some mountainous systems with maximum elevations of 851 m.a.s.l, mainly represent the relief. the major fluvial systems are formed of the drainage catchments of the chone and portoviejo rivers (winckell et al., 1997). the lithology consist of more than 90% poorly consolidated sedimentary rocks. this is represented by rock outscrops belonging to the following lithological units: piñón formation (cretaceous; basaltic lavas, tuff and breccias), cayo formation (cretaceous; grauwackes, shales and basaltic mantles), san mateo formation (eocene; sandstones and conglomerates), zapallo formation (eocene; conglomerates, sandstones and shales), dos bocas formation (oligo-miocene; shales and mudstones), villingota formation (miocene; shales and sandstones), angostura formation (miocene; coquinas, sandstones and mudstones), onzole formation (miocene; shales and siltstones), borbón formation (pliocene; tobaceous sandstones), balzar formation (pliocene; sands, conglomerates and clays), san tadeo formation (pliocene; volcanic fan and lahars), tablazo formation (pleistocene; bioclastic marine terraces) and marine terrace (holocene; estuarine marine clays) (ministerio de minería [mm], instituto nacional de investigación geológico minero metalúrgico [inigmm], 2017; enríquez, 2018). figure 1. location of the study area, manabí province, ecuadorian coastal region. residual soils are composed primarily of very plastic and expansive clays. there are also large areas that are covered with silt, which has a high degree of collapsibility and scattering materials. these materials are easily mobilized and lose their strength as soon as they are excavated, thus making them easily erodible by runoff action (aguirre and chávez, 2009). the coverage of the native forest in the ecuadorian coastal region has been reduced and continues to decrease significantly in habitat extension and quality. this has caused an irreparable loss of its biodiversity, as a consequence of intensive deforestation and other human activities (cornejo, 2011). the land use and cover is primarily agricultural and pasture and it is the first-ranked province in the country as regards total agricultural land with 1,643,428 ha (inec, 2015). this represents 21.16% of the country’s productive activities in the economic sector, which accounts for 14% of provincial gross added value (inec, 2016), and which represents 37.1% of the economically active population (inec, 2010). this productivity is mainly supported by family farming occupying average areas of 4 ha per farmer (monteros and salvador, 2015). material and methods soil erosion risk the soil erosion risk was estimated by employing a combination of the precipitation, the r factor of the usle, topography with a slope map from 295soil erosion risk zoning in the ecuadorian coastal region using geo-technological tools ri = mean rainfall of month i (mm) rt = annual mean rainfall (mm) the information from each station was then employed to carry out a spatial interpolation in the gis using the ordinary kriging method according to the criteria of plouffe et al. (2015) and panagos et al. (2016). the values of the r were subsequently employed to reclassify the map of rainfall erosivity into 8 classes by adapting criteria of rivera and gómez (1991), páez (1999) and olivares et al. (2013). slope (s) the slope is one of the main dynamic factors to activate mass movements because it determines the kinetic and potential energy of an unstable mass. this increases susceptibility to the production of gravitational movements as the ground tilt increases. this instability is determined by the characteristics of relief, which have a significant influence on many variables involved in ecosystems dynamics and equilibrium (sánchez et al., 2002). the slope is, overall, one of the main conditioning factors of erosive processes, sediment transport and runoff (as a result of both surface flow and subsurface flow). the ground slope was calculated using the gis spatial analysis tools, taking the aster dem of 30 meters of resolution as a basis. the software slope tool has predefined algorithms with which to calculate the change in the maximum rate of elevation value of each cell with regard to its eight neighbours. the slope map was generated and reclassified according to the criteria suggested by pacheco (2012), which is based on a detailed analysis of the ground slope distribution using topographic profiles that allowed the determination of clinometric ranges. normalized difference vegetation index (ndvi) the variable related to vegetation was obtained from calculations of the ndvi using landsat le70110612003128edc00.tif satellite image. the image was processed with atmospheric and geometric corrections (chávez, 1989; chávez and kwarteng, 1989) in order to apply a cloud mask that allowed us to obtain the ndvi. this index was calculated by considering only two of the original bands provided by the sensor according to the expression (chuvieco, 1996): ndvi = −( ) +( )pir vis pir vis/ (4) where: ndvi = normalized difference vegetation index pir = near infrared band vis = visible red band the application of this equation enables an image with normalized values between 1 and -1 to be generated (karaburun, 2010; pacheco, 2012). the values close to 1 represent very dense vigorous vegetation, while negative and positive values near zero indicate regions devoid of vegetation. the method resulted in a regression model between the vegetation cover values measured in the field or obtained from the guide tables and ndvi values derived from the satellite image. this information was then used to reclassify the standard deviation of the ndvi values (intervals = 1/3 of the deviation). the results show 6 classes, of which less than 4 correspond to water bodies, and whose variations are possibly owing to sedimentation, depth and vegetation. the highest values correspond to vegetation in different states of health, density and size. land use (lu) the land use factor was evaluated and weighted by employing the map of coverage and land use of ecuador elaborated by ministerio del ambiente de ecuador (mae) – ministerio de agricultura, ganadería, acuacultura y pesca (magap) (2015) at scale of 1:100,000 as input. land use information from the map in question was obtained from field work and the analysis and interpretation of satellite imagery and high resolution orthophotographs. the spatial units of use include forests, agricultural lands, shrub and herbaceous vegetation, water bodies, anthropic areas (populated areas and infrastructure) and other types of land. relative weights were assigned to each type of use depending on its influence on the activation of erosive processes. the criteria utilized for the allocation of weights and reclassification of land use map were the dem aster, vegetation cover using ndvi, land use and infiltration capacity variables from ministerio de agricultura, ganadería, acuacultura y pesca [magap] (2016). the variables were reclassified into five classes of soil erosion risk, as follows: 1 (very low), 2 (low), 3 (moderate), 4 (high) and 5 (very high). the variables were combined using a weighted ratio, obtained from expert judgment by means of a multi-criteria evaluation (starkloff and stolte, 2014; vulević et al., 2015), and using software for the analysis and interpretation of satellite images and gis, specifically those related to map algebra. the weighting process of the factors determined the following results: 20% (0.2) for the usle r factor, 30% (0.3) for the topographic slope, 30% (0.3) for the ndvi, 10% (0.1) for land use, and 10% (0.1) for soil texture, expressed as follows: soil erosion risk r* s* ndvi* lu* st*= ( ) + ( ) + ( ) + ( ) +0 2 0 3 0 3 0 1 0. . . . ..1( )(1) where: r = usle r factor (rainfall) (mj mm ha-1 h-1 y-1) s = slope (°) ndvi = normalized difference vegetation index (dimensionless; the ndvi values range between -1 and 1, where the negative values correspond to areas without vegetal coverage and the positive values at different levels of vegetal coverage, with 1 being the highest density of vegetation) lu = land use (dimensionless; the types of land uses were reclassified into 5 classes according to the following criteria: current land use, land use change and land protection policies) st = soil texture (dimensionless; the types of soil textures were reclassified into 5 classes according to the following criteria: coarse, moderately coarse, medium, fine and very fine) the erosion risk map was reclassified into four classes according to the author’s criteria. for the validation and calibration of the results obtained for this map, 50 control points were georeferenced in the field, which were superimposed onto the map in question. usle r factor the usle r factor was determined from processing rainfall records weather stations camposano, chone, el carmen, jama, julcuy, olmedo, pedernales, portoviejo and rocafuerte, provided by the instituto nacional de meteorología e hidrología (2016) for the period 1984-2011. any missing rainfall data were estimated by using rational and average substitution methods, while homogenization and consistency were estimated by means of the “rachas” or round method (soares et al., 2012). the influence of rainfall on erosive processes was estimated by considering the equation proposed by echeverri and obando (2010). this was generated and validated by carrying out a linear regression analysis of the modified fournier index (mfi) and the monthly average i30 reported for 15 year records, with a determination index of 0.84 and a significance level of 5%. a spearman correlation coefficient matrix was produced after considering the cases of significant and highly significant correlations. the simple linear regressions were determined to a significance level of 5%. the models with a determination index of ≥ 0.75 and a low variation coefficient (≤ 12%) were selected according to the expression: r = ( ) +∑ 12 1 30 4 28 3. .mfi (2) where: r = erosivity (mj mm ha-1 h-1 y-1) mfi = modified fournier index the mfi was estimated according to the equation proposed by arnoldus (1977) to represent climate aggression: mfi = ∑ 12 1 2r ri t/ (3) where: 296 henry antonio pacheco, williams méndez, armando moro based on those proposed by paroissien et al. (2015), who consider current land use, land use change and land protection policies. soil texture (st) the soil texture map developed by magap (2016) at scale of 1:1,250,000 was used as the basis for the evaluation and weighting of this factor. the information presented on the map was obtained from field work and an analysis of soil samples, discriminated by lithological and geomorphological units. the textural groups presented on the map are coarse, moderately coarse, medium, fine and very fine (ditzler et al., 2017). the assignment of relative weights to textural classes was based on expert judgment. the map was reclassified on the basis of the criteria proposed by de lannoy et al. (2014) and cisty et al. (2015), who considered hydric models with high precision algorithms and techniques. results and discussion usle r factor the rainfall amounts (annual means) and their spatial distribution in the manabí province can be observed in table 1 and figure 2. it is characterized by a progressive increase in rainfall from the coast towards the mountains in a westeast direction, with annual amounts of less than 400 mm on the coast, which is distinctive from the arid areas in the region located between montecristi, manta, jaramijó and sucre. amounts significantly increase to annual values of 1,600 mm towards the north and south of the province, while maximum amounts close to 3,000 mm have been recorded in the north-eastern part, as registered in el carmen canton. the erosivity shows the same spatial distribution of the rainfall with the presence of low, moderate, high and very high classes. according to rivera and gómez (1991) páez (1999) and olivares et al. (2013), the territory is mostly covered by class 4, which is categorized as moderate erosivity (table 1), and extends from north to south across the province as a central strip. this is followed by the high class to the center and east of the territory in a north-south direction. the low class is toward the coastal area in the jama, rocafuerte, portoviejo and manta axis. finally, the very high class that covers the smallest area is concentrated to northeast of the manabí province. the usle r factor is a measure of the erosive power of the rain that reflects its kinetic energy, it is directly correlated with the loss of soil and the amount of sediment carried by runoff and can, therefore, be considered as a good indicator of the erosion risk. actions for the practice of sustainable agricultural activities should, therefore, be considered in sectors with high and very high erosivity. however, in areas with low erosivity levels, it is possible to intensify erosion processes triggered by extraordinary rainfalls. in this sense, according to data from portoaguas (2017), with a total rain sheet of 70 mm for a duration of 24 hours, a turbidity in the portoviejo river of 25,000 ntu was measured, which reflects a high erosion rate associated with that particular event, as well as a high amount of sediment carried. slope (s) the low slopes are in the coastal areas and intermountain valleys (table 2 and figure 3), which serve as settlements for the largest cities in the province, of which manta, portoviejo, chone, calceta, junín and tosagua stand out. the erosive processes in these locations are minimal while, conversely, floods and sediment accumulation predominate. these areas account for over 74.65% of the territory and are categorized as having a low and very low influence on the activation of erosive processes. the high and very high slopes cover 6% and the moderate slopes 19% of the province (table 2) and are mainly located in the central mountain range region (flavio alfaro, chone, bolívar, junín and santa ana cantons) (fig. 3) and, to a lesser extent, in the pedernales and jama northern and jipijapa and puerto lópez southern cantons. it is in these cantons where the slope conditions greatly influence the genesis of intense erosive processes. table 1. annual mean values of rainfall and erosivity in the manabí province, ecuadorian coastal region. station annual rainfall (mm) erosivity r (mj mm ha-1 h-1 y-1) area (km2) area (%) erosivity class description rocafuerte 471.04 3,162.12 4,149.97 21.36 3 lowporto viejo 603.85 3,797.41 jama 687.52 3,990.93 pedernales 1,049.18 5,767.14 8,067.56 41.53 4 moderatejulcuy (jipijapa canton) 869.16 6,024.03 chone 1,144.15 6,407.31 camposano (paján canton) 1,364.27 7,815.62 6,187.25 31.85 5 high olmedo 1,625.53 9,001.34 el carmen 2,945.62 12,683.72 1,022.86 5.26 6 very high total 19,427.64 100.00 figure 2. spatial distribution of the annual rainfall in the province of manabí, ecuadorian coastal region. 297soil erosion risk zoning in the ecuadorian coastal region using geo-technological tools table 2. area and percentage distribution of the slope classes in the manabí province, ecuadorian coastal region. class slope (°) description area (km2) area (%) 1 ≤ 5 very low 5672.87 29.20 2 4 – 15 low 8829.86 45.45 3 15 – 25 moderate 3772.85 19.42 4 25 – 35 high 1023.84 5.27 5 ≥ 35 very high 128.22 0.66 total 19,427.64 100.00 figure 3. spatial distribution of the topographic slopes in the province of manabí, ecuadorian coastal region. normalized difference vegetation index (ndvi) the ndvi has values ranging between -0.26 and 0.64 (table 3 and figure 4), thus indicating the presence of uncovered or few vegetation covered areas with below zero values. the majority of these areas are located in the coastal cantons, mainly in the center and south of the province. the presence of the densest vegetation cover with values above 0.5, although not completely vigorous, is observed with increased abundance in the cantons located in the eastern sector of the province. li et al. (2015) reported the recovery of vegetal cover in regions which ecological engineering projects were been carried out. this could be assessed by changes observed in the ndvi on satellite images. their findings have shown that vegetation is effective in protecting soil from erosive processes. land use (lu) according to the inec (2015), manabí is located between the ecuadorian provinces with the largest areas of agricultural activity (permanent, temporary and fallow crops, natural and cultivated pastures). the cultivated and natural pastures comprise 886,633 ha, which allows them to provide habitats for the largest livestock production in the country, with 17.76% of the national total. land use is a dynamic process that spans several decades and has a significant influence on the activation of erosive processes. this is understandable when considering the soil disturbance prompted by local people on the mountainous hillsides, and that adversely affects the ecosystem functions. the annual and seasonal crops are those most prone to being affected by vulnerability, resulting in serious problems such as the indiscriminate felling of native trees and the pollution of water resources. these activities take place in order to obtain greater extensions of land devoted to the planting of agricultural products or livestock breeding (rodríguez and sandino, 2010). the values shown in the table 4 and figure 5 indicate that more than 50% of the territory is predominately covered by different crops and pastures, located in the northern center of the province in the portoviejo, santa ana, chone, flavio alfaro, jama and pedernales cantons. these uses condition the activation of erosive processes. on a small area of the province the forest is preserved, in which the effects of erosion are decreased by the protection that this biome offers to the soil (garcía-ruiz et al., 2008; nadal-romero et al., 2012). another class includes the combined coverage of pastures and crops where there is bare soil during the dry season. this condition is highly prone to the activation of erosive processes with the first rains in the winter season, which move sediments from eroded hillsides toward the water bodies and urban areas, thus increasing the possibility of overflows and floods in addition to table 3. area and percentage distribution of erosion risk according to the ndvi, manabí province, ecuadorian coastal region. class ndvi description area (km2) area ( % ) 1 > 0.60 very low 835.39 4.30 2 0.40 – 0.60 low 3122.02 16.07 3 0.30 – 0.40 medium 4509.16 23.21 4 0.20 – 0.30 high 6469.40 33.30 5 < 0.20 very high 4491.67 23.12 total 19,427.64 100.00 figure 4. spatial distribution of ndvi in the province of manabí, ecuadorian coastal region. 298 henry antonio pacheco, williams méndez, armando moro decreasing the quality and quantity of water intended for human consumption and the practice of agricultural activities. areas of greater disturbance resulting from uses associated with anthropic activities condition more severe soil erosion responses (high runoff coefficients) owing to their greater sensitivity to the impact of high intensity rainfall events (regüés et al., 2014). soil texture (st) the distribution percent in the textural composition of soils (table 5 and figure 6) reflects the fact that fine (58.21%) and very fine (6.55%) material predominates almost 65%, which reduces the infiltration capacity and favors both the accelerated increase in runoff and the erosion rates. the hydraulic parameters of the soil are strongly linked to its texture (de lannoy et al., 2014). for example, sandy soils drain more easily than clay soils and have, given their moisture content, a much higher hydraulic conductivity that maximizes when soil physicochemical characteristics are altered by the occurrence of forest fires (inbar et al., 2014). these fires occur frequently in the manabí province, these fires are generated frequently, as is reported by some residents of rural parishes. table 4. area and percentage distribution of the land use classes, manabí province, ecuadorian coastal region. land use class description area (km2) area ( % ) 100 % artificial water body 1 very low 2,273.03 11.70 100 % natural water body 100 % urban area water 100 % sand banks 2 low 423.52 2.18100 % shrimp farms permanent crops forest 3 medium 10,710.46 55.13 forest + cultivated forest forest + crops forest + grass forest + shrubby vegetation parks protection shrubby vegetation cultivated forest 4 high 2,109.84 10.86 cultivated forest + grass crops + shrubby vegetation grass + shrubby vegetation transients transients + grass crops + grass 5 very high 3,910.78 20.13 grass total 19,427.64 100.00 figure 5. spatial distribution of the land uses in the province of manabí, ecuadorian coastal region. table 5. area and percentage distribution of the soil texture, manabí province, ecuadorian coastal region. textures class description area (km2) area (%) urban area 1 very low 64.11 0.33 water bodies 25.26 0.13 coarse texture 48.57 0.25 beaches, barrier islands, sand bank 1.94 0.01 shrimp farms 2 low 184.56 0.95 moderately coarse texture 1,466.69 7.55 medium texture 3 medium 5,055.07 26.02 fine texture 4 high 11,308.83 58.21 very fine texture 5 very high 1,272.51 6.55 total 19,427.64 100.00 spatially, there is a fine texture from north to south throughout the province with a compact presence in the western sector, while the very fine texture is concentrated in the chone and tosagua cantons. the mean texture is most predominant in the upper part of the chone and portoviejo river basins, and in the vicinity of the floodplain of the latter. the percentage of coarse and moderately coarse textures is concentrated in the el carmen canton of the north-eastern sector of the province, and in some lots isolated from the upper and middle parts of the portoviejo river basin. erosion risk map as part of the validation and calibration process of the results obtained for the soil erosion risk map, 50 control points were georeferenced in the field in 299soil erosion risk zoning in the ecuadorian coastal region using geo-technological tools which evidence of erosive processes was observed, such as gullies and erosion grooves. these control points were superimposed onto this map, thus allowing us to discover a correspondence of more than 90% with the high and very high categories. this map (figure 7) allows us to visualize the spatial distribution of soil erosion risk in the province of manabí, and also enables us to relate the different risk classes with the factors that have the greatest weight as regards their determination by interest areas (compared by the superposition of the final erosion map onto the map of each of the factors). this has allowed us to identify that the factors with the greatest influence on soil erosion risk are the slope and the soil coverage, thus corroborating the point made by vulević et al. (2015). in manabí, the cantons (flavio alfaro, chone, bolívar and pichincha) with highest erosion risk are located in the central-eastern sector of the province, concurring with the highest slopes, deforested areas and agricultural activity with inappropriate intensive use practices, such as the felling of native forests and the systematic burning of crop waste. the main cantons with medium erosion risk (portoviejo, santa ana, 24 de mayo, junín and olmedo) are located in the southeast part of the province, coinciding with low to moderate slopes and moderate precipitation amounts. the cantons with the least erosion risk (manta, montecristi and jaramijó) correspond to those located in coastal areas of center of the province, where gentle slopes prevail, and there are low amounts of rainfall and urban land use, protected by asphaltic and concrete infrastructures. in the area distribution of erosion risk shown in table 6, it observed that a low percentage of the territory shown a very high erosion risk. however, when considering the real extent that this percentage represents, government agencies consider that the erosion risk may have a high impact, and this is reflected in their planned protection activities. as an example, the project for provincial forestation and reforestation developed by the provincial government of manabí, which must invest for more than 7 years in order to cover solely this portion of manabí territory, fulfills the goal of 10,000 ha reforested per year. for this program to be successful it is necessary to propose alternative technological changes in agricultural practices so as to avoid and reduce, for example, the felling of forests and fires in crop areas, which cause changes in the characteristics of the soil surface (stoof et al., 2015). this modifies the roughness, thus increasing the erosion risk by surface flow, particularly when considering the projections of routschek et al. (2014), who claim that one of the climate change scenarios reflects the increased intensity of rainfall, with a severe impact on soil loss, until at least 2050. additionally, paroissien et al. (2015) reported that, in order to ensure soil resource sustainability, changes in their use must be taken into account. they proved that the soil erosion rate and its life expectancy are more sensitive to land use changes than to variations in rainfall. they also found that for simulations of the a1b scenario proposed by the intergovernmental panel on climate change (ipcc), the erosion rate was quantified as being 4.2 t.ha-1.y-1 with a soil life expectancy of 249 years. this reflects a decrease of 24 years when compared to current conditions. the manabí province should be of particular interest in this scenario, changes in land use associated with agricultural activities conditioned by the effects of el niño are reported, which are attributed to substantial losses of economic resources by farmers of permanent crops such as coffee and cocoa. these farmers have seen the need to migrate to short-cycle crops and pastures in order to achieve livelihoods. according to ministerio del ambiente de ecuador [mae] (2012), deforestation maps for 1990-2008 show an accelerated process of this type of anthropic intervention, with a rate over 10,000 ha.y-1 in the coastal region, conditioning the existence of soils with scarce or no vegetal protection in areas of steep slopes, moderate rainfalls and extensive agricultural activities. these features determine the strong action of erosive processes and loss of soil resources, with the consequences of a low yield in the agricultural production, figure 6. spatial distribution of the soil textures in the province of manabí, ecuadorian coastal region. figure 7. spatial distribution of soil erosive risk in the province of manabí, ecuadorian coastal region. table 6. area and percentage distribution of the soil erosion risk in the manabí province, ecuadorian coastal region. class value description area (km2) area (%) 1 ≤ 2 low 2,737.35 14.09 2 2 – 3 medium 12,925.21 66.53 3 3 – 4 high 2,989.91 15.39 4 ≥ 4 very high 775.16 3.99 total 19,427.64 100.00 300 henry antonio pacheco, williams méndez, armando moro siltation and the pollution of water bodies, and an increased hazard of floods and waterborne diseases. the combination of the erosion risk with the massively high rate of deforestation in the province has caused an increase in flood risks, not only because of the material volumes carried but also owing to the silting of riverbeds. the impacts are thus manifested in the flooded areas of the manabí province, located in the lower parts of the bolívar, chone, tosagua and portoviejo cantons and as an exceptional case in manta. the erosion risk combined with anthropic factors such as deforestation, indiscriminate land use, road construction, excavation and overloads generated by the development of structures on steep ground, similarly have significant influences on the genesis of mass removal processes. the cantons located in the central mountain range of the coast, are in this respect, those with an increased susceptibility to this morphodynamic. a direct consequence of erosion is the loss of good quality soils, thus decreasing agricultural production and signifying that farmers must use agrochemicals to improve yields or decide to fell additional extensions of forest, making the problem a vicious circle. the sediments generated by erosion are incorporated into the discharge of the rivers, creating operational problems in hydraulic projects, such as the chone-carrizal multipurpose (hydraulic infrastructure for irrigation, human consumption, hydroelectric energy and flood control, in the bolívar, chone and tosagua cantons), which is less effective owing to sedimentation, and prolonging the problem associated with the lack of water in the province. the increased sediment load in the rivers affects biodiversity because it prevents the natural reproduction of some species of fish and plants, thus causing an imbalance in the ecology of natural systems such as mangroves and tidelands, among others, and generating important losses as regards fisheries and aquaculture. finally, environmental degradation affects the socio-economic activities of the inhabitants of this territory, inducing demographic processes such as migration, which can lead to what the organización de las naciones unidas [onu] (2015) has called “environmental refugees” (human groups that move driven by environmental degradation) as the result of deforestation, rising sea levels, expanding deserts and catastrophic climatological events. all these environmental problems exist in the manabí province, where the number of environmental refugees could exceed the number of refugees caused by wars (federación internacional de la cruz roja [ficr] y la sociedad de la media luna roja [smlr], 2012). the environmental problems dealt with in this paper are of current and global importance, as was understood in the post-2015 schedule of the onu (2015), with the proposal of sustainable development goals (sdg), which emphasizes environmental concerns and economic growth. these objectives are accompanied by goals that are operationalized by means of indicators focused on measurable results whose aim is to promote to global and universal actions in nature that are applicable to all countries, considering the particular characteristics that distinguish them. conclusions the git allow the processing of useful, timely, current, accurate and abundant information regarding geospatial features. in this particular case, they have allowed us to zone the soil erosion risk in the manabí province. the rainfall erosive factor was determined by using the mfi. the ndvi proved to be a good indicator of the vegetation cover spatial distribution, and it correlates very well with the risk of soil erosion. a total of 365,872.95 ha that represent 19.38% of manabí territory are in the categories of high and very high erosion risk. the cantons of manabí with the highest soil erosion risk are located in the central-eastern sector of the province, such as flavio alfaro, chone, bolívar and pichincha, where the highest slope values with deforested areas and agricultural use with inappropriate intensive practices are. the cantons with medium erosion risk are located in the southeast part of the province, standing out among them portoviejo, santa ana, 24 de mayo, junín and olmedo. the level of risk in these areas is determined mainly by the predominance of low to moderate slopes and moderate rainfall amounts. the lower erosion risk corresponds to the coastal areas in the center of the province that are highly represented in the manta, montecristi and jaramijó cantons, with predominantly gentle slopes, low amounts of rainfall and urban land use protected by an asphaltic and concrete infrastructure. strong inter-institutional support is required in order to promote erosion risk mitigation actions. it is recommended that the decentralized autonomous governments (dag) and productive sectors make use of this information to plan the optimal use and occupation of geographical space in view of their agroecological conditions. in this respect, it is important to promote the development of programs for forestation and reforestation, such as those put forward by the provincial government of manabí. acknowledgements the authors are very grateful to the secretaría nacional de educación superior, ciencia y tecnología (senescyt) of ecuador for the financial support of the research project from which this paper has been derived. the authors would also like to thank the gobierno provincial de manabí (gpm) and the universidad técnica de manabí (utm) for the logistic support of the project as host institutions. the authors are also grateful to professor yuly asención (phd) from the university of arizona and mrs. sally newton for the reviewing the english in the manuscript of this paper. references aguirre, m. & chávez, m. (2009). susceptibilidad al deslizamiento de los suelos y rocas, provincia de manabí, ecuador. artículo de tesis de grado, escuela superior politécnica del litoral, facultad de ingeniería en ciencias de la tierra, guayaquil, ecuador. https://www.dspace. espol.edu.ec/handle/123456789/149 alves, m. r., demattê, j. a. m. & barros, p. p. s. (2015). multiple geotechnological tools applied to digital mapping of tropical soils. revista brasileira de ciência do solo, 39(5), 1261-1274. http://dx.doi. org/10.1590/01000683rbcs20140410 amsalu, t. & mengaw, a. (2014). gis based soil loss estimation using rusle model: the case of jabi tehinan woreda, anrs, ethiopia. natural resources, 5(11), 616-626. http://dx.doi.org/10.4236/nr.2014.511054 arnoldus, h. m. j. (1977). methodology used to determine the maximum potential average annual soil loss due to sheet and rill erosion in morocco. fao soil b, 34, 39-48. http://agris.fao.org/agris-search/ search.do?recordid=xf8001961 chávez jr., p. s. (1989). radiometric calibration of landsat thematic mapper multispectral images. photogrammetric engineering and remote sensing, 55(9), 1285-1294. https://pubs.er.usgs.gov/ publication/70016102 chávez jr., p. s. & kwarteng, a. y. (1989). extracting spectral contrast in landsat thematic mapper image data using selective principal component analysis. photogrammetric engineering and remote sensing, 55(3), 339-348. https://www.researchgate.net/profile/andy_ kwarteng/publication/262301549_extracting_spectral_contrast_ in_landsat_thematic_mapper_image_data_using_selective_ principal_component_analysis/links/0deec53745b103723b000000/ extracting-spectral-contrast-in-landsat-thematic-mapper-imagedata-using-selective-principal-component-analysis.pdf chirala, u., gurram, m. k. & kinthada, n. r. (2015). mapping of soil erosion zones of meghadrigedda catchment, visakhapatnam, india for conservation – a geospatial approach. international journal of geosciences, 6(4), 326-338. http://dx.doi.org/10.4236/ijg.2015.64026 chuvieco, e. (1996). fundamentos de teledetección espacial (3a ed). rialp, madrid, españa, 449 pp. cisty, m., celar, l. & minaric, p. (2015). conversion between soil texture classification systems using the random forest algorithm. air, soil and water research, 8, 67-75. http://dx.doi.org/10.4137/aswr. s31924 cornejo, x. (2011). endemismo en la región litoral. en: s. león, r. valencia, n. pitman, l. endara, c. ulloa & h. navarrete (eds.). libro rojo de las plantas endémicas del ecuador 2000 [online]. pontificia universidad católica de ecuador, quito, ecuador. https://bioweb.bio/floraweb/ librorojo/litoral/ 301soil erosion risk zoning in the ecuadorian coastal region using geo-technological tools de lannoy, g. j. m., koster, r. d., reichle, r. h., mahanama, s. p. p. & liu, q. (2014). an updated treatment of soil texture and associated hydraulic properties in a global land modeling system. journal of advances in modeling earth systems, 6(4), 957-979. http://dx.doi. org/10.1002/2014ms000330 de noni, g. & trujillo, g. (1986). degradación del suelo en el ecuador. cultura, 24, 383-394. http://horizon.documentation.ird.fr/exl-doc/ pleins_textes/cc-2010/26531.pdf desinventar. (1994). sistema de inventario de efectos de desastre [database]. https://www.desinventar.org/es/ ditzler, c., scheffe, k. & monger, h. c. (2017). soil survey manual (agriculture handbook nº 18). unites states department of agriculture, soil science division staff, usa, 587 pp. echeverri, l. & obando, f. h. (2010). erosividad de las lluvias en la región centro-sur del departamento de caldas, colombia. revista facultad nacional de agronomía medellín, 63(1), 5307-5318. http://www. redalyc.org/articulo.oa?id=179914617009 enríquez, a. (2018). zonificación de la susceptibilidad de fenómenos de remoción en masa provocados por terremotos en la ciudad de portoviejo. trabajo de titulación previo a la obtención del título de ingeniería en geología, universidad central del ecuador, facultad de ingeniería en geología, minas, petróleos y ambiental, quito, ecuador. http://www.dspace.uce.edu.ec/handle/25000/15636 farhan, y., zregat, d. & farhan, i. (2013). spatial estimation of soil erosion risk using rusle approach, rs, and gis techniques: a case study of kufranja watershed, northern jordan. journal of water resource and protection, 5(12), 1247-1261. http://dx.doi.org/10.4236/ jwarp.2013.512134 federación internacional de la cruz roja y la sociedad de la media luna roja. (2012). informe mundial sobre desastres – migración forzosa [website]. http://www.ifrc.org fonseca, d., durigon, v. l., homen, m. a., sampaio, w. & sánchez, p. t. (2014). predicting soil erosion using rusle and ndvi time series from tm landsat 5. pesquisa agropecuária brasileira, 49(3), 215224. http://dx.doi.org/10.1590/s0100-204x2014000300008 gaatib, r. e. & larabi, a. (2014). integrated evaluation of soil erosion hazard and risk management in the oued beht watershed using remote sensing and gis techniques: impacts on el kansra dam siltation (morocco). journal of geographic information system, 6(6), 677-689. http://dx.doi.org/10.4236/jgis. 2014.66056 ganasri, b. p. & ramesh, h. (2016). assessment of soil erosion by rusle model using remote sensing and gis – a case study of nethravathi basin. geoscience frontiers, 7(6), 953-961. http://dx.doi. org/10.1016/j.gsf.2015.10.007 garcía-ruiz, j. m., regües, d., alvera, b., lana-renault, n., serrano-muela, p., nadal-romero, e., navas, a., latron, j., martí-bono, c. & arnáez, j. (2008). flood generation and sediment transport in experimental catchments affected by land use changes in the central pyrenees. journal of hydrology, 356, 245-260. inbar, a., lado, m., sternberg, m., tenau, h. & ben-hur, m. (2014). forest fire effects on soil chemical and physicochemical properties, infiltration, runoff, and erosion in a semiarid mediterranean region. geoderma, 221-222, 131-138. http://dx.doi.org/10.1016/j.geoderma.2014.01.015 instituto nacional de estadística y censos. (2010). censo de población y vivienda [online data]. http://anda.inec.gob.ec/anda/index.php/catalog instituto nacional de estadística y censos. (2015). encuesta de superficie y producción agropecuaria [online data]. www.ecuadorencifras.gob.ec/ encuesta-de-superficie-y-produccion-agrop instituto nacional de estadística y censos. (2016). estadísticas económicas [online data]. http://www.ecuadorencifras.gob.ec/estadisticaseconomicas-3/ instituto nacional de meteorología e hidrología. (2016). anuarios meteorológicos [online data]. www.serviciometeorologico.gob.ec/ biblioteca/ karaburun, a. (2010). estimation of c factor for soil erosion modeling using ndvi in buyukcekmece watershed. ozean journal of applied sciences, 3 (1), 77-85. http://www.academia.edu/408871/estimation_ of_c_factor_for_soil_erosion_modeling_using_ndvi_in_ buyukcekmece_watershed lahlaoi, h., rhinane, h., hilali, a., lahssini, s. & khalile, l. (2015). potential erosion risk calculationusing remote sensing and gis in oued el maleh watershed, morocco. journal of geographic information system, 7(2), 128-139. http://dx.doi.org/10.4236/jgis.2015.72012 lech-hab, k. b. h., issa, l. k., raissouni, a., el arrim, a., tribak, a. a. & moussadek, r. (2015). effects of vegetation cover and land use changes on soil erosion in kalaya watershed (north western morocco). international journal of geosciences, 6 (12), 1353-1366. http://dx.doi. org/10.4236/ijg.2015.612107 li, s., yang, s., liu, x, liu, y. & shi, m. (2015). ndvi-based analysis on the influence of climate change and human activities on vegetation restoration in the shaanxi-gansu-ningxia region, central china. remote sensing, 7(9), 11163-11182. http://dx.doi.org/10.3390/ rs70911163 ministerio de agricultura, ganadería, acuacultura y pesca. (2016). actualización tema: ecuador continental – mapa de suelos 2003 – variable textura, versión 2, escala de trabajo 1:250.000, escala gráfica 1:1.000.000. quito, ecuador. ministerio del ambiente / ministerio de agricultura, ganadería, acuacultura y pesca. (2015). mapa de cobertura y uso de la tierra de ecuador continental 2013-2014, escala 1:100.000. quito, ecuador. http://sni. gob.ec/mapa-cobertura-uso ministerio de minería, instituto nacional de investigación geológico minero metalúrgico. (2017). mapa geológico de la república del ecuador, escala 1:1.000.000. quito, ecuador. https://www.geoinvestigacion. gob.ec/mapas-geologicos-2017/ ministerio del ambiente de ecuador. (2012). mapa de deforestación histórica. quito, ecuador. http://mapainteractivo.ambiente.gob.ec monteros, a. & salvador, s. (2015). rendimientos de maíz duro seco en el ecuador verano 2014 (junio – noviembre). ministerio de agricultura, ganadería, acuacultura y pesca, coordinación general del sistema de información nacional, dirección de análisis y procesamiento de la información. https://docplayer.es/41869297-rendimientos-de-maizduro-seco-en-el-ecuador-verano-2014-junio-noviembre.html moura-bueno, j. m., dalmolin, r. s. d., ten caten, a., ruiz, l. f. c., ramos, p. v. & dotto, a. c. (2016). assessment of digital elevation model for digital soil mapping in a watershed with gently undulating topography. revista brasileira de ciência do solo, 40, 1-15. http:// dx.doi.org/10.1590/18069657rbcs20150022 nadal-romero, e., lana-renault, n., serrano-muela, p., regües, d., alvera, b. & garcía-ruiz, j. (2012). sediment balance in four catchments with different land coverin the central spanish pyrenees. zeitschrift für geomorphologie, 50(3), 147-168. olivares, b., lobo, d. & verbist, k. (2013). estimación del factor erosividad de la lluvia en la zona semiárida de san pedro de chile. geominas, 41(62), 135-140. https://www.researchgate.net/profile/deyanira_ lobo_lujan/publication/48223843_estimacion_del_factor_ erosividad_de_la_lluvia_en_la_zona_semiarida_de_san_pedro_ chile/links/5616669c08ae0f21400711a0.pdf organización de las naciones unidas. (2015). proyecto de documento final de la cumbre de las naciones unidas para la aprobación de la agenda para el desarrollo después de 2015. http://www.un.org/es/comun/ docs/?symbol=a/69/l.85 pacheco, h. a. (2012). el índice de erosión potencial en la vertiente norte del waraira repano, estado vargas, venezuela. cuadernos de geografía, 21(2), 85-97. http://dx.doi.org/10.15446/rcdg páez, m. (1999). diseño de prácticas de conservación con la ecuación universal de pérdida de suelo. cidiat, mérida, venezuela, 125 pp. 302 henry antonio pacheco, williams méndez, armando moro panagos, p., ballabio, c., borrelli, p. & meusburger, k. (2016). spatio-temporal analysis of rainfall erosivity and erosivity density in greece. catena, 137, 161-172. http://dx.doi.org/10.1016/j.catena.2015.09.015 paroissien, j. b., darboux, e., couturier, a., devillers, b., mouillot, f. & le bissonnais, y. (2015). a method for modeling the effects of climate and land use changes on erosion and sustainability of soil in a mediterranean watershed (languedoc, france). journal of environmental management, 150, 57-68. http://dx.doi.org/10.1016/j. jenvman.2014.10.034 parveen, r. & kumar, u. (2012). integrated approach of universal soil loss equation (usle) and geographical information system (gis) for soil loss risk assessment in upper south koel basin, jharkhand. journal of geographic information system, 4(6), 588-596. http://dx.doi. org/10.4236/jgis.2012.4 6061 plouffe, c. c. f., robertson, c. & chandrapala, l. (2015). comparing interpolation techniques for monthly rainfall mapping using multiple evaluation criteria and auxiliary data sources: a case study of sri lanka. environmental modelling & software, 67, 57-71. http://dx.doi. org/10.1016/j.envsoft.2015.01.011 portoaguas. (2017). registros diarios de turbiedad del agua en los meses enero y febrero de 2017 [unpublished data]. portoaguas, laboratorio agua potable “cuatro esquinas”, portoviejo, manabí, ecuador. prasannakumar, v., vijith, h., abinod, s. & geetha, n. (2012). estimation of soil erosion risk within a small mountainous sub-watershed in kerala, india, using revised universal soil loss equation (rusle) and geoinformation technology. geoscience frontiers, 3(2), 209-215. http:// dx.doi.org/10.1016/j.gsf.2011.11.003 regüés, d., arnáez, j., badía, d., cerdá, a., echeverría, m. t., gispert, m., lana, n., lasanta, t., león, j., nadal, e. & pardini, g. (2014). analysing surface runoff and erosion responses to different land uses from the ne of iberian peninsula through rainfall simulation. geophysical research abstracts, 16, egu2014-11061. http://meetingorganizer. copernicus.org/egu2014/egu2014-11061.pdf rivera, j. & gómez, a. (1991). erosividad de las lluvias en la zona cafetera central colombiana (caldas, quindío y risaralda). cenicafé, 42(2), 37-52. rizzo, r., demattê, j. a. & lacerda, m. p. c. (2015). espectros vis-nir do solo e fuzzy k-médias aplicados na delimitação de unidades de mapeamento de solos em topossequências. revista brasileira de ciência do solo, 39(6), 1533-1543. http://dx.doi.org/10.1590/01000683rbcs20140694 rodríguez, a. & sandino, m. (2010). comunidades locales vulnerables y cambio climático, estudio de caso: cuenca del río cane, iguaquecampohermoso, boyacá. medio ambiente, observatorio grupo de estudios en economía política y medio ambiente, universidad sergio arboleda, bogotá, colombia. routschek, a., schmidt, j. & kreienkamp, f. (2014). climate change impacts on soil erosion: a high-resolution projection on catchment scale until 2100. engineering geology for society and territory, 1, 135-141. http://dx.doi.org/10.1007/9 78-3-319-09300-0_26 sánchez, r., urrego, l., mayorga, r. & vargas, g. (2002). modelo para el pronóstico de la amenaza por deslizamientos en tiempo real. en: simposio latinoamericano de control de erosión, bucaramanga, colombia, 1-8. soares, j., coelho, c., carvalho, t., oliveira, e. & valente, s. (2012). the use of rainfall simulations to assess land degradation and soil erosion produced by an slm technology, portugal. geophysical research abstracts, 14, egu2012-885. http://adsabs.harvard.edu/ abs/2012eguga..14..885s starkloff, t. & stolte, j. (2014). applied comparison of the erosion risk models erosion 3d and lisem for a small catchment in norway. catena, 118, 154-167. http://dx.doi.org/10.1016/j.catena.2014.02.004 stoof, c. r., ferreira, a. j. d., mol, w., van den berg, j., de kort, a., drooger, s., slingerlan, e. c., mansholt, a. u., ferreira, c. s. s. & ritsema, c. j. (2015). soil surface changes increase runoff and erosion risk after a low–moderate severity fire. geoderma, 239-240, 58-67. http://dx.doi. org/10.1016/j.geoderma.2014.09.020 tahiri, m., tabyaoui, h., tahiri, a., el hadi, h., el hammichi, f. & achab, m. (2016). modelling soil erosion and sedimentation in the oued haricha sub-basin (tahaddart watershed, western rif, morocco): risk assessment. journal of geoscience of environment protection, 4(1), 107-119. http://dx.doi. org/10.4236/gep.2016.41013 uddin, k., dhakal, m. & joshi, g. (2014). an optical high and medium spatial resolution approach for erosion-prone areas assessment in mustang, nepal. international journal of geoscience, 5(4), 383-393. http:// dx.doi.org/10.4236/ijg.2014.54037 vulević, t., dragović, n., kostadinov, s., belanović, s. & milovanović, i. (2015). prioritization of soil erosion vulnerable areas using multicriteria analysis methods. polish journal of environmental studies, 24(1), 317-323. http://dx.doi.org/10.15244/pjoes/28962 winckell, a., zebrowski, c. & sourdat, m. (1997). los paisajes naturales del ecuador: las regiones y paisajes del ecuador (tomo iv, vol. 2, geografía física). en: geografía básica del ecuador, centro ecuatoriano de investigación geográfica, instituto panamericano de geografía e historia (sección ecuador), instituto francés de investigación científica para el desarrollo en cooperación, instituto geográfico militar, quito, ecuador, 417 pp. zeng, z. y., cao, j. z., gu, z. j., zhang, z. l., zheng, w., cao, y. q. & peng, h. y. (2013). dynamic monitoring of plant cover and soil erosion using remote sensing, mathematical modeling, computer simulation and gis techniques. american journal of plant sciences, 4(7), 1466-1493. http://dx.doi.org/10.4236/ajps.2013.47180 contenido efecto de la rugosidad del terreno en las secciones sismicas: modelamiento por elementos finitos robinson quintana, sandra cespedes, alfredo ghisays, gabriel perez y luis montes vides 5 amenaza sismica por metoda determlntsta en el trianqulo de fallas valera-bocono-oca en el occidente de venezuela 9 raymi a. castilla, frank a. audermard m. y jose antonio rodriguez cambios temporales del valor de b: lposibles correlaciones con la actividad hidrotermal del volcan tungurahua? c. indira molina p., mario c. ruiz 17 monitoreo de emisiones de gas radon vinculado a la actividad microsismica en la zona de sampacho, cordoba, argentina virginia grosso, romina daga, guillermo sagripanti y diego villalba 2a dispersion sismica por presencia de capas someras de alta velocidad: modelamiento con diferencias finitas 2s sandra cespedes, robinson quintana, alfredo ghisays, gabriel perez y luis montes analisis espacial v temporal de q-coda en el piedemonte l1anero (colombia) juan p. duran, carlos a. vargas y luis a. briceno 33 determinacion de mecanismos focales para temblores pequeiios 0 moderados: region del caribe nororiental victor a. huerfano moreno 40 resultados del uso de geodesia satelital para estudios geodinamicos en colombia 43 hector mora-paez, robert trenkamp, james kellogg, jeff freymueller y milton ordonez-villota t.eves de atenuaclon de la intensidad macrosismica en colombia augusto antonio gomez capera y elkin de jesus salcedo hurtado 53 soluclon a un problema estatlco en un area con topografia abrupta usando una velocidad de reemplazamiento variable luis antonio castillo lopez y carlos pedraza 62 optlmizacion en la soluclon del problema inverso en geofisica usando el algoritmo de entrenamtento supervisado de levenberg-marquardt yaqueline figueredo baez y ramiro leon diaz campos 66 modelamiento slsmtco utilizando teo ria de raves en medios transversalmente tsotroptcos: teo ria v ejemplos carlos piedrahita. trino salinas y saul guevara 70 issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v26n2.68391 earth sciences research journal earth sci. res. j. vol. 26, no. 2 (june, 2022): 173 182 m e t e o r o l o g y detecting climatic and hydrological changes in gaoyou lake, eastern china, over the last millennium using pollen, grain size, and lithology records shuheng li1,2*, wei guo3, na gao1 , jianfeng du1,2 1. school of urban and environmental sciences, northwest university, xi’an 710127, china. 2. shaanxi key laboratory of earth surface system and environmental carrying capacity, northwest university, xi’an 710127, china. 3. department of environmental sciences and technology, xi’an jiaotong university, xi’an 710049, china. *corresponding author. e-mail: lish@nwu.edu.cn abstract: numerous lakes formed in the area where the huai river and the grand canal converged during the historical period, and it played a substantial role in the evolution of the regional environment. gaoyou lake is a vital detention reservoir in the lower reaches of the huai river. variations in hydrology and basin-wide environmental changes were analyzed based on pollen assemblages, grain size data, and lithology from sediment cores collected at gaoyou lake. this study focused on variations in the hydrodynamic processes of gaoyou lake that were responses to changes in climate, regional hydrological events, and human activity. the high percentages of pollen from terrestrial herbs and results of principal component analysis suggested that the environment of eastern china was arid from ad 900 to 1300 and that this corresponded to the dry conditions of the medieval climate anomaly. the subsequent period between ad 1300 and 1650 was more humid, as revealed by an increase in aquatic pollen and a decrease in grain size. a rise in xerophytic herb taxa (particularly artemisia and chenopodioideae) indicated that the climate became arid again from ad 1650 to 1850, during the little ice age. the modern period (ad 1850 onwards) was characterized by another increase in moisture, as reflected by a decrease in herb pollen and grain sizes. the shifts in the course of the yellow river in ad 1194 and ad 1855 were recorded by abrupt changes in lithology, sediment characteristics, and the percentages of alien pollen taxa. intense human impacts were demonstrated by cereal-type pollen and an increase in pollen of other cultivated plants such as cruciferae and poaceae. this study provides a foundation for further analysis of the dynamic mechanisms between lake development, climate change, and the effects of significant river changes. keywords: paleoclimate; lake sediment; last millennium; pollen; gaoyou lake. detección de los cambios climáticos e hidrológicos en el último milenio del lago gaoyou, al oriente de china, a través del polen, tamaño del grano y registros litológicos resumen varios lagos se formaron en el área donde convergen el río huai y el gran canal durante el período histórico, lo que juega un rol importante en la evolución del entorno regional. el lago gaoyou es un importante reservorio en los segmentos bajos del río huai. en este trabajo se analizaron las variaciones en hidrología y los cambios ambientales de la cuenca amplia con base en los grupos de polen, tamaño del grano y litología de las muestras de sedimento recolectadas en el lago gaoyou. este trabajo se enfoca en las variaciones de los procesos hidrodinámicos del lago gaoyou que fueron las respuestas a los cambios del clima, los eventos hidrológicos regionales, y la actividad humana. los altos porcentajes de polen de las hierbas terrestres y los resultados de los análisis de componentes principales sugieren que el ambiente de la china oriental fue árido entre los años 900 a 1300 dc y esto se corresponde con las condiciones de sequía del período cálido medieval. el período subsecuente entre 1300 y 1650 fue más húmedo, como se revela por un incremento en el polen acuático y en el decrecimiento del tamaño del grano. un incremento en el taxón de las hierbas xerófitas (especialmente artemisias y chenopodioideaes) indica que el clima se tornó arido de nuevo entre 1650 y 1850, durante la pequeña edad de hielo. el período moderno (de 1850 en adelante) se caracterizó por otro incremento en la húmedad, como se refleja en un decrecimiento del polen de hierbas y en los tamaños del grano. los cambios en el curso del río amarillo entre 1194 y 1855 se registran por las variaciones abruptas en la litología, las características de los sedimentos y los porcentajes en los taxones de polen alienígena. los impactos humanos intensos se demuestran con la ocurrencia de polen tipo cereal y un incremento en el polen de otras plantas cultivadas como las cruciferas y las poáceas. este estudio presenta una base para análisis posteriores de los mecanismos dinámicos entre el desarrollo del lago, el cambio climático y los efectos de los mayores cambios de los ríos. palabras clave: paleoclimate; sedimentos de lagos; último milenio; polen; lago gaoyou. record manuscript received: 18/10/2020 accepted for publication: 07/06/2022 how to cite item: li, guo, gao, & du. (2022). detecting climatic and hydrological changes in gaoyou lake, eastern china, over the last millennium using pollengrain size and lithology records. earth sciences research journal, 26(2), 173-182. https://doi.org/10.15446/esrj.v26n2.68391 https://doi.org/10.15446/esrj.v26n2.68391 mailto:lish@nwu.edu.cn https://doi.org/10.15446/esrj.v26n2.68391 174 shuheng li, wei guo, na gao, jianfeng du introduction the last millennium was a critical period for understanding environmental change because of the occurrence of major changes in climate, particularly during crucial periods such as the medieval climate anomaly (mca) (approximately ad 900-1300) (soon et al., 2003; gao et al., 2015; peteet et al., 2020) and the little ice age (lia) (approximately ad 1300-1850) (mann et al., 2009; sun et al., 2020). hydroclimatic variability demonstrates significant regional differences divided by the summer monsoon boundary in china compared to increasing trends in temperature (chen et al., 2010; chen et al., 2015). most researches analyzing the variability of moisture and precipitation in eastern china over the last 1000 years have been based on historical documents (zheng and wang, 2005; zheng et al., 2006). multi-proxy records from lake sediments are main sources of evidence for paleoenvironmental changes. sedimentary lithology gives an overview about the compositional variability of the sediment records and the occurrence of unconformities. palynological investigations of lacustrine sediments are useful tools for reconstructing past vegetation and hydrologic changes in lake ecosystems that were responses to variations in climate (djamali et al., 2009). however, reconstructing climate based solely on pollen analysis could be biased due to human influence (li et al., 2014). the sizes of lacustrine sediments grains are controlled primarily by water dynamics and sediment sources, which in turn are influenced by various factors such as local climate changes, regional hydrological transitions, and human activity (lerman, 1978). therefore, the environmental significance of variations in grain size is complex. the combined study of lithology, pollen assemblages and grain size has the potential to provide different yet complementary information about the paleoenvironment (li et al., 2009). during the historical period, several lakes have developed in the area where the huai river, the yellow river, and the jing-hang grand canal converge. this vitally contributed to the water conservation, flood control, and social economy development. multi-proxy limnologic studies of these lakes using different time scales revealed that the evolution of the regional environment was frequently affected by the interaction between the large rivers, which formed the remaining lacustrine deposits, creating important archives for use in the study of the regional hydrological and environmental evolution (he et al., 2005; zhang et al., 1999; xia et al., 1995; chen et al., 2013). located in the lower reaches of the huai river and west of the jinghang grand canal, gaoyou lake is the sixth largest fresh-water lake of china (fig. 1). early studies largely based on historical records show that gaoyou lake was comprised of lagoons during the early holocene (ling, 2001). the area of gaoyou lake was diminished to a shallow bay approximately 8,000 bp, when sea transgression reached its shorelines. interfluve land was present between the gaoyou and baoying counties until 6,000 bp. in the historical period, gaoyou lake was part of the waterways of the yellow river and huai river and served as the water resource to the grand canal. as a result, the evolution of gaoyou lake was closely linked with the large river system and regional climate changes in eastern china, which made it an ideal site for the study of the regional hydrological and environmental evolution. researches during the modern period have studied the changes in the water area, biological characteristics, and water quality of gaoyou lake (wei et al., 2017; guan et al., 2019; zhang et al., 2020). however, the paleoenvironmental studies of gaoyou lake sediments have not received adequate attention. our research team combined grain size, magnetic susceptibility, and geochemical proxies in gaoyou lake sediments to infer regional climate fluctuations (li et al., 2007; li et al., 2015). yet, biological proxies such as pollen, one of the most sensitive paleoenvironmental indicators, have not been studied. this study presents the first pollen records collected from gaoyou lake, as well as grain size and lithology analyses, to provide a better understanding of the lake evolution and to deduce regional climate changes during the last millennium. the exploration of the interaction between the lake development, climate changes, and large river effects will provide scientific basis for regional ecosystem protection and construction. the aims of this paper can be summarized as: 1. reconstructing the climatic and hydrological changes of gaoyou lake during the last millennium (past 1,100 years), 2. detecting the evidence of human activities and historical events such as the course shifts of the yellow river from the core sediments of gaoyou lake. study area gaoyou lake (32°42’n–33°04’n, 119°06’e–119°25’e) spans approximately 39 km from north to south and approximately 17.3 km from west to east. it covers a 674.7 km2 area, with the lowest elevation at 3.3 m above sea level (a.s.l) and an average elevation of 4.4 m a.s.l of the lake bottom. it has an average water depth of 1.4 m and maximum depth of approximately 2.4 m (wang and dou, 1998). the average annual runoff into the lake is 2.2 × 1010 m3. water from the huai river contributes 95% of the total input for the south of the lake via the water way to the yangtze river (chen et al., 2017). several small creeks and streams–such as the tonglong river, xincun river, baita river, qinnan river, and wangqiao river–also feed the lake (fig. 1). figure 1. (a) map showing the location of gaoyou lake in china. (b) the core sampling sites of gaoyou lake the current catchment area has a northern subtropical monsoon climate, with an average annual temperature of 16.7℃ and average annual precipitation of 1,029 mm according to the meteorological records at gaoyou from 1981 to 2010 (fig. 2). precipitation during the monsoon season (june-september) accounts for 56.4% of the annual total. july (mean temperature of 27.2 ℃) and january (2.8℃) are the warmest and the coldest month, respectively. the zonal vegetation in the catchment area is a mixed forest consisting of deciduous and evergreen broad-leaved trees. the current vegetation around gaoyou lake is mostly composed of farm and economic crops. trees in the genera salix, ulmus, populus, and paulownia are commonly used in artificial cultivation (wang and yang, 1990). there are 38 genera of planktonic algae in gaoyou lake, and these are dominated by chlorophyta (40.0%), cyanophyta (24.2%), and bacillariophyta (13.64%) (wei et al., 2017). figure 2. meteorological observations of gaoyou lake (1981-2010). data collected from the national meteorological information center of china (http:// data.cma.cn) materials and methods coring, dating and grain size in november 2010, four cores (gy05, gy06, gy07, and gy08) were taken from gaoyou lake using a piston corer within a 2-m long and 80-mm diameter fitted steel coring tube. the cores were transported to the laboratory and stored at 4 °c. each core was split, and the sediments were photographed 175detecting climatic and hydrological changes in gaoyou lake, eastern china, over the last millennium using pollen, grain size, and lithology records and described in detail. one half of each core was archived and the other half was used for analysis. in this study, we focused on the pollen, particle size and lithology analysis of samples from core gy08 (158 cm long) collected from the central part of the lake (32°53’36”n, 119°21’6”e) in 1.78 m of water. the lithology of core gy07 (32°55’6”n, 119°19’39”e), collected at a water depth of 1.8 m and located approximately 12 km from gy08 was described and the core was sampled for grain size and chronological control analysis. grain size analyses were completed for 186 samples from gy08 at 0.5cm interval along the top 30 cm, and 1-cm intervals between 30 and 158 cm. h2o2 and hcl were added to each sub-sample prior to analysis to remove organic and inorganic carbon. then, the sediment was dissolved in a (napo3)6 solution for 24 h. grain size distributions were determined by a laser particle size analyzer (mastersizer 2000, malvern panalytical, united kingdom). the grain size parameters were calculated using the excel macro-program gradistat (blott and pye, 2001), based on the following equations s1 (folk and ward, 1957). the age-depth model previously developed by li et al. (2015) for the cores sampled at gaoyou lake was used for the study. in total, six calibrated accelerator mass spectrometry (ams) 14c dates were presented in this study in two batches. radiocarbon ages of four samples from core gy 07 were determined at the beta analytic radiocarbon dating laboratory in miami, usa using ams in 2013. two samples (one from core gy07 and one from core gy08) were dated using the ams system at the xi’an accelerator mass spectrometry center, china, in 2020 (table 1). the radiocarbon dates were calibrated using the software calib 6.0 and the intcal20 curve. table 1. radiocarbon chronology from gaoyou lake sediments depth /cm core dated materials δ13c (‰) calibrated 14c age 2σ /a bp (ad) lab site lab no. 7 gy07 shells -11.6 modern miami 326289* 98 gy07 shells -9.7 660-550 bp (ad 12801400) miami 326291* 150 gy07 bulk sediments -28.1 800-690 bp (ad 11601260) miami 320812* 160 gy07 bulk sediments -27.6 1060-1030 bp (ad 900-920) miami 320811* 84 gy07 shells -6.8 670-630 bp (ad 12901320) xi’an xa50651 138 gy08 bulk sediments -25.5 800-765 bp (ad 11551185) xi’an xa50652 the data of the samples with * were quoted from li et al., 2015. pollen analysis pollen samples were collected at 5-cm interval. therefore, 30 samples were obtained with approximately 150 pollen grains in each sample. the samples were processed following the standard hf technique (fægri and iversen, 1989). this involves treating 1-2 g of sediment with 10% hcl, 10% koh, and 36% hf to remove carbonates, humid acids, and silicate, respectively. lycopodium tablets were added to the samples to calculate the pollen concentrations. pollen was counted and identified using an olympus bx51 microscope at the nanjing institute of geology and paleontology, chinese academy of sciences. pollen percentage and concentration calculations were performed using the software tilia 1.5.12. only major pollen taxa with an occurrence frequency exceeding 2% in at least three samples were included in the numerical analyses and the pollen diagram. the coniss program was used to divide the pollen diagram by constrained minimum-variance cluster analysis based on percentages of pollen (grimm, 1987). pca and dca analysis principal component analysis (pca) is considered a powerful data reduction technique that produces a visual scatterplot for the qualitative assessment of the resemblances and differences among samples (ter brakk, 1988; jiang et al., 2013; geng et al., 2019). to ordinate the pollen data from core gy08 and detect patterns of vegetation development, detrended correspondence analysis (dca) and pca were conducted based on the square root transformed pollen percentage data using canoco 4.5. dca was initially performed to estimate the underlying linearity of the data. the preliminary dca yielded a gradient length of 1.176 (< 2.0) standard units, which indicates a possible linear response of pollen assemblages to environmental variables. therefore, pca was chosen to ordinate the pollen data for the gaoyou lake sediment cores. the pollen and spore data were reduced to a two-dimensional plot and the resulting data set was displayed as a plot for samples and taxa based on pca. figure 3. lithology, grain size, and ams 14c dates at gy07 and gy08. data of gy07 were extracted from li et al., 2015 figure 4. age-depth model for the core gy08 and core gy07 from gaoyou lake sediments. data of core gy07 were quoted from li et al., 2015 results core lithology and chronology the lithology of gy08 has obvious layers, which may resulted from the changing paleosedimentary environment, paleoclimate and different material components. the upper 50 cm of gy08 was composed of brown and grey mud, with plant remains at the top 0-15 cm. the 50-100 cm tier had a grey and brown mud layer, and had horizontal-wavy bedding at the 80-90 cm tier. the 100-137 cm tier was dark grey to grey clay with a lot of shells and other organic fragments at the bottom of the segment. the 137-158 cm tier was grayish black and dark brownish clay, containing rich brownish organic matter 176 shuheng li, wei guo, na gao, jianfeng du and humus. figure 3 showed the obvious stratigraphic similarities between cores gy08 and gy07. core gy07 consisted of brownish-grey mud, grey clay, and dark brownish clay. the core showed wave and horizontal bedding at 85 cm, contained plant fragments on the top segment, and contained abundant shells at 144-161 cm. both cores had the dark organic layer at the bottom (137158 cm at gy08, 144-161 cm at gy07), marked as a characteristic layer for corresponding relation between the two cores. li et al. (2015) presented an age-depth model of the sediment sequence that was based on four ams radiocarbon dates obtained for gy07 in 2013 (table 1). to establish a more accurate chronology, two more samples (one from each gy07 and gy08) were used to supplement that ams dating in 2020. here, the chronology of gy08 was developed by extrapolating the calibrated radiocarbon dates from gy07 to gy08, which was corroborated by the date at 138 cm in gy08 (fig. 4). core gy08 was formed in the 1,100 year period from ad 900 to 2010, at an average sedimentation rate of 0.15 cm/yr. the sedimentation rate was especially high during ad 1200-1400. the increase of depositional rate was assumed to be related with the result of the course change of the yellow river in ad 1194 which brought heavy sediment into gaoyou lake. the interpretation was also justified by the abrupt increasing of particle size and pollen concentration, as discussed in detail below. grain size the sediment sequence of gy08 was dominated by silt, which had a mean content of 84.5%. the remainder consisted of clay, fine sand, and medium and coarse sand–the latter two found in very small amounts. the grain size data showed a relatively stable median value and variable coarsest value. the sand fraction fluctuated greatly, varying between 0 and 40.7% at an average of 6.2%. the mean grain size of the sediment samples ranged from 4.4 to 37.2 μm, with an average of 12.5 μm. the value of sorting varied between 2.1 and 7.3, with a mean value of 3.6. the value of skewness changed from -0.5 to 0.6, averaging -0.1 while kurtosis oscillated between 2.3 and 0.8, with an average of 1.1(table 2). the results indicate that the particle sorting was poor and finer relatively. the sediment distribution was asymmetric which shows that the samples may be combined from multiple sources. according to the results of the grain size analysis, suspended loads are the main component of sediments in gaoyou lake, whereas strong inflow occurred constantly. figure 5 showed several obvious variations of most parameters at depths around 100-135 cm, 65 cm, and 20 cm indicating fluctuant sedimentary environment. table 2. the maximum (max), minimum (min) and average values of clay (%), silt (%), sand (%), mean (μm), sorting, skewness and kurtosis of gy08 value clay (%) silt (%) sand (%) mean (μm) sorting skewness kurtosis max 23.0 96.0 40.7 37.2 7.3 0.6 2.3 min 2.9 56.4 0 4.4 2.1 -0.5 0.8 average 9.2 84.5 6.2 12.5 3.6 -0.1 1.1 figure 5. mean (μm), clay (%), silt (%), sand (%), sorting, skewness and kurtosis of gy08, based on the grain-size data figure 6. percentage pollen diagram for gy08 sediment core 177detecting climatic and hydrological changes in gaoyou lake, eastern china, over the last millennium using pollen, grain size, and lithology records pollen record a total of 96 pollen and spore taxa were identified. to interpret the pollen spectra, all pollen types were classified into five eco-groups according to the life form of their parent plants. these eco-groups included tree and shrub pollen (including pinus, quercus, cyclobalanopsis, ulmus, trema, celtis, liquidambar, salix, carpinus, and pterocarya); terrestrial herb pollen (including poaceae, asteraceae, chenopodioideae, cruciferae, artemisia, polygonum, fagopyrum, plantago, sesamum indicum, and zea mays); aquatic pollen (including typha and cyperaceae); fern spores (including selaginella, hicriopteris, pteris, dicranopteris, ceratopteris); and algae (including pediastrum and concentricystes). the photographs of the main pollen, spores, and algae in core gy08 are presented in figure s1. overall, terrestrial herb and tree & shrub pollen were the dominant pollen types in the entire profile with an average percentage of 46.5% and 34.8%, respectively. the pollen data are shown in figure 6. based on the changes of pollen types, percentages and concentrations of the major taxa, and coniss results, the pollen spectra can be divided into four pollen assemblage zones. the lowest pollen zone (phase a, 158-100 cm) is characterized by the lowest percentage of tree and shrub pollen with the average value of 31.2%. out of the 27 trees and shrubs pollen taxa identified, only 14 taxa were present in phase a including pinus, quercus, cyclobalanopsis, and ulmus. terrestrial herb pollen dominated with an average percentage of 50.3% and consisted mainly of chenopodioideae (4.12%) and artemisia (3.45%). among fern spores, ceratopteris reached its highest percentage in this zone. pollen concentrations ranged from 26 to 195 grains/drop, with an average concentration of 58.3 grains/drop. phase b (100-45 cm) was notable for a clear increase in pollen contents for most taxa. the contents of aquatic pollen (12.3%) and fern spores (11.0%) significantly increased compared to the entire core. several pollen types first appeared in this zone, such as zea mays, sesamum indicum, liquidambar, salix, and pediastrum. herb decreases to 39.0%, but they were still the dominant pollen. pollen concentrations varied from 27 to 176 grains/drop, with an average concentration of 57.9 grains/drop. the remarkable characteristic of phase c (45-20 cm) was the terrestrial herbs pollen percentages increase to their highest average value of 51.3% at the expense of the fern spore percentages (6.6%). pediastrum disappeared in this zone. the percentage of cyperaceae decreased to its minimum of 2%, while the percentages of artemisia (10.4%) and chenopodioideae (6.7%) increased to their peak values over the whole core. the average pollen concentration was at the lowest value. herbs were still the dominant pollen in phase d (20-0 cm) with percentages slightly decreasing to 49.7% compared to phase c. tree and shrub pollen percentages increased to 37.3% with relatively high content of quercus (2.7%), while the percentage of ulmus dropped to 2%. the percentages of fern spores and algae both increased in comparison to phase c, with the average values being 8.1% and 0.8%, respectively. the average pollen concentration increased to the highest value. pca analysis the result showed that cyperaceae, polygonum, typha, ceratopteris and other algae adapting to wet environments tend to spread in the positive values of the first axis, while artemisia, chenopodioideae and asteraceae–which prefer an arid environment–were distributed towards the opposite direction. this indicated that the first principal component mainly reflected the changes in humidity, with the samples on axis 1 showing positive values related to wet conditions and negative values related to dry conditions (fig. 7). taxa such as dicranopteris, cruciferae, selaginella, labiatae, sesamum, and pediastrum were influenced strongly by human agricultural activity (li et al., 2008; xie et al., 2008). they were mainly distributed at positive values along the second axis, implying that the second axis may represent the intensity of human activity. the pca of the pollen assemblages of the luonan basin also showed that the first axis reflected the variation in humidity and the second axis may relate to human activity (zhang et al., 2014). figure 7. results of pca analysis in the pollen record of core gy08. liq(liquidambar); hic(hicriopteris); sal(salix); que(quercus); sesi(sesamum indicum); oth-t(other trilete spore); oth-m(other monolete spore); sel(selaginella); ped(pediastrum); lab(labiatae); cru(cruciferae); dic(dicranopteris); typ(typha); pla(plantago); zea-(zea mays); tre(trema); cel(celtis); pte(pterocarya); art(artemisia); ulm(ulmus); hum(humulus); cyc(cyclobalanopsis); che(chenopodioideae); fag(fagopyrum); com(compositae); pin(pinus); pteris(pteris); cer(ceratopteris); car(carpinus); poa(poaceae); oth-a(other algae); pol(polygonum); cyp(cyperaceae) discussion proxy interpretation and discussion the assessment of lacustrine sediment lithology may provide a clue to reconstruct lake levels and infer environmental conditions. a highly variable local environment was suggested by rapidly changing lithology. in terms of sediment transport dynamics, the presence of coarse grains in the lithology may reflect climate change or results of mass movements (kliem et al., 2013). moreover and most important, the sedimentation lithology offers the detection of flood deposits, which are marked by a high content of soil organic matter and appear as dark-brown layers that can be easily differentiated from the background (simonneau et al., 2013; zhang et al., 2015). particle size analysis has been used as a well utilized tool for reconstructing paleo-sedimentary environments based on grain size parameters such as mean diameter, sorting, skewness and kurtosis (folk and ward, 1957; folk, 1974; sun et al., 2021). grain size of the lacustrine sediment is related to hydraulic processes associated with transporting, sorting and deposition (guo et al., 2021). the particle size chiefly depends on the hydrodynamic energy of the lake water, which is strongly linked to water level, energy of the input, and the source domains (li et al., 2009). on one hand, grain size can be regarded as a lake level index that is sensitive to regional hydrological conditions. previous studies (chen et al., 2003) show that during wet periods, water levels rise and lakes expand. hence, the distance between the lake shore and coring site would increase. it is more difficult for coarse particles to reach the core site, which leads to a decrease in grain size. conversely, during dry periods, coarse sediments can more easily reach the coring site due to the diminished lake area. as a result, the mean value of particle size increases. on the other hand, transport ability of runoff was important factor that controls the grain size of sediment materials. the change of sediment source, intensity of erosion, human activities could affect the particle size by modifying the runoff transport. thus, grain size has 178 shuheng li, wei guo, na gao, jianfeng du been widely used in the studies to trace and separate material sources (ghilardi et al., 2008; jia et al., 2012; sun et al., 2015). identification of the sources of the pollen types is essential for interpretation of the pollen signal and reconstruction of palaeovegetation. the evolution of gaoyou lake is affected by the interaction of the river-lake system, thus the pollen of the lacustrine sediments may have multiple sources. the waterborne pollen by the inflowing rivers may also make a substantial contribution to the pollen influx to the lake. thus, pollen in alluvium originates from both the realm of the rivers and the lake itself. former studies of alluvial pollen on the north china plain demonstrated that most of the arboreal pollen which derived from mountain trees, such as pinus, picea, quercus, betula, tilia and castanea, were transported by rivers from the mountains to the plain. by contrast, the appearance of salix, ulmus, populus and ailanthus, which were treated as plain trees, may derived from the plain itself. most of the nonarboreal pollen were derived from the plain around the lakes. high percentages of pinus pollen can be used as an indicator of flooding because the larger and heavier pollen grains were mostly deposited by flood. artemisia and chenopodioideae pollen percentages mainly have close relations with ecological regions (xu et al., 2007). cyperaceae pollen has a strong correlation with plant cover. vegetation and climate reconstruction using pollen records are often based on the presence and percentage abundance of selected indicator taxa (herzschuh et al., 2009). artemisia spp. and chenopodioideae are widespread in temperate regions of the northern hemisphere. they are common in temperate steppes, meadows, and uplands especially in dry areas (zhao et al., 2021). thus, increases in their pollen percentages may indicate climate changes in more arid conditions. pediastrum spp. is green algae common in freshwater and lives in shallow ponds, lakes, or small rivers. when pediastrum increases, it suggests a warm and humid climate, or increased nutrients in the lake system (tong et al., 2009). pollen assemblages are the proxies commonly used in paleoclimate reconstruction to identify human influences on the environment. detecting human impacts via pollen analysis is based not only on abrupt changes in pollen concentration and richness, but also by the occurrence of cereal-type and cropland weed pollen (li et al., 2008). cereal-type pollen has been used as an ideal indicator of human activity. the occurrence and increase of cerealtype pollen such as rice, corn, and wheat in sediments is direct evidence of an agricultural area (djamali et al., 2009). poaceae pollen is ubiquitous and widely used as a paleoclimatic indicator for increased regional aridity. the duration and intensity of human occupation are also variables affecting the percentages of poaceae pollen (bush, 2002). lake evolution and environmental reconstruction a synopsis of the interpretation of grain size and pollen combined with core lithology results in the following most probable scenario of the paleoenvironmental history of gaoyou lake. during stage a (158-100 cm; ad 900-1300), the percentages of tree and shrub pollen were the lowest of all phases. terrestrial herbs such as artemisia, asteraceae, and chenopodioideae showed relatively higher percentages than the average values of the whole core, indicating a drier climate. in addition, the low humidity was implied by negative pca-1 scores. the particle size data from gaoyou lake were characterized by marked fluctuations. based on the principle of sediment sorting, the high and greatly changed coarse component suggested a low and fluctuating lake level. the synthesis of the proxies indicated low water level and unstable climate during the deposition of this layer. this relatively dry interval can be attributed to an arid situation in the southern part of eastern china during the mca period. the results were in agreement with previous studies (zheng et al., 2006; mann et al., 2009). ge et al. (2014) also found the dry period to occur during the northern song, southern song, and yuan dynasties (ad 960-1368) in eastern china. the lithology of the bottom part of this layer consisted of dark brownish clay with a large amount of organic matter and humus. previous studies demonstrated that the color of the alluvial sediments appeared brown, which is a notable characteristic thereof, and a layer with abundant biogenic debris is a signal of flood events (sirocko et al., 2013). according to historical documents, the yellow river battered its banks and began to shift southward, encroaching the huai river in ad 1128, due to human disturbance for military purposes (xu, 1993). as a result, the yellow river took over the lower channel of the huai river heading for the yellow sea in ad 1194 (fig. 8). since then, a huge sediment load from the yellow river has entered gaoyou lake. the change of material sources also led to the change of pollen concentration. consistent with the result, figure 5 and figure 6 showed the increase in sand content and pollen concentration around ad 1200. former researchers found that suspended sand increases the buoyancy of the river water. due to the high sand content of the water from the yellow river, it contains more pollen (xu et al., 1996). figure 8. sketch map showing the yellow river’s course changes since ad 900 (after xu, 1993). 1. the channel after the major course change in ad 1048; 2. the channel after the major course change in ad 1194; 3. the channel after the major course change in ad 1494; 4. the channel after the major course change in ad 1855 stage b (100-45 cm; ad 1300-1650) was separated from the previous zone by the clear increase in the proportion of aquatic pollen, accompanied by the decreasing value of mean particle size, both indicating that climate conditions were more humid. positive pca-1 denoted that the climate became wetter at this stage. it was consistent with previous studies that relatively warm and humid climate prevailed in north jiangsu before the end of 16th century (chen et al., 1998). historical documents recorded 91 heavy floods during the yuan and ming dynasties (ad 1279-1643) (yang and wang, 1995). the presence of cereal-type pollen such as zea mays (65 cm, around ad 1530) and sesamum indicum (90 cm, around ad 1340) is important as it may mark the expansion of agricultural activities during this period. productive american crops such as maize were introduced and popularized in china at the beginning of the sixteenth century, which was confirmed by the peak in maize pollen percentage at 65 cm (around ad 1530). the earliest recorded planting of maize in china can be found during ad 1511 in “ying state records”, the neighboring anhui province. the record of maize cultivation in the gaoyou lake basin first appeared in the “xinghua county annals” in ad 1559 (zhang, 1990). few pine species are found in the modern vegetation of the lake basin. consequently, pinus pollen mainly should be an alien component transported to gaoyou lake. in this study area, fluvial flow is more capable than wind to transport large quantities of exotic pollen over long distances, especially for pinus pollen because of its larger size. as a result, increase of pinus spp. content suggested that the enhanced fluvial flow or the regional climate turned wet (jiang et al., 2013). the high pinus percentage of this stage indicated a more exogenous input to the lake. since ad 1194, the yellow river has encroached the huai river basin. as a result, huge quantities of water with high sediment 179detecting climatic and hydrological changes in gaoyou lake, eastern china, over the last millennium using pollen, grain size, and lithology records load from the yellow river have poured into the lower reaches of the huai river, leading to a rapid expansion of gaoyou lake (xu, 1993). finally, the lake enlarged gradually and amalgamated to form the present gaoyou lake around ad 1600 (liao, 1992). stage c (45-20 cm; ad 1650-1850) was marked by an upsurge in terrestrial herb taxa, especially dry-tolerant plants such as artemisia, asteraceae and chenopodioideae. the negative pca-1 scores, extreme low percentages of aquatics, and absence of algae indicated further drying during this stage that was apparently connected with the lia. previous research demonstrated that the lia was characterized by a cold and dry climate throughout china from the late 16th century until the end of the 19th century, especially in the midseventeenth century (zhu, 1973; chen et al., 1998). the abrupt decrease of mean particle size and increase of sand content at 20 cm in the core (dated around ad 1860) was prominent in this stage. it coincided with the decrease in percentages of several types of tree pollen, such as quercus, ulmus, celtis, and pinus. this transition may be due to the northward shift of the yellow river in ad 1855. historical documents showed that the yellow river broke its northern bank at tongwaxiang and changed to a northeast course into the bohai sea in ad 1855 (fig. 8), ending a 700-year long occupation of the huai river (xu, 1993). thus, proportions of coarse particles dropped sharply. the reduced stream inflow may have led to the decreased percentage of exotic tree pollens. stage d (20-0 cm; ad 1850-2010) was separated from the preceding zone by the notably decreasing value of the mean particle size and the most positive pca-1 scores, suggesting a more humid period. the decreased herb pollen percentages as well as the increased proportions of fern spores and algae also confirmed this inference. the anomalous warming over china in the 20th century is well known (yang et al., 2002). previous studies also reported that moister conditions in china occurred since 1890 (peng et al., 2009). the most conspicuous feature of this stage was the appreciable increase in anthropogenic herbs, such as fagopyrum and sesamum indicum, indicating intensified cereal cultivation. the high amounts of poaceae and cruciferae were also evidence of human disturbance (li et al., 2008). the rise of pediastrum is interpreted to be related to increased nutrients in the lake. it is documented that lake reclamation in the gaoyou lake area started in 1857 (wang and yang1990). in 1988, the area of gaoyou lake shrank by 25% compared to the 1950s. the population had almost doubled during the four decades from the 1950s to the 1990s (li et al., 2011). as a result, water quality has degraded since the 1990s with accelerated urbanization. conclusion 1. this study indicated that the main component of the sediments in core gy08 was silt. terrestrial herbs were the dominant pollen in the entire profile of core gy08. 2. pollen assemblages, sediment grain size and lithology of core gy08 recorded four main phases in the climatic and hydrological changes of gaoyou lake over the last millennium. during ad 900-1300, the climate was dry, and gaoyou lake experienced a stage of low water levels. the lake began to expand under more humid conditions after ad 1300. subsequently, the water levels dropped because of the dry climate during ad 1650-1850 in conjunction with the lia. after ad 1850, a wet climate prevailed in the study area. 3. major regional events such as course changes of the yellow river during ad 1194 and ad 1855 were reflected in abrupt variations of sediment characteristics and the changing percentage of river-transported pollen from exotic plants. the first appearance of maize pollen during ad 1560 as well as the increase in pollen of plants associated with human activities in the modern era reflected a widespread human interference with the natural vegetation and an intense cultivation of the lake area. acknowledgements the authors acknowledge the national natural science foundation of china (40901101) for funding the research. we thank the reviewers for their valuable comments and suggestions. dr. yin yong, tang wei, li yanjun and li haiqing are acknowledged for their assistance during field work and experimental measurements. we are grateful to dr. mao mili for the support of pollen analysis. thanks are also extended to professor yue ming for help with plant taxonomy. the authors also would like to thank wang jiachuan and yang qi for their kind help in improving the figures. references blott, s. j., & pye, k. (2001). gradistat: a grain size distribution and statistics package of the analysis of unconsolidated sediments. earth surface processes and landforms, 26(11), 1237-1248. https://doi.org/10.1002/ esp.261. bush, m. b. (2002). on the interpretation of fossil poaceae pollen in the lowland humid neotropics. palaeogeography, palaeoclimatology, palaeoecology, 177(s 1-2), 5-17. https://doi.org/10.1016/s00310182(01)00348-0. chen, f. h., chen, j. h., holmes, j., boomer, i., austin, p., gates, j. b., wang, n. l., brooks, s. j., & zhang, j. w. (2010). moisture changes over the last millennium in arid central asia: a review, synthesis and comparison with monsoon region. quaternary science reviews, 29(7-8), 1055-1068. https://doi.org/10.1016/j.quascirev.2010.01.005. chen, j. h., chen, f. h., feng, s., huang, w., liu, j. b., & zhou, a. f. (2015). hydroclimatic changes in china and surroundings during the medieval climate anomaly and little ice age: spatial patterns and possible mechanisms. quaternary science reviews, 107, 98-111. https://doi. org/10.1016/j.quascirev.2014.10.012. chen, j. j., jiang, t., & xu, p. z. (1998). climatic changes during the last 2000 years in jiangsu province. scientia geographica sinica, 18, 219-226. https://doi.org/10.13249/j.cnki.sgs.1998.03.219 chen, j. a., wan, g. j., zhang, f., david, d. z., & huang, r. g. (2003). lacustrine sediments records of different time scales: a case study of sediment grain size. science in china (series d), 33(6), 563-568. https://doi. org/10.3321/j.issn:1006-9267.2003.06.010. chen y. y., chen s. y., liu j. z., yao m., sun w. b., & zhang q. (2013). environmental evolution and hydrodynamic process of dongping lake in shandong province, china, over the past 150 years. environmental earth sciences, 68, 69-75. https://doi.org/10.1007/s12665-0121716-x. chen, y., guan y. q., miao j. z., & zhang, d. r. (2017). determination of the ecological water-level and assuring degree in gaoyou lake, northern jiangsu with long-term hydrological alteration. journal of lake sciences, 29(2), 398-408. https://doi.org/10.18307/2017.0216. djamali, m., de beaulieu, j. l., andrieu-ponel, v., berberian, m., miller, n. f., gandouin, e., lahijiani, h., shah-hosseini, m., ponel, p., salimian, m., & guiter, f. (2009). a late holocene pollen record from lake almalou in nw iran: evidence for changing land-use in relation to some historical events during the last 3700 years. journal of archaeological science, 36(7), 1364-1375. https://doi.org/10.1016/j.jas.2009.01.022. faegri, k., kaland, p. e., & krzywinski, j. (1989). textbook of pollen analysis. john wiley & sons, chichester. https://doi.org/10.2307/3038005. folk, r. l., & ward, w. c. (1957). brazos river bar: a study in the signification of grain size parameters. journal of sedimentary research, 27 (1), 3-26. https://doi.org/10.1306/74d70646-2b21-11d78648000102c1865d. gao, j. h., xu, x. a., jia, j. j., kettner, a. j., xing, f., wang, y. p., yang, y., qi, s. h., liao, f. q., li, j., bai, f. l., zou, x. q., & gao, s. (2015). a numerical investigation of freshwater and sediment discharge variations of poyang lake catchment, china over the last 1000 years. the holocene, 25(9), 1470-1482. https://doi.org/10.1177/0959683615585843. ge, q. s., fang, x. q., & zheng, j. y. (2014). learning from the historical impacts of climatic change in china. advances in earth science, 29(1), 23-29. https://doi.org/10.11867/j.issn.1001-8166.2014.01-0023. geng, r., zhao, y., cui, q., & qin, f. (2019). reprensentation of modern pollen assemblages with respect to vegetation and climate in northeast china. https://doi.org/10.1002/esp.261 https://doi.org/10.1002/esp.261 https://doi.org/10.1016/s0031-0182(01)00348-0 https://doi.org/10.1016/s0031-0182(01)00348-0 https://doi.org/10.1016/j.quascirev.2010.01.005 https://doi.org/10.1016/j.quascirev.2014.10.012 https://doi.org/10.1016/j.quascirev.2014.10.012 https://doi.org/10.13249/j.cnki.sgs.1998.03.219 https://doi.org/10.13249/j.cnki.sgs.1998.03.219 https://doi.org/10.3321/j.issn:1006-9267.2003.06.010 https://doi.org/10.3321/j.issn:1006-9267.2003.06.010 https://doi.org/10.1007/s12665-012-1716-x https://doi.org/10.1007/s12665-012-1716-x https://doi.org/10.18307/2017.0216 https://doi.org/10.1016/j.jas.2009.01.022 https://doi.org/10.2307/3038005 https://doi.org/10.1306/74d70646-2b21-11d7-8648000102c1865d https://doi.org/10.1306/74d70646-2b21-11d7-8648000102c1865d https://doi.org/10.1177/0959683615585843 https://doi.org/10.11867/j.issn.1001-8166.2014.01-0023 180 shuheng li, wei guo, na gao, jianfeng du quaternary international, 532, 126-137. https://doi.org/10.1016/j. quaint.2019.11.003. ghilardi, m., kunesch, s., styllas, m., & fouache, e. (2008). reconstruction of mid-holocene sedimentary environments in the central part of the thessaloniki plain (greece), based on microfaunal identification, magnetic susceptibility and grain-size analyses. geomorphology, 97(3-4), 617-630. https://doi.org/10.1016/j.geomorph.2007.09.007. grimm, e. c. (1987). coniss: a fortran 77 program for stratigraphically constrained cluster analysis by the method of incremental sum of squares. computers & geosciences, 13(13), 13–35. https://doi. org/10.1016/0098-3004(87)90022-7. guan, y., yu, g., han, r., xu, j., huo, d., & wang, q. (2019). changes in water area of gaoyou lake (including shaobo lake) and the influencing factors in the past 30 years. research of environmental sciences, 32(12), 2057-2064. doi: 10.13198/j.issn.1001-6929.2019.06.27 ming, g. d., zhou, w. j., wang h., shu p. x, cheng p., liu t. b., & zhou j. (2021). grain size variation in two lakes from margin of asian summer monsoon and its paleoclimate implications. palaeogeography, palaeoclimatology, palaeoecology, 567, 110295. https://doi.org/10.1016/j. palaeo.2021.110295. he, c. h., ding, h. y., zhang, z. k., shi, x. d., li, s. h., & mao, l. j. (2005). grain-size characteristics and their environmental significance of hongze lake sediment. scientia geographica sinica, 25(5), 590-596. https://doi. org/cnki:sun:dlkx.0.2005-05-012. herzschuh, u., kramer, a., mischke, s., & zhang, c. j. (2009). quantitative climate and vegetation trends since the late glacial on the northeastern tibetan plateau deduced from koucha lake pollen spectra. quaternary research, 71(2), 162-171. https://doi.org/10.1016/j. yqres.2008.09.003. jia, j. j., gao, j. h., liu, y. f., gao, s., & yang y. (2012). environmental changes in shamei lagoon, hainan island, china: interactions between natural processes and human activities. journal of asian earth sciences, 52, 158168. https://doi.org/10.1016/j.jseaes.2012.03.008. jiang, q. f., ji, j. f., shen, j., matsumoto, r., tong, g. b., qian, p., ren, x. m., & yan, d. z. (2013). holocene vegetational and climatic variation in westerly-dominated areas of central asia inferred from the sayram lake in northern xinjiang, china. china earth sciences, 56(3), 339-353. https://doi.org/10.1007/s11430-012-4550-9. kliem, p., enters, d., hahn, a., ohlendorf, c., lise-pronovost, a., st-onge, g., wastegard, s., & zolitschka, b. (2013). lithology, radiocarbon chronology and sedimentological interpretation of the lacustrine record from laguna potrok aike, southern patagonia. quaternary science reviews, 71, 54-69. https://doi.org/10.1016/j.quascirev.2012.07.019. lerman, a. (1979). lake: chemistry, geology, physics. science, 204(4395), 825826. https://doi.org/10.1126/science.204.4395.825. li, j. y., zhao, y., xu, q. h., zheng, z., lu, h. y., luo, y. l., li, y. c, li, c. h., & seppa, h. (2014). human influence as a potential source of bias in pollen-based quantitative climate reconstructions. quaternary science reviews, 99(9), 112-121. https://doi.org/10.1016/j.quascirev.2014.06.005. li, s. h., fu, g. h., guo, w., he, h. c., & zhang, z. k. (2007). environmental changes during modern period from the record of gaoyou lake sediments, jiangsu. journal of geographical sciences, 17(1), 62-72. https:// doi.org/10.1007/s11442-007-0062-5. li, s. h., guo, w., & mitchell, b. (2011). evaluation of water quality and management of hongze lake and gaoyou lake along the grand canal in eastern china. environmental monitoring and assessment, 176(1-4), 373–384. https://doi.org/10.1007/s10661-010-1590-5. li, s. h., guo, w., yin, y., jin, x. y., & tang, w. (2015). environmental changes inferred from lacustrine sediments and historical literature: a record from gaoyou lake, eastern china. quaternary international, 380, 350357. https://doi.org/10.1016/j.quaint.2015.01.010. li, y., wang, n., morrill, c., cheng, h. y., long, h., & zhao, q. (2009). environmental change implied by the relationship between pollen assemblages and grain-size in n.w. chinese lake sediments since the late glacial. review of palaeobotany & palynology, 154(1-4), 54-64. https://doi. org/10.1016/j.revpalbo.2008.12.005. li, y. y., zhou, l. p., & cui, h. t. (2008). pollen indicators of human activity. chinese science bulletin, 53(9), 1281-1293. https://doi.org/10.1007/ s11434-008-0181-0. liao, g. m. (1992). formation and evolution of the gaoyou lake, jiangsu province. acta geographica sinica, 47(2), 139–145. https://doi. org/10.11821/xb199202005. ling, s. (2001). palaeogeographic changes of the lixiahe district since the holocene. scientia geographica sinica, 21(5), 474–479. https://doi.org/10.3969/j.issn.1000-0690.2001.05.017. michael, m. e., zhang, z. h., rutherford, s., bradley, r. s., hughes, m. k., shindell, d., ammann, c., faluvegi, g., & ni, f. (2009). global signatures and dynamical origins of the little ice age and medieval climate anomaly. science, 326(5957), 1256-1260. https://doi.org/10.1126/ science.1177303. peng, y. b., xu, y., & jin, l. y. (2009). climate changes over eastern china during the last millennium in simulations and reconstructions. quaternary international, 208(1-2), 11-18. https://doi.org/10.1016/j. quaint.2009.02.013. peteet, d., nichols, j., pederson, d., kenna, t., chang, c., newton, b., & vincent, s. (2020). climate and anthropogenic controls on blue carbon sequestration in hudson river tidal marsh, piermont, new york. environmental research letters, 15(6), 065001. https://doi.org/10.1088/17489326/ab7a56. simonneau, a., chapron, e., vanniere, b., wirth, s. b., gilli, a., di giovanni, c., anselmetti, f. s., desmet, m., & magny, m. (2013). mass-movement and flood-induced deposits in lake ledro, southern alps, italy: implications for holocene palaeohydrology and natural hazards. climate of the past, 9(2), 825-840. https://doi.org/10.5194/cp-9-825-2013. sirocko, f., dietrich, s., veres, d., grootes, p. m., schaber-mohr, k., seelos, k., nadeau, m. j., kromer, b., rothacker, l., rohner, m., krbetschek, m., appleby, p., hambach, u., rolf, c., sudo, m., & grim, s. (2013). multi-proxy dating of holocene maar lakes and pleistocence dry maar sediments in the eifel, germany. quaternary science reviews, 62(62), 56-76. https://doi.org/10.1016/j.quascirev.2012.09.011. soon, w., bahunas, s., idso, c., idso, s., & legates, d. r. (2003). reconstructing climatic and environmental changes of the past 1000 years: reappraisal. energy and environment, 14, 233-296. https://doi. org/10.1260/095830503765184619. sun, l. q., liang, x., ma, b., li j., zhang, x., song, c., & li, l. (2021). characteristics of sedimentary environment since quaternary in northern jianghan basin, china: reconstruction of paleoenvironments. quaternary international, 589, 12-24. https://doi.org/10.1016/j.quaint.2021.03.034. sun, z. y., li, g., & yin, y. (2015). the yangtze river deposition in southern yellow sea during marine oxygen isotope stage 3 and its implications for sea-level changes. quaternary research, 83(1), 204-215. https://doi. org/10.1016/j.yqres.2014.08.008. sun. z., yuan, k., hou, x., ji, k., li, c., wang, m., & hou, j. (2020). centennial-scale interplay between the indian summer monsoon and the westerlies revealed from ngamring co, southern tibetan plateau. the holocene, 30(8), 1163-1173. https://doi. org/10.1177/0959683620913930. ter braak, c. j. f. (1988). canoco a fortran program for canonical community ordination by [partial] [detrended] [canonical] correspondence analysis, principal components analysis and redundancy analysis, version 2.1. agricultural mathematics group, wageningen, the netherlands. tong, x. f., xiao, x. y., yang, x. d., wang, s. m., & xiao, j. y. (2009). climatic changes and human activities revealed by pollen records in lake taibai, https://doi.org/10.1016/j.quaint.2019.11.003 https://doi.org/10.1016/j.quaint.2019.11.003 https://doi.org/10.1016/j.geomorph.2007.09.007 https://doi.org/10.1016/0098-3004(87)90022-7 https://doi.org/10.1016/0098-3004(87)90022-7 https://doi.org/10.1016/j.palaeo.2021.110295 https://doi.org/10.1016/j.palaeo.2021.110295 https://doi.org/10.1016/j.yqres.2008.09.003 https://doi.org/10.1016/j.yqres.2008.09.003 https://doi.org/10.1016/j.jseaes.2012.03.008 https://doi.org/10.1007/s11430-012-4550-9 https://doi.org/10.1007/s11430-012-4550-9 http://lib.cqu.edu.cn/asset/search?key=a%3dkliem%252c%2bpierre http://lib.cqu.edu.cn/asset/search?key=a%3denters%252c%2bdirk http://lib.cqu.edu.cn/asset/search?key=a%3dhahn%252c%2bannette http://lib.cqu.edu.cn/asset/search?key=a%3dohlendorf%252c%2bchristian http://lib.cqu.edu.cn/asset/search?key=a%3dzolitschka%252c%2bbernd https://doi.org/10.1016/j.quascirev.2012.07.019 https://doi.org/10.1126/science.204.4395.825 https://doi.org/10.1016/j.quascirev.2014.06.005 https://doi.org/10.1016/j.quascirev.2014.06.005 https://doi.org/10.1007/s11442-007-0062-5 https://doi.org/10.1007/s11442-007-0062-5 https://doi.org/10.1007/s10661-010-1590-5 https://doi.org/10.1016/j.quaint.2015.01.010 https://doi.org/10.1016/j.revpalbo.2008.12.005 https://doi.org/10.1016/j.revpalbo.2008.12.005 https://doi.org/10.1007/s11434-008-0181-0 https://doi.org/10.1007/s11434-008-0181-0 https://doi.org/10.11821/xb199202005 https://doi.org/10.11821/xb199202005 https://doi.org/10.3969/j.issn.1000-0690.2001.05.017 https://doi.org/10.3969/j.issn.1000-0690.2001.05.017 https://doi.org/10.1126/science.1177303 https://doi.org/10.1126/science.1177303 https://doi.org/10.1016/j.quaint.2009.02.013 https://doi.org/10.1016/j.quaint.2009.02.013 https://doi.org/10.1088/1748-9326/ab7a56 https://doi.org/10.1088/1748-9326/ab7a56 https://doi.org/10.5194/cp-9-825-2013 https://doi.org/10.1016/j.quascirev.2012.09.011 https://doi.org/10.1260/095830503765184619 https://doi.org/10.1260/095830503765184619 https://doi.org/10.1016/j.quaint.2021.03.034 https://doi.org/10.1016/j.yqres.2014.08.008 https://doi.org/10.1016/j.yqres.2014.08.008 https://doi.org/10.1177/0959683620913930 https://doi.org/10.1177/0959683620913930 181detecting climatic and hydrological changes in gaoyou lake, eastern china, over the last millennium using pollen, grain size, and lithology records hubei province, the middle and lower reaches of the yangtze river region over the past 1500 years. journal of lake sciences, 21(5), 732-740. https://doi.org/10.3321/j.issn:1003-5427.2009.05.019. wang, h., & yang, j. (1990). annals of gaoyou county. nanjing: jiangsu people’s press. p149. wang, s., & dou, h. (1998). annals of lakes in china. science press, beijing. wei, w. x., chen, r. m., wang l. f., & fu, l. x. (2017). spatial distribution of crustacean zooplankton in a large river-connected lake related to trophic status and fish. journal of limnology, 76(3), 546-554. https://doi. org/10.4081/jlimnol.2017.1622. xia, w. l., wang, y. f., & pan, h. x. (1995). modern sedimentation rate dating and environmental implication of nushan lake. journal of lake science, 7(4), 314-320. https://doi.org/cnki:sun:flkx.0.1995-04-003. xie, y. y., li, c. a., wang, q. l., & ying, h. f. (2008). palynological records of early human activities in holocene at jiangling area, hubei province. scientia geographica sinica, 28(2), 276-281. https://doi.org/10.3969/j.issn.1000-0690.2008.02.026. xu, j. x. (1993). a study of long term environmental effects of river regulation on the yellow river of china in historical perspective. geografiska annaler. series a, physical geography, 75(3), 61-72. https://doi. org/10.2307/521025. xu, q. h., li, y. c., yang, x. l., & zheng, z. h. (2007). quantitative relationship between pollen and vegetation in northern china. science in china series d: earth sciences, 50(4), 582-599. https://doi.org/10.1007/ s11430-007-2044-y. xu, q., yang, x., chen, w., meng, l., & wang, z. (1996). alluvial pollen on the north china plain. quaternary researh, 46, 270-280. https://doi. org/10.1006/qres.1996.0066. yang, b., brauning, a., johnson, k. r., & shi, y. f. (2002). general characteristics of temperature variation in china during the last two millennia. geophysical research letters, 29(9), 1324. https://doi.org/10.1029/ 2001gl014485. yang, d., & wang, y. f. (1995). on change of geographic environment and flood damage along the huaihe river basin during the last 2000 years. journal of lake science, 7(1), 1-7. https://doi.org/10.18307/1995.0101. zhang, c., zhou, a. f., zhang, x. n., wu, d., & hao, s. t. (2015). identification of paleaoflood events by lacustrine archives and their links to climatic conditions. progress in geography, 34(7): 898-908. https://doi. org/10.18306/dlkxjz.2015.07.011. zhang, l., cheng, y., niu, y., & jiang, j. (2020). analysis and prediction of eutrophication for advanced warning of the water quality concerns in gaoyou lake. water supply, 20(1), 186-196. https://doi.org/10.2166/ ws.2019.148. zhang, w. c., li, c. h., lu, h. y., tian, x. h., zhang, h. y, lei, f., & tang, l. y. (2014). relationship between surface pollen assemblages and vegetation in luonan basin, eastern qinling mountains, central china. journal of geographical sciences, 24(3), 427-455. https://doi.org/10.1007/ s11442-014-1098-y. zhang, y. h. (1990). the history of introduction and cultivation of ancient chinese maize. seed world, 40(12), 40. https://doi.org/ cnki:sun:sjzz.0.1990-12-042. zhang, z. k., wang, s. m., shen, j., wu, y., xia, w. l., & zhang, z. l. (1999). river channel changes recorded by lake sediments in nansihu lake, the lower reaches of the yellow river. journal of lake sciences, 11(3), 231236. https://doi.org/10.18307/1999.0307. zhao, y. t., miao, y. f., fang, y. m., li, y., lei, y., chen, x. m., dong, w. m., & an, c. b. (2021). investigation of factors affecting surface pollen assemblages in the balikun basin, central asia: implications for palaeoenvironmental reconstructions. ecological indicators, 123, 107332. https:// doi.org/10.1016/j.ecolind.2020.107332. zheng, j. y., wang, w. c., ge, q. s., man, z., & zhang, p. y. (2006). precipitation variability and extreme events in eastern china during the past 1500 years. terrestrial atmospheric & oceanic sciences, 17(3), 579-592. https://doi.org/10.3319/tao.2006.17.3.579(a). zheng, j. y., & wang, s. w. (2005). assessment on climate change in china for the last 2000 years. journal of geographical sciences, 60(1), 21-31. https://doi.org/10.3321/j.issn:0375-5444.2005.01.003. zhu, k. z. (1973). a preliminary study on the climatic fluctuations during the past 5000 years in china. science sinica, 16(2), 226-256. https://doi.org/10.3321/j.issn:1003-5427.2009.05.019 https://doi.org/10.4081/jlimnol.2017.1622 https://doi.org/10.4081/jlimnol.2017.1622 https://doi.org/cnki:sun:flkx.0.1995-04-003 https://doi.org/10.3969/j.issn.1000-0690.2008.02.026 https://doi.org/10.3969/j.issn.1000-0690.2008.02.026 https://doi.org/10.2307/521025 https://doi.org/10.2307/521025 https://doi.org/10.1007/s11430-007-2044-y https://doi.org/10.1007/s11430-007-2044-y https://doi.org/10.1006/qres.1996.0066 https://doi.org/10.1006/qres.1996.0066 https://doi.org/10.1029/2001gl014485 https://doi.org/10.1029/2001gl014485 https://doi.org/10.18307/1995.0101 https://doi.org/10.18306/dlkxjz.2015.07.011 https://doi.org/10.18306/dlkxjz.2015.07.011 https://doi.org/10.2166/ws.2019.148 https://doi.org/10.2166/ws.2019.148 https://doi.org/10.1007/s11442-014-1098-y https://doi.org/10.1007/s11442-014-1098-y https://doi.org/cnki:sun:sjzz.0.1990-12-042 https://doi.org/cnki:sun:sjzz.0.1990-12-042 https://doi.org/10.18307/1999.0307 https://doi.org/10.1016/j.ecolind.2020.107332 https://doi.org/10.1016/j.ecolind.2020.107332 https://doi.org/10.3319/tao.2006.17.3.579(a) https://doi.org/10.3319/tao.2006.17.3.579(a) https://doi.org/10.3321/j.issn:0375-5444.2005.01.003 https://doi.org/10.3321/j.issn:0375-5444.2005.01.003 ole_link15 ole_link7 ole_link14 ole_link5 ole_link6 _hlk106115164 ole_link62 ole_link63 ole_link61 ole_link1 ole_link2 ole_link3 under freeze-thaw cycles, the relationship between rock microstructure deterioration and its macroscopic mechanical characteristics has drawn extensive attention from engineers. with the objective to incorporate freeze-thaw cycle experiment into headrace tunnel engineering, in the present study two groups of andesite rock samples in different states are tested under the conditions of the lowest freezing temperature of –40 ºc and the thawing temperature of 20 ºc. damage detection was performed by magnetic resonance imaging for the interior microstructure of rock samples subject to different freeze-thaw cycles, and the relationship between the sample mechanical properties and gradual deterioration of rock microstructures was discussed. the results demonstrate evident influence of freeze-thaw cycle on the damage and deterioration of internal pore structure in andesite, and the rock uniaxial compressive strength and elasticity modulus exhibit a decreasing trend with the increase of freezethaw cycles. after 40 cycles, the strength of naturally saturated rock samples decreases by 39.4% (equivalent to 69.4 mpa) and the elasticity modulus drops by 47.46% (equivalent to 3.27 gpa). for rock samples saturated by vacuum, 40 freeze-thaw cycles lead to a decrease of 36.86% (equivalent to 58.2 mpa) in rock strength and a drop of 44.85% (equivalent to 2.83 gpa) in elasticity modulus. therefore, the test results quantitatively elucidate the substantial influence of freeze-thaw cycle on the damage and deterioration of internal structure in andesite. abstract keywords: freeze-thaw cycle; microstructure; progressive deterioration; strength damage the relationship between mechanical properties and gradual deterioration of microstructures of rock mass subject to freeze-thaw cycles issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n1.66108 haibo jiang* college of hydraulic and architectural engineering, shihezi university, xinjiang 832000, china. earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018): 53-57 en el ciclo de congelación y descongelación, la relación entre el deterioro de la microestructura de la roca y sus características mecánicas macroscópicas ha atraido un amplio interés de los investigadores. con el objetivo de incorporar las pruebas del ciclo congelación y descongelación del canal de un túnel, en este estudio se evaluaron dos grupos de muestras de andesitas en diferentes estados bajo las condiciones de congelación a -40 ºc y de descongelación a 20 ºc. la detección del daño se realizó por el escaneo de resonancia magnética en el interior de la microestructura de las muestras de roca sujetas a diferentes ciclos de congelación y descongelación, y se analizó la relación entre las propiedades mecánicas de las muestras y el deterioro gradual de las microestructuras de la roca. los resultados muestran una influencia evidente del ciclo de congelación y descongelación en el daño y deterioro de la estructura interna de los poros de la andesita y la fuerza compresiva uniaxial de la roca, mientras el módulo de elasticidad presenta una tendencia decreciente con el incremento de los ciclos aplicados. después de 40 ciclos, la fuerza de las muestras de roca saturadas naturalmente cae en un 39,4 % (el equivalente a 69,4 mpa) y el modulo de elasticidad baja en un 47,46 % (equivalente al 3,27 gpa). para las muestras saturadas por aspiradora, los cuarenta ciclos de congelación y descongelación llevan a un decrecimiento del 36,86 % (equivalente a 58,2 mpa) en la fuerza de la roca y una bajada del 44,85 % (equivalente a 2.83 gpa) en el módulo de elasticidad. además, los resultados de la evaluación cuantitativa esclarecen la influencia sustancial de los ciclos de congelación y descongelación en el daño y deterioro de la estructura interna de la andesita. resumen palabras clave: ciclo congelacióndescongelación; microestructura; deterioro progresivo; daño a la resistencia de la roca. record manuscript received: 05/07/2017 accepted for publication: 21/03/2018 how to cite item jiang, h. (2018). the relationship between mechanical properties and gradual deterioration of microstructures of rock mass subject to freeze-thaw cycles. earth sciences research journal, 22(1), 53-57 doi: http://dx.doi.org/10.15446/esrj.v22n1.66108 g e o p h y si c s 54 haibo jiang introduction the expansion of water during freezing process and its shrinking during thawing process inside rock mass is the primary factor for freezingthawing damage in underground engineering projects in cold area (wen, 2013). the temperature differences caused by the alternation of seasons induce repeating freeze-thaw cycles of water in rock mass, lead to enormous volume expansion force and result in the damage of rock mass (javier, 2013). the macroscopic brittle failure and strength damage of rock under freeze-thaw cycles are caused by the germination, expansion, growth and connection of many micro pore structures in the rock (rostasy, 1980; li, 2013). many laboratory experiments and detection results show that the damage of rock is caused by the continuous development of microscopic pore structure in rock mass, and rock mass experiences damages during the process in which the large pore volume increases and small pore volume decreases (mutlutürk et al, 2004). therefore, the detection and analysis of gradual deterioration progress of rock microstructures are of great significance to both the construction of correct macroscopic constitutive relation and the accurate measurement of rock strength characteristics (yan, 2015). the research on the relationship between the macroscopic mechanical characteristics and the gradual deterioration progress of rock mass microstructures under freeze-thaw cycles can greatly benefit and facilitate underground engineering projects. over the years many researchers developed the understanding through engineering practice that the effects of freeze-thaw cycle of water in rock pores change the microscopic pore structure of rock mass and cause variations in physical and mechanical properties, e.g. elastic modulus, strength and poisson’s ratio of rock mass. through a thermal expansion straining test under single freeze-thaw cycle, yamabe (2001) studied the relationship between the mechanical characteristics of rock and freeze-thaw cycle. the test proved that a freeze-thaw cycle causes the sample to deform elastically in axial direction and the saturated sample to deform plastically. they also found that there existed strong effect of freeze-thaw cycles on the damage of saturated rock. de argandona (1999) detected the damage of dolomite using x-ray, found serious weathering of rocks in a few short cycles, and assessed the damage according to the experimental results. nicholson (2000) performed the freeze-thaw cycle test for ten types of rock, and studied the impacts of lithology on the deterioration degree of rock during freeze-thaw cycles. zhang (2008) summarized the factors that cause the damage and deterioration of rock during freeze-thaw cycles, and studied the corresponding mechanism and mechanical properties of rock mass during freeze-thaw cycles. liu (2014) held the view that the elastic modulus of intact rocks decreases with the increase of freeze-thaw cycles, and through laboratory test they comparatively analyzed the freeze-thaw damage change rule of joint rock mass. the strength loss of limestone in cold regions subject to freeze-thaw recycles was studied by fatih (2012). from the studies above, the indoor freeze-thaw test demonstrates significant effects of freeze-thaw cycles on the macroscopic mechanical characteristic of the rock. the macroscopic strength decrease of rock mass has a close relation with the gradual deterioration of micro pore structure of rocks. there are many crisscross microscopic pore structures of various sizes within the rocks. the temperature differences caused by the alternation of seasons induce repeated freeze-thaw cycles of water in rock mass, and then lead to the germination, expansion, growth and connection of many micro pore structures in rocks, and ultimately lead to macroscopic brittle failure and even strength damage of rocks. under different hydration conditions, the relationships between damage and deterioration of microstructure and macroscopic mechanical characteristics of rock have drawn great attention from engineers. the recent research on rock freeze-thaw mainly concentrates on the laboratory test, but the combination of laboratory findings to practical engineering projects is still scarce. the freezethaw cycle test and the detection of micro pore structure of rock samples after the recycle were closely linked to headrace tunneling practice. in the present work the freeze-thaw tests were carried out for two groups of andesite sample rocks in different states under the conditions of the lowest freezing temperature of 40 ºc and the thawing temperature of 20 ºc. damage detection was performed by magnetic resonance imaging for the interior microstructure of rock samples subject to different freeze-thaw circles, and the relationship between the mechanical properties and the gradual deterioration of microstructures of rock mass was also discussed. the damage and deterioration mechanism of micro pore structure and macroscopic mechanical characteristic of rocks under different hydration conditions were analyzed. the results provide theoretical reference for the construction and engineering safety in cold regions. test of rock microstructure gradual deterioration subject to freeze-thaw cycles selection of samples table 1. physical mechanical property of rock samples. andesite samples were taken from a headrace tunnel project of a hydropower station in xinjiang and these sample lumps had dark gray surfaces and regular structures. most rock fracture had closed and fresh surfaces with a few with slight color-fade. occurrence was 270° ~ 285° ne <70° ~ 75° and < 275° ~ 285° ne <25° ~ 30°, smooth straight sliding surface, filled quartz vein and calcium. the physical mechanical properties of rock samples are shown in table 1. the headrace tunnel of the hydropower station is deeply buried, which allowed most of the tunnel surrounding rocks embedded in fresh rock mass. the onsite engineers tested the in-situ deformation of rock mass and obtained the field data of its mechanical characteristics, which provided references for the excavation of surrounding rocks and the design of lining structure. the deformation and elastic modulus of the tunnel rock in different states of stress were measured as shown in table 2. table 2. in-situ deformation of tunnel rock in the headrace hydropower station. note: the elastic modulus surface is located at a depth of 9.0 m in tunnel pd6. the rock was andesite with a smooth surface, and it was saturated for 7 days. the force on the elastic modulus surface was perpendicular to the bottom of tunnel. experiment program to understand the gradual deterioration progress of micro pore structure of rock mass under freeze-thaw cycles, the damage caused to rock samples was detected by magnetic resonance imaging. the results allowed us to understand how the micro pore structure of rock mass changes and to analyze the gradual deterioration of microstructure of rock mass under freezing-thawing cycles. the test procedure is stated as follows. the rock samples were generated by water drilling which drilled the cores along the direction perpendicular to the bedding plane of rock mass. after cutting and further processing, ten standard cylindrical rock mass samples were made which had a height-to-diameter ratio of 2 : 1, a diameter of 50 ± 2 mm and a height of 100 mm, according to the rock mechanics test regulations. in 55the relationship between mechanical properties and gradual deterioration of microstructures of rock mass subject to freeze-thaw cycles addition, the mass and size of the specimens were measured and each rock sample was visually inspected, as shown in figure 1. figure 1. the first group of andesite samples. (b) the samples were divided into two groups with five samples in each group. the first group (a1, a2, a3, a4, a5) were in natural saturation state, while the second group (b1, b2, b3, b4, b5) reached the saturation state with the aid of vacuum saturating device at 0.1 mpa vacuum pressure for 4 h of pumping. then both groups of samples were soaked in distilled water for 24 h. (c) based on the climate conditions at the construction site and the operating procedure of the freeze-thaw cycle test of rock mass, the freezing temperature was set at – 40 ºc and the freezing time was 4 hours. then the samples were thawed in water at temperature of 20 ºc for 4 hours. this constituted a freeze-thaw cycle, and the cycle was repeated. the two groups of rock samples were subjected to repeating freeze-thaw cycles, and data were acquired at the cycle number of 10, 20, 30 and 40. tds-300 freeze-thaw cycle machine was used for the repeating cycle test. (d) upon the completion of 10, 20, 30 and 40 cycles, the rock samples were measured by nuclear magnetic resonance relaxometry and nuclear magnetic resonance (nmr) system. the distribution curves of t2 spectrum of rock samples were obtained for two states and various freezethaw cycles. according to the area difference of sample t2 spectrum before and after the designated cycles, the rock porosity and its change rate were analyzed for different freeze-thaw cycles. (e) the freeze-thaw deterioration process of rock samples was observed and recorded, and the test results were analyzed. analysis of test results during the freezing process, the moisture in rock mass was frozen to ice. the accompanying volume expansion pushed and pressed the pore surface, which led to the extension of pore structure in rock and the generation of new pores (matsuoka, 2001). under saturation condition, pore structures in rock were filled with moisture. therefore the change of pore structure in rock mass can be characterized by the signal strength of water indicated by the nuclear magnetic resonance (nmr) system, and the relationship between nmr t2 spectrum distribution and pore size was established. nmr t2 spectrum also reflected the distribution of pores. in the t2 spectrum distribution, the peak position related to poresize, and the peak area related to the quantity of pore with corresponding aperture (li, 2012). the samples were imaged by nuclear magnetic resonance, and the transversal relaxation time of t2 spectrum distributions was obtained. the relaxation time of t2 is proportional to the pore diameter, while the signal strength reflects the quantity of pore with different apertures (zhou, 2014). by observing the changes in t2 spectrum distribution, the change of pore structure in rock can be analyzed. the t2 spectrum distributions of rock samples subject to different freeze-thaw cycles are shown in figures 2 and 3. as shown in figures. 2 and 3, after the freeze-thaw cycles, the t2 spectrum distribution of the rock samples exhibited two primary peaks. the morphological distribution of t2 spectrum shifted to the upper-left with the increase of freeze-thaw cycles, which indicates the enhancement in the t2 spectrum signal of small pores and the generation of new micro-pore structures in rock under the freeze-thaw cycles. therefore, the freeze-thaw cycles gradually deteriorated the rock microstructure. by analyzing the t2 spectrum distribution in figure 2 and 3, it was found that the first peak in the t2 spectrum distribution is significantly higher than the second for all the samples tested. the trend shows that new micropore structures were generated in rock samples. during freeze-thaw cycles, the moisture migrated to new pore structure, and the t2 spectrum signal strength was enhanced. the results show that under the freeze-thaw cycles, pore structures of all sizes expanded inside rock mass. not only were new pore structures generated, but also the original micro-pore structures extended, with a growth in the size of pore structure. this caused further damage to the rock microstructure. 0 50 100 150 200 250 300 350 400 450 500 0.01 0.1 1 10 100 1000 10000 relaxation time (t2) after freezing and thawing cycles/ms si gn al a m pl it ud e. without freezing-thawing cycles after 10 freezing-thawing cycles after 20 freezing-thawing cycles after 30 freezing-thawing cycles after 40 freezing-thawing cycles figure 2. t2 spectrum distribution of naturally saturated rock samples. 0 50 100 150 200 250 300 350 400 450 0.01 0.1 1 10 100 1000 10000 relaxation time (t2) after freezing and thawing cycles/ms si gn al am pl itu de . without freezing-thawing cycles after 10 freezing-thawing cycles after 20 freezing-thawing cycles after 30 freezing-thawing cycles after 40 freezing-thawing cycles figure 3. t2 spectrum distribution of rock samples saturated by vacuum. the peak of t2 spectrum distribution for naturally saturated samples was obviously higher than that of the samples saturated by vacuum. this indicates that naturally saturated rock samples had many micro-pore structures generated after the freeze-thaw cycles, and had faster generation and expansion speeds of fracture structure. these results explain the further deterioration of rock microstructure. table 3. the rock porosity and its change rate under different freezethaw cycles. 56 haibo jiang table 3 shows the rock sample porosity and its change rate after 10, 20, 30 and 40 freeze-thaw cycles. as shown in table 3, the porosity of the rock increased with the freeze-thaw cycles and there was small variation of amplitude among the samples of each group. for naturally saturated samples, their porosity increased by 42.56% ~ 72.72% after 20 freeze-thaw cycles, 145.45% ~ 103.81% after 30 cycles and 146.71% ~ 180.24% after 40 cycles. for samples saturated by vacuum, the porosity increased by 38.46% ~ 67.74% after 20 freeze-thaw cycles, 99.67% ~ 142.12% after 30 cycles and 149.0% ~ 188.97% after 40 cycles. it is noted that the number of freeze-thaw cycles has essential impact on the porosity of rock sample. the increase of freeze-thaw cycles resulted in faster generation and expansion speeds of fracture structure, as well as more evident change of porosity and gradual deterioration of rock microstructure. macroscopic mechanical characteristics of the rock mass subject to freeze-thaw cycles test method and scheme to establish the relationship between macroscopic mechanical characteristics and gradual deterioration of rock mass microstructure, the quality of rock samples subject to freeze-thaw cycles was measured by nuclear magnetic resonance (nmr) technique in order to investigate the rapid non-destructive detection characteristics. the samples subject to 0, 10, 20, 30, 40 cycles were tested for compression strength. the relationship between microscopic gradual deterioration and macroscopic mechanical characteristics of rock samples was analyzed for different freeze-thaw cycles. after the rock samples were detected by nuclear magnetic resonance (nmr), compression strength test and elastic modulus measurement were immediately performed using a universal qw-w300 microcomputercontrolled low-temperature electron test equipment. the data in each group were recorded and the results were analyzed systematically. the analysis of uniaxial compression characteristics table 4. strength and elastic modulus of andesite under different freezethaw cycles. due to the difference in lithology, mineral composition, porosity and water content, etc., the damage and degradation patterns in rocks subject to freeze-thaw cycles are also different. during the freeze-thaw cycles, the respective strength damage laws of rock also varied. table 4 lists the uniaxial compression strength and elastic modulus of andesite subject to different freeze-thaw cycles. figure 4 and figure 5 show the strength damage curves and elastic modulus curves of two groups of andesite under different freezethaw cycles. as shown in figure 4, with fewer freeze-thaw cycles, the compression strength of andesite is stronger. the increase of freezethaw cycles results in decreasing compression strength. the detection of microscopic pore structure of the rock samples uncovered that new microscopic pore structures were generated with the increase of freezethaw cycles. during freeze-thaw cycles, moistures migrated to new pore structures. continuous freeze-thaw cycles resulted in gradual deterioration of microstructure and larger compression strength of the rock samples. 0 50 100 150 200 0 10 20 30 40 cycles of freezing-thawingu n ia x ia l co m p re ss io n s tr en g th /m p a naturally saturated saturated by vacuum 0 2 4 6 8 0 10 20 30 40 cycles of freezing-thawing e la s ti c m o d u lu s /g p a naturally saturated saturated by vacuum figure 4. uniaxial compression strength of andesite. figure 5. elastic modulus of andesite. during the test, no pore structure change was observed after small number of freeze-thaw cycles. after 10 ~ 20 cycles, the original micropore structures began to extend. in the process of freeze-thaw cycles, the newly generated micro-pore structures were observed through nuclear magnetic resonance (nmr) detection. during freeze-thaw cycles, the damage kept deepening and the cracks extended with the increase of cycles. the test data and figure 4 show that, the strength of andesite rock samples decreased by different amount under various freeze-thaw cycles. the strength of naturally saturated rock samples decreased by 7.66% (equivalent to 13.5 mpa) after 10 cycles, 16.5% (equivalent to 29.1 mpa) after 20 cycles, 30.17% (equivalent to 53.2 mpa) after 30 cycles and 39.4% (equivalent to 69.4 mpa) after 40 cycles, respectively. for rock samples saturated by vacuum, the strength decreased by 7.03% (equivalent to 11.1 mpa) after 10 cycles, 14.88% (equivalent to 23.5 mpa) after 20 cycles, 25.84% (equivalent to 40.8 mpa) after 30 cycles and 36.86% (equivalent to 58.2 mpa) after 40 cycles. based on the test data and figure 5, the elastic modulus of naturally saturated rock samples decreased by 21.77% (equivalent to 1.5 gpa) after 10 cycles. the reduction continued for more cycles, which is 31.35% (equivalent to 2.16 gpa) after 20 cycles, 41.51% (equivalent to 2.86 gpa) after 30 cycles and 47.46% (equivalent to 3.27 gpa) after 40 cycles. for rock samples saturated by vacuum, the reductions of elastic modulus are 18.54% (equivalent to 1.17 gpa) after 10 cycles, 29.95% (equivalent to 1.89 gpa) after 20 cycles, 38.67% (equivalent to 2.44 gpa) after 30 cycles and 44.85% (equivalent to 2.83 gp) after 40 cycles, respectively. conclusions the macroscopic mechanical properties of andesite deteriorate gradually with the increase of freeze-thaw cycles. the deterioration mainly shows that, both the strength and the elastic modulus decrease gradually with the increase of freeze-thaw cycles. the macroscopic strength damage of rock mass is closely related to the gradual deterioration of its microstructure. the microscopic pore structures in rocks continuously develop and expand with the increase of freeze-thaw cycles, accompanied by the migration of moisture. continuous freeze-thaw cycles cause the gradual deterioration of microstructure and larger strength damage of the rock samples. (2) the gradual deterioration test of micro-structure of rock samples under freeze-thaw cycles suggests that the development and expansion speed of pores vary in different freeze-thaw stages. the nuclear magnetic resonance area of each rock sample increases with the freeze-thaw cycles. there are evident and distinct changes of nuclear magnetic resonance areas of all rock samples after 10, 20 and 30 cycles, which demonstrates different rates of freeze-thaw damage in different stages. (3) the influence of freeze-thaw cycles on the mechanical properties of andesite rock samples under different conditions is substantial. with the increase of freeze-thaw cycles, the uniaxial compression strength and elasticity modulus of rock present a clear reduction trend. the strength of 57the relationship between mechanical properties and gradual deterioration of microstructures of rock mass subject to freeze-thaw cycles naturally saturated rock samples decreased by 39.4% (equivalent to 69.4 mpa) and the elasticity modulus decreased by 47.46% (equivalent to 3.27 gpa) after 40 cycles. in contrast, the strength of rock samples saturated by vacuum decreased by 36.86% (equivalent to 58.2 mpa) and the elasticity modulus decreased by 44.85% (equivalent to 2.83 gpa) after 40 cycles. the results demonstrate evident influence of freeze-thaw cycle on the damage and deterioration of internal structure in andesite. it should be noted that the strength damage and even deterioration caused by freeze-thaw cycles in cold regions is a complicated process in rock engineering, and this is a complex thm multi-physics problem at low temperature. the phase change of pore water and ice in the rock mass is the key factor of rock freeze-thaw damage. by microscopic detection, the relationship between the macroscopic mechanical characteristics and the gradual deterioration of microstructure of rock mass under freezethaw cycles was investigated in this work. the study was based on the microscopic dimensions of the gradual degradation of rock microstructure under freeze-thaw cycles, and then extended to the effect of freeze-thaw cycles on macroscopic mechanical characteristics of rock mass. the present work contributed to discovering the damage mechanism of frozen rocks. the present work suggests that further study on rock freeze-thaw cycle damage should be based on the microscopic dimensions, extend to the effect of freeze-thaw cycles on the development of rock mass fracture network, and focus on the macro mechanical characteristics of rock. acknowledgements this research was financially supported by the national natural science foundation of china. (grant no. 51408377) and shihezi university outstanding young scholar training scheme (grant no. 2015zrkxjq06). references demirdag s. (2013). effects of freezing-thawing and thermal shock cycles on physical and mechanical properties of filled and unfilled travertines. construction and building materials, 47, 1395-1401. de argandona, v. g. r., rey, a. r. & celorio, c. (1999). characterization by computed x-ray tomography of the evolution of the pore structure of a dolomite rock during freeze-thaw cyclic tests. physics and chemistry of the earth, part a –solid earth geodesy, 24, 633637. fatih b. (2012). predicting mechanical strength loss of natural stones after freeze-thaw in cold regions. cold regions science and technology, 83-84, 98-102. javier, m. m., david, b., & miguel, g. h. (2013). non-linear decay of building stones during freeze-thaw weathering processes. construction and building materials, 38, 443-454. kubicar, l., vretenar, v., & bobac. (2006). thermophysical analysis of sandstone by pulse transient method. international journal of thermophysics, 27, 220-234. li, j., zhou, k., & zhang, y. (2012). experimental study of rock porous structure damage characteristics under condition of freezing-thawing cycles based on unclear magnetic resonance technique. chinese journal of rock mechanics and engineering, 31, 1208-1214. li, x., lu, y. & wang, y. (2013). research on damage model of single jointed rock masses under coupling action of freezethaw and loading. chinese journal of rock mechanics and engineering, 32, 23082315. liu, h., liu, y., & xing, c. (2014). test study of damage failure of jointed rock mass under freezing-thawing cycles. rock and soil mechanics, 35, 1547-1554. matsuoka, n. (2001). microgelivation versus macro-gelivation: towards bridging the gap between laboratory and eld frost weathering. permafrost and periglacial processes, 11, 299-313. mutlutürk, m., altindag, r. & türk, g. (2004). a decay function model for the integrity loss of rock when subjected to recurrent cycles of freezing-thawing and heating-cooling. international journal of rock mechanics & mining sciences, 41, 237-244. nicholson, h., dawn, t. & nicholson, f. (2000). physical deterioration of sedimentary rocks subjected to experimental freezing-thawing weathering. earth surface processes and landforms, 25, 1295-1308. rostasy, r. s., weib, r. & wiedemann, g. (1980). changes of pore structures of cement mortars due to temperature. cement and concrete research, 10, 157-164. wen, l., li, x., & yin, y. (2013). study of physico-mechanical properties of granite porphyry and limestone in slopes of open-pit metal mine under freezing-thawing cycles and their application. journal of glaciology and geocryology, 36, 632-639. yamabe, t. & neaupane, k. m. (2001). determination of some thermo mechanical properties of sirahama sandstone under subzero temperature conditions. international journal of rock mechanics & mining sciences, 38, 1029-1034. yan, x., liu, h., & xing, c. (2015). constitutive model research on freezingthawing damage of rock based on deformation and propagation of microcracks. rock and soil mechanics, 36, 3489-3499. yavuz, h. effect of freeze-thaw and thermal shock weathering on the physical and mechanical properties of an andesite stone. bulletin of engineering geology and the environment, 70, 187-192. zhang, j., miao, l. & yang, z. (2008). research on rock degradation and deterioration mechanisms and mechanical characteristics under cyclic freezing-thawing. chinese journal of rock mechanics and engineering, 27, 1688-1694. zhang, s. j., lai, y. m. & zhang, x. f. (2004). study on the damage propagation of surrounding rock from a cold–region tunnel under freeze-thaw cycle condition. tunneling and underground space technology, 19, 295-302. zhou, k. p., hu, z. x. & li, j. l. (2014). study of marble damage evolution laws under unloading conditions based on nuclear magnetic resonance technique. chinese journal of rock mechanics and engineering, 33, 3523-3530. earth sciences research journal eart sci. res. j. vol. 18, no. 1 (june, 2014): 74 letter from editor welcome to earth sciences research journal, volume 18, number 1. more than 20 years have passed since we decided to be part of the scientific world in geology, seismology, geophysics, geodesy, meteorology, oceanography, planetary sciences, and so many other fields. we raised our voice to say to the world that in colombia, there is a journal committed to providing the highest level of manuscript in earth sciences. it has been not an easy road, but we are here. this 2014 has been full of challenges moving us to find the best ways to present new manuscripts. our volume 18 was made possible by the non-profit cooperation of authors and reviewers working hand-inhand with us. in this volume, we are presenting to the eyes of scientific world nine papers coming from three different continents, with contributions by people from seven countries. our first paper comes from egypt, about an aeromagnetic data analysis in the region of fayoum-cairo. this work regarding the october 12, 1992 earthquake, suggests how the new tectonics of africa’s northeast is predominant in this area. from turkey, our second manuscript obtained a new, important, more up-to-date and more reliable relationship between surface wave magnitude and rupture length for earthquakes in that country. the results obtained in this study will give some useful and up-to-date information to authors whose studies are focused in this area. in the field of hydrogeology, the third article analyzes the hydrogeochemistry of shallow groundwater in the yellow river alluvial plain. the results suggest that the shallow groundwater in the study area is fresh to moderately mineralized water. the assessment of samples shows how this water is acceptable for agricultural use but is not potable. from china, too, an interesting paper about the petrography and mineralogy of biotites within the wushan granodiorite is presented. the formation conditions of these biotites are also determined and the significance of these minerals in terms of petrogenesis and mineralization is discussed. from india, where climate change is a topic trend, a study attempts to provide information on climate variability and its impacts on runoff in the kosasthaliyar sub-basin. from nigeria, a work related to aromatic biological markers are used for assessing thermal maturity of source rocks in the campano-maastrichtian mamu formation. results of this study show how these maturity values indicate that the coal and shale extracts are marginally mature for hydrocarbon generation. we invite you to read also three south american papers. from colombia, the manuscripts titled “enhancing time resolution by stabilized inverse filter and q estimated on instantaneous spectra” and “hydrocarbon production scenarios in colombia. review of field sizes, hydrocarbon reserves and expectations of conventional and unconventional resources” are presented. from chile comes the paper “uncertainty in rainfall input data in a conceptual water balance model: effects on outputs and implications for predictability”. thanks to every one of you and enjoy this issue of the esrj. carlos a. vargas. editor earth sci. res. j. vol. 11, no. 2 (december 2007): 108-116 accurate gravity anomaly interpolation: a case-study in cameroon, central africa j. kamguia1, c. t. tabod2, j. m. tadjou1, e. manguelle-dicoum2, r. nouayou2 and l. h. kande1 1 national institute of cartography (nic) – cameroon, po box 157, yaounde – cameroon 2 department of physics, faculty of science, university of yaounde – cameroon corresponding author: joseph kamguia: kjerryfr@yahoo.fr; j.kamguia@geomaticsystem.com; 0023799440426 abstract many treatments in geodesy and geophysics require regularly gridded gravity anomalies. the gridding of gravity data needs interpolation. for the predicted data to be accurate, the smoothest type of gravity anomaly should be used along with the most indicated prediction method. this paper presents the comparison of various prediction methods applied on different types of gravity anomalies and considering the relative geological complexity of the study area. many algorithms are tested and the suitability of each type of anomaly and each prediction method discussed in a case-study in cameroon (central africa), using a set of 43,000 gravity data points to determine the must accurate prediction technique. key words: prediction, gravity anomaly, residual free-air anomalies, global geopotential model, complex geology, geophysics. resumen muchos tratamientos en geodesia y geofísica requieren que las anomalías gravimétricas se encuentren en grillas regulares. esta disposición de los datos de gravedad requiere interpolación. para que los datos predichos sean exactos, la anomalía gravimétrica más fina debe ser utilizada junto con el método más indicado para predicción. este artículo presenta la comparación de varios métodos de predicción aplicados a diversos tipos de anomalías gravimétricas y considerando la relativa complejidad geológica del área del estudio. se probaron una gran cantidad de algoritmos y la conveniencia de cada tipo de de anomalía y método de predicción fueron discutidos en una aplicación en camerún (áfrica central), usando un conjunto de 43.000 datos gravimétricos para determinar la técnica de predicción más adecuada.. 108 earth sciences research journal manuscript received july 15 2007. accepted for publication december 5 2007. palabras claves: predicción, anomalía gravimétrivca, anomalías residuales libres de aire, modelo geopotencial global, geología compleja, geofísica. introduction interpolation is a process widely used in earth science. it estimates the value of a parameter at a point from neighbourings. the process may be extrapolation in some areas (heiskanen and moritz, 1967). in geodesy, gravity anomalies are interpolated when computing geoid undulations or the quasi-geoid height anomalies. in geophysics, prediction is used to compute regular grids of gravity anomalies required for presentation purposes or further treatments. the accuracy of the predicted gravity anomalies depends mainly on their smoothness and on the method used. it also depends the data density, their geographical distribution and the grid intervals. in general, the smoother the gravity anomalies, the more accurate the predicted values. many interpolation methods are now proposed (el abbas et al., 1990), with some algorithms available in open sources such as the generic mapping tools or gmt (wessel and smith, 1995). comparisons have already been made between surface fitting algorithms, in order to check their reliability in predicting bouguer gravity anomalies (el abbas et al., 1990). however, many types of gravity anomalies are routinely used in geodesy and in geophysics. it may sometimes be difficult to determine which type is a priori suited for prediction. a constant crustal density value generally used in different computations might be too far from the reality below the earth topographic surface in complex geological areas. residual anomalies that are deduced from free-air (faa), simple bouguer (ba), complete bouguer (cba) and isostatic (ia) anomalies might also be smoother than the original anomalies in some areas. these anomalies are now easily computed using the spherical harmonic coefficients of a global geopotential model (ggm) (kamguia et al., 2007). this paper presents the results of comparison of various types of gravity anomaly predicted with some methods commonly and in different areas of cameroon. these anomalies were computed from a homogeneous database containing 43,000 data points. the aim is to give some hints as to the most appropriate gravity anomaly prediction strategy according to the density of the gravity net, their geographical distribution and the relative complexity of the geology of the study area. to achieve this, four areas with different geological characteristics are chosen. in these areas of different extension and gravity net density, six prediction algorithms are tested: the minimum curvature splines in tension (smith and wessel, 1990); the least square polynomial fitting (two variants); krigging (krige, 1978) and the inverse distance to a power method (two variants). 2. gravity anomaly prediction 2.1. characteristics of gravity anomalies different types of gravity anomaly are used in earth sciences. gravity data recorded on land must be adjusted for elevation above or below sea level to yield the free-air anomalies (faa). they are nearly equivalent to what would be observed if all the topographic masses were condensed onto this reference surface (blakely, 1996). simple and complete bouguer gravity anomalies (ba and cba) are obtained by completely removing the masses that exist between the level of observation and sea level. they are in theory smoother and then more accurate gravity anomaly interpolation: a case-study in cameroon, central africa 109 kamguia et al. esrj vol. 11, no. 2. december 2007 110 suitable for prediction than the faa (heiskanen and moritz, 1967). however, in complex geological contexts, the gravity field may have special characteristics. therefore, ba and cba may be rougher than faa. the isostatic reduction regularizes the earth’s crust according to a model of isostasy. depending on the complexity of the geology of the study area, these anomalies may be less or more suited in prediction processes than faa, ba and cba. from spherical harmonic coefficients of global geopotential models (ggm), the long wavelength components of faa, ba, cba and ia are easily computed. after subtraction from respective total field gravity anomalies, the residual anomalies obtained may be smoother in some areas. the long wavelength anomalies ( )ggmg∆ are computed from the harmonic coefficients of the ggm using equation (1) and subtracted from the total field gravity anomalies )( g∆ (heiskanen and moritz, 1967): (1) where gm is the geocentric gravitational constant; ),,( qlr the spherical coordinates of the computation point; g the normal gravity on the reference ellipsoid; a the equatorial radius of the earth; (sin )n mp q the fully normalized associated legendre functions for degree n and order m ; mn c∆ and mn s∆ the normalized egm-ggm harmonic coefficients, reduced for the even zonal harmonic for the ellipsoid and complete to degree and order 360max =n . 2.2. prediction of gravity anomalies and statistical analysis the gravity is a regionalized field, since it varies from place to place in a continuous manner but with no possibility of associating a specified mathematical function. the most accurate gravity anomaly prediction technique is the one that maintains the broad and fine features of the original gravity data processed, without introducing undue distortions. therefore, the predicted and the measured values must at the same point should be mathematically and/or physically related. to achieve this, the criterion of crain and bhattacharyya (1967) is important. data point grid node with a reference point of gridded data (i) input data in gridding algorithms (ii) gridded cells waiting for gridded data at nodes   (i) (ii) figure 1. choice of reference points of gridded data   max 2 2 0 ( 1) cos sin (sin ) nn n ggm n m n m n m n m gm a g n c m s m p r r                to apply the criterion, some data points must be selected to serve as reference points (fig. 1). a reference point is a data point with coordinates nearly coinciding with those of a grid node. the predicted and known values are therefore comparable in a simple statistical analysis of their differences. 3. a case-study in cameroon (central africa) 3.1. characteristics of the gravity field in cameroon free-air gravity anomalies are smoother than ba and cba in cameroon, as is indicated in table 1. in this table, the standard deviation (std) of the ba and cba are nearly 10 mgal greater than the faa’s. this means that the gravity field in cameroon possesses special features. one can therefore conclude that: ● a constant density )67.2( =r used in the bouguer reduction does not effectively suppress the effect of the topography above the reference surface in cameroon; ● the disturbing geological masses responsible of the observed gravity anomalies cannot be easily determined here; ● the geology of cameroon is more complex than it looks like. in this context, interpolation of gravity anomalies may be subject to many errors if the more indicated method and especially the suitable type of gravity anomaly are not used. table 1: statistical comparison of bouguer and free-air anomalies in cameroon (unit: mgal) four test areas ),,,( 4321 zzzz (fig. 2) were selected in order to precise the special features of the gravity field and its behaviour in prediction processes. their selection is based on the differences in geological characteristics, the density and distribution of the data points, and the relative roughness of the topography relief of the area. the principal characteristics of these areas are summarized in table 2. table 2: characteristics of the four test areas the first test area 1z is located in the northern sedimentary region of cameroon. the area is characterized by a long wavelength relatively positive bouguer anomaly attributed by adighije (1981), and fairhead and okereke (1987) to an uplift of the moho. the topographic relief has a mean orthometric height of about 1000 m. gravity data are more denser in this test area. the second test area 2z is situated on the north eastern extension of the cameroon volcanic line (cvl). it is characterised by large negative bouguer anomalies oriented e-w from western cameroon to western c. a. r. these are attributed to a light mass anomaly structure in the upper mantle beneath the adamawa uplift (fairhead and okereke, 1987; poudjom et al., 1992). the mean elevation is greater than 1000 m. the gravity data density is the smallest in this test area. 3z is located on the continental part of the cvl. this line is associated with a series of nesw trending gravity lows not well defined by the reconnaissance survey of collignon (1968). the mean height of the area is nearly 2000 m, with a maximum of 4010 m for mount-cameroon. the gravity data density is intermediate to those of 1z and 2z . the area 4z is situated in southern cameroon. the mean height is lower than in the 111 anomaly max. min. mean std variability free-air 315,5 –85,4 –0,03 25,0 smoother simple bouguer 253,4 –125,5 –26,5 34,4 rougher complete bouguer 254,2 –121,4 –25,5 33,6 rougher number oftest area points used control points data mean density 1z  ee  1613  ;  nn  118  2179 36 1 point per 25 km 2z  ee  1513  ;  nn  85  672 16 1 point per 2100 km 3z  ee  119  ;  nn  74  1465 9 1 point per 250 km 4z  ee  1412  ;  nn  52  987 16 1 point per 2100 km accurate gravity anomaly interpolation: a case-study in cameroon, central africa 112 previous test areas and the gravity data density is comparable to what is observed in 3z . the distributions of data in each test area are shown in figures 3 to 6. the reference points are well distributed in 1z . the numbers of these points in the test areas are enough for an informative analysis of predicted gravity anomalies. table 3 shows the statistics of the anomalies to be interpolated in the test areas. from the analysis based on the std, one can see that the total field bouguer and isostatic anomalies are not always the smoothest. the gravity field has special characteristics in cameroon and its surroundings. the high frequency components of this field, created by the topography relief, are not entirely removed after the bouguer and isostatic reductions. by smoothing faa, ba, cba and ia with subtraction of the longer wavelength components computed using equation (1), four residual types of anomaly specified by (res) in table 3 are computed. table 3: statistics based on std of different types of anomaly in the test areas (unit: mgal) the differences between the measured and interpolated anomaly at the reference points are computed in each area and for each type of anomaly. a statistical comparison is made. the analysis is made on the standard deviations (std) of the differences between anomalies. where the std are small, the interpolation technique is said to be most accurate. the ggm used in residual anomaly computations is egmggm, the most representative of gravity data in the central african subregion (kamguia et al., 2007). the residual free-air anomalies faa(res) are smoother than all the other types of anomaly, except in area 2z where the total gravity bouguer anomalies (ba and cba) are the smoothest (std = 13 mgal). the difference between the std of the residual anomalies is nearly 1 mgal. the std of residual bouguer and isostatic anomalies are greater than the std of total gravity bouguer and isostatic anomalies and than the std of residual free-air anomalies in some areas. in conclusion, cba and ba are the smoothest anomalies in 2z while faa(res) is the smoothest in 1z , 3z and 4z . faa(res) is therefore smoothest type of anomaly in cameroon. it is the only residual anomaly used in the analysis made below. 3.2. comparison of prediction results the faa(res) were treated with the total gravity free-air, bouguer (simple and complete) and isostatic anomalies with the above interpolation methods (and their variants). table 4 shows the results of statically analysis based on std of the differences between predicted and known data at the reference points. table 4: summary of the std of interpolation results in test areas (unit = mgal) from table 4, one can conclude that the accuracy of predicted anomalies depends both maximum minimum mean standard deviation anomaly z1 z2 z3 z4 z1 z2 z3 z4 z1 z2 z3 z4 z1 z2 z3 z4 faa 71 89 155 67 -30 -60 -40 -30 2,4 26 25 9,9 13 22 32 20 ba 3,7 -46 105 -12 -73 -126 -119 -105 -41 -90 -37 -62 13 13 42 22 cba 4,3 -43 106 -9 -72 -121 -95 -102 -39 -83 -30 -59 12 13 41 20 ia 68 85 153 65 -30 -63 -44 -32 0,9 23 22 7,7 13 22 39 20 faa (res) 56 57 87 27 -34 -84 -84 -37 -1,6 5,8 -6 -1,2 11 19 29 11 ba (res) 3 -55 34 -40 -94 -163 -187 -109 -44 -110 -68 -73 15 20 58 13 cba (res) 4 -52 34 -37 -86 -153 -172 -106 -43 -102 -60 -70 14 18 54 12 ia (res) 53 53 85 26 -36 -87 -87 -39 -3,1 2,4 -8,4 -3,5 12 20 30 12 kamguia et al. esrj vol. 11, no. 2. december 2007 faa ba cba ia faa (res) methods z1 z2 z3 z4 z1 z2 z3 z4 z1 z2 z3 z4 z1 z2 z3 z4 z1 z2 z3 z4 minimum curvature 4 6 7 2 3 4 3 3 4 4 3 3 4 6 7 2 3 6 3 1 polynomial quadratic 17 27 29 16 12 12 13 19 12 13 16 19 17 26 29 16 10 26 11 13 polynomial cubic 16 27 27 15 11 12 13 18 11 13 15 19 17 27 27 15 10 26 11 12 krigging 3 3 6 2 3 3 2 4 3 3 2 4 3 3 6 2 2 3 2 1 inverse distance (1/r2) 5 5 7 2 4 4 4 4 4 4 4 4 5 5 7 3 3 5 2 1 inverse distance (1/r3) 4 4 7 2 3 4 3 4 3 3 3 4 4 4 7 2 3 4 2 1 113 longitude la tit ud e z1 niger cameroon nigeria chad c. a. r. gabon p. r. congo e. guinea atlantic ocean z2 z4 z3 study area lac chad figure 2. location of the four test areas 13 14 15 16 8 9 10 11 figure 3. data and control point distributions in + = data point ; ● = reference point accurate gravity anomaly interpolation: a case-study in cameroon, central africa 114 kamguia et al. esrj vol. 11, no. 2. december 2007 13 14 15 5 6 7 8 figure 4. data and control point distributions in + = data point ; ● = reference point 9 10 11 4 5 6 7 figure 5. data and control point distributions in + = data point ; ● = reference point 115 accurate gravity anomaly interpolation: a case-study in cameroon, central africa on the interpolation method, the variability of the anomaly used and the test area considered. however, the choice of the type of anomaly to interpolate is more pertinent than the method and the area. the smoother the anomaly interpolated, the less are the interpolation errors, represented here by the std. for the same type of anomaly interpolated, the krigging and the inverse distance to power 3 methods gave the most accurate results, with approximately the same precision. the best precisions are 2 mgal, 3 mgal, 2 mgal and 1 mgal in test areas ),,,( 4321 zzzz respectively. hence, when the best gridding method is used in conjunction with the smoothest anomalies, a precision of about 2 mgal is obtained after interpolation. 4discussions and conclusion accurate interpolated gravity anomalies are needed in geodetical and geophysical analyses. these anomalies are obtained if the interpolation method and the type of anomaly used are carefully selected. theoretically, bouguer gravity anomalies are smoother and therefore suited for prediction processes than free-air anomalies. however, this is only true in an area with simple geology where a constant crustal density of 67.2 is nearest to the real value. for specific geological, geophysical and geodetical studies in such areas, if bouguer anomalies are interpolated, the information deduced from the gravity interpretations (such as subsurface modelling, basin exploration, geoid computation …) will be biased. residual free-air anomalies can be smoother than total gravity free-air, bouguer and isostatic anomalies in many areas. they are easily computed using the spherical harmonic coefficients of a global geopotential model. after reconstruction of the total gravity free-air anomalies from interpolated residual anomalies, any type of gravity anomaly may be deduced. these residual free-air anomalies are associated 11 12 13 14 2 3 4 5 figure 6. data and control point distributions in + = data point ; ● = reference point 116 kamguia et al. esrj vol. 11, no. 2. december 2007 to smaller errors in gravity anomaly prediction in cameroon. the krigging method and the inverse distance to power 3 performed approximately equally as best prediction methods. the geology of cameroon seems more complicated and only a digital density model may introduce adjusted crustal density values into bouguer gravity reductions. the number of interpolation methods tested is very limited, just for clarity reasons. if the data were very dense in cameroon, many grid intervals would have been tested. acknowledgments the authors would like to thank the following individuals and organizations: mr. sylvain bonvalot of ird (institut de recherche pour le developpement), for giving the ird gravity data during the computation of the first geoid model of cameroon, and mr. bernard langellier of bgi (bureau de gravimetrie international) for making these data accessible. references adighije, c. i., 1981. a gravity interpretation of the benue trough, nigeria. tectonophysics, 79, pp. 109-128. blakely, j. r., 1996. potential theory in gravity and magnetic applications. cambridge university press. 441 pages. collignon, f., 1968. gravimetrie de reconnaissance, cameroun, rapport orstom, 35 p. crain, i. k. and bhattacharyya, b. k., 1967. treatment of non-equispaced two dimensional data with a digital computer. geoexploration, vol 5, pp. 173-194. el abbas, t., jallouli, c., albouy, y. and diament, m., 1990. a comparison of surface fitting algorithms for geophysical data. terra-nova, vol. 2, n° 5, pp. 467-475. fairhead, j. d. and okereke, c. s., 1987. a regional gravity study of the west african rift system in nigeria and cameroon and its tectonic interpretation. tectonophysics, 143, pp. 141-159. heiskanen, w. a. and moritz, h., 1967. physical geodesy. freeman, sanfrancisco, 364 pages. kamguia, j., manguelle-dicoum, e., tadjou, j. m., tabod, c. t. and kande, h. l., 2007. the first geoid model of cameroon (central africa). accepted for publication in the nordic journal of surveying and real estate research (njsr). krige, d. g., 1978. geostatistics for ore evaluation. south african institute of mining and metallurgy, johannesburg, south africa. 50 pp. poudjom-djomani, y. h., diament, m. and albouy, y., 1992. mechanical behaviour of the lithosphere beneath the adamawa uplift (cameroon, west africa) based on gravity data. journal of african earth sciences, 15 (1), pp. 81-90. smith, w. h. f. and wessel, p., 1990. gridding with continuous curvature splines in tension. geophysics, vol. 55, n°3, pp. 293-305. wessel, p. and smith w. h. f., 1995. new version of gmt released: transactions of the american geophysical union 72 (441), pp. 445-446. articulo 5.indd earth sciences research journal archaeology earth sci. res. sj. vol. 15, no. 2 (december, 2011): 115 120 quaternary dating by electron spin resonance (esr) applied to human tooth enamel eduar carvajal1, luis montes2 and ovidio a. almanza1 1 departamento de física, universidad nacional de colombia, cra 30 nº 45-03, bogotá 2 departamento de geociencias, universidad nacional de colombia. cra 30 nº 45-03, bogotá. corresponding author: ovidio almanza m, tel. (57 1) 3165000 ext 13048; fax (57 1) 3165635. e-mail oaalmanzam@unal.edu.co abstract this paper presents the results obtained from using electron paramagnetic resonance (epr) to analyse tooth enamel found at the aguazuque archaeological site (cundinamarca, colombia), located on the savannah near bogota at 4° 37’ north and 74°17’ west. it was presumed that the tooth enamel came from a collective burial consisting of 23 people, involving men, women and children. the tooth enamel was irradiated with gamma rays and the resulting free radicals were measured using an electron spin resonance (esr) x-band spectrometer to obtain a signal intensity compared to absorbed doses curve. fitting this curve allowed the mean archaeological dose accumulated in the enamel during the period that it was buried to be estimated, giving a 2.10 ± 0.14 gy value. rosy software was used for estimating age, giving a mean 3,256 ± 190y before present (bp) age. these results highlight epr’s potential when using the quaternary ancient ruins dating technique in colombia and its use with other kinds of samples like stalagmites, calcite, mollusc shells and reefs. resumen en este trabajo se presentan los resultados obtenidos por datación mediante el uso de resonancia paramagnética electrónica en muestras de esmalte dental provenientes del sitio arqueológico aguazuque, ubicado en la sabana de bogotá en coordenadas 4° 37’ de latitud norte y 74°17’ de longitud oeste de greenwich. se presume que el esmalte dental proviene de un entierro colectivo compuesto de 23 individuos entre mujeres, hombres y niños. el esmalte dental fue irradiado con rayos gamma y los radicales libres producidos fueron medidos usando un espectrómetro de resonancia del spin electrónico (esr), banda x, con el fin de realizar una curva de intensidad de la señal versus la dosis absorbida por el esmalte. el ajuste de esta curva permitió obtener la dosis media arqueológica acumulada en el esmalte (d a), durante el periodo de enterramiento de la muestra. el valor obtenido fue de 2.10 ± 0.14 gy. para estimar la edad de la muestra se usó el software rosy, obteniendo una edad media de las muestras de esmalte de 3,256 ± 190y bp. los resultados obtenidos muestran el potencial uso de la técnica epr como herramienta de datación de muestras cuaternarias en colombia y abre la posibilidad de usarla en otro tipo de muestras como estalagmitas, calcitas, conchas de moluscos y corales. palabras claves: epr, esmalte dental, datación arqueológica. keywords: epr/esr, tooth enamel, colombia, archaeological dating. record manuscript received: 11/05/2011 accepted for publications: 30/11/2011 introduction electron spin resonance (esr) analysis has recently become an alternative c14 and thermo-luminescence dating method which can be applied to a variety of problems in geology, archaeology and paleoanthropology (renfrew c et al., 1992; jonas 1997; walker 2005; grun 1989). esr spectroscopy is the only method for detecting, identifying and quantifying free radicals (halliwell & gutteridge 2007; punchard & kelly 1996). for instance, this method is used for estimating the free radical level and paramagnetic centres produced in some materials by ionising radiation. the some free radicals’ long mean life (above 100 my) make it possible to reveal them in materials as old as 2 my without the risk of overcoming the sample’s saturation limits. esr allows some materials to be dated because free radical concentration is a measurement of the total radiation dose absorbed by the sample during the time it was exposed to radiation and hence their burial time (jonas 1977). esr has been used for dating the formation and more recent re-crystallisation of three types of gypsum samples: massive, bedded and fracture filling gypsum; in this case, the estimated upper miocene–pliocene age agreed with that soil’s stratigraphy (ülkü 2004). molodkov (2001) used esr to date eight terrestrial shell samples from treugolnaya (triangular) cave (northern caucasus), ages ranging from 583,000 to 393,000y old. shigueo et al., (2003) dated a calcite formation found on a rock wall painting at toca da bastiana, rock shelter in serra da capivara national park, piaui, brazil, obtaining an age of 35,000 to 43,000y, indicating that humans lived there before 35,000 years ago (this being the date for the first humans to arrive in brazil). otheduar carvajal, luis montes and ovidio a. almanza116 er papers dating stalagmites, calcite, mollusc shells and reefs are renfrew et al., (1992), jonas (1997), walker (2005) and grun (1989). the present study was aimed at using the esr method for dating tooth enamel from the aguazuque archaeological site in soacha, colombia. theoretical aspect the aguazuque archaeological site (cundinamarca, colombia) is located on the savannah near bogota at 40 37’ north and 74°17’ west (figure 1). the human tooth used in this work was extracted from a skeleton (labelled ag-i) taken from a collective burial site where 23 people were found arranged in a foetal position; men, women and children were identified. the burial site was circular shaped (4.5 m diameter) and as burnt bones and lacerated skeletons were found outside the circle, partially painted red and black, it was deduced that cannibal practices were common by the group responsible for the burial (correal 1990). archaeological research concerning the early and middle holocene period has led to figure 1. location of aguazuque archaeological site (cundinamarca, colombia, 40 37' north and 74°17' west). (taken from groot, a.m., 1992). figure 2. energy levels of an electron placed in a magnetic field. the arrow shows the transitions induced by microwave radiation. reconstructing the way of life for human groups in the region, thereby showing a transition from hunters and gatherers to agricultural societies (groot 1992). a way of life based on hunting several sized mammals has been recorded during a lapse of 11,000 to 5,000 years before today, with rock shelters providing favourites housing places, such as the archaeological sites of el abra, tequendama and nemocón 4 in cundinamarca (correal 1979). sites such as chía i (ardila 1984) and aguazuque (correal 1990) have been investigated and the aforementioned research has led to establishing changes in the form of settlement and adaptation mechanisms (groot, 1992). electron spin resonance (esr) consists of the resonant absorption of electromagnetic energy during electron-spin transitions. a static magnetic field should be applied to resolve different electron-spin levels. unpaired electrons from free radicals have spin equal to ½. there are two magnetic levels (+½ and –½) in a magnetic field, both having different energy (figure 2); the transition between these two levels is possible in the following resonance condition: hν=gβh where ν is resonance frequency, h is plank’s constant, g is the g-factor (which is a constant = 2 for ½ spin), β is the bohr magneton which is an electron magnetic dipole moment and h is the magnetic field. an important conclusion derived from this formula is the linear dependence between the applied magnetic field and resonance frequency. the most frequently used microwave energies drop in what is called the x band. this is because it is a good compromise between sensitivity, sample size and water content effects; ν ≈ 9.8 ghz and h ≈ 350 mt for the x band. esr dating and dosimetry are based on the fact that ionising radiation produces paramagnetic centres having long lifetimes in a number of materials. such centres’ concentration in a given sample is a measure of the total radiation dose to which a particular sample has been exposed. this effect can be used to determine the length of time of such exposure and hence, in many cases, a sample’s burial time (jonas 1997). esr has been used with tooth enamel for determining the age of palaeolithic archaeological sites, mainly because tooth enamel is ubiquitous in these contexts and because it allows dating beyond the 14c dating time range (40–45 ka max.). it is thus also very useful in important sites regarding human evolution studies, sites having palaeontological interest within a quaternary time frame and because many sites lack the volcanic materials needed for k – ar and 40ar/39ar dating (rink 1997). (1) quaternary dating by electron spin resonance (esr) applied to human tooth enamel 117 the aim is to determine the accumulated dose (da) to which a sample has been exposed whilst being buried; the exposure time is then determined by comparing d a to the average dose rate over such period: t= da (d) however, no significant difference in the accuracy of dose reconstruction has yet been found using different calibration methods (wieser et al., 2000). each sample’s radiation sensitivity is individually calibrated for dose assessment by the additive dose method; each sample is incrementally irradiated with at least 14 additional doses up to 100 gy or more (avoiding the sample becoming saturated). a linear regression analysis is applied to the esr measurements at each dose increment and the original measurement (laboratory-un-irradiated sample). the original d a in the sample is obtained from the negative intercept of the regression line with the dose axis; is estimated as being the total radiation provided by all radioactive elements present in a sample and soil and also cosmic radiation (ikeya 1993; jonas 1997; walker 2005; grun 1989). materials and methods the dentine was initially removed from teeth using a dental drill and water cooling. the enamel was then placed in a 30% naoh solution for one day to disinfect it and separate any remaining dentine (iaea report 2002); 1.4 mm mean enamel thickness was obtained for ag-i. the sample was air-dried at room temperature for three days; a dental drill was used to strip around 50 ± 5 μm from the inside and outside of the enamel surface to ensure that alpha radiation had no effect. the molar enamel powder was divided into nine aliquots. each aliquot was placed inside a plastic capsule for irradiation using 60c standard radiation equipment at room temperature and 8gy/min mean dose rate; the doses ranged from 0 to 740 gy. after irradiation, the esr signal was measured with a bruker esp300 x-band spectrometer. the g  and g  signals’ peak to peak height was used for calculating amplitude. the spectrometer parameters used were: 336 mt central field, 0.02 t scan range, 0.1 m t amplitude modulation, 100 khz modulation frequency, 0.328 s time constant and 20 mw microwave power. the aliquots were then irradiated with additional doses and the same process was repeated until getting 9 data points from esr signal amplitude as a function of γ-ray dose. soil samples were also collected from the aguazuque site and sent for u, th and k content analysis by gamma spectrometry. radiometric analysis is widely used for determining natural radioisotopes in geological samples by means of spectroscopic methods and, especially, for quantifying the uranium, thorium and potassium present in samples. these samples were analysed at ingeominas’ nuclear technology group’s environmental radiometrics’ laboratory which has a low-level gamma spectrometry system with 15 cm lead shielding and high-resolution gehp hyperpure germanium detector (canberra model gc2019), having 1.89% absolute detection efficiency for 661.6 kev. results and discussion figure 3 shows the esr spectra for enamel sample ag-i having an irradiation dose ranging from 54 to 324 gy. a linear model was used for fitting esr signal intensity to dose, as has been done by other authors (baffa et al., 2000; hefne et al., 2002; kinoshita et al., 2001a, 2008b; mascarenhas et al., 1982); very good agreement with linear fit was shown (figure 4). the archaeological d a obtained using this method was da = 2.10 ± 0.14 gy. rosy software (brennan et al., 1999) was used for calculating the enamel sample’s age, based on the given d a, cosmic radiation dose rate (determined at 251 ± 15 μgy/year), u, th and k content obtained from figure 3. esr spectra of tooth enamel ag-i in a range of 54 to 324 gy. (2) eduar carvajal, luis montes and ovidio a. almanza118 soil sample analysis and enamel (1.4 mm ag-i thickness and 5 mm dentine thickness) in sample. soil water content was 12.1%. the software’s default values were used for the remaining parameters. the cosmic radiation dose rate was obtained using the equation for luminescence and esr dating (prescott & hutton 1994). finding the dose rate to convert d a into an age is a crucial step after da has been found. rosy software allows age to be calculated from an enamel sample by taking da due to radioisotopes from adjacent layers into account. the software computes age based on three uranium absorption models (linear, early uptake and a combination of both). table 1 shows the sample concentration (ppm) 238u 232th 40k aguazuque 1.86 ± 0.19 3.72 ± 0.37 1.66 ± 0.09 alpha beta gamma cosmic gamma+cosmic dentine 0.00 0.00 enamel 0.00 0.00 cementum 0.00 0.00 sediment 0.00 46.55 347.31 uranium 0.00 34.54 188.50 thorium 0.00 12.01 158.81 potassium 0.00 0.00 total 0.00 46.55 347.31 251.00 598.31 (alpha+beta) dose rate (µgy/y) = 46.55 total dose rate (µgy/y) = 644.86 table 1. uranium, thorium and potassium concentration as determined by gamma spectrometry. table 2. dose rate obtained from radioisotope concentration, local cosmic radiation and total dose rate. alpha radiation from sediment was not used in the calculation, as about 50 μm was removed from the outer side of the tooth. figure 4. signal versus dose growth showing the linear fitting used. archaeological dose for ag-i was ad = 2.91 ± 0.14 gy; correlation coefficient was 0.9937. soil analysis results. the present work did not determine u, th and k concentration in enamel and dentine as their percentages have been negligible for calculating age in previous work, especially in dates for the holocene period (kinoshita et al., 2008; hefne et al., 2002). typical soil density, enamel and dentine values were used for calculating age (2.00, 3.00 and 2.82 g/cm3, respectively). the layer of enamel stripped off from each side by acid etching was about 50±5 mm; this prevented a possible contribution towards the dose having been made by alpha particles deposited in the sample. using these values and the radioisotope concentrations listed in table 1, rosy software was used for obtaining the alpha, beta and gamma radiation dose rate contribution for each radioisotope in sediment (the results are listed in table 2). these results were the same for any uranium absorption model. considering a 251 ± 15 μgy/year dose rate for cosmic radiation, a 3,256 ± 190y before present (bp) age was thus found for the ag-i sample. it should be noted that rosy software calculates these ages using three uranium accumulation models: linear, early and a combination of both (ikeya, 1993; grun, 1989; fattibene et al., 2010). the way that uranium accumulates in a tooth is established by these models; accumulation increases at the same rate as time in the linear model and accumulation occurs in a short period of time in the early model, relative to a tooth’s age and then remains constant after this short period of time. the tooth’s age was the same when calculated by these three methods; this result was consistent if taken into account that the accumulation of radioactive elements inside the tooth was negligible. conclusions age determined at many countries’ archaeological sites is the result of radiocarbon (14c) analysis of particular items found around the burials and not direct dating of individuals found there. the results regarding either physical or geological phenomena may not be chronologically equal, hence the importance of having techniques like esr which can directly quaternary dating by electron spin resonance (esr) applied to human tooth enamel 119 and absolutely determine the age of individuals of interest. the age of a human burial site found in colombia was determined in this work from a single tooth, a mean age of 3,256 ± 190y bp being found. this result agreed with stratigraphic analysis at sites providing soil ages ranging from 2,800 to 3,500y bp for 50 and 80 cm depths (correal, 1990). esr dating accuracy was obtained relatively straightforwardly, using just a few grams of tooth enamel, thereby highlighting some of the advantages and characteristics of the esr dating method. the results so obtained have contributed towards joining links in the long chain of colombian and american history (where there is still much to do, due to the rich pre-hispanic and colonial past). this paper thus represents a pioneering effort aimed at promoting the dating of our ancestors. acknowledgments we would like to thank the dib, universidad nacional de colombia, for providing financial support (project 8003348), m. peña from ingeominas for soil sediment analysis, the universidad nacional de colombia’s physical anthropology laboratory and its director, j.v. rodriguez, for providing facilities and allowing the use of his samples, as well as am groot (archaeologist) for her support during this research. references a.f. borgonovea, a. kinoshitaa,b, f. chena, p. nicoluccia, o. baffaa, (2007): energy dependence of different materials in esr dosimetry for clinical x-ray 10mv beam. radiation measurements, 42: 1227– 1232. anatoly molodkov; (2001): esr dating evidence for early man at a lower palaeolithic cave-site in the northern caucasus as derived from terrestrial mollusc shells; quaternary science reviews 20: 1051 -1055 ardila, g.i, (1984): chía, un sitio precerámico en la sabana de bogotá. fundación de investigaciones arqueológicas nacionales. banco de la república. bogotá d.c. baffa, o., brunetti, a., karmann, i., neto, c.m.d., (2000): esr dating of a toxodon tooth from a brazilian karstic cave. appl. rad. isot., 52: 1345-1349. brennan, b.j., rink, w.j., rule, e.m., schwarcz, h.p., prestwich, w.v., (1999): the rosy esr dating program, ancient. tl. 17: 45-53. correal, g., (1979): investigaciones arqueológicas en abrigos rocosos de nemocón y sueva. fundación de investigaciones arqueológicas nacionales. banco de la república. bogotá d.c. correal, g., (1990): aguazuque : evidencias de cazadores, recolectores y plantadores en la altiplanicie de la cordillera oriental. fundación de investigaciones arqueológicas nacionales, banco de la república. bogotá d.c. groot, a.m., (1992): checua. una secuencia cultural entre 8500 y 3000 años antes del presente. fundación de investigaciones arqueológicas nacionales. banco de la república. bogotá. grün, r., (2006): direct dating of human fossils. yearb. phys. anthropol. 49: 2-48. halliwell b. and gutteridge j. 2007. free radicals in biology and medicine. oxford university press, usa. hefne, j., yamani, a., al-dayel, o., ikeya, m., al-osaimi, s., (2002): esr dating of tooth from pre-islamic sitin saudi arabia. advances in esr applications. 18: 119-121. iaea report., (2002): use of electron paramagnetic resonance dosimetry with tooth enamel for retrospective dose assessment. report of a coordinated research project. iaea-tecdoc-1331. vienna. ikeya, m., (1993): new applications of electron spin resonance: dating, dosimetry and microscopy, world scientific publishing company, singapore. jonas, m., (1997): concepts and methods of esr dating. radiat. meas. 27: 943-973. kinoshita a., braga f.j.h.n., graeff c.f.o., baffa o., (2001): esr dosimetry of 89srand 153sm-in bone. appl. rad. isot. 54: 269-274. kinoshita, a., figueiredo, a.m.g., felice, g.d., lage, m.c.s.m., guidon, n. baffa, o., (2008): electron spin resonance dating of human teeth from toca da santa shelter of são raimundo nonato, piauí, brazil. nucl. instrum. meth b. 266: 635-639. mascarenhas s., baffa filho o., ikeya m., (1982): electron spin resonance dating of human bones from brazilian shell-mounds (sambaquis). am. j. phys. anthropol. 59, 413-417. prescott, j.r., hutton, j.t., (1994): cosmic ray contributions to doserates for luminescense and esr dating: large depths and long-term time variations. radiat. meas. 23, 497-500. punchard n. and kelly f. 1996. free radicals: a practical approach. oxford university press, usa. renfrew, c. y bahn, p. arqueología. teorías, métodos y práctica (1992): ediciones akal s.a. pp 107147, rink, w.j., (1997): electron spin resonance (esr) dating and esr applications in quaternary science and archaeometry. radiat. meas. 27: 975-1025. shigueo watanabe, walter elias feria ayta and henrique hamaguchi; (2003): some evidence of a date of first humans to arrive in brazil. journal of archaeological science 30: 351–354. ülkü ulusoy; (2004): esr studies of anatolian gypsum; spectrochimica acta part a 60: 1359–1365. walker m. (2005): quaternary dating methods. john wiley & sons ltd, the atrium, southern gate, chichester. chapters 1, 2, 3 and 4. wieser, a., mehta, k., amira, s., aragno, d., bercea, s., brik, a.,bugai, a., callens f., chumak v., ciesielski b., debuyst r., dubovsky s., duliu o.g., fattibene p., haskell e.h., hayes r.b., ignatiev e.a., ivannikov a., kirillov v., kleschenko e., nakamura n., nather m., nowak j., onori s., pass b., pivovarov s., romanyukha a., scherbina o., shames a.i., sholom, s., skvortsov v., stepanenko v., tikounov d.d., toyoda s. (2000): the second international intercomparison on epr tooth dosimetry. radiat. meas, 32, 549–557. beryl crystals are found within stream sediments transecting schists in the northeast of eskişehir, western anatolia. this paper studied the eskişehir beryl crystals with optical microscopy, scanning electron microscopy (sem-edx), infrared spectroscopy (ir) and geochemical analyses. beryl is accompanied by garnet, glaucophane, quartz, epidote, muscovite and chlorite in the stream sediments. the crystals are euhedral emerald (green gem beryl) and light bluishgreen aquamarine, with ideal sharp ir bands. wet chemical analysis of eskişehir beryl yielded 61.28% sio2, 15.13% al2o3, 12.34% beo, 0.18% cr2o3, 1.49% mgo, 1.69% na2o, 0.98% fe2o3, and 0.008% v2o3, resulting in the formula (al1.75cr0.01mg0.22fe0.08)(be2.90si6.00)(na0.32)o18. large ion lithophile elements (lile) (barium, strontium), some transition metals (cobalt, except nickel) and high field strength elements (hfse) (niobium, zirconium, and yttrium) in stream sediments that are associated with beryl exhibited low content about metamorphic rocks. beryl formation appears to be controlled by upthrust faults and fractures that juxtaposed them with cr-bearing ophiolitic units and a regime of metasomatic reactions. such beryl crystals have also been found in detrital sediments that are derived from the schists. cristales de berilio fueron encontrados en sedimentos de corrientes que atraviesan en esquistos en el noreste de eskisehir, al oeste de anatolia. en este artículo se presentan resultados sobre el estudio de los cristales de berilio de eskisehir con microscopio electrónico de barrido (sem-edx, del inglés scanning electron microscopy), espectroscopia infrarroja y análisis geoquímicos. el berilio estaba acompañado de granate, glaucofana, cuarzo, epidota, moscovita, y clorito en las corrientes sedmientarias. los cristales son esmeraldas de formas definidas (gema verde de berilio) y aguamarinas color verde celeste, con bandas de espectroscopia infrarroja de buena nitidez. el análisis químico húmedo del berilio de eskisehir mostró 61.28 % de sio2, 15.13 % de al2o3, 12.34 % de beo, 0.18 % de cr2o3, 1.49 % de mgo, 1.69 % de na2o, 0.98 % de fe2o3, y 0.008% de v2o3, lo que resulta en la formula (al1.75cr0.01mg0.22fe0.08)(be2.90si6.00)(na0.32)o18. los elementos litófilos de ion grande (bario, estroncio), algunos metales de transición (cobalto, excepto níquel) y los elementos de gran campo de fuerza (niobio, circonio e itrio) en las corrientes de sedimentos que están con berilio mostraron un bajo contenido sobre las rocas metamórficas. las formaciones de berilio aparecen controladas por fallas de cabalgamiento y fracturas que se yuxtaponen con las unidades ofiolíticas relacionadas con cromo y un régimen de reacciones metasomáticas. estos cristales de berilio se han encontrado también en sedimentos detríticos que se derivan de los esquistos. earth sciences research journal earth sci. res. j. vol. 20, no. 3 (september, 2016 ) : a1 a10 abstract resumen keywords: beryl, kaymaz, schist, sem-edx, ir. palabras clave: berilio, kaymaz, esquistos, microscopio electrónico de barrido, espectroscopia infrarroja. occurrence of cr-bearing beryl in stream sediment from eskişehir, nw turkey record manuscript received: 04/12/2014 accepted for publication: 26/08/2016 how to cite item erkoyun, h., & kadir, s. (2016). occurrence of cr-bearing beryl in stream sediment from eskişehir, nw turkey. earth sciences research journal, 20(3), a1-a10 doi:http://dx.doi.org/10.15446/esrj.v20n3.47677 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n3.47677 hülya erkoyun and selahattin kadir * eskişehir osmangazi university, deparment of geological engineering, tr−26480 eskişehir, turkey herkoyun@ogu.edu.tr skadir.euroclay@gmail.com *corresponding author o r e d e p o s it s presencia de berilios relacionados con cromo en corrientes sedimentarias de eskisehir, noroeste de turquía hülya erkoyun and selahattin kadir a2 introduction the study area is in the village of halilbağı, approximately 72 km northeast of the eskişehir, western anatolia (fig.1). the area was previously explored to understand its structural geology and stratigraphy (erentöz, 1975; bingöl, 1976, kulaksız, 1981; gözler et al., 1996). groat et al. (2008) and arif et al. (2010) has classified the genetic types of beryl deposits. worldwide, beryl deposits occur 1) in association with granitic pegmatite and hydrothermal veins in mafic-ultramafic rocks, with or without schist in contact zones (kazmi and snee, 1989; sinkankas, 1994; dereppe et al., 2000; schwarz and giuliani, 2001; sabot, 2002). emerald deposits in poona (australia) occur in quartz and pegmatite veins related to metasomatic reactions between quartz-muscovite or quartz-topaz-greisens and ultramafic units (groat et al., 2008); 2) in schist-type deposits that are controlled by thrust faults and shear zones, with or without pegmatites (sinkankas, 1994; schwarz and giuliani, 2001; zwaan, 2006). the emeralds from santa terezinha (brazil) formed in metasomatized mica schist related to hydrothermal fluids along thrust fault associated with pegmatite veins (giuliani et al., 1997a,b); 3) related to black shale with veins and breccias controlled by thrusts and faults (beus, 1979; dereppe et al., 2000; schwarz and giuliani, 2001). the colombian (cordillera basin) emerald is hosted in the lower cretaceous sediments (sandstone, limestones, black shales and evaporates) associated with stratiform breccias and mobilized black shales. firstly, calcite, pyrite and muscovite filling veins related to hydrothermal fluids developed along faults. secondly, extensional faulting and folding resulted in the formation of emeralds in voids accompanied by fluorite, dolomite, and quartz as breccia filling of defects (giuliani et al., 2000; pignatelli et al., 2015). 4) about oceanic suture zones (dereppe et al., 2000). the emerald occurrences in panjshir valley (afghanistan) associated with continental suture zone within intrusion of tourmaline-bearing leucogranite into serpentinite units that result of phlogopite bands containing jewels (sabot et al., 2000); 5) in carbonate rocks (sabot, 2002). the swat valley (pakistan) emeralds occur as veins and stockworks within fractured quartz-bearing magnesite rocks and chromium-rich tourmaline in limonitized shear zones (schwarz and giuliani, 2002); 6) in granitic cupolas (dereppe et al., 2000). the emeralds from nigeria formed in pegmatite having contact with alkaline granite bedrock where the chromium is originated from basement schists or younger volcanic units (schwarz et al., 1996); and 7) rarely, in alluvial-fluvial placer deposits (dill, 2010). the gem deposits (beryl, chrysoberyl, amethyst, corundum, etc.) of the ratnapura (sri lanka) developed by transportation from gem-bearing source rocks and deposition in old stream terraces and flood plains such as well-rounded gravel or lenses of sand (dissanayake et al., 2000). there has been no prior research on the chromium-rich beryl [be3al2(sio3)6] in the schists from the halilbağı area (which are in tectonic contact with ophiolites), although prime ministry archives (1993) reported that emeralds (a green gem variety of beryl) were discovered and mined in the sivrihisar area during the ottoman empire. therefore, the beryl crystals in proximal metamorphic rocks prompted the present study, with emphases on geochemistry, the genetic relationship between the beryl and metamorphic rock, and the mineralogy and geochemistry of beryl crystals that occur in sediments derived from the metamorphic rocks. figure 1. geological map of the study area (modified from gültekin et al., 2003; sarıfakıoğlu et al., 2009). occurrence of cr-bearing beryl in stream sediment from eskişehir, nw turkey a3 figure 2. photomicrographs showing views of a) garnet-glaucophane schist (plane-polarised light) (h1); b) garnet-lawsonite schist (plane-polarised light) (h6); c) epidote-chlorite schist (plane-polarised light) (h2); d) epidotite (plane-polarised light) (h7). geological setting the basement of the study area consists of triassic phyllites, phyllitic schists and partially metamorphosed clastic rocks (metaarkose, metagraywacke) (kahya and kuşcu, 2014)(fig. 1). triassic metasediments are overlain unconformably by jurassic conglomerates and sandstones and cretaceous sivrihisar massif blueschists (garnet-glaucophane schist, lawsoniteglaucophane schist, epidote-chlorite schist, epidotite, and chlorite-lawsonite schist) (davis and whitney, 2006). the blueschist facies comprise of sodic amphibole + garnet + phengite + lawsonite and/or epidote ± omphacite ± quartz characterizing a broad range of pressure-temperature conditions (12-16 kbar, 380500 °c) (okay, 2002). these units are tectonically overlain by upper cretaceous ophiolitic rocks (harzburgites, dunites, wehrlites, pyroxenites, gabbros, and serpentinites). these units are overlain by the oligo-miocene rocks comprising silicified and altered trachytic tuffs as well as fresh intrusive alkaline trachytes and phonolites. 40ar/39ar geochronological analyses of sanidine crystals from the phonolites have yielded an age of 25 ma (late oligocene-early miocene) for the alkaline volcanic rocks (sarıfakıoğlu et al., 2009). temel (2001) concluded that the alkali basaltic and trachytic rocks of the sivrihisar (eskişehir) area are products of post-collisional miocene alkaline volcanism in an extensional tectonic regime. pliocene and quaternary sediments overlie these units. structural features indicate that the study area is affected by major and minor e-wand n-s-trending faults that have been under the tectonic influence of the north anatolian fault zone (şengör, 1979; şengör and yılmaz, 1981; gültekin et al., 2003). materials and methods twenty-nine sediment samples and twenty-nine rock samples were collected along creek valleys that transect the metamorphites, phonolites, trachytes, and pegmatites (outside the study area) in compliance with the principles of geochemical prospecting. transparent and opaque minerals were separated from sediment samples using a stereomicroscope (nikon smz-1b), and these minerals were grouped based on color and crystal form. transparent, colored minerals considered to be beryl were separated, and the refractive indices (r.i.) of these minerals were determined under polarized light using 0.002 r.i.-spaced immersion oils; other optical properties were also measured. the samples were analyzed for their mineralogical characteristics by polarized light microscopy (leitz laborlux 11 pol), x-ray powder diffractometry (xrd) (rigaku-geigerflex), and scanning electron microscopy (sem-edx, jeol jsm 84a-edx). xrd analyses were performed at 40 kv and 30 ma with cukα radiation; slits with divergence 1°, scatter 0.15°, receiving 1° and monochrometer 0.30 mm; and a scanning speed of 1° 2θ/min. un-oriented mounts of four powdered sediment samples were scanned to determine the mineralogy of bulk samples. representative beryl crystal samples were prepared for sem-edx analyses by adhering each beryl sample onto an aluminum sample holder that had been covered with double-sided tape and coated with a thin film (~ 350 å) of gold using a giko ion coater at eskișehir osmangazi university. infrared spectroscopy was performed using pressed pellets of powdered beryl crystal samples (2 mg of < 2 µm) mixed with 200 mg kbr and a perkinelmer 100 ft-ir spectrometer; scans were made at 4 cm-1 resolution. data were collected over the range 400-4000 cm-1. major element analyses and trace element analyses of four metamorphic sediment samples and one metamorphic rock sample were performed using the inductively coupled plasma atomic emission spectroscopy (icp-aes) method at acme laboratory ltd. (canada) to ascertain their beryllium contents. the detection limits for these analyses were between 0.01 and 0.1 weight percent for major elements, 0.1 and 0.5 ppm for trace elements, and 0.01 and 0.1 ppm for rare earth elements. the detection limit for beryllium was 1 ppm. hülya erkoyun and selahattin kadir a4 figure 3. various views of study-area beryl collected from stream sediments: a) stereomicroscopic view of crystals of aquamarine (a) on the left and emerald (z) on the right; b) emerald crystal viewed using a stereomicroscope; c) view of a beryl grain in an immersion fluid with an r.i. of 1.60; d) emerald crystal (plane-polarized light); e) basal cleavage (0001) in aquamarine (plane-polarized light). occurrence of cr-bearing beryl in stream sediment from eskişehir, nw turkey a5 one emerald crystal was analyzed for all major and trace elements using both inductively coupled plasma – optical emission spectroscopy (icp-oes) and inductively coupled plasma – mass spectroscopy (icp-ms) methods in the analytical services laboratory of the university of greenwich (uk). the sample was analyzed by icpoes and icp-ms after dissolution using a lithium metaborate fusion (jarvis and jarvis, 1992). these methods are accredited to iso 17025, by ukas. in essence, 0.25 g of each sample was mixed with 1.25 g of lithium metaborate flux and fused at 900 °c. the molten bead was poured figure 4. sem images of a) beryl crystals collected from stream sediments; b) a close-up view of a hexagonal prismatic beryl crystal. figure 5. infrared spectroscopy spectra for an eskişehir beryl crystal. figure 6. aluminum versus the sum of mg+cr+fe-site cations of octahedral beryl in atoms per formula unit; from beryl analyses from the literature (leitmeier, 1937; abdalla and mohamed, 1999; groat et al., 2002; seifert et al., 2004; gavrilenko et al., 2006). into weak nitric acid and stirred until dissolved. the resultant solution was made to volume. the analysis of major elements and some trace elements was undertaken using a thermo icap 6500 icp-oes with calibration via matrix-matched synthetic standards. certified reference materials and internal qc powders were included within the analytical batch. analysis of the remaining trace elements and rare earth elements was completed using a thermo x series 2 icp-ms, again with calibration via matrix-matched synthetic standards and the use of crm s and internal qc powders. data quality was assessed post analysis via the use of shewhart charts. hülya erkoyun and selahattin kadir a6 table 2. chemical compositions of ophiolitic units, metamorphic sediments, and rocks that are associated with beryl in the study area. table 1. mineralogical variation in stream sediments of the study area. acc: accessory; +: relative abundance of mineral. occurrence of cr-bearing beryl in stream sediment from eskişehir, nw turkey a7 table 3. the chemical composition of an eskişehir octahedral emerald in schist compared to another emerald in metamorphic rocks of similar nature. hülya erkoyun and selahattin kadir a8 results petrography and mineralogy the metamorphic units comprise garnet-glaucophane schist (lepidoporphyroblastic texture), epidote-chlorite schist (lepidoblastic texture), garnet-lawsonite-glaucophane schist and epidotite (nematoblastic texture) (fig. 2ad). these units consist of glaucophane, garnet, lawsonite, muscovite, epidote, chlorite, augite, quartz, sphene and zircon. locally, garnet crystal exhibit chloritization and carbonatization developed in microfracture. the results of the stereomicroscopic and xrd analyses of stream sediments showed lawsonite, quartz, garnet, augite, glaucophane, muscovite, epidote and sphene. beryl is associated with garnet, glaucophane, quartz, epidote, muscovite and chlorite in the stream sediment. sediment samples from the kaymaz stream, which transects the metamorphic rocks, were defined as emerald and aquamarine by transparency and color, with crystal sizes between 200 µm and 1 mm (e.g., h1 sample) (fig. 3). the examination of green crystals using a stereomicroscope revealed that the hexagonal morphology (fig. 3a-b); the refractive indices of these grains were found to be no= 1.568 and ne= 1.584 (fig. 3c). based on optical studies of the emeralds (fig. 3d) and light bluish green aquamarine (fig. 3e) using polarized light and immersion oils, the analyzed grains were determined to be uniaxial (-), euhedral hexagonal-prismatic crystals. stereomicroscopically determined beryl crystals were also analyzed by sem (fig. 4a-b); these crystals exhibit euhedral hexagonal prismatic forms, with locally corroded crystal edges. infrared spectra bands at 3594, 1634 and 1602 cm-1 correspond to the characteristic stretching vibrations of oh and h2o molecules in the channel coordinated to be2+ cations in tetrahedral coordination and al3+ cations in octahedral coordination along the c-axis direction of the beryl crystal (łodziński et al., 2005; makreski and jovanovski, 2009) (fig. 5). 1184, 1144, 952, 749, 519 and 441 cm-1 bands are characteristic of the si-o stretching vibration (farmer, 1974; łodziński et al., 2005). bands at 808, 749, 680 and 650 cm-1 correspond to be-o stretching, and bands at 519 and 492 cm-1 correspond to al-o vibrations (hofmeister et al., 1987; viana et al., 2002; łodziński et al., 2005; makreski and jovanovski, 2009). geochemistry major element and trace element analyses of selected schist, schist sediment, peridotite and pyroxenite samples are given in table 2. bulk stream sediment samples from the karakız stream, which cuts the metamorphic rocks, are composed of high cao (8.58-18.75%), fe2o3 (4.18-7.11%), moderate mgo (3.14-5.56%), chromium (356-554 ppm), nickel (229-306 ppm), strontium (133-140 ppm), vanadium (103 ppm) and zirconium (7-91 ppm). the metamorphic rocks (garnet-glaucophane schist, lawsonite schist and lawsoniteepidote schist) contain appreciable amounts of fe2o3 (13.12%), mgo (8.279.20%), moderate al2o3 (11.91-18.46%), nickel (97-291 ppm), chromium (196619 ppm), zirconium (117-285 ppm), cobalt (50-111 ppm), zinc (33-134 ppm), vanadium(18-228 ppm) and low beryllium (0.5-21 ppm). large ion lithophile elements (lile) (potassium, rubidium, and lead) of the metamorphic rocks are depleted, whereas the metamorphic rocks are enriched with barium (119190 ppm) and strontium (280-490 ppm). in general, lile (barium, strontium except for lead, potassium), some transition metals (cobalt, except nickel) and hfse (niobium, zirconium, yttrium except thorium) elements are relatively low or depleted compared to metamorphic rocks. however, chromium and nickel are enriched in stream sediments about the metamorphic rocks. ophiolitic rocks (peridotite, pyroxenite) of the area contain 20.6-37.84 wt% mgo, 6.276.56 wt% fe2o3, 5515-7116 ppm chromium and 1369-2103 ppm nickel. beryl chemistry a chemical analysis of an eskişehir beryl crystal is given in table 3. this study reveals that the eskişehir beryl has crystal chemistry of emerald compared to the composition of similar occurrences of octahedral beryl in metamorphic units of similar nature (table 3). the calculated structural formula for this eskişehir beryl is ( a l 1 . 7 5 c r 0 . 0 1 m g 0 . 2 2 f e 0 . 0 8 ) ( b e 2 . 9 0 s i 6 . 0 0 ) ( n a 0 . 3 2 ) o 1 8 . the high chromium value compared to vanadium and the ratio of fe2+ to fe3+ indicates that the green coloration of the eskişehir emeralds is derived from chromium. the atoms per formula unit (apfu) for the emerald crystal were calculated based on 18 oxygen. the small aluminum value (1.746 apfu) in the eskişehir emerald versus a high total value for the octahedral cations (mg2+, cr3+, fe2+), compared to the compositions of other schist-hosted emeralds, shows that the eskişehir emerald is poor in aluminum (al<1.75 apfu), similar to canadayukon emeralds, as reported by groat et al. (2002). magnesium (0.217 apfu) and iron (0.08 apfu) cations are substituted for aluminum at the octahedral site of the emeralds structure, thus allowing this to be characterized as an octahedral beryl (aurisicchio et al., 1988) (fig. 6). sodium was assigned to the channel site. discussion beryl rarely occurs in stream sediments of the study area as a result of resistance to alteration processes via short stream transportation distances and accumulates as euhedral to subhedral crystals in alluvial-fluvial placer deposits (dill, 2010). similarly, developed occurrences of beryl and other gem crystals in stream sediment by chemical and physical alteration of pegmatite, metamorphic and greywacke type units, were described in the state of minas gerais, brazil (proctor, 1984); the ratnapura gem deposit, sri lanka (munasinghe and dissanayake, 1981; dissanayake et al., 2000); and ilakaka river gravels in madagascar (pezzotta and simmons, 2001). beryl-bearing sediments of the study area that are derived from metamorphic units consist of garnet-glaucophane schist, lawsoniteglaucophane schist, epidote-chlorite schist, epidotite, and chlorite-lawsonite schist (kulaksız, 1981) reflecting blueschist facies (bucher and frey, 1994; davis and whitney, 2006). green and light bluish-green, hexagonal prismatic beryl crystals occur in stream sediment that transects the metamorphic units, suggesting that the chromium-rich beryl developed within the metamorphic rocks consisting of garnet, glaucophane, epidote, chlorite, feldspar, quartz, lawsonite, muscovite, augite and sphene. its formation in the metamorphic units was controlled by synto post-tectonic deformation due to the upthrust of chromium-rich ophiolitic units, which resulted in metasomatic reactions. thus, the concentration of magnesium-rich and beryllium-rich fluid favored the crystallization of beryl (turner et al., 2005; grundmann and morteani, 2008; groat et al., 2008). the host rock of this type of beryl occurrence is similar to beryl (emerald) deposits that are formed via synto post-tectonic reactions under low-grade regional metamorphism and within biotite schist tectonically overlain by ophiolite, such as in southern egypt (abdalla and mohamed, 1999) and habachtal, austria, (grundmann and morteani, 1989) where the source of chromium is believed to be metasomatized ophiolitic rocks (groat et al., 2008). the green coloration of the eskişehir emerald is due to the presence of chromium and, potentially, the ratio of fe2+ to fe3+, similar to the emeralds of zambia, canada, egypt, austria and brazil (table 3). aurisicchio et al. (1988) documented substitutions involving fe2+, mg2+, mn2+, fe3+, cr3+, v5+, and sc3+, as well as v5+ and ti4+, for octahedrally coordinated aluminum, as well as lithium for tetrahedrally coordinated beryllium. additionally, eskişehir beryl can be classified as chromium-containing beryl according to the gemological institute of america (sinkankas, 1994). the presence of chromium and nickel in the eskişehir emerald crystals and the small values of barium, strontium, and rubidium suggest that crystallization of this beryl (emerald) crystals was due to a metasomatic-fluid reaction regime controlled by the upthrust of ophiolitic units onto the schists. eskişehir beryl contains high na2o (1.69%) compared to k2o (<0.29%) and indicates alkali-rich, sodic beryl occurrence of cr-bearing beryl in stream sediment from eskişehir, nw turkey a9 (łodziński et al., 2005). the presence of type-ii h2o molecules narrow 3594 cm-1 and strong 1634 cm-1 bands corresponds to the presence of alkalis in the structural channels of beryl, similar to that described for beryls of the sudety mountains and macedonian regions (łodziński et al., 2005; makreski and jovanovski, 2009). these stretching bands decrease, and type-ii h2o disappears when the alkali content of the beryl falls (manier-glavinaz et al., 1989). magnesium is the main constituent of emerald that is derived from metamorphic rocks. the high magnesium contents of the metamorphic units may indicate a genetic relationship to beryl (schwarz and giuliani, 2001). the chromium (619 ppm) and nickel (681 ppm) contents in the garnet-glaucophane schist can be explained by the close association with ophiolitic rocks, which contribute chromium and nickel via leaching (grundmann and morteani, 2008). furthermore, lile (barium, strontium), transition metals (cobalt, except nickel) and hfse (niobium, zirconium, and yttrium) element values that were low in stream sediments relative to metamorphic rocks suggest that these elements were transported during weathering of primary minerals, similar to what has been reported for lile values in sri lanka (chandrajith et al., 2000). the low beryllium content of stream sediment (1 ppm) and rocks (0.46-21.09 ppm) in the study area is similar to that of other beryl-bearing metamorphic rocks and sediments in nature (grundmann and morteani, 1989; grew, 2002). all major oxide (except sio2 and cao) values decrease in stream sediment, possibly during transportation from altered metamorphic units. beryllium-bearing fluid associated with sio2 and al2o3 may indicate thermodynamic conditions (p, t) that are related to the alkalinity of melts and fluids through reactions (barton and young, 2002). thus, beryl is stable under high silica activities and temperature (> 600°c) conditions. by contrast, a decrease of the silica activity of beryl may result in replacement with other beryl crystals, such as chrysoberyl and phenakite, which are absent in the study area. the occurrence of alluvial gemstones was identified for the first time in the eskişehir region (turkey) during this study. this research may be significant for future use in investigations of beryl crystals that are associated with metamorphics and related sediments. additionally, this study will also help identify the effect of tectonism and metasomatism on the formation of beryl crystals near ophiolite and metamorphic units and the local occurrence of beryls in sediment under short distance transportation from metamorphics. conclusions the eskişehir green color beryl crystals formed in schists via the upthrust in the upper cretaceous of chromium and magnesiumrich metasomatized ophiolites onto metamorphic units. locally, the contact with ophiolites metamorphic unit is transected by streams, the sediments of which contain micrometric accessory beryl crystals. the euhedral, hexagonal beryl crystals have low refractive index value and between 200 µm and 1 mm sizes. the eskişehir beryl’s chemical composition is characterized by high chromium, iron and magnesium, reduced sodium, cesium and low potassium, vanadium and rubidium. acknowledgements this research comprises further work on the first author’s m.sc. study, supervised by the second author. the authors are indebted to doctors güneş kürkçüoğlu (eskişehir osmangazi university) and david wray (university of greenwich) for conducting the infrared spectra and chemical analyses of the emerald crystals, respectively. we also thank professor rifat bozkurt (eskişehir osmangazi university) for his help during the m.sc. study. the authors are much indebted to professor fahri esenli (i̇stanbul technical university) and an anonymous reviewer for their extremely careful and constructive reviews that improved the quality of the paper significantly. we are also incredibly grateful to editor in chief, professor andrés felipe torres-villegas for his insightful editorial comments and suggestions. references abdalla, h. m, & mohamed, f. h. (1999). mineralogical and geochemical investigation of emerald and beryl mineralisation, pan-african belt of egypt: genetic and exploration aspects. journal of african earth sciences, 28, 581-598. arif, m., henry, d. j. & moon, c. j. (2010). cr-bearing tourmaline associated with emerald deposits from swat, nw pakistan: genesis and its exploration significance. american mineralogist, 95, 799–809. aurisicchio, c., fioravanti, g., grubessi, o. & zanazzi, p. f. (1988). reappraisal of the crystal chemistry of beryl. american mineralogist, 73, 826-837. barton, m. d. & young, s. (2002). non-pegmatitic deposits of beryllium: mineralogy, geology, phase equilibria and origin. reviews in mineralogy and geochemistry, 50, 591-691. beus, a. a. (1979). sodium: a geochemical indicator of emerald mineralization in the cordillera oriental, colombia. journal of geochemical exploration, 11, 195–208. bingöl, e. (1976). batı anadolu’nun jeotektonik evrimi. mta dergisi, 86, 14-43 [in turkish, unpublished]. bucher, k. & frey, m. (1994). petrogenesis of metamorphic rocks. springer, berlin, 318 pp. chandrajith, r., dissanayake, c. b. & tobschall, h. j. (2000). the stream sediment geochemistry of the walawe ganga basin of sri lanka-implications for gondwana mineralization. gondwana research, 3, 189-204. davis, p. b. & whitney, l. (2006). petrogenesis of lawsonite and epidote eclogite and blueschist, sivrihisar massif, turkey. journal of metamorphic geology, 24, 823-849. dereppe, j. m., moreaux, c., chauvaux, b. & schwarz, d. (2000). classification of emeralds by artificial neural networks. journal of gemmology, 27, 93-105. dill, h. g. (2010). the “chessboard” classification scheme of mineral deposits: mineralogy and geology from aluminum to zirconium. earthscience reviews, 100, 1-420. dissanayake, c. b., chandrajith, r. & tobschall, h. j. (2000). the geology, mineralogy and rare element geochemistry of the gem deposits of sri lanka. bulletin of the geological society of finland, 72, parts 1-2, 5-20. erentöz, c. (1975). 1/500 000 ölçekli türkiye jeoloji haritası. mta, ankara. farmer, v. c. (1974). layer silicates, in farmer, v.c. (ed), infrared spectra of minerals, mineralogical society, london, 331-363. gavrilenko, e. v., pérez, b. c., bolibar, r. & castroviejo, garcia del amo d. (2006). emeralds from the delbegetey deposit (kazakhstan): mineralogical characteristics and fluid-inclusion study. mineralogical magazine, 70, 159-173. giuliani, g., cheilletz, a., zimmermann, j.l., ribeiro-althoff, a.m., francelanord, c. & feraud, g. (1997b). les gisements d’émeraude du brésil: genése et typologie. chronique de la recherche miniére, 526, 17-61. giuliani, g., france-lanord, c., cheilletz, a., coget, p., branquet, y. & laumomnier, b. (2000). sulfate reduction by organic matter in colombian emerald deposits: chemical and stable isotope (c, o, h) evidence. economic geology, 95, 1129-1153. giuliani, g., france-lanord, c., zimmermann, j. l., cheilletz, a., arboleda, c., charoy, b., coget, p., fontan, f. & giard, d. (1997a). fluid composition, δd of channel h2o and δ 18o of lattice oxygen in beryls: genetic implications for brazilian, colombian and afghanistani emerald deposits. international geology review, 39, 400-424. gözler, m. z., cevher, f., ergül, e. & asutay, h. j. (1996). orta sakarya ve güneyinin jeolojisi. mineral research and exploration (mta) raport no. 9973 (unpublished). grew, e. s. (2002). beryllium in metamorphic environments (emphasis on aluminous compositions). in: grew, e.s. (ed), reviews mineralogy and geochemistry, 50 mineralogical society of america, washington dc, 487-537. hülya erkoyun and selahattin kadir a10 groat, l. a., giuliani, g., marshall, d. d. & turner, d. (2008). emerald deposits and occurrences: a review. ore geology reviews, 34, 87-112. groat, l. a., marshall, d. d., murphy, d. c., piercey, s. j., jambor, j. l., mortensen, j. k., ercit, t. s., gault, r. a., matte, d. p., schwarz, d., maluski, h., wise, m. a., wengzynowski, w. & eaton, d. w. (2002). mineralogical and geochemical study of the regal ridge emerald showing, southeastern yukon. the canadian mineralogist, 40, 1313-1338. grundmann, g. & morteani, g. (1989). emerald mineralisation during regional metamorphism: the habachtal (austria) and leydsdorp (transvaal, south africa) deposits. economic geology, 84, 1835-1849. grundmann, g. & morteani, g. (2008). multi-stage emerald formation during panafrican regional metamorphism: the zabara, sikait, umm kabo deposits, south eastern desert of egypt. journal of african earth sciences, 50, 168-187. gültekin, a. h., örgün, y. & suner, f. (2003). geology, mineralogy and fluid inclusion data of the kızılcaören fluorite-barite-ree deposit, eskişehir, turkey. journal of asian earth sciences, 21, 365-376. hofmeister, a. m., hoering, t. c. & virgo, d. (1987). vibrational spectroscopy of beryllium aluminosilicates: heat capacity calculations from band assignments. physics and chemistry of minerals, 14, 205-224. jarvis, i. & jarvis, k. e. (1992). plasma spectrometry in earth sciences: techniques, applications and future trends, in jarvis, i. & jarvis, k.e. (eds), plasma spectrometry in earth sciences. chemical geology, 95, 1-33. kahya, a. & kuşcu, m. (2014). source of the mineralizing fluids in ultramafic related magnesite in the eskişehir area, northwest turkey, along the i̇zmir-ankara suture: a stable isotope study. turkish journal of earth sciences, 23, 1-15. kazmi, a. h. & snee, l. w. (1989). geology of world emerald deposits: a brief review, in kazmi, a.h. & snee, l. w. (eds), emeralds of pakistan: geology, gemology, and genesis, van nostrand reinhold, new york, 165-236. kulaksız, s. (1981). sivrihisar kuzeybatı yöresinin jeolojisi. yerbilimleri, 8, 103-120. leitmeier, h. (1937). das smaragdvorkommen in habachtal in salzburg und seine mineralen. tschermak’s mineralogische und petrographische mittteilungen, 49, 245-368. łodziński, m., sitarz, m., stec, k., koznecki, m., fojud, z. & jurga, s. (2005). icp, ir, raman, nmr investigations of beryls from pegmatites of the sudety mts. journal of molecular structure, 744-747, 1005-1015. makreski, p. & jovanovski, g. (2009). minerals from macedonia xxiii. spectroscopic and structural characterization of schorl and beryl cyclosilicates. spectrochimica acta part a, 73, 460-467. manier-glavinaz, v., couty, r. & lagache, m. (1989). the removal of alkalis from beryl: structural adjustments. the canadian mineralogist, 27, 663-671. munasinghe, t. & dissanayake, c. b. (1981). the origin of gem stones of sri lanka. economic geology, 76, 1216-1224. okay, a. i. (2002). jadeite-chloritoid-glaucophane-lawsonite blueschists in north-west turkey: unusually high p/t ratios in continental crust. journal of metamorphic geology, 20, 757-768. pezzotta, f. & simmons, wm. b. (2001). field course on the rare element pegmatites of madagascar. technical program and field trip guidebook 1-9. pignatelli, i., giuliani, g., ohnenstetter, d., agrosi, g., mathieu, s., morlot, c. & branquet, y. (2015) colombian trapiche emeralds: recent advances in understanding their formation. gems and geology, 51, 222-259. prime ministry archieves (1993). başbakanlık devlet arşivleri genel müdürlüğü osmanlı arşivi daire başkanlığı 438 numaralı muhasebe-i vilayet-i anadolu defteri (937/1530) i kütahya, karahisar-i sahib, sultan önü hamid ve ankara livaları. ankara, 414. proctor, k. (1984). gem pegmatites of minas gerais, brazil: exploration, occurrence, and aquamarine deposits. gemological institute of america. gems and gemology summer, 78-100. sabot, b. (2002). classification des gisements d’emeraude: apports des etudes petrographiques, mineralogiques et geochimiques. ph. d. thesis (unpublished), institut national polytechnique de lorraine, france, 172pp. sabot, b., cheilletz, a., de donato, p., banks, d., levresse, g. & barrés, o. (2000). afghan emeralds face colombian cousins. chronique de la recherche miniére, 541, 111-114. sarıfakıoğlu, e., özen, h. & hall, c. (2009). petrogenesis of extension related alkaline volcanism in karaburhan (sivrihisar-eskişehir), nw anatolia, turkey. journal of asian earth sciences, 35, 502-515. schwarz, d., & giuliani, g. (2001). emerald deposits − a review. australian gemmologist, 21, 17-23. schwarz, d. & giuliani, g. (2002). emeralds from asia. extra lapis english, 2, 60-63. schwarz, d., kanis, j. & kinnaird, j. (1996). emerald and green beryl from central nigeria. journal of gemmology, 25, 117-141. seifert, a. v., žáček, v., vrána, s., pecina, v., zacharıáš, j. & zwaan, j. c. (2004). emerald mineralization in the kafubu area, zambia. bulletin of geosciences, 79, 1-40. sendir, h. (2004). karaburhan (sivrihisar-eskişehir) yöresi platin grubu elementlerinin mineralizasyonu ve dağılımı. esogü yüksek lisans tezi, 56pp. şengör, a. m. c. (1979). the north anatolian transform fault: its age, offset and tectonic significance. journal of the geological society of london, 136, 269-282. şengör, a.m.c. & yılmaz, y. (1981). tethyan evolution of turkey: a plate tectonic approach. tectonophysics, 75, 181-241. sinkankas, j. (1994). emerald and other beryls. geoscience press, prescott, arizona, 665 pp. temel, a. (2001). post-collision miocene alkaline volcanism in the oglakçi region, turkey: petrology and geochemistry. international geology review, 43, 640. turner, d., groat, l. a. & wengzynowski, w. (2005). mineralogical and geochemical study of the true blue aquamarine showing, shark property, southern yukon, in emond, d.s. & wengzynowski, w. (eds), yukon exploration and geology, 2004, yukon geological survey, 275-285. viana, r. r., jordt-evangelista, h., magela da costa, g. & stern, w.b. (2002). characterization of beryl (aquamarine variety) from pegmatites of minas gerais, brazil. physics and chemistry of minerals, 29, 668-679. zwaan, j.c. (2006). gemmology, geology and origin of the sandawana emerald deposits, zimbabwe. scripta geologica, 131, 1-211. search | earth sciences research journal correo electrónico dninfoa sia bibliotecas convocatorias identidad u.n. english español revistas.unal.edu.co servicios skip to main content skip to main navigation menu skip to site footer earth sciences research journal register login you are in: portal de revistas un / earth sciences research journal / search earth sciences research journal english español current archives announcements about about the journal submissions editorial team personal data processing policy contact search search search search articles for advanced filters published after 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 published before 1992 1993 1994 1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 january february march april may june july august september october november december 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 by author search no results scimago journal & country rank (sjr) indexed and registered   keywords earth sciences research journal - universidad nacional de colombia (sede bogotá). facultad de ciencias. departamento de geociencias editor: alexander caneva  correo electrónico: earthjour_fcbog@unal.edu.co issn impreso: 1794-6190 issn línea: 2339-3459 doi: 10.15446/esrj   indexada en: ibn publindex el índice bibliográfico nacional publindex es un sistema colombiano para la clasificación, actualización, escalafonamiento y certificación de las publicaciones científicas y tecnológicas. es regido por colciencias y el icfes en colombia. directory of open access journals doaj aumenta la visibilidad y la facilidad de uso de las revistas científicas y académicas de acceso abierto, pretende ser global y abarcar todas las revistas que utilizan un sistema de control de calidad para garantizar el contenido. scielo colombia scielo colombia es una librería virtual para américa latina, el caribe, españa y portugal, fue creada por fapesp en el año de 1997 en sao pablo brasil, actualmente en colombia es gestionada por la universidad nacional de colombia. redib (red iberoamericana de innovación y conocimiento científico) redib es una plataforma de agregación de contenidos científicos y académicos en formato electrónico producidos en el ámbito iberoamericano. redib cuenta con una clara vocación de promoción de la innovación tecnológica de las herramientas de producción editorial. estas facilitan el acceso, la difusión y la puesta en valor de la producción científica generada en los países de su ámbito de actuación, especialmente en los diversos idiomas que les son propios. los destinatarios de esta información son tanto la comunidad académica como la sociedad en general, así como los responsables, gestores y analistas de políticas científicas. science citation index expandedtm sci de thomson reuters es un prestigio sistema de indexación en línea que incorpora información bibliográfica y de citación de publicaciones científicas alrededor del mundo. scopus scopus es una base de datos bibliográfica de resúmenes y citas de artículos de revistas científicas. cubre aproximadamente 19.500 títulos de más de 5.000 editores internacionales, incluyendo la cobertura de de 16.500 revistas. latindex latindex es producto de la cooperación de una red de instituciones latinoamericanas que funcionan de manera coordinada para reunir y diseminar información bibliográfica sobre las publicaciones científicas seriadas producidas en la región. nivel nacional amazonía bogotá caribe manizales medellín orinoquía palmira bibliotecas catálogo recursos repositorio institucional portal de revistas facebook twitter youtube email rss régimen legal talento humano contratación ofertas de empleo rendición de cuentas concurso docente pago virtual control interno calidad buzón de notificaciones correo institucional mapa del sitio redes sociales faq quejas y reclamos atención en línea encuesta contáctenos estadísticas glosario contacto página web: avenida el dorado no. 44a-40, ed. 571, piso 4o. hemeroteca nacional universitaria bogotá d.c., colombia (+57 1) 316 5000 ext. 20 004 © copyright 2014 algunos derechos reservados. digital@unal.edu.co acerca de este sitio web actualización: 31/08/19 this issue of earth sciences research journal we at the earth sciences research journal hope all our readers have a happy and productive 2018. our last issue of esrj in 2017 presents a number of papers on different topics within the earth sciences. in alvarez-ysabel et al. a 150+ year database of tropical cyclone activity is analyzed for the dominican republic (dr). a total of 112 events have tracks at most 300 km from the dr are studied in terms of their seasonal distribution and long-term distribution, their frequency and landfall probability. the authors also present a review of trocial cyclone activity in the region. ogly gahramanov uses a combination of geological and geochemical information to model the formation of hydrocarbon deposits in the south caspian basin, in particular for the south caspian depression. the authors argue that large pressures and tectonic history have a large effect on the appropriate selection of potential drilling sites. using high-resolution aeromagnetic survey data in combination with other geological-geophysical thematic maps, batista takes advantage of gis tools for identifying different tectonic regimes in the sabinas basin, northeastern mexico. regions where the basin is deepest and the delineation of faults allows for the determination of block boundaries as well. he et al. use a modified direct shear test apparatus to determine shear strength parameters of unsaturated soils under various conditions. the paper proposes that the direct shear test analysis is an efficient way to analysis soil samples. zhuo and chen explore the improved water retention capacity of sandy soils by adding nanocarbon made from coconut shells. different amounts and thicknesses of nanocarbon were added to the soil, increasing water content with increased nanocarbon. loess deposits can have very large thicknesses, and for engineering purposes it is important to better understand the geotechnical response of the loess deposits. wang et al. investigate the collapsibility of loess deposits in the xixian new region using compressions tests under different pressure and water content conditions. results presented may have important application in future geotechnical studies in the region. germán a. prieto editor-in-chief th is is su e o f ea rt h s c ie n c es r e se a r c h j o u r n a l earth sciences research journal earth sci. res. j. vol. 21, no. 4 (december, 2017) vargas et al.indd earth sci. res. j. vol. 11, no. 1 (june 2007): 5-19 geoinformation density: a criterion on anh block negotiation carlos a. vargas j.1,2 – armando zamora r. 1 – andres pardo 1,3 1 agencia nacional de hidrocarburos – anh 2 departamento de geociencias, universidad nacional de colombia – sede bogotá 3 departamento de geología, universidad de caldas corresponding author: carlos a. vargas j., e-mail: cavargasj@unal.edu.co abstract based on the general information included within the anh exploration and production information service, the information distribution densities of the colombian territory were determined. these densities are referred in seismic length per square kilometer and amounts in magnetic-gravimetric and geochemical information data or drill-hole length per sq. km. a probabilistic distribution was assessed along the density distribution and cost distribution for each variable. the variables, as information layers, were cross-referenced in order to define relative weights that assess information with respect to the presence or absence of information in the treated area. this procedure could be regarded as a methodological support proposal for area negotiation of the oil industry. a better approach to the quandary should include: new input of data into the system, a division of the territory in smaller areas adjusting to complex geometries, and considering the particular market conditions for each basin. nevertheless, it is apparent that the obtained results favor the consolidation of a conceptual framework that at medium term will allow a conscious approach towards block negotiation between the petroleum industry and the anh. key words: density of information, oil exploration, negotiation, colombia. resumen la clasificación automática de señales sísmicas se ha llevado a cabo típicamente sobre representaciones el uso de la información de carácter general incluida dentro del servicio informativo de la exploración y de la producción de anh permitió la determinación de densidades de distribución de información del territorio colombiano. estas densidades se refieren a longitud de sísmica por el kilómetro cuadrado, cantidades de datos de información magnético-gravimétrica y geoquímica, así como longitud de perforación por kilómetro cuadrado. una distribución probabilística fue entonces estimada para la distribución de la densidad y de la distribución de costos para cada variable. las variables como capas de información fueron pesadas relativamente para determinar la presencia o 5 earth sciences research journal manuscript received march 13 2007. accepted for publication june 20 2007. ausencia de la información en el área tratada. este procedimiento se podría mirar como una propuesta metodológica para la ayuda en la negociación de áreas de interés en la industria del petróleo. un acercamiento mejor al problema debería incluir: nueva entrada de datos en el sistema; una división del territorio en áreas más pequeñas que se ajusten a las geometrías complejas; y considerar las condiciones de mercado particulares en cada cuenca. sin embargo, es evidente que los resultados obtenidos favorecen la consolidación de un marco conceptual que en el mediano plazo permitirá un acercamiento consciente hacia la negociación de bloques de interés exploratorio para la industria petrolera y la anh. palabras claves: densidad de información, exploración de petróleo, negociación, colombia. introduction a negotiation is a communication framework where the assessed object considers an interaction between value estimations of each participating transaction party. this practice may become subjective, as each participant may imprint a differential value based on its own cognizance. evidently, he who has the utmost knowledge of the object will have a better position to assess the cost benefit relationship and go appropriately further in the negotiation. the strategy is to set up a large sum of negotiation criteria to establish block allocation mechanisms within the practice of exploratory promotion developed by the anh. one of these criteria may be supported on the hypothesis of geoinformation density, that is, “the possibility that information distributed in a specific area may be considered as typical and a criterion to quantify its relative investment”. accepting this statement as starting point, it can be demonstrated that geological, geophysical, geochemical, and in general all geospatial information, may be conceived as an objective tool for block appraisement. in this line of thought, the goal of this paper is guided towards defining a conceptual tool that will allow supplementing block negotiation undertaken by the anh within its mission outline. methodological aspects based on the general information included within the anh exploration and production information service epis such as longitude and location of several seismic programs, gravimetric, magnetometric, and geochemical location coverage along with well distribution performed during the exploratory history of colombia, information distribution densities have been determined from territory area discretization (e.g., 25km x 25km). these densities are referred in seismic length per square kilometer and magnetic-gravimetric and geochemical information data amounts or drillhole length per square kilometer. once the information densities have been established for the area, the probabilistic distribution is assessed along with the inferred relative weighing factors to the density distribution and cost distribution for each variable. the variables, as information layers, have been cross-referenced in order to define relative weights that assess information weighters with respect to the presence or absence of information in each treated area. figure 1 depicts a synthetic outline establishing the procedures followed on this paper. data it could be stated that a number of the geospatial variables analyzed show strong contrasts with respect to their distribution in colombia. figure 2 shows the location of seismic information depicting a dense coverage towards the eastern plains (llanos orientales) basin including the eastern mountain range foothills, putumayo and magdalena valley (high, medium, and low), pacific off-shore coast (central and south region), atlantic off-shore coast, and sinú-san jacinto and cesar-ranchería basins. in contrast, low information densities are seen towards the orinoquía and amazon regions, north of the department of chocó, and along the mountain range axis. a similar situation is seen with the geoinformation density: a criterion on anh block negotiation 6 vargas et al., esrj vol. 11, no. 1. june 2007 7 magnetic and gravimetric information, well distribution, and geochemical controls (figures 3, 4, and 5). even though the epis database has information voids in some periods and there are time differences in data entry in the same database, probably as a testimony of the change of administration, its current structure and contents remains representative and may guarantee that the database is suitable to assess the exploratory tendencies and ensure an appropriate criterion for block negotiation. analysis and results distribution function to understand how spatial distribution of geoinformation density could be analyzed, it is necessary to begin from an analysis of the distribution function that better suits the observations. for instance, seismic density registered in colombia could be used as a representative variable for this procedure. figure 6 shows the distribution of density frequencies and the best fit representing two models of statistical distribution: normal and gamma. for the first event, a strong unbalance is seen, suggesting that the distribution could not comply with a normal distribution process. usually, offer and demand processes follow normal distributions. in this event, although it is a similar scenario for area allocation, it follows a gamma pattern fit, specifically an exponential distribution (ayyub and mccuen, 2003). exponential distributions are useful to assess processes through time (blaesid and granfeldt, 2003). however, this condition is not reflected in our issue because this is a data set encompassing the whole observation period, and illustrates a scenario where the main biasing frequency almost dominates the distribution below the mode and in turn allows values greater than frequency. a checkmark for this distribution may be seen in the simple and cumulative probability curves (figures 7 and 8). the behavior is similar to “club admittance”, in other words there is a minimum fee to be a member, but any sum above the minimum is also admissible and very exceptionally members below the fee will be presented. this is situation fundamentally imposed by the petroleum industry, which possibly will promote increases in admission quotas when the business tends to be more attractive. in effect, for a greater observation time and under the same energy requirements in colombia, it could be expected that the seismic information density as a criterion for block allocation (admission quota) tends to be greater. table 1 shows the parameters for each distribution. table 1. statistical parameters for types of distributions applied to seismic data density per unit area. a similar situation is seen in distribution functions for information densities associated to the amount of magneto-gravimetric points and drill-hole length (figures 9 and 10). for well data (m/km2), this index depicts in certain degree, the inversion rate executed to know a given area. likewise, information density related with the number of geochemical stations should take into account the dissimilarity in the level of tests for each station (toc, pyrolysis, organic petrography, sulphur content, metal content, rock extracción, liquid gas chromatography, and biomarkers). this information is conditioned to the existence of an analysis sequence protocol as a function of the results obtained from certain initial testing such as toc and pyrolysis. therefore, its representability only grants a partial nature (figure 11). a representation of these variables in the colombian territory is shown in density information maps associated with the analyzed variables (figures 12, 13, 14, and 15). table 2 summarizes the exponential distribution parameters converted into relative weights, describing all the observed information. normal exponential domain median 0.387563 0.387563 variance 0.219149 0.150205 generally, empirical adjustments with confidence levels can be seen, widely reproducing the behavior for the amount of data for the observed period. estimation of a relative weighting factor for area negotiation the exponential probability distribution function and probability estimation have been defined as (blaesid and granfeldt, 2003): (1), (2), where is the arithmetic mean and x the discrete value of the distribution of the estimated probability. for all previously analyzed information densities, it is necessary to shorten the equation (2) to a relative weight factor that promotes the generation of new knowledge in areas where data is scarce: (3), the weighting factor ( ) is a tool that relatively contemplates initiatives directed towards new surveys under different block contracting employed by anh. according to equation (3), areas with plenty of information will be penalized with low weighting factors and viceversa. obviously, upon promoting strategies encouraging enforcement of new scientific and technological procedures on areas of interest, the need to estimate other weighting factors with the equation (3) is a must. information layer weighting as can be expected, the relative sum of related weights with all variables, allows having an approach to the spatial distribution of information density. although average costs for each variable are very different and entail another relative weighting to market averages. table 3 shows colombian seismic kilometer average costs, unitary values for each magneticgravimetric and geochemical point and linear drill-hole foot (with more frequent analyses), as developed approximate amounts. it could be seen that seismic is the most expensive exploratory variable, apart from being the most widely used. likewise, the technical link between seismics and wells is noticeable, being these two representable variables of areas where the major exploration investments have been executed.therefore, superimposing the drill-hole density map on top of the seismic density information map, or at least one of these two variables, may be a criterion enough to estimate a map of information density relative to the weighting providing support for block negotiation processes. nevertheless, equation 4 is an empirical approach that assumes the superimposition of all variables taken into consideration and weighting the relative cost of each variable. the estimates seen in equation (4) enabled performing a weighted geoinformation density map (figure 16). this map highlights the areas where the effect of new exploratory information has less effect. possibly, the current setting may vary as new data is added to the information foundation that engineered it. according to figure 16, there are areas where no exploratory efforts have been undertaken as those analyzed, namely those in the colombian eastern border, the borders with venezuela and brazil, in the serranía of san lucas, the junction of the pastos, the sierra nevada of santa marta, and the lower cauca river valley. this state of 8 geoinformation density: a criterion on anh block negotiation (4), 9 inform ation d ensity form ula seism ic (km /km 2) m agnetic-gravim etric stations (am ount/km 2) d rill-hole length (ft/km 2) g eochem ical stations (am ount/km 2) table 2. e xponential distribution param eters converted into relative w eights, describing all the observed inform ation vargas et al., esrj vol. 11, no. 1. june 2007 figure 1. general aspects on the methodology used in this work. density weight data for four information layers were weighted to obtain one layer of useful relative weight geoinformation in support of block negotiation processes. figure 2. distribution of seismic information associated to all hydrocarbon exploration projects in colombia. geoinformation density: a criterion on anh block negotiation 10 figure 3. distribution of the information associated to potential fields (aerial and terrestrial aeromagnetometry and gravimetry) product of several exploration mining and energy projects in colombia. figure 4. distribution of wells associated to several hydrocarbon exploration projects in colombia. 11 vargas et al., esrj vol. 11, no. 1. june 2007 figure 5. distribution of geochemical stations related to several hydrocarbon exploration projects in colombia. geoinformation density: a criterion on anh block negotiation figure 6. seismic data density relative frequency distribution and best fit representing two models of statistical: normal and gamma (exponential). 12 figure 7. cumulative probability curves for seismic density information fitted to typical normal and exponential distribution curves. figure 8. simple probability curve for seismic information density fitted into normal and exponential distribution typical curves. 13 vargas et al., esrj vol. 11, no. 1. june 2007 figure 9. simple probability curve for magnetometric and gravimetric information density adjusted to typical normal and exponential distribution curves. figure 10. simple probability curve for drill-hole length information density adjusted to typical normal and exponential distribution curves. geoinformation density: a criterion on anh block negotiation 14 figure 12. distribution map for information densities for the colombian territory. figure 11. simple probability curve for geochemical stations density information adjusted to typical normal and exponential distribution curves. 15 vargas et al., esrj vol. 11, no. 1. june 2007 figure 13.distribution map for information densities for drill-hole lengths for the colombian territory. figure 14. distribution map for gravimetric and magnetometric information densities for the colombian territory. geoinformation density: a criterion on anh block negotiation 16 figure 15. distribution map for geochemical station information densities for the colombian territory. figure 16. colombian territory distribution map for geochemical station information densities. 17 vargas et al., esrj vol. 11, no. 1. june 2007 18 affairs can be explained in view of the current geological knowledge. these areas do not seem to be favored by their inherent geological conditions and possibly have no mineral-energy potential of interest. in contrast, some areas have vast amounts of information density, evidenced by the widespread of exploratory activity such as the eastern plains, putumayo, and the magdalena mid-valley. on the other hand, some belts and areas with high information density are suddenly interrupted making it uncertain to know the potential continuity of the prospects that have not been adequately analyzed as the northern part of the pacific coast and towards the north of putumayo. although it is necessary to further analyze the information densities of these areas, it is also clear that the socio-environmental conditions of these areas are intricate and have hindered the possible development of potential prospects. conclusions the development of the present knowledge management assignment based on the epis database should be regarded as a methodological support proposal for area negotiation whose quantitative results rise above the preliminary results. a better approach to the dilemma should include: new input of data into the system as a consequence of an exhaustive search of data to complement the current petroleum knowledge of the country. the discretization of the territory in smaller areas adjusted to complex geometry, showing the current conditions for block negotiation. 1. 2. an approach of the problem considering the particular market conditions for each basin. data entry that establishes new information layers allowing to improve the relative weighting of geoinformation such as seismic information reprocessing and dating. breaking up information in basic data that allows to establish information layers with differing weights (e.g., differentiation of stratigraphic wells into different diameters and considering supplementary surveys such as electrical registry and petrophysical testing). nevertheless, it is apparent that the obtained results favor the consolidation of a conceptual framework that at medium term will allow a conscious approach towards block negotiation between the petroleum industry and the anh. acknowledgments to the staff at schlumberger, particularly to edgar medina and liliana salcedo, for enduring two months of my intense requirements over the available epis system information. references ayyub, b.m. and mccuen, r.h. (2003). probability, statistics, and reliability for engineers and scientists. new york: chapman & hall/crc press. balakrishnan, n. and basu, a. p. (1996). the exponential distribution: theory, methods, and applications. new york: gordon and breach. blaesild, p. and granfeldt, j. (2003). statistics 3. 4. 5. exploratory variable unit cost range mode (approx.) amount performed (2006) seismic (linear km) u$12000 – u$40000 u$18000 277.329 magneto-gravimetric station u$1.5 – u$80 u$15 1.155.624 geochemical station u$25 – u$1500 u$80 27.655 drilling (linear foot) u$350 – u$890 u$390 53.890.746 table 3. commercial values and amounts performed up to date for the exploratory variables used for this study. this data was approximated from anh existent information. geoinformation density: a criterion on anh block negotiation with applications in biology and geology. new york: chapman & hall/crc press. spiegel, m. r. (1992). theory and problems of probability and statistics. new york: mcgrawhill, p. 119. 19 vargas et al., esrj vol. 11, no. 1. june 2007 untitled earth sciences research journal earth sci. res. j. vol. 13, no. 1 (june 2009): 74-81 magnetic and palaeomagnetic studies as an aid in deciphering groundwater flow: a case study from deccan traps s.k.g. krishnamacharyulu1 and g.v.s. poornachandra rao2 1school of earth sciences, srtm university, nanded -431 606, maharashtra, india 2national geophysical research institute, hyderabad 500 007, andhra pradesh, india email: skgkchary@gmail.com, gvsprao@rediffmail.com abstract this paper reports the utility of palaeomagnetic investigations first time in the ground water studies. the magnetic anomalies along with palaeomagnetic studies are used to locate barriers obstructing groundwater flow in a basaltic terrain. magnetic anomalies in an area of around 4 sq. km in the university campus, nanded, maharashtra, india brought out two distinct anomaly closures that can be attributed to bodies with different directions of magnetizations. palaeomagnetic investigations reveal that the basalts are almost uniformly magnetized, asserting that the anomalies come form the subsurface and the anomalous sources are attributed to weathering of basal basaltic flows. the difference between the measured palaeomagnetic directions of the surface rocks and the observed magnetic anomalies, in corroboration with local geomorphologic conditions and the nature of the basaltic flows in thin basaltic terrains suggest that subsurface weathering plays a major role in the ground water movement. this paper presents a novel idea on the utilization of the palaeomagnetic studies coupled with ground magnetic anomalies in deciphering groundwater flow. key words: ground water, palaeomagnetic, magnetic anomalies resumen este artículo reporta por primera vez, la utilidad de las investigaciones paleomagnéticas en el estudio de aguas subterráneas. las anomalías magnéticas junto con estudios paleomagnéticos se usan para localizar barreras que obstruyen el flujo de aguas subterráneas en terrenos basálticos. las anomalías magnéticas observadas en un área de 4 km2 del campus de la universidad strm en nanded maharashtra, india, indicaron dos cierres anómalos atribuidos a sendos cuerpos con diferentes direcciones de magnetización. las investigaciones paleomagnéticas 74 manuscript received: january 02th, 2009. accepted for publication: june 10th, 2009. revelaron que los basaltos fueron magnetizados casi uniformemente, lo cual indica que las anomalías provienen del subsuelo y que las fuentes anómalas se atribuyen a flujos basálticos basales meteorizados. la diferencia entre las direcciones paleomagnéticas medidas en el afloramiento y las anomalías magnéticas observadas, corroboradas con condiciones geomorfológicas locales y con la naturaleza de los flujos basálticos en laminas delgadas, sugieren que la meteorización del subsuelo juega un papel clave en el movimiento de aguas subterráneas. este artículo presenta una idea novedosa para el uso de estudios paleomagnéticos conjuntamente con anomalías magnéticas para descifrar flujos en aguas subterráneas. palabras claves: aguas subterráneas, paleomagnetismo, anomalías magnéticas. introduction the utility of magnetic method in ground water studies, although limited, to trace the floor of the basins or locating the structures that obstruct or control the groundwater flow is well known. however, palaeomagnetic investigations with unchallenged contributions to calculate the pole positions, to construct the history of the earth’s magnetic field and to decipher the plate motions were never sought after in groundwater investigations. in an aimed finding the clues for anomalous quantities of groundwater in a localized area, the palaeomagnetic data is found to be of utmost value to supplement the magnetic anomalies in identifying the barriers obstructing the groundwater flow. this paper presents the case study of this investigation, where the palaeomagnetic studies are utilized probably for the first time in ground water studies. geology and location of the area figure 1 presents location and general geology of the area. the area lies in the toposheet no. 56 e/8 of survey of india (190 061n & 770 171e) on the banks of river godavari. the area is covered with deccan lava flows with exposures of massive and columnar-jointed basalt, which believed to be erupted during cretaceous-eocene period over the precambrian basement rocks. in general, this area comprises of three flows (kumar and naik, 1990; bhaskar, 1990). approximately 60m thick flow i is of compound pahoehoe type and highly weathered, exhibiting fragmentary tops at some places. this flow i is overlain by columnar aa type flow ii. the present area is exposed with many columnar basalts. flows iii, is a massive basalt exposed in limited places as hillocks. in this area, the elevated places are typically covered by massive basalts, while low relief areas are covered by columnar basalts. table i shows the general stratigraphic sequence of the region and the present study area represents flow ii with exposures of columnar basalts. mostly the ground water in this region is in the fractures and vesicular basalts. the water bearing capacity of the vesicular basalt largely depends on the size and percentage of vesicles, density and degree of interconnection. generally pahoehoe flows exhibit good porosity and permeability with uniform vesicles. in the massive compact basalts the weathering, fractures, joints, shearing leads to secondary porosity. the pretrappean topography of this area was irregular and the lower flows are abutting against the hillocks of the pre-trappean period (raychaudhuri et al., 1984). foote (1876) reported rude massive blocks at basal basaltic flows formed due to weathering. according to him the massive blocks are common striking features in thin basaltic layers particularly in the margin areas of the deccan basalts and decomposition of the trap rocks under certain circumstances gives rise to eminently clayey products containing large percentage of iron. several e-w joints/fractures were identified from photo-interpretation, showing the curvilinear nature on the ground. the godavari river in this region presents a corrugated shape with u turns and right angle bends the central ground water depart75 magnetic and palaeomagnetic studies as an aid in deciphering groundwater flow: a case study from deccan traps ment identified three sets of lineaments in this area in nw-se, n-s and almost e-w directions. (fig.1; cgwb aap report, 1996). the elevation map of the area is presented in figure 2. there is a geomorphic denuded feature in approximately e-w direction (shown by arrow in figure 2), flanked by columnar massive basalts to the north and south, corresponding to flow ii. a perennial lake forming cal de sac to this denuded feature can be seen on the eastern side (figure 2). hydrogeologically this is supposed to be a highly favourable condition for the groundwater. however, the existing bore wells (shown as star symbol in figure 2) in this area established only poor supply of groundwater. the bore wells went up to a depth of 76 s.k.g. krishnamacharyulu and g.v.s. poornachandra rao alluvium deccan trap lineaments study area 19 060 1 77 170 1 figure 1: general geology and location map of the area (after cgwd aap report, 1996) 77 magnetic and palaeomagnetic studies as an aid in deciphering groundwater flow: a case study from deccan traps table i: general stratigraphic sequence of the area covered under toposheet no. 56e/8 (after kumar and naik, 1990; bhaskar, 1990) elevation (msl) (m) flow no. type of flow characters flow iii compound, mixed (61m) forming hilly terrain 405 m red bole/intertrappeans flow ii aa flow (5-47m) highly columnar —355 m red bole flow i compound/mixed (60m) massive fine grained, vesicular and highly weathered 0 200 400 600 800 1000 1200 1400 1600 distance x (m) 0 200 400 600 800 1000 1200 1400 1600 1800 d is ta n c e y (m ) lake figure 2: elevation contour map of the study area (arrow shows the denuded geomorphic feature, dashed lines show the interpreted subsurface blocks and star symbol show the locations of failed bore wells. around 70 m. some open wells dug up to a depth of around 10-15 m within the top flow produce some water. this paper is an attempt to find out the probable cause for the obstruction of the ground water in the subsurface formations. magnetic survey and anomalies: to investigate the anomalous behaviour of the bore wells on the ew denuded area, a magnetic survey was conducted covering an area of approximately 4 sq.km. figure 3 shows the observed total field magnetic anomaly map with a station spacing of 200m. the magnetometer used in this study is a proton precession magnetometer of eg & g geometrics, canada, with a reading accuracy of 1 gamma. the magnetometer readings were corrected for the earth’s normal field and diurnal variation by making repeated measurements at a chosen base station the magnetic anomaly map shows two conspicuous pairs of positive and negative magnetic anomaly contour clusters. these two anomalous zones show almost a similar range of anomalies, with -1400 nt to 2500 nt on the northern anomalous zone and -2200 nt to 1600 nt on the southern zone; but they exhibit different trends, one in almost n-s direction and the other in e-w direction. since the interest is to 78 s.k.g. krishnamacharyulu and g.v.s. poornachandra rao 0 200 400 600 800 1000 1200 1400 1600 1800 distance x (m) 0 200 400 600 800 1000 1200 1400 1600 1800 a a b b d is ta n c e y (m ) figure 3: observed total field magnetic anomaly map of the university campus, nanded. (continuous lines indicate positive anomalies and dashed lines indicate the negative anomalies. the anomalies are expressed in nt and distances in meters. aa and bb are interpreted profiles). have the order of the depths, two-dimensional modeling is attempted though the anomaly pattern suggests a three-dimensional nature. two profiles aa and bb, one from each anomalous zone for quantitative interpretation. various models tried using the algorithm developed by radhakrishna murthy (1998) and found that the dyke model is best-fitted model for these profiles. it is observed that for this model the objective function is minimum and the fit between observed and calculated anomalies is good. the interpretation revealed that the anomalous bodies are situated at depths of 37 m and 40 m respectively under aa and bb. while the intensity of magnetization needed for both is almost the same, the directions of magnetization differ: the values being -74 and 172 degrees respectively. the lava flows of same age are expected to show same magnetization and hence these values are attributed to subsurface features altered probably due to weathering. further the calculated depth values interestingly correspond to the base of the top flow and clearly rule out the possibility of the anomalies coming form the magnetization changes within the top flow. to further rule out of this possibility, twenty two oriented rock samples from the surface were collected from the exposed rock units covering both the anomalous zones, and were subjected to palaeomagnetic and rock magnetic studies. rock magnetism and palaeomagnetism cylindrical specimens of standard dimension were drilled from the oriented samples and their natural remnant magnetism (nrm) was measured using a spinner magnetometer (model dsm-2, schonstedt, usa) at ngri, hyderabad, india. nrm vectors of these samples exhibit some scatter (fig. 4a) with downward pointing inclination in the se quadrant. however, the specimens in each sample reveal good grouping. the sample mean nrm intensity (jn) varies between 0.74 and 41.07 a/m. the susceptibility of these samples is very uniform varying over a limited range of 9.68 to 23.5 x 10-3 si units. stability of the remnant vector in these samples has been examined by subjecting the samples to af and thermal demagnetization on a pilot basis. a peak field of 20-30 mt has been applied to these samples to recover the characteristic remnant magnetization (chrm) vectors in these samples (figure 4b). when the samples were subjected to thermal demagnetization on a pilot basis up to 580 °c at intervals of 50 to 100 °c, the remnant vector in them revealed very stable nature. the samples revealed a blocking temperature of 500 °c where the chrm can be recovered, indicating the presence of titanomagnetite as the carrier of remnant magnetization in these rocks. typical demagnetization characteristics in case of some specimens to af and thermal fields are shown in figure 4c and 4d respectively. the specimen remnant vectors obtained after laboratory af demagnetization have been averaged to obtain the sample mean chrm vectors using fisher’s (1953) method. sample mean chrm improved as result of demagnetization by removal of viscous magnetization acquired by these rocks with the precision parameter (k) increasing from 4.64 to 25.74. the mean chrm vector for all the specimens together is dm = 148.1 and im = + 43.5 (k = 25.74, á95 = 5.88) and are presented in figure 4 b. the chrm vectors of samples from the two anomalous zones, the northern and southern zones in the vicinity of aa and bb are grouped separately and derived the mean remnant magnetic vectors from each anomalous zone. the northern zone showed a mean remnant magnetic direction of dm = 138.1 and im = + 42.24 (k = 29.21, á95 = 7.81) and the southern zone showed a mean remnant magnetic direction of dm = 158.3 and im = + 43.9 (k = 33.72, á95 = 7.27). thus the nrm directions of both the zones are practically the same. discussion and conclusion the inversion of observed magnetic anomaly suggests that two different bodies responsible for the anomalies are differently magnetized in direction, one practically horizontally and the other vertically. since the palaeomagnetic directions of the rocks from the two anomalous zones are almost the same, 79 magnetic and palaeomagnetic studies as an aid in deciphering groundwater flow: a case study from deccan traps and the surface rocks exhibit uniform susceptibility, the cause of the magnetic anomalies need to be expected to lie in the subsurface. in the peripheral areas of deccan basalts there are reports of rude block formations in the basal flows, which in turn forming iron rich clay bodies due to weathering. (foote, 1876). therefore, the magnetic anomalies can be attributed to the two different iron rich clay bodies formed at the base of the flow i. these clay bodies might be acting as local barriers in the subsurface and may be obstructing the flow of ground water, resulting poor ground water conditions in the area down below the lake. 80 s.k.g. krishnamacharyulu and g.v.s. poornachandra rao (a) (b) © (d) figure 4: palaeomagnetic results of the study area: (a) & (b) stereographic plot of sample mean vectors of nrm level and chrm level. (c) and (d) typical af and thermal demagnetization characteristics of specimens from the study area. acknowledgements we thank the director, national geophysical research institute (ngri), hyderabad and prof. i. v. radhakrishna murthy, andhra university, visakhapatnam for their encouragement. we acknowledge the financial support by dst, new delhi (16/165/2002). references bhaskar, b. (1990) geology in part of nanded district of maharashtra, geological survey of india records, v.123, part 6, pp. 83-85. cgwd annual action program report (1996), groundwater resources & potential development of nanded district, maharashtra, central ground water department, central region, govt. of india, nagpur. fisher, r.a. (1953) dispersion on a sphere, proc. roy. soc. london, a 217, pp. 295-315. foote, r.b. (1876) the deccan trap and associated formations, mem. geol. survey of india no.12, pp. 171-199., which reprinted in geol. survey of india special publication no.38., annals of deccan trap study and bibliography on deccan traps (1996) pp.43-65. kumar d. and naik, k..k. (1990) geology in parts of nanded district of maharashtra, geological survey of india records, v.123, part 6, pub. by govt. of india, pp. 81-83. radhakrishna murthy, i.v. (1998) gravity and magnetic interpretation in exploration geophysics, memoir no.40, geological society of india, pp. 363. raychaudhuri, j.k., saxena, n.p., sinha, a.k. and sen, d. (1984) deccan trap flows in parts of nanded and yeotmal districts, maharashtra, geological survey of india special publication no.14, pp. 6-8. 81 magnetic and palaeomagnetic studies as an aid in deciphering groundwater flow: a case study from deccan traps geociencias-vol 14-1 2010.vp earth sciences research journal earth sci. res. j. vol. 14, no. 1 (june 2010): 76-87 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt magdy a. atya1, olga a. khachay2, mamdouh m. soliman1, oleg yu khachay3, ahmed b. khalil1, mahmoud gaballah1, fathy f. shaaban1,4 , ibrahim a. el hemali1 1 national research institute of astronomy and geophysics (nriag), 11722 helwan, egypt. 2 institute of geophysics, russian academy of science (ras), ural division, russia. 3 ural state university, russia. 4 king khaled university, faculty of science, physics department, abha, saudia arabia. magdy@igse-eg.com abstract in the present work, we involve the control source electromagnetic (csem) technique to image the dynamic migration center of the near surface fractures, fissures, and cracks in a new dwelling area at 15th of may city close to cairo. this area forms the center of the zonal weakness of the subsoil, which in turn, interact with the weight center of the construction leading to catastrophic collapses. the control source electromagnetic technique has been developed recently to monitor the migration of the weakness center represented as the accumulative fissures and cracks in the near surface. three composite profiles of wide and planshet mesh data have been collected in 2008. this survey has been followed by performing two profiles in 2010; one of these profiles has been repeated to observe the situation changes, and the second profile has been measured between the other two profiles of 2008. the last profile had been performed to study the mutual relation between the measurements of the two cycles. the objectives of the whole process are to validate the capability of the technique to pick the minor changes of the weakness center, consequently, study its relation to the weight center of the adjacent construction to produce a recommended procedure to minimize the destruction resources at the site of investigation. the measured data has been interpreted and represented in graphs showing the distribution of the heterogeneity of the geoelectric parameters in the subsoil, furthermore, in a series of geoelectric cross section representing the applied frequencies used during the survey. the study concludes that (1) the center of the cracked zone is moving upward closer to the surface of the ground and heaver, (2) the water content is moving downward producing soil dryness at the shallow depths, and (3) the site became more stable in 2010 than 2008, however, the destruction resources remain warning with collapse events. furthermore, the situation reflects the relation between the water content and the changes in the weakness center. key words: electromagnetic technique, en helwan, weakness center, control source electromagnetic (csem) 76 manuscript received: 28/03/2010 accepted for publication: 20/05/2010 resumen en el presente trabajo, aplicamos la técnica de fuente electromagnética controlada (csem) para obtenerla imagen del centro dinámico de la migración de las fracturas, fisuras y grietas de la superficie somera en una nueva área de vivienda en la ciudad de 15 de mayo cerca a el cairo. esta zona constituye el centro de la debilidad del subsuelo, que a su vez, interactúacon el centro de peso de la construcción que ha llevado a colapsos catastróficos. la técnica de control de fuente electromagnética se ha desarrollado recientemente para monitorearla migración del centro de la debilidad, representados por la acumulación de fisuras y grietas de la superficie somera. tres perfiles compuestos se han recogido en 2008. este estudio se ha seguido realizando con dos perfiles en 2010. uno de estos perfiles se ha repetido para observar los cambios, y el segundo perfil se ha medido entre los otros dos perfiles de 2008. el último perfil se ha realizado para estudiar la relación mutua entre las mediciones de los dos ciclos. los objetivos de todo el proceso es validar la eficacia de la técnica para recoger los mínimos cambios del centro de la debilidad, y así, estudiar su relación con el centro de peso de construcciones cercanas, para recomendar un procedimiento que reduzca al mínimo la destrucción de los recursosen el sitio de la investigación. los datos medidos se haninterpretado y representado en gráficos que muestran la distribución de la heterogeneidad de los parámetros geo-eléctricos en el subsuelo; además, en una serie de seccionescruzadas geoeléctrica se representa las frecuenciasutilizadasdurante el estudio. el estudio concluye que (1) el centro de la zona agrietada se mueve hacia arriba cerca de la superficie de la tierra, (2) el contenido de agua está bajando, produciendola sequedad del suelo apoca profundidad, y (3) el sitio se volviómás estable en 2010 que en 2008, sin embargo, los producto de la destrucción siguen susceptiblesa colapsos.además, la situación refleja la relación entre el contenido de agua y los cambios en el centro de debilidad. palabras clave: técnica electromagnética, en helwan, centro de debilidad, control de código fuente electromagnética (csem) 1. introduction the idea behind the present paper reflects the capability of the applied and environmental geophysics to be involved in the community problems. accordingly, it is strongly recommended to establish a monitoring system for the detection of possible occurrence of disasters. currently, various kinds of geophysical systems are being operated for environmental monitoring purpose and detecting the subsurface temporal changes. the problem under investigation is related to the unstable subsurface. the subsurface heterogeneity caused by the intensity of the fractures, the fissures, the cracks, and the water content, is a factor of the instability level of the base ground beneath the constructions. such instability of the rock status resulted in several destructions at many places such as the stone block fallen at al deweaka (fig. 1) and at the second level of al mokkatam plateau, furthermore, the constructions deformation like that happened in the 15th of may city, quarter 27 (fig 2), and land subsidence like that occurred at several parts of wadi hof province and along the autostrade form al maadi to the margins of 15th may towards al tebbeen. the research layout is devoted to observe and monitor the near surface structures and to follow the growth of the intensity of the fractures, the fissures and the cracks. the water content and its salinity is also an important factor to be observed and followed. it is planned that, the observation and monitoring phase will run arbitrary at the start of the work to calculate the optimum measurement circulation comparable to the cyclic stress forces or loads affecting the near surface structures. then the situation will be subject of vector analysis to provide a management procedure to reduce the destruction resources and overcome or at least minimize the collapse. geophysics can easily detect the presence of different objects in wide range of applications. however, if those geophysical measurements are repeatedly obtained and compared, additional temporal dimension analysis makes interpretation and characterization of target much easier (hachay, 2004, hachay et al., 2007). the csem approach involves the possibility to survey the subsurface in volume block as wide profiling and planshet pickets along profiles. therefore, the technique is capable to produce an image of the subsurface representing 77 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt the near surface structural setting of the fissures, cracks, fractures, and also the hydrologic situation. getting enough information about the structural image of the surface including the heterogeneity intensity and the concentration depth of the fractures, cracks, and the development of their growth, and also the water content will help to provide an objective procedure assures the best management of the site, therefore, the catastrophic and dramatic accidental results will be minimized. the technique is proposed to be applied on a site with destruction resources that is warning with collapse, the site has been measured in 2008 and repeated in 2010. the comparative study of the subsoil images obtained in the two cycles represents enough information about the subsurface structural changes occurred during the time interval between the two cycles. the study concludes that, the center of weakness zone is moving upwards and became denser as the water saturation became less in 2010 than 2008. that means that the fracture contacts at the subsurface became closer to the surface and became heavier, while the water is moving downwards. the technical impact of this result is the site became more stable but still dangerous and is subject to collapse. 2. objectives the expected objectives and outcome of the present paper could be declared in the following items: 1) full understanding of the site of investigation, proposed here which is the quarter 27 at 15th of may city, and a more instable site in al mokkatam plateau. 2) a documentary archive of the situation including the hydrologic setting, the subsurface structure, the geoelectric heterogeneity move of the center, and upgradable conclusions. 3) vector analysis with locations of the forces applied to the area in reference to the center of heterogeneity. 4) a monitoring protocol and a management procedure for the site of investigation. 5) and finally the main object of the research which is the overcome or at least reduce the catastrophic impact via pre read of the destruction resources and probably prediction of the accident. 78 magdy a. atya, olga a. khachay, mamdouh m. soliman, oleg yu khachay, ahmed b. khalil, mahmoud gaballah, fathy f. shaaban, ibrahim a. el hemali figure 1. accidental fall down of stone block ad al deweaka, 2008. 3. geological background and site description the city of 15th of may is constructed over a limestone plateau on the eastern bank of the river nile (fig. 3). this plateau is covered mainly with wadi garawi and el-qurn formations of middle eocene age. el-qurn formation is of about120 m thick and is composed of chalky and marly limestone intercalated with shale, sandy marls and shale banks (strougo, 1985; said, 1990). in addition, these formations are highly jointed, highly fossiliferous, vuggy and sandy dolomitic limestone. structurally, the area is affected by three major sets of faulting systems striking mainly in nw-se, e-w, and ne-sw directions. all of them are normal faults, with dip ranges from 7u to 40u, where the beds are locally dragged. in addition to faults, there are at least many sets of joints dissect the area (moustafa et al., 1985). these joints trend in e-w and nw-se directions (moustafa et al., 1985; farag, and ismail, 1959). the major fractures in the area around cairo and its vicinity were interpreted based on satellite image of large scale (el-shazly et al., 1980). these fractures were analyzed statistically and each trend of fractures sets was treated separately. it was found that there are three main trends which are ew, wnw and nw. 4. theory and mathematics of csem 4.1 principal the csem provides an enhanced geophysical technique and device, in addition to, the precise analytical interpretation of the geophysical complexes for subjects such as the topic of this research, especially when the primary information about the site is not enough. the device has been established to use the planshet method of electromagnetic induction (hachay, 1997a, hachay and bodin, 1997b, hachay et al., 1999, and 2000, hachay, 2002) in the frequency domain. the method was adapted to map and monitor the high complicated geological mediums, to determine the structural factors and criteria of stability of the rock massif in the mine subsurface. the field observation and the way of interpretation make the new technology differ from other known earlier methods of field raying or tomography (atya et al., 2010, hachay and novgorodova, 1997c, hachay et al., 1999, and 2000, lau and cheng, 1977). the concept to construct the 3d geoelectric medium is based on interpreting the alternating electromagnetic field in the frame of a block-layered isotropic medium with inclusions over three stages (hachay 1997a, and 2002); in the first stage, the geoelectric parameters of the horizontal block-layered medium, which includes the block heterogeneities, are defined, in the second stage, a geometrical model of the different local heterogeneities or groups inside the block-layered medium is constructed based on the data of local geoelectrical heterogeneities, while in the third stage, the surfaces of the searched heterogeneities could be calculated in account of the physical parameters of the anomalous objects. 4.2 data collections the csem s*data collection is a measure of the geoelectric parameters of a medium, through which the electromagnetic waves propagates from the transmitter to the receiver. at each location of the transmitter and re79 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt divergence joints figure 2. constructions showing the joints of divergence at quarter 27, fifteenth of may city. ceiver, the three components of the electromagnetic field are measured. figure (4) explains the layout of the acquisition process. (a) shows an arbitrary position of the transmitter and receiver during a wide profile survey, (b) shows the measurement of the vertical component z, while the receiver coil is placed vertical, (c) shows the measuring of the first horizontal component h1, while the receiver coil is placed horizontal at right angle with the line between the transmitter and the receiver, and (d) shows the measuring of the second horizontal component, while the receiver coil is placed horizontal inline coaxial with the line between the transmitter and the receiver. the transmitter can inject more than a controlled frequency to scan different depths at the survey area. the measuring procedure could be classified into tow configurations, in correspondence to the objects of the survey. planshet data collections; is a procedure to collect the electromagnetic response over a 2d plan underneath the line between the transmitter and the receiver. figure (5) shows the process of gathering the planshet data. in this procedure, the transmitter and receiver are placed on the same line. the transmitter is placed at its first station, as the receiver moves stepwise at pre defined pickets along the profile. then the transmitter is moved to the next station and the receiver repeat the stepwise movement along the profile. this process is repeated until the last transmitter station. wide profile data collection; is a procedure employed to measure the electromagnetic response over a volume, therefore, it may be called volume csem. in this procedure, the transmitter is placed on a line, while the receivers are placed on a parallel line. both of the transmitter and receiver stations on the lines are marked as pickets. the survey volume is margined with the two lines. the data collected on two phases, the first is measures as the transmitter is placed on a line and the receiver is placed on a parallel line (fig. 6a), while in the second phase, the lines are exchanged and the transmitter is placed on the receiver line and vise versa (fig. 6b), the combination of the two phases produces the final image of the volume block. 4.3 interpretation for each array and fixed frequency ù two interpretation parameters are defined: � � � � �eff r r hz hr r h hr( ) ( / ) ( ) ( / ) %� � 2 100 where ñeff (r) is the apparent resistivity, r is the distance between the source and the point of observation, ù is the frequency f multiplied by 2ð, �hz� is the module of the vertical magnetic component, �hr� is the horizontal component of magnetic field oriented to the source, ä(r) is the parameter of geoelectrical heterogeneity, and �hö� is the second horizontal component of magnetic field perpendicular to the direction of the source. 80 magdy a. atya, olga a. khachay, mamdouh m. soliman, oleg yu khachay, ahmed b. khalil, mahmoud gaballah, fathy f. shaaban, ibrahim a. el hemali quarter 30 299.5n 299.5n 299n 299n 0 200m 111.0 112.0 111.9 110.0 110.0 105.0100.7108.0 108.0 120.0 59.2 112.0 103.0 100.0 100.0 buildings roads mediterranean sea cairo 15th of may city nile delta quarter 27 3 4 3 e survey site 3 4 3 .5 e su rv ey sit e 3 4 3 .5 e egypt location map s uez g ulf 3 4 3 e figure 3: location map of quarter 27 at 15 th of may city. that data are the information base for the further interpretation. on the first stage we define the geoelectrical parameters of the 1-d section for each array and each frequency after the preliminary filtration of the data: ñeff (r): ä(r)<a where, a is a level of data filtration to the area of inverse problem operator for one-dimensional medium (hachay, 1997a). the interpretation is made in a frame of n-layered model for each array and planshet location. after that each point of the planshet is associated with one and only column of layers thicknesses and corresponding fixed column with resistance of the medium in that layers. gathering information of all planshets together we obtain a many-valued function of each point – distribution of thicknesses and resistances of the medium layers. then we calculate the average value for these distributions for each point of the observation set. thus we obtain the unique distribution of thicknesses of horizontal layers and 81 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt (a) (b)(c)(d) transmitter receiver (a) the transmitter receiver layout. (b) measuring the vertical component, the coil is vertical. (c) measuring the first horizontal component, the coil is horizontal and prependicular to the trannsmitter. (d) measuring the second horizontal component, the coil is horizontal and inline to the trannsmitter. se co nd ho riz on tal co mp on en t fir st ho riz on tal co mp on en t ve rtic al c om po ne nt tr lay ou th for on e r ea din g figure 4. collection of the three component electromagnetic field in csem technique. ground surface subsoil medium interface transmitter receiver tr rays figure 5. the layout of the planshet data collection. resistances, which corresponds to the medium model as a cylinder with vertical axis and with a rectangle at the bottom and with a point of observation located in its center. thus we change over layered model to a block-layered model. then, gathering the values of thicknesses and resistances for all points of observation, located on one and the same profile we obtain the file of an average cross-section along the profile. the next step is combining the neighboring blocks with close-range values of resistance to one block. that operation is made according to the fixed scale of resistance. the second stage of interpretation is used to define the geometrical characteristics of conductive inclusions and their equivalent moments, which are proportional to the ratio of the conductivity difference in the host rock and in the inclusion to the conductivity in the host rock. here the approximation principle is used for alternating electromagnetic fields. the initial model of the inclusion is a current line of fixed length. that approximation construction is used for fitting of the average parameter of geoelectrical heterogeneity, which is calculated as an average value of ä in each point of the profile, located in each point of the profile (hachay, 2002). when the survey has been repeated to estimate the changing of the rock state, during the interpretation, we take all parameters from the previous cycle as the initial values for interpretation data for the next cycle. 5. field measurements in the present paper, two comparative data sets have been collected; the first data set has been obtained during the spring of 2008, and the second has been collected in the winter of 2009/2010. the first data set is composed of three wide profiles (fig. 7) employed to investigate the subsurface among the deformed buildings at quarter 27, fifteenth of may city, the applied frequencies are 80.15, 40.62 and 20.3. the second data set is composed of two wide profiles (fig. 7); the first profile is a repetition profile and the second is a new measured profile laid between two profiles of 2008. 82 magdy a. atya, olga a. khachay, mamdouh m. soliman, oleg yu khachay, ahmed b. khalil, mahmoud gaballah, fathy f. shaaban, ibrahim a. el hemali t1t2t3t4t5t6 transmitters line receivers line r1r2rn r1r2rn receivers line t1t2t3t4t5t6 transmitters line a b figure 6. volume csem data collection along wide profile. the applied frequencies to the system during the two cycles were: 40.62, 20.3, and 10.15 khz. 6. interpretation the two data sets have been analyzed and converted into graphs and cross sections representing the distribution of the geoelectric heterogeneity of the subsoil. in the present paper we give more consideration to the comparative tasks to estimate the discrepancies in the electromagnetic image that might have been occurred with the time interval of the two cycles. therefore, two comparative studies will be discussed; the first will be the repeated profile that has been measured in 2008 and 2010. the second will be between a profile from the data set of 2008 and another from the data set of 2010 measured between the profiles of 2008. firstly; the two profiles represent the same medium while the measurements have been taken over the same pickets in 2010 as in 2008. figure (8) shows the comparative distribution of the average parameter of the geoelectric heterogeneity, it elucidates that the level is higher in range for the cycle of 2010 than the cycle of 2008. this could be interpreted in reference to the soil material that the subsurface beneath the measured profile became less conductive and more resistive, which might be contributed to the migration of the underground water far away either laterally or into deeper depths. the graphs in figure (9) represent the deeper frequency 20.3 khz for the wide profile measured in 2008 and 2010. it could be easily noticed that, the medium became more resistive in 2010 than 2008. the center of the crack concentration at each picket became more regular, parallel, and closer to the ground surface. the start of the profile shows an area (pickets 1-7) where the cracks became denser making the zone more instable, while around the end of the profile; the crack situation produced more stability. the two geoelectric sections of the frequency 40.62 khz (fig. 10), they show that the irregularity is, generally, less, furthermore, it became more regular, but denser and closer to the ground surface in 2010 than 2008.the water content of the general medium became also less producing more resistive soil. secondly, the comparative study between a profile measured in 2008 and another measured in 2010 representing two different mediums. figures (11, 12 and 13) show that, the cross sections of the wide profile (7-8) are more conductive and the disintegration zones are more intensive and are located deeper than of the wide profile (3-4) (fig. 7). the state of the massif 83 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt 1 3 5 7 9 11 13 15 17 19 21 4 10 2 4 6 8 10 12 14 16 18 20 1 7 1 3 5 7 9 11 13 4 1 3 5 7 9 11 13 15 17 4 10 0 10 20 30 40 50 60 70 80 90 100 110 120 0 10 20 30 40 50 60 70 80 90 100 110 120 m wide profile (1-2) wide profile (7-8) wide profile (3-4) 2010 2008-2010 2008 2 0 0 8 w id e p ro fil e (5 -6 ) m figure 7. location map of the data sets measured in 2008 and 2010. along the new profile can be estimated as less stable than for the wide profile (1-2) and for the wide profile (3-4). the cross sections give a quantitative informative layout about the structure and state of the massif. having more cycles of observation, a stability table of stability of the rock state will be constructed. 7. conclusion the present work involves the control source electromagnetic approach to use the multi-frequency function for observing the near surface parameters and monitoring the change of geoelectric heterogeneity distribution. the techniques utilizes a series of fixed frequencies to probe the subsurface using the planshet picket profiling to measure the integrated image in depth plan beneath the transmitter to receiver distance, and the wide profiling to investigate the 3d block of the electromagnetic induction between the transmitter line to the receiver line. this survey way provides high resolution informative image about the depth of investigation. the area of investigation is laid at the quarter 27, fifteenth of may city, where the shallow surface of the ground is highly fractured producing a level of danger on the constructions and the block buildings at many sites of the quarter. the move of the fracture center and its combination with the center of weight applied on the ground by the construction loads results in the change of the destruction resources which might lead to collapse, landslide, or divergence over the joints of the blocks. two cycles of measurements have been collected in 2008 and 2010; the first was composed of three wide profiles investigated the area among the block buildings of quarter 27. the selected location of the profiles has been selected to cover areas where the constructions show divergence over the block joints and landslide and fractures. the second cycle of observation has been collected in the winter of 2009/2010 in form of two wide profiles; one of them is a repeat of one of the profiles of 2008 while the second is a new measurement, but lay between the profiles of 2008. 84 magdy a. atya, olga a. khachay, mamdouh m. soliman, oleg yu khachay, ahmed b. khalil, mahmoud gaballah, fathy f. shaaban, ibrahim a. el hemali 5 10 15 20 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 n 5 10 15 20 20 40 60 80 100 120 140 160 180 200 220 240 260 280 300 320 340 360 380 400 n 2 0 1 0 20.3 khz 40.62 khz 2 0 0 8 20.3 khz 40.62 khz deeper shallower figure 8. the comparison of the distribution of the average parameter of geoelectrical heterogeneity for 2 cycles of monitoring 2008 and 2010, quarter 27. 85 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt -2 -7 -12 -17 -22 -27 -32 -37 -42 0 10 20 30 40 50 m 60 70 80 90 1 3 5 7 9 11 pk 13 15 17 19 1 3 5 7 9 11 pk 13 15 17 19 -2 -7 -12 -17 -22 -27 -32 -37 -42 0 10 20 30 40 50 m 60 70 80 90 mo~ less .1 .1 .2 .2 .5 .5 .7 .7 1 1 2 2 4 4 6 6 10 10 or more frequency 20.62 khz 2008 0 20 50 90 150 200 300 400 500 750 1000 2000 5000 or more (0 m m). frequency 20.62 khz 2008 figure 9. the repetition wide profile at quarter 27, fifteenth of may city with 20.3 khz. 0 20 50 90 150 200 300 400 500 750 1000 2000 5000 or more (0 m m). 1 3 5 7 9 11 pk 13 15 17 19 -2 -7 -12 -17 -22 -27 -32 -37 -42 -2 -7 -12 -17 -22 -27 -32 -37 -42 0 10 20 30 40 50 m 60 70 80 90 frequency 40.62 khz 2008 frequency 40.62 khz 2008 mo~ less .1 .1 .2 .2 .5 .5 .7 .7 1 1 2 2 4 4 6 6 10 10 or more 0 10 20 30 40 50 m 60 70 80 90 1 3 5 7 9 11 pk 13 15 17 19 figure 10. the repetition wide profile at quarter 27, fifteenth of may city with 40.62 khz. 500 450 400 350 300 250 200 150 100 50 0 2 4 6 8 10 12 14 16 18 20 n 81.25 khz 40.62 khz 20.3 khz a figure 11. distribution of the average parameter of the geoelectric heterogeneities using the electromagnetic induction multi-frequency technique, a) wide profile (3-4) of 2008, and b) wide profile (7-8) of 2010. 220 200 180 160 140 120 100 80 60 40 20 0 40.62 khz 20.3 khz 10.15 khz 2 4 6 8 10 12 14 16 n b the data of the two cycles has been analyzed and comparatively interpreted to show the subsurface situations and enhance the relation between the two cycles. a series of 2d sections and analytical graphs have been constructed to show the distribution of the average geoelectric parameters and the fracture and crack distribution considering their weight over the depth of investigation. the interpretation of the two cycles showed that the technique is capable to image the subsurface change recorded in the two data sets. the results showed that the fractures and cracks became denser and their center is moving upwards to the ground surface, while the water content became less and moved deep ward. the rock state became more resistive and stable in 2010 than 2008, but is still suffering from the destruction resources. 8. recommendations it is high recommended to repeat the measurement over arbitrary time interval until reaching the optimum cycling of the force applied on the ground surface and cyclic motion of the center of the cracks and fractures. 86 magdy a. atya, olga a. khachay, mamdouh m. soliman, oleg yu khachay, ahmed b. khalil, mahmoud gaballah, fathy f. shaaban, ibrahim a. el hemali 1 3 5 7 9 11 13 15 17 19 21 -42 -37 -32 -27 -22 -17 -12 -7 -2 0 10 20 30 40 50 60 70 80 90 100 pk m m 0 20 50 90 150 200 300 400 500 750 1000 2000 5000 or more (om"m) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 -42 -37 -32 -27 -22 -17 -12 -7 -2 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 pk m m 0 20 50 90 150 200 300 400 500 750 1000 2000 5000 or more (om"m) a b mo~ less .1 .1 .2 .2 .5 .5 .7 .7 1 1 2 2 4 4 6 6 10 10 or more figure 12. geoelectrical section: a) along the wide profile (3-4) of 2008), and b) wide profile (7-8) of 2010) using the frequency 40.62 khz. 1 3 5 7 9 11 13 15 17 19 21 -42 -37 -32 -27 -22 -17 -12 -7 -2 0 10 20 30 40 50 60 70 80 90 100 pk m m 0 20 50 90 150 200 300 400 500 750 1000 2000 5000 or more (om"m) 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 -42 -37 -32 -27 -22 -17 -12 -7 -2 0 5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 pk m m 0 20 50 90 150 200 300 400 500 750 1000 2000 5000 or more (om"m) ba mo~ less .1 .1 .2 .2 .5 .5 .7 .7 1 1 2 2 4 4 6 6 10 10 or more figure 13. geoelectrical section: a) along the wide profile (3-4, 2008), and b) wide profile (7-8, 2010) using the frequency 20.3 khz. 9. acknowledgement the work has been done in cooperation of researchers from the national research institute of astronomy and geophysics (nriag), helwan, egypt and researchers from the institute of geophysics (igf), ural division (ud) of russian academy of sciences (ras), russia. the work has been financed by the local resources of nriag and the instrument belongs to igf. the authors thank the cooperation of the city office of 15th of may city. also, thanks are given for miss mona abu shady and the helping team for collecting the data of the first cycle. 10. references atya m. a., hachay o. a., abdel-lattif a. a., khachay o. yu., el-qady g. m., and taha a. i., 2010: geophysical contribution to evaluate the hydrothermal potentiality in egypt: case study: hammam faraun and abu swiera, sinai, egypt, earth sciences research journal, in press. el-shazly, e.m., el-ghawaby, m.a., salamn, a.b., khawas, s.m., el-amin, h., el-rakaiby, m.m, el-aassy, i.e., abdel-megid, a.a., and mansour, s.i., 1980: geological map of egypt, remote sensing center, academy of scientific research and technology, cairo, egypt. farag, i., and ismail, m., 1959: contribution to the stratigraphy of the wadi hof area (northeast helwan): bull. fac. sci., cairo univ., 34, 147–168. hachay o. a., 1997a: the three stage concept of common interpretation for 3-d electromagnetic and seismic fields and some results of its practical realization, engineering and environmental geophysics for the 21 st century, china, chengdu 286-292. hachay o. a., bodin v. v., novgorodova e.n., bodin vd. v., hachay a. yu., and hinkina t.a., 1997b: a new complex near-surface electromagnetic and seismic technique for 3-d research of inhomogeneous and unstationary medium, engineering and environmental geophysics for the 21 st century. china, chengdu 181-189. hachay o. a. and novgorodova e. n., 1997c: the experience of area induction research of sharp heterogeneous geoelectrical medium, fizika zemli 1997, 5, 60 -64. hachay o. a., novgorodova e. n, and bodin v. v., 1999: about the problems of near surface geoelectricity and some results of their solution, fizika zemli, 1999, 5, 47–53. hachay o. a., novgorodova e. n.and hachay a. yu., 2000: mapping of 3-d conductive zones with use of area systems of observation in a frame of 3-d frequency-geometrical method, geology and geophysics, 41, 9, 1331–1340. hachay o. a., 2002: new conception of active and passive electromagnetic 3-d monitoring of volcanoes and active faults, iii international workshop on magnetic, electric and electromagnetic methods in seismology and volcanology, moscow, ras 169 –171. hachay o.a., 2004: to the question of structure and state research of geological heterogeneous nonstationary medium in a frame of discrete and hierarchic model, russian geophysical journal, 33, 34, 32-37. hachay o.a., and hachay o. yu., 2007: research estimation and classification of stability of geological medium with use of data of active geophysical monitoring on the base of physical mesomechanics idea, physical mesomechanics, 10, 2, 87-92. hachay o.a., 2007: research of nonstability of the rock massive with use the method of active electromagnetic monitoring, physics of the earth, 4, 65-70. lau k. h. and cheng p., 1977: the effect of dike intrusion on free convection in conduction-dominated geothermal reservoirs, intern .j. heat and mass. trans., 20, 11, 197–209. moustafa, a., yehia, a., and abdel tawab, s., 1985: structural setting in the area east of cairo, maadi and helwan: middle east res. centre, ain shams univ., sci. res. series, 5, 40–64. said, r., 1990: geology of egypt, balkema pub., rotterdam, netherlands. strougo, a., 1985: eocene stratigraphy of the eastern greater cairo (gebel mokattam-helwan) area: middle east res. centre, ain shams univ., sci. res. series, 5, 1–39 87 csem imaging of the near surface dynamics and its impact for foundation stability at quarter 27, 15 th of may city, helwan, egypt geociencias-vol 12-1 sep.vp earth sciences research journal earth sci. res. j. vol. 12, no. 1 (june 2008): 107-117 seismic velocity determination in gravel and sands using piezocrystals armando luis imhof1 and juan carlos santamarina2 1 instituto geofísico sismológico volponi / facultad de ciencias exactas físicas y naturales. ignacio de la roza y meglioli. rivadavia. c.p. 5400, san juan, república argentina. fax: (54) 264 4234980. tel.: 54 264 4945015. e-mail: aimhof@unsj.edu.ar. 2 center for applied geomaterials research. atlanta, georgia. usa. http://geosystems.gatech.edu postal address: instituto geofísico sismológico volponi / ruta 12, km. 17. c.p. 5413 rivadavia san juan argentina abstract the exact determination of seismic waves’ propagation velocities has great importance in the geotechnics due to from that it is possible to determine, among other parameters, the dynamic ones: elasticity e, rigidity g, poisson !, compressibility b; as well as to reach a knowledge on the stress-strain behavior for the studied soil samples. the seismic waves transmission considered in tests at laboratory scale carried out in the present work is a phenomenon that produces very small deformation, and so doesn’t disturb the material. this allows to apply the results in a more general scale to study the behavior of soils in situ and to predict their answer to stress. with the purpose to study the response of particulate material subjected to seismic excitements at small scale, samples of gravels and sands were successively introduced in an odometric cell, exciting them with impulsive signals and registering the corresponding seismograms through general purpose piezoelectric transducers embedded in ends of the cell. the distance source-receiver was interval increased, which enabled, from the corresponding regression straight lines, to calculate in precise form the propagation velocities (for p waves). the tests were carried out in samples of dry alluvial soil with three different grain sizes. the respective frequency spectra of the signals were determined for two packing modes: loose and compact, what added information on the medium characteristics. 107 manuscript received april 18, 2008. accepted for publication june 3, 2008. the results showed that general purpose piezocrystals can be adapted for design and construction of a complete low cost acquisition system that brings great resolution in time, facilitating as consequence very precise calculations of the transmission velocities that take in a later stage to reliable determination of dynamic parameters of the soil. keywords: piezocrystals, seismic waves, compressional waves, signal processing. resumen la determinación precisa de velocidades de propagación de ondas sísmicas reviste gran importancia en la geotecnia debido que a partir de las mismas es posible determinar, entre otros parámetros, los dinámicos: elasticidad e, rigidez g, poisson �, compresibilidad b; así como alcanzar un conocimiento sobre el comportamiento tenso-deformacional de las muestras de suelo estudiadas. la propagación de las ondas sísmicas considerada en los ensayos de laboratorio llevados a cabo en el presente trabajo es un fenómeno que produce ínfima deformación, por lo que no disturba el material. ello permite aplicar los resultados en una escala más general para estudiar el comportamiento de los suelos in situ y predecir su respuesta a esfuerzos. con el propósito de estudiar la respuesta de materiales particulados sometidos a excitaciones sísmicas a pequeña escala, se introdujeron sucesivamente muestras de gravas y arenas en una celda odométrica, excitándolas luego con señales impulsivas y registrándose los correspondientes sismogramas a través de transductores piezoeléctricos de uso general adosados en extremos de la celda. si incrementó por intervalos la distancia emisor-receptor lo que posibilitó, a partir del trazado de las rectas de regresión correspondientes, calcular en forma precisa las velocidades de propagación (ondas p). los ensayos se llevaron a cabo en muestras de suelo aluvional seco con tres tipos diferentes de granulometría. se determinaron los respectivos espectros de frecuencia de las señales para dos estados de compactación considerados, lo que permitió aportar información sobre las características del medio en estado suelto y compacto. los resultados mostraron que los piezocristales de uso general pueden ser adaptados para diseño y construcción de transductores que funcionen tanto como emisores como receptores de ondas y que permitan obtener registros con gran resolución en tiempo, posibilitando como consecuencia de ello cálculos muy precisos de las velocidades de propagación que lleven en una etapa posterior a determinación confiable de parámetros dinámicos del suelo. palabras clave: ondas sísmicas, piezocristales, ondas compresionalesm, procesamiento de señales. introduction this investigation presents the development and implementation of geophysical techniques based in transmission of elastic waves using low cost instruments developed for these work. the techniques were applied for sand and gravels samples at laboratory scale. the precise determination of seismic wave velocities has great importance in civil engineering because from that it is possible to determine the dynamic modulus (e.g. rigidity g and compressiblility or bulk, b). see for example, the following expression (sheriff, 1994; sheriff et al, 1995): 108 armando luis imhof and juan carlos santamarina v b g p � � 4 3 " (1) which express compressional wave velocity (vp) related to bulk (b) and rigidity (g) dynamic modules, and density (") of the medium traversed by the wave. the fact to consider particulate media (i.e. not continuum) adds some level of complexity to equation (1), which transforms in (lee, 2003): v b g n s b s b n b p soils sk sk w a g � � � � � � � � �� � � � � � � � � �� #4 3 1 1 $ % % & ' ( ( � ( )1 1 n nsg w" " (2) where s is the degree of humidity of the soil sample; " density, n porosity, bsk bw bg bulk modules for skeleton (solid phase), water and grain respectively, and vp the propagation velocity of p waves. seismic waves propagation in these tests is a small deformation phenomenon, so does not disturbs the material involved. thereforre, it is possible to extrapolate the methodology to field to study gravel and sands behaviour in situ and predict their response to stress at major scale. the performance of geophysical tests in granular media at laboratory scale, particularly of coarse grain (e.g. alluvial gravels), faces difficulties related to the transducer-medium coupling, to the type of sources, the selection of transducers for the reception, the inherent dispersion (i.e. velocity variation with frequency) and the selective attenuation of frequencies with regard to the size of the particles, among others. test assembly main characteristics of emission-reception transducers general purpose piezocrystals (disk cell type; apci, 2000) of 20mm of diameter were used as sources and receivers of seismic waves, figure 1(a) . their reversible characteristic allows them to work converting electric energy in mechanical and vice versa so they can be implemented as emitters and receivers. an electric impulse makes vibrate the crystal mainly to its resonance frequency (wells, 1977), which is inversely proportional to the thickness of the piezoelectric material (being approximately a half of the wavelength, )/2). besides, the bandwidth of the registered signal also possesses an inverse relationship with the duration of the excitement (lee, 2003). the signal created by the signal generator (pulser) is not identical to that verily emitted by the transducer (that is a mechanical one), since it also depends on the proper characteristics of the medium. therefore it is important to carry out a monitoring of the real transmitted waveform. with that purpose it was implemented the system represented in figure 1 (b). the true signal emitted by the piezocrystal (pzt1 for brevity) is monitored by the digital storage oscil109 seismic velocity determination in gravel and sands using piezocrystals figure 1. (a) disk cells. (b) transducers connection to pulser and dso for real transmitted wave monitoring. loscope (dso) by means of other piezoelectric (pzt2), stuck to it. figure 2 presents on one hand the real form of the output pulse; and for other the frequency content of it. it’ll be studied what happens to the last as the waves transverse the medium under study. system assembly figure 3 shows the cylindrical steel cell where the samples were introduced, and the outline of system connection is visualized. as signal generator a pulser was built, which generated impulsive excitements of short duration (10*s) in periodic form (each 10 ms) and with high amplitudes (range 40 400v). connected to the pzt receiver, an operational amplifier fet tl084 was used of up to 80 db gain, fed with 12 vcc, with 100hz highpass filter. as seismograph, a 60mhz digital real time oscilloscope tektronix tds 210 of two channels and one additional for external trigger was used, with sampling rate up to 1 gsample/sec. the instrument possess an rs232 interface. a basic algorithm was designed and implemented to transfer data to pc, for later signal processing. as wave emitters e and receivers r, general purpose piezoelectric ceramics of 4.1khz of resonance frequency transducers were used, direct arrivals to be able to calculate the transmission velocities, it is necessary the determination of the first arrivals accurately (direct waves). therefore in confined media it is essential to know the types of possible arrivals in order to eliminate the undue ones (i.e. the no-direct ones). imhof (2007) studied several types of wave propagation inside cylindrical test steel probes: direct, refracted, reflected and transmitted through the walls of the cell, verifying that: (a) the waves transmitted by the steel walls will arrive much faster that the other ones, due to the high propagation velocity there (vsteel = 6000m/s; santamarina et al, 2001); (b) for distances h>0.13m and diameters + = 0.125m the wall refracted waves will reach r before the other types in all the cases ; (c) although the wall reflected 110 armando luis imhof and juan carlos santamarina figure 2. output signal from pulser. (a) real output wave. (b) spectral density. waves don’t arrive before the direct ones in any case, however as h increases the arrival times for both will approach to each other. this will imply interference in the complete registration of the signal. studying the directivity of this type of transducers, imhof (2007) demonstrated that, for distances h greater than 0.13m (figure 3) the wall refractions were not detected due to the fact that the critical angles were inside the area of crystal directivity shade, which attenuated them almost in 70-80%. at last, to avoid detection of waves that spread through the cell walls, so the base as the cover were isolated with rubber o-rings, that produced uncoupling of waves. methodology tests were carried out using dry samples of sands (s) and gravels (g) (sp and gp related to unified soil classification system, uscs, by american society of testing materials, astm) with three different sizes (#20 #50; #8 #20 and #2 #8, astm) and two packing modes: loose (a) and dense (c). table 1 summarizes sample characteristics which brought to six complete tests. at first velocities were determined, starting from a sample of sand (sp) with grain sizes among 0.297mm (#50) and 0.841mm (#20), repeating the determinations later for the other ones. the sample was introduced in the test cell (figure 3) until reaching the distance h = 70 mm. in the base a couple of pzt (source e, pzt1 and trigger to channel 1, dso, pzt2) was mounted, covering then the material with a steel cap with the receiving, r transducer stuck in its central part with an epoxi elastic mastic, and connected to channel 2, dso through the amplification system. after obtaining the first record the cap was removed and more sample material was added, until the next distance h, repeating the acquisition. in total there were six recordings for each sample and packing mode, for distances h=70, 90, 110, 130, 150, and 170mm. all the records were transmitted to pc through rs232 interface. the signal to noise (s/r) relationship was improved using stacking with the purpose to cancel or minimize the noises randomly and so enhancing the incoming signal. the number of stacks commonly used was 128, but on occasions, mainly for the #2-#8 samples, reached 1024. the sample was placed in two successive packing forms: loose a, poured with tablespoon taking care that the grains remained with the greatest volume of holes; and dense c compressing it in mechanical form in order to minimize the poral space without breaking the grains. 111 seismic velocity determination in gravel and sands using piezocrystals pulser dso amplifier (60 db) e r sample trigger cap (steel) h figure 3. cillindric cell for tests and connection scheme. table 1. volumetric form and shallow texture of soil samples used. (courtesy: instituto de materiales y suelos, fi-unsj) size particle diameter esfericity texture #20 #50 0.841 mm-0.297 mm 0.8 subrounded #8 #20 2.38 mm-0.841 mm 0.7 subangled to subrounded #2 #8 10 mm-2.38 mm 0.8 subrounded to rounded results the signals quality diminishes as the size of the particles grows, (e.g. figure 4 (a) vs figure 4 (c)). particularly it was not necessary to perform stacking to get the records for figure 4 (a)-(b), while the number of stacks to obtain the records shown in figures 4 (c)-(d) were of 128. finally to get legible records for sample #2-#8 (not shown) the number of stcaks ascended to 1024. the explanation of this phenomenon will be elucidated later when analyzing the frequency spectra for every signal. after the first arrivals’ picking for all signals, distance and time data were recorded in matrix form. linear regression was calculated for each one of them and then represented graphically together with corresponding travel times versus h distances (figure 5). the inverse of regression line slope constitutes the propagation velocity. 112 armando luis imhof and juan carlos santamarina 0 0.00125 0.00188 0.0025 0 0.5 p waves. sample #20-#50c 0 6.25 10� -4 0.00125 0.00188 0.0025 0 0.2 0.4 0.6 p waves. sample #20-#50a time [sec] 0 0.00125 0.00188 0.0025 0.05 0 0.05 0.1 p waves. sample #8-#20a 0 0.00125 0.00188 0.0025 0 0.1 0.2 0.3 p waves. sample #8-#20c (a) 6.25 10� -4 (b) 6.25 10� -4 time [sec] (a) (b) 6.25 10� -4 h=70 mmh=70 mm h=90 mmh=90 mm h=110 mmh=110 mm h=130 mmh=130 mm h=150 mmh=150 mm h=170 mmh=170 mm time [sec] time [sec] figure 4. compressional waves for increasing source-receiver distances (h). (a)-(b): sample #20-#50 (a and c); (c)-(d): sample #8-#20 (a and c). 113 seismic velocity determination in gravel and sands using piezocrystals 0 0.05 0.1 0.15 0.2 0 0.001 0.0015 measured fitting measured fitting linear regression.#20-#50 a e-r distance [m] t ra v e l t im e [s e c ] 0 0.05 0.1 0.15 0.2 0 0.001 0.0015 measured fitting measured fitting linear regression.#20-#50 c e-r distance [m] t ra v e l t im e [s e c ] 0 0.04 0.08 0.12 0.16 0.2 0 0.001 0.0015 measured fitting measured fitting linear regression.#8-#20 a e-r distance [m] t ra v e l t im e [s e c ] 0 0.04 0.08 0.12 0.16 0.2 0 0.001 0.0015 measured fitting measured fitting linear regression.#8-#20 c e-r distance [m] t ra v e l t im e [s e c ] 0.05 0.07 0.09 0.11 0.13 0.15 0 0.001 0.0015 measured fitting measured fitting linear regression.#2-#8 a e-r distance [m] t ra v e l t im e [s e c ] 0.05 0.07 0.09 0.11 0.13 0.15 0 0.001 0.0015 measured fitting measured fitting linear regression.#2-#8 c e-r distance [m] t ra v e l t im e [s e c ] 5 10� -4 5 10� -4 5 10� -4 5 10� -4 5 10� -45 10� -4 figure 5. linear regression. determination of propagation velocity (p wave) in samples #20-#50 (a-b); #8-#20 (c-d) y #2-#8 (e-f). (a y c). applying fast fourier transform (fft), the power spectra were evaluated for each one of the registered signals and represented in figure 6. finally, figure 7 shows the frequency content variation with distance h for the different cases under study. 114 armando luis imhof and juan carlos santamarina 70 mm. 90 mm. 11 0 mm. 130 mm. 150 mm. 170 mm. 0 6250 signal spectra sample #20-#50 c frequency [hz] 0 6250 1.25 10� 4 signal spectra sample #20-#50 a frequency [hz] p o w e r s p e c tr a signal spectra sample #8-#20 csignal spectra sample #8-#20 a signal spectra samples #2-#8 csignal spectra sample #2-#8 a p o w e r s p e c tr a 1.87 10� 4 2.5 10� 4 1.25 10� 4 1.87 10� 4 2.5 10� 4 (a) (b) p o w e r s p e c tr a 0 6250 frequency [hz] 0 6250 1.25 10� 4 frequency [hz] 1.87 10� 4 2.5 10� 4 1.25 10� 4 1.87 10� 4 2.5 10� 4 © (d) frequency [hz] 0 5000 1 10� 4 frequency [hz] 1.5 10� 4 2 10� 4 © (d) 0 5000 1 10� 4 1.5 10� 4 2 10� 4 figure 6. power spectra for increasing h. samples #20-#50; #8-#20 y #2-#8 (a and c). discussion compressional wave propagation the quality of the signals got worse as particle size increased. on one hand the fact to augment the particle size decremented teh contact area between the sample and the tyranbsducer (i.e. transducer-sample coupling). on the other the typical wavelength calculated for the samples (see below power spectra of signals) approaches to particles size, appearing the diffraction phenomenon (potts & santamarina, 1993). this turned critic in the case of sample #2-#8. in general it was observed that the measured travel times presented minimum dispersion, being adjusted very well to a straight line, because the material is considered homogeneous and isotropic (at laboratory small scale). the propagation velocities in all cases are lower than those of wave in air: 343 m/s to 20 ºc, 1atm (carmichael, 1982,1989; weast, 1988; guéguen & 115 seismic velocity determination in gravel and sands using piezocrystals 60 80 100 120 140 160 180 2000 3750 5500 7250 9000 frecuency response #20-#50 a y c h [mm] f re c u e n c y [h z ] 60 80 100 120 140 160 180 2000 3375 4750 6125 7500 frecuency response #8-#20 a y c h [mm] f re c u e n c y [h z ] 60 80 100 120 140 160 180 1000 2750 4500 6250 8000 sample a sample c sample a sample c frecuency response #2-#8 a y c h [mm] f re c u e n c y [h z ] figure 7. comparative diagram showing frequency response versus source-receiver distances. samples #20-#50; #8-#20 y #2-#8 (a and c). palciauskas, 1994), fluctuating in the range 150 215 m/s. it was verified that the propagation velocities in loose media are always lower than those in compacted samples. this appears as a logical result because the material with smaller volume of holes presents a more similar behavior to the solid (rock), that possesses greater propagation velocity that the wave in the air. the vp also diminishes as the grain size increases (compare figure 5 a,c,e). regarding sample #2-#8, it is very difficult to detect the first arrivals accurately. the explanation of this phenomenon is analized in the following section. power spectra of signals. the frequency content of the signals for the loose state was lower than that for the compact one, probably because the material in the last case resembles more to a continuous medium (compare in figure 6 the couples: (a)-(b) for sample #20-#50; (c)-(d) for sample #8-#20 and (e)-(f) for sample #2-#8). the maximum frequency content was obtained for the sample #20-#50 (a and c); being located in the range 4000hz (a) 8000hz (c), decreasing as the particle size increased to the interval 2500hz (a) 6000hz (c) for sample #8-#20 and rounding 2500hz fot sample #2-#8. therefore, as the particle size increases, decreases the sample frequency response, being accentuated the filtering of higher ones (low-pass filter). in the case of the sample #2-#8 a particular phenomenon takes place. according to the spectrum displayed at figure 6 (c)-(d), the average wave frequency is about f~4000 hz (h = 0.07m). considering a velocity of propagation vp~140 m/s: wavelength vp f m s s m mm� � � � �) 140 40001 0 035 3 5 / / . . also, the maximum amplitude values take place for smaller frequencies as h increases and therefore � also grows. the grain size for sample #2-#8 is among 2.38mm and 10mm; therefore ) rounds into the order of particles size. this implies the beginning of diffraction phenomenon that will hinder the detection of first arrivals due to scattering (potts et al, 1993; santamarina et al, 2005). figure 7 shows frequency response versus distance h for the samples #20-#50; #8-#20 and #2-#8 a and c, showing that the bandwith is greater in c that in a packing modes. besides, the frequency content decreases as h increases, logical result due that earth responds as a low-pass filter to seismic waves. only the sample #2-#8 presents irregular behavior, due to the described phenomenon. conclusions general purpose piezocrystals (disk cells) can be adapted for the construction of transducers to generate and detect seismic p waves at laboratory scale using soil samples (gravel and sands), in the range of frequencies that allows an appropriate balance between resolution and penetration. the propagation velocities calculated for the three samples in loose and compacted media increase in the second case; the typical frequency contents increase too. compact soil samples filter less the high frequencies than do the loose ones. the spectrograms amplitude peaks moves toward lower values of frequency as the grain size increases. finally, it was proved the efficiency of these type of piezocrystals to perform high resolutive and low cost laboratory experiments. references american piezo com international (2000) piezoelectric ceramics: principles and applications. apci. carmichael, r.s. (1982) handbook of physical properties of rocks. crc press, boca raton, 345 pp. 116 armando luis imhof and juan carlos santamarina carmichael, r.s. (1989) crc practical handbook of physical properties of rocks and minerals. crc press, 741 pp. guéguen, y. & palciauskas, v. (1994) introduction to the physics of rocks. princeton university press. princeton; 294 pp. imhof, a. l. (2007) “caracterización de arenas y gravas con ondas elásticas: tomografía sísmica en cross hole”. ph.d. thesis. universidad nacional de cuyo. mendoza argentina. lee, j.s. (2003) high resolution geophysical techniques for small-scale soil model testing. ph.d. thesis. georgia institute of technology. atlanta. usa. potts, b.d. & santamarina, j.c. (1993) geotechnical tomography: the effects of diffraction geotechnical testing journal. 16, nº1, 510-517. santamarina, j.c. & fratta, d.. (2005) discrete signals and inverse problems. an introduction for engineers and scientists. ed. wiley & sons.uk. santamarina, j.c.; klein, k.a. & fam, m.a. (2001) soils and waves. wiley & sons. ltd. england. sheriff, r.e. & geldard, l.p. (1995) exploration seismology. 2nd edition. cambridge university press; ny. 592 pp. sheriff, r.e. (1994) encyclopedic dictionary of exploration geophysics. geophysical reference series. society of exploration geophysics (seg). tulsa, ok. usa. weast, r.c. (1988) crc handbook of chemistry and physics. 1st student edition. crc press, boca raton. wells, p.n.t. (1977) biomedical ultrasonics. academic press. new york, 635 pp. 117 seismic velocity determination in gravel and sands using piezocrystals letter from editor a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombiacarlos alberto ríos reyes, oscar mauricio castellanos alarcón and akira takasu 7 material assimilation in a shallow diapiric forceful intrusion: evidence from microstructures and csd analysis in a porphyritic intrusive body, “la línea” tunnel, central cordillera, colombialorena rayo and carlos a. zuluaga 31 mineralization controls and petrogenesis of the rare metal pegmatites of nasarawa area, central nigeriaakintola, o.f. and adekeye, j.i.d. 44 denoising resistivity phosphate “disturbances” using haar mother wavelettransform (sidi chennane, morocco)saad bakkali and mahacine amrani 62 petrogenesis of the metacarbonate and related rocks of the silgará formation, central santander massif, colombian andes: an overview of a “reaction calcic exoscarn”ríos, c.a., castellanos, o.m., gómez s.i. and ávila, g.a. 72 seismic velocity determination in gravel and sands using piezocrystalsarmando luis imhof and juan carlos santamarina 107 orozco et al.indd earth sci. res. j. vol. 10, no. 2 (december 2006): 57-65 dissimilarity-based classification of seismic signals at nevado del ruiz volcano mauricio orozco1,3, marcelo e. garcía 2, robert p.w. duin 3, and césar g. castellanos1 1 universidad nacional de colombia sede manizales, grupo de control y procesamiento digital de señales, campus la nubia, km 7 vía al magdalena, manizales, colombia. 2ingeominas observatorio vulcanológico y sismológico de manizales, avenida 12 de octubre no. 15-47, manizales, colombia. 3information and communication theory group, mekelweg 4, 2628 cd delft, delft university of technology, the netherlands. corresponding author: mauricio orozco, e-mail: morozcoa@unal.edu.co abstract automatic classification of seismic signals has been typically carried out on feature-based representations. recent research works have shown that constructing classifiers on dissimilarity representations is a more practical and, sometimes, a more accurate solution for some pattern recognition problems. in this paper, we consider bayesian classifiers constructed on dissimilarity representations. we show that such classifiers are a feasible and reliable alternative for automatic classification of seismic signals. our experiments were conducted on a dataset containing seismic signals recorded by two selected stations of the monitoring network at nevado del ruiz volcano. dissimilarity representations were constructed by calculating pairwise euclidean distances and a non-euclidean measure on the normalized spectra, which is based on the difference in area between spectral curves. results show that even though euclidean dissimilarities have advantageous properties, non-euclidean measures can be beneficial for matching spectra of seismic signals. key words: classification, dissimilarity, nevado del ruiz volcano, seismic signals. resumen la clasificación automática de señales sísmicas se ha llevado a cabo típicamente sobre representaciones de características. trabajos de investigación recientes han mostrado que construir clasificadores sobre representaciones de disimilitud es una solución más práctica y, algunas veces, más precisa para ciertos problemas de reconocimiento de patrones. en este artículo consideramos clasificadores bayesianos construidos sobre representaciones de disimilitud. mostramos que tales clasificadores son una alternativa viable y confiable para la clasificación automática de señales sísmicas. nuestros experimentos fueron llevados a cabo sobre 57 earth sciences research journal manuscript received july 5 2006. accepted for publication october 24 2006. una base de datos que contiene señales sísmicas detectadas por dos estaciones seleccionadas de la red de monitoreo del volcán nevado del ruiz. las representaciones de disimilitud fueron construidas mediante el cálculo de distancias euclidianas y de una medida no euclidiana sobre los espectros normalizados, ésta última está basada en la diferencia de área entre curvas espectrales. los resultados muestran que aunque las disimilitudes euclidianas tienen propiedades ventajosas, las medidas no euclidianas pueden resultar benéficas para comparar espectros de señales sísmicas. palabras claves: clasificación, disimilitud, volcán nevado del ruiz, señales sísmicas. introduction nevado del ruiz volcano (nrv) is capped by a large volume of snow and ice, forming a glacier which has a volume of about 1200~1500 million cubic meters. nrv has three craters: arenas —the currently active vent—, and two parasite craters: olleta and piraña. since the start of digital recording at the volcanological and seismological observatory in manizales (vsom), a large and increasing amount of data has been recorded by the monitoring networks. classification of seismic signals is a crucial issue in order to discover the interaction between volcanic earthquakes and volcanic processes. the database available from vsom is suitable for applying automatic classification/learning techniques. in this study, we consider three classes of seismic signals originating from nrv: volcanotectonic (vt) earthquakes, long-period (lp) earthquakes and icequakes (ic); of course, for every seismically monitored volcano, seismologists use their own classification with more detailed descriptions of every subtype of earthquakes (zobin, 2003). vsom staff currently classifies volcanic earthquakes by visual inspection; such a method imposes a great deal of workload for the seismic analysts. in consequence, an automatic classification tool dramatically reduces this arduous task and also makes the classification reliable and objective, removing errors associated to tedious evaluations and changing of personnel. among the applications of pattern recognition techniques to volcanic-seismic signals, two recent works are highlighted: automatic classification of seismic signals at mt. vesuvius volcano, italy (scarpetta et al., 2005) and automatic classification of seismic events at soufrière hills volcano, montserrat (langer et al., 2006). both of them propose the application of artificial neural networks (ann) to classify seismic events. the former work uses a multilayer perceptron (mlp) to distinguish between vt events and transient signals due to other sources such as underwater explosions, quarry blasts, and thunders; spectral features and amplitude parameters are used for characterization. in the latter work, an ann is used to classify five fundamental types of signals: vt events, regional (re) events, lp events, hybrid (hb) events and rockfalls (roc); autocorrelation functions, high order statistical moments and amplitude ratios are introduced as features to the input nodes; a mismatch rate of 30% is reported, which was reduced up to 20% after a manual revision of the original a-priori classification. typically, in the context of volcanic seismology, neural networks have been preferred rather than other classical statistical pattern recognition methods; they are still being used for discrimination of seismic signals, including modifications in the feature-based representation (e.g. the modified approach used in benbrahim et al., 2005). the popularity of neural networks models to solve pattern recognition problems has been primarily due to their seemingly low dependence on domain-specific knowledge and also because of the availability of efficient learning algorithms (jain et al., 2000). recently, a number of studies showed advantages of learning from dissimilarity representations as opposed to learning from feature-based representations (duin et al., 1998; pękalska et al., 2001; pękalska and duin, 2002; paclík and duin, 2003; pękalska and duin, 2005). a dissimilarity representation of objects, seismic events in our particular case, is based on pairwise comparisons and can be expressed as a dissimilarity matrix , where each entry corresponds to dissimilarity between pairs of objects. dissimilarity representations are more general dissimilarity-based classification of seismic signals at nevado del ruiz volcano 58 than feature-based representations; in fact, the notion of dissimilarity is more fundamental than that of a feature (pękalska and duin, 2005). for dissimilarities the geometry is contained in the definition, giving the possibility to include physical background knowledge; in contrast to feature-based representations which usually suppose a euclidean geometry. this paper is devoted to explore dissimilarity representations to classify volcanic-seismic signals. in dealing with this particular problem, we advocate the dissimilarity-based classification of seismic signals as an advantageous and feasible alternative to the feature-based classification. dataset the signals were selected from data collected by the monitoring network deployed by vsom staff on nrv. the stations of the nrv monitoring network are located strategically; for instance near to glaciers and craters. signals from two stations (olleta crater station and glacier station) have been selected for the experiments because, according to the experiences reported by vsom staff, these two stations are a reference for the volcanic and ice-related events. the stations are located at distances of 4.08 km and 1.8 km from the active crater, respectively. signals were digitized at 100.16 hz sampling frequency by using 12 bits analogue-to-digital converter. a description of the nrv data set is provided in table 1. typical waveforms are shown in figure orozco et al., esrj vol. 10, no. 2. december 2006 figure 1. typical waveforms of the three classes considered: vt, lp and ic events. 59 1. table 1. composition of the data set (number of events per class). vt and lp events were detected by the olleta crater station. ic events were detected by the glacier station. vt lp ic 483 580 782 vt and lp earthquakes are commonly detected at nrv. in our experiments, the vt and lp events are analyzed as a separate problem (the ruiz-lp,vt two-class problem), and we also analize all classes together (the multi-class ruiz-all problem). dissimilarity representations and classifiers dissimilarity representations can be derived in many ways, e.g. from raw (sensor) measurements such as images, histograms or spectra or, from an initial representation by features, strings or graphs (pękalska et al., 2006); nonetheless, the particular way in which dissimilarities are computed is crucial, and relies on the additional knowledge that experts —volcanic seismologists— have about the problem. the spectra of seismic signals are commonly used for classification and monitoring of volcanic activity. since differences in spectral content allow a visual discrimination of different types of volcanic earthquakes (zobin, 2003), we have calculated the spectrum for each signal by using two different approaches: (i) n-point fast fourier transform (fft) and, (ii) parametric estimation of the power spectral density (psd). in such a way, we explore the difference between deriving dissimilarities from a data-based spectral estimation and from a model-based spectral estimation such as the yule-walker ar method. dc bias was removed before computing the spectra; in addition, when spectra are directly compared, they are required to be normalized and to have the same length. in consequence, considering the length of the shortest event and a length-resolution trade-off, we calculated 128point spectra. two different dissimilarities measures have been computed between spectra: (i) pointwise euclidean distance and (ii) area difference: the area of non-overlapping parts ( -norm) as shown in figure 2. dissimilarity-based classification of seismic signals at nevado del ruiz volcano figure 2. dissimilarity measured as the difference between normalized spectra. 60 the k nearest neighbor classifier (k-nn) k-nn is considered a direct approach for dissimilarity-based classification. this rule classifies a new object by assigning it the class label most frequently represented among the k nearest prototypes (e.g., by finding the k neighbors with the minimum distances between the new object and all the prototypes). for , the rule is called 1-nn. even though k-nn is asymptotically optimal in the bayesian sense, it is sensitive to noise and erroneously labelled prototypes. linear and quadratic normal density based classifiers previous studies (pękalska et al, 2001; pękalska and duin, 2002; paclík and duin, 2003) have shown that bayesian (normal density-based) classifiers, particularly the linear (ldc) and quadratic (qdc) normal based classifiers, perform well in dissimilarity spaces. for a 2-class problem, the ldc based on the representation set is given by (1) and the qdc is derived as (2) where is the sample covariance matrix, and are the estimated class covariance matrices, and and are the mean vectors, computed in the dissimilarity space , and are the class´ a-priori probabilities. if is singular, a regularized version must be used. in this study, the following regularization is used: . (3) we have fixed λ to be 0.01. nonetheless, this regularization parameter should be optimized in order to obtain the best possible results for the normal density-based classifiers. experimental results experiments were conducted to compare the results of the k-nn rule and the ldc and qdc classifiers built on the dissimilarity representations described above. experiments were performed 25 times for randomly chosen training and test sets. since we are particularly interested in accuracy recognition rather than computational complexity and storage requirements, the entire training set has been used as the representation set . nonetheless, may be properly reduced by prototype selection procedures (pękalska et al., 2006). training and testing sets were generated by selecting equal partitions for the classes. the results of our experiments are shown in figures 3 and 4. they present the generalization errors as a function of the number of randomlychosen training objects. figure 3 presents the results for four dissimilarity representations of the ruiz-vt,lp two-class problem; similarly, the results for the ruiz-all problem are shown in figure 4. standard deviations for averaged test errors decrease rapidly, varying around 0.02 after at least 10 training objects per class become available; for clarity reasons, standard deviations are not presented in figures 3 and 4. final errors and their standard deviations are summarized in table 2. orozco et al., esrj vol. 10, no. 2. december 2006 61 figure 3. ruiz-vt, lp data. average classification error of the bayesian classifiers and the 1-nn classifier in different dissimilarity spaces ( fft+euclidean, fft+areas, psd+euclidean and psd+areas) as a function of the number of prototypes per class. dissimilarity-based classification of seismic signals at nevado del ruiz volcano 62 figure 4. ruiz-all data. average classification errors of the bayesian classifiers and the 1-nn classifier in different dissimilarity spaces (fft+euclidean, fft+areas, psd+euclidean, and psd+areas) as a function of the number of prototypes per class. orozco et al., esrj vol. 10, no. 2. december 2006 63 table 2. averaged final classification error (in %) with its standard deviation for the four representations and the three classifiers considered. ruiz-vt,lp two-class problem representation k-nn ldc qdc fft + euclidean 19.85 (1.84) 14.79 (1.55) 16.36 (2.15) fft + areas 19.90 (2.04) 14.94 (1.00) 19.35 (1.98) psd + euclidean 17.99 (2.03) 13.35 (1.24) 15.45 (1.38) psd + areas 17.48 (2.09) 13.40 (1.06) 16.23 (2.05) ruiz-all problem fft + euclidean 24.53 (1.66) 18.57 (1.13) 19.31 (1.33) fft + areas 24.28 (1.52) 18.94 (1.02) 26.38 (2.73) psd + euclidean 27.76 (2.08) 21.19 (0.80) 24.10 (1.47) psd + areas 26.70 (1.67) 20.11 (1.25) 22.99 (1.85) discussion and conclusions we have explored and tested a dissimilaritybased strategy for classifying three different types of volcanic-seismic signals recorded by the monitoring network at nrv. two classification problems were conducted: a two-class problem including vt earthquakes and lp seismic events, and a multi-class problem including ice-related seismic events. four dissimilarity representations were derived by combining two different approaches for spectral estimation: n-point fft and parametric psd estimation, as well as two dissimilarity measures: euclidean distance and area difference between spectral curves. these dissimilarity representations allowed the usage of traditional statistical decision rules, particularly normal density based classifiers. the 1-nn rule was employed as a reference for performance comparison. the two-class ruiz-vt, lp problem seems the easiest because it contains signals detected and identified by the same station (olleta crater station); thus it is expected that sensor and noise conditions are the same influencing the subsequent steps for representation and classification. in addition, it is well know that, in general, multi-class problems are more difficult to deal with. for the two-class problem, experiments based on parametric psd estimation outperform those based on the fft. this makes sense because event lengths are, in general, short and, consequently, a parametric spectral estimation yields a higher resolution; in addition, the autoregressive methods (ar) tend to adequately describe spectra of peaky data, which is precisely the spectral nature of many volcanic-seismic signals. in contrast, for the multi-class problem, the fft yields better results but, in these particular cases, differences are not significant. our experiments confirm that bayesian classifiers outperform the 1-nn classifier, when a sufficient number of prototypes are provided. the ldc constructed on the different dissimilarity representations, for both ruiz-vt,lp and ruizall problems, always outperforms the 1-nn rule. ldc accuracies for the ruiz-vt,lp problem vary between 85% and 87% when the average classification error curve reaches a steady state. similarly, classification accuracies for the ruizall problem vary between 81% and 84%. qdc shows accuracy loss when certain number of prototypes is provided. therefore, a further study on a proper regularization for the qdc should be conducted in order to obtain an improvement of this classifier. ldc accuracies could be an intrinsic limit of our classification problem; however, a further study on other dissimilaritybased classifiers is needed as well as a re-analysis of the original a-priori classification, in order to find more suitable classifiers and to confirm the labels assigned by the experts. dissimilarity-based classification of seismic signals at nevado del ruiz volcano 64 acknowledgements this work was partially supported by the research project: “técnicas de computación de alto rendimiento en la interpretación automatizada de imagines médicas y bioseñales”. we thank the vsom staff for providing the raw data set, prof. ricardo henao for his valuable suggestions, and two anonymous reviewers for their useful comments. references benbrahim, m., daoudi, a., benjelloun, k., and ibenbrahim, a. (2005). discrimination of seismic signals using artificial neural networks, 2nd world enformatika conference, istanbul, turkey, february, 4-7. duin, r. p. w., de ridder, d., and tax, d. m. j. (1998). featureless pattern classification, kybernetika. 34 (4). 399-404. jain, a. k., duin, r. p. w., and mao, j. (2000). statistical pattern recognition: a review, ieee trans. pattern anal. machine intell. 22 (1). 437. langer, h., falsaperla, s., powell, t., and thompson, g. (2006). automatic classification and a-posteriori analysis of seismic event identification at soufrière hills volcano, montserrat, journal of volcanology and geothermal research. 153. 1-10. paclík, p. and duin, r. p. w. (2003). dissimilaritybased classification of spectra: computational issues, real time imaging. 9. 237-244. pękalska, e., paclík, p., and duin, r. p. w. (2001). a generalized kernel approach to dissimilarity based classification, j. mach. learn. res. 5. 175211. pękalska, e. and duin, r.p.w. (2002). dissimilarity representations allow for building good classifiers, pattern recognition lett. 23. 943-956. pękalska, e. and duin, r. p. w. (2005). the dissimilarity representation for pattern recognition: foundations and applications. world scientific, singapore. pękalska, e., duin, r. p. w., and paclík, p. (2006). prototype selection for dissimilaritybased classifiers. pattern recognition. 39. 189208. scarpetta, s., giudicepietro, f., ezin, e. c., petrosino, s., del pezzo, e., martini, m., and marinaro, m. (2005). automatic classification of seismic signals at mt. vesuvius volcano, italy, using neural networks. bulletin of the seismological society of america. 95 (1). 185196. zobin, v. (2003). introduction to volcanic seismology. elsevier, the netherlands. orozco et al., esrj vol. 10, no. 2. december 2006 65 the reservoir structural fractures have excellent fractal characteristics, as well as self-similarities. based on the fractal theory, the surface fractal characteristics of faults and the fractal characteristic of fractures in the core of the xujiahe formation in the fault-fold belt of the central yuanba area were studied, and a quantitative relationship was set up between them. based on the fractal characteristics of faults, predictions were made of the favorable fracture zone, which provides a new idea for the research of fracture, as well as offers theoretical references for exploring the fracture development zone during oil-gas exploration. the results show the following: the seismic value of reflection fault fractal dimension of the xujiahe formation is 1.5284; the correlation coefficient r2 is bigger than 0.9901; the capacity dimension linear regression correlation coefficient of the fracture in core of the xujiahe formation is bigger than 0.98; the fractal dimension d can well reflect the fault and fracture development degree, as well as the complexity of the fracture system; it can quantitatively calculate the density of the fracture of the reservoir in the area; the areas of capacity dimension bigger than 1.45 are the fracture development zones in the fault-fold belt of the central yuanba area; the oil and gas enrichment degree is high; the areas with the fractal dimension value between 0.95 and 1.45 are the fracture relatively-developed zones; the fractal dimension with values smaller than 0.95 are the lack of fracture areas. abstract keywords: fractal characteristics, structural fracture, xujiahe, formation, central yuanba fractal characteristics of reservoir structural fracture: a case study of xujiahe formation in central yuanba area, sichuan basin issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n2.72250 cunhui fan1,2, qirong qin1,2*, dongfeng hu4, xiaolei wang1,2, mengyue zhu1,2, wei huang1,2, yuxi li1,2, muhammad aqeel ashraf 3 1school of geoscience and technology; southwest petroleum university; chengdu 610500; china 2state key laboratory of oil and gas reservoir geology and exploitation, southwest petroleum university, chengdu, 610050, china 3international water, air & soil conservation society iwascon 59200 kuala lumpur, malaysia 4exploration company sinopec chengdu 610041, china corresponding author: 986336692@qq.com earth sciences research journal earth sci. res. j. vol. 22, no. 2 (june, 2018): 113-118 las fracturas estructurales de yacimientos tienes excelentes características fractales y autosimilares. basados en la teoría fractal, se estudiaron las características fractales superficiales de las fallas y las características fractales de las fracturas en el núcleo del cinturón falla-pliegue de la formación xujiahe, área central de yuanba (china). luego se estableció una relación cuantitativa para estas relaciones. según las características fractales de las fallas, se predijo la zona favorable de fractura, lo que ofrece una nueva idea para el conocimiento de esta y a la vez brinda referencias teóricas para la exploración de la zona de desarrollo de la fractura durante la exploración de petróleo y gas. los resultados muestran lo siguiente: el valor sísmico de la dimensión fractual de la falla en la formación xujiahe es de 1.5284; el coeficiente de correlación r2 es mayor a 0.99901; la dimensión del coeficiente de correlación de regresión lineal en el núcleo de la fractura de la formación xujiahe es mayor a 0.98; la dimensión fractal d puede reflejar la falla y el grado de desarrollo de la fractura, al igual que la complejidad del sistema de fractura; también puede calcular cuantitativamente la densidad de la fractura del yacimiento en el área; las áreas con capacidad mayor a 1.45 son las zonas de desarrollo de la fractura en el cinturón falla-pliegue del área central de yuanba; el grado de enriquecimiento de petróleo y gas es alta; las áreas con valor de dimensión fractal entre 0.95 y 1.45 son las zonas de fractura de desarrollo relativo; la dimensión fractal con valores menos a 0.95 son las áreas que carecen de fracturas. resumen palabras clave: características fractales; fracturas estructurales; formación xujiahe, yuanba central. record manuscript received: 14/11/2017 accepted for publication: 27/06/2018 how to cite item fan, c. h., qin, q. r., hu, d. f., wang, x. l., zhu, m. y., huang, w., li, y. x., & ashraf, m. a.(2018). fractal characteristics of reservoir structural fracture: a case study of xujiahe formation in central yuanba area, sichuan basin. earth sciences research journal, 22(2), 113-118. doi: http://dx.doi.org/10.15446/esrj.v22n2.72250 características fractales de fracturas estructurales de yacimientos: caso de estudio de la formación xujiahe en el área central de yuanba, cuenca de sichuan st r u c t u r a l g e o l o g y 114 cunhui fan, qirong qin, dongfeng hu, xiaolei wang, mengyue zhu, wei huang, yuxi li, muhammad aqeel ashraf 1. introduction oil and gas explorations concentrate on fracture-developed zones. tight sandstone reservoirs were regarded as their target areas. this type of reservoir comprises fractures with tectonic origins as their main seepage path and reservoir space (bahrami et al., 2012; abija & nwankwoala, 2018). the formation, development, and evolution of structural fractures are the critical issues and difficulties in oil and gas exploration and development, especially in the study of the characteristics and division of the formation stages of reservoir fractures. this research area has received considerable attention from domestic and foreign experts and scholars (larsen et al., 2010; nelson, 2001; ali et al., 2018). mandelbrot, a french mathematician, proposed the fractal theory to study a complicated phenomenon with similar features and suggested making a quantitative description of fractal complexity with fractal dimension (mandelbrot, 1982; roslee, 2018). scholz & aviles (1985) employed the fractal theory in the fracture study for the first time, providing a new mean for quantitative characterization of spatial distribution and the geometry of fracture structure. barton & hsieh (1989) made systematic research on the statistical law of rupture and cracks on field outcrops, in which he found fractal features of fracture distribution. zhang et al. (1996) reported that fractal dimension of tectonic fissures in an oil and gas reservoir reflects fracture development degrees of different lithological reservoirs. both fractures and cracks are a geological phenomenon and that rocks generate cracks or displacement under stresses. a self-similar structure is formed in a particular scale range (deng et al., 2006; rahim et al., 2018). the fractal dimension of fault and core cracks could contribute a quantitative description of the spatial development of cracks in reservoirs. recently, the fraction theory is widely applied in the geological field. it provides a new means for the quantitative characterization of spatial distribution and the geometry of fault structure, as well as a new idea for studying crack distribution in reservoirs (takayuki et al., 1987; usman et al., 2018). the yuanba area is located in the northeastern part of the sichuan basin. yuanba spans several administrative divisions of bazhong city, as well as the counties of cangxi, tongjiang, and nanjiang, among others. remarkable breakthroughs in oil and gas exploration have been achieved through perennial geological research in the continental and marine strata. commercial oil and gas were found in the wells of several series of gas-bearing formations, which includes: the lei-3 section of the leikoupo formation; second, third, and fourth sections of the xujiahe formation; zhenzhuchong and daanzhai sections of the ziliujing formation; and the middle jurassic qianfoya formation (zeng et al., 2010; nwankwoala & oborie, 2018). the productive natural gas flow was obtained in multiple wells. among these, the yl7 well in the third section of the xujiahe formation was tested and produced a gas flow of approximately 120.3×104 m3/d. these discoveries indicate good prospects for gas exploration. however, gas productivity varies dramatically from well to well. the daily gas output of a single well ranges from 0.1×104 to 120×104 m3/d. in the study area, the reservoir has low porosity and low permeability with fracturedominated productivity (zeng & li, 2005; zeng et al., 2008; jiang et al., 2004; veeraragavan et al., 2018). structural fractures have relative development due to the multi-stage tectonic deformation they undergo. fractures in different stages dominate and regulate hydrocarbon migration and accumulation. the presence of these fractures is likewise an indispensable condition for high gas production (zeng, 2008; raj & prabhakaran, 2018). therefore, the author used the fractal theory to study fractal features of cracks in reservoirs of the xujiahe formation, in the fault-fold belt of the central yuanba area. the capacity dimension on the t3x fracture and fractal dimension (d) of cracks on 12 cores were also calculated. fractal features of cracks were explored based on the relationship between core crack density of a cored well and the capacity dimension of cracks. the fractal characteristics of fractures and the threshold of the fractal dimension (d) of cracks in the study area were determined by combining the verification of oil and gas statistics of the study area. on this basis, crack distribution in the study area was predicted. 2 geological characteristic figure 1. map showing the general structural belts in the study area (large map) in regional tectonics, the yuanba area is located in the transition region between the frontal zone of micangshan and the gentle fold belt of the central sichuan basin, which connects the micangshan-dabashan foreland fold-thrust belt on the north and the longmenshan orogenic belt in the northwest (guo, 2011; lin et al., 2011; len et al., 2018) (fig.1). central yuanba lies close to the subsurface of yuanba, which connects the jiulongshan structure in the north and the tongnanba structure in the northwest. proximity to the subsurface causes the structural features of the study area to have a close genetic relation to its nearby regional structure. the surrounding structural belts of the longmenshan, dabashan, and micangshan exert varying degrees of influence on tectonic formation and evolution. ever since the indosinian movement, these surrounding structural belts have been reformed and superposed by multi-stage tectonic shifts. the study area and southern qinling in the yanshanian and himalayan movement stages were intensively compressed from the north to the south. the structural movement patterns of the himalayan scene inherit the characteristics of the yanshanian stage. the study area was thrust continually from the north to the south. thus, the structure was ultimately formed in the late-himalayan stage. the principal structures of the study area can be divided into two secondorder units: namely, low-flattish fold-thrust belts and syncline structural belts. the degree of the folds’ development is weak, and the structural height is not extremely obvious when compared with that of the nearby jiulongshan and tongnanba structures (roslee & tongkul, 2018). the major faults developed are mostly of three sets: namely, the nearly sn thrust faults and the nw and ne compresso-shear faults, which are mainly middle-shallow and middle-small faults. most faults can only break through the xujiahe formation because of the small scale and range. finally, the faults disappear upward in the zhenzhuchong formation and downward in the leikoupo formation. the faults that appeared on the seismic reflection profiles are mainly high-angle reverse faults. the reservoir lithology of the xujiahe formation in the central yuanba area mainly consists of medium-grained lithic sandstone, quartz sandstone, and siltstone. primarily, the sedimentary environment is a lake-deltaic sedimentary system, and the reservoir thickness is generally from 270 to 700m. the pore space of the reservoir is mainly composed of residual intergranular pore, intergranular dissolved pore, repitation, and fractures. the distribution of the reservoir porosity ranges from 0.61% to 6.88%, and the average value is 2.46%. permeability distribution ranges from 0.0012×10-3 to 26.0086×10-3 µm2 with an average value of 0.2944×10-3. the reservoir has the typical low porosity and low permeability characteristics of a tight sandstone reservoir (nwankwo & nwankwoala, 2018). the distribution of the middle and shallow hydrocarbon 115fractal characteristics of reservoir structural fracture: a case study of xujiahe formation in central yuanba area, sichuan basin figure 2. faults systems diagram of xujiahe formation in fault-fold belt of central yuanba area figure 3. characteristic of fractures of xujiahe formation in fault-fold belt of central yuanba area, a) low angle shear fracture yl2, t3x and b) low angle shear fracture yl2, t3x reservoir is controlled by a large area of overlapping, deltaic sand bodies. the development of fractures and faults provide favorable transportation conditions for hydrocarbon migration. some sand bodies that are blocked by a reverse-fault offer a favorable trap condition for hydrocarbon enrichment and accumulation (wali et al., 2018). located in the gently low porosity and low permeability hydrocarbon reservoir, the multiphase petroleum system with a complicated oil-water-air relationship finally forms. 3 self-similarity of middle xujiahe formation in yuanba reservoirs of the xujiahe formation in the intermediate zone of the fault-fold belt of the central yuanba area are common pore-crack reservoirs. crack development mainly determines oil-gas enrichment and productivity. as viewed from the causes, fault, and fracture are different fractural structures of rocks with different cracking degrees and displacement under the effect of a uniform, tectonic stress field. they have self-similarity and meet the fractal theory. 3.1 self-similarity of fault and fracture azimuth t3x reflects faults in the xujiahe formation of fault-fold belt of the central yuanba area, which mainly suffers tectonic extrusion of the longmenshan tectonic zone, dabashan fault zone, and micangshan tectonic zone. it is dominated by the nearly sn-oriented and nnw-oriented fracture system, followed by the ne-oriented faults (fig. 1). in the northeast area of the study area, considering the effect of the dabashan arcuate tectonic belt, the advantageous fault azimuth is in the nw direction. the advantageous azimuths of the yl3 and yb4 cracks in this region are in the nw direction and the nnw direction. the nw direction is likely to be the advantageous azimuth. influenced by the micangshan tectonic zone, the fault azimuth at the middle area is close to the sn and nnw direction. the advantageous azimuth is close to the sn direction, while the advantageous azimuths of the yl1 and yl5 cracks that are located in the middle zone are close to the sn direction (fig. 2). to sum up, the central crack azimuth of each well conforms suitably to the fault strike surrounding the borehole, which indicates the apparent self-similarity between fault and fracture layout. 3.2 self-similarity between fault and fracture combinations according to the rock sound emission test and analysis on homogenization temperature and the tectonic structure of inclusion, the tectonic system in this area was mainly developed during the late yanshan movement-early himalayan movement. during the late yanshan movement-early himalayan tectonic movement, this area, together with the south qinling mountains, suffered sturdy extrusion from the north to the south and formed the nw-oriented and ne-oriented fracture system under the nearly sn-oriented tectonic extrusion. during the middle and late himalayan tectonic movement, the roughly snoriented fault system was developed under the ne-oriented tectonic extrusion of the dabashan arcuate tectonic belt, which reconstructed and superposed the previous nw-oriented and ne-oriented fault system. as a result, fault often represents the convergence of the ne-oriented fault tail and the sn-oriented fault, or the convergence of the nw-oriented fault tail and the sn-oriented fault. the result of the faults shows a distinctive zigzag structure. this result reflects the recombination and superposition of two tectonic movements. moreover, the apparent x-shaped combination of faults could be seen along different fault extension directions. tectonic rupture and fault are material reflections at different stages of evolution from the same tectonic stress field. tectonic fracture has evident direction. it often extends along the fault zone or forms a sharp angle with the fault zone. the discovery of the fracture azimuth of the xujiahe formation was introduced based on imaging logging data of yl5 and yl2. fractures still extend mainly toward sn, nnw and ne directions, conforming to the fracture azimuth and height on the t3x reflection tectonic structure. core observation found that there are mostly two types of fractures. one is shear cracks. they often have two groups, and most are perpendicular to bedding. they extend stably with apparent striations and have broad distributions, presenting an “x-shaped” conjugate (fig. 3a). the other type of fracture is tension fissures. they are often indentation, manifested in tracing and reconstruction of the early two groups of shear cracks (fig. 3b). the self-similarity between fault and core fracture demonstrates that fault combinations are the product of the typical stress field. 4. fractal features of fault and core fracture 4.1 principle and process of fractal dimension according to the definition of mandelbrot (1982), the fractal has a similar composition to the whole in some way. he applied fractal dimension (d) to make a quantitative description of fractal complexity. for quantitative research on the fractal problem, a fractal set could be defined as the following: (1) (2) where nn is the number of objects with appointed features, rn is the radius of the particular object, d is the fractal dimension, and c is the proportional constant. calculate the natural logarithm to both sides of the equation (1), then, it can be seen from equation (2) that lnnn and lnrn are in a linear relationship, and d is the slope of the straight line. a physical model was designed according to the definition of fractal dimension and equation (1), and thereby d of each entrapment was calculated. there are a lot of methods to calculate the distribution fractal of reservoir fault and fractures, such as the box covering method, the cantors’ dust method, 116 cunhui fan, qirong qin, dongfeng hu, xiaolei wang, mengyue zhu, wei huang, yuxi li, muhammad aqeel ashraf the box-counting method. in this paper, the fractal dimension of fault and fractures was calculated by using the box covering method, which is easy to be implemented on the computer. elementary steps of this method include the following: use a square grid (side length=r) to cover the study area to make statistics on the number of squares covering fault zones and fractures (nn). next, change r gradually and make a statistic on corresponding nn. establish a rectangular coordinate system taking lnrn as the x-axis and lnnn as the y-axis, in which a regression analysis of statistical data is implemented by using the least square method. if fault and fracture distribution in the study area have fractal features, lnrn and lnnn shall meet the linear relationship of d=lnnn/δlnrn. at this moment, the slope of this regression line is d of each fault distribution. d of the fault and fracture is calculated similarly except for the different standard of r. figure 4. calculation of information dimension of faults table 1. fractal statistics of fractures of the reservoir in the study area figure 5. contour map of the distribution of faults capacity dimension of t3x 4.2 fractal features of fault this paper carried out different-scale gridding and calculated corresponding fractal dimensions to the t3x fault tectonic map (fig.3) of the fault-fold belt of the central yuanba area using the box covering method. in the t3x fault tectonic map, faults were vectorized with coredrawx6 and divided into 200 sub-regions (grids with an actual side length of 1.5km). the capacity dimension of each sub-region was calculated. subsequently, the plane distribution map was drawn by using sufer8.0. when calculating the fractal dimension of each sub-region, the subregion was covered by same-scale (r=1, 0.8, 0.6, 0.4, 0.25 and 0.2km) gridding. d of this tectonic map was calculated as 1.5284, and the correlation coefficient was r2=0.9901. when the minimum size of the box reaches the length of the smallest fault and the correlation coefficient is higher at 0.92, it can be viewed meeting requirements of statistical self-similarity. therefore, t3x faults in the study area have fractal features. figure 4 shows calculated results of the subregion in which the yb3 well lies. the absolute value of the straight slope is 1.5213, which is the capacity dimension of faults in this sub-region. after the information dimension of all sub-regions is calculated, the horizontal distribution of the information dimension of t3x faults in the study area could be drawn (fig. 5). 4.3 fractal features of fractures fracture pictures were collected from drum scanning of 18 rock samples of the t3x in the study area. the side length of the square gridding was set as r=2, 3, 4, 5, 8, 10, 14, 16, 18, 20, 25 and 30mm. the physical number of squares containing fractures was counted. the other calculation process is same as the calculation of the fractal dimension of fault. finally, the capacity dimension of fractures on each rock sample dh was calculated (table 1). according to the calculated results, all correlation coefficients are higher than 0.98, indicating the desirable fractal feature of the general fracture distribution of the core samples. therefore, distribution law of fractures could be predicted by using the fractal theory. due to different lengths of cores and different contributions of each section to the overall core density of the studied core segment, as well as the lack of a representative core section, the weighted average method was applied to calculate fracture density of the overall coring section of each well. in other words, the contribution of each rock sample to the fracture density of the whole well-section being studied was taken into account. this could be expressed as: where pi=li/l; pi is the weight of each rock sample; ρ1 is the coring density of the whole well-section being studied; ρ1i is the density of each rock sample; l is the total coring length of the whole study well-section, and li is the length of each coring. coring fracture densities of yl2 and yl3 were calculated from the equation, 2.01 fractures/dm2, and 3.43 fractures/dm2, respectively. 4.4 quantitative relationship between fault and fracture fractal features since fault and fracture have the same causes and self-similarity in term of fractal significance, fracture distribution could be predicted from fault distribution, as long as some internal quantitative relationship between them was discovered. the capacity dimension of the fault on the sub-region and the capacity dimension of fractures on the coring section were connected through fracture density on the core. finally, capacity dimension distribution of fault on the t3x tectonic map was transformed into the capacity dimension distribution of fractures. through statistical fitting on coring data (fig.7), it was discovered that there is some relationship between the fracture density of different wellsections on the core and the capacity dimension of corresponding fractures: where k is the fracture density on the core cross-section (fracture/dm2). dh = 0.0012k2 + 0.086k + 1.1422 the fracture density of the coring well and the capacity dimension of the fault (3) (4) 117fractal characteristics of reservoir structural fracture: a case study of xujiahe formation in central yuanba area, sichuan basin figure 6. the fitting relation between fracture face density and fracture capacity dimension figure 8. prediction graph of fracture favorable zones of t3x figure 7. the fitting relation between faults capacity dimension and fracture face density k = 8.6596d‘ – 11.014 on the sub-region of this coring well have the following fitting relationship: where d‘ is the capacity dimension of the fracture on the sub-region. the capacity dimension of the macro fault can be transformed into the capacity dimension of the micro-fracture through equation (1) and (2). fig. 8 shows the capacity dimension of the fracture transformed from that of fault. comparing fig. 6 and fig. 8, the region with higher d of fracture has higher d of fracture, indicating the higher development of fractures; the vice versa. 5 fractal dimension prediction and oil & gas productivity according to the distribution of discovered commercial hydrocarbon flow and oil & gas wells, regions with higher than 1.45 fractal dimension of fault often have better oil & gas flows. the fractal dimension of regions predicted favorable for fracture development from the above two formulas shall be higher than 1.4. hence, regions with d > 1.4 are defined as favorable regions for fracture development. regions with 0.95 < d < 1.4 are defined relatively favorable regions for fracture development, and regions with d < 0.95 are unfavorable regions for fracture development. in fig.8, regions with fracture development in the xujiahe formation of yuanba mainly concentrate within the isoline region (d > 0.95) and surrounding areas. for example, well-areas like yb3, yl3 and yl4 show a higher than 1.5 fractal dimension of fracture and enjoy high productivity of commercial hydrocarbon flow. this agrees well with fracture prediction. wells in the isoline region (d < 0.95), that is, predicted secondary fracture development regions, show no oil and gas. this confirms that the fractal theory is a useful technology to predict fracture development law. 6 conclusions 1.fault distribution and fracture distribution controlled by fault have remarkable self-similarity and fractal features. therefore, studying fracture development law with the fractal theory is a reliable, technological idea and an effective approach. 2.the capacity dimension of the fault and fracture distributions (d) is calculated based on the fractal theory by using the box covering method, which could reflect the development degree of fault and fracture, as well as the complexity of the fracture system. as a result, d could be used as one quantitative parameter to study spatial development and the distribution characteristics of fractures. 3.according to the prediction results, fault distribution and fracture distribution have evident fractal features. their correlation coefficients are all higher than 0.99. higher d represents dense fracture development. regions with d > 1.45 are favorable regions for fracture development and have reservoirs with good physical properties. these regions are favorable for oil and gas enrichment. regions with 0.95 < d < 1.45 are relatively favorable for fracture development. regions d < 0.95 are unfavorable regions for fracture development. acknowledgments this work was supported by open fund (pln 201718) of state key laboratory of oil and gas reservoir geology and exploitation (southwest petroleum university) references abija, f.a., & nwankwoala, h.o. (2018). characterization of aquifers in parts of abia state southeastern nigeria. earth sciences pakistan, 2(1), 18-22. ali, w., nasir, m.s., nasir, a., rashid, h., ayub, i., gillani, s.h., & latif, m.j. (2018). assessment of carbon footprints in terms of co2 of diesel generator, pakistan. earth sciences pakistan, 2(1), 15-17. bahrami, h., rezaee, r., & clennell, b. (2012). water blocking damage in hydraulically fractured tight sand gas reservoirs: an example from perth basin, western australia. journal of petroleum science and engineering, 88-89, 100-106. barton, c. c., & hsieh, p. a. (1989). physical and hydrologic flow properties of fractures. 28th international geological congress (5) 118 cunhui fan, qirong qin, dongfeng hu, xiaolei wang, mengyue zhu, wei huang, yuxi li, muhammad aqeel ashraf field trip guidebook t385, american, geophysical union, las vega s, nv. deng, p., chen, m., & yang, y. (2006). the application of fractal approach to the quantitative estimation research and evaluation of fractured reservoir (in chinese). petroleum geology & oilfield development in daqing, 25(2), 18-20. guo, t. l. (2011). reservoir characteristics and its controlling factors of the changxing formation reservoir in the yuanba gas field, sichuan basin, china (in chinese). acta petrologica sinica, 27(8), 2381-2391. jiang, l., gu, j., & guo, b. c. (2004). characteristics and mechanism of low permeability clastic reservoir in chinese petroliferous basin. acta sedimentologica sinica, 22(1), 14-18. larsen, b., grunnaleite, i., & gudmundsson, a. (2010). how fracture systems affect permeability development in shallow-water carbonate rocks: an example from the gargano peninsula, italy. journal of structural geology, 32(9), 1212-1230. len, n.l.s., bolong, n., roslee, r., tongkul, f., mirasa, a.k., & ayog, j.l. (2018). flood vulnerability of critical infrastructures review. malaysian journal of geosciences, 2(1), 31-34. lin, l. b., chen, y. q., dan, y., zhong, y. j., & xu, s. l. (2011). development characteristics of the xujiahe formation bauxite in the northeast of sichuan basin and its response to structural movement (in chinese). acta petrologica sinica, 27(8), 23922401. mandelbort, b. b. (1982). the fractal geometry of nature. san francisco, w. h. freeman & co. nelson, r. a. (2001). geologic analysis of naturally fractured reservoirs. boston: gulf professional publishing, 89-94. nwankwo, c., nwankwoala, h.o. (2018). gully erosion susceptibility mapping in ikwuano local government area of abia state using gis techniques. earth sciences malaysia, 2(1), 08-15. nwankwoala, h.o., & oborie, e. (2018). geological and hydrogeological characterization of a hydrocarbon impacted site in the niger delta. pakistan journal of geology, 2(1), 12-17. rahim, i.a., tahir, s., musta, b. & roslee, r. (2018). urbanization vs. environmental quali ty: some observation in telipok, sabah, malaysia. geological behavior, 2(1), 12-17. raj, n.j., & prabhakaran, a. (2018). lineaments of kodaikanal-palani massif, southern granulitic terrain of tamil nadu, india: a study using srtm dem and landsat satellite’s oli sensor’s fcc. geology, ecology, and landscapes, 2(3), 188-202. roslee, r. (2018). geohazards in sandakan town area, sabah, malaysia. geological behavior, 2(1), 18-23. roslee, r., & tongkul, f. (2018). engineering geological assessment (ega) on slopes along the penampang to tambunan road, sabah, malaysia. malaysian journal of geosciences, 2(1), 06-14. scholz, c. h., & aviles, c. a. (1985). fractal dimension of the 1906 san andreas fault and 1915 pleasant valley faults. earthquakes notes, 55, 20 takayuki, h., takashi, s., keisuke, i. (1987). fractal structure of spatial distribution of microfracturing in rock. geophysical journal international, 90(2), 369-374. usman, m., khalid, m.b., yasin, h., nasir, a., & arslan, c.h. (2018). impact of industrial effluents on ground water qualitya case study of gujranwala, pakistan. pakistan journal of geology, 2(1), 18-20. veeraragavan, s., duraisamy, r., & mani, s. (2018). prevalence and seasonality of insect pests in medicinally important plant senna alata l. under tropical climate in the coromandel coast of india. geology, ecology, and landscapes, 2(3), 177-187. wali, e., phil-eze, p.o., & nwankwoala, h.o. (2018). saltwater freshwater wetland ecosystem and urban land use change in port harcourt metropolis, nigeria. earth sciences malaysia, 2(1), 01-07. zeng, l. b. (2008). fissure and its seepage characteristics in lowpermeable sandstone reservoir (in chinese). chinese journal of geology (scientia geologica sinica), 39(1), 12-17. zeng, l. b., & li, x. y. (2005) fractures in sandstone reservoirs with ultra-low permeability: a case study of the upper triassic yanchang formation in the ordos basin, china (in chinese). aapg bulletin 27(7), 14–16. zeng, l. b., & li, y. (2010). tectonic fractures in tight gas sandstones of the upper triassic xujiahe formation in the western sichuan basin, china. acta geologica sinica 84(5), 1230-1238. (in chinese) zeng, l. b., gao, c. y., qi, j. f., wang, y. k., li, l., & qu, x. f. (2008). the distribution rule and seepage effect of the fractures in the ultra-low permeability sandstone reservoir in east gansu province, ordos basin (in chinese). science in china, series d: earth sciences, 51, 44-52. zhang, j., tian, g., & liu, j. (1996). a fractal analysis on structure fractures of reservoirs (in chinese). petroleum exploration and development, 23(4), 65-68. keywords: soil gas radon; emanation; lima; lr115 detector; niño costero; river floods; alluvial deposits; igneous rocks. palabras clave: radón en gas del suelo; detector lr115; emanación; lima; niño costero; inundaciones fluviales; depósitos aluviales; rocas ígneas. how to cite item vilcapoma, l., lopez-herrera, m. e., pereyra, p., palacios, d. f., perez, b., rojas, j., sajobohus, s. (2019). measurement of radon in soils of lima city peru during the period 20162017. earth sciences research journal, 23(3), 171-183. doi: https://doi.org/10.15446/esrj. v23n3.74108 lima city is situated on alluvial fan deposits of rivers flowing through geological formations that contain different levels of uranium. in this paper, a study is made on the average spatial and temporal behavior of radon gas in soils of lima city. radon concentration was determined using the lr-115 type 2 track detector during 36 periods, of 14 days each, in twenty holes distributed in the fifteen districts of lima city. radon concentration in soil pores ranged from 0.1 to 64.3 kbq/m3 with an average value of 5.6 kbq/m3. the average radon concentration in soil gas was about two times lower in winter than in the other seasons. high radon values during october/november 2017 were related to the earthquakes perceived in lima city in that period. the highest radon concentrations were found in areas of alluvial deposits whose parental material has been removed from the quilmaná and huarangal volcanics by the chillón and huaycoloro rivers. soil gas radon concentrations were even higher in areas closer to volcanic and less distant from rivers. during the period of maximum flooding of the chillón, rímac and lurín rivers, due to the natural phenomenon “el niño costero”, anomalous high soil radon concentrations were observed in most of the measurement sites located near rivers. these high radon values were associated with ground vibrations caused by rock and debris avalanches in rivers and creeks. abstract measurement of radon in soils of lima city peru during the period 2016-2017 medición del radon 222 en suelos de la provincia de lima perú durante el período 2016-2017 issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n3.74108 la ciudad de lima está situada sobre depósitos de abanicos aluviales de ríos que fluyen a través de formaciones geológicas que contienen diferentes niveles de uranio. en este trabajo se realiza un estudio sobre el comportamiento espacial y temporal promedio del gas radón en suelos de la ciudad de lima. la concentración de radón se determinó utilizando el detector de trazas lr-115 tipo 2 durante 36 períodos, de 14 días cada uno, en veinte hoyos distribuidos en los quince distritos de la ciudad de lima. la concentración de radón en los poros del suelo estuvo comprendida entre 0.1 y 64.3 kbq/m3 con un valor promedio de 5.6 kbq/m3. la concentración promedio de radón en el gas del suelo fue aproximadamente dos veces menor en invierno que en las otras estaciones. los altos valores de radón durante octubre/noviembre de 2017 se relacionaron con los terremotos percibidos en la ciudad de lima en ese período. las concentraciones más altas de radón se encontraron en áreas de depósitos aluviales cuyo material parental ha sido removido de los volcánicos quilmaná y huarangal por los ríos chillón y huaycoloro. las concentraciones de radón en el gas del suelo fueron incluso mayores en áreas más cercanas a los volcánicos y menos distantes de los ríos. durante el período de máxima inundación de los ríos chillón, rímac y lurín, debido al fenómeno natural "el niño costero", se observaron altas concentraciones anómalas de radón en el suelo en la mayoría de los sitios de medición ubicados cerca de los ríos. esos altos valores de radón se asociaron con las vibraciones del suelo causadas por avalanchas de rocas y escombros en ríos y quebradas. resumen record manuscript received: 08/08/2018 accepted for publication: 21/06/2019 earth sciences research journal earth sci. res. j. vol. 23, no. 3 (september, 2019): 171-183 lázaro luís vilcapoma1, maría elena lópez herrera1, patrizia pereyra1, daniel francisco palacios1, bertin pérez1, jhonny rojas1, laszlo sajo-bohus2 1pontificia universidad católica del perú, lima, peru 2universidad simón bolívar, caracas, venezuela * corresponding author: dpalaciosf@pucp.edu.pe en v ir o n m en ta l g eo lo g y 172 lázaro luís vilcapoma, maría elena lópez herrera, patrizia pereyra, daniel francisco palacios, bertin pérez, jhonny rojas, laszlo sajo-bohus introduction radon gas is produced by the radioactive decay of 226ra, which in turn is derived from the 238u series. uranium is found in small quantities in all soils and rocks, although the amount varies from place to place. since its discovery, at the beginning of the 20th century, the radon gas 222rn (indicated as rn or radon in the text), has awakened considerable interest due to its harmful effects on human health and its applications in geosciences. in the present study two areas of interest related to radon are combined: as a health hazard and as a natural tracer. the connection of radon with uranium makes it useful as a pathfinder element when searching for uranium rich minerals and indirectly, to identify radon-prone areas that may have the potential to generate high indoor radon levels. radon generated within the mineral grains of rocks and soils can escape from them by recoil into air or fluid filled cracks and into the contiguous pore space, respectively. this can be quantified by the emanation coefficient, which depends on the availability of 226ra, porosity, moisture content and temperature (nazaroff, 1992). subsequent transport from the place of escape into the gas/fluid phase towards the earth surface occurs by diffusion and convection/advection. radon transport is influenced not only by local geology as rock types, but by other geophysical and geochemical parameters as joints, fractures, hydrology, porosity, permeability and presence of other gas carriers. these characteristics are, in turn, determined by the character of the bedrock, glacial deposits, or transported sediments from which the soil was derived (gundersen et al., 1992; alonso et al., 2019). soil-gas radon concentrations can vary in response to climatic and weather changes on hourly, diurnal, or seasonal time scales (sundal et al., 2008; zafrir et al., 2012). the most important meteorological factors appear to be barometric pressure, wind, relative humidity, rainfall, and temperature. geology is the most important factor controlling the source and distribution of radon (appleton & miles, 2010). the release of radon from rocks and soils is controlled largely by the types of minerals in which uranium and radium occur. granite, metamorphic rocks, black shales, and some volcanic rocks may be enriched in uranium; these rocks, and the soils derived from them, are the most common sources of radon gas (sprinkel & solomon, 1990). uranium concentrations in igneous rocks are usually higher compared to sedimentary ones, excepting shale, deep-sea sediments and phosphate rocks, while metamorphic rocks have concentrations similar to those of the rocks from which they were derived. the content of uranium in igneous rocks depends on the origin and composition of the rock, for example, high silica igneous rocks contain relatively higher u concentrations compared to such basic low silica igneous rocks as basalt (iaea, 2014). because of its inert gas geochemistry and its radioactivity, radon has been recognized as a potential tracer of fault systems (al-tamimi & abumurad, 2001; walia et al., 2008), as an important tool to forecast earthquakes and volcanic eruptions (zmazek et al., 2005; neri et al., 2006; giammanco et al., 2007; pérez et al., 2007), as a good indicator of hydrocarbon leakage from reservoirs (matolin & stráník, 2006; palacios et al., 2016), to wide range of applications as a tracer in marine and hydrological settings. several studies have demonstrated that radon concentrations in soil and groundwater are controlled not only by types of rocks and geohydrological characteristics of rock mass but also by tectonic processes (choubey et al., 1999; choubey et al., 2010). landslides, which result due to geodynamic processes, have also been considered as a source of radon (purtschller et al., 1995). recent studies revealed that spatial distribution of indoor radon concentration in lima city can be represented by two different areas: one with concentrations <100 bq/m3 and the other in the range (100 300) bq/m3 (pereyra et al., 2018). several factors may have influenced on this distribution, however the two major factors determining indoor radon concentration are the entry rate and the ventilation rate. the entry rate is the most important factor in determining houses with high radon levels (nazaroff & nero, 1988). it depends on soil gas radon concentration, on soil physical characteristics such as permeability, on meteorological circumstances such as temperature and pressure, and on building-related factors. geogenic radon potential (grp) provides information about the source of indoor radon concentration and depends on the equilibrium activity concentration (rn concentration in soil gas at about 1 m depth), and soil permeability. criteria have also been established for radon hazard assessment considering only the soil radon as the source of high indoor radon concentration (åkerblom et al., 1984; snih, 1992). most of the area of lima city is built on alluvial deposits originating by the chillón and rímac rivers, which eroded, transported and deposited materials from the geological formations present in their beds. throughout their courses, these rivers interact with rock formations of different geological origin, and different chemical (e.g., uranium content) and mineralogical composition. it is to be expected, then, that one of the main source of variation of radon concentration in soil pores beneath and around dwellings in lima city could be related to the type of rock originating the alluvial deposits. thus, some association could be found between radon indoors and main parent material of the alluvial soil on which dwellings are located. indoor radon measurements in lima city shows the highest concentration in houses located at northeast of the rímac district (pereyra et al., 2018). in this area, landslides and avalanches of rocks, debris and sediments caused by the floods of the rímac and huaycoloro rivers have historically occurred. although the alluvial parent material in the lima city area is related to the occurrence of two lithologies, with rocks containing different amounts of uranium, the relationships between uranium content in parent material and soil gas radon in a given area is not obvious, since many factors and processes may contribute to influence the latter. in this paper a study is made on the average spatial and temporal behavior of radon in alluvial soils of lima city in order to evaluate the possible relationship between soil radon concentration and the origin of deposited parent material. in addition, to evaluate if temporal variations of soil gas radon were related to the seismic events and to the natural phenomenon “el niño costero” occurred during measurements. material and methods characteristics of the study area lima city is located at the west central peru, in a desert stripe between the pacific ocean and the andes, mostly on flat terrain of the peruvian coastal plain within the valleys of the chillón, rímac and lurín rivers (fig. 1). the chillón river is to the north, the rímac river in the center and the lurín river is to the south. these rivers have contributed significantly to the erosion and transport of loose materials, resulting from the erosion of hills present along their channels. the geology of the lima region reflects the subduction of the nazca plate under the south american plate and consequent uplift and volcanic activity. this subduction zone is constantly generating seismic activity. the geomorphology of lima is a tectonic and plutonic result superimposed by the processes of geodynamics. according to the lithology, in the northern area of lima there are volcanic rocks along the chillón river resulting from the presence of the chain of huarangal and quilmará volcanoes, formed at the end of the mesozoic and beginning of the cenozoic age (80 million years ago) and currently all are inactive (hakl et al., 1992). the alluvial cones were the result of the accumulation of debris formed by blocks, clasts and gravels of volcanicsedimentary origin with silty sandy matrix. in the central part of lima, where the rímac river flows, we find intrusive and metamorphic rocks, with little presence of plutonic rocks, predominating granodiorites that are found along the course of the huaycoloro river (tributary of the rímac river). from the upper part to the dejection cone, the rímac river and its tributaries have eroded shale sedimentary rocks, paleogene-neogene volcanics, and middle cretaceous to upper and triassic-jurassic limestones. at lower elevations and farther to the west, intrusives emerge at the edges of the dejection cone of the rímac river, mostly of upper cretaceous intermediate composition. in the area of lima city, volcanic rocks of the middle cretaceous casma formation, compact sedimentary rocks of sandstones, shales, and lower cretaceous limestones emerge (palacios et al., 1992). unlike the main parent material deposited by the chillón river, most of the material deposited in the alluvial fan of the rímac river generally has low uranium content. in the south, the lurín river, with similar characteristics to the rímac river, transports meteorite material from the hills to the lurín valley (farmland) until it reaches the sea (see fig. 1). lima city has an arid climate, with deficiency of rainfall throughout the year, maintains a semi-warm temperature and a humidity of 64% to 75%. because it usually does not rain and the atmospheric pressure practically does not change, the observed seasonal changes are possibly due to changes in temperature. 173measurement of radon in soils of lima city peru during the period 2016-2017 radon measurement method a hole of about 1.8 m deep was drilled at each site where radon will be monitored. this depth ensures the measurement of the equilibrium activity concentration, a determining factor of the grp. the hole was sheathed with a pvc tube approximately 2 m long. the tube was inserted inside the hole and dropped to its bottom. the space left between the tube and the subsoil is filled with part of the extracted soil material. an integrating track detector, cellulose nitrate lr-115 type 2, was fixed at a distance of 20 cm from the bottom of the hole staying its sensitive side facing the soil radon emanating surface. in this configuration, the lr-115 detector was exposed just to the volume of gas within 20 cm under it, registering only the alpha particles emitted from the effective volumes of each airborne radionuclide. the effective volume of 222rn is located approximately between 1 3 cm from detector surface, but approximately between 3 6 cm and 6 8 cm for the most energetic radon and thoron progenies, respectively (rojas et al., 2018). fig. 2 shows details of the installation of pvc tube in the hole and the sketch of the monitoring system. during the exposure period of detector, a cap screwed tightly on top of the tube housing isolates it from the outside. bore-holes closed from environmental air exchange do not allow the soil gas to flow freely into the atmosphere, therefore, the rate of diffusion controls the radon migration process (shweikani & hushari, 2005; papp et al., 2008). the detector assembly enclosed in a pvc pipe housing reduces the influence of environmental factors such as temperature, air pressure, wind and moisture. using similar procedures, other authors have considered the method as stable since the atmospheric parameters have little influence on the radon concentration inside the hole (ciotoli et al., 1998; king & minissale, 1994; papp et al., 2008). the distance of the lr-115 detector to the bottom of the hole causes negligible contribution of thoron gas to the track density because it decays before reaching the effective volume. the distribution of 220rn atoms exhaling uranium occurrences in peru are distributed according to different geological environments (magmatic-sedimentary) that do not obey exclusively to a geological time. in lima region, five prospects of uranium deposits of relatively low average grade (0.060) were identified, totalling 10,556 metric tons of u3o8 (rivera, 2010). they are located in the area of volcanic rocks northeast of lima city (fig. 1), particularly in andesitic lavas of quilmaná volcanic (upper cretaceous) that emerge in the chillón river (palacios et al., 1992; perales, 1994). in this area, which has the casma group as the main host rock, deposits of massive volcanogenic sulphides of pb-zn-cu were formed (chira et al., 2018), confirming the behavior of these elements as pathfinders associated with uranium (kurt et al., 2009). the quilmaná and huarangal formations present in the chillón and huaycoloro river basins are part of the casma group, which consists mainly of volcanic materials including rocky fragments of shales and sandstones, and andesitic lavas (villacorta et al., 2015). previous analysis indicates that erosion materials deposited in the dejection cone of chillón and rímac rivers are associated to rock formations with different uranium contents. due to the natural phenomenon “el niño costero”, intense anomalous rainfall caused multiple phenomena of mass movements (debris and sludge flows, landslides, rock falls, river erosion of hillsides, etc.) and floods (huaycos), which affected a large part of lima city in summer of 2017. the erosion of slopes in the outcrops of volcanic and intrusive rocks, very fractured and weathered, provided material for transport and subsequent deposition. the most important forces and phenomena related to the physical processes of huaycos might be gravitational acceleration down slopes, the resistance experienced by the flow and the processes of erosion and deposition of particles, by which the flow exchanges mass with the underlying boundary. however, fundamental understanding of the mechanisms of particle erosion from an underlying substrate by a huayco is lacking (hogg, 2018). figure 1. map of lima city quadrangle showing its lithology (palacios et al., 1992), the main rivers that formed the alluvial fan deposits on which lima city is situated, and spatial distribution of the 20 measurement points locations (holes) in the 15 districts. 174 lázaro luís vilcapoma, maría elena lópez herrera, patrizia pereyra, daniel francisco palacios, bertin pérez, jhonny rojas, laszlo sajo-bohus spatial distribution of measurement sites radon measurements in soil pores were carried out in places near and far to chillón, rímac and lurín rivers. the selection of the observation sites was conceived in order to cover all districts of lima city. to get an idea of the potential contribution of soil to indoor radon, 20 measurement sites were distributed in the 15 districts of lima city. detectors located inside the pvctubes were coded from s1 to s20. table 1 shows the coding of each one and the distances of each hole to the nearest riverbank. the minimum and maximum distances between a hole and the riverbank were 0.4 km and 9.4 km, respectively. the holes were drilled in ground of residential houses to ensure the ease and safety of access for the recovery and periodic change of detectors. figure 1 also shows the locations of the measurement sites (tip of the arrows). previsions were taken so that measurement sites were not affected by rains. exposure time and detector processing all detectors were exposed simultaneously during periods of 14 days in the pvc tubes installed permanently in soil. data collection was conducted over 36 consecutive periods in the 20 holes from april 2016 to september 2017 (table 2). after exposure, detectors were chemically etched in 2.5 n naoh solution at 60°c during 90 minutes (fleischer et al., 1975; salama et al., 2006; mehta et al., 2014; pereyra et al., 2015). etched detectors were analyzed in an automatic reading system (polytrack) to determine the net track density. the calibration factor that converts the track density rate to time-averaged radon concentration was determined in a radon chamber. environmental conditions within the chamber were similar to those existing in measurement holes. reference radon concentration in the chamber was continuously measured by means of the alphaguard df 2000 monitor, traceable to the primary rn standard in physikalischtechnische bundesanstalt (ptb), braunchweig, germany. from soil surface decreases exponentially with the distance to the bottom of the hole, in a manner analogous to what happens in front of a wall (godoya et al., 2002; hadler & paulo, 1994; chauhan et al., 2017) or inside a diffusion chamber (urosevic et al., 2008; nikezic & stevanovic, 2007). since the main mechanism of gas transport within the pvc tube is diffusion, very few thoron atoms can reach 20 cm from the source because their diffusion length is approximately 2-3 cm for the commonly used air diffusion coefficient values. for example, assuming a diffusion coefficient in air of 1.2·10-5 m2/s, the diffusion lengths (l d =  ) for 220rn and 222rn would be 3.1 cm and 239.2 cm, respectively (huxol et al., 2012). thus, at a distance (d) of 20 cm from a surface only 0.16% of the exhaled 220rn atoms would arrive ( n n d l0 = −     exp ), while 92% of the exhaled 222rn atoms could reach the same distance (bonotto & andrews, 1999). the first 220rn progeny (216po) behaves as its progenitor due to its very short half-life. the rest of the 220rn progeny, and practically all 222rn daughters, are almost completely deposited on all internal surfaces within the pvc tube due to the plate-out effect (hadler & paulo, 1994; paulo et al., 2001; nikezic & stevanovic, 2007) such as the bottom hole, inner surfaces of the pvc wall and lid, and surfaces of detector support. in short, due to very short half-life (t1/2=55.6 s), 220rn atoms cannot be homogenously distributed into the pvc tube and decreases exponentially with the distance from the bottom of the hole in opposite to the 222rn atoms (t1/2=3.82 d) (urosevic et al., 2008). previous considerations seem to indicate that radon is the main contributor to track density in the lr-115 detector, so in this study no filtering membrane was used at the entrance of the radon cup. kumar et al. (2017) used the same principle for soil radon measurement and discrimination of radon and thoron concentrations in soil gases. furthermore, some negative aspects of the filtering membranes, such as large variation of their gas permeability with temperature and reduction of detector sensitivity to radon (fleischer et al., 2000; tommasino, 2016), supported the decision of not using them. in summary, using the described setup the measurement of 222rn is practically not influenced by the 220rn soil gas concentration. figure 2. installation of a measuring point in san juan de lurigancho district (left), and schematic view of the hole with closed pvc tube and the radon measurement system (right). 175measurement of radon in soils of lima city peru during the period 2016-2017 results and discussion comments on results of soil gas radon concentrations and on the measurement methodology the minimum and maximum values of soil gas radon concentration in the study area were 0.1 kbq/m3 and 64.3 kbq/m3, respectively, which show great variability of the data. the average radon concentration ranges from 1.5 kbq/m3 in winter to 11.9 kbq/m3 in spring. the average concentration of rn in the 20 holes during the entire observation time was 5.6 kbq/m3. there is no history of rn measurements in soil gases in lima city. the obtained results are compatible with those reported in similar studies, both using active (choubey et al., 2005; szabó et al., 2013; duggal et al., 2014; cinelli et al., 2015) and passive methods (kareem, 2016). they are also similar to those found in zones of fluvial deposits and poorly developed soils whose associated lithologies correspond to plutonic rocks, volcanic magmatic rocks and differentiated magmatic rocks (alonso et al., 2019). nevertheless, the average radon concentration of 5.6 kbq/m3 is quite low compared to that reported by cothern & smith (1987). the relatively low value could be table 1. codes of measurement sites and distances of each detector hole to the nearest riverbank. site code district latitude longitude nearest river distance (km) s1 chorrillos -12.17 -77.02 surco 0.73 s2 huaycan -12.02 -76.82 rímac 3.00 s3 santa eulalia -11.90 -76.66 santa eulalia 0.17 s4 cuspanca (chosica) -11.89 -76.65 santa eulalia 0.06 s5 villa el salvador -12.24 -76.92 lurín 2.81 s6 villa maría del triunfo (1) -12.19 -76.94 lurín 7.48 s7 villa maría del triunfo (2) -12.15 -76.93 lurín 8.72 s8 san juan de lurigancho (3) -11.94 -76.99 rímac 9.40 s9 san juan de lurigancho (1) -11.97 -77.00 rímac 6.13 s10 san juan de lurigancho (2) -12.01 -77.01 rímac 1.90 s11 chaclacayo -11.98 -76.80 rímac 0.40 s121 ate -12.04 -76.93 rímac 2.06 s13 manchay -12.10 -76.88 lurín 7.06 s14 independencia -11.98 -77.04 chillón 6.84 s15 carabayllo -11.86 -77.06 chillón 2.67 s16 puente piedra -11.93 -77.10 chillón 1.56 s172 san martín de porres -11.99 -77.09 rímac 4.45 s18 pueblo libre -12.07 -77.07 rímac 4.47 s19 jesús maría -12.09 -77.06 pacific coast 2.44 s203 miraflores -12.11 -77.04 surco 4.74 1also at 3.9 km of confluence of rivers huaycoloro and rímac 2also at 5.34 km from chillón river 3also at 1.00 km from pacific ocean coast table 2. codes of measurement periods. code from/to/year code from/to/year code from/to/year p1 ap. 17/may 01/2016 p13 oct. 2/oct. 16/2016 p25 mar. 19/ap. 2/2017 p2 may 1/may 15/2016 p14 oct. 16/oct. 30/2016 p26 ap. 2/ap.16/2017 p3 may 15/may 29/2016 p15 oct. 30/nov. 13/2016 p27 ap. 16/ap. 30/2017 p4 may 29/jun. 12/2016 p16 nov. 13/nov. 27/2016 p28 ap. 30/may 14/2017 p5 jun. 12/jun. 26/2016 p17 nov. 27/dec. 11/2016 p29 may 14/may 28/2017 p6 jun. 26/jul. 10/2016 p18 dec. 11/dec. 25/2016 p30 may 28/jun. 11/2017 p7 jul. 10/jul. 24/2016 p19 dec. 25/2016/jan. 8/2017 p31 jun. 11/jun. 25/2017 p8 jul. 24/aug. 7/2016 p20 jan. 8/jan. 22/2017 p32 jun. 25/jul. 9/2017 p9 aug. 7/aug. 21/2016 p21 jan. 22/feb. 3/2017 p33 jul. 9/jul. 23/2017 p10 aug. 21/sep. 4/2016 p22 feb. 3/feb. 19/2017 p34 jul. 23/aug. 6/2017 p11 sep. 4/sep. 18/2016 p23 feb. 19/mar. 5/2017 p35 aug. 6/aug. 20/2017 p12 sep. 18/oct. 2/2016 p24 mar. 5/mar. 19/2017 p36 aug. 20/sep. 3/2017 176 lázaro luís vilcapoma, maría elena lópez herrera, patrizia pereyra, daniel francisco palacios, bertin pérez, jhonny rojas, laszlo sajo-bohus explained considering that most measurement sites meet one or more of the following points: 1. disequilibrium between radium and uranium due to leaching action (valid for all alluvial surfaces), 2. although the permeability in alluvial soils is fairly high, low emissions of radon can occur in deposits of parent material of volcanic origin due to the low specific surface area of the uranium minerals, 3. generally fairly low uranium concentration in the parent material of plutonic origin, 4. very low bedrock, 5. low moisture content of the soil most of the time. one might also think that the relatively low value is because measurements made with passive nuclear track detectors do not represent the exact radon concentration in soil gas. however, szabó et al. (2013) found even lower values of average radon concentrations in soils with similar characteristics to the alluvial soils of lima city, and measurements were made by an active method. furthermore, the relative low average radon concentration in soil pores is in agreement with the average radon concentration in dwellings of lima city (83 ± 11 bq/ m3), also relatively low (pereyra et al., 2018). the minimum value, (11 ± 6) bq/m3, was found in coastal districts away from rivers, while the maximum, (407 ± 55) bq/m3, was in the rímac district located to the northeast of lima province, at the right margin of the rímac river, in an area formed predominantly by andesitic volcanic rocks interbedded with sandstones and siltstones. the obtained maximum value (64.3 kbq/m3) is much higher than the median of made ground, typical of urban areas, even higher than the median of undisturbed soil according to values reported by wiegand (2001). elío et al., (2015) found lower values of radon by passive etched track detectors than those obtained using an active monitor, but, unlike the conditions in our study area, radon measurements were carried out for monitoring geological storage of co2, i.e., detection of co2 leakages, where co2 acts as carrier gas. other authors have used active and passive systems to measure soil gas radon and radon emanation factor, and results were in general agreement with each other (giammanco et al., 2009, lópez-coto et al., 2009). the values measured by different methods depend, among others, on the physical characteristics of soils. for example, ruckerbauer & winkler (2001) found no significant differences for active and passive procedures in a porous soil. nevertheless, passive procedures yield soil radon concentrations significantly lower than obtained with the active procedure in an impermeable clay soil. it is important to note that soils of the study area in lima city are mostly porous and permeable. a more detailed discussion on the use of active and passive methods for radon measurements in soil gas can be found in palacios et al., (2016). time series of soil gas radon concentrations relationships whit seasons and some meteorological parameters time series of the average soil radon concentrations obtained biweekly from the twenty positions are shown in fig. 3. medium-term (the measurement periods) and seasonal changes in rn activity concentration in borehole air were observed. apparently, seasonal changes in radon activity concentration in borehole air were associated with atmospheric temperature changes. winter is characterized by approximately twice less average soil gas radon concentration than the average for the other seasons. during the total period of measurements the atmospheric pressure remained almost constant, and rainfall occurred only during the “el niño costero” event. descriptive statistics of radon concentration and some meteorological parameters is presented in table 3. meteorological parameters represent average values of data reported by nine meteorological stations distributed nearly uniformly in the study area. radon levels in soil gas and atmospheric pressure were the variables that showed the highest and lowest variability, respectively. variation of meteorological parameters (normalized to their maximum values) is shown in fig. 4. table 3. descriptive statistics of radon concentration and meteorological variables. variable mean sd min max cov(%) rn concentration (kbq/m3) 5.6 2.9 1.5 11.9 51.8 humidity (%) 76.4 5.1 66.6 83.1 6.7 temperature (°c) 19.8 3.3 15.9 25.9 16.7 pressure (mb) 985.4 1.3 983.1 987.0 0.1 dispersion graphs and the moderate correlation coefficients show the low relation between radon and meteorological variables (fig. 5). from fig. 5 it is clear that the measured radon concentration shows low negative, positive, and negative correlations with relative humidity, temperature, and atmospheric pressure, respectively. these results support 13 12 11 10 9 8 7 6 5 4 3 2 1 0 r n (k b q/ m 3 ) observacion period p1 p2 p3 p4 p5 p6 p7 p8 p9 p1 0 p1 1 p1 2 p1 3 p1 4 p1 5 p1 6 p1 7 p1 8 p1 9 p2 0 p2 1 p2 2 p2 3 p2 4 p2 5 p2 6 p2 7 p2 8 p2 9 p3 0 p3 1 p3 2 p3 3 p3 4 p3 5 p3 6 figure 3. average rn concentration per observation period in the 20 wells. the horizontal solid line represents the radon concentration value averaged over the entire measurement period and the dotted lines represent the deviation (s) from the average value. 177measurement of radon in soils of lima city peru during the period 2016-2017 that the variation of soil gas radon concentration in time is only weakly influenced by the meteorological conditions at the surface. however, the level of significance (p) was less than 0.05 in all cases, indicating statistically significant correlations. pca analysis of the data was performed to investigate the variability, both seasonal and throughout the study period, of radon concentration, relative humidity, temperature and barometric pressure. the results are summarized in table 4. in most cases the variability of the four parameters showed differences between different seasons of the same year and between the same seasons of different years. from the higher variability of radon concentration compared to the variability of humidity, temperature, and barometric pressure it can be concluded that there should be other agents governing the radon concentration variation besides the analyzed meteorological variables. 1.5 1.4 1.3 1.2 1.1 1 0.9 0.8 0.7 0.6 0.5n or m al iz ed t em pe ra tu re , h um id ity a nd p re ss ur e observation period p 1 p 2 p 3 p 4 p 5 p 6 p 7 p 8 p 9 p 10 p 11 p 12 p 13 p 14 p 15 p 16 p 17 p 18 p 19 p 20 p 21 p 22 p 23 p 24 p 25 p 26 p 27 p 28 p 29 p 30 p 31 p 32 p 33 p 34 p 35 p 36 figure 4. plot showing the temporal variation of average values of pressure, humidity and temperature. table 4. results of the principal component analysis. time interval season percentage of variance (%) radon concentration humidity temperature barometric pressure p1 p5 autumn 87.3 11.3 1.4 0.1 p6 p12 winter 42.6 28.1 20.6 8.7 p13 p18 spring 73.9 20.1 5.9 0.1 p19 p24 summer 66.2 21.9 10.6 1.4 p25 p31 autumn 87.5 7.9 4.1 0.5 p32 p36 winter 58.7 33.8 7.2 0.3 p1 p36 all periods 76.0 20.0 2.8 1.2 14 12 10 8 6 4 2 0 -2 -4 14 12 10 8 6 4 2 0 -2 -4 10 15 20 25 3060 66 72 78 84 90 r = 0.35 p = 0.031 r = -0.36 p = 0.027 temperature (˚c) r n (k b k/ m 3 ) r n (k b k/ m 3 ) humidity (%) 14 12 10 8 6 4 2 0 -2 -4 982 984 986 988 990 r n (k b k/ m 3 ) pressure (mb) r = -0.38 p = 0.023 figure 5. dispersion graphs, pearson correlations (r) and p-values (p) for soil gas radon and meteorological variables. confidence level for ellipses is 95%. 178 lázaro luís vilcapoma, maría elena lópez herrera, patrizia pereyra, daniel francisco palacios, bertin pérez, jhonny rojas, laszlo sajo-bohus other factors having influenced the temporal variation of radon concentration in soil pores closeness to volcanic formations as can be seen in fig. 3, the average radon concentration in the 20 holes has high values in the first two measurement periods. the greatest contribution to these high values came from the hole located in carabayllo district (s15). in those two periods, two of the highest concentrations measured throughout the study (53.1 kbq/m3 and 21.1 kbq/m3) were found in that hole. the s15 hole is near the chillón river and is also the closest measurement site to the uraniumrich volcanic rocks of quilmaná and huarangal formations. no seismic activity or particular meteorological conditions were observed that could explain the high average values found during those periods. occurrence of earthquakes the high values observed in october and november 2016 (p13 – p16 in fig. 3) could be associated with the earthquakes occurred in lima during those months. in addition to the large number of reports on radon anomalies prior to earthquakes, anomalies have also been registered during their occurrence (petraki et al., 2015). a possible cause of the apparent simultaneity could be related to the time resolution of the radon measurement system. the time resolution of active systems, such as continuous radon monitors, can be as small as 10 minutes. however, passive systems, such as nuclear track detectors, integrate radon concentrations over long time intervals (at least > 1 4 weeks), i.e., they provide coarse time-series estimations. for example, as an illustration, suppose that the empirical equation of sultankhodzhayev et al. (1980), which relates precursor time with earthquake magnitude and epicentral distance, is fulfilled as a first approximation. thereby, for earthquakes of magnitudes 4 and 5, occurring at epicentral distances between 50 and 300 km, the maximum precursor times would be approximately between 1 and 28 days. most of the seismic events perceived in lima city during the study had magnitudes between four and five, and their epicenters were at distances from approximately 50 to 300 km from lima city. since soil gas radon concentrations are integrated in periods of about two weeks, anomalies could be recorded in periods prior to the earthquake or in the period in which the event occurred. as can be seen in fig. 6a, the number of earthquakes perceived in lima city is greater during two consecutive periods in october-november 2016 (seismic bulletins, 2017), and in the period from the end of june to the beginning of july 2017. however, the magnitudes of the earthquakes of the last period were smaller, being much greater the energy released during the october earthquakes (fig. 6b). the two earthquakes perceived most intensely in lima city during the period of measurements (according to the modified mercalli scale) occurred precisely in 2016, on october 19 and 28 (fig. 7). a more detailed study on the association of seismic events with radon in soil gases is planned to be carried out in the near future. rock and debris avalanches due to flooding of rivers during the “el niño costero” phenomenon a significant peak in radon concentration was also observed during the exposure period of detectors from march 5 to 19, 2017 (p24 in fig. 3). this could be associated with the natural phenomenon “el niño costero” which greatly affected several areas in lima city. heavy rains during this period triggered flows of detritus, rock falls and landslides. they also caused overflows and floods of the chillón, rímac and lurín rivers, where large avalanches of rocks, debris and mud were generated. the increased flow in these rivers caused the dragging of rocks and debris. rock motions, debris flows, rock and debris avalanches, rocks impacting on the ground, and rocks impacting or rolling on the channel bed generate ground vibrations, which could cause the increase of radon emanation in holes located near rivers and creeks. our assumption is supported by studies on the influence of vibrations on emanation and exhalation of soil gas radon, and on ground vibrations produced by rock motions and debris flows. moharram (2000) found increased radon concentration in soil pores and enhanced radon exhalation in areas nearby railway and motor vehicle traffic tracks. the author considers the increased delivery of radon from the subsoil due to the enhanced upward diffusion. cai et al. (2018) and cai et al. (2019) studied the influence of low-frequency vibrations on the radon exhalation from uranium samples. the results showed that the radon exhalation increased with the rising vibration frequency until frequencies of about 50 to 70 hz, but fell slowly after reaching the peak. studying ground vibrations produced by rock motions and debris flows, huang et al. (2007) found that ground vibrations caused by individual rocks were 10-150 hz and that larger stones generate ground vibrations with lower peak frequencies. the frequency of the debris flows is lower than 50 hz, but after the front passed, the frequency is between 50 and 100 hz. in these frequency ranges there is an increase in radon emanation and exhalation according to cai et al. (2018) and cai et al. (2019). these authors confirmed that one of the main sources of ground vibration caused by debris flows is the interaction of rocks or boulders with the channel bed. the following facts also support our assumption about the cause of the increased radon concentration in soil gas during the “el niño costero” event: 1all the holes had the same depth; 2detectors were placed at the same distance from the bottom of the hole; 3in no hole there was water infiltration or water vapor condensation on detector surface that could affect the measurements; 4in each measurement period all stations experienced approximately the same meteorological conditions since the measurements were carried out simultaneously; 5overflowing of natural channels of rivers during floods did not affect any measurement site; 6the lowermost pvc tube was never submerged because of rising groundwater levels; the groundwater level in the measurement sites is very deep (inrena, 2004) and its possible fluctuations should have affected little the soil radon emanation; 7if below the bottom of the hole the soil water content would have been very high, contrary to what happened, the radon concentration should have decreased 6 5 4 3 2 1 0 n um be r o f e ar th qu ak es e ne rg y re le as ed (t on s of t n t ) p 1 p 2 p 3 p 4 p 5 p 6 p 7 p 8 p 9 p 10 p 11 p 12 p 13 p 14 p 15 p 16 p 17 p 18 p 19 p 20 p 21 p 22 p 23 p 24 p 25 p 26 p 27 p 28 p 29 p 30 p 31 p 32 p 33 p 34 p 35 p 36 period of observation period of observation p 1 p 2 p 3 p 4 p 5 p 6 p 7 p 8 p 9 p 10 p 11 p 12 p 13 p 14 p 15 p 16 p 17 p 18 p 19 p 20 p 21 p 22 p 23 p 24 p 25 p 26 p 27 p 28 p 29 p 30 p 31 p 32 p 33 p 34 p 35 p 36 30.000 25.000 20.000 15.000 10.000 5.000 0 figure 6. number of earthquakes felt in the lima city (left) and energy released by earthquakes (right) during the period of measurements. 179measurement of radon in soils of lima city peru during the period 2016-2017 considerably; 8during the period p24 there was only one earthquake on 03/13/2017 at local time 10:46:00, latitude -12.63 and longitude -76.97, and 53 km depth (seismic bulletins, 2017). although its magnitude was 4.5, it was perceived in lima city with the lowest intensity an earthquake could be reported (ii level) according to the modified mercalli scale. combined spatial and temporal behavior of soil gas radon concentrations radon concentration in different holes during the period of maximum flow of rives due to phenomenon “el niño costero” to help interpret the results, in particular regarding the effect of floods on the increase of radon emanation from the base of the tube holes, the average radon concentration in each of the 20 holes is shown in fig. 7 for three cases: 1. during the entire period of measurements (epm), 2. during the entire period of measurements except periods of the “el niño costero” event (epm-pec), and 3. during the period of maximum flood of chillón, rímac and surco rivers (mfr). as described, lima city is situated on alluvial fan deposits of the chillón and rímac rivers. alluvial soils of the study area are mainly differentiated by deposited parent material, which is associated with the fluvial erosion of geological formations that contain different uranium content. consequently, figure 7 should provide information on the spatial distribution of deposited parent material, given by the spatial radon distribution, and on the temporal variation of radon in the most important considered periods. radon concentrations were generally higher than the total average concentration in most holes near rivers (s1, s10, s11, s12, s15, s16) regardless of the period of time in which they were averaged. in contrast, radon concentrations were below the total average concentration in most holes located far from rivers and they were approximately the same for the different averaging periods. table 5 and the scatter plot in fig. 8 show a significant negative correlation between soil radon concentration and distance from the hole to the nearest river bank for the different averaging periods, however, the correlation and gradient were higher for the period of maximum flow of rivers. for analysis, the 2-tailed test of significance was used. as can be seen in figure 7, radon concentrations averaged during epm and during epm-pec were very similar in most of the holes. there figure 7. isoseismal maps of the two most intense earthquakes perceived in lima city during soil gas radon measurements (hernando et al., 2016). 30 25 20 15 10 5 0 r n (k b q/ m 3 ) observation well s1 s2 s3 s4 s5 s6 s7 s8 s9 s1 0 s1 1 s1 2 s1 3 s1 4 s1 5 s1 6 s1 7 s1 8 s1 9 s2 0 1 2 3 figure 8. average radon concentration in each hole during the entire period of measurements. (1). during the entire period of measurements except periods of “el niño costero” event. (2). and during the period of maximum flow of chillón and rímac rivers (3) (bars with red inclined lines correspond to holes where very large increases in radon concentration occurred). horizontal and dotted lines represent the same as in figure 3. table 5. pearson and spearman correlation coefficients between soil gas radon concentration and distance from the measurement site (hole) to the nearest river bank for the different averaging periods. pearson corr. p-value spearman corr. p-value mfr -0,67* 0,0049 -0,76* 0.0007 epm -0.60* 0.010 -0.61 0.0089 epm-pec -0.55* 0.0213 -0.61 0.0089 *correlation is significant at the 0.05 level 180 lázaro luís vilcapoma, maría elena lópez herrera, patrizia pereyra, daniel francisco palacios, bertin pérez, jhonny rojas, laszlo sajo-bohus was a large increase in radon concentration in the mfr period in most holes where radon concentrations averaged during the epm were similar or greater than the total average concentration; those holes were located relatively close to rivers. hence, the largest source of temporal variation of radon in holes located relatively close to chillón and rímac rivers was associated with the mfr period, where large floods occurred with avalanches of rocks, debris and mud. the hole located near the surco river in chorrillos district (s1), the holes located along the channel of the rímac river in san juan de lurigancho (s10), chaclacayo (s11) and ate (s12) districts, and the holes situated near the chillón river in puente piedra (s16) and san martin de porres (s17) districts, showed large increase in radon concentration during the period of maximum flow (higher flood) of these rivers. although the s17 hole was relatively far from rivers, it is located in an area historically affected by the overflow and flooding of the chillón river. rocks and debris transported during the period of maximum flow of the chillón river reached an area close to this measurement site. the hole located in independencia district (s14) showed relatively high radon concentration during the epm, but the increase in radon concentration was not as great as in other holes during the mfr period. this hole was located far from rivers, however, the nearby steep and short distance creeks were activated during the period of higher intensity of rainfall, producing debris flows accumulated on their slopes and beds. possibly, the intensity and frequency of resulting vibrations were different from those produced by the large avalanches of rocks and debris occurred during the floods of chillón and rímac rivers; this could be the cause of the lower increase in radon concentration in the hole. radon concentration in each hole averaged over all measurement periods results of radon concentration averaged over the entire study period could be related to the parent material of alluvial soils in the monitoring sites. fig. 7 shows that the highest average radon concentration was registered in ate hole. this may be related to the origin of alluvial deposits in the area around the hole, which are most likely associated with the material transported during the floods of the huaycoloro river in the pleistocene era (about 2.5 million years ago). this river has eroded in its course volcanic rocks rich in 238u (ingemmet, 1999; sebrier & macharé, 1980). the high average values of radon emanation found in carabayllo (s15) and san martín de porres (s17) holes, are due to the presence of alluvial deposits of pleistocene and recent (less than 0.13 million years ago), corresponding to the floods of chillón river that eroded the chain of volcanic hills located on its slopes. soils around these holes are shallow and they have developed from hard, coherent and consolidated rock. the lithological composition of these soils is variable such as quartzites, sandstones, limestones, shales and volcanic rock; some of them commonly present high 226ra content. differently, the floods of rímac river have contributed less in the deposition of materials with relatively high content of radon parents, since the predominant geological formation along its channel and slopes has lower uranium content compared to the casma group (where the chillón and huaycoloro rivers flow). the inferred permeability of the medium confirms the direct association between the measured radon levels and the origin of the alluvial deposits. most of the area of lima city is covered by an alluvial gravel deposit which is composed of a medium to dense coarse gravel with a sandy matrix. these materials are mostly from the alluvial fan of the rímac river. in such permeable deposits radon moves readily and could reach the measurement site from relatively large distances. in these circumstances, high radon levels would be generated at the measurement site if the emanation power of the medium is high. since the measured radon concentrations were relatively low, the low emanation power of materials deposited in the alluvial fan of the rímac river is confirmed. however, in the northern districts of puente piedra, comas and carabayllo, and in the eastern districts of la molina and ate vitarte, there 40 30 20 10 0 -10 -20 40 30 20 10 0 -10 -20 r n (k b k/ m 3 ) -8 -4 -2 0 2 4 6 8 10 12 14-8 -4 -2 0 2 4 6 8 10 12 14 distance to the river (km)distance to the river (km) r n (k b k/ m 3 ) 40 30 20 10 0 -10 -20 r n (k b k/ m 3 ) -8 -4 -2 0 2 4 6 8 10 12 14 distance to the river (km) a b c figure 9. scatter plot of soil gas radon concentration averaged during the epm (a), epm-pec (b), and mfr (c) periods vs. distance from the hole to the nearest riverbank. the confidence level for ellipse was 95%. 181measurement of radon in soils of lima city peru during the period 2016-2017 are silty sand deposits. in these less permeable deposits, the measured radon comes from shorter distances, thus providing more local information about the emanation power of the medium. since the highest levels of radon were found in these sites, the high emanation power of materials deposited by the chillón and huaycoloro rivers is confirmed. semkov (1990) and sutherland (1994) have also suggested correlation between specific surface area and emanating powers of radon. thus, particle-size fractions due to fracturing and crushing have possibly played a major role in increasing the radon emanation at the measurement sites of districts indicated above. a significant contribution to the radon concentration was found when volcanic materials were present at the measurement sites. these materials have been transported and deposited by the chillón and rímac rivers and come from the disaggregation by meteorization and erosion of the casma group’s volcanic rocks. however, measurements made directly on slope of hills of volcanic (s2) or plutonic origin (s8, s9, s13), yielded the lowest values of radon concentration. in these places there is no means of transport and thus erosion is slower compared to sites located in the riverbeds. according to ennemoser et al. (1993), the emanation rate of radon from soil depends not only on the concentration of uranium and radium content but to some extent on emanating power and the diffusion coefficient of radon in soil. the alluvial deposits, compared to the rocky areas on the hill slopes, has an increased emanating power and diffusion coefficient due to intense fracturing, and crushing down to the grain and sub-grain size, resulting in an increased surface area and porosity (purtschller et al., 1995). radon measurements in the studied area have shown that the highest levels of radon are in accordance with the type of lithological material present in riverbeds and with its disintegration, erosion, transport and distribution by the action and flooding of rivers. the grain-size reduction due to fracturing and further weathering of transported parent materials expose the uraniumrich mineral thus giving rise to high values of radon concentration in the zones of higher accumulation. floods of rivers associated with “el niño costero” have been repeated over thousands of years and have great responsibility in the geomorphology of the current lima. thus, the results of this study provide information not only on the effects of the event on soil gas radon during the period of measurements, but also help in understanding the causes of differentiated geogenic radon potential in the study area. conclusions 1. total average soil gas radon concentration (spatial and temporal) was 5.6 kbq/m3, while the minimum and maximum values for the whole set of measurements were 0.1 kbq/m3 and 64.3 kbq/m3, respectively. these results are included within values reported in similar studies. 2. total average value of radon concentration in alluvial soil of lima city is in correspondence with the relatively low radon values reported in dwellings. however, more studies are required to obtain statistical evaluation of spatially geogenic radon potential with indoor radon. 3. the average radon concentration in soil gas is about two times lower in winter than in the other seasons. although there is a moderate relation between soil radon concentration and meteorological variables, it was shown that other agents controlled the radon concentration variation. 4. high values of soil gas radon concentration during october/november 2017 apparently were related to the earthquakes perceived in lima city in that period. 5. the highest radon concentrations were found in areas of alluvial deposits whose parental material has been removed from the volcanic quilmaná and huarangal by the chillón and huaycoloro rivers flowing through them. soil gas radon concentrations were even higher in areas closer to volcanic and less distant from rivers. 6. in general, the lowest average concentrations of radon in soil gas were found in areas far from rivers, on hills of volcanic origin, and in geological formations where intrusive rocks predominate or in alluvial fans of rivers flowing through them (rímac and lurín rivers). 7. anomalous high soil gas radon concentrations were found during the period of maximum flooding of rivers due to “el niño costero” phenomenon. they were associated with ground vibrations caused by rock and debris avalanches in rivers and creeks. 8. the study of radon emanation in pores of alluvial soils allowed to identify the sources of deposited materials. thus, the application of this methodology could in principle help in the characterization of geodynamic events in geological and geomorphological contexts similar to that of our study area. 9. the identification and distribution of the parent material in alluvial deposits of lima city, and the influence on them of the historical occurrence of “el niño costero”, help to interpret the geogenic radon hazard variation defined by radon indoors studies. acknowledgements this study was carried out as part of the “measurement of radon-222 in soils, using nuclear track detectors” project (cap 339-2016-pucp). references åkerblom, g., andersson, p. & clevensjö, b. (1984). soil gas radon a source for indoor radon daughters. radiation protection dosimetry, 7(1-4), 49-54. alonso, h., rubiano, j.g., guerra, j. g., arnedo, m. a., tejera, a. & martel, p. (2019). assessment of radon risk areas in the eastern canary islands using soil radon gas concentration and gas permeability of soils. science of the total environment, 664, 449-460. al-tamimi, m. h. & abumurad, k. m. (2001). radon anomalies along faults in north of jordan. radiation measurements, 34, 397-400. appleton, j. d. & miles, j. c. (2010). a statistical evaluation of the geogenic controls on indoor radon concentrations and radon risk. journal of environmental radioactivity, 101(10), 799-803. bonotto, d. m. & andrews, j. n. (1999). transfer of radon and parent nuclides 238u and 234u from soils of the mendip hills area, england, to the water phase. journal of geochemical exploration, 66, 255-268. cai, z., li, b., hong, c., li, x., yuan, j. & wang, h. (2018). laboratory experimental laws for the radon exhalation of similar uranium samples with low-frequency vibrations. sustainability, 10(8), 2937. cai, z., li, x., bo, l., hong, c., ming, l. & qiong, w. (2019). experimental study on influence of low-frequency vibration on radon exhalation from thermal porous medium. in: chang x. (editor) proceedings of the 11th international mine ventilation congress. springer, singapore, 1036-1047. chauhan, r. p., kumar, a. & chauhan, n. (2017). distribution of indoor thoron in dwellings under normal and turbulent flow conditions using cfd simulation technique. nuclear technology & radiation protection, 32(2), 180-184. chira, j., rios, c., trelles, g. & villareal, e. (2018). estimación del potencial minero metálico del perú y su contribución económica al estado, acumulado al 2050. ingemmet, 90 pp. choubey, v. m., ahmad, i., kamra, l. & ramola, r. c. (2010). radon variations in soil and groundwater of bhilagana valley, garhwal himalaya, india. japanese journal of health physics, 45(3), 278-283. choubey, v. m., bartarya, s. & ramola, r. (2005). radon variations in an active landslide zone along pindar river, in chamoli district, garhwal lesser himalaya, india. environmental geology, 47, 745-750. choubey, v. m., bist, k. s., saini, n. k. & ramola, r. c. (1999). relation between soil-gas radon variation and the different lithotectonic units of garhwal himalaya, india. journal of applied radiation and isotopes, 51(5), 587-592. cinelli, g., tositti, l., capaccioni, b., brattich, e. & mostacci, d. (2015). soil gas radon assessment and development of a radon risk map in bolsena, central italy. environmental geochemistry and health, 37, 305-319. ciotoli, g., guerra, m., lombardi, s. & vittori, e. (1998). soil gas survey for tracing seismogenic faults: a case study in the fucino basin, central italy. journal of geophysical research solid earth, 103(b10), 23781-23794. cothern, c. r. & smith, j. e. (1987). environmental radon. environmental science research, vol. 35, plenum press, new york, 363 pp. duggal, v., rani, a. & mehra, r. (2014). measurement of soil-gas radon in some areas of northern rajasthan, india. journal of earth system science, 123(6), 1241-1247. 182 lázaro luís vilcapoma, maría elena lópez herrera, patrizia pereyra, daniel francisco palacios, bertin pérez, jhonny rojas, laszlo sajo-bohus elío, j., ortega, m. f., nisi, b., mazadiego, l. f., vaselli, o., caballero, j., quindós-poncela, l. s., sainz-fernández, c. & pous, j. (2015). evaluation of the applicability of four different radon measurement techniques for monitoring co2 storage sites. international journal of greenhouse gas control, 41, 1-10. ennemoser, o., ambach, w., brunner, p., schneider, p., oberaigner, w., purtschller, r. & stingl, v. (1993). unusually high indoor radon concentrations. atmospheric environment, 27a(14), 2169-2172. fleischer, r. l, price, p. b. & walker, r. m. (1975). nuclear tracks in solids: principles and applications. univ. of california press, berkeley, 605 pp. fleischer, r. l., giard, w. r. & turner l. g. (2000). membrane-based thermal effects in 222rn dosimetry. radiation measurements, 32, 325-328. giammanco, s., imme, g., mangano, g., morelli, d. & neri, m. (2009). comparison between different methodologies for detecting radon in soil along an active fault: the case of the pernicana fault system, mt. etna (italy). applied radiation and isotopes, 67, 178-185. giammanco, s., sims, k. w. w. & neri. m. (2007). measurements of 220rn and 222rn and co2 emissions in soil and fumarole gases on mt. etna volcano (italy): implications for gas transport and shallow ground fracture. geochemistry, geophysics, geosystems, 8(10), 1644. godoya, m., hadler, j. c., iunes, p. j., mestanza, s. n., oliveira, r. a., osorio, a. m. & paulo, s. r. (2002). effects of environmental conditions on the radon daughters spatial distribution. radiation measurements, 35, 213-221. gundersen, l. c. s., schumann, r. r., otton, j. k., dubiel, r. f., owen, d. e. & dickinson, k. a. (1992). geology of radon in the united states. in: gates, a.e. & gundersen, l.c.s., (editors). geologic controls on radon. geological society of america, special paper 271, 1-16. hadler, j. c. & paulo, s. r. (1994). indoor radon daughters contamination monitoring: the absolute efficiency of cr-39 taking into account the plate-out effect and environmental conditions. radiation protection dosimetry, 51 283-296. hakl, j., hunyadi, i., csige, i., geczy, g., lenart, l. & torocsikll, i. (1992). outline of natural radon occurrences on karstic terrains of hungary. radiation protection dosimetry, 45(1-4), 183-186. hernando, tavera, h., agüero, c. & fernández, e. (2016). catálogo general de isosistas para sismos peruanos. dirección de ciencias de la tierra sólida/unidad de sismología, lima, perú. hogg, a. j. (2018). predicting huaycos and lahars: physical models, mathematics and uncertainty, international workshop, strengthening capacity for mitigation of huaico (flash flood), impacts in peru lima and arequipa, peru, october, 25-27. huang, c. j., yin, h. y., chen, c. y., yeh, c. h. & wang, c. l. (2007). ground vibrations produced by rock motions and debris flows. journal of geophysical research earth surface, 112, f02014(1)f02014(20). huxol, s., brennwald, m. s., hoehn, e. & kipfer, r. (2012). on the fate of 220rn in soil material in dependence of water content: implications from field and laboratory experiments. chemical geology, 298-299, 116-122. iaea (2014). the environmental behaviour of radium. revised edition. technical reports series no 476, vienna, austria, 267 pp. ingemmet (1999). carta geológica nacional 501 cuadrángulos geológicos digitales de la carta nacional 1960 1999, república del perú, sector energía y minas, instituto geológico minero metalúrgico, lima, perú. inrena (2004). inventario de fuentes de agua subterránea en el valle del río chillón. república del perú, ministerio de agricultura, instituto nacional de recursos naturales, intendencia de recursos hídricos, administración técnica del distrito de riego chillón-rímac-lurín. lima, perú. kareem, h. a. (2016). measurement of radon concentration and the effective dose rate in the soil of the city of karbala, iraq. journal of radiation and nuclear applications, 1(1), 17-23. king, c. y. & minissale, a. (1994). seasonal variability of soil-gas radon concentration in central california. radiation measurements, 23(4), 683-692. kumar, g., kumari, p., kumar, a., prasher, s. & kumar, m. (2017). a study of radon and thoron concentration in the soil along the active fault of nw himalayas in india. annals of geophysics, 60(3), s03291s032912. kyser, k., alexandre, p., polito, p., djouka-fonkwe, m., lahd geagea, m. and uvarova, y. (2009). exploration strategies for uranium deposits, 24th international applied geochemistry symposium, at fredericton, new brunswick, canada, 439-443. lópez-coto, i., mas, j. l., san miguel, e. g., bolivar, j. p. & sengupta, d. (2009). a comparison between active and passive techniques for measurements of radon emanation factors. applied radiation and isotopes, 67, 849-853. matolin, m. & stráník, z. (2006). radioactivity of sedimentary rocks over the ždánice hydrocarbon field. geophysical journal international, 167(3), 1491-1500. mehta, v., singh, s. p., chauhan, r. p. & mudahar, g. s. (2014). measurement of indoor radon, thoron and their progeny levels in dwellings of ambala district, haryana, northern india using solid state nuclear track detectors. romanian journal of physics, 59(7-8), 834-845. moharram, b. m. (2000). the influence of mechanical vibrations of railway and car traffics on the radon exhalation using track detector technique. in: aly, h.f. (editor.). seventh conference of nuclear sciences and applications vol. 2, egyptian society of nuclear sciences and applications, cairo, egypt, 968-980. nazaroff, w. w. (1992). radon transport from soil to air. reviews of geophysics, 30, 137-160. nazaroff, w. w., moed. b. a. & sextro, r. g. (1988). soil as a source of indoor radon: generation, migration, and entry. in: nazaroff w.w. & nero j.a.v. (editors). radon and its decay products in indoor air. usa, john wiley & sons, inc., new york, 57-112. neri, m., behncke, b., burton, m., galli, g., giammanco, s., pecora, e., privitera, e. & reitano, d. (2006). continuous soil radon monitoring during the july 2006 etna eruption. geophysical research letters, 33(24), l24316. nikezic, d. & stevanovic, n. (2007). behavior of 220rn progeny in diffusion chamber. nuclear instruments and methods in physics research a, 570, 182-186. palacios, d., fusella, e., avila, y., salas, j., teixeira, d., fernández, g., sajóbohus, l., greaves, e., barros, h., bolívar, m. & regalado, j. (2016). soil gas radon and thoron measurements in some venezuelan oilfields. journal of radioanalytical and nuclear chemistry, 307(1), 801-810. palacios, o., caldas, j. & vela, ch. (1992). geología de los cuadrángulos de lima, lurín chancay y chosica. ingemmet-boletín no. 43 serie a: carta geológica nacional. papp, b., deak, f., horvath, a., kiss, a., rajnai, g. & szabo, cs. (2008). a new method for the determination of geophysical parameters by radon concentration measurements in bore-hole. journal of environmental radioactivity, 99, 1731-1735. paulo, s. r., neman, r., iunes, p. j. & hadler, j. c. (2001). simulating radon daughters diffusion through the air and their depletion on material surfaces. radiation measurements, 34, 517-519. perales, c. f. (1994). glosario y tabla de correlación de las unidades estratigráficas del perú. lima, perú, 177 pp. pereyra, p., lópez, m. e. & vilcapoma, l. (2015). concentration measurements of rn indoors in lima-peru. international journal of physics, 3(4), 165-169. pereyra, p., lópez, m. e., palacios, d., canchos, v., guevara, c., liza, r. & sajo bohus, l. (2018). mediciones de radón 222 en la ciudad de lima y callao utilizando detectores de huellas nucleares, xi congreso regional de seguridad radiológica y nuclear, congreso regional irpa, la habana, cuba, abril, 1259-1268. pérez, n. m., hernández, p. a. padron, e., melian, g., marrero, r., padilla, g., barracks, j. & mohasco, d. (2007). precursory subsurface 222rn and 220rn degassing signatures of the 2004 seismic crisis at tenerife, canary islands. pure and applied geophysics, 164, 2431-2448. 183measurement of radon in soils of lima city peru during the period 2016-2017 petraki, e., nikolopoulos, d., panagiotaras, d., cantzos, d., yannakopoulos, p., nomicos, c. & stonham, j. (2015). radon-222: a potential shortterm earthquake precursor. journal of earth science and climatic change, 6, 282(1)-282(11). purtschller. f., pirchl, p., sieder, g., stingl, v., tessadri, t., brunner, p., ennemoser, o. & schneider, p. (1995). radon emanation from giant landslides of koefels (tyrol, austria) and langtang himal (nepal). environmental geology, 26, 32-38. rivera, r. (2010). realidad y perspectivas de los minerales radioactivos en el perú. ingemmet. http://www.ingemmet.gob.pe/ documents/73138/202784/p11_expo_uranio.pdf rojas, j., palacios, d., pereyra, p., pérez, b., sajo-bohus, l. & lópez, m.e. (2018). a semi-empirical approach to estimate the parameters determining the lr-115 detector response in radon measurements. radiation measurements, 118, 36-42. ruckerbauer, f. & winkler, r. (2001). radon concentration in soil gas: a comparison of methods. applied radiation and isotopes, 55, 273-280. salama, t. a., seddik1, u., hegazy, t. m. & morsy a. a. (2006). direct determination of bulk etching rate for lr-115-ii solid state nuclear track detectors. pramana journal of physics, 67(3), 529-534. sebrier, m. & macharé, j. (1980). observaciones acerca del cuaternario de la costa del perú central. bulletin de l’institut français d’etudes andines, 9(1-2), 5-22. seismic bulletins (2017). national center for seismic monitoring, geophysical institute of peru (gip), boletín n° 001-boletín n° 354. semkov, t. m. (1990). recoil-emanation theory applied to radon release from mineral grain. geochimica et cosmochimica acta, 54, 425-440. shweikani, r. & hushari, m. (2005). the correlations between radon in soil gas and its exhalation and concentration in air in the southern part of syria. radiation measurements, 40, 699-703. snihs, j. o. (1992). swedish radon programme. radiation protection dosimetry, 42(3), 177-184. sprinkel, d. a. & solomon, b. j. (1990). radon hazards in utah: utah geological and mineral survey, circular 81, utah, united states, 24 pp. sultankhodzhayev, a. n., latipov, s. u., zakirov, t. z. & zigan, f. g. (1980). dependence of hydrogeoseismal anomalies on the energy and epicentral distance of earthquakes. doklady akademii nauk uzbekskoj ssr, 5, 57-59. sundal, a. v., valen, v., soldal, o. & strand, t. (2008). the influence of meteorological parameters on soil radon levels in permeable glacial sediments. science of the total environment, 389(2-3), 418-428. sutherland, d. s. (1994). radon workshop-geology, environment, techniques. geoscientist, 4(2), 27-29. szabó, k. z., jordán, g., horváth, á. & szabó, c. (2013). dynamics of soil gas radon concentration in a highly permeable soil based on a long-term high temporal resolution observation series. journal of environmental radioactivity, 124, 74-83. tommasino, l. (2016). concealed errors in radon measurements and strategies to eliminate them, 8th conference on protection against radon at home and at work, prague, czech republic, september, 25. urosevic, v., nikezic, d. & vulovic, s. (2008). a theoretical approach to indoor radon and thoron distribution. journal of environmental radioactivity, 99, 1829-1833. villacorta, s., núñez, s., tatard, l., pari, w. & fidel, l. (2015). peligros geológicos en el área de lima metropolitana y la región callao, ingemmet. boletín serie c: geodinámica e ingeniería geológica, lima, perú, 59, 156 pp. walia, v., mahajan, s., kumar, a., singh, s., bajwa, b. s., dhar, s. & yang, t. f. (2008). fault delineation study using soil-gas method in dharamsala area, nw himalayas, india. radiation measurements, 43, s337-s342. wiegand, j. (2001). a guideline for the evaluation of the soil radon potential based on geogenic and anthropogenic parameters. environmental geology, 40, 949-963. zafrir, h., barbosa, s. & malik, u. (2012). differentiation between the effect of temperature and pressure on radon within the subsurface geological media. radiation measurements, 49, 39-56. zmazek, b., zivcic, m., todorovski, l., dzeroski, s., vaupotic, j. & kobal i. (2005). radon in soil gas: how to identify anomalies caused by earthquakes, applied geochemistry, 20, 1106-1119. http://www.ingemmet.gob.pe/documents/73138/202784/p11_expo_uranio.pdf http://www.ingemmet.gob.pe/documents/73138/202784/p11_expo_uranio.pdf palabras clave: galería en muro; análisis numérico; estabilidad subterránea. how to cite item ozdogan, m. v., & gonen, a. (2019). determination of distance required to ensure stope and footwall-drift non-interaction zone based on geological strength index. earth sciences research journal, 23(1), 17-25. doi: https://doi.org/10.15446/esrj.v23n1.71879 in the bakibaba copper mine, the longhole stoping method is used in the production of copper ore. stability problems have occurred at times on the footwall drift due to the interaction between the footwall drift and stope. in this study, we propose a method for estimating the minimum distance necessary to ensure a non-interaction zone between the footwall drift and stope. we used the finite element method and various distances between the footwall drift and stope and the displacements over drifts as parameters. we also performed analyses on various geological strength index values from low to high to determine the effect of the rock mass on the interaction between the footwall drift and stope. abstract keywords: footwall drift, numerical analysis, underground stability. determination of distance required to ensure stope and footwall-drift non-interaction zone based on geological strength index determinación de la distancia requerida para asegurar el rebaje y las galerías en muros de la zona de no interacción en minería subterránea con base en el índice de resistencia geológica issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v23n1.71879 en la mina de cobre de bakibaba se utiliza el método de minería por perforación de barrenos largos para extraer el mineral. los problemas de estabilidad han ocurrido generalmente en las galerías en muro debido a la interacción entre la galería y el rebaje. en este estudio se propone un método para la estimación de la distancia mínima necesaria para asegurar la zona de no interacción entre la galería en muro y el rebaje. se utilizó como parámetro el método del elemento finito y varias distancias entre la galería en muro y el rebaje y los desplazamientos sobre las galerías. también se realizaron análisis con varios índices de resistencia geológica de abajo hacia arriba para determinar el efecto de la masa rocoa en la interacción entre la galería en muro y el rebaje. resumen record manuscript received: 24/04/2018 accepted for publication: 24/08/2018 earth sciences research journal earth sci. res. j. vol. 23, no. 1 (march, 2019): 17-25 mehmet volkan ozdogan, alper gonen dokuz eylul university engineering faculty department of mining engineering. * corresponding author: mehmet.ozdogan@deu.edu.tr g eo lo g y 18 mehmet volkan ozdogan, alper gonen introduction the stability of the openings in underground mines is a major factor affecting mining safety and sustainability. as such, safety and economic restrictions are the main design parameters considered in underground excavations. the geometry of an underground opening (shape and size), toe between openings and length of openings are functions of safety and cost. decisions regarding these mining parameters are generally empirically based and at times unexpected failures occur. the distance between the footwall drift and the stope is an important factor that affects the stability of the footwall drift. the strength of the rockmass and the mining depth significantly influence the distance required between the footwall drift and stope. in deep mines, the rockmass is highly stressed and stability problems occur around underground openings. although several studies have addressed the interaction between underground tunnels (ghaboussi & ranken, 1977, jamshid & randall, 1977; yamaguchi et al. 1998; addenbrooke & potts, 2001; gercek, 2005; karakus et al. 2007; wang et al., 2007; wang & li, 2008; chakeri et al. 2011; wang et al., 2017), few have addressed the influence of the stope on the footwall drift in underground mines. abdellaha et al. (2011) examined the interaction between the haulage drift and mine stope in a sublevel-stoping mine. the authors considered six excavation sequences in one stope and evaluated the support required for the haulage drift after each excavation. purwanto et al. (2013) studied the influence of stope design on the stability of the hanging wall decline in cut-and-fill mines. the authors numerically analysed the stability of the hanging wall decline for various stope geometries. kabwe & bowa (2016) investigated the effect of the mining method on the stability of the footwall drift for the same orebody by numerically modelling the stress distribution over the stope and footwall drift for three different mining methods. on the other hand, none of the studies mentioned above have effort on determination of non-influenced zone distance between stope and footwall drift in longhole stoping mines. the main scope of this paper is to estimate the distance between the footwall drift and stope that is necessary to ensure the existence of a noninteraction zone in longhole stoping mines. we also consider the effect of geotechnical conditions. in our study, we examined the footwall drift of the bakibaba underground copper mine, which has experienced stability problems, and we used numerical analyses to determine the displacements over the footwall drift for various geological strength index (gsi) values.. mining method the bakibaba underground copper mine is located in kure province in north-western turkey, near the city of kastamonu and the black sea. in this mine, a longhole-stoping and post-backfill method is used and the mine has an annual production capacity of 1,000,000 tonnes. the orebody lies in a north-easterly direction, is 100–300 m in length and averages 40 m in width. the ore deposit is 500 m in depth and the dip of the ore is 70°. the orebody is accessed via a 15% declining ramp. a footwall drift running parallel to the orebody is 5-m wide by 5-m high. crosscuts are driven from the footwall across the orebody to the vein. the footwall drifts are positioned 40–50 m from where the footwall makes contact with the orebody. from the boundary of the ore deposit, 7-m wide by 5-m high sill drifts are driven until contact is made with the hanging wall. the upper and lower sill drifts are connected at 15-m intervals by a slot raise, and then widen to a 7-m drift width. parallel blast holes 76 mm in diameter are drilled downward between two sill drifts. after blasting, a remote-controlled load-haul-dump (lhd) vehicle is used to muck the ore from the lower sill drift and transport it to the ore pass. after the entire stope is mined, the open stope is backfilled from the upper sill. in the bakibaba underground mine, primary stopes are paste-filled or cementedrock-filled and secondary stopes are rock-filled. figure 1 shows schematic illustrations of the stope production cycle. numerical model in the mining industry, the numerical finite element method (fem) is widely used to solve geotechnical problems. fem can accommodate material heterogeneity, non-linear deformability, complex boundary conditions, in-situ stresses and gravity (jing & hudson, 2002; sharma, 2009). the utilisation figure 1. stope production cycle (a: stope development, b: production drilling, c: ore mucking, d: backfilling 19determination of distance required to ensure stope and footwall-drift non-interaction zone based on geological strength index of fem-based software is one of the most preferred methods, particularly in underground design analyses (ozdogan et al. 2017). in this study, phase2 twodimensional (2d) fem software program developed by rocscience (2007) was used to determine the influence of the distance between the stope and the footwall drift. figure 2 shows the model used in our analysis, in which we used 10-m-interval distances between the footwall drift and stope up to the stabilisation of the displacement over the drift. laboratory tests in accordance with isrm standards were performed to determine the geomechanical properties of the rock unit, including the uniaxial compressive strength, unit weight, young’s modulus, poisson’s ratio, cohesion and internal friction angle of the rock unit (table 1). to determine the rock-mass input parameters for the fem analyses, we performed both laboratory and field studies and used hoek and brown criteria. hoek and brown criteria developed by hoek et al. 2002, is an empirical method to determine the strength of rock masses in terms of major and minor principal stresses and widely used criterion in geotechnical projects. hoek and brown criteria estimates the strength envelopes determined from laboratory triaxial tests of intact rock by using, uniaxial compressive strength of intact rock (sigci), the intact rock parameter (mi), the geological strength index (gsi) and the disturbance factor (d). the hoek and brown criteria is expressed as;      1 3 3' ' ' = + +      ci ci a bm s (1) where, σci, is the uniaxial compressive strength and mb is the reduced value of material constant mi which is calculated as; m m exp gsi db i = − −     100 28 14 (2) s and a are constants for rock mass given by; s exp gsi d = − −     100 9 3 (3) a e e gsi = + −         − − 1 2 1 6 15 20 3 (4) in order to determine rock mass parameters from laboratory test results rocdata software was used. table 2 shows the strength parameters of the rock mass derived from our laboratory and field studies. also using a chart derived by sheorey (1994), we calculated a total stress ratio of 1.25 for the bakibaba copper mine, based on the elasticity modulus and depth. results and discussion fem analyses were performed to investigate the effect of the distance between the extracted stope and the footwall drift on drift stability, and examined various distances between them. figure 2. base model used in analyses table 1. test results for rock units parameter, symbol, unit basalt (hanging wall & footwall) ore zone unit weight, δ, kn/m3 28.04 28.80 uniaxial compressive strength, σc, mpa 50.48 60.28 brazilian tensile strength, σt, mpa 4.88 5.25 young’s modulus, ei, gpa 12520 14280 poisson’s ratio, ν 0.25 0.24 cohesion, c, mpa 9.82 10.16 internal friction angle, ф, ° 44.5 43.25 table 2. input parameters used in fem analysis parameters, symbol, unit basalt (hanging wall & footwall) ore zone geological strength index (gsi) 55 60 rock type, mi 20 20 disturbance factor, d 0 0 cohesion of rock mass, c, mpa 3.29 3.95 internal friction angle of rock mass, ф, ° 38 48 tensile strength of rock mass, σt, mpa 0.08 0.10 rock mass deformation modulus, ei, mpa 9429 10329 20 mehmet volkan ozdogan, alper gonen two evaluation criteria, including the yield zone, i.e. failure area, and the displacement over the footwall drift were considered. the analyses were stopped when the displacements over the drift had not changed in two consecutive analyses. the red zones in figure 3 indicate the 100% yield zone around underground spaces due to mining extraction when the distance from the stope to the footwall drift is 10 m. in figure 4, the numbers on left-side wall (lw), right-side wall (rw) and roof (rf) of the drift indicate the displacement that occurred over the drift for a 10-m distance between the stope and footwall drift. similarly, figures 5–7 show the changes in the yield zone and displacements for greater distances between the stope and footwall drift. as expected, as the distance between the stope and footwall drift increases, the yield zone and displacements over drift decrease. when the distance is 10 m or less, the yield zone of the stope and footwall drift overlap and unstable drifts occur. the model results show that when the distance between the stope and footwall drift is 10 m, the displacement on the roof of the upper drift is 0.021 m and that of the bottom drift is 0.048 m. the displacement on the roof decreases to 0.015 m for the upper drift and 0.033 m for the bottom drift for a distance of 20 m. table 3 shows the change in the displacements over the footwall drift, in which we can see that when the distance between stope and the footwall drift is 70 m, the footwall drift is totally removed from the influence zone. figure 8 shows the relation between distance between the stope and footwall drift and the displacements over the footwall drift. our numerical analyses indicate that, for the given geotechnical circumstances, the displacements over the drift become stable when there is a distance of 70 m from the stope. we conducted further numerical analyses for various gsi values to determine the effect of the gsi value on the critical distance between the stope and footwall drift. table 4 shows the displacements around the footwall drifts for various gsi values and various stope distances. as expected, the interaction between the stope and footwall drift decreases with increases in the gsi value. however, a gsi value of 55 is the limit value regarding interaction between the stope and footwall. higher gsi values have no influence on the non-interaction zone distance. at distances of 70 m and more between the stope and footwall drift, we observed no interaction for gsi values higher than 55. variation of non-influenced zone distance based on gsi were shown in figure 9. figure 3. yield zone for 10-m interval between stope and footwall drift figure 4. displacements over footwall drift for 10-m interval between stope and footwall drift 21determination of distance required to ensure stope and footwall-drift non-interaction zone based on geological strength index figure 5. yield zones and displacements for various stope–footwall-drift distances (a: yield zone for 20-m distance, b: displacements over drifts for 20-m distance, c: yield zone for 30-m distance, d: displacements over drifts for 30-m distance, e: yield zone for 40-m distance, f: displacements over drifts for 40-m distance) 22 mehmet volkan ozdogan, alper gonen figure 6. yield zones and displacements for various stope–footwall-drift distances (g: yield zone for 50-m distance, h: displacements over drifts for 50-m distance, i: yield zone for 60-m distance, j: displacements over drifts for 60-m distance, k: yield zone for 70-m distance, l: displacements over drifts for 70-m distance) 23determination of distance required to ensure stope and footwall-drift non-interaction zone based on geological strength index figure 7. yield zones and displacements for various stope–footwall-drift distances (m: yield zone for 80-m distance, n: displacements over drifts for 80-m distance) table 3. change in displacements over the footwall drift distance between stope and footwall drift (m) upper drift (m) bottom drift (m) left wall right wall roof left wall right wall roof 10 0.018 0.030 0.021 0.042 0.042 0.048 20 0.012 0.024 0.015 0.030 0.033 0.033 30 0.009 0.021 0.015 0.021 0.030 0.024 40 0.006 0.021 0.012 0.012 0.024 0.018 50 0.003 0.018 0.012 0.009 0.021 0.015 60 0.003 0.018 0.012 0.006 0.021 0.012 70 0.003 0.018 0.012 0.003 0.018 0.012 80 0.003 0.018 0.012 0.003 0.018 0.012 0 0,01 0,02 0,03 0,04 0,05 0,06 0 20 40 60 80 100 upper drift lw upper drift rw upper drift roof bottom drift lw bottom drift rw bottom drift roof d is pl ac em en t ( m ) distance between stope and footwall drift (m) figure 8. relation between distance between stope and footwall drift and displacement over drift 24 mehmet volkan ozdogan, alper gonen table 4. displacements around footwall drifts for various gsi values and various stope distances gsi distance between stope and footwall drift (m) upper drift (m) bottom drift (m) left wall right wall roof left wall right wall roof 10 10 1.000 1.800 1.600 1.800 2.000 2.000 20 0.800 1.600 1.400 0.800 1.600 1.400 30 0.800 1.400 1.000 0.200 1.400 1.000 40 0.600 1.200 1.000 0.400 1.200 1.000 50 0.400 1.200 0.800 0.400 1.000 0.800 60 0.400 1.000 0.800 0.400 1.000 0.800 70 0.400 1.000 0.800 0.200 1.000 0.800 80 0.200 0.800 0.600 0.200 1.000 0.600 90 0.200 0.800 0.600 0.200 0.800 0.600 100 0.200 0.800 0.600 0.200 0.800 0.600 35 10 0.075 0.150 0.105 0.165 0.180 0.195 20 0.060 0.120 0.090 0.105 0.150 0.135 30 0.045 0.105 0.075 0.075 0.120 0.105 40 0.030 0.090 0.075 0.060 0.105 0.075 50 0.030 0.090 0.060 0.045 0.090 0.060 60 0.015 0.075 0.060 0.030 0.090 0.060 70 0.015 0.075 0.060 0.015 0.090 0.060 80 0.015 0.075 0.045 0.015 0.075 0.045 90 0.015 0.075 0.045 0.015 0.075 0.045 55 10 0.018 0.030 0.021 0.042 0.042 0.048 20 0.012 0.024 0.015 0.030 0.033 0.033 30 0.009 0.021 0.015 0.021 0.030 0.024 40 0.006 0.021 0.012 0.012 0.024 0.018 50 0.003 0.018 0.012 0.009 0.021 0.015 60 0.003 0.018 0.012 0.006 0.021 0.012 70 0.003 0.018 0.012 0.003 0.018 0.012 80 0.003 0.018 0.012 0.003 0.018 0.012 75 10 0.004 0.007 0.005 0.013 0.012 0.014 20 0.002 0.007 0.004 0.009 0.010 0.009 30 0.002 0.006 0.004 0.005 0.008 0.006 40 0.001 0.005 0.003 0.004 0.007 0.005 50 0.001 0.005 0.003 0.002 0.006 0.004 60 0.001 0.005 0.003 0.002 0.006 0.003 70 0.001 0.005 0.003 0.001 0.005 0.003 80 0.001 0.005 0.003 0.001 0.005 0.003 90 10 0.002 0.003 0.002 0.006 0.005 0.006 20 0.001 0.003 0.002 0.004 0.004 0.004 30 0.001 0.002 0.002 0.003 0.003 0.003 40 0.001 0.002 0.002 0.002 0.003 0.003 50 0.001 0.002 0.001 0.001 0.003 0.002 60 0.001 0.002 0.001 0.001 0.002 0.002 70 0.001 0.002 0.001 0.001 0.002 0.001 80 0.001 0.002 0.001 0.001 0.002 0.001 25determination of distance required to ensure stope and footwall-drift non-interaction zone based on geological strength index conclusions in this study, 3d finite element method was used to investigate the interaction between the stopes and footwall drifts of the bakibaba underground copper mine. we determined the range of influence between the stope and footwall drift for the currently existing circumstances. then, the gsi value was systematically increased while keeping all the other parameters constant to determine the effect of the gsi value on this range of influence. our series of numerical simulations revealed that a distance of 70 m is necessary to ensure the existence of a non-interaction zone between the stope and footwall drift at the bakibaba copper mine. our analyses also show that the gsi value of the rock mass has a significant effect on this non-interaction zone distance up to a gsi value of 55, after which higher gsi values do not increase the required non-interaction-zone distance.. references abdellaha, w., mitria, h., & thibodeau, d. (2011). assessment of mine haulage drift safety using probabilistic methods of analysis. procedia engineering, 26, 2099–2111. doi: https://doi.org/10.1016/j. proeng.2011.11.2412 addenbrooke, t. i., & potts, d. m. (2001). twin tunnel interaction: surface and subsurface effects. international journal of geomechanics, 1(2), 249–271 chakeri, h., hasanpour, r., hindistan, m. a., & unver, b. (2011). analysis of interaction between tunnels in soft ground by 3d numerical modeling. bulletin of engineering geology and the environment, 70, 439–448. doi: http://dx.doi.org/10.1007/s10064-010-0333-8 gercek, h. (2005). interaction between parallel underground openings. the 19th international mining congress and fair of turkey, imcet ghaboussi, j. & ranken r. e. (1977). interaction between two parallel tunnels. international journal for numerical and analytical methods in geomechanics, 1(1), 75-103. doi: https://doi.org/10.1002/ nag.1610010107 hoek, e., carranza-torres, c. & corkum, b. (2002). hoek-brown failure criterion2002 edition. proceedings narmas-tac conference, toronto, 1,267-273 jing, l., & hudson, j. a. (2002). numerical methods in rock mechanics. international journal of rock mechanics and mining, 39, 409–427. kabwe, e., & bowa, v. m. (2016). determination of the appropriate geometry of footwall drifts using numerical analysis technique. geotechnical and geological engineering, 34, 1955–1969. doi: https://doi. org/10.1007/s10706-016-0076-9 karakus, m., ozsan, a., & basarir, h. (2007). finite element analysis for the twin metro tunnel constructed in ankara clay-turkey. bulletin of engineering geology and the environment, 66, 71–79. doi: http:// dx.doi.org/10.1007/s10064-006-0056-z ozdogan, m. v, yenice, h., gonen, a., & karakus, d. (2017). optimal support spacing for steel sets: omerler underground coal mine in western turkey. international journal of geomechanics, 18(2), 05017003 1-12. http://dx.doi.org/10.1061/(asce)gm.1943-5622.0001069 purwanto, a., shimada, h., sasaoka, t., wattimena, r. k., & matsui, k. (2013). influence of stope design on stability of hanging wall decline in cibaliung underground gold mine. international journal of geosciences, 4, 1-8. doi: http://dx.doi.org/10.4236/ijg.2013.410a001 sharma, k. g. (2009). numerical analysis of underground structures. indian geotechnical journal, 39(1), 1–63 sheorey p. r. (1994). a theory for in situ stresses in isotropic and transversely isotropic rock. international journal of rock mechanics and mining sciences, 31(1), 23-34. doi: https://doi.org/10.1016/0148-9062(94)92312-4 wang, b., & li, s. (2008). a complex variable solution for stress and displacement field around a lined circular tunnel at great depth. international journal for numerical and analytical methods in geomechanics, 33, 939-951. doi: https://doi.org/10.1002/nag.749 wang, h. n., zenga, g. s., utilic, s., jiangd, m. j., & wua, l. (2017). analytical solutions of stresses and displacements for deeply buried twin tunnels in viscoelastic rock. international journal of rock mechanics & mining sciences, 93, 13–29. doi: https://doi.org/10.1016/j. ijrmms.2017.01.002 wang, j., milne, d., wegner, l. & reeves, m. (2007). numerical evaluation of the effects of stress and excavation surface geometry on the zone of relaxation around open stope hanging walls. international journal of rock mechanics & mining sciences, 44(2), 289-298. doi: http:// dx.doi.org/10.1016/j.ijrmms.2006.07.002 yamaguchi, i., yamazaki, i., & kiritani, k. (1998). study of ground-tunnel interactions of four shield tunnels driven in close proximity, in relation to design and constructions of parallel shield tunnels. tunnelling and underground space technology, 13(3):289–304. doi: https://doi. org/10.1016/s0886-7798(98)00063-7 50 55 60 65 70 75 80 85 90 95 0 20 40 60 80 100 gsi d is ta nc e be tw ee n st op e an d fo ot w al l d ri ft re qu ir ed fo r e ns ur in g no nin te ra ct io n zo ne ( m ) figure 9. variations in distance between stope and footwall based on gsi value untitled earth sciences research journal earth sci. res. j. vol. 13, no. 1 (june 2009): 40-53 rio grande cone tectono-stratigraphic model – brazil: seismic sequences castillo., l.l.a¹, kazmierczak, t. de s.², and chemale., jr., f.³ ¹ profesor, curso de geofísica, departamento de geociencias – universidad nacional de colômbia bogota. ² schlumberger ltd, brazil. ³ professor, instituto de geociêcias, universidade federal do rio grande do sul, porto alegre – brazil. corresponding author email: lacastillol@unal.edu.co, thaissk@povo.net and farid.chemale@ufrgs.br resumen análisis sísmicos integrados con técnicas de interpretación, procesamiento, información de velocidades y datos geofísicos permiten modelar diferentes rasgos de tipo tectónico, estructural o geomorfológico. este trabajo se constituye en una primera aproximación a un modelo tridimensional de un rasgo geomorfológico ubicado en el fondo oceánico. el modelo inicial comprende una configuración bidimensional de una grilla interpretada a partir de secciones sísmicas llevadas a un volumen tridimensional, esto mediante la concepción de una aproximación dos y medio dimensional. además de un cambio de configuración, se hizo un análisis de atributos sobre las imágenes de las secciones y se determinaron las características presentes desde un área de la plataforma, hasta el offshore del sudeste brasilero. el rasgo geomorfológico a tratar es conocido como el cono de rio grande, el cual pudo ser caracterizando con sus principales estructuras tectónicas y estratigráficas, mediante la cartografía y modelamiento tridimensional del subsuelo, con la integración geológica, geofísica y geomorfológica. la geomorfología del cono, se ve influenciada por procesos gravitacionales, corrientes de fondo, estructuras de tipo tectónico y composición litológica, como sus principales controladores que están presentes a partir del neógeno. así, a partir de la interpretación sísmica pueden ser extractados diferentes estructuras tectónicas, sedimentarias y geoformas, por ejemplo presencia de fallas, pliegues, canales, levee naturales, contornitos, etc. palabras clave: sismoestratigrafía, modelamiento, tectonoestratigrafía, interpretación geofísica. 42 manuscript receiver: february 03th, 2009. accepted for publication: june 01th, 2009. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:47 p p p composite 133 lpi at 45 degrees abstract the seismic analyses integrated to interpretation techniques, processing, velocity model and geophysical data have permitted modeling different tectonic, stratigraphy or geomorphology features. this paper is the first approach to geomorphological features for a three dimensional depth model. the initial model comprises a seismic section grid with bidimensional configuration conversed to three dimensional subvolumes. these include a two and half dimensional model. also, an attribute analyses in image section is necessary to determine the most relevant characteristics. the geomorphologic feature corresponds to the named rio grande cone, which is characterized by principal tectonic and stratigraphic structures. that is why mapping and subsurface three dimensional modeling, with geology, geophysical and geomorphologic integration is necessary. the cone geomorphology is characterized by gravitational processes, bottom currents, tectonic structures and lithological compositions which are the principal factors that have been influencing it since neogene. then, the seismic interpretation could extract different tectonic, sedimentary structures and geoforms, for instance, faults, folds, channels, natural levee, contourites, etc. key words: seismic stratigraphy, modeling, tectono stratigraphy, geophysical interpretation. introduction seismic stratigraphy method has improved since development of acquisition, processing and interpretation processes. they include loading, processing, visualization and modeling tools that have allowed the manipulation of format data, screen and the creation of a high quality model. data could be transformed to different configurations like 2d, 2.5d and 3d. the first concepts related to seismostratigraphy interpretation refer to geometry and stratal terminations (mitchum et al. 1994).in the brazil’s southeast area, the rio grande cone has been influenced by tectonic episodes since the cretaceous until the recent. two and half dimensional seismic interpretation on some sections along the surface and the subsurface show the presence of some features located on shelf. they include a fault system, geoforms and folds that influence the slope, proximal and distal shelf and the sea floor. the tectono-stratigraphyc analysis make the description of some geological aspects in subsurface possible. as a consequence, a geological model was obtained by means of the seismic information and a velocity function within seismic interpretation (sarta, 2004). the geological mapping was obtained by means of the seismic stratigraphic interpretation on the rio grande cone . this interpretation allowed the division of the deep sedimentary package into several sequences and a driven-model from a sequential stratigraphy classification could be made. to define seismic units (chronoestratigraphic unit referred as sequences), the first step was to define the unconformities and then recognize the unit according to the facies, the seismic expression and the attributes analyses. 43 rio grande cone tectono-stratigraphic model – brazil: seismic sequences figure 1. rio grande cone location within the pelotas basin – brazil (modified from google earth). julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:48 p p p composite 133 lpi at 45 degrees geomorphology, geophysics and computational disciplines are powerful tools that could be integrated in order to obtain a 3d approach by using a driven-model method. this model allows the generation of a more realistic feature of the subsurface geomorphology. in this preliminary research a 3d model obtained from seismic interpretation in the rio grande cone will be documented. location rio de la plata and rio grande cone were formed during the maastrichtian/danian atlantic transgression in south america. the continental shelf and the depression are flooded by the sea. they are typical examples of geoforms can be seen on the uruguay and brazilian platforms. the rio grande cone (rgc) is a relevant geoform that extends 28900 km² on the brazilian offshore (figure 1). it is a subsurface landscape semicircular-shaped noticeable feature that comprises a transition from shelf to slopes with a high amount of shale supply. its source is the highlands and the brazilian craton. seismic data in this analysis, the data corresponds to 3577km of seismic lines that were acquired from brazil’s anp, and other old printed sections (fontana, 1996; porto, 2007). structural and straigraphy models were generated by means of software modules: petrel 2008, matlab (v. 7.0.1), voxler, seismic unix and open source seismic interpretation software (opendtect v3.03e). the dominant frequency seismic information could be obtained from attributes analyses, i.e. an instantaneous frequency map, which exhibits a 30-35hz dominant frequency and 90 µs/ft slowness in the area. a 4ms sample rate, 5 to 10s record length and 28ms windows time analyses are comprised. in order to establish the wavelength, geophysical parameters on pelotas basin whose average velocity is 3050 m/s and frequency of 30hz need to be considered. the expression for wavelength is given by: ë= v/f. (1) then, ë = 3048m/s / 40 hz = 76,2m, by the ë/4 => 76,2 m/4 = 19,05m => 62,5 feet. these calculations determined that the obtained thickness for a minimum layer is between 20 to 100 feet (7 to 30m), otherwise it is not tuned. the seismic resolution limit in section data establishes what can and cannot be seen on subsurface. attribute analysis all information extracted from seismic data is known as attribute and the interpretation depends on the combination of attributes, the quality of data, and the interpreter’s experience. the attribute is used to highlight subsurface features in order to delimit horizons, zones, seismic facies, and geometries by increasing the ability to define structural models and stratigraphic analysis. the advantage of seismic data is the multiplicity, areal extension and depth penetration as it permits obtaining information from seismic attributes (taner et al. 1977, 1979). seismic attributes are calculated from at least two trace input and provide information about lateral variations in set data. an attribute is not restricted to structural or stratigraphic analysis; it has been used to estimate petrophysic properties and geomorphological elements by combining them with well data. historical development of seismic tools and techniques improve the execution of the different stages of seismostratigraphic analysis. seismic stratigraphy and attribute analysis could be considered as tools for geomorphological, geological and geophysical modeling (figure 2). seismic attributes application seismic attributes were introduced in the seventy’s first to display seismic data, then to obtain different derived seismic measures, and finally turned into an analytical tool to predict lithological characterization. although a direct relationship between the attributes and the geological characteristics of the 44 castillo., l.l.a, kazmierczak, t. de s., and chemale., jr., f. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:48 p p p composite 133 lpi at 45 degrees earth has not been established, almost all analyses describe several uses of seismic attributes i.e. classes’ classification discriminator. recently, the seismic attribute application includes surfaces, horizon, geomorphology mapping, sequence stratigraphic interpretation, modeling subsurface and geobody information. they all have been accepted as tools to interpret old strata and processes through the use of modern survey (posamentier and kolla, 2003). then seismic geomorphology is an evolving field, building on the historic approaches of seismic stratigraphy, sequential stratigraphy, fluvial geomorphology and three dimensional modeling. seismic attributes are used in structural, stratigraphy and geomorphologycal interpretation (figure 3). an example of this is the improvement of image resolution thanks to the sophistication of computational systems that allow determining reservoir properties and their lateral continuity. figure 3. dip seismic line display of seismic attribute analysis and seismic stratigraphy interpretation 45 rio grande cone tectono-stratigraphic model – brazil: seismic sequences adquisition processing time line interpretation 1960 1970 1980 1990 2000 now 1 d 2 d 2.5 d 3 d 4 d c o n f i g u r a t i o n cmp configuration air gun marine digital recording digital seismic processing bright spot technology digital migration comercial seismic attribute comercial in colors color display 3 d shot vertical seismic profile introduction to seismic attribute concept (anstey) energetic crises dip moveout geophysicist stratigraphy avo analysis attribute proliferation seismic attributes expansionocean deep cable seismic attributes analysis texture attributes analysis of multi-attributes attributes and horizons classes synthetic seismogram workstation complex traces analyses trace complex seismic stratigraphy seismic inversion analyses horint attributes dip, zimur 3d autotracking attributes respose 3d seismic (brown) subsalt imgen seismic stratigraphy sequences attribute analysis attribute classification spectral decomposition coerence attribute introduction neural network geomorfología sísmica volumentric attributes spectral inversion figure 2. historical development of the seismic method: acquisition, processing, interpretation and modeling (adapted from liner, 2008; chopra and marfurt, 2005 and friedman, 1998). julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:48 p p p composite 133 lpi at 45 degrees in the seismic images processing, local seismic attributes analyses (lsaa) are being used to measure seismic signal characteristics of neighboring points. this technique has been applied in different steps of multicomponent seismic image acquisition (fomel, 2007). in this paper attributes were used to delineate structural (dip/azimute, ant-tracking, dip deviation, local structural dip, structural smoothing, variance, etc), stratigraphy (iso-frequence component, local flatness, acoustic impedance, etc) and geomorphologic (gradient, coherence, strata slices attributes, horizon slice attributes etc) features (figures 2 and 3).coherence attribute measures the similarity of a trace with its neighbors and display discontinuities, faults and channels (marfurt et al. 1998). this criterion guides the interpreter in the establishment of seismic reflection attributes and seismic facies patterns. instantaneous amplitude and frequency, which are the most frequently used, show changes in lateral continuous reflections. some attribute analyses were applied to seismic lines belonging to the brazilian offshore. the most important features are the lineaments and fractures found by means of local flat analyses and relative acoustic impedance that showed normal faults influencing the sedimentary package into the rio grande cone. root meters square permits delineating strong reflectors that could be correlated and interpreted by 46 castillo., l.l.a, kazmierczak, t. de s., and chemale., jr., f. structural stratigraphicseismic attributes sequencial stratigraphy interpretation local flat rms windows 9 dip seismic line acoustic impedance rms windows 3 schematic sequence smoothisofrequence figure 3. dip seismic line display of seismic attribute analysis and seismic stratigraphy interpretation julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:50 p p p composite 133 lpi at 45 degrees a sequence stratigraphic model, and delineated by geomorphological elements like channels. seismostratigraphy and sequence stratigraphy of the rio grande cone rio grande cone is located in a passive margin basin characterized by several progradational systems supplied by fluvial and cratonic sediments influenced by sea level fluctuations. some geological features have been reported by authors like alves (1977); fontana (1996) and porto (2007). seismic section and attributes permit identifying structural, sedimentology, and geomorphology features and they allow determining structural and the stratigraphic sequences elements. tectonic structures description the southeast of the brazilian offshore is composed by sequences influenced by rift faults causing a half-graben configuration on the basal sequence. the rio grande cone comprises pos-rift sequences. the post-rift stage starts during aptian, in which the adiastrophic tectonic developed, influencing basal sequences, i.e., the aptian-albian sequence. above it more than 3000 m of sediments belonging to rio grande cone were laid (fontana, 1996). structural sequences are influenced by tectonic and sedimentary structures. relevant structures correspond to active system faults until pleistocene -late wisconsin (alves, 1977). from a structural point of view, fault systems are the most relevant factor that affects clastic sequences. the first domain contains a faults system, including listric faults, thrust and their detachment planes. those structures are characterized by normal faults that were originated by distensive stress generating displaced blocks located on the proximal area (northeast-southwest trends and vergence to the northwest). there are some small faults on the cone’s distal section which structural style changes to inverse faults. this style is located to the end of the section (figure 4). some seismic section permit identifying gravitational deposits associated to fault systems, generating deformational and tectonic structures. studies have 47 rio grande cone tectono-stratigraphic model – brazil: seismic sequences figure 4. three-dimensional structural model of the rio grande cone, brazilian southeast. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:51 p p p composite 133 lpi at 45 degrees determined that some lineaments and faults were originated by fluids decompactation, and have influenced on the generation of fluid escape structures visible on the sea bottom’s surface (fontana, 1996; porto, 2007). the rio grande cone constitutes a huge sedimentary package characterized by the result of channel system morphology, sediment waves and contourities that have been influenced on the upper sequences by a distensional fault system that originated normal faults to the north and inverse faults to the south. stratigraphic sequence in the rio grande cone the seismic stratigraphic interpretation of the pelotas basin includes at least sixteen sequences (butler, 1970; fontana, 1996 and porto, 2007). the cone’s area has geoforms, internal structures and velocity models that can be described with geophysical methods, for instance seismic refraction. the refraction method revealed along to rio grande do sul and the uruguay’s continental margin a 1.8 km/s wedge, and a velocity model for the sedimentary sequence (alves, 1977). the wedge sedimentation lasted from middle miocene to pleistocene. stages of deposition and erosion were caused by eustatic sea level fluctuations, originating the development of four sedimentary sequences. the depocenters distribution suggests a source migration towards the continent, maybe due to marine transgression (alves, 1977). a different interpretation for the pelotas basin and florianopolis shelf established eight sedimentary sequences using seismic reflection (gonçalves et al. 1979). martins (1983) described the rio grande cone as a sedimentary-originated deep sea feature supplied by rio grande do sul highlands, with progradational deposition and gravitational processes (turbidities and another flux) modeled by bottom current. some researchs reports hydrate gas in the pelotas basin, i.e., fontana (1989), rosa et al. (2006). 48 castillo., l.l.a, kazmierczak, t. de s., and chemale., jr., f. figure 5. seismic dip line (di) interpretation of surface (su: unconformity; mrs: maximum regressive and mfs: maximum flooding surfaces) sequences (hnr: high normal regressive, lnr: lowstand normal regressive ). julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:10:52 p p p composite 133 lpi at 45 degrees the generated trasgressive (t), highstand (hnr) and lowstand regressive (lnr) could be identified on seismic sections (figure 5). the shelf edge suffered erosion resulting in the truncation of the prograding sediments. the eroded sediments fed it via suspension or gravitational processes. this feeding was four times greater than holocene rates. seismo-stratigraphy and sequence stratigraphy description in order to describe the evolution of the rio grande cone, it is important to describe its tectono-stratigraphy. the cone comprises several continental and post-rift (drift) sequences. the middle and upper sequences belong to the rio grande cone (figures 5 and 6). the middle sequences correspond to onlap transgressive sequences and downlap basinward. all sequences permit identifying montiforms with bidirectional downlaps that represent the lowstand fan unit deposited on the basin’s deeper area. this is the first lowstand stage in which a rapid decrease on eustatic curve inflection occurs. few continuous reflections and variable amplitudes are found on paleoslopes. middle sequences with erosion surfaces changed from cretaceous to tertiary. geological mapping considerations comprise sequences that have been influenced by different structural and 49 rio grande cone tectono-stratigraphic model – brazil: seismic sequences figure 6. sea level variations (haq et al, 1987) and cronostratigraphy interpretation correlation in the rio grande cone (alves, 1977; fontana, 1996; porto, 2007 and this paper). julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:07 p p p composite 133 lpi at 45 degrees stratigraphic styles dominated by facies variations that were caused by a high quantity of sediments supply. the upper sequences that belong to the rio grande cone are characterized by changes from retrogradational to progradational sequences. transgressive (t) and regressive features (lnr) are the result of sea level fluctuations that occurred from aptian. in the holocene transgression (t), the southeast continental margin has not received any significant quantity of terrigenous supply. two processes are still active: widespread pelagic and sedimentation geostrophic. contour current activity developed along lower continental rise. bottom-currents deposits result from upslope flowing processes, while gravity deposits result from downslope process. recent drift sequences allow identifying slope features deposited by bottom-currents, slump and gravity deposits process (alves, 1977). due to the instability of the upper sequences originated by steep slopes, the seismic has established slumps geometric characteristics that lead to a progradational system. the sediment supply in a decline plane influenced by a higher sea level produces a mass flow deposit. the highstand (hnr) is constant from paleocene to middle eocene and oligocene, through regressive intermittent cycles that finished with an oligocene regression (this is shown by hard layering, as result of paleoshelf). there are regional terraces and an erosive surface covered by deltaic events until upper pleistocene on rio de la plata, (martins, 1983; martins et al. 1990). they have similar geometries to the rio grande cone. sedimentary effects in the paleocene cycles could be evident in quiescence tectonic causing stability. in the final cycle, as result of andean tectonics, a regional tilt from west south american took place and 50 castillo., l.l.a, kazmierczak, t. de s., and chemale., jr., f. figure 7. 3d model obtained from seismic stratigraphic interpretation showing paleoshelf and several stratigraphic sequences and highlights the tectono stratigraphic features of the rio grande cone. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:09 p p p composite 133 lpi at 45 degrees caused the beginning of the sea level decrease and the deposition of progradantional sequences. this event could be coincident with haq curve (figure 6). rio grande modeling and mappingthe boundaries and horizon picking obtained in 2,5-d configuration permitted constructing the three-dimensional subsurface image of the rio grande geoform (figure 7). seismic stratigraphic analyses and basal sequence mapping permit identifying platforms with regressive progradational sediments. the sequences are comprised by sequential systems in a marginal sag type since late cretaceous. these correspond to the pelotas basin sequences. the semicircular-shaped plan-view morphology extends to the southeast and strike to the north. the thick recent sequences are crossed by faults systems and by the presence of geophysics anomalies, for instance a reflector that simulates the bottom surface (bsr). this reflector is an anomaly generated by the high impedances caused by hydrate gas presence. below 500 ms, a system faults extends cutting all upper sequences. near offset or proximal sector it includes some clinoform set which has been separated by internal downlap surfaces and are not faulted. visualization and 3d modeling has allowed defining fault systems geometries (figure 7), showing details of normal faults. this process includes fault propagation and could be interpreted as a polygonal faults system due to contraction factors and early fluid expulsion during possible burial, where compaction acts in different directions on clay rich sediments.the rio grande cone progradational layering produce continuous offlap, evidencing features like the continental slope, the marginal cone, progradational systems, and the upper interval that constitute recent continental shelf (urien et al. 2003). the last stage during the inferior holocene sedimentary dispersion included erosive process through submarine sedimentation. eustatic changes are less evident during the neogene. the highstand and lowstand alternation could be aged miocene and pliocene. they are associated to eustatic episodes with sea level changes that are controlled by andean orogenic pulses. conclusionthe feature named rio grande cone could be defined as a huge semicircular shape geological body with a thick package of sediments. it is mainly a shale geoform. the area is influenced by complex structures that affect the whole sedimentary package. the rio grande cone is overlying an older inferior and a middle sedimentary sequence that constitute the pelotas basin sequences. the cone comprises the basin’s younger sequence which is the upper sequence.the three dimensional analysis comprises an academic research of seismic data applied to the modeling and visualization for interactive understanding. this will permit a paleogeomorpholgy reconstruction of rio grande cone. a methodology and a typical sequence of seismic interpretation are given in this paper. the purpose is to show the importance of visualization, attribute analysis and modeling, especially in sequential stratigraphy. it is important to emphasize the use of different computational technologies as tools for data interpretation, visualization, and the recovery of information for the subsurface image by means of a velocity model. acknowledgments this article was written for the doctorate program research in the federal university of rio grande do sul (ufrgs porto alegre, brazil) sponsored by national university of colombia. thanks to national petroleum agency for providing some of the seismic lines used in this paper. thanks to schlumberger for the software contribution, especially to the technical support for petrel’s (module and tools 2008), processing, visualization and seismic interpretation with the geophysical integration data. references al-husseini, m., glover j. and barley b. (1981). dispersion patterns of the ground roll in east51 rio grande cone tectono-stratigraphic model – brazil: seismic sequences julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:10 p p p composite 133 lpi at 45 degrees ern saudi arabia, geophysics, vol 46, p. 121-137. alves, e. c. (1977). estrutura rasa do talude e sopé da margem continental do rio grande do sul e uruguai. tese de mestrado. instituto de geociências, universidade federal do rio grande do sul, porto alegre/rs, 77p. al-yahya, k.m. (1991). application of the partial karhunen-loève transform to suppress random noise in seismic sections: geophys. prosp., 39, p. 77-93. anjos-zerfass, g., souza, p. a., chemale jr. f. (2008). biocronoestratigrafia da bacia de pelotas: estado atual e aplicação na geologia do petróleo. revista brasileira de geociências 38(2), 42 – 62. butler, l. w. (1970). shallow structure of the continental margin southern brazil and uruguay. geological society of america bulletin, new york, n.y., v.81, p. 1079-1096catuneanu, o. (2006). principles of sequence stratigraphy. elsevier’s sciences & technology. british library, p. 375. chopra, s. and marfurt, k. j. (2005). seismic attributes – a historical perspective. society exploration geophysicists 75th anniversary. geophysics, vol. 70, n. 5. 26 p. d’agosto, c., marfurt, k. and steven, j. (2003). modeling and removal of ground roll from horizontal component of c-waves. seg. 1981-1983. dias j. l., silveira, d. p., sad a. r. e., and latgé m. a. l. (1994). estágio atual do conhecimento geológico. boletim de geociências da petrobras, 8(1), p. 235-245. fomel, s. (2007). local seismic attributes. geophysics. vol. 72. n. 3. pag. a29-a33. fontana, r. l. (1989). evidências geofísicas da presença de hidratos de gás na bacia de pelotas – brasil. primer cong. soc. bras. geofísica, 7(1):93. fontana, r. l. (1996). geotectônica e sismoestratigrafia da bacia de pelotas plataforma de florianópolis. tese de doutorado, universidade federal do rio grande do sul, porto alegre/rs. 214p. friedman, m. g. (1998). sedimentology and stratigraphy in the 1950’s to mid-1980s: the story of a personal perspective. episodes. vol. 21. no. 3. new york. usa. 172-177. gomide, j. (1989). bacia de pelotas – bioestratigrafia baseada em nanofóeseis calcáreos. in: sbp, congresso brasileiro de paleontologia, anais 11, 338-351. gonçalves, a., oliveira, m.a. and oliveira, m. s. (1979). geología da bacia de pelotas e da plataforma de florianópolis. boletim técnico da petrobrás, rj. 79:157-174.haq, b. u., hardenbol, j., and vail, p. r. (1987). chronology of fluctuations sea levels since the triassic (250 years ago to present): science, v. 235, p.1156-1167. jones, i. f., and levy, s. (1987). signal noise enhancement in multichannel seismic data processing: geophys. prosp., 35, p. 12-32. karsh, h., and bayrak, y. (2004). using the wiener – levinson algorithm to suppress ground roll. journal of applied geophysics 55, p 187-197. levy, a. and linderbaurn, m. (2000). sequential karhunen-loève basis extraction and its application to images, ieee transactions on image processing, vol 9, n. 8, p. 1371-1374. liner, c. (2008). time line data. american association of petroleum geologist – explorer. november. p 31. luo, y., marhoon, m., al dossary, s., and alfaraj, m. (2002). edge-preserving smoothing and applications: the leading edge, 22, 136, 138, 141-158. marfurt, k.j., kirlin, r. l.,farmer, s., l. and bahorich, m.s. (1998). 3-d seismic attributes using a semblance-based coherence algorithm: geophysics, 63, 1150-1165. 52 castillo., l.l.a, kazmierczak, t. de s., and chemale., jr., f. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:10 p p p composite 133 lpi at 45 degrees martins, l. r.; uruen, c. m.; buttler, l. w. (1972). províncias fisiográficas e sedimentos modernos da margem continental atlântica. congresso brasileiro de geologia, 26o, belém-pará. anais...sociedade brasileira de geologia, p.105-114. martins, i. l. (1983). modelo sedimentar do cone de rio grande. tese de doutoramento. instituto de geociências, universidade federal de rio grande do sul, porto alegre, rs. 205p. martins, l. r.; martins, i. r. and urien, c.m. (1990). episódios dinâmicos de curta duração na margem continental do rio grande do sul: registro e importância. acta geológica leopoldensia. 29 (2): 9-22. são leopoldo. brasil. mitchum, m. r. jr., sangree, j., vail, p., and wornardt w. (1994). reconizing sequences and system tracts from well logs, seismic data, and biostratigraphy: examples from the late cenozoic of the gulf of mexico. in: weimer, p. and posamentier, h. w. (1994). siliciclastic sequence stratigraphy: recent developments and applications. american association of petroleum geologist, memoir, 58. p. 163-197. porto, r. a. (2007). interpretação sismoestratigrafica da porção da bacia de pelotas que engloba o cone de rio grande e a avaliação do seu potencial petrolífero. tese de doutorado. universidade estadual do norte fluminense. macaé/rj. 264 f. posamentier, h., and kolla, v. (2003). geomorphology and stratigraphy of depositional elements in deep-water settings. journal of sedimentary research; v.73; no. 3; p. 367-388. rosa, m. l. c., ayup-zouain, r. n., & barboza, e. g. (2006). utilização de seções sísmicas 2d na identificação de zonas de escapes de fluídos, gravel, n. 4. p. 109-118. porto alegre. scales, j. and snider, r. (1998). ¿what is noise? geophysics, vol 63, p. 1122-1124. sarta, e. (2004). implementación de un software de pre proceso y análisis de datos sísmicos seismic un. geociencias – universidad nacional de colombia. 163 p. taner, m. t., and sheriff, r. e. (1977). applications of amplitude, frequency and other attributes to stratigraphic and hydrocarbon determination: american association petroleum geologist, memoir 26, 301-327. taner, m. t., kochler, f., sheriff, r. e. (1979). complex seismic trace analyses. gephysics, 44, no. 6, 1041-1063. urien, c. m., martins, l. r. and martins, i.r. (2003). paleoplataformas e programação deltaica do neógeno na margem continental do uruguai e norte a argentina. gravel n. 1. 40 46. porto alegre brasil. yilmaz, o. (1987). seismic data processing: society of exploration geophysicists, tulsa, p. 9-79. zywicki, d., (1999). traditional seismic surface wave tests. georgia tech research institute (gtri), p. 11.34 53 rio grande cone tectono-stratigraphic model – brazil: seismic sequences julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:10 p p p composite 133 lpi at 45 degrees this paper investigates the potential of data mining techniques to predict daily soil temperatures at 5-100 cm depths for agricultural purposes. climatic and soil temperature data from isfahan province located in central iran with a semi-arid climate was used for the modeling process. a subtractive clustering approach was used to identify the structure of the adaptive neuro-fuzzy inference system (anfis), and the result of the proposed approach was compared with artificial neural networks (anns) and an m5 tree model. result suggests an improved performance using the anfis approach in predicting soil temperatures at various soil depths except at 100 cm. the performance of the anns and m5 tree models were found to be similar. however, the m5 tree model provides a simple linear relation to predicting the soil temperature for the data ranges used in this study. error analyses of the predicted values at various depths show that the estimation error tends to increase with the depth. abstract keywords: soil temperature; data mining; m5 tree model; anfis; ann. estimation of daily soil temperature via data mining techniques in semi-arid climate conditions issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n2.49829 earth sciences research journal earth sci. res. j. vol. 21, no. 2 (june, 2017): 85 93 m. taghi sattari1*, esmaeel dodangeh2, john abraham3 *1department of water engineering, faculty of agriculture, university of tabriz, tabriz, iran. email: mtsattar@gmail.com (corresponding author), 2department of natural resources, sari university of agriculture and natural resources, sari, iran. email: smaeel.dodangeh@gmail.com 3university of st. thomas, minnesota, school of engineering 2115 summit avenue st. paul, minnesota 55105, usa. jpabraham@stthomas.edu este artículo investiga el potencial de las técnicas de búsqueda y procesamiento de datos para pronosticar las temperaturas diarias del suelo a profundidades que van de los 5 a los 100 cm con propósitos agrícolas. se utilizó la información climática y de temperatura del suelo de la provincia ishafan, ubicada en el centro de irán y de clima semiárido, para el proceso de modelamiento. se usó un enfoque de agrupamiento sustractivo para identificar la estructura del sistema de inferencia neuronal difuso adaptado (anfis, del inglés adaptive neuro-fuzzy inference system) y el resultado del acercamiento propuesto se comparó con redes artificiales neuronales (ann) y el modelo tipo árbol m5. los resultados sugieren un desempeño mejorado al usar el enfoque anfis en la predicción de las temperaturas del suelo en varios puntos de profundidad, excepto en los 100 cm. el desempeño de las redes artificiales neuronales y los modelos de árbol m5 fueron similares. sin embargo, el modelo tipo árbol m5 provee una relación linear simple para predecir los rangos de datos de la temperatura del suelo utilizados en este estudio. los análisis de error de los valores predichos a varias profundidades muestran que la estimación de error tiende a incrementarse con la profundidad. resumen palabras clave: temperatura del suelo; minería de datos; modelo tipo árbol m5; anfis; ann record manuscript received: 26/03/2015 accepted for publication: 31/05/2017 how to cite item: sattari, m. t., dodangeh, e., & abraham, j. (2017). estimation of daily soil temperature via data mining techniques in semi-arid climate conditions. earth sciences research journal, 21(2), 85 93. doi:http://dx.doi.org/10.15446/esrj.v21n2. 49829 estimación de la temperatura diaria del suelo a través de técnicas de búsqueda y procesamiento de datos en condiciones climáticas semiáridas m e t e o r o l o g y 86 m. taghi sattari, esmaeel dodangeh, john abraham 1. introduction soil temperature prediction is important for various agricultural purposes, especially in arid and semi-arid regions, such as iran. temporal patterns of soil temperatures in these regions show large seasonal and daily fluctuations. these variations in soil temperature affect plant growth directly through their effect on physiological activities and indirectly through the effect on soil nutrient availability (tuntiwaranurk et al., 2006). for example, root growth and biological soil activity are directly influenced by soil temperature (kang et al., 2000). soil temperature fluctuations affect various processes within the soil such as microbial decomposition, p and k absorption, soil-moisture content (elshorbagy and parasuraman, 2008) and soil respiration (gaumont-guay et al., 2006). the success of seeding efforts depends greatly on the spatial and temporal distribution of soil temperature. thus, the temperature predictions help to improve the understanding of the dynamics of vegetation (kang et al., 2000). it also helps agronomists and engineers to decide the proper plantation date, design drainage and irrigation systems, and to optimize the application of pesticides and fertilizers to reduce chemical pollution of soils and groundwater. for these reasons, understanding of the variation in soil temperature is vital. however there are very few climatologic stations in arid and semi-arid regions where soil temperatures are recorded at various depths. it is usually quite difficult to measure temperature at depth. to aid in this measurement process, application of data-driven models are used to estimate daily soil temperatures in ungaged homogeneous regions. temperatures in soil are influenced by a number of factors, such as meteorological conditions (i.e. solar radiation and air temperature), site topography, soil water content, and whether the surface is covered by litter and canopies of plants. many conceptual models have been proposed to model soil temperature based upon the meteorological parameters such as surface global radiation and air temperature (kang et al., 2000; shannon et al., 2000; timlin et al., 2002), soil physical parameters, such as water content and texture, topographical variables such as elevation, slope and aspect (kang et al., 2000), and other surface characteristics such as leaf area index (lai) and ground litter stores. other models such as multiple regression and fourier analysis have also been suggested with some modification. most of these models are based on several assumptions and boundary conditions resulting in the limitations of their use in practice. for instance, kang et al. (2000) developed a hybrid soil temperature model based on heat transfer physics and a relationship between air and soil temperature to predict daily spatial patterns of soil temperature in a forested landscape. they incorporated the effects of topography, canopy and ground litter. despite of the availability of different models to predict the soil temperature, these models are found to work well only for specific climatic and agronomic conditions under which they were originally developed. over past decades, remote sensing techniques have also been utilized to measure and predict soil temperature over large areas but a major drawback of this approach is the availability of soil temperature data in the top few centimeters (elshorbagy and parasuraman, 2008). the temperature of the soil profile with increasing depth is difficult to predict so that the abovementioned techniques are limited to shallow soils (tyronese et al., 2008). within the last decade, artificial intelligent (ai) systems such as fuzzy logic (fl) and artificial neural networks (anns) have effectively been used to model nonlinear and non-stationary process (shiri and kisi, 2011). anns are mathematical models consisting of a network of computation nodes called neurons with established connections between them. fuzzy logic is an alternative technique capable of generating models that incorporate expert knowledge and available measurements for a system by using a set of easily comprehensible rules in the form of a fuzzy inference system (fis) (zadeh, 1965). a fis is a nonlinear mapping of a given input vector to an output using fuzzy logic based on a set of membership functions and rules. improved performance can be obtained by integrating fuzzy systems and the ann approach to deal with large and imprecisely defined complex systems. an adaptive neuro-fuzzy inference system (anfis) is one of the most successful schemes which combine the benefits of these two powerful paradigms into a single model. the goal of the anfis is to find a model or mapping that will correctly associate the inputs with the output. data mining refers to the process of searching for and discovering various patterns in data and of summarizing a set of known values to obtain the most important information (quinlan, 1992). tree-based methods are one data mining technique and their output is a model having the structure of a tree with input and output data. the m5 model tree was introduced by quinlan in 1992 and is a subset of data mining methods. the m5 algorithm is the most common classification used in the family of decision-making tree models. a decision tree model is essentially a decision-making tree in which linear regression equations at the leaves replace terminal class values. within the last decade, several studies reported the use of data mining techniques such as the m5 tree model for water resource issues applications (solomantine and dulal, 2003; bhattacharya and solomatine, 2005; stravs and brilly, 2007; pal et al., 2012; sattari et al., 2013a; sattari et al., 2013b; sattari, et al., 2014; esmailzadeh and sattari, 2015; biabani et al., 2016; shortridge et al., 2016; adnan et al., 2017; sayagavi et al., 2016; schnier, 2016). however to the best knowledge of the authors, there has not been an application of the m5 tree model to predict soil temperatures at various soil depths, especially in arid and semi-arid climates. our primary motivation in this study is to simulate soil temperatures at various depths. in this manner, it is possible to predict future soil temperatures by simply acquiring the climatic data from the meteorological stations. this is particularly important when planning future agriculture practices. this study will evaluate the performance of various data mining techniques such as anfis, anns and the m5 tree model for estimating daily soil temperature at different depths in semi arid regions such as iran. 2. materials and methods 2.1 study area and data used located in the central arid region of the country, the isfahan province of iran lies between 30o 42’ to 34o 30’n and 49o 36’ to 55o e (figure 1). the altitude in this area varies from 707 to 4000 m. the significant change in altitude and its effect on climate provide various habitats and diverse plant species. the experimentally obtained daily soil temperature and other meteorological parameters were measured at the weather station of isfahan from 1992 to 2005. the soil temperature data were obtained for soil profiles at various depths (i.e. 5, 10, 20, 30, 50 and 100 cm). meteorological parameters including: daily mean, minimum and maximum air temperature (ave t, min t, max t), evaporation (ev), daily sunshine hours (sunh) and radiation (ra) were also considered as inputs. figure 1. location of the study area 87estimation of daily soil temperature via data mining techniques in semi-arid climate conditions the descriptive statistics min, max, mean, standard deviation, coefficient of skewness (cs), coefficient of kurtosis (ck) and coefficient of variation (cv) of the soil temperature and meteorological data time series, are provided in table 1. table 1. descriptive statistics of the soil temperature data the available data were split into two parts. the first part consists of 3028 samples between june 1992 and october 2004 and was used to train different models. the second part, consisting of 335 samples between october 2004 and december 2005 was used for testing. the coefficient of determination (r2) and root mean square error (rmse) statistics were used to compare the performance of various modeling approaches used in this study. it is often useful to scale the input and output parameters before using them with anns. in the present work, input and output data were scaled to a range from -1 to +1, which is preferable when tan-sigmoid activation functions are used with the neural network. the following normalization equation was used (1) where z is standardized input values lying in the range of [-1, +1], and xmin and xmax are minimum and maximum input values, respectively. 2.2 adaptive neuro-fuzzy inference system (anfis) fuzzy logic represents knowledge using if–then rules in the form of “if x and y then z” (zadeh, 1965). fis mainly consists of fuzzy rules and membership functions and fuzzification and de-fuzzification operations (jang, 1993). figure 2 shows a typical architecture of an anfis. in this figure, circles represents fixed nodes, whereas squares indicate adaptive nodes. the input and output nodes represent the meteorological parameters and soil temperature, respectively. the nodes in the hidden layers act as membership functions (mfs) and rules. for simplicity, it is assumed that the examined fis has two inputs and one output. for a first-order sugeno fuzzy model, a typical rule set with two fuzzy ‘‘if-then’’ rules can be expressed as follows: (2) (3) where x and y are two crisp inputs, and ai and bi are the linguistic labels associated with the node function. the anfis has the multiple layers, as displayed in figure 2. figure 2. architecture of the anfis layer 1: all the nodes in this layer are adaptive nodes which mean that the outputs of the nodes depend on the parameters pertaining to these nodes. each node corresponds to a linguistic label which has a membership function that may be gaussian or any other mf. layer 2: every node in this layer is a fixed node labeled as ii, representing the firing strength of each rule. layer 3: every node in this layer is a fixed node labeled as n, representing the normalized firing strength of each rule. layer 4: every node i in this layer is an adaptive node with a node function defined as: (4) where oi4 is node output, wi is the normalizing firing strength from layer 3 and {pi, qi, ri} are the parameter set of this node. layer 5: the single node in this layer is a fixed node labeled σ which computes the overall output by summing all incoming signals and is the last step of the anfis. the output of the system is calculated as: (5) where o15( node output), is the weighted sum of right hand side polynomials in equation 5. the implementation of anfis consists of two major phases; the structure identification phase and the parameter identification phase. the anfis parameter estimation can be carried out by training the 88 m. taghi sattari, esmaeel dodangeh, john abraham anfis system by using a hybrid learning algorithm. the hybrid learning algorithm of anfis consists of the two parts: (a) the learning of the premise parameters by back-propagation and (b) the learning of the consequence parameters by least-squares estimation. in the forward pass of the hybrid learning algorithm, functional signals move forward to layer 4 to calculate each node output. the premise parameters in layer 2 remain fixed in this pass. the consequent parameters are then identified by the least-squares estimate. in the backward pass, the error rates propagate backward from the output towards the input, and the premise parameters are updated by the gradient descent (shu, et al., 2008). 2.2.1 subtractive clustering subtractive clustering is an automated data-driven approach to generate primary fuzzy models. the subtractive clustering algorithm is used in case the number of clusters is not clear. this algorithm depends on the structure of the data and can be used as a dimensionality reduction tool. this algorithm can also be used to generate a fuzzy system with the minimum number of rules required to distinguish the fuzzy qualities associated with each of the clusters. subtractive clustering is based on a measure of the density of data points in the feature space. the idea here is to find regions in the feature space with high densities of data points. consider a collection of n data points {x1,…,xn}, subtractive clustering algorithm assumes each data point as a potential cluster center. a density measure at a data point xi is then defined as: (6) where di is the density measure and the cluster radius rα is a positive constant (rα > 0) defining the neighborhood radius for each cluster center. thus, a data point that has many neighboring data points will have a high potential of being a cluster center. data points existing outside of this radius have little or no effect on the density measure. the choice of rα is crucial in determining the cluster numbers. large value of rα will generate a limited number of clusters, while small values of rα will generate a large number of clusters. after calculation of the potential of each vector, the one with the highest potential is selected as the first cluster center. suppose xc1 is the point selected and dc1 is its density measure. the density measure for each data point xi is revised by the formula: (7) where rb is a positive constant (rb > 0) that represents the radius of the neighborhood for which considerable potential reduction will happen in density measure. in order to avoid obtaining closely spaced cluster centers, the constant rb is usually 1.5 times that of rα. finally, the clusters’ information is used to determine the initial number of rules and antecedent membership function that is used for identifying the fis (chiu, 1994). 2.3. artificial neural networks (anns) the anns are alternative artificial intelligent (ai) methods and employed in this study to predict soil temperature. a number of network and training algorithms are reported in the literature. the multi-layer perceptron (mlp) is one of the mostly used ann in many research areas. this study uses mlp possessing a three-layer learning network consisting of an input layer, one hidden layer, and one output layer. the input layer accepts values of the input variables and the output layer provides estimations. the hidden layer which lies between the input and output layers contains the processing elements called as neurons. the hidden layer and nodes play very important roles for successful application of the neural network. the nodes in the hidden layer allow neural networks to detect the feature, to capture the pattern in the data, and to perform complicated non-linear mapping between input and output variables. it has been suggested that only one hidden layer is sufficient for anns to approximate any complex nonlinear function within desired degree of accuracy. in the case of the hidden layer, many studies suggest "2m+1" (hecht-nielsen, 1990; lippmann, 1987), "2m" (wong, 1991) and "m" (tang and fishwick, 1993) hidden neurons for better forecasting accuracy, where m is the number of input nodes. in the current study, a large number of trials were carried out to select the optimal number of nodes in the hidden layer. the transfer functions used in this study include the tan-sigmoid in the hidden layer and the linear transfer function in the output layer. the levenbergmarquardt training algorithm was used for the ann models because this technique is more powerful and faster than the gradient descent algorithm. 2.4 m5 model tree model trees generalize the concepts of regression trees and are analogous to piece-wise linear functions. a m5 model tree is a binary decision tree having linear regression function at the terminal nodes, which can predict continuous numerical attributes (quinlan, 1992). the m5 model tree is an algorithm for making numerical predictions, and the selected tree nodes have the attribute of maximum expected error that is a function of the standard deviation in the output parameters. a model tree based regression approach works in two different stages. in the first stage, a splitting criterion is used to create a decision tree. the splitting criterion for the m5 model tree algorithm is based on treating the standard deviation of the class values that reach a node as a measure of the error at that node and calculating the expected reduction in this error as a result of testing each attribute at that node. the formula for computing the standard deviation reduction (sdr) is: (8) (9) and t is a set of samples entering each node. the symbol ti represents a subset of the samples that have the ith result of the potentiality test, sd is the standard deviation, yi is the numerical value of the target attribute of sample i, and n the total number of data points (alberg et al., 2012). the splitting process allows the data in child nodes to have a lower standard deviation compared to their parent node and can be considered as more pure. after examining all the possible splits, m5 chooses the one that maximizes the expected error reduction. the division of training data with a m5 model tree produces a large tree-like structure which may cause overfitting of the data. a pruning algorithm is used to prune back the tree, for example by replacing a sub tree with a leaf in order to remove the problem of overfitting. thus, the second stage in the design of the model tree involves pruning the overgrown tree and replacing the sub trees with linear regression functions. this technique of generating the model tree splits the parameter space into areas (subspaces) and builds in each of them a linear regression model. for further details of m5 model tree, readers are referred to related studies (pal and surinder, 2009; quinlan, 1992). 3. results and discussion analysis of table 1 suggests a noticeable change in coefficient of variation (cv) value of soil temperature from the surface to the depth of 100 cm. the highest value of the coefficient of variation was observed at a 5 cm soil depth with a continuous decline in its value with increasing depth of the soil. other statistical properties such as the maximum and mean soil temperatures are also decreasing with increasing soil depth. 89estimation of daily soil temperature via data mining techniques in semi-arid climate conditions it can be observed that the variability in thermal behavior of the soil profile can be a cause of this change. the higher values of coefficients of variation in the surface layers indicate that higher variability of the soil temperature at this level is possibly due to the variety of causal mechanisms influencing soil temperature. figure 3 shows the variation of soil temperature at different soil depths in the study area. as can be seen from the figure, a wide range of fluctuations exists for the surface layers which decreases with increasing depth. 0 500 1000 1500 2000 2500 3000 3500 0 10 20 30 40 50 time (day) so il te m pe ra tu re (c o ) dep th 5cm dep th 10cm dep th 20cm dep th 30cm dep th 50cm dep th 100cm figure 3. time series of observed soil temperatures at different soil depths (5-100 cm) the three modeling techniques; anfis, ann and m5 tree model were used to predict the soil temperature at varying depths. to obtain optimal network architecture in anns, the number of neurons in the hidden layer was determined by trial and error. based on the number of input neurons (m = 6 input nodes are included here), the number of hidden nodes were varied from 1 to 2m+1 to find optimal number of nodes in hidden layer. the network was trained for 300 epochs using the levenberg-marquardt training algorithm. based on a trial and error process (by exploring 1 to 2m+1 hidden neurons), hidden layers with 3, 4, 5, 4, 5 and 3 neurons were found to be optimal for the 5, 10, 20, 30, 50 and 100 cm soil depths, respectively (see table 2 for the 10 and 50 cm soil depths). table 2. tested model structures and rmse values of the anns models. the parameters of the subtractive clustering algorithm should be specified in anfis models to predict soil temperatures at various depths. the clustering radius (ra) is the most important parameter in the subtractive clustering algorithm and is optimally determined through a trial-and-error procedure. the values of ra ranging between 0.2 and 1 with a step size of 0.01 were examined to minimize the root mean squared error (table 3). for the values below 0.2, network training was found to be difficult and for the values above 1, no remarkable change in rmse values was achieved. clustering radius rb was selected as 1.5ra and default values of the other parameters given in the matlab were used. table 3. resulting model structures and rmse values of the anfis models. *shows optimal clustering radius (ra) gaussian membership functions were used for each fuzzy set in the fuzzy system. the number of membership functions and fuzzy rules required for a particular anfis model were determined through the subtractive clustering algorithm. parameters of the gaussian membership function were optimally determined using the hybrid learning algorithm. each anfis model was trained for 100 epochs. the test results of the anns, anfis and m5 model trees are compared with respect to r2 and rmse in table 4. table 4. the test results of the optimal anns, anfis and m5 model trees. results indicates a high correlation (r2 value changes from 0.98 at 5 cm depth to 0.80 at 100 cm depth) between the actual and predicted soil temperatures suggesting that all three models achieved acceptable results in predicting the soil temperatures at varying depths. time variation and scatterplots of the test results obtained by m5 tree model, ann and anfis models are illustrated in figures 4-6. 90 m. taghi sattari, esmaeel dodangeh, john abraham plot of observed and predicted soil temperature values suggests that all three models are suitable in modeling soil temperatures. figure 5. performance of ann model at soil depths; 5 cm and 100 cm, with test data set 91estimation of daily soil temperature via data mining techniques in semi-arid climate conditions as can be seen from the table 4, the rmse values also increase with the increasing depth. the lowest and highest rmse values achieved by the anfis, ann and m5 tree model are 1.86, 1.95, 1.97 (observed at 5 cm soil depths) and 2.39, 2.44, 2.37 (observed at 100 cm soil depths) respectively. the plots of actual and predicted soil temperature in figures 4, 5 and 6 as well as table 4 suggest a slightly improved performance by anfis in predicting soil temperature in comparison to ann and m5 models except at the depth of 100 cm. a comparison of results from table 4 also suggests that both the ann and anfis perform better than the m5 model tree approach in predicting the surface soil temperature, however at depth of 100 cm, m5 tree model seems to be most robust. figure 6. performance of m5 model tree at soil depths; 5 cm and 100 cm, with test data set these figures depict good agreement between the actual and predicted soil temperature values of the surface layers in comparison to the layers at increasing depths. all three models tend to provide less biased estimates for the surface layers in comparison to the predictions for deeper layers (figure 7). figure 7. rmse values of the evaluated models for different soil depths analysis of predicted soil temperature values by different modeling approaches (figure 7) shows increasing values of rmse with increasing soil depth, thus indicates that the soil temperatures predicted by anfis, ann and m5 models are more accurate for the surface temperatures. the reason for increasing rmse value with increasing depth may be mainly due to the reduction in correlation between the input climatic variables and the soil temperature at increasing depth. the superior performance of the anfis and ann in modeling the surface soil temperatures may be attributed to the increase in the network nonlinearity and the better correlation between input and output values (gao et al., 2007). moreover it may be noted that a trial-anderror procedure has to be adopted to select suitable user-defined parameters for the ann model; this process is time consuming. on the other hand, no such procedure is required to develop an anfis model. although error analysis of predicted values confirmed better performance for anfis and ann approaches for the surface soil temperatures, the m5 model tree had slightly better results with increasing depth. it is worth noting that m5 model trees being analogous to piecewise linear functions, provides a simple linear relation to model the soil temperatures, as described mathematically in figure 8. 92 m. taghi sattari, esmaeel dodangeh, john abraham figure 8. linear functions for predicting temperature at 5 cm soil depth based on m5 tree model. 4. conclusions the anfis, anns and m5 tree model approaches have been used to predict the daily soil temperature with increasing depth in this study. the results presented here are quite encouraging and confirm that all three approaches work well in predicting soil temperatures at different depths. a comparison among the models indicates that the anfis model provides more accurate estimates of soil temperature than the anns and m5 tree model. the results also suggests that both ann and m5 tree model approaches work well in predicting daily soil temperatures, however, m5 tree model has simple linear relations that can be easily used to predict the daily soil temperature data by field engineers. error analysis of temperature predictions at different soil depths indicates that all three models perform well in predicting surface soil temperature data rather than at deeper depths. the reason behind this may be the fact that there is a strong relationship between climatic parameters and surface soil temperature. the results of the present study demonstrate that the proposed anfis model is quite efficient in predicting soil temperature for the surface layers however further investigation is needed with different data sets to compare proposed approach and other mathematical methods such as time series modeling to model soil temperatures in deeper layers. 5. references adnan, r. m., yuan, x., kisi, o., & anam, r. (2017). improving accuracy of river floe forecasting using lssvr with gravitational search algorithm. advances in meteorology. doi.org/10.1155/2017/2391621. bhattacharya, b., & solomatine, d. p. (2005). neural networks and m5 model trees in modeling water level–discharge relationship. neurocomputing, 63, 381–396. biabani, r., meftah halaghi, m., & ghorbani, k. h. (2016). m5 model tree to predict temporal evolution of clear water abutment scour. journal of geology, 6, 1045-1054. bonng, b. & vanclev, k. (1992). soil temperature nitrogen mineralization and carbon source-sink relationships in boreal forests. canadian journal of forest research, 22, 629-639. chio, j. s., fermanian, t. w., wehner, d. j. & spomer, l. a. (1988). effect of temperature, moisture and soil texture on dcpa degradation. agronomy journal, 80, 108-111. chiu, s. (1994). fuzzy model identification based on cluster estimation. journal of intelligent and fuzzy systems, 2, 762–767. elshorbagy, a., & parasuraman, k. (2008). on the relevance of using artificial neural networks for estimating soil moisture content. journal of hydrology, 362, 1– 18. esmaeilzadeh, b., & sattari, m. t. (2015). monthly evapotranspiration modeling using intelligent systems in tabriz, iran. agriculture science developments, 4(3), 35-40. gao, z., bian, l., hu, y., wang, l. & fan, j. (2007). determination of soil temperature in an arid region. journal of arid environments, 71, 157-168. gaumont-guay, d., black, t. a., griffis, t. j., barr, a. g., jassal, r. s. & nesic, z. (2006). interpreting the dependence of soil respiration on soil temperature and water content in a boreal aspen stand. agricultural and forest meteorology, 140, 220–235. hecht-nielsen, r. (1990). neurocomputing. addison-wesley, menlo park, ca, usa. jang, j. s. r. (1993). anfis: adaptive-network-based fuzzy inference system. ieee transactions systems, man and cybernetics, 23(3), 665–685. kang, s., kim, s., oh, s. & lee, d. (2000a). predicting spatial and temporal patterns of soil temperature based on topography, surface cover and air temperature. forest ecology and management, 136, 173-184. lippmann, r.p. (1987). an introduction to computing with neural nets. ieee assp magazine, 4–22. pal, m. & surinder, d. (2009). m5 model tree based modelling of reference evapotranspiration. hydrological processes, 23, 1437–1443. pal, m., singh n. k., & tiwari, n. k. (2012). m5 model tree for pier scour prediction using field dataset. ksce, 16(6), 1079-1084. quinlan, j. r. (1992). learning with continuous classes. in: proc. ai’92 (fifth australian joint conf. on artificial intelligence) (ed. by a. adams & l. sterling), 343–348. world scientific, singapore. raju, k. g. (2001). prediction of soil temperature by using artificial neural networks alghorithms. nonlinear analysis, 47, 1737-1748. sattari, m. t., pal, m., apaydin, h., & ozturk, f. (2013a). m5 model tree application in daily river flow forecasting in sohu stream, turkey. water resources, 40(3), 233-242. sattari, m. t., pal, m., yurekli, k., & unlukara, a. (2013b). m5 model trees and neural network based modelling of et0 in ankara, turkey. turkish journal of engineering and environmental sciences, 37, 211-219. sayagavi, v. g., charhate, s., & magar, r. (2016). estimation of discharge using ls-svm and model trees. journal of water resources and ocean science, 5(6), 78-86. shannon, e. b., kurt, s. p., david, d. r. & andrew, j. b. (2000). predicting daily mean soil temperature from daily mean air temperature in four northern hard wood forest stands. forest science, 46, 297-301. shiri, j., & kishi, o. (2011). comparison of genetic programming with neurofuzzy systems for predicting. computers & geosciences, 37, 1692-1701. shortridge, j. e., guikema, s. d., & zaitchik, b. f. (2016). machine learning methods for empirical streamflow simulation: a comparison of model accuracy, interpretability, and uncertainty in seasonal watersheds. hydrology and earth system sciences, 20, 2611-2628. schnier, s. t. (2016). data driven analyses of watersheds as coupled humannature systems. unvisesity of illinois, urbana, illinois. shu, c., & ouarda tbmj. (2008). regional flood frequency analysis at ungauged sites using the adaptive neuro-fuzzy inference system. journal of hydrology, 349, 31-43. siek, m. & solomantine, d. p. (2007). tree-like machine learning models in hydrologic forecasting: optimality and expert knowledge. geophysical research abstracts, 9, 2-5. 93estimation of daily soil temperature via data mining techniques in semi-arid climate conditions solomantine, d. p. & dulal, k. n. (2003). model trees as an alternative to neural networks in rainfall-runoff modeling. hydrological sciences journal, 48, 455-472. stravs, l., & brilly, m. (2007). development of a low flow forecasting model using the m5 machine learning method. hydrological sciences journal, 52, 466–477. tang, z., & fishwick, p. a. (1993). feedforward neural nets as models for time series forecasting. orsa journal of computing, 5(4), 374–385. timlin, d. j., pachepsky, y., acock, b. a., s'imunek, j., flerchinger, g. & whisler, f. (2002). error analysis of soil temperature simulations using measured and estimated hourly weather data with 2dsoil. agricultural systems, 72, 215–239. tuntiwaranuruk, u., thepa, s., tia, s., & bhumiratana, s. (2006). modeling of soil temperature and moisture with and without rice husks in an agriculture greenhouse. renewable energy, 31, 1934–1949. tyronese, j., katrina, m., mohamed, s., tommy, c. & peter, r. (2008). measuring soil temperature and moisture using wireless mems sensors. measurement, 41, 381–390. winegardner, d. l. (1996). an introduction to soils for environmental professionals. boca raton, fla.: lewis publishers. wong, f. s. (1991). time series forecasting using backpropagation neural networks. neurocomputing 2, 147–159. yang, c. c., prasher, s. o., mehuys, g. r. & patni, n. k. (1997). application of artificial neural networks for simulation of soil temperature. transactions of the asae, 40, 649-656. zadeh, l. a. (1965). fuzzy sets. information control, 8, 338–353. sensors. measurement, 41, 381–390. winegardner, d. l. (1996). an introduction to soils for environmental professionals. boca raton, fla.: lewis publishers. wong, f. s. (1991). time series forecasting using backpropagation neural networks. neurocomputing 2, 147–159. yang, c. c., prasher, s. o., mehuys, g. r. & patni, n. k. (1997). application of artificial neural networks for simulation of soil temperature. transactions of the asae, 40, 649-656. zadeh, l. a. (1965). fuzzy sets. information control, 8, 338–353. an artificial neural network was used for forecasting of long-term wind speed data (24 and 48 hours ahead) in la serena city (chile). in order to obtain a more effective correlation and prediction, a particle swarm algorithm was implemented to update the weights of the network. 43800 data points of wind speed were used (years 20032007), and the past values of wind speed, relative humidity, and air temperature were used as input parameters, considering that these meteorogical parameters are more readily available around the globe. several neural network architectures were studied, and the optimum architecture was determined by adding neurons in systematic form and evaluating the root mean square error (rmse) during the learning process. the results show that the meteorological variables used as input parameters, have influential effects on the good training and predicting capabilities of the chosen network, and that the hybrid neural network can forecast the hourly wind speed with acceptable accuracy, such as: rmse=0.81 [m·s−1], mse=0.65 [m·s−1]2 and r2=0.97 for 24-hours-ahead wind speed prediction, and rmse=0.78, mse=0.634 [m·s−1]2 and r2=0.97 for 48-hours-ahead wind speed prediction. abstract keywords: wind speed; time series forecasting; artificial neural network; particle swarm optimization; meteorological data. long-term prediction of wind speed in la serena city (chile) using hybrid neural network-particle swarm algorithm predicción a largo plazo de la velocidad de viento en la ciudad de la serena (chile) utilizando un algoritmo híbrido de red neuronal-enjambre de partículas issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v21n1.50337 juan a. lazzús* and ignacio salfate departamento de física y astronomía, universidad de la serena, casilla 554, la serena, chile *jlazzus@dfuls.cl una red neuronal artificial fue utilizada para la predicción de datos de la velocidad de viento a largo plazo (24 y 48 horas en adelanto) en la ciudad de la serena (chile). para obtener una efectiva correlación y predición, se implementó una optimización de enjambre de particulas para actualizar los pesos de la red. se emplearon 43800 datos de velocidad de viento (años 2003-2007), y los valores pasados de velocidad del viento, humedad relativa y temperatura del aire fueron utilizados como parámetros de entrada, considerando que estos parámetros meteorológicos se encuentran fácilmente disponibles en todo el mundo. se estudiaron varias arquitecturas de redes neuronales y la arquitectura optima se determine añadiendo neuronas de forma sistemática y evaluando la raíz del error cuadrático medio (rmse) durante el proceso de aprendizaje. los resultados muestran que las variables meteorológicas utilizadas como parámetros de entrada, tienen un efecto positivo sobre el correcto entrenamiento y capacidades predictivas de la red, y que la red neural híbrida puede pronosticar la velocidad del viento horaria con una precisión aceptable, como un rmse=0.81 [m·s−1], mse=0.65 [m·s−1]2 y r2=0.97 para la predicción de la velocidad del viento de 24 horas en adelanto, y un rmse=0.78, mse=0.634 [m·s−1]2 and r2=0.97 para la predicción de la velocidad del viento de 48 horas en adelanto. resumen palabras clave: velocidad del viento; predicción de series de tiempo; redes neuronales artificiales; optimización de enjambre de particulas; datos meteorológicos. record manuscript received: 30/04/2015 accepted for publication: 17/02/2017 how to cite item lazzus, j. a. & salfate, i. (2017). long-term prediction of wind speed in la serena city (chile) using hybrid neural network-particle swarm algorithm. earth sciences research journal, 21(1). 29-35. doi:http://dx.doi.org/10.15446/esrj.v21n1.50337 m e t e o r o l o g y earth sciences research journal earth sci. res. j. vol. 21, no. 1 (march, 2017): 29 35 30 juan a. lazzús and ignacio salfate 1. introduction energy derived from wind has played a vital role in the history of mankind and is again receiving considerable attention because of its free and non-polluting character. with the development of wind energy technologies and the decrease of wind power production cost, wind power has rapidly developed around the world in recent years (akdağ and guler, 2011). as a clean energy source wind is considered an alternative to fossil fuels, which actually accelerate global warming. the first scientific research to utilize wind for generating electricity, was initiated by the danish in the 1960s. the 1973 energetic crisis, forced many governments to realize the value of wind, as a renewable and independent energy source (hanağasioğlu, 1999). electricity generation using wind energy has been well recognized as environmentally friendly, socially beneficial, and economically competitive for many applications (monfared et al., 2009). prediction of wind speed (ws) at the surface or near the surface, is essential in many areas of science and technology, e.g., wind energy generation, aviation, space vehicle launching, weather forecasting, and agro-meteorology (kulkarni et al., 2008). wind field prediction at the level of wind farm, is still a challenging problem. different methods have been developed (kallos et al., 2007); and several studies have been performed to estimate the wind potential in different parts of the world (çam and yildiz, 2006). there are various strategies for wind speed prediction that can be classified into two categories: (1) statistical methods that can be subdivided into numerical weather prediction (nwp) and persistence and (2) artificial intelligence techniques that have subdivisions such as artificial neural networks (ann) and fuzzy logic (monfared et al., 2009). with the developments made in chaos theory, researchers have looked for the determinism in various seemingly chaotic-looking fluctuations from different disciplines such as physics, chemistry, hydrology, atmospheric sciences, etc (karunasinghe and liong, 2006). the time series prediction is one of the most important aspects in chaos theory. time series contain much information about dynamic systems (han and wang, 2009). these systems are usually modeled by delay-differential equations. some of them, for example, the mackey–glass equation (mackey and glass, 1977), the ikeda equation (ikeda, 1979), and equation for an electronic oscillator with delayed feedback (chua et al., 1992), are standard examples of time-delay systems (bezruchko et al., 2001). the main problem of the time series study consist of predicting the next value of a series known up to a specific time, using the known past values of the series. in time series prediction, this is usually first embedded in a state space using delay coordinates: where x(t) is the value of the time series at time t, τ a suitable timedelay and d the order of the embedding. this embedded vector is then used to predict the next value of the series x(t + τ). therefore, the non-linear dependence of the level of a series on previous data points is of interest, partly because of the possibility of producing a chaotic time series. note that short-term prediction for chaotic time series have been widely investigated by several techniques (karunasinghe and liong, 2006), however, the long-term prediction has not been widely studied in the literature. in this work, chaotic time series data taken from the mackey–glass differential equation were used to develop a neural network. in order to still obtain a more effective correlation and prediction, particle swarm algorithm has been introduced to update the weights of all layers of the network. next, this hybrid algorithm was used in the long-term prediction of the next 24 and 48 hours of the wind speed time series. to the best of the authors’ knowledge, there is no application for the prediction of the wind speed that includes the long-term prediction, such as the one presented here. 2. computational method a feed-forward neural network was used to represent non-linear relationships among variables. this ann was implemented by replacing standard back-propagation algorithm with particle swarm optimization (pso). pso is a population-based optimization tool, where the system is initialized with a population of random particles and the algorithm searches for optima by updating generations (eberhart and kennedy, 1995). in each iteration, the velocity of each particle j is calculated according to the following formula (lazzús, 2011): where s and v denote a particle position and its corresponding velocity in a search space, respectively. k is the current step number, ω is the inertia weight, c1 and c2 are the acceleration constants, and r1, r2 are elements from two random sequences in the range (0,1). kjs is the current position of the particle, ψ k jy is the best one of the solutions that this particle has reached, and ψg is the best solutions that all the particles have reached. in general, the value of each component in v can be clamped to the range [–vmax ,vmax] control excessive roaming of particles outside the search space (kennedy et al., 2001). after calculating the velocity, the new position of each particle is: the total steps to calculate the output values, using the input values of the network were as follows (lazzús et al., 2014): where xi is the input variables i, min ix and max ix are the smallest and largest value of the data, thus the input data are normalized using this equation. next, the net inputs (n) are calculated for the hidden neurons coming from the inputs neurons. for a hidden neuron: where pi is the vector of the inputs of the training, , h i jw is the weight of the connection among the input neurons with the hidden layer h, and the term , h i jb corresponds to the bias of the neuron of the hidden layer h, reached in its activation (freeman and skapura, 1991). the pso algorithm is very different from any of the traditional methods of training (lazzús, 2011). each neuron contains a position and velocity. the position corresponds to the weight of a neuron while the velocity is used to update the weight . starting from these inputs, the outputs (yi) of the hidden neurons are calculated, using a transfer function f h associated with the neurons of this layer (freeman and skapura, 1991). to minimize the error, the transfer function f should be differentiable. in the ann, the hyperbolic tangent function (tansig) was used as all the neurons of the ann have an associated activation value for a given input pattern; the algorithm continues finding the error that is presented for each neuron, except those of the input layer. after finding the output values, 1 1k+ k k+ j j js = s + v 31long-term prediction of wind speed in la serena city (chile) using hybrid neural network-particle swarm algorithm the weights of all layers of the network are actualized by pso, using equations 2 and 3 (lazzús et al., 2014). the velocity is used to control how much the position is updated. on each step, pso compares each weight using the data set. the network with the highest fit ness is considered the global best. the other weights are updated based on the global best network rather than on their personal error or fitness (pérez ponce et al., 2012; lazzús et al., 2014). in this article, we used the mean square error (mse) to determine network fitness for the entire training set: where yi is the output value obtained from the normalized output (yi) of the network. this process was repeated for the total number of patterns to training. for a successful process the objective of the algorithm is to modernize all the weights by minimizing the total root mean squared error (rmse): figure 1 presents a block diagram of the ann+pso algorithm developed in this study. in pso, the inertial weight ω, the constant c1 and c2, the number of particles npart and the maximum speed of particle summarizes the parameters to synchronize for their application in a given problem. an exhaustive trial-anderror procedure was applied for tuning the pso parameters. table 1 shows the selected parameters for this hybrid algorithm. 3. simulations 3.1. mackey-glass time series to evaluate the capability of the proposed hybrid algorithm in the long-term prediction, the mackey–glass time series was used. thus, a set of data points were generated from the mackey–glass time-delay differential equation (mackey and glass, 1977; farmer, 1982) which is defined by: (11) where t is a variable, x is a function of t, and τ is the time delay. the initial values of the time series are α = 0.2, β = 0.1, and x(0)=1.2. if τ ≥ 17, the time series show the chaotic behaviour (farmer, 1982; mirzaee, 2009). the goal of the task is to use known values of the time series up to the point x=t to predict the value at some point in the future x=t+t. the standard method for this type of prediction is to create a mapping from d points of the time series spaced apart, that is , to a predicted future value x(t+t). in order to solve the mackey–glass equation, the fourth-order runge–kutta method was applied to find the numerical solution. the time series was obtained evaluating the solution of eq. (11) at each integer points. step size of 0.1 was used to generate a time series, and x(t) is thus derived for with x(t)=0 for t <0 in the integration. four non consecutive points in the time series are given to generate each input vector xi (where i=1, 2, …, n) of the input matrix x, as: (12) a similar criterion was used to create the output matrix, as: (13) figure 1. flow diagram for training of the ann using pso algorithm table 1. parameters used in the hybrid ann+pso algorithm. 32 juan a. lazzús and ignacio salfate then, the ann+pso method was used in the long-term prediction. this hybrid algorithm was trained to predict the future value x(t+84) from the current value x(t) and the past values, using the standard form applied in the literature, for d=4 and ∆=t=6 (chng et al., 1996; mirzaee, 2009). (14) one thousand data points of the above format were collected. the first 500 were used for training while the others were used for testing the ann+pso method. then, the following case was simulated with α = 0.2, β = 0.1, x(0) = 1.2 and τ = 17. from this case, several network architectures were tested. the most basic architecture normally used for the analysis of chaotic time series involves a neural network consisting of three or four layers (lazzús et al., 2014). the input layer contains one neuron for each input parameter: x(t), x(t–6), x(t–12), and x(t–16). the output layer has one node generating the scaled estimated value of the chaotic time series x(t+84). the number of hidden neurons needs to be sufficient to ensure that the information contained in the data utilized for training the network is adequately represented (lazzús et al., 2014). there is no specific approach to determine the number of neurons of the hidden layer, many alternative combinations are possible. the optimum number of neurons was determined by adding neurons in systematic form and evaluating the mse and rmse of the sets during the learning process (pérez ponce et al., 2012; lazzús et al., 2014). for our case the optimum architecture was 4-12-1. the results obtained with the ann+pso method present a mse=0.000063 and rmse=0.0079 for training set, and mse=0.000065 and rmse=0.0080 for prediction set. these results show that the ann+pso model can be accurately trained and that the chosen architectures can predict the long-term x(t+84) with acceptable accuracy. table 2 shows a comparison between some computational methods found in the literature (martinetz et al., 1993; whitehead and choate, 1996; bersini et al., 1997; awad et al., 2009) and the result obtained with the ann+pso method. this comparison was made using the normalized root mean squared error (nrmse), defined as: (15) table 2. comparison between computational methods found in the literature for the long-term prediction. 3.2. wind speed time series once the capability of the hybrid algorithm was proved, it was used to forecast the long-term of wind speed time series. this study is based in data collected from a meteorological stations located on the semi-arid norte chico of chile (29º54’ s; 71º15’ w; 10 m), located at south of the hyper-arid atacama desert. the region is characterized by complex topography with altitudes varying from sea level until 5000 m at the high andes cordillera. the climatic characteristics are influenced by the south-eastern pacific subtropical anticyclone and the cold humbolt current which results in low precipitation rates (kalthoff et al., 2002). the zone is one of the most sensitive areas in south america (kalthoff et al., 2006), and recent studies of oceanic and atmospheric variability have confirmed the implications of the dynamics of el niño-southern oscillation (enso) cycle on the local climate of this zone (meinen and mcphaden, 2009). 43800 data points (years 2003-2007) were used. the future values of wind speed was predicted using the past values of the time series of wind speed ws(m·s−1), relative humidity rh(%), and air temperature t(k). figure 2 shows the time series of the selected meteorological data used. the data ranges and the properties of interest are listed in table 3. as seen in this table, hourly ws cover wide ranges, going from ≈0 to 10 (m·s−1). other wide range for the input parameters are: t from 270 to 305 (k), and rh from 2 to ≈100 (%). the influence of the selected meteorological data in this study (t, rh, and ws) over the climate of the semi-arid zone of the atacama desert has been revised and evaluated in other communications (kalthoff et al., 2002; kalthoff et al., 2006). figure 3 shows ws as a function of the selected meteorological data: t and rh. fig. 3a shows ws as a function of t with a coefficient of linear correlation (r2) of 0.5979. fig. 3b shows ws as a function of rh with r2 of 0.2757. note that the coefficient of linear correlation in this figure shows a non-linear relationship between table 3 summary of data used in this study. figure 2. time series of the meteorological data used in this study. (a) wind speed ws/ms-1, (b) air temperature t/k, and (c) relative humidity rh/%. 33long-term prediction of wind speed in la serena city (chile) using hybrid neural network-particle swarm algorithm ws and the input parameters for this climate zone. then, the relationship between ws and these meteorological data is highly non-linear, and consequently an ann is the best alternative to model the hourly ws. two new cases were studied with this methodology, the long-term prediction of the next 24 hours ws(t+24) and the long-term prediction of the next 48 hour ws(t+48). to select the best input parameters for solving the problem, the hourly data from the current value to 23 past hours (t-23, t-22, t-21,… , t), were considered. then, the sum of absolute values of weights (savw) was used (lazzús, 2013). figure 4 shows the input with more contribution for the prediction of future values of ws. this figure shows the great significance of the past values (t-18), (t-12), (t-6), and the current value (t) on the three meteorological data (ws, t, and rh). figure 3. experimental data of ws from 2003 to 2007 as a function of the selected meteorological parameters. a) ws vs t →r2=0.5979; (b) ws vs rh →r2=0.2757. thus, the optimum input vector was: (16) where τ is the future value to predict: 24 and 48 for the long-term prediction. figure 4. influence of several points of time series on prediction of future values of wind speed. bars represent the sumsof absolute values of weights (savw) of the ann+pso. the leave-20%-out cross-validation method was used to estimate the predictive capabilities of the model. 34996 data points were used in the training set, and 8760 data points (not used in the training step) were used in the prediction set. 4. results and discussion several network architectures were tested for the long-term wind speed prediction (t(t+24) and t(t+48), separately). the optimum architecture was checked using the objective function (eq. 10). figure 5 shows mse found in correlating the ws as function of the number of neurons in the hidden layer (nn). fig. 5a shows the best topology found for the prediction of the ws for the next 24 hours t(t+24) with a network architecture 12-28-1. fig. 5b shows the prediction of the ws for the next 48 hours t(t+48) with an optimum network architecture of 12-42-1. once the best architectures were determined, the optimum weights and biases required to carry out the estimate of future values of ws were obtained. figure 6 shows a comparison between real data (black line) and calculated values (points) of long-term prediction of ws(t+24). fig. 6a shows the forecasting of ws(t+24) with a correlation coefficient (r2) of 0.974 for the training set (years 2003-2006) with a slope of the curve (m) of 0.965 (expected to be 1.0), and with rmse=0.787 [m·s−1] (mse=0.620 [m·s−1]2 and msemax=10.028 [m·s-1]2). fig. 6b shows the forecasting of ws(t+24) with r2=0.973 during the prediction step (year 2007) with m=0.963 (also expected to be 1.0), and with rmse=0.807 [m·s−1] (mse=0.651 [m·s−1]2 and msemax=7.110 [m·s−1]2). figure 5. deviations found in the correlation of wind speed as a function of the number of neurons in the hidden layer for: (a) long-term prediction ws(t+24), and (b) long-term prediction ws(t+48). in both graphics, training step (■) and prediction step (○). 34 juan a. lazzús and ignacio salfate figure 6. comparison between real and calculated values of ws(t+24) using the proposed ann+pso model. (a) training set, period 2003-2006; and (b) prediction set, period 2007. figure 7 shows a comparison between real data (black line) and calculated values (points) of long-term prediction of ws(t+48). fig. 6a shows the forecasting of ws(t+48) with r2=0.970 for the training set with m=0.963, and with rmse=0.788 [m·s−1] (mse=0.622 [m·s−1]2 and msemax=9.386 [m·s−1]2). fig. 6b shows the forecasting of ws(t+48) with r2=0.969 for the prediction set with m=0.962, and with rmse=0.796 (mse=0.634 [m·s−1]2 and msemax=8.210 [m·s−1]2). a comparison was made with a neural network with standard back-propagation (bpnn) algorithm (hagan and menhaj, 1994), and similar architecture and database. this bpnn show results of mse higher than 1 [m·s−1]2 and r2 lower 0.8 for the forecasting of ws(t+24) and ws(t+48). and other comparison was made with a multiple linear regression (mlr) method, and similar database. the mlr method shows mse higher than 4 [m·s−1]2 and r2 lower 0.7 for both cases. note that, the predictions with the proposed ann+ pso method shows mse a little higher than 0.6 [m·s−1]2 and r2 higher than 0.97. table 4 summarizes the deviations obtained in the long-term prediction using the proposed method versus bpnn and mlr methods. these results show that the ann+pso can be accurately trained and that the chosen topologies can estimate the future values of ws with acceptable accuracy. these results represent a tremendous increase in accuracy for forecasting this important meteorological property and show that not only the optimum architecture obtained was crucial, also the appropriate selection of the independent parameters (t and rh). this is important because air temperature and relative humidity are commonly available parameters. note that the coefficients of linear correlation of these parameters show a non-linear relationship with the ws for the climate of several geographic zones. then, the relationship between ws and these meteorological data is highly non-linear, and consequently the ann+pso is a good tool for modeling ws for several applications. figure 6. comparison between real and calculated values of ws(t+48) using the proposed ann+pso model. (a) training set, period 2003-2006; and (b) prediction set, period 2007. the results obtained by other models and other sites can be usefully compared with our results. zhang et al. (2012) shows the performance analysis of four modified approaches for wind speed forecasting for four observation sites in gansu (china), with mse higher than 2 [m·s−1]2. liu et al. (2013) shows the forecasting models for wind speed using wavelet, wavelet packet, time series and artificial neural networks with mse higher than 1 [m·s−1]2. for spain, the wind speed estimation was made using a multilayer perceptron with mse higher than 1 [m·s−1]2 and r2 below 0.75 (velo et al., 2014). recently, wang et al. (2014) shows the mean hourly wind speed prediction in the hexi corridor of china based on the seasonal adjustment method (sam), exponential smoothing method (esm), and radial basis function neural network (rbfn), with rsme higher than 0.7 [m·s−1]. it must be mentioned that these results were obtained from different sites and based on different weather conditions, and the results cannot be compared directly with one another. however, results from the different methods show that the accuracy of ann+pso model employed in this study is good. 5. conclusions in this work, a neural network was used for the forecasting of longterm wind speed time series. in order to obtain a more effective correlation and prediction, particle swarm algorithm has been introduced to update the weights of all layers of the network. 43800 data points (years 2003table 4. summary of the deviations obtained with the ann+pso algorithm for the long-term prediction of wind speed.. 35long-term prediction of wind speed in la serena city (chile) using hybrid neural network-particle swarm algorithm 2007) of wind speed were used. to distinguish between the different values of hourly data considered in this study, so that the network can discriminate and learn in optimum form, the following past values of the time series of meteorological data were used as input parameters: wind speed ws(m·s-1), relative humidity rh(%), and air temperature t(k). based on the results and discussion presented in this study, the following main conclusions are obtained: i) the results show that the proposed ann+pso can be properly trained for predicting the hourly wind speed, with acceptable accuracy; ii) the meteorological variables used (ws, t, and rh), have influential effects, on the good training and predicting capabilities, of the chosen network; iii) the low deviations found with the proposed ann+pso method indicate that it can predict the future values of ws with better accuracy than other methods; and iv) the values obtained with the proposed method are believed to be sufficiently accurate for engineering calculations, among other uses. acknowledgements the authors thank the direction of research and development of the university of la serena (diduls) through the research project peq16141, and the department of physics of the university of la serena (dfuls) for the special support that made possible the preparation of this paper. special acknowledgements to dr. pedro vega (uls) for the support that made possible the implementation of the meteorological station and to mg. julio marín (uls) for providing the meteorological data set. references akdağ, s.a. and güler ö. (2011). a comparison of wind turbine power curve models. energy sources part a, 33, 2257–2263. awad, m., pomares, h., rojas, i., salameh, o. and hamdon, m. (2009). prediction of time series using rbf neural networks: a new approach of clustering. international arab journal of information technology, 6, 138–143. bersini, h., duchateau, a. and bradshaw, n. (1997). using incremental learning algorithms in the search for minimal effective fuzzy models. in: proceedings of 6th international conference on fuzzy systems, pp. 1417–1422. bezruchko, b.p., karavaev, a.s., ponomarenko, v.i. and prokhorov, m.d. (2001). reconstruction of time-delay systems from chaotic time series. physical review e, 64, 056216. çam, e. and yildiz, o. (2006). prediction of wind speed and power in the central anatolian region of turkey by adaptative neuro-fuzzy inference systems (anfis). turkish journal of engineering and environmental sciences, 30, 35–41. chng, e.s., chen, s. and mulgrew, b. (1996). gradient radial basis function networks for nonlinear and nonstationary time series prediction. ieee transactions on neural networks, 7, 190–194. chua, l.o., kocarev, l., eckert, k. and itoh, m. (1992). experimental chaos synchronization in chua’s circuit. international journal of bifurcation and chaos, 2, 705–708. eberhart, r.c. and kennedy, j. (1995). a new optimizer using particle swarm theory. in: proceedings of 6th international symposium on micro machine and human science, nagoya. new york, ny, usa: ieee, pp. 39–43. farmer, j.d. (1982). chaotic attractors of an infinite-dimensional dynamical system. physica d, 4, 366–393. freeman, j.a. and skapura, d.m. (1991). neural networks: algorithms, applications and programming techniques. computation and neural systems series. massachusetts, usa: addison-wesley. hagan, m.t. and menhaj, m.b. (1994). training feedforward networks with the marquardt algorithm, ieee transactions on neural networks, 5, 989–993. han, m. and wang, y. (2009). analysis and modeling of multivariate chaotic time series based on neural network. expert systems with applications, 36, 1280–1290. hanağasioğlu, m. (1999). wind energy in turkey. renewable energy, 16, 822–827. ikeda, k. (1979). multiple-valued stationary state and its instability of the transmitted light by a ring cavity system. optics communications, 30, 257–261. kallos, g., galanis, g. and katsafados, p. (2007). local wind speed forecasting and applications to power prediction. geophysical research, 9, 93–99. kalthoff, n., bischoff-gauß, i., fiebig-wittmaack, m., fiedler, f., thürauf, j., novoa, e., pizarro, c., castillo, r., gallardo, l., rondanelli, r. and kohler, m. (2002). mesoscale wind regimes in chile at 30º s. journal of applied meteorology, 41, 953–970. kalthoff, n., fiebig-wittmaack, m., meißner, c., kohler, m., uriarte, m. and bischoff-gauß, i. (2006). the energy balance, evapotranspiration and nocturnal dew deposition of an arid valley in the andes. journal of arid environments, 65, 420–443. karunasinghe, d.s.k. and liong, s.y. (2006). chaotic time series prediction with a global model: artificial neural network. journal of hydrology, 323, 92–105. kennedy, j., eberhart, r.c. and shi, y. (2001). swarm intelligence. san diego, ca, usa: academic press. kulkarni, m.a., patil, s., rama, g.v. and sen, p.n. (2008). wind speed prediction using statistical regression and neural nerwork. journal of earth system science, 117, 457–463. lazzús, j.a. (2011). predicting natural and chaotic time series with a swarmoptimized neural network. chinese physics letters, 28, 110504. lazzús, j.a. (2013). neural network-particle swarm modeling to predict thermal properties. mathematical and computer modelling, 57, 2408–2418. lazzús, j.a., salfate, i. and montecinos, s. (2014). hybrid neural network–particle swarm algorithm to describe chaotic time series. neural network world, 24, 601–617. liu, h., tian, h., pan, d. and li, y. (2013). forecasting models for wind speed using wavelet, wavelet packet, time series and artificial neural networks. applied energy, 107, 191–208. mackey, m.c. and glass, l. (1977). oscillation and chaos in physiological control systems. science, 197, 287–289. martinetz, t.m., berkovich, s.g. and schulten, k.j. (1993). neural-gas network for vector quantization and its application to time-series prediction. ieee transactions on neural networks, 4, 558–569. meinen, c.s. and mcphaden, m.j. (2000). observations of warm water volume changes in the equatorial pacific and their relationship to el niño and la niña. journal of climate, 13, 3551–3559. mirzaee, h. (2009). linear combination rule in genetic algorithm for optimization of finite impulse response neural network to predict natural chaotic time series. chaos, solitons & fractals, 41, 2681–2689. monfared, m., rastegar, h. and madadi-kojabadi, h. (2009). a new strategy for wind speed forecasting using artificial intelligent methods. renewable energy, 34, 845–848. pérez ponce, a.a., lazzús, j.a. and palma-chilla, l. hybrid neural network– particle swarm method to predict global radiation over the norte chico (chile). journal of renewable and sustainable energy, 4, 023108. velo, r., lópez, p. and maseda, f. (2014). wind speed estimation using multilayer perceptron. energy conversion and management, 81, 1–9. wang, j., zhang, w., wang, j., han, t. and kong, l. a novel hybrid approach for wind speed prediction. information sciences, 273, 304–318. whitehead, b.a. and choate, t.d. (1996). cooperative-competitive genetic evolution of radial basis function centers and widths for time series prediction. ieee transactions on neural networks, 7, 869–880. zhang, w., wu, j., wang, j., zhao, w. and shen, j. (2012). performance analysis of four modified approaches for wind speed forecasting. applied energy, 99, 324–333. letter from the editor our first issue of 2018 is available now. we have 9 articles in this issue, in topics ranging from soil characterization, petrophysical properties of rocks, signal processing of images and gnss station to asteroid impact effects. mechanical soil properties in sandy-pebble soil are studied as a function of grain size content in lu et al. satellite-derived soil moisture estimation is discussed in thanabalan and vidhya, based on a semi-empirical approach and backscattering images. macroscopic mechanical characteristics of rocks depend on a number of factors, including microstructure damage. under changing temperature conditions (freeze-thaw) rock samples studied by jiang show significant strength decrease, which has important consequences in engineering. advanced signal processing methods are used in zeng et al. for image retrieval applied to remote sensing data, using a bayesian network approach. similarly, oktar and erdogan use linear trend and wavelet analysis to continuous gnss data showing both displacements due to tectonic as well as atmospheric and hydrologic effects. debris flows can in some cases become serious hazards because they can block river flow as a debris flow dam. chen et al. propose a method to identify the formation of such dams, with an example from the er river in taiwan. mamaseni et al. study the petrophysical properties of three formations in the duhok basin, northern iran, based on well-log data. results suggest a significant thickness with good moveable hydrocarbons in the study area. methane adsorption and gas content are strongly influenced by shale composition. zhu et al. show that total organic content has a stronger influence on methane adsorption and gas content than the mineral composition, studying samples from the southern sichuan basin. our last contribution in this issue discussed a large asteroid impact in eastern colombia. the impact would have affected the environment and landscape, but the degree of the effect depends on a number of parameters including size of the impactor, angle of impact, type of rocks in the crust, and much more. a model is proposed by hernandez and alexander, that can predict both ejecta deposit distance as well as seismic shaking, and of course thermal radiation. results of this study provide additional tools for searching for evidences of direct impact craters. germán prieto editor in chief l e t t e r f r o m t h e e d it o r earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018) enero%2031-geociencias-vol%2012-2%20dic-en%20baja[1].pdf earth sciences research journal earth sci. res. j. vol. 12, no. 2 (december 2008): 194-212 intraseasonal variability of rainfall over northern south america and caribbean region j.d. pabón1 and j. dorado2 1 oficina 317, edificio 212 (aulas de ciencias humanas). department of geography, national university of colombia. e-mail address: jdpabonc@unal.edu.co 2 posgraduate program in meteorology, department of geosciences, national university of colombia abstract basedondecadal (amounts foreach tendays)precipitationdata from meteorologicalstationssituated innorthernsouthamericaandcaribbeanregion,adecadalprecipitationindex(dpi)wascalculatedinorder tostudythe intraseasonal variability (isv) of regional rainfall. the spectral analysis of dpi allows to identify signals with 20-25,30,40and50-60daysperiod.according to theanalysisof their spatialdistribution thesesignalsarewell definedover thecaribbean islandandcoastal sector such as in somesectors of theandean region; the60-days signal is presented onlyover caribbean region and in someplaces in thepacific sector; in the eastern lowlands of orinoco and amazon basin these signals are not clearly expressed. exploring the relationshipbetween regional isvandmadden-julianoscillationcorrelation analysiswasmade. due to the presence of signals different of 30-60 days, the correlation coefficients were very low. considering this situation,high frequencysmoothingwasapplied todpi timeseries; after that, a relativecorrelationwasdetected between smoothed dpi and madden-julian index (mji). keywords: intraseasonal variability, madden-julian oscillation, rainfall. resumen con base en datos de precipitación decadal (acumulados de diez días) provenientes de estaciones meteorológicas localizadas en el norte de suramérica y en el caribe, se calculó un índice de precipitación decadal (ipd) para estudiar la variabilidad intraestacional (vis) de la precipitación de ésta región. el análisis espectral del ipd muestra señales con períodos de 20-25, 30, 40 y 50-60 días. de acuerdo con el análisis de la 194 manuscript received: october 10th, 2008. accepted for publication: november 22th, 2008. distribuciónespacial, estas señalesestánbiendefinidas sobre las islasdelcaribey lazonacostera, así comoen algunos sectores de la región andina; la señal de 60 días se presenta únicamente en el caribe y sobre algunos lugares del pacífico; en las tierras bajas de las cuencas del orinoco y amazonas las señales no se expresan claramente. seanaliza tambiénlarelaciónentre lavisregionalde laprecipitaciónylaoscilacióndemadden-julian.debido a la presencia de señales diferentes a las de 30-60 días en la precipitación, los coeficientes de correlación obtenidos son muy bajos. por esto, se realizó la suavización de las altas frecuencias en las series del ipd y se calcularon nuevamente los coeficientes de correlación del ipdcon el índice madden-julian, después de lo cual hubo un notorio aumento de los coeficientes de correlación. palabras clave: variabilidad intrestacional, oscilación maden-julian, lluvia. 1. introduction extremephases of climatevariabilitybring todifferent regionswarmorcoldperiods, rainy(moreprecipitation than normal or more frequent heavy rainfall events)ordryconditions, andso.thisvariability impacts in several ways ecosystems and economic systemsof the countries around the world, producing in somecasesdisasters. in theclimatesystemmanyprocesses generate this variability. for example, the tropical pacific phenomena el niño (warm condition) and la niña (cold conditions) are the cause of 2-7years timescaleoscillationsof climaticvariables known as enso cycle (philander, 1990; hastenrath, 1996; see also enso bibliography in coaps, 2006). in addition to the enso cycle, signals such as quasi-biennial component (ropelewski et al., 1992; meehl, 1997; baldwin et al., 2001), and fluctuations in the period interval of 20-90 days called intraseasonal oscillations (knutson & weickman, 1987; bantzer & wallace, 1996; nogués-paegle et al., 2000;krishnamurti&shukla,2000;goswami& mohan, 2001; bond & vecchi, 2003; krishnamurti & shukla, 2007) have been identified. today the most studied signal of climate variability is that causedbyenso.therearemanyworks relatedto theeffectsofenso inmonthlyprecipitation of different regions in the world (ropelewski et al., 1986; ropelewski & halpert, 1987; pabón & montealegre, 1992; peel et al., 2002; poveda, 2004; and many others). currently, seasonal climate prediction schemes are based on the knowledge about particularities of enso cycle in a given region, however, because theydonot includeothermodesof climate variability, prediction fails frequently, especially in month-to-month range and less (see for examplehendon et al., 2000; jones& schemm,2000; jones et al., 2004c). asourceof fails inprediction in month-to-month range is associated to the no inclusion of intraseasonal variations in the schemes. in fact, the phases of intraseasonal fluctuations activate and deactivate rainfall for periods of a couple of weeks lasting or forwarding the beginning or end of rainy season, or breaking it. the rainy phase of intraseasonalvariabilityalsoactivatesheavyprecipitation events and related to them disasters (flashfloods, landslides, etc). due to practical value to improve subseasonal predictability (waliser et al. 2003; webster & hoyos, 2004), the interest on intraseasonal modes of climate variability has been increasing in last decade and manyefforts have been doing to study this variability especially the associated tomadden-julianoscillation(madden&julian, 1994), the dominant mode in intraseasonal climate variability. several authors have been studied the intraseasonal variability (hereafter isv) in precipitation for different geographical regions of the world. krishnamurti & shukla, (2000, 2007), for example, found modes with 45 and 20 days period in precipitation in india. wang et al. (1996) explored isv of precipitation in china finding 12, 21 and 43daysperiod.analysiswasmadealso forafintraseasonal variability of rainfall over northern south america and caribbean region 195 rica (janicot & sultan, 2001; mathews, 2004) and signals over 10-25 and 25-60 days period were found inconvectionandprecipitation in thewestern region (sultan et al., 2003; mounier & janicot, 2004); satatistically significant spectral peaks over 15and40daysperiodwere found forsahel precipitation (janicot & sultan 2001). jones et al. (2004a) using outgoing long wave radiation data developed a climatology for tropical intraseasonal convective anomalies. also, ye & cho (2001), analyzed precipitation data for united states, and found 24 and 37 days signals. isvof convection and precipitation for different regions of south america has been studied by garreaud (2000), petersen et al. (2002), misra (2005). exploring the causes of isv of precipitation many researchers have been paying special attention to its relationship to madden-julian oscillation (mjo),because themjo is thedominantmodeof tropical isv. thus, bantzer & wallace (1996) analyzed temperatureandprecipitationdatausingsatellitedata and found a40-50 days component, close to mjo period. liebman et al. (1994) investigated the relationship between tropical cyclones of the indian and western pacific oceans and the mjo and found that cyclones preferentially occur during the convective phaseof theoscillation;but theynoted,however, that the increase in cyclone activityduring activeperiods of convection is not restricted to mjo activity and concluded that the lastdoesnot influence tropicalcyclones in a unique fashion (this situation maybe due to the existence of other modes of isv). a similar analysiswasdonebymaloney&hartman (2000a,b) for hurricanes of eastern north pacific and gulf of mexico (informationon caribbean is also included). kayano & kousky (1999) studied the mjo in the global tropicsusingpentad-means for the1979-1995 period computed for 200and 850-hpa zonal winds, 200-hpa velocity potential, 500-hpa geopotential height and pressure vertical velocity, 925-hpa temperature and specific humidity, slp and total precipitablewater (pw); theyfound inallvariablesan eastward traveling large-scale oscillatory regime with a period of approximately45 days. in the other hand, jones et al. (2004b) using pentadal precipitation data based on global precipitation climatology project (gpcp) confirmed thatover indianocean, indonesia, western pacific, eastern south america, western north america, northeast africa, the middle east, and eastern china, extremes precipitation events increaseswith thepresenceof active (convective) phase of mjo. barlow et al. (2005) analyzing dailyprecipitation forsouthwestasia found that this variable is modulated by mjo activity in the eastern indian ocean, with strength comparable to the interannual variability. bond & vechi (2003) found a relationship between mjo and precipitationof oregon and washington states. isvwas detected in convectiveprocessesoveramazonregionbypetersen et al. (2002). theclimatevariability fornorthernsouthamerica and caribbean region has been studied mainly in interannual scale (hastenrath, 1976; pabón & montealegre, 1992; enfield, 1996; alfaro et al., 1998; enfield & alfaro, 1999; montealegre & 196 j.d. pabón and j. dorado figure 1. northern south america and caribbean region anddistributionofmeteorological stationsusedforanalysis (the number corresponds to station listed in table 1). intraseasonal variability of rainfall over northern south america and caribbean region 197 table 1. list of meteorological stations used in the study. id station name latitud longitude altitude (m.s.n.m.) country 1 montego bay 18,50 –77,91 1 jamaica 2 3 las américas aeropuerto san juan intl 18,43 18,41 –69,66 –66,00 18 19 dominican rep. puerto rico 4 5 barahona kingston 18,20 17,93 –71,10 –76,78 3 3 dominican rep. jamaica 6 aeropuerto el embrujo 13,37 –81,35 1 colombia 7 aeropuerto sesquicentenario 12,58 –81,72 1 colombia 8 aeropuerto queen beatrix 12,50 –70,01 18 aruba 9 10 aeropuerto hato aeropuerto flamingo 12,20 12,15 –68,96 –68,28 9 6 netherl antills netherl antills 11 12 manaure coro 11,77 11,41 –72,45 –69,66 1 16 colombia venezuela 13 14 matitas esc agr carraipia 11,25 11,22 –73,05 –72,37 20 118 colombia colombia 15 apto. simón bolívar 11,13 –74,23 4 colombia 16 san lorenzo 11,07 –74,03 4 colombia 17 aeropuerto ernesto cortissoz 10,88 –75,30 14 colombia 18 19 maiquetía apto. s. bolívar aeropuerto rafael núñez 10,60 10,45 –66,98 –75,52 43 2 venezuela colombia 20 21 22 23 maturín san bernardo d. aeropuerto rafael barvo lorica 9,75 9,37 9,33 9,27 –63,18 –75,97 –75,26 –75,82 65 22 166 30 venezuela colombia colombia colombia 24 aeropuerto las flores 9,07 –73,98 34 colombia 25 tocumen 9,05 –79,36 45 panamá 26 guanare 9,01 –69,73 163 venezuela 27 28 marcos a. gelabert el salado 8,98 8,92 –79,55 –75,58 13 40 panamá colombia 29 30 31 32 david aeropuerto camilo daza san fernando de apore aeropuerto arauca 8,40 7,93 7,90 7,06 –82,41 –72,52 –67,41 –71,22 29 250 47 128 panamá colombia venezuela colombia 33 aeropuerto otu 7,03 –75,20 630 colombia 34 aeropuerto yariguíes 7,02 –73,80 126 colombia 198 j.d. pabón and j. dorado id station name latitud longitude altitude (m.s.n.m.) country 35 el cucharo 6,87 –73,37 975 colombia 36 tulio ospina 6,53 –75,92 1438 colombia 37 aeropuerto olaya herrera 6,37 –75,97 1490 colombia 38 panamericana 6,22 –77,40 4 colombia 39 aeropuerto puerto carreño 6,18 –67,48 50 colombia 40 41 tuparros boca torno surbata bonza 6,05 5,82 –67,52 –73,07 250 2485 colombia colombia 42 43 u.p.t.c. isla del santuario 5,57 5,47 –73,22 –73,73 2690 2580 colombia colombia 44 aeropuerto matecaña 5,37 –77,08 53 colombia 45 aeropuerto el carano 5,12 –77,08 53 colombia 46 la camelia 5,08 –76,45 1670 colombia 47 48 aeropuerto la nubia aeropuerto el edén 5,03 4,78 –75,47 –76,28 20,80 1204 colombia colombia 49 50 tibaitatá las gaviotas 4,70 4,55 –74,22 –70,93 2543 171 colombia colombia 51 52 centro ad. la unión el salto 4,53 4,47 –76,05 –74,77 920 450 colombia colombia 53 aeropuerto santiago villa 4,28 –75,33 266 colombia 54 aeropuerto el dorado 4,12 –74,15 2547 colombia 55 aeropuerto farfán 4,02 –76,37 955 colombia 56 57 palmira ica aeropuerto buenaventura 3,87 3,85 –76,53 –76,97 975 14 colombia colombia 58 59 60 61 aeropuerto benito salas aeropuerto alfonso bonilla vista hermosa bonanza 3,62 3,55 3,03 2,95 –75,50 –76,38 –73,73 –77,88 439 961 325 10 colombia colombia colombia colombia 62 aeropuerto g. l. valencia 2,43 –76,58 1730 colombia 63 la sierra 2,37 –77,37 1870 colombia 64 la macarena 2,30 –73,13 350 colombia 65 66 paleterá aeropuerto antonio nariño 2,08 1,67 –76,05 –77,47 2900 1796 colombia colombia 67 68 69 70 aeropuerto san luis aeropuerto artunduaga obonuco valparaiso 1,42 1,36 1,30 1,15 –78,13 –75,32 –77,05 –75,42 2961 244 2710 270 colombia colombia colombia colombia pabon, 2000; giannini et al., 2000; chang & stephenson, 2000; chang & taylor, 2002; taylor et al., 2002; poveda, 2004, nobre et al., 2006), especially the associated to enso, with purpose to improve seasonal-interannual climate prediction. the isv of precipitation have been less studied, however someattemptshavebeencarriedout by povedaetal. (2002), whoanalyzed thedaily cycleof precipitation ofcolombianandesandfoundasignificant relationshipbetweenmjoanddailyprecipitation, suchasbetween mjo and amplitude of daily cycle. pabón (2007), explored isv using decadal (ten days amounts) precipitation data for different regions of colombiaandfound20-25and50-70daysperiodical components; searching the relationship of the analyzed timesserieswithmjo itwas founda lowcorrespondencebecause thepresenceofothermodeof isv different to mjo mode. these works show the evidences of isv modes in climate of the region. taking in account the state of knowledge about the isv and the regional importance for improving climate prediction and to strength the disaster prevention systems, especially in the component related toheavyrainfall, thispaper try toanalyze inmoredetail the characteristics of isv of precipitation in northern south america and caribbean region. 2. data and methodology for the analysis in the current study as basic data was used daily precipitation for the 1978-2004 period from meteorological stations distributed over northern south america and caribbean region as showed infigure1 (the75are listed intable1).selection of the meteorological stations was done considering criteria as representativeness of agiven region, length of record period at least 20 years and minimal gaps in data series. considering the complex topography over colombian territory that generates a rich climate diversity, it was necessary to include a relatively high density of the network for this region. an initial check was carried out in order to test thequalityofdata.after thisqualitycontrol adecadal (amounts for each ten days period) precipitation series were organized and an index (dpi) was calculated using the equation: dpi p p p = s (1) (or simply, the z-score of decadal precipitation) where:p–decadalprecipitation; multianualprecipitationaverage for respectivedecade; standarddeviation for the series of agiven decade (timesequences of firstdecades,or seconddecadesof theyear andso on). to facilitate the presentation of analysis and results the report was organized for five sectors of the region: islands in caribbean sea region, continental plain lowland of the caribbean coastal zone, mountainous (andean) region, eastern lowlands of orinoco-amazon basin, and pacific sector. to identify signals of isv of precipitation spectral analysis (wilks, 1995) was applied to timeseries intraseasonal variability of rainfall over northern south america and caribbean region 199 id station name latitud longitude altitude (m.s.n.m.) country 71 villagarzón 1,05 –76,60 440 colombia 72 aeropuerto mariscal sucre –0,13 –78,48 28,11 ecuador 73 manta –0,95 –80,68 13 ecuador 74 aeropuerto gral. m. serrano –3,25 –69,93 84 colombia 75 aeropuerto vásquez cobo –4,16 –69,93 84 colombia 200 j.d. pabón and j. dorado estación san juan (puerto rico) -2 -1 0 1 2 years (in decadal resolution) estación manaure (colombia) -2 -1 0 1 2 d estación apto la nubia (colombia) -2 -1 0 1 2 estación maturin (venezuela) -2 -1 0 1 2 2 d estación apto buenaventura (colombia) -2 -1 0 1 2 d 2 0 0 4 1 9 9 6 1 9 9 7 1 9 9 8 1 9 9 9 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 1 9 9 6 1 9 9 7 1 9 9 9 1 9 9 8 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 d e c a d a l p re c ip it a ti o n in d e x years (in decadal resolution) d e c a d a l p re c ip it a ti o n in d e x 2 0 0 4 1 9 9 6 1 9 9 7 1 9 9 9 1 9 9 8 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 years (in decadal resolution) years (in decadal resolution) years (in decadal resolution) d e c a d a l p re c ip it a ti o n in d e x d e c a d a l p re c ip it a ti o n in d e x d e c a d a l p re c ip it a ti o n in d e x 2 0 0 4 1 9 9 6 1 9 9 7 1 9 9 9 1 9 9 8 2 0 0 0 2 0 0 1 2 0 0 2 2 0 0 3 2 0 0 4 1 9 9 6 1 9 9 7 1 9 9 8 2 0 0 0 2 0 0 1 2 0 0 2 figure 2: intraseasonal fluctuations of dpi. intraseasonal variability of rainfall over northern south america and caribbean region 201 figure 3. spectral density for dpi of stations located on the islands (left) and coastal sector (right) of caribbean region. 202 j.d. pabón and j. dorado cont. figure 3. of dpi using commercial software that calculates the spectral density. considering that madden-julian oscillation induces the most outstanding signal of isv in the tropics, an attempt to associate the regional isv of precipitation with mjo was done; therefore the madden-julian index (mji) for 120 and 40°w was compared with dpi series. for that, dpi was smoothed using moving averages to filtrate high frequency modes. mji data was taken from noaa/ncep/cpc web page (see http://www.cpc.ncep.noaa.gov/products/precip/cwlink/daily_mjo_index/details.shtml) finally, correlationcoefficients for mjiandoriginal and smoothed dpi were calculated. 3. analysis and discussion figure 2 shows the 3-points (30 days) moving averages of dpi for five stations (one for each delimited sector) in the region; this presentation visualizes the intraseasonal fluctuations of dpi. the mji for 120 and 40°w is also presented to compare with dpi series. the lines that correspond to mji over both 120 and 40°w has similar fluctuations with a noticeable delay caused by the eastward propagation of mjo, however in1997-1998 (during strongelniñoevent) this concordance was disrupted. it is possible to observealso thatduringelniñoevents (1997-1998and 2002-2003 in the analyzed period) the mji tends to have the lowest values, while during cold events la niña (1996, 1999-2000, and 2003-2004) the highest values are presented. at firstglance, in thefigure2 it ispossible toobserve too, that some extreme events of dpi are synchronic indifferent regions in spite theydiffer by theirmagnitude.comparing dpiand mji evolution is not possible to identify anypatternof correlation because insomeperiods theyare in thesamephase, and in timesare in counter phase. however, a correspondence may be detected for extreme rainfall events: maximums values of dpi are associated with minimum of mji; at least the most outstanding rainfall (heavyrainfall)eventsareassociatedto lowvaluesof mji. the figures 3-5 present spectra of several stations of the regions analyzed in this paper (due to limitation of space, is not possible to present the spectra for all 78 stations). also, it is necessary to consider that in thesegraphics theperiods less than 2decades (20days)arenotobserved,because2decades coincide with 0.5 frequency (niquist frequency), under which is not possible to represent the spectrum. theanalysisoffigure3(left side) shows that for east and central caribbean region the graphics are similar: all spectra have peaks at 20, 30, 45 and 60 days period. in the western sector ( aeropuerto el embrujo–providence island, and aeropuerto sesquicentenario – san andres island) these peaks arenotmarkedand just the20,30and60daysperiod are slightly noticeable. in several spectra peaks are presentedevenover80-90-daysperiod,however this interval approaches to seasonal scale. the spectra over caribbean coastal sector (figure 3, right) show also peaks at 20, 30, and 45-days period, but the signal over 60-days period is very weak or is missing. in the mountainous sector of the region (figure 4) there are some places were spectra did not presented an outstanding signal or the signals are weak (airportlanubia,airportanariñoandobonuco in pasto, colombia), however many of them (airport camilo daza – cúcuta, uptc and eldorado, also in colombia) have the 20, 30, and 45-days periods are well defined thefigure4 (right side) showsspectra formeteorological stations of the eastern lowlands of orinocoandamazonbasin.these spectra in the sector of intraseasonal frequencies are very noisy and it isdifficult to identifyclearlydefinedpeaksatagiven period. thesamesituationoccurs inseveralplaces in the pacific sector (figure 5), however there are points liketocumen(panama),panamericanaandbonanza (colombia) and manta (ecuador) where peaks outstand over 20-25, 30 and 45-days period. also, a intraseasonal variability of rainfall over northern south america and caribbean region 203 204 j.d. pabón and j. dorado figure 4. same as figure 3, but for the stations located on andean (left) and eastern plain (right) zone region. intraseasonal variability of rainfall over northern south america and caribbean region 205 cont. figure 4. markedpeak appears over the interval of 50-60-days period in panamericana and buenaventura (colombia) such as in manta (ecuador). summarizing in awhole the region, the signals 20-25, 30and 45-days period are the most noticeable of the isv of precipitation. a 50-60-days sig206 j.d. pabón and j. dorado figure 5. same as in figure 3, but for the stations of pacific zone. intraseasonal variability of rainfall over northern south america and caribbean region 207 f ig u r a 6 . s p at ia l di st ri b u ti o n of sp ec tr al de n si ty fo r 2 0 -, 3 0 -, 4 5 (t o p , fo rm le ft to ri g h t) , 5 5 -, 6 0 an d 9 0 -d ay s (b o tt o m ) . 208 j.d. pabón and j. dorado f ig u r e 7 .s p at ia l di st ri b u ti o n of co rr el at io n co ef fi ci en ts b et w ee n m ji ov er 12 0° w (t op ) an d 40 °w (b ot to m ) an d th e d p i o ri g in al se ri es (l ef t) ,i ts 3 -p o in ts sm o o th ed d p i (c en te r) , an d 5p o in ts sm o o th ed d p i (r ig h t) . nal appears clearly in the caribbean region, but is not important in others regions. the mapspresented in figure 6 show the spatial distribution of spectral density of 20-, 30-, and 45-dayssignals. it ispossible toconclude that inspite the20-dayssignal isobserved inall region, themajor spectral density isobservedoverwesterncaribbean and southwestern sector (ecuador and pacific ocean). for 30and45-days signals there is a similar distribution. the relationship between dpi and mji was explored calculating the correlation coefficients. these coefficients are very low when the correlation is calculated for original (not smoothed) dpi time series, and increase as the high frequency modes are smoothing by moving averages. this fact suggests that the high frequency of isv of regional precipitation is controlled by processes different from mjo. the spatial distribution of correlation coefficients are presented in figure 7. 4. conclusions the analysis made above shows that in the isv of precipitation over northern south america and caribbean region there are signals with 20, 30, 45 and 60-days period. the three first are persistently observed in all the zones of the analyzed region, while the last is observed only in both the caribbean islands and some places of pacific sector. searching therelationshipbetweenintraseasonal variability of regional precipitation and madden-julianoscillationitdidnot findadefinedassociation pattern and even the correlation coefficients betweenmjianddpiwereverylow;however,wasestablished that theheavy rainfall events areassociated with low values of mji. the low values of correlation coefficients and their increasing with smoothing of dpi suggest that the regional isv is controlled not only by mjo. it is necessary to explore the nature of high frequency (20-days, for example) modes. acknowledgments this paper partially reports the results of the research project “analysis of colombian climate variability generated by processes different from el niño-la niña-southern oscillation cycle” supported by both colciencias (colombian research support agency; grant no. 1118-05-16900, rc 178-2004) and research division of national university of colombia (project numbers dib-20100004448 (805253) and dib-20101004929). references alfaro, e., cid, l. & enfield, d. (1998). relaciones entreel inicioyel términode laestación lluviosa en centroamérica y los océanos pacífico y atlántico tropical. revista de investigaciones marinas 26, 59-69. baldwin m.p., gray l.j., dunkerton t. j., hamilton k., haynesp.h.,randelw. j.,holton j.r.,alexanderm. j.,hirota i., t.horinouchit., jones d.b.a., kinnersleyj.s.,marquardtc.,satok., takahashim., 2001:thequasi-biennialoscillation. reviews of geophysicis, 39,2, pp. 179-229 bantzer c h., wallace j m., 1996: intraseasonal variability in tropical mean temperature and precipitation and their relation to the tropical 40-50 day oscillation. j. of atmos. sc., v. 53, no. 21 (nov ), pp. 3032-3045. barlowm., wheelerm.,lyonb.,cullenh.,2005: modulation of daily precipitation over southwest asia by the madden–julian oscillation monthly wea. review, 133 pp3579-3594. bondn.a.,vecchi g.a., 2003:the influenceof the madden-julian oscillation on precipitation in oregon and washington. wea. forecasting, v. 18, pp. 600-613. coaps, 2006: a comprehensive bibliography on the el nino phenomena, center for ocean-atmospheric prediction studies (coaps), florida state university, url: http://www.coaps. intraseasonal variability of rainfall over northern south america and caribbean region 209 fsu.edu/lib/biblio/enso-bib-intro.html (accesed at december 2006). chen a.a., stephenson t.s., 2000: analyzing and understanding climate variability in the caribbean islands. university of the west indies, kingston 7, jamaica. chen, a. a., and m.taylor, 2002: investigating the linkbetweenearly seasoncaribbean rainfall and theelniño+1year. int. j. climatology, v. 22 (1), pp.87-106. enfield, d. b., 1996: relationships of inter-american rainfall to tropical atlantic and pacific sst variability. geophys. res. letters. 23, 3305-3308. enfield, d. & alfaro, e. (1999). the dependence of caribbean rainfall on the interactionof the tropical atlantic and pacific oceans. journal of climate 12, 2093-2103. garreaud r.d., 2000: intraseasonal variability of moisture and rainfall over the south american altiplano. monthly wea. rev., 128, pp. 337-3346. giannini, a., y. kushnir y., cane m.a., 2000: interannual variability of caribbean rainfall, enso, and theatlanticocean. journal of climate, 13, pp. 297–311. goswami, b. n., and r. s. a. mohan, 2001: intraseasonal oscillations and interannual variability of the indian summer monsoon. j.climate, 14, 1180–1198. hastenrath, s., 1976: variations in low-latitude circulationandextremeclimaticevents in the tropical americas. j. atmos. sci., 33, 202-215. hastenrath s.,1996: climatedynamics of the tropics. updated edition from climate and circulation of the tropics. atmospheric sciences library. kluwer academic publishers. dordrecht, netherlands, 488 p. hendonh.h.,liebmannb.,newmanm.,glickj.d., 2000: mediu-range forecast errors associated withactivesepisodesof themadden-julianoscillation. mon. wea. review, v. 128, pp. 69-86. janicot s., sultan b., 2001: intra-seasonal modulationofconvection in thewestafricanmonsoon. geophys. res. lett., 28, 523–526. jones c., carvalho l.m.v., higgins r.w., waliser d.e., schem j.k.e., 2004a: climatology of tropical intraseasonal convective anomalies: 1979-2002. j. of climate, v. 17, pp. 523-539. jones,c., andschemmj.-k.e., 2000:the influence of intraseasonal variations on medium-range weather forecasts over south america. mon. wea. rev., 128, 486–494. jones c., waliser d.e., lau k.m., stern w., 2004b: globaloccurrence of extremeprecipitation and madden-julian oscillation: observations and predictability. j. of climate,v.7,pp.4575-4589. jones c., waliser d.e., lau k.m., stern w., 2004c: the madden–julian oscillation and its impact onnorthernhemisphereweatherpredictability. mon. wea. rev., 115, pp. 1462–1471. kayano m.t., kousky v.e., 1999: intraseasonal (30–60 day) variability in the global tropics: principalmodesandtheirevolution.tellus,51a, pp. 373-376. knutson,t.r., and k. m.weickmann, 1987: 30–60 day atmospheric oscillations:composite life cycles of convection and circulation anomalies. mon. wea. rev., 115, pp. 1407–1436. krishnamurti v., shukla j., 2000: intraseasonal and interanual variability of rainfall over india. j. of climate, v. 13, pp. 4366-4377. krishnamurti v., shukla j., 2007: intraseasonal and seasonally persisting patterns of indian monsoon rainfall. j. of climate, v. 20, pp. 3-20. liebmann b., h. h. hendon, and j. d. glick, 1994: the relationship between tropical cyclones of the western pacific and indian oceans and the madden–julian oscillation. j. meteor. soc. japan, 72, 401–411. 210 mehmet c. demirel, ercan kahya and diego rivera maddenr.a., julianp.r.,1994:observationsof the 40-50-day tropical oscillation— a review. mon. weather rev., v. 122, no. 5, pp. 814-837. maloney e.d., hartmann d.l., 2000a: modulation ofhurricaneactivity in thegulfofmexicobythe madden–julian oscilllation. science, 287, issue 5460, pp. 2002–2004. maloney e.d., hartmann d.l., 2000b: modulation ofeasternnorthpacifichurricanesby themadden-julian oscillation. j. of climate, v. 13, no. 9, pp. 1451-1460. meehl g.a., 1997: the south asian monsoon and the tropospheric biennial oscillation (tbo). j. climate, 10, p. 1921–1943. misrav.,2005:simulationof theintraseasonalvariance of the south americansummermonsoon., monthly wea. rev., 133, pp. 663-676. montealegre j.e., pabón j.d., 2000: la variabilidad climática interanual asociadaal cicloelniño-la niña-oscilación del sur y su efecto en el patrón pluviométrico de colombia. meteorología colombiana, nº 2, pp. 7-21. mounier, f., and s. janicot (2004), evidence of two independent modes of convection at intraseasonal timescale in thewestafricansummer monsoon, geophys. res. lett., 31, l16116, doi:10.1029/2004gl020665. nobre p., marengo j.a., calvacanti i.f.a., obregón g., barros v., camilloni i., camposn., ferreira a.g., 2006: seasonal-to-decadal predictability andpredictionofsouthamericanclimate. j. of climate, 19, pp. 5988-6004. nogués-paegle, j., l. a. byerle, and k. mo, 2000: intraseasonal modulation of south american summer precipitation. mon. wea. rev., 128, 837–850. pabón j.d., montealegre j.e., 1992: interrelación entre el enos y la precipitación en el noroccidente de suramérica. boletín erfen, no. 31, p. 12. pabón j.d., 2007: improving climate prediction schemeswith intraseasonal variability: a key tool toward hydrometeorological disasters reduction in tropical america. in proceedings of internationalroundtablemeetingonlessons fromnaturaldisasterspolicy issuesandmitigation strategies (vellore, india, 8 – 12 january 2007), 10 p. peel m.c., mcmahon ., t.a., finlayson b.l, 2002: variability of annual precipitation and its relationship to el niño-southern oscillation. j. of climate, 15 (6), pp. 545-551. petersenw.a., nesbitt s.w., blaskesleer.j.,cifelli r.,heinp.,rutledges.a.,2002:trmmobservations of intraseasonal variability in convectiveregimesover theamazon. j. of climate,15, pp.1278-1294. philander s.g.h., 1990: el niño, la niña and southern oscillation. academic press, 291p povedag.,mesao.,agudelop.,álvarezj.,ariasp., morenoh., salazarl.,torov.,vieiras., 2002: influencia de enso, oscilación madden-julian, ondasdeleste,huracanesyfasesde lalunaenel ciclo diurno de la precipitación en los andes tropicales de colombia. meteorología colombiana. no. 5, pp.3-12. povedag.,2004:lahidroclimatologíadecolombia: una síntesisdesde laescala inter-decadalhasta la escala diurna. rev. acad. colomb. cien., 28 (107): 201-222. ropelewski, c. f., and m. s. halpert, 1986: north american precipitation and temperature patterns associatedwith theelniño/southernoscillation (enso). mon. wea. rev., 114, 2352–2362. ropelewskic.f.,halpertm.s.,1987: globalandregional scale precipitation patterns associated with the el niño/southern oscillation. mon. wea. rev., 115, pp. 1606-1626. ropelewski c.f., halpert m.s., x. wang 1992: observed tropospheric biennial variability and its intraseasonal variability of rainfall over northern south america and caribbean region 211 relationship to the southern oscillation. j. of climate, 5 (6), pp. 594-614. sultan b., janicot s., diedhiou a., 2003: the west african monsoon dynamics. part i: documentation of intraseasonal variability. j. of climate, 16, pp. 3389-3406. taylor m.a., enfield d.b., chen a.a:, 2002: influence of the tropical atlantic versus the tropical pacificoncaribbean rainfall. j. of geoph. res., v. 107, no. c9, 3127, doi: 10.1029/2001jc001097, 2002 walisser d.e., stern w.,schubert s., lau k.m., 2003: dynamic predictability of intraseasonal variability associated with the asian summer monsoon. q. j. r. meteorol. soc., 129, pp. 2897–2925. wang x.l., corte-real j., zhang x., 1996: intraseasonal oscillations and associated spatial-temporal structures of precipitation over china. j. of geophysicar research, v. 101, no. d14, pp. 19035-19042. webster p.j., hoyos c., 2004: prediction of monsoon rainfall and river discharge on 15-30-day time scales. bull. of the american meteo. soc., v. 85, pp. 1745-1765 wilks d.s., 1995: statistical methods in atmospheric sciences. academic press. san diego. 467 p. yeh.,choh.-r., 2001:spatial and temporal characteristics of intraseasonal oscillations of precipitation over the united states. theoretical and applied climatology. v.68, pp. 51-66. 212 j.d. pabón and j. dorado the influence of shale composition on methane adsorption capability and gas content is investigated using 14 samples from well ys8 in the southern sichuan basin, china. the results show that the langmuir adsorption capacity of the longmaxi shale is mainly a function of the total organic carbon (toc) content. when toc is ~1.1%, 50% ch4 is adsorbed onto the surface of the organic matter. the mineral content has limited control on the adsorption capacity of the longmaxi shales. organic matter is also a major control on gas content when toc content is <1.0%. when toc is >1.0%, gas content remains constant, indicating that gas preservation is more important than gas generation and rock adsorption capacity. scatter plots of toc versus gas content and, langmuir adsorption capacity shows that when toc is <2.0%, ch4 occurs both as free and absorbed gas, and ch4 occurs mainly as absorbed gas when toc is >2.0%. abstract influence of black shale composition on methane adsorption and gas content: implications for gas storage in the longmaxi black shales issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n1.70539 haihua zhu1,2,3*, tingshan zhang1,2, jun lang1,2, jianli zeng1,2, xing liang4, yong he4, gaocheng wang4, junjun li4 1 stake key laboratory of oil and gas reservoir geology and exploitation, chengdu,china 2 school of geoscience and technology, southwest petroleum university, chengdu, china 3 unconventional oil and gas assessment center, southwest petroleum university, chengdu, china 4petrochina, zhejiang oil field company, hangzhou, china *email of corresponding author:zhhswpu@163.com earth sciences research journal earth sci. res. j. vol. 22, no. 1 (march, 2018): 59-63 la influencia de la composición de lutitas en la capacidad de absorción del metano y el contenido de gas se analiza en este trabajo a partir de 14 muestras del pozo ys8 en la cuenca de sichuan, al suroeste de china. los resultados muestran que la capacidad de absorción langmuir para las lutitas de longmaxi es principalmente una función del contenido de carbono orgánico total (toc, del inglés total organic carbon). cuando el índice toc es ~1.1%, el 50 % del ch4 (capacidad máxima de absorción de la función langmuir) es absorbido por la superficie de materia orgánica. el contenido mineral tiene un control limitado en la capacidad de absorción de las lutitas de longmaxi. la materia orgánica también tiene un control mayor en el contenido de gas cuando el índice toc es <1.0 %. cuando el toc es >1.0 %, el contenido del gas es constante, lo que indica que la preservación del gas es más importante que la generación del gas, y la capacidad de absorción de la roca. diagramas de dispersión del toc frente al contenido de gas y de la capacidad de absorción langmuir muestran que el toc es <2.0 %, que el ch4 se alcanza tanto en gas libre como en gas absorbido y que este ocurre principalmente cuando el gas absorbido tiene un nivel toc de >2.0 %. resumen how to cite item zhu, h., zhang, t., lang, j., zeng, j.,liang, x., he, y., wang, g., & li, j. influence of black shale composition on methane adsorption and gas content: implications for gas storage in the longmaxi black shales. earth sciences research journal, 22(1), 59-63 doi:http://dx.doi.org/10.15446/esrj.v22n1.70539 influencia de la composición de lutitas negras en la absorción de metano y el contenido de gas: implicaciones para el almacenamiento de las lutitas negras de longmaxi keywords: shale; methane; adsorption; toc; sichuan. palabras clave: lutitas; metano; adsorción, carbono orgánico total. g e o c h e m is t r y record manuscript received: 14/11/2017 accepted for publication: 21/03/2017 60 haihua zhu, tingshan zhang, jun lang, jianli zeng, xing liang, yong he, gaocheng wang, junjun li 1. introduction the wufeng–longmaxi black shale in the sichuan basinis the first shale gas reservoir to be commercially produced in china. by 2016, four national demonstration areas of marine shale gas production (the weiyuan, changning, zhaotong, and jiaoshiba shale gas plays) had beenestablished in sichuan, chongqing, and yunnan.however, the heterogeneity of black shalesconcerning the total organic carbon(toc) content and mineral composition, which control gas generation and reservoir quality, limits thepredictability of theshale gas ‘sweet spot(chen et al., 2015; farrooqi et al., 2017). therefore,it is necessary to evaluate the control of shale composition on storage capabilityand shale gas content. in this study, 14samples from well ys8 in the southern sichuan basin areanalyzed to investigate the geochemistry, petrology, and methane adsorption capacityof the shale by toc testing, xrd analysis, and methane adsorption experiments. by discussing the relationship between shale composition and methane adsorption capacity, andshale gas content, we hope to identify the main factors that influence the storage capability and gas content of shale reservoirs,which can then be applied to other shale reservoirs with a similar preservation condition to that of the longmaxi shale. 2. geological setting the study area is located in the changning shale gas field in the southern sichuan basin (figure 1). the sichuan basin is located on the northwestern yangtze plateand is the remnant of the lower paleozoic upper yangtze cratonic basin(lu et al., 2006; karami et al., 2017). the wufeng–longmaxi formations were formed in a continental shelf sea(zhang et al., 2012; liang et al., 2016) during the late ordovician and early silurian(su et al., 2009; sunny et al., 2018).the thickness of the strata varies from several meters to >300 m, with depocenters located in the southern and eastern parts of the basin(guo et al.,2014; nwankwo and nkwankwoala, 2018). thewufeng– longmaxi shale is now mainly thermally over-mature (liang et al.,2011). figure 1. study area and paleogeography of the wufeng and longmaxi formations 3. samples and methods fourteen shale core samples with burial depths ranging from 2382.43 to 2512.42m were selected for analysis from shale gas wellys8 in the study area.the whole-rock mineral compositionswere determined by xrd analysis. the samples were crushed to powder, and xrd analyses were performedusing an x’pert pro x-ray diffractometer with a cu anode at 40 kv, 20 ma, and 0.154 nm wavelength. the toc was measured with a leco cs-400 carbon– sulfur analyzer (combustion at temperatures of >800°c). before analysis, carbonate from the samples wasremoved usinghydrochloric acid (1:1). the methane adsorption capacity was investigated using a pctproe&e high-pressure adsorption–desorption instrument. the samples were dried for 48 h at 110°cand tested in asample cell with a temperature of 30°c (±0.2°c) and a pressure of 0–11mpa.the room temperature was 30°c and humiditywere 1%–2%. the methane adsorption capacity was calculated in m3/t (stp: 0°c, 101.325 kpa) using the material balance principle and the volumetric method (krooss et al., 2002; ali et al., 2018). the shale gas content of core samples comprisesthe desorbed gas, residual gas, and lost gas. desorbed gas and residual gas weremeasuredusingcanister desorption of the fresh cores. the lost gas was calculated using usbm methods. 4. results 4.1 gas content the gas content ofthe 14 samples ranges from 1.49 to 3.43 m3/twith an average of 2.59 m3/t. gas content varies with depth (table 1),showing a decreasing upward trend. the shale samples with a gas content of>2.0 m3/t are found mainly between 2486.77 and 2512.42m depth. 4.2 toc content the toc content of the 14 samples ranges from 0.34% to 4.72% with an average of 1.81%. the maximum toc(>2.0%) was obtained for sixshale sampleswith a burial depth between 2491.48 and 2512.15m(table 1). the samples from depths of <2481m have toc contents of<1.0%. a similarupward decreasingtrend in the toc content of shalehas been reportedelsewhere in the upper yangtze region (e.g.,liang et al., 2011; guo et al., 2014; asghar et al., 2018). table 1. toc, gas content, and langmuir adsorption capacity of the longmaxi shales in ys8 4.3 mineral composition the xrd data show that the 14 black shale samples from well ys8 are dominated by carbonate, clay minerals, and quartz, with subordinate feldspar and pyrite (table 2). the carbonate mineral content ranges from 15.6% to 62.0%(average 32.2%), the clay mineral content ranges from 14.1% to37.2%(average 28.1%), and the quartz contentranges from 15.3% to 41.7%(average28.3%). the feldspar contentranges from 4.2% to16.4%(average 8.0%) and the pyrite content from 1.5% to 9.2%(average3.4%). mineral composition data show vertical variations. the three samples (ys8-1 to ys8-3) withthe minimum burial depths are characterized by high carbonate and quartz contents, moderate clay mineral contents, and low toc contents (hasan, 2018).the fourdeepest samples (ys8-11–ys8-14) contain high carbonate and toc contents, and low clay contents. theseven middle samples (ys8-4 to ys8-10) contain low carbonate, high clay, and moderate toc contents. 61influence of black shale composition on methane adsorption and gas content: implications for gas storage in the longmaxi black shales table 2. mineral compositions of longmaxi shale samples from well ys8 4.4 methane adsorption methane adsorption isotherms are shown in figure 2. the measured isotherms at different temperatures can be fitted well to the langmuir function. the langmuir maximum ch4 adsorption capacities range from 0.95 to 4.11 m3/t, with highercapabilitiesfoundfor samples ys8-9 to ys8-14 (>3.1m3/t).the adsorption capacities of samples ys8-1 to ys8-7 are <1.8 m3/t. figure 2. methane adsorption isotherms of the studied samples 5 discussion 5.1 controls on ch4 adsorption in black shales figure 3 shows scatter plots of the langmuir maximum ch4 adsorption (γl) versus shale composition (organic matter and minerals).γlshows astrong positive correlation with toc. figure 3.langmuir adsorption capacity versus shale composition equation (1) shows the result of linear regression betweenγl and toc content, based on the experimental measurements of this study. γl= 0.8137×toc + 0.891 (1) whereγl is in units of m 3(ch4)/t rock and toc is the total organic carbon content in units of wt%. the regression provides an empirical basis for estimatingthe langmuir maximum ch4 adsorption capacity from toc content in a shale reservoir.a similar correlation between gas sorption capacity and toc content at 30°c and 6 mpa pressure has been observed in previous studies (chalmers and bustin, 2007a,b; ross and bustin, 2007, 2009; zhang et al., 2012; msarah and alsier, 2018). the strong correlation betweenγland toc content is due to the massive surface areaof organic matter for ch4 adsorption.caoet al. (2015) reported thatkerogen of the longmaxi shale in the sichuan basin hasa surface area of ~161.2m2/g, which is much higher than that of quartz and carbonate minerals. the proportion of ch4 adsorbedonto organic matter can be estimated byequation (2). γl = toc×γorg. + mineral content×γrock (2) where γorg is thelangmuir ch4 adsorbed onto organic matter and,γrock is thelangmuir ch4 adsorbed onto minerals. mineral content is the total mineral content in wt%, toc is the total organic carbon content in wt%, and mineral content+toc =100%. in the model, we assume that i) ch4 is adsorbed onto two types of solid material (organic matter and minerals), and ii) γrock remains constant, and the change of γrock due to the variations in type and relative content of minerals is negligible compared with the influence of toc content. base on equation (2), when tocis 0%all the ch4 is adsorbed onto inorganic mineral surfaces, with 0.891 m3of ch4 adsorbed for every ton of inorganic minerals. as toc content increases and minerals content decreases, the proportion of ch4 absorbed onto organic matter increases, and when toc is ~1.05%, ~50% of the ch4 is absorbed onto thesurface of organic matter (figure 4). figure 4.relative proportion of ch4 adsorbed on organic matter and minerals 5.2 methane absorption capacity and shale gas content due to the high-to-over thermal maturity of organic matter and strong tectonic deformation,the gas content of the longmaxi shalesis controlled by many factors, especiallythe preservation condition, natural matter abundance, and reservoir quality(e.g.,zou et al., 2010; rahim et al., 2018). in the shale samples from well ys8,the preservation conditions and thermal evolution processes of organic matterare similar, thus enablingan evaluation ofthe influence of shale composition on shale gas content. 62 haihua zhu, tingshan zhang, jun lang, jianli zeng, xing liang, yong he, gaocheng wang, junjun li scatter plots of gas content versus shale composition are shown in figure5. poor relations are found between gas content versus mineral composition.regular change of shale gas content with toc content is observed. as toc content increases, gas content first increases and then remains constant. this relationship indicates that organic matter is a significant controlon gas content when toc content is <2.0%. when tocis >2.0%, the adsorption surface area is sufficient, and other factors such as gas preservation condition may influence the amount of residue gas (indan et al., 2018).a similar pattern isobtained when gas contentis plotted against pyrite content. pyrite content is considered an indicator of redox conditionsand is usually found in larger concentrations when toc content is high, which may explain thesimilar correlation of gas content versus pyrite and toc contents. figure 5. shale gas content versus shale composition the scatter plots of toc content versus gas content, and langmuir adsorption capacity shows that when toc is<2.0% the gas contents are higher than the langmuir adsorption capacity of shale(figure 6). therefore, ch4 partially occurs as free gas within the macropores of the shale (nwankwoala and oborie, 2018; nwankwoala and ememu, 2018).when tocis >2.0%, the gas contents are less than the langmuir adsorption capacity of shale,indicating that ch4 tends to occur mainly as absorbed gas, as the surface area of organic matter is sufficient for gas absorption. figure 6. the scatter plots of toc content versus gas content and langmuir adsorption capacity 6. conclusions the langmuir adsorption capacity of the longmaxi shale is mainly a function of toc content. the regression shows that when toc content is ~1.1%, 50% of the ch4 is adsorbed ontothe organic matter. the mineral content has control on the adsorption capacity of the longmaxi shales. organic matter is a major control on gas content when toc content is<2.0%. when toc is>2.0%, gas content remains constant and is a function of organic matter. scatter plots of gas content and langmuir adsorption capacity versus toc contentshow that when toc is<2.0%, ch4 occurs as both free and absorbed gas andthat ch4 occurs mainly as absorbed gas when tocis >2.0%. 7. acknowledgments this work was funded by the young scholars development fund of southwest petroleum university (swpu) (201499010046), china postdoctoral science foundation (2015m580797), national natural science foundation of china(41772150) and national science and technology major project (2017zx05063002-009). references ali, w., nasir, m. s., nasir, a., rashid, h., ayub, i., gillani, s. h., & latif, m. j. (2018). assessment of carbon footprints in terms of co2 of diesel generator, pakistan. earth sciences pakistan, 2(1), 15-17. asghar, z., ali, w., & nasir, a. (2018). atmospheric monitoring for ambient air quality parameters and source apportionment of city faisalabad, pakistan. earth sciences pakistan, 2(1), 01-04. cao, t. t., song, z. g., & wang, s. b. (2015). a comparative study of the specific surface area and pore structure of different shales and their kerogens. science china earth sciences, 58(4), 510–522. chalmers, g. r. l., & bustin, r.m. (2007). the organic matter distribution and methane capacity of the lower cretaceous strata of northeastern british columbia, canada. international journal of coal geology, 70, 223–239. chen, l., lu, y., & jiang, s. (2015). heterogeneity of the lower silurian longmaxi marine shale in the southeast sichuan basin of china. marine and petroleum geology, 65, 232–246. farooqi, z. u. r., nasir, m. s., & nasir, a. (2017). evaluation and analysis of traffic noise in different zones of faisalabad an industrial city of pakistan. geology, ecology, and landscapes, 1(4), 232-240. guo, t. l., & zhang, h. r. (2014). formation and enrichment mode of jiaoshiba shale gasfield, sichuan basin. petroleum. exploration & development, 41, 31–40. hasan, m. (2018). effect of rhizobium inoculation with phosphorus and nitrogen fertilizer on physico chemical properties of the groundnut soil. environment and ecosystem science, 2(1), 04-06. indan, e., roslee, r., & tongkul, f. (2018). earthquake vulnerability assessment (evas): analysis of environmental vulnerability and social vulnerability in ranau area, sabah, malaysia. geological behavior, 2(1), 24-28. karami, a., karamshahi, a., & shahi, e. (2017). effects of forestry practices on the regeneration and biodiversity of woody plants in the northern forest ecosystems of iran. geology, ecology, and landscape, 1(4), 264-270. krooss, b. m., bergen, f. v., & gensterblum, y. (2002). high-pressure methane and carbon dioxide adsorption on dry and moisture-equilibrated pennsylvanian coals. international journal of coal geology, 51(2), 69-92. liang, c., jiang, z., & cao, y. (2016). deep-water depositional mechanisms and significance for unconventional hydrocarbon exploration: a case study from the lower silurian longmaxi shale in the southeastern sichuan basin. aapg bulletin, 100(5), 773–794. liang, x., ye, x., & zhang, j. h. (2011). reservoir forming conditions and favorable exploration zones of shale gas in the weixin sag, dianqianbei depression. petroleum exploration and development, 38(6), 693–699. lu, k., zhu, x., & qi, j. (2006). petroleum-bearing basin analysis. china university of petroleum press, beijing. pp. 424. (in chinese). msarah, m., & alsier, a. (2018). protein digestibility and amino acid content of malaysian local egg protein prepared by different methods. environment and ecosystem science, 2(1), 07-09. nwankwo, c., & nwankwoala, h. o. (2018). gully erosion susceptibility mapping in ikwuano local government area of abia state using gis 63influence of black shale composition on methane adsorption and gas content: implications for gas storage in the longmaxi black shales techniques. earth sciences malaysia, 2(1), 08-15. nwankwoala, h. o., & ememu, a. j. (2018). hydrogeochemical signatures and quality assessment of groundwater in okpoko and environs, southeastern nigeria. pakistan journal of geology, 2(1), 06-11. nwankwoala, h. o., & oborie, e. (2018). geological and hydrogeological characterization of a hydrocarbon impacted site in the niger delta. pakistan journal of geology, 2(1), 12-17. rahim, i. a., tahir, s., & musta, b. (2018). urbanization vs. environmental quality: some observation in telipok, sabah, malaysia. geological behavior, 2(1), 12-17. ross, d. j. k., & bustin, r. m. (2007). shale gas potential of the lower jurassic gordondale member, northeastern british columbia, canada. bulletin of canadian petroleum geology, 55, 51–75. ross, d. j. k., & bustin, r. m. (2009). the importance of shale composition and pore structure upon gas storage potential of shale gas reservoirs. marine and petroleum geology, 26, 916–927. su, w., huff, w. d., & ettensohn, f. r. (2009). k-bentonite, black-shale and flysch successions at the ordovician–silurian transition, south china: possible sedimentary responses to the accretion of cathaysia to the yangtze block and its implications for the evolution of gondwana. gondwana research, 15(1), 111–130. sunny, a. a., omowumi, a., & chris, o. a. (2018). improved magnetic data analyses and enhancement techniques for lithological and structural mapping around akure, southwestern nigeria. earth sciences malaysia, 2(1), 16-21. zhang, c., zhang, w., & guo, y. (2012). sedimentary environment and its effect on hydrocarbon source rocks of longmaxi formation in southeast sichuan and northern guizhou. earth science frontiers, 19(1), 136–145. zou, c., dong, d., & wang, s. (2010). geological characteristics and resource potential of shale gas in china. petroleum exploration and development, 37(6), 641–653. earth sci. res. j. vol. 11, no. 2 (december 2007): 139-154 hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan adrian h. gallardo 1,2,*, atsunao marui 1 1aist, geological survey of japan, quintessa japan, yokohama 220-6007 2hytec co. yodogawa-ku, miyahara 2-11-9, osaka 532-0003, japan *corresponding author. a. h. gallardo. e-mail: ad.gallardo@aist.go.jp tel/fax: +81-29-861-3684 abstract hydraulic characteristics of sediments were investigated at j-parc for the purpose of site characterization in relation with the construction of japan’s largest proton-accelerator. a total of 340 samples extracted from 9 exploratory wells were examined by standard laboratory tests and complemented with statistical analyses to quantitatively determine the main terrain attributes. two main hydro-geological units were recognized, although a number of embedded layers defined a multilevel aquifer. grain-size distribution derived from sieve analysis and the coefficient of uniformity showed that soils are poorly sorted. on the other hand, hydraulic conductivity was measured by a number of parameters such as a log-normal distribution. conductivity was also predicted by empirical formulas, yielding values up to three orders of magnitude higher. discrepancies were explained in terms of soil anisotropy and intrinsic differences in the calculation methods. based on the shepherd’s approach, a power relationship between permeability and grain size was found at 2 wells. hydraulic conductivity was also correlated to porosity. however, this interdependence was not systematic and therefore, properties at many parts of the profile were considered to be randomly distributed. finally, logs of electrical conductivity suggested that variations of soil hydraulic properties can be associated to changes in water quality. in spite of the remaining uncertainties, results yielded from the study are useful to better understand the numerical modelling of the subsurface system in the site. key words: drill core, soil properties, heterogeneity, site characterization, proton accelerator resumen las características hidráulicas de los sedimentos fueron investigadas en j-parc con el proposito de caracterizar sitios relacionados con la construcción del acelerador de protones más grande de japón. un total de 340 muestras extraídas a partir de 9 pozos exploratorios fueron examinadas a partir de pruebas de laboratorio estándar y complementadas con análisis estadísticos para determinar cuantitativamente las cualidades principales del terreno. dos unidades hidrogeológicas principales fueron reconocidas, aunque un número de capas encajadas definieron un acuífero de niveles múltiples. 139 earth sciences research journal manuscript received june 30 2007. accepted for publication november 20 2007. la distribución del tamaño del grano derivada de análisis de tamices y del coeficiente de uniformidad demostró que los suelos están mal clasificados. por otra parte, la conductividad hidráulica fue medida a partir de un número de parámetros tales como una distribución logarítmica normal. la conductividad también fue predicha por medio de fórmulas empíricas, mostrando valores de hasta tres órdenes de magnitud. las discrepancias fueron explicadas en términos de la anisotropía del suelo y diferencias intrínsecas en los métodos de cálculo. basados en la aproximación de shepherd, se encontró una fuerte relación entre la permeabilidad y el tamaño de grano en 2 pozos. la conductividad hidráulica también fue correlacionada con la porosidad. sin embargo, esta interdependencia no era sistemática y por lo tanto, propiedades en muchas partes del perfil eran consideradas como distribuidas aleatoriamente. finalmente, los registros de la conductividad eléctrica sugirieron que las variaciones de las propiedades hidráulicas del suelo se puedan asociar a los cambios en la calidad del agua. a pesar de las incertidumbres restantes, los resultados del estudio son útiles para entender mejorar el modelamiento numérico del sistema subterraneo en el sitio. palabras claves: nucleos, propiedades del suelo, heterogeneidad, caracterización del sitio, acelerador de protones. 1. introduction the construction of japan`s largest proton accelerator has been initiated within j-parc (japan proton accelerator research complex) at the coasts of tokaimura, about 100 km from tokyo. in this context, hydraulic conditions of the main sedimentary units were investigated as a pre-requisite to simulate the flow conditions at the site. groundwater is the primary pathway by which radionuclides can migrate from the underground to the biosphere (fairhurst, 2004), so understanding the flow and the type and characteristics of the geologic medium plays a fundamental role to assess the safety of the accelerator. soils are heterogeneous systems with regions more or less favorable to flow distributed spatially in intricate patterns (giménez et al., 1999). lithofacies usually show high spatial variability and as a result, the associated hydrologic and physical properties can be of appreciable different character. as the problem of adequately modeling subsurface becomes more difficult with increasing heterogeneity, determinations based on a large amount of samples have great importance to draw reliable conclusions. an extensive analysis of drill cores appears to be a powerful way to delineate sedimentary structures and to define the formation properties. impervious materials are the most favorable for radio-nuclides isolation and consequently, determination of hydraulic conductivity becomes of paramount importance. pumping and slug tests are frequently used to measure it in porous media; however, the analysis of textural parameters is a comparatively cost-effective and quick technique which has also the advantage of providing information at a more detailed scale. in this line of thought, saturated hydraulic conductivity can be approximated either from laboratory measurements or by empirical formulas derived from the size of the granular sediments (vukovic and soro 1992). moreover, shepherd (1989) derived a power function between permeability and particle size which does not depend on the temperature of the medium, constituting thus an alternative to other techniques. this approach was also followed to predict the hydraulic conductivity of soils at j-parc. porosity changes will influence permeability and hence the possible transport of the radio-nuclides (de craen et al., 2004). for soils of similar origin, the relationship of permeability to porosity can hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan 140 gallardo and marui. esrj vol. 11, no. 2. december 2007 141 be strong (ahuja et al., 1989); therefore, the possible interdependence of these properties was also examined. summing up, the present work presents the results obtained from nine exploratory boreholes within j-parc, evaluating the sedimentary sequence and investigating the main properties of the soils to effectively characterize the hydrogeology of the site. the study attempted to find the existence of meaningful relationships among parameters, providing background information relevant for predictions of groundwater flow in the area. 2. site location and geological setting the j-parc complex is located at the tokaimura, ibaraki prefecture, which is approximately 100 km northeast of tokyo. the investigated wells are essentially concentrated around the proton accelerator, a circular area of about 500 m of diameter close to the mouth of the shinkawa river (fig. 1). the regional geology has been extensively studied by sakamoto et al. (1972). these authors identified the basement of the area as a marine sequence of turbidite and slump deposits from the late cretaceous. the strata reach up to 1,500 m of thickness. above it, the neogene tertiary is occasionally exposed on sea and terrace cliffs. the taga formation from the late miocene outcrops along the shinkawa river and was reached by drilling at about 50 m below the ground. this unit consists mainly of sandy siltstones with abundant mollusk fossils. intercalated tuffs would correspond to sub-aqueous pyroclastic flows. their thickness ranges from nearly 10 m at the inland wells to a minimum of 1 m by the seaside. the miwa formation is in discordance and constitutes the base of the pleistocene. its lower member reaches a maximum thickness of 30 m, and is composed of gravels and mud that filled the valleys at the early phase of the last interglacial transgression. the upper end of the formation consists of sands and pebbly sands deposited at the maximum stage of the transgression. its thickness ranges between 15 and 20 m. in addition, a buried channel with numerous levels of marine and fluvial terraces has been observed in the upper levels of the formation, right below the accelerator site (gallardo, 2006). overlying the miwa formation, the fluvial gravels of the nukada terrace were deposited in unconformity during a regressive period. further regression was responsible of the formation of subsequent terraces (takai and tsuchi, 1963). recent alluvium is present along the streams, while a belt of sand dunes extends along the figure 1. location of the research area and boreholes shoreline. 3. methodology 3.1 monitoring wells and coring nine exploratory wells were drilled in and around the accelerator ring to get basic information about field conditions. soil specimens were recovered continuously at intervals of 1 m by the hybrid boring method (takeda and komiya, 2006). this technique was originally designed for subsurface exploration in damns but is currently being tested in other underground works in japan. as in standard methods, sampling consists in driving a hard bit into the desired depth and coring the material directly into a double steel tube with an internal diameter of 68 mm. however, the success of the hybrid boring relies in the circulation fluid. a dry-foam based on air bubbles originates a fluid that improves considerably the cores recovery even in non-cohesive sediments. soil recovery was close to 100% up to depths of 200 m. after retrieving, cores were wrapped in humid polyethylene to prevent desiccation, and finally packaged in sealed wooden-boxes. while all cores were numbered, logged, and photographed, a subset of 340 samples was used for the physical analyses. at individual wells, the number of analyzed samples increased systematically with the length of the column, bringing more confidence for comparison purposes. 3.2 soils physical analysis the grain-size distribution in soils was determined by sieve analysis for particles above 63 μm, and by an analyzer shimadzu sa-cp3l for finer sizes. the coefficient of uniformity (cu) was derived from the formula: 10 60 d d cu = (1) where d60 and d10 represent the diameters corresponding to the percent finer of 60% and 10% respectively. for conductivity measurements, a 100 cc can was driven into the cores avoiding fracturing or shrinkage. it was left overnight in water to reach saturation, weighted, and finally tested by the falling head method. the results were corrected to a water temperature of 15 ºc in accordance with standard methods. after the analysis, samples were oven-dried at 105 ºc during 24 h, re-weighted, and total porosity (n) was calculated by the relation: s w v v n 100 = (2) where n: total porosity of the sediment (as percentage); vw: volume of water filling the sample voids; and vs: total volume of the sample. the weight of the water removed was converted to volume considering its density as 1 g/cm3. although it is quite difficult to completely saturate a sample (freeze and cherry 1979), this simple method produced quick and satisfactory results. hydraulic conductivity was also calculated from the grain-size distribution. the method of creager et al. (1945) provided a straightforward relation between the effective grain-diameter of 20% weight content (d20) and conductivity. kasenow (2002) presented a detailed compilation of empirical formulas for hydraulic conductivity determination. based on the effective grain diameters of the sediments 142 hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan 143 involved, the equations of slichter and sauerbrei were considered the most appropriate for the analysis. the sauerbrei formula is applicable when the grain-size diameter is no greater than 0.5 mm. it can be written as 1722 3 )1( d n n v g k z tb       − = (3) while the slichter formula can be applied for an effective grain diameter (de) ranging between 0.01 and 5 mm. it takes the form of: ( ) 210)( dnv g k s jb= (4) in eqs. (3) and (4), k, hydraulic conductivity (l/t); g, gravitational constant (l/t2); ν, kinematic viscosity for a given temperature (l2/t); n, porosity (unitless); βz and βs, constants; τ correction for temperature; j (n) porosity function; and d, effective grain diameter (l). 4. results and discussion 4.1 hydrological setting the top of the hydrostratigraphic sequence is characterized by a complex of sand and conglomerates with a thickness that varies from 13 to about 55 m (fig. 2). the uppermost 10-13 m of the unit contains basement pebbles and rests of shells attributed to beach and fluvial deposits. the water table is found at an average of 6 m below the ground, sustaining a great number of residential wells used for irrigation and domestic needs. the successive fluctuations of the sea level over geologic times have enhanced the intercalation of thin horizons of silts and clays that cause local confinement at specific sites. siltstones of the taga formation constitute the bottom of the unconfined aquifer. there is some presence of interbedded tuffs and sandy lenses with a thickness ranging from a few cm to nearly 5 m. a somewhat different stratigraphic scheme is seen at well 0402 where fine sandstones occur throughout the borehole to a depth of 200 m. the hydraulic contact between both quaternary and tertiary sandstones might suggest the existence of an irregular unconfined aquifer extending along the coastal plain. 4.2 physical properties grain size is a fundamental physical parameter of sediments and, as such, is a useful descriptive property (friedman et al., 1992). a frequency distribution of the whole population showed that the sediments tend to be unimodal, with a mode of ɸ 2.2 (fine sands), and a negative skewness toward very fine sands (fig. 3). for a better interpretation, soils were discriminated between tertiary and quaternary origin. results indicated that the shallow fluvial and beach deposits have a relatively larger mode, in the order of ɸ 1.1 (medium sand), whereas there are no significant differences in the values and distribution for the rest of the strata. differences in grain size between shallow and deep sediments might be associated with variations in the depositional environments. fluvial sediments generally have coarser grains than units deposited in coastal environments, and their mean and maximum grain size commonly reflects the average energy of the depositional medium (boggs jr., 1987). particles over 2 mm normally consist of shell fragments and basement-derived pebbles transported under conditions of high flow. although simple, the two distinguished sections of the profile supported the distinction between terrigenous deposits near the surface, and gallardo and marui. esrj vol. 11, no. 2. december 2007 144 figure 2. simplified stratigraphy based on the drill cores figure 3. grain sizes frequency distribution obtained by sieve analysis. units: phi (φ) hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan 145 transitional to marine sediments at depth. the degree of sorting of a soil can be estimated based on the range of its standard deviation. following the formula of folk and ward (1957), the calculated standard deviation varied from 2 to 3.5 ɸ, which correspond to very poorly to extremely poorly sorted sediments (folk, 1974). a single parameter may not be adequate to distinguish sediments from different wells; therefore, the coefficient of uniformity (cu) was also calculated. soils having a coefficient smaller than about 2 would be considered uniform (lambe, 1951). such a threshold was exceeded in several orders by the vast majority of the measured samples, demonstrating the good gradation (or poor sorting) of the sediments concerned (fig. 4). in general, there were no substantial differences in the arithmetic mean of cu among boreholes with exception of the well 0403, which showed a mean more than twice the value obtained for the rest of the samples. three groups with different sorting could be qualitatively defined when evaluating the range of cu at each well. as expected, shallow wells presented the best (although still poor) sorting, as they usually intercepted a limited number of formations also associated to higher homogeneity in their sedimentary attributes. the size range reflects the shortterm random variation of the fluid forces involved in sediment transport (bridge, 1981). therefore, it can be assumed that even though shallow sediments would have been deposited in an environment with sufficient energy to transport a wide range of particle sizes, they could have undergone important reworking especially by tides. finer particles may have been removed by the return flow moving up the front of a beach, leaving behind a population of relatively well-sorted, coarser grains (boggs jr., 1987). no meaningful relationship between sorting and geology could be found for wells 0402 and 15-1, while wells 0401 and 0403 presented the widest grain-size distribution throughout the site. in effect, the deeper the well the higher the non-uniformity, as longer soil profiles included more layers and sedimentary horizons, resulting consequently in larger deviations of the cu. 4.3 hydraulic conductivity saturated hydraulic conductivity (ks) is arguably one of the most important hydrologic properties of soils (mbagwu, 1995). conductivity values ranged from 4.5 x 10-2 to 1.3 x 10-6 cm/sec, with an isolated sample in the order of 10-7 cm/sec within a 1.5 m horizon of fine sands and organic silts at well no. 1. the mean value exhibited an irregular pattern along the profile, except for wells 0405 and tokai where ks decreased with depth (fig. 5). hydraulic conductivity at the upper horizons is two to four times higher than values near the wells bottom, in coincidence with the presence of 20-25 m gravel-filled paleochannels. in addition, roots and root channels related to the original vegetation covering the site could promote water movement through the topsoil layers. variances calculated for the ln(ks) at each well ranged between σ 2 = 1.44 and 7.24 confirming that the variability of the hydraulic parameter within the soil columns is rather distinct. the distribution of permeability in porous media has been usually reported to approximate a log-normal distribution (freeze, 1975; sudicky, 1986). a plot for the entire data set shows that hydraulic conductivity in j-parc is close to this distribution (fig. 6). this statement was validated by the shapiro-wilk test (one sample vs. normal) at the 1% significance level. the test did not reject the null hypothesis of normality (w: 0.99) and gallardo and marui. esrj vol. 11, no. 2. december 2007 146 consequently, the sample can be considered to follow a log-normal distribution. to reduce the probability of correlation, the statistical distribution should ideally be evaluated on a random subsample of the data set (turcke and kueper 1996). again, the log-normality of ks was evaluated by the shapiro-wilk test for a random set of 100 samples (w: 0.96), confirming that the assumed distribution is correct. values derived from core samples represent ks in the vertical direction. however, it is widely known that in heterogeneous aquifers the permeability tends to be different along each of the principal directions. differences were examined introducing the ratio of horizontal (kh; from empirical formulas) to vertical conductivity (kv; from direct measurements). in general, empirical formulas yielded higher values of ks than direct measurements, with differences sometimes as much as three orders of magnitude. the slichter formula gave the lowest differences, with a mean kh/kv ratio of 8.8 (table 1). this mean ratio applied to most of the stratigraphic columns, defining soils of relatively small anisotropy, with some occasional and random discrepancies presumably caused by local heterogeneities in the mineral content or pore distribution. on the other hand, conductivity calculated by the creager and the sauerbrei approaches showed a poorer agreement with core measurements. although, a mean kh/kv ratio of 24.8 and 29 respectively is still fairly consistent, considering differences between the conductivity tensors may range from 1 to 42 (muskat 1937; bear 1972). conductivity showed substantial variations along depth, with differences of more than two orders of magnitude respect kv measurements, and kh calculated by the slichter solution. the greatest values of kh were found on the upper 50 m of the sequence, between 70 and 130 m, and at about 160 m. it is reasonable to accept larger variations in conductivity at shallow depths, where unconsolidated fluvial sediments exist. these kind of deposits have been described to be strongly heterogeneous (jussel et al., 1994), and would be also affected by surface desiccation and human activities, which promote degradation of the upper horizons and enhance the anisotropy of the soils (regalado and muñoz, 2004). in contrast, there was no reliable relationship to substantiate the differences in conductivity results at deeper soils. it is possible that sediments that appear to be relatively homogeneous are in fact affected by micro-scale bedding and microscopic textural variations, which imprint the profile its anisotropic characteristics. with respect to the empirical formulas, values derived from the slichter equation were nearly three times smaller than results obtained by the methods of creager and sauerbrei. the origin of the differences should be searched within the formulas themselves. grain size is the fundamental independent parameter that controls hydraulic conductivity in unconsolidated sediments (pryor, 1973; morin, 2006) but empirical formulas were elaborated introducing different relations and therefore, it is not unusual they yield divergent results. since each formula bases its calculations on a different effective grain-diameter, more emphasis is made on a determined size fraction, which indefectibly leads to discrepancies in the conductivity distribution. thus, defining the reliability of the determinations remains problematic. although empirical formulas gave just approximate values of hydraulic conductivity and results must be interpreted with caution, they are consistent with the hypothesis of an anisotropic medium with the major conductivity tensor oriented along the horizontal plane, and the minor direction along the vertical. hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan 147 figure 4. distribution of the coefficient of uniformity (cu) for the whole dataset figure 5. profiles of hydraulic conductivity at some of the wells. variance in parenthesis. gallardo and marui. esrj vol. 11, no. 2. december 2007 148 figure 6. distribution of hydraulic conductivity for the entire dataset figure 7. electrical conductivity near the freshwater/seawater interface. hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan 149 as summarized by shepherd (1989), the correlation of permeability and grain size results in a power expression of the form y = axb (5) where (y) is permeability when plotted against grain size (x), the coefficient a is the value of y at x = 1, and b is the slope of the line that is fitted to the data. in contrast to other formulas, this equation does not depend on the temperature of the medium (vrbka et al., 1999). in an attempt to improve predictions, hydraulic conductivity measured at each well was plotted against grain size to find an empirical correlation between both parameters. the best fit was calculated by the least-squares regression. from the nine sets of data only two yielded acceptable values of r-squared in coincidence with the coarse-grain dominated shallow wells (table 2). the calculated slopes were in agreement with other reports (stearns, 1929, rose and smith, 1957) although they were somewhat low, which may be an indication of texturally immature sediments and miscellaneous samples from different environmental populations (shepherd, 1989). no significant relation was seen for the rest of the datasets, possibly due to the mixture of distinct grain sizes. however, a strong correlation has been reported for a subset of samples at well 4-1 (sankoh consultants, 2004) suggesting the theoretical expectations might be confirmed within certain facies of the soil profile. the hydraulic conductivity of granular deposits is also dependant upon porosity (morin, 2006). therefore, the existence of any trend was investigated for each well dataset by the pearson correlation-coefficient. the null hypothesis (there is no relationship) was rejected at the 1% level of significance at three wells, and at the 2% level for one more borehole (table 3). in all cases, the relationship was negative, suggesting that the permeability of the soil would decrease with an increase in the fraction of fines. despite that it is reasonable to have a negative correlation under certain grain-size distributions and packing arrangements (beard and weyl, 1973; morin, 2006), it was not possible to find a consistent pattern all over the vertical cross section. measurements displayed variable curves, and while the relationship was visible at some points, there seemed to be no arrangement at other depths. it is also feasible that migration of colloidal particles could significantly affect the permeability of the porous medium at depth (khilar et al., 1983). in effect, the diameter of colloidal matter has been found to increase with depth in association with the higher groundwater salinity (düker and ledin, 1998). this phenomenon could contribute to clog the aquifers causing an important reduction in permeability. changes are likely to be accompanied by reductions in porosity as well, which may explain the occasional deviations in the negative correlation. it can be concluded that there may be numerous causes for the changes in conductivity in the area, and while the negative trend between conductivity and porosity is significant at many units of the vertical profile, this correlation cannot be generalized to the whole sequence because it is not systematic. the variability of soil properties is so complex that only parts of the structure can be described in a deterministic way, while the rest probably have to be seen as random realizations (heuvelink and webster, 2001; herbst et al., 2005). 4.4 profiles of electrical conductivity gallardo and marui. esrj vol. 11, no. 2. december 2007 changes in the physical properties of soils may modify the rates and chemical composition of water entering the well. figure 7 shows the logs of electrical conductivity (ec) for two wells near the seawater/freshwater interface (marui, 2005). values increased with depth, with a marked change in concentrations at 100 and 160 m in well 0402, and 170 m for well 0401. shallow waters would relate to freshwaters of low ec, while high conductance at depth would be associated either with fossil waters or seawater intrusion. although not shown here, data collected from other wells indicated the lowest salinities occur in the upper 40 m below the ground, where alluvial deposits of higher hydraulic conductivity promote a rapid inflow of meteoric waters and the existence of a freshwater aquifer. in this way, a preferential path for submarine groundwater discharge has been reported at a depth of 30 m (itoh et al., 2006). geology may also explain the changes in ec at a depth of 100 m in well 0402, where an increase in ions concentration coincided with the intercalation of a thin layer of tuffs. unlike the upper aquifer, flow would occur in the opposite direction, toward the land. thus, the change in water chemistry would reflect an increase in na+ and clconcentrations associated with seawater intrusion. at other depths, water quality compared less favorably with physical parameters. density-driven flow can be expected to occur along preferential flow paths surrounded by stagnant zones (schincariol, 1998). however, mixing of water inside wells could have altered the quality of water at different depths. it is possible that seawater flowing through favorable sediments sank due to higher density when entering the wells, which would explain why peaks of salinity occur scattered in the profile. on the other hand, the effect of the wells screen may be another factor obscuring changes in conductance as a function of hydraulic properties. in effect, waters from different depths may enter long-screened wells, so it is not possible to know if poor-quality water is from deeper or shallower sources (izbicki et al., 2005). thus, even though ec logs can be considered a quick and inexpensive tool to delineate major changes of hydraulic properties, the method can not be preferred over direct measurements of soil physical parameters, at least without a multilevelwell system screened at short intervals all over the profile. 5. summary and conclusions drill cores from 9 wells were analyzed to a depth of 200 m at the coastal area of j-parc, and the main lithological facies and hydraulic properties of the sedimentary sequence determined. deposits are strongly heterogeneous, although two main hydrogeological units could be recognized: quaternary sand and gravels constitute an unconfined aquifer with a thickness up to 55 m at the top of the profile, while silty sediments from the tertiary define the lower part of the section. embedded throughout the sequence, there are a number of horizontal layers with variable thickness and composition which define a multi-aquifer system. as expected, the high variability in the sedimentary structure and the inevitable mixing of samples with different origins was associated with important irregularities in the parameters distribution. basic properties as hydraulic conductivity, porosity, and grain-size presented complex patterns difficult to be distinguished by laboratory determinations alone and therefore, results were complemented with statistical approaches to provide a more rigorous interpretation of the field conditions. since soil properties not always 150 hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan table 1. ratio of hydraulic conductivity calculated by empirical formulas (kh) in relation to core measurements (kv) for 317 samples table 2. power relation between grain size and hydraulic conductivity for three of the wells table 3. correlation of porosity to hydraulic conductivity for the analyzed wells 151 gallardo and marui. esrj vol. 11, no. 2. december 2007 presented a systematic relationship, parts of the spatial structure would better be explained as random distributions. the hydraulic conductivity was described by a lognormal distribution. results from empirical calculations and laboratory tests differed in up to three orders of magnitude, leading to the assumption that conductivity in the horizontal direction is appreciably higher than in the vertical. the approach of shepherd was also applied to improve predictions. the regressions provide a simple relationship between grain size and conductivity although good estimates were limited to just two wells, probably in coincidence with texturally-immature sediments. a negative trend between porosity and hydraulic conductivity was inferred at four wells. it is intuitive to think about a decrease of permeability with an increase in the fraction of fines at depth however, this relationship was not constant and therefore, it is likely that other processes as colloidal migration, surface weathering or soil compaction may be responsible for the changes in the parameters distribution. acknowledgements the authors are very grateful to dr. roger morin, usgs, for providing essential material to accomplish this work. special thanks are extended to dr. diego tanaka, school of system and information engineering, university of tsukuba, for his advises on statistics. finally, the authors want to thank the anonymous reviewers, whose comments helped improve the manuscript. references ahuja, l.r., cassel, d.k., bruce, r.r., and b.b. barnes (1989). evaluation of spatial distribution of hydraulic conductivity using effective porosity data. soil science 148 (6), 404-411. aimrun w., amin, m.s.m., and s.m. eltaib (2004). effective porosity of paddy soils as an estimation of its saturated hydraulic conductivity. geoderma 121, 197-203. bear, j (1972). dynamics of fluids in porous media. dover pubs. new york, 784 pp. beard, d.c., and weyl (1973). influence of texture on porosity and permeability of unconsolidated sand. am assoc petrol geol bull 57 (2), 349-369. boggs jr, s (1987). principles of sedimentology and stratigraphy. merril publishing co. usa. 800 pp. bridge, j.s (1981). hydraulic interpretation of grain-size distributions using a physical model for bedload transport. j sed petrol 51, 1109-1124. creager, w.p., justin, j.d.ɸand j, hinds (1945). engineering for dams vol iii, earth, rock-fill, steel and timber dams. john wiley, new york. de craen, m., van geet, m., wang, l., and m. put (2004). high sulphate concentrations in squeezed boom clay pore water: evidence of oxidation of clay cores. phys chem earth 29, 91-103. düker, a., and a. ledin (1998). properties of the particulate phase in deep saline groundwater in laxemar, sweden. wat res 32 (1), 186-192. fairhurst, c (2004). nuclear waste disposal and rock mechanics: contributions of the underground research laboratory (url), pinawa, manitoba, canada. j rock mech & mining sci 41, 12211227. folk, r.l (1974). petrology of sedimentary rocks. hemphill, austin, texas folk, r.l, and w.c. ward (1957). brazos river bar: a study in the significance of grain-size parameters. j sed petrol 27, 3-26. 152 hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan freeze, r.a (1975). a stochastic-conceptual analysis of one-dimensional groundwater flow in nonuniform homogeneous media. water resour res 11, 725-742. freeze r.a., and j.a. cherry (1979). groundwater. prentice-hall, inc., englewood cliffs, nj. 604 pp. friedman, g.m., sanders, j.e., and d.c. kopaskamerkel (1992). principles of sedimentary deposits. macmillan coll. div. usa. 717 pp. gallardo, a.h (2006). analysis of groundwater flow at the local scale based on the grid refinement method. geological survey of japan, geobarrier group, technical report (in japanese). gimenez, d., rawls, w.j., and j.g. lauren (1999). scaling properties of saturated hydraulic conductivity in soil. geoderma 88, 205-220. herbst, m., diekkrüger, b., and h. vereecken (2006). geostatistical co-regionalization of soil hydraulic properties in a micro-scale catchment using terrain attributes. geoderma 132 (1-2), 206221. heuvelink, g.m.b., and r. webster (2001). modelling soil variation: past, present, and future. geoderma 100, 269-301. itoh, n., marui, a., miyakoshi, a., and a.h. gallardo (2006). evaluation of the shape and movement of seawater/freshwater interface by multipoint time-series observations. spring meeting of the japanese assoc. of groundwater hydrol, tokyo, japan, may, 72-73. (in japanese) izbicki, a., christensen, a.h., newhouse, m.w., smith, g.a., and r.t. hanson (2005). temporal changes in the vertical distribution of flow and chloride in deep wells. ground water 43 (4), 531544. jussel, p., stauffer, f., and t. dracos (1994). transport modeling in heterogeneous aquifers: 1. statistical description and numerical generation of gravel deposits. water res res 30 (6): 1803-1817. kasenow, m (2002). determination of hydraulic conductivity from grain size analysis. water resources publications, highlands ranch. colorado. khilar, c.k., fogler, h.s., and j.s ahluwalia (1983). sandstone water sensitivity: existence of a critical rate of salinity decrease for particle capture. chem eng sci 38 (5): 789-800. lambe, t.w (1951). soil testing for engineers. john wiley & sons inc. 165 pp. marui, a (2005). survey on the conditions and shape of the freshwater/saltwater interface. geological survey of japan, geo-barrier group, technical report (in japanese). mbagwu, j.s.c (1995). saturated hydraulic conductivity in relation to physical properties of soils in the nsukka plains, southeastern nigeria. geoderma 68, 51-66. morin, r.h (2006). negative correlation between porosity and hydraulic conductivity in sand-andgravel aquifers at cape cod, massachusetts, usa. j hydrol 316 (1-4), -52. muskat, m (1937). the flow of homogeneous fluids through porous media, mc graw-hill, new york pryor, w.a (1973). permeability-porosity patterns and variations in some holocene sand bodies. am assoc petrol geol bull 57 (1), 162-189. regalado, c.m., and r. muñoz-carpena (2004). estimating the saturated hydraulic conductivity in a spatially variable soil with different permeameters: a stochastic kozeny-carman relation. soil & tillage research 77, 189-202. rose, h.g., and h.f. smith (1957). a method for determining permeability and specific capacity from effective 153 gallardo and marui. esrj vol. 11, no. 2. december 2007 grain size. illinois state water survey circular no. 59. sakamoto, t., tanaka, k., soya, t., noma, t., and k. matsuno (1972). geology of the nakaminato district. geological survey of japan, quadrangle series, tokyo (8), no. 11. sankoh consultants (2004). comparison experiments on core quality according to boring construction. geological survey of japan, geo-barrier group, technical report (in japanese). schincariol, r.a (1998). dispersive mixing dynamics of dense miscible plumes: natural perturbation initiation by local-scale heterogeneities. j contam hydrol 34, 247-271. shepherd, r.g (1989). correlations of permeability and grain size. ground water 27 (5), 633-639. stearns, n.d (1929). laboratory tests on physical properties of water-bearing materials. us geol surv water supply paper 596, 121-176. sudicky, e (1986). a natural gradient experiment on solute transport in a sand aquifer: spatial variability of hydraulic conductivity and its role in the dispersion process. water resour res 22 (13), 2069-2082. takai, f., and r. tsuchi (1963). the quaternary. in: takai, f., matsumoto, t., and r. toriyama (eds), geology of japan. the university of tokyo press, pp 173-196. takeda, s., and k. komiya (2006) from air bubble boring: the hybrid method to a core sampling system of high quality. chiban kougaku journal 54 (4), pp 16-18 (in japanese). turcke, m.a., and b.h. kueper (1996). geostatistical analysis of the borden aquifer hydraulic conductivity field. j hydrol 178, 223-240. vrbka, p., ojo, o.j., and h. gebhardt (1999). hydraulic characteristics of the maastrichtian sedimentary rocks of the southeastern bida basin, central nigeria. j african earth sci 29 (4), 659-667. vukovic, m., and a. soro (1992). determination of hydraulic conductivity of porous media from grain-size composition. water resources publications, littleton, colorado 154 hydraulic characteristics of sedimentary deposits at the j-parc proton-accelerator, japan the band-e narges deposit is located about 70 km northeast of the city of badrud, northern isfahan province. band-e narges ore deposit is mining for magnetite. to release valuable minerals, crushing and grinding implemented for separation ore from the gangue. magnetic separation and flotation methods for upgrading magnetite iron ore were carried out in different experimental conditions with varied parameters. the particle size of the initial content was 74 microns for flotation, and 150 microns for magnetic separation. the initial samples, with the iron grade of 43.4% and sulfur of 1.9%, are individually subjected to upgrading by floatation and magnetic separation during which the affecting parameters for each method were optimized. the improvement in the optimal condition for magnetic separation culminated in 60.85% for the iron grade, 85.21% iron for recovery and 1.08% for sulfur content. the upgrading by floatation in the optimal mode produced 60.02% iron grade, 80.41% iron recovery and 0.95% sulfur content. to determine the best method for the pre-concentration stage of ore, the content gained from each technique passed reclining for grad improvement. the final content obtained from the magnetic separation of was undergone the floatation test yielded to a content with 64.3% iron grade, 77.15% iron recovery and 0.7% sulfur content. the use of magnetic separation as a pre-concentration stage for floatation method is proposed as an economical method for improving the grade of the iron and reduce the sulfur content and to avoid the high cost of grinding, which is costly part of processing procedures. abstract keywords: magnetite iron ore; floatation; magnetic separation; reduction of processing costs. characterization of band-e narges magnetite iron ore for mineral processing issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v22n2.73438 amir pazoki1*, reza rashidi-khabir1, reza jahanian2, ali pourbahaadini2 1assistant professor, mine engineering department, lorestan university, khorramabad, iran *e-mail: pazoki.a@lu.ac.ir 2ms student, mine engineering department, lorestan university, khorramabad, iran earth sciences research journal earth sci. res. j. vol. 22, no. 2 (june, 2018): 145-148 el yacimiento de hierro band-e narges se ubica a 70 kilómetros al noroeste de la ciudad de bad rud, en la norteña provincia de isfahan, irán. en este yacimiento se extrae magnetita. para la extracción de minerales valiosos, se utiliza la separación por molienda y triturado desde la ganga. se utilizaron los métodos de separación magnética y flotación con diferentes parámetros para la recuperación de la magnetita. el tamaño de la partícula inicial fue de 74 micrones en flotación y de 150 micrones en la separación magnética. las muestras iniciales, con un 43.4 % de hierro y 1.9 % de súlfuro, fueron sujetas individualmente al mejoramiento por flotación y separación magnética, proceso donde se optimizaron los parámetros afectados por cada método. el mejoramiento por la separación magnetica registró un 60.85 % de hierro, mientras que por recuperación fue 85.21 % de hierro y 1.08 % de contenido sulfúrico. el mejoramiento por flotación en condiciones óptimas produjo un 60.02 % de hierro, con un 80.41 % de hierro recuperado y un 0.95 % de contenido sulfúrico. para determinar el mejor método en la etapa de preconcentración de la mena, la ganancia de cada técnica estuvo reclinada para mejorar el contenido. el contenido final obtenido de la separación magnética se sometió a la prueba de flotación con un 64,3 % de hierro, con el 77.15 % de recuperación de hierro, y un 0.7 % de contenido sulfúrico. el uso de la separación magnética como una etapa de pre-concentración para el método de flotación se propone como un medio económico para mejorar el grado del hierro y reducir el contenido sulfúrico además de evitar el alto costo de la molienda, que es la parte más costosa de los procedimientos. resumen palabras clave: yacimiento de hierro; flotación; separación magnética; reducción de los costos de producción. record manuscript received: 18/01/2018 accepted for publication: 13/06/2018 how to cite item pazoki, a., rashidi-khabir, r., jahanian, r., & pourbahaadini, a. (2018). characterization of band-e narges magnetite iron ore for mineral processing. earth sciences research journal, 22(2), 145-148. doi: http://dx.doi.org/10.15446/esrj.v22n2.73438 o r e g e o l o g y caracterización del yacimiento de hierro band-e narges (irán) para el procesamiento mineral 146 amir pazoki, reza rashidi-khabir, reza jahanian, ali pourbahaadini introduction methods available for upgrading iron or aree physical (magnetic), physicochemical (floatation) or chemical methods, amongst which the physical methods are the most cost effective (rezaei, 1999). when the particle size of the impurities is large, upgrading is possible in larger dimensions and the cost will decrease. rock upgradation ability and upgrading costs are important factors in the evaluation of iron ore (cohen, 1986). the type of processing and its stages depend on the ore and its complexity. in general, magnetite iron ore upgrades initially include two or three stages of crushing and dry magnetic separation (svoboda and fujita, 2003). for mechanical floatation to function effectively the energy should be sufficient to meet the requirements of floatation; therefore, optimal energy use is critical. some of the energy consumed is transmitted across the cell as kinetic energy available in the vortices of various sizes, but most of it is lost through micro-interactions between the water, air and solid phases (schubert, 1999; shahbazi et al., 2015; schubert and bischofberger, 1978; ahmed and jameson, 1985; deglon et al., 1998; pyke et al., 2003; newell and grano, 2006; amini et al., 2013). after grinding to release the minerals, the valuable minerals are separated from the gangue. floatation is a commonly used separation process (chau, 2009; vaqar, 1997). magnetic separation is a physical and low-cost upgradation method for minerals having magnetic properties. the use of flotation depends on the surface characteristics of the minerals. because rock is composed of different minerals, different methods must be employed to process them according to their economic viability and in terms of the percentage of recovery and improvement in the grade. the current study was undertaken to determine the optimal state of the parameters affecting floatation and magnetic separation and the sequence of use of each method is investigated. research method the study area is the band-e narges iron ore mine located in northern isfahan province about 70 km northeast of the city of badrud. for analysis of the use of floatation and magnetic separation, the samples were crushed by jaw, cylindrical and conical crushers into dimensions of less than 2 mm. the product was then sieved with 18, 30, 50, 80, 100, 120, 150, 170, 200 and 230 mesh. representative samples were collected from each sieve class to make a smooth section. analysis of the smooth sections allowed determination of the degrees of freedom (dof) and the dispersion of mineral particles. testing was then conducted for different sizes and intensities to determine the parameters affecting magnetic and floatation separation. after determining the optimum parameters and obtaining concentrates using each method, the floatation concentrate was further processed by magnetic separation and the primary magnetic separation concentrate was further processed by floatation and the results were compared in terms of grade, recovery and sulfur content. discussion degree of freedom economic recovery of valuable minerals depends on several factors. grind ability is the most important characteristics determining the grinding mechanism, which includes crushing and smelting. one goal of these processes is the release of valuable minerals from waste material in the largest possible particle size. for present study, the samples were crushed using jaw and conical crushers to dimensions of less than 2 mm and were then run through mesh of different sizes (18, 30, 50, 80, 100, 120, 150, 170, 200, and 230). from each size of sieve, samples were collected to determine the dof and prepare smooth sections. analysis of the smooth sections showed that the particles from the 120-mesh sieve class (125-150 micron) featured the required dof of d80. analysis of the smooth sections showed that the magnetite from the 100 mesh at 150 micron had the highest amount of release (d80); therefore, this range of granulation was used for processing. the mineral ore must be upgraded after grinding in accordance with its mineralogical characteristics. however, in smooth sections at this range of granulation, there is contact between the magnet and disturbing minerals; thus, the processing orbit appropriate to the type and characteristics of minerals should be carefully selected. if the processing orbit is designed precisely, the cost and time of the process will be minimized at the operational stages. because the samples tested in this study were identified to be magnetite, a magnetic separation technique was used to upgrade it as the most cost-and time-efficient method. the magnetic properties of different materials generally depend on the amount of electrical charge and the magnetic torque (spin) of the electrons. the motion of an electron cloud around an atomic nucleus in the form of opposing orbitals creates a magnetic dipole torque. the sum of the magnetic torque generated by the spins when moving around the nucleus and the magnetic dipole torque caused by an external magnetic field gives the induction dipole torque (skoog et al., 2013). davis tube testing was used to determine the optimal dof with the highest grade at the best field intensity. it was concluded that the optimal dof for this ore occurred at a mesh size of +150, although better grades of fractions were produced at +200 and +230. these latter sizes required more grinding and more expense although the grades obtained at +150, +200 and +230 did not differ significantly. to reduce the cost of grinding and in accordance with the results of davis tube testing, the optimum dof for the ore was chosen to be that for the +150 mesh. dry magnetic separation tests particles with the dimensions of 0 to 150 micron produced the highest dof and particles with 1000 to 2000 micron in size with an intensity of 2000 gauss had the highest percentage of recovery. the difference in recovery was not very high for these two granulation ranges. in terms of grade, the grade of the particles 0 to 150 micron in size was significantly higher for those 1000 to 2000 micron in size; thus, the optimal size and field intensity were selected to be 0 to 150 micron and 2000 gauss, respectively. the grade of magnetite considered suitable for the industry is more than 60%. after one stage of magnetic separation, a grade of 54.74% was obtained with particles of 0 to 150 micron in size; thus, more than one stage of magnetic separation was required to reach the appropriate grade for the industry. because the optimal dimensions were 0 to 150 micron and optimal magnetic field intensity was 2000 gauss, further other magnetic separation was carried out for these dimensions and field intensity. in order to increase the grade, the concentrate obtained from the magnetic separation of particles 0 to 150 micron in size with a magnetic field intensity of 2000 gauss was run through two more magnetic separation stages. the results are listed in table 1. table 1. results of stages 2 and 3 of magnetic separation of 0 to 150 μm particles after three stages of magnetic separation at a 2000-gauss field intensity on particles of 0-150 micron, a product with a grade of 60.85% and a recovery of 85.21% was obtained that is usable in the industry. the sulfur content of this product was 1.08%, so chemical methods such as floatation had to be used to reduce the sulfur content. use of floatation on magnetic separation concentrate since after three stages of magnetic separation, the sulfur content was 1.08%, the concentrate was subjecting to a next grounding with 147characterization of band-e narges magnetite iron ore for mineral processing table 2. results of floatation on magnetic separation concentrate table 3. steps of magnetic separation on floatation concentrate table 4. final specifications after magnetic separation on floatation concentratea mill to reach to gain size of less than 74 micron. the flotation was carried out in three stages to reach a concentrate with an iron grade of 67.9%, recovery of 76.39% and sulfur content of 0.48%. the results are shown in table 2. as seen, after floatation on the magnetic separation concentrate, the amount of sulfur decreased to 0.48% while the iron content increased. use of magnetic separation on floatation concentrate when the floatation parameters were optimized, five stages of floatation were performed and a product with a grade of 60.02%, a recovery of 80.41% and a sulfur content of 0.95% was obtained. the concentrate was next subjected to magnetic separation. because floatation is performed only on particles of less than 74 micron, magnetic separation could only be performed on particles of that size. it was possible, however, to examine the intensity of the field at 1000, 1500 and 2000 gauss. it was determined that, for a floatation concentrate of less than 74 micron in size, the highest grade was produced at 1000 gauss; thus, three stages of magnetic separation were performed on the concentrate at 1000 gauss. the results are shown in table 3. after three stages of magnetic separation a product was attained with a grade of 64.3% and a final recovery of 77.15% iron and a sulfur content of 0.74% (table 4). conclusion the type and complexity of ore determines the type of processing used. in addition, factors such as the simplicity, cost-effectiveness, speed and level of environmental pollution of the various processes must be taken into account. the results of this study showed that the use of the magnetic separation method as a pre-concentration method for floatation is more suitable than using the flotation method as a pre-concentration method for magnetic separation. the following conclusions were drawn. 1. the most appropriate dimensions for magnetic separation are equivalent to a dof of 150 micron, while the proper dimensions for floatation are 74 micron. this indicates that the use of magnetic separation before floatation will reduce the cost of grinding. 2. the optimal magnetic field intensity for magnetic separation when it is used for pre-concentration was 2000 gauss for a particle size of up to 150 micron. when flotation was used as a pre-concentration before magnetic separation, the optimal field intensity was 1000 gauss because the particle size was a maximum of 74 micron. this occurred because, when small particles are subjected to high field intensities, they create magnetic masses from the soft particles created. these masses trap waste and reduce the grade at high intensity; thus, the intensity of the field is reduced. 3. the use of floatation method as pre-concentration for magnetic separation required a total of five stages. this increases the cost of chemicals consumed, the difficulty of the process and amount of environmental pollution produced. 4. the use of magnetic separation before floatation for preconcentration creates a product with a higher grade, recovery and sulfur content. the high sulfur content should be reduced by chemical methods such as floatation. 5. when magnetic separation is used as pre-concentration, its product has a higher grade and a lower sulfur content, which is a drawback. although the recovery was less, its high grade and low sulfur content justifies this approach. references ahmed, n., jameson, g. j. (1985). the effect of bubble size on the rate of flotation of fine particles. international journal of mineral processing, 14(3), 195-215.‏ amini, e., bradshaw, d. j., finch, j. a., brennan, m. (2013). influence of turbulence kinetic energy on bubble size in different scale flotation cells. minerals engineering, 45, 146-150.‏ chau, t. t. (2009). a review of techniques for measurement of contact angles and their applicability on mineral surfaces. minerals engineering, 22(3), 213-219.‏ cohen, h. e. (1986). magnetic separation. in mineral processing at a crossroads (pp. 287-315). springer, dordrecht. deglon, d. a., o'connor, c. t., pandit, a. b. (1998). efficacy of a spinning disc as a bubble break-up device. chemical engineering science, 53(1), 59-70.‏ newell, r., grano, s. (2006). hydrodynamics and scale up in rushton turbine flotation cells: part 2. flotation scale-up for laboratory and pilot cells. international journal of mineral processing, ‏.65-78 ,(2)81 pyke, b., fornasiero, d., ralston, j. (2003). bubble particle 148 amir pazoki, reza rashidi-khabir, reza jahanian, ali pourbahaadini heterocoagulation under turbulent conditions. journal of colloid and interface science, 265(1), 141-151.‏ rezaei, b. (1999). mineral processing technology (magnetic upgradation). amir kabir university of technology publishing center. first edition. schubert, h. (1999). on the turbulence-controlled microprocesses in flotation machines. international journal of mineral processing, ‏.257-276 ,(1-4)56 schubert, h., bischofberger, c. (1978). on the hydrodynamics of flotation machines. international journal of mineral processing, 5(2), 131‏.142 shahbazi, b., rezai, b., koleini, s. m. j., noaparast, m. (2015). estimation of gas holdup and input power in froth flotation using artificial neural network. iranian journal of materials science and engineering, 12, 12-20.‏ skoog, d. a., west, d. m., holler, f. j., crouch, s. (2013). fundamentals of analytical chemistry. nelson education.‏ svoboda, j., fujita, t. (2003). recent developments in magnetic methods of material separation. minerals engineering, 16(9), 785-792.‏ vaqar, r. (1997). preparation of the loads of iron and steel furnaces, tehran university press. rainfall intensity-duration-frequency (idf) curves are of particular importance in water resources management, for example, in urban hydrology, for the design of hydraulic structures and the estimation of the flash flood risk in small catchments. idf curves describe rainfall intensity as a function of duration and return period, and they are significant for water resources planning, as well as for the design of hydraulic constructions and structures. in this study, scaling properties of extreme rainfall are examined to establish the scaling behavior of statistical non-central moment over different durations. idf curves and equations are set up for all stations by using the parameter obtained from scaling behavior, the location and scale parameters μ24 and σ24 of the gumbel distribution (evi) sample of annual maximum 1440 min rainfall data. in another hand, we have established the idf curves for ten selected rain gauge stations in the northern (oueme valley) parts of benin republic, west africa by using the simple scaling approach. analysis of rainfall intensities (5 min and 1440 min rainfall data) from the ten rainfall stations shows that rainfall in north-benin displays scales invariance property from 5 min to 1440 min. for time scaling, the statistical properties of rainfall follow the hypothesis of simple scaling. therefore, the simple scaling model applies to the rainfall in (oueme valley). hence, the simple scaling model is thought to be a viable approach to estimate idf curves of hourly and sub-hourly rainfall form rainfall projections. the obtained scaling exponents are less than 1 and range from 0.23 to 0.59. the empirical model shows that the scaling procedure is a good estimator as it is more efficient and gives more accurate estimates compared with the observed rainfall than the traditional method which only consists the gumbel model in all stations for lower return periods (t<5 years) but not for higher return periods. las curvas de precipitación intensidad-duración-frecuencia (idf) son de particular importancia en el manejo de los recursos hídricos, como es el caso de la hidrología urbana o para el diseño de estructuras hidráulicas y la estimación del riesgo de crecidas en pequeñas captaciones. las curvas idf describen la intensidad de las precipitaciones como una función con períodos de duración y recurrencia, lo que las hace significativas en la planeación de recursos hídricos así como en el diseño de construcciones y estructuras hidráulicas. este estudio examina las propiedades de escala en precipitaciones extremas para establecer un comportamiento en momentos estadísticos marginales en diferentes períodos de duración. se establecieron las curvas idf y las ecuaciones para todas las estaciones a partir del parámetro obtenido del comportamiento de escala, la ubicación y los parámetros de escala μ24 and σ24 de la muestra de información de precipitación máxima anual de 1440 minutos de la distribución de gumbel (evi). por otro lado, se establecieron las curvas idf para 10 estaciones pluviométricas seleccionadas en el valle oueme, al norte de la república de benín (áfrica occidental), con el uso de aproximación simple de escala. el análisis de las intensidades de precipitación en las diez estaciones pluviométricas muestra que la precipitación en el norte de benín expone propiedades de poca variación en la escala 5 min y 1440. en el tiempo de escala, las propiedades estadísticas de precipitación confirman la hipótesis de escala simple; además, este modelo so corresponde a la precipitación del valle oueme. por lo tanto, el modelo de escala simple se considera una aproximación viable para estimar las curvas idf en las proyecciones de precipitación de cada hora y sub-hora. los exponentes de escala obtenidos son menores a 1 y oscilan de 0,23 a 0,59. el modelo empírico muestra que el procedimiento de escala es un buen estimativo, más eficiente y con cálculos más exactos que el método tradicional, el cual consiste solamente en el modelo gumbel aplicado en todas las estaciones pluviométricas en períodos de menor recurrencia (t<5 años) pero no en lapsos de mayor recurrencia. abstract resumen keywords: simple scaling, idf curves, annual maximum rainfall, frequency analysis, oueme valley. palabras clave: escala simple, curvas de intensidadduración-frecuencia, precipitación máxima anual, análisis de frecuencia, valle oueme. estimation of idf curves of extreme rainfall by simple scaling in northern oueme valley, benin republic (west africa) record manuscript received: 02/03/2015 accepted for publication: 08/01/2016 how to cite item agbazo, m., koton'gobi, g., kounouhewa, e., alamou, e. & afouda, a. (2016). estimation of idf curves of extreme rainfall by simple scaling in northern oueme valley, benin republic (west africa). earth sciences research journal, 20(1), d1-d7. doi: http://dx.doi.org/10.15446/esrj.v20n1.49405 issn 1794-6190 e-issn 2339-3459 http://dx.doi.org/10.15446/esrj.v20n1.49405 estimación de las curvas idf de extrema precipitación por escala simple en el valle oueme, al norte de la república de benín (africa occidental) m e t e o r o l o g y earth sciences research journal earth sci. res. j. vol. 20, no. 1 (march, 2016): d1 d7. m.agbazo1, g. koton’gobi2, b. kounouhewa2, e. alamou3, a. afouda3 1 international chair in physics mathematics and applications (cipma-chair unesco), university of abomey-calavi, benin bp: 526 uac benin 2 radiation physics laboratory (lpr), university of abomey-calavi, bp: 526 uac benin 3 applied hydrology laboratory (lha), university of abomey-calavi, bp: 526 uac corresponding author: m.agbazo; e-mail: agbazomedard@yahoo.fr d2 m.agbazo, g. koton’gobi, b. kounouhewa, e. alamou and a. afouda 1. introduction west african countries have been struck these last years by floods and inundations of unprecedented magnitude (danida, 2008; zannou, 2011; g. panthou, 2014). for example we can cite the inundations that occurred from 2007-2009 in benin, niger and burkina-faso (world bank, 2009) which caused drastic consequences in more field of development agriculture (destroyed cultivation), aquatic ecosystems, economy, urban drainage systems, energy distribution and social security by displacing people (danida, 2008, descroix et al., 2009). likewise, the emblematic niger river floods of 2010 and 2012 are the two most important floods ever recorded at the niamey station since the beginning of the observation in 1929, causing heavy casualties to people living close to the river (panthou, 2014). projection in the northwest of benin, one of the high-risk regions, predicts trends including lower average annual rainfall (15% up to 2025); more variation in precipitation; more frequent extreme climatic elements such as heavier rains and more powerful storms; shorter, delayed monsoons with irregular precipitation; and higher long-term average temperatures (danida, 2008). this illustrates that extreme events can appear in dry or moderately wet years anywhere in west africa where benin republic is located and in which wet years can occur without any extreme rainy event being recorded. in urban drainage systems, short duration rainfall events are one of the most critical elements for hydrological investigation (bougadis and adamoski, 2006), useful for the control of storm run-off such as flood detention, reservoirs, sewer systems etc. in particular, rainfall extremes with high temporal resolution (one hour or shorter) are necessary for the design of drainage systems, usually characterized by fast response. the temporal resolution of rainfall data usually available for practical applications are often lower than the data requested for the design procedures or mathematical models application that greatly affect their reliability (nhat et al, 2008) because they depend on many assumptions such as the distribution selection for durations. rainfall data generally require a large number of parameters and are time dependent. such information is often expressed in intensityduration-frequency (idf). the establishment of such relationships goes back to the 1930s (bernard, 1932). since then, different forms of idfs have been constructed for several regions of the world. in africa we can cite the studies which utilized statistical method to construct idf curves: in congo (mohymont et al., 2004; mohymont & demarée, 2006; van de vyver & demarée, 2010), in ghana (endreny & imbeah, 2009), in cote d’ivoire (soro et al., 2008; soro et al., 2010) and in nigéria (oyegoke & oyebande, 2008; okonkwo & mbajiorgu, 2010). traditionally the idf curves are constructed by performing statistical analysis on annual maxima series (ams) or partial duration series (pds) by fitting probability distribution for several pre-selected rainfall durations (bougadis and adamowski, 2006). such a procedure has several disadvantages including fitting distribution and parameters estimations for each duration, and requirements to extrapolate result to different time-scales. the usefulness and accuracy of such procedures are limited because of fitting uncertainties and lack of ability to adequately describe rainfall properties at different time scales (van nguyen and wang, 1996). in addition, this traditional method has certain limitations. for example, a high number of parameters are involved, which makes it none-parsimonious from the statistical point of view (baghirathan and shaw, 1978; bell, 1969; sherman, 1931; bernard, 1932; chen, 1983; garcia-bartual and schneider, 2001; takara, 2005; nhat et al., 2008). traditional method cannot take into consideration characteristics of rainfall for different durations; and it is based on the annual extreme data available at local site only. it is therefore advantageous to develop a model, which could adequately describe rainfall characteristics through a continuum of times scales including inferences at time resolutions that may not have been observed and to reduce the number of parameters to be estimated in order to increase their reliability. consequently, alternative approaches to construct idf relationships are used. for example the fractal properties of rainfall which implies scaling invariance were developed (e.g., menabde, et al., 1999) where the scaling properties of the statistical moments of rainfall in south africa over a range of durations from 30 min to 1440 min were examined. yu, et al., (2004) used regional idf formulas for estimating scaling exponents for 46 non-recording rain gauge stations in taiwan. molnar and burlando (2005) investigated the variability of the scaling properties from 62 rain gauge stations in switzerland and özger, et al. (2010) focused on the scaling properties of rainfall at 43 rain gauge stations in texas. recently, scaling formulas are proposed to extend the idf relationships from daily time scale to shorter time intervals based on scaling properties. gupta and waymire (1990) studied the concepts of simple and multiple scaling to characterize the probabilistic structure of the precipitation process, koutsoyiannis and foulfoula-georgiu (1993) used a scaling model to predict storm hyetographs. menabde et al. (1999) showed that based on the empirically observed scaling properties of rainfall and some general assumptions about the cumulative distribution function for the annual maxima of mean rainfall intensity, it is possible to derive simple idf relationships. bendjoudi, et al. (1997) used a multifractal point of view on rainfall idf curves. rosso and burlando (1990) and later burlando and rosso (1996) used this concept to study traditional forms of depth-durationfrequency relationships. de michele et al. (2002) developed idf curves design for storms in milan (italia); yu, et al. (2004) developed regional idf formulas for non-recording sites in taiwan, molnar and burlando (2005) examined the variability of scaling exponents in alpine mountainous region of switzerland; nhat, et al. (2007) developed regional relationship for ungauged locations based on the scaling theory in japan. molnar and burlando (2008) investigated the variability of the scaling properties from 62 rain gauge stations in switzerland, acar, et al. (2008) used a multilayer perception artificial neural network model to assess idf relationships for short duration rainfalls. recently, özger, et al. (2010) focused on the scaling properties of rainfall at 43 rain gauge stations in texas; ceresseti (2011) used a multifractal point of view on rainfall idf curves in france to characterize idf curves; ghanmi (2014) developed regional relationship for ungauged locations based on the scaling theory in tunisia. scaling properties of extreme rainfall has never been investigated in benin republic. the objectives of the paper in this study area are: (i) to examine the scaling properties of extreme rainfall in the oueme valley (benin), (ii) to establish scaling behavior of statistical non central moment over different durations of rainfall intensity and (iii) to derive the idf relationships for short-duration rainfall from daily data using the scaling methodology and to compare with empirical method. 2. methodology the scaling or scale-invariant models enable transforming hydrologic information from one temporal or spatial to another one, and thus, help overcome the difficulty of inadequate hydrologic data. a natural process fulfills the simple scaling property if the underlying probability distribution of some physical measurements at one scale is identical to the distribution at another scale. the basic theoretical development of scaling has been investigated by many authors and considerable amount of studies were devoted to extreme rainfall and its scaling properties, including waymire and gupta (1981); waymire, et al.(1984); rodriguez-iturbe, et al. (1984); marien and vandewiele (1986); sivapalan and wood (1987); gupta and waymire(1990); rosso and burlando (1990); smith (1992); koutsoyiannis and foulfoula-georgiu (1993); burlando and rosso (1996); veneziano et al. (1996), bendjoudi et al. (1997); willems (2000); hubert et al. (2002) ; de michele et al. (2002); molnar and burlando (2005); ceresetti (2011); panthou (2013); ghanmi (2014). in this work the simple scaling hypothesis is adopted to test the scaling behavior of rainfall in northern of benin, following the methods described in menabde et al. (1999); yu et al. (2004); kuzuha et al. (2002); nhat et al. (2007); bara et al. (2009) and ghanmi (2014) and are briefly outlined in the ensuing sections. 2.1 scaling characteristics of extreme rainfall let id and id denote the annual maximum rainfall intensity series for time durations d and d, respectively. in the model proposed by menabde et al. (1999); yu et al. (2004); veneziano and furcolo (2002); bara et al. d3estimation of idf curves of extreme rainfall by simple scaling in northern oueme valley, benin republic (west africa) (2009) rainfall intensity id of duration d is said to exhibit simple scale invariance behavior if the following equation holds true. (1) hold true (2) the equality refers to identical propability distributions in both sides of the equations; denotes a scale factor and k is a scaling exponent. it follows that by raising both sides of equation (1) to the power of q and taking the ensemble average that is, taking the qth moment of both distributions, we obtain where k (q) represents the scaling exponent of order q. in order to obtain the value of k (q), we simply take logarithm of both sides of equation (2), which transforms it into: (3) (8) (10) (9) (4) (5) (7a) (7b) (6) when the moments are plotted on a logarithmic chart versus the scale λ for various moments order q, each slope will represent k (q) for the respective q. the various slope k(q) are then plotted against q: if the graph is linear, it implies that k is a constant and i(d) exhibits characteristics for simple scaling, while in other cases, the multi-scaling approach has to be considered. 2.2 estimation of idf rainfall extremes the idf curves are often fitted to the extreme value type i, evi developed by gumbel and it is still the most often used distribution by many national meteorological services in the world to describe rainfall extremes. it’s also used in this study along with the method of moments. the annual maximum rainfall intensity i(d) has a cumulative probability distribution (cdf) (gumbel, 1958). according to the scaling theory, the idf formula can be derived (menabde et al, 1999; nhat et al., 2007 and ghanmi, 2014) as: where and are respectively the location and the scale parameter for gumbel model at d hour data series; and represent the rainfall intensity for a duration d and d for a return period t. λ denotes a scale factor and k is a scaling exponent. it was also shown in menabde et al. (1999); nhat et al. (2007) and ghanmi (2014) that: where the location parameter μd and scale parameter ϭd of gumbel model are calculated from daily data series by using the equations of landwehr et al. (1979) given as: and are probability weight moment (pwm) for order 0 and 1, and are estimated by using the daily annual maximum rainfall intensity. the idf relationships can be derived from longer duration data series based on three parameters: scale exponent, the location and scale parameter of evi. the location and scale parameters (μ24 and σ24) of the evi (gumbel model) sample of annual maximum 1440 min rainfall data are calculated using equation (7a) and (7b). the results are recorded in table 2. 2.3 model performance as an indication of goodness of fit between the observed and predicted values, the coefficient of determination (r2), the root mean square error (rmse), the mean absolute percent error (mape) are calculated. the rmse and mape are defined as: -the root mean square error (rmse): and -the mean absolute percent error (mape): xobs,i is the ith observed data point, xest,i is the ith estimated value, n is the number of data point. idf curves are plotted together with the actual observed data, and shown in fig.4. to evaluate the goodness-of-fit graphically, idf curves derived by fitting gumbel extreme value distribution to actual observed data (of various durations d) are also plotted on the same graphs (fig.4). the statistical root mean square error (rmse) and mean absolute percent error (mape) are also used to evaluate the performance of the simple scaling model. the results are presented in table 3 are for 5 min, 10 min, 15 min, 20 min, 30 min and 60 min idf curves simple scaling. 2.4 study sites and data analysis the hydro meteorological observatory of upper oueme valley (ohhvo) has an area of about 14,000 km2 and is located in northern of benin republic, an area of humid tropical climate of sudanese type and delimited by latitude 9°n-10°n and longitude 1.5°e-3°e. this area was instrumented during the catch (coupling of tropical atmosphere and hydrological cycle) and amma (multidisciplinary analyses of african monsoon) projects. ohhvo is an area characterized by a single rainy season from mid-march to late october, the heart of the rainy season is between july and august with an interannual average rainfall of about 1200 mm (lawin, 2007). the study area has a subdued topography with an elevation of about 200 m (le barbé et al., 1993). the location of ohhvo in benin is shown in fig. 1. for a good representativeness of the samples, ten rain stations of ohhvo which operating rate (the ratio of the total number of days of operation to the total number of days in the year) during the year is greater than 90% have been chosen. the locations of these stations are shown in fig. 1. the data used are time series of rainfall intensity 5 min and 1440 min, measured from 1999 to 2012. matlab (v10.2) programming code is used to accomplish the tasks, allowing the aggregation of data over periods of 5 min and more and longer (i.e. 10 min, 20 min, 30 min, 60 min, 90 min, 120 min, 180 min, 240 min, 1440 min), using disjoint windows. missing in these raw data in the sets are provided by interpolating the neighboring values using the proposed formula of hoang, (2008), given as: p is the intensity of the calculated length, n is the total number of time steps of 5 min from the calculated time, n’ is the number of time steps of 5 min measured from the calculated time and p’i is the ith measured intensity pitch 5 min i. d4 m.agbazo, g. koton’gobi, b. kounouhewa, e. alamou and a. afouda table1: name of stations fig. 1. study site: (a) benin republic location in west africa, (b) ohhvo location in benin republic and (c) rain gauge location in ohhvo (1 = affon, 2 = corpargo, 3 = djougou, 4=wewe, 5= koko, 6= pélébina, 7= sarmanga, 8= sonoumon, 9= tobre and 10= wewe). 3. results and discussions 3.1 scale invariance properties of rainfall the property of simple scaling of rainfall intensities in the wide sense is demonstrated in the ten selected sample stations. the scaling exponents are derived by including all the analyzed durations of rainfall (5 min, 10 min, 15 min, 20 min, 30 min, 60 min, 90 min, 120 min, 180 min and 1440 min) to the analysis. fig. 2 displays the relationships between the log-transformed values of moments of various orders q (q values of 0.5 to 5 at increments) against various rainfall durations at copargo station. it can be observed that, for all the considered moment orders, log (moment) and log (duration) have a linear relationship for durations between 5 min and 1440 min. it’s not broken between the short and long duration data. linear regression performed for the regime confirmed the linear relationship, the correlation is consistently strong with coefficient of determination r2, greater than 0.94. the behavior (linearity) found from copargo station is similar to the other examined stations in the study area. the linearity of the moment from this study is also observed in others region of world. for example ghanmi (2014) in tunisia (northern of africa) find that data from (tunis) present linearity, ceresseti (2011) in france; bara, et al. (2009) in slovakia; nhat (2006) in japan, bougadis and admowski (2006) in canada have found linearity in such study. fig. 2. relationships between the log-transformed values of moments of various orders q and various rainfall durations at the station copargo. the orders q1 – q10 have values from 0.5 – 5 in steps of 0.5 respectively. the scaling exponents of the moments of various orders are estimated as the slopes of the linear regression between these moments and the rainfall duration. fig. 3 shows the relationships between the scaling exponents of the moments and the orders of the moments at copargo station. it’s obvious that the scaling exponents decrease with the order of moments and a linear relationship exists between scaling exponents and the moment orders. this implies that the property of wide sense simple scaling of rainfall intensity exists in this station. the scaling exponents and order of moments were linearly related with r2 value of 0.9932 for copargo station. data from other gauges of the 10 stations showed similar scaling relationships, indicating that scaling may be applicable in all the stations. fig.3 relationship between k (q) and order of moment q (q= [0.5:0.5:5]) for range, 5 min to 1440 min at copargo station. table 1 presents the characteristics of the examined stations, showing the stations and results of the scaling exponent factor k of the 10 stations in ohhvo. catchments are shown with high coefficients of determination ranging from 0.98 to 1. table 2: characteristics of examined stations: the values scale exponent k, and gumbel’s parameters of statistical analysis for 1440 min (σ24, μ24) the result indicates that for all the stations, the scaling exponents are less than 1 and in ranging from 0.23 to 0.59 in all the stations, showing considerable variability. the highest (the lowest) scaling exponent was found at sonoumon (pelebina) station. it is likely that the variability of the scaling exponents is influenced by the geographical location of the stations and/or the different mechanisms of rainfall generation. however, the results of olsson et al. (1992) which identified [0.6 0.8] as an adequate interval of scaling exponents at 45 min are not coherent with ours. these differences in the number of regime of scaling can be attributed to the differences of rainfall threshold detection of different used devices. comparing the value of scaling exponents k to those found by bara et al. (2009) in slovakia and chang (2013) in china, our estimation for all the 10 stations are different to other authors results, because these workers found the scaling exponents k higher than 6.2; but for few stations (sarmanga, wewe and sonoumon) where the value of k is approximately equal to those obtained by nhat (2006) d5estimation of idf curves of extreme rainfall by simple scaling in northern oueme valley, benin republic (west africa) in japan, bougadis et al. (2006) in canada and ghanmi (2014) in tunisia. the results indicate a strong validity of the simple scaling properties of the extremes rainfall in time series for the stations. 3.2 estimation of idf curves and evaluation of model performance the figures (fig.4) show the idf curves estimated by simple scaling and by fitting evi to the observations. it can be observed that the estimated idf curves of the short-duration rainfall by simple scaling fit the observed data point well at lower return periods (t<5 years) but not as well at higher return periods. this result may be due to the short observation period (only 13 years) of qualitative available rainfall data. from table 2, the location parameter μ24 and scale parameter σ24 calculated using equation (7) for the 10 stations is shown. subsequently, the mean μd and standard deviation σd for various sub-daily durations d are calculated from μ24 and σ24 using equations (5) and (6). fig. 4 idf curves estimated by simple scaling and by fitting evi to the observations for: (a) 5 min, (b) 10 min, (c) 15 min, (d) 20 min), (e) 30 min and (f) 1440 min rainfall. table 3. rmse and mape for 5 min, 10 min, 15 min, 20 min, 30 min, and 1440 min idf curves estimate using (a) simple scaling assuming evi and (b) fitting evi to observed data. the table 3 shows that mape values (scaling assuming evi) are in the range of 3.7-10%, which is slightly higher than the mape values when fitting evi to the observed data, ranging from 3.4-6.3%. the rmse values are in the range of 1-2.5 which is approximately higher than the rmse values when fitting evi to the observed data, ranging from 0.4-2.4. these results validate the performance of the simple scaling model in estimating the idf curves of short-duration rainfall in ohhvo valley (northern benin). 4. conclusion properties of the time scale invariance of selected rainfall data at 10 rainfall station in ohhvo valley (northern benin republic) are studied. a simple analytical formulation for rainfall idf relationship, which utilizes the scaling behavior, is presented. the benefit of using the principals of scaling d6 m.agbazo, g. koton’gobi, b. kounouhewa, e. alamou and a. afouda reduces the amount of parameters required to compute the quantiles. the main result of the study can be summarized as follows: (i) rainfall displays scale invariance property in one scaling regime from 5 min to 1440 min, (ii) the properties of rainfall follow the hypothesis of simple scaling, following the result of menabde et al. (1999), (iii) it is possible to derive the rainfall idf curves with only three parameters (k, μ24 and σ24) for duration shorter than a day. from only the location and scale parameters μ24 and σ24 of the evi sample of annual maximum 1440 min rainfall data and the scaling exponents, the idf curves of short-duration rainfall for low return periods (t<5years) could be estimated but may not be a reliable estimate for higher return period. in future studies, we should consider to exclude higher return periods, (iv) for the all stations the scaling exponents are less than 1 and range from 0.23 to 0.59, showing a considerable variability in values for the all stations. since it was not possible to find any clear dependence of the variability of the scaling exponents, it is believed that this variability may be accounted for by either for the geographical location of the stations or it the result of it is the sampling variability of the data (different mechanisms of rainfall generation), (v) idf relationships show acceptable result in comparison with the idf curves obtained from at-site short duration rainfall data. the key advantage is that idf curves of shorter rainfall duration for low return periods (t<5years) that are not measured or unavailable can be derived solely based on the statistical properties of the annual maximum 1440 min rainfall. this is relevant for benin republic daily data, available for long duration period, but data for short duration are often not available for the required site. the finding from this study may have significant bearing beyond the study area. acknowledgements the authors would like to thank the general directorate of amma for providing the data. we also thank the anonymous reviewers for their suggestions to improve the manuscript. we also thank the international foundation for science (ifs) for its financial support. 5. references acar, r., celik, s., senocak, s. (2008). rainfall intensity-duration-frequency model using an artificial neural network approach. jsir 198–202. baghirathan, v.r., shaw, e. m. (1978). rainfall depth-duration-frequency studies for srilanka. jh 37, 223–239. bara m. (2007). analysis of short-term rainfall intensities the simple scaling approach. in: 20th conference of young hydrologists, shmi, bratislava, cd, 10 p. (in slovak). bara, m., gaal, l., kohnova, s., szolgay, j., hlavcova, k. (2009). simple scaling extreme rainfall in slovakia a case study. meteorological journal 11, 153–157. bell, f.c. (1969). generalized rainfall-duration-frequency relationships. asae 95, 951–960. bendjoudi, h., hubert, p., schertzer d., lovejoy s. (1997). multifractal point of view on rainfall intensity-duration-frequency curves. c.r.acad.sci.paris earth planet sci 323–326. bernard, m. m. (1932). formulas for rainfall intensities of long durations. transactions of the american society of civil engineers, 96, 592-624. bougadis, j. and adamowski, a. (2006). scaling model of a rainfall intensityduration-frequency relationships. hp 3747–3757. burlando, p. and rosso, r. (1996). scaling and multiscaling models of depthduration-frequency curves for storm precipitation. jh 45–65. ceresetti d. anquetin s., molinié g., leblois e., et creutin j. d. (2011), severity diagrams: a new approach for the multi-scale evaluation of extreme rainfall events. weather and forecasting. chen, c. l. (1983). rainfall intensity duration frequency formulas. jhe 109, 1603–1621. danida, rapport final, (2008). appréciation des impacts des changements climatiques sur les programmes de développement de la coopération avec le bénin. de michele, c., kottegoda, n. t., rosso, r. (2002). idaf (intensityduration-area-frequency) curves of extreme storm rainfall: a scaling approach. wst 25, 83–90. descroix, l., mahé, g., lebel, t., favreau, g., galle, s., gautier, e., olivry, jc., albergel, j., amogu, o., cappelaere, b., dessouassi, r., diedhiou, a., le breton, e., mamadou, i., sighomnou, d. (2009). spatio-temporal variability of hydrological regimes around the boundaries between sahelian and sudanian areas of west africa. a synthesis. j. hydrol. 375: 90–102. endreny, t. a., and imbeah, n. (2009). generating robust rainfall intensity-duration-frequency estimates with short-record satellite data. jh 371, 182–191. garcia-bartual m, schneider h. (2001) estimating maximum shortduration rainfall intensities from extreme convictive storms. physics and chemistry of the earth, partb: hydrology, oceans and atmosphere, 26(9), 675-681. gumbel, e. j., 1958. statistics of extremes (thèse de doctorat). columbia university, columbia. gupta, v. k. and waymire, e. (1990). multiscaling properties of spatial and river flow distributions. jgr 95, 1999–2009. ghanmi, h., (2014). estimation des courbes intensité-durée-airefrequence (idaf) de la region de tunis dans un contexte multifractal. université de versailles saint-quentin-enyvelines et université de tunis el manar, à l’uvsq-latmos (guyancourt-france) thèse de doctorat. hubert p., bendjoudi h., schertzer d., lovejoy s. (2002): multifractal taming of extreme hydrometeorological events. the extremes of the extremes (proceeding of reykjavik symposium, july 2000). iahs publication no. 271, 51–56. hoang c.t. (2008). classic multifractal frequency analysis with high resolution 10 series rainfall. m.sc. thesis, university p. & m. curie, paris, 729 50 pp. koutsoyiannis d., foufoula-georgui e. (1993). a scaling model of storm hyetograph. water resour res 29, 2345–2361. landwehr, j., matalas, n. and wallis, j. (1979). probability weighted moments compared with some traditional techniques in estimating gumbel parameters and quantiles. wrr 1055–1064. lawin e. a., (2007) statistical and climatologycal analysis of rainfall patterns of the upper oueme valley rainfall data from ammacatch-benin, phd thesis, and inpg/uac 221p le barbé l.,alé g.,millet b., texier h., borel y., and gualde r. (1993) the surface water resources of the republic of benin orstom/ dh, orstom, paris(france), isbn: 2-7099-1168-x, 540p. marien j. l., vandewiele g. l. (1986): a point rainfall generator with internal storm structure. water resour. res., 22, 4, 5231–5238. menabde m., seed a., pegram g. (1999). a simple scaling model for extreme rainfall. water resour res 35, 335–339. mohymont b, demarée g r. (2006) courbes intensité-durée-fréquence des précipitations de yangambi, congo, au moyen de différents modèles de types montana. hydrol. sci. j., 51(2), 236-253. mohymont, b., demarée, g. r. and faka, d. n. (2004). establishment of idf-curves for precipitations in the tropical area of central africa. comparison of techniques and result. natural hazards and earth system science 4, 375–387. molnar, p. and p. burlando, (2008) variability in the scaling properties of high resolution precipitation data in the alpine climate of switzerland, water resour. res., 44, w10404, doi: 10.1029/2007wr006142. molnar, p., and burlando, p. (2005). preservation of rainfall properties in stochastic disaggregation by a simple random cascade model. atmos. res., 77, 137-151. nhat l.m., tachikawa y., sayama t., takara k. (2006). establishment of intensity-duration-frequency curves for precipitation in monsoon area of vietnam. annuals of disas. prev, res, inst, kyoto univ. nhat, l. m., tachikawa, y., sayama, t., and takara, k. (2007): a simple scaling characteristic of rainfall in time and space to derive d7estimation of idf curves of extreme rainfall by simple scaling in northern oueme valley, benin republic (west africa) intensity duration frequency relationships, annual journal of hydraulic engineering, jsce, vol.51, 73-78. nhat l.m., tachikawa y., sayama t., takara k. (2008) regional rainfall intensityduration-frequency relationships for ungauged catchments based on scaling properties. annuals of disas. prev, res, inst, kyoto univ 50, 33–43. okonkwo, g. i. and mbajiorgu, c. c. (2010). rainfall intensity duration frequency analysis for southeastern nigeria. cigr journal 12, 22–30. oyegoke, s. o. and oyebande, l. (2008). a new technique for analysis of extreme rainfall for nigeria. erj 2, 7–14. özger, m., mishra, a.k., singh, v.p. (2010). scaling characteristics of precipitation data in conjunction with wavelet analysis. journal of hydrology, 395, 279−288 olsson j., niemczynowicz j., berndtsson r. and larson m. 1992, an analysis of the rainfall time structure by box counting-some practical implications. journal of hydrology, 137: 261-277. panthou, g. (2013). analyse des extrêmes pluviométriques en afrique de l’ouest et de leur évolution au cours des 60 derniers années. thèse de doctorat, université de grenoble, ecole doctorale terre univers environment. panthou, g., vischel, t., lebel t., quantin, g. and molinié, g. (2014). characterising the space–time structure of rainfall in the sahel with a view to estimating idaf curves. hydrol. earth syst. sci., 18, 5093–5107, 2014 www.hydrol-earth-syst-sci. net/18/5093/2014/ doi:10.5194/hess-18-5093-2014 rodriguez-iturbe i., waymire e.c., gupta v.k. (1984). scale considerations in the modeling of temporal rainfall. wrr 20, 1611–1619. rosso r. and burlando p. (1990). scale invariance in temporal and spatial rainfall. aegs 23–27. sivapalan, m., wood, e.f. (1987). a multidimensional model of non-stationarity space time rainfall at the catchment scale. wrr 23, 1289–1299. sherman, c. w. (1931). frequency and intensity of excessive rainfall at boston. asce 95, 951–960. smith, j.a. (1992). representation of basin scale in flood peak distributions. water resour. res., 28, 11, 2993–2999. soro, g. e., goula, b. t. a., kouassi, f.w., koffi, k., kamagate, b., doumouya, i., savane, i. and srohourou, b. (2008). intensité durée fréquence des précipitations en climat tropical humide: cas de la région d’abidjan (côte d’ivoire). europeen journal of scientific research 21, 394–405. soro, g. e., goula, b. t. a., kouassi, f.w., srohourou, b. (2010). update of intensity-duration-frequency curves for precipitation of short duration in tropical area of west africa (cote d’ivoire). jas 10, 704–715. takara, (2005). report on data availability and idf procedures: situation in japan. annex japan country report. van nguyen vt, wang h. (1996). regional estimation of short duration rainfall distribution using available daily rainfall data. in advances in modeling the management of stormwater impacts, james w (ed.). published by chi: guelph. van de vyver, h. and demarée, g. r. (2010). construction of intensity-durationfrequency (idf) curves for precipitation at lubumbashi, congo, under the hypothesis of inadequate data. hsj 55, 555–564. veneziano, d., bras r.l., niemann j.d. (1996). non linearity and self-similarity of rainfall in time and a stochastic model. jgr 101, 371–392. veneziano d. and furcolo p. (2002), multifractality of rainfall and scaling of intensity-duration-frequency curves. water resources research 38, n° 12. waymire, e.c., gupta v.k. (1981). the mathematical structure of rainfall representation, 1, a review of stochastic rainfall model, 2, a review of the point processes theory, 3, some applications of the point process theory to rainfall processes. wst 17, 1261–1294. waymire, e.c., gupta v.k., rodriguez-iturbe i. (1984). a spetial theory of rainfall intensity at the meso-scale. wst 20, 1453–1465. willems, p. (2000). compound intensity/duration/frequency relationships of extreme precipitation for two seasons and two storm types, j. hydrol., 233: 189-205. world bank, 102p, (2009). world-bank, report on the status of disaster risk reduction in the sub-saharan africa region. yu p. s., yang t. c. et lina c. s. (2004), regional rainfall intensity formulas based on scaling property of rainfall. journal of hydrology 295, n° 1-4: 108–123. zannou, b. a. (2011). analysis and modeling of the continental hydrological cycle for integrated water resources management in benin: beterou basin at oueme. university of abomey-calavi, cipma. phd thesis. palabras clave: experimento de columna predicción del transporte de solutos; columna de suelo no perturbado. how to cite item godoy, v. a., napa-garcía, g. f., & zuquette, l. v. (2018). determination of the sodium retardation factor using different methods: analysis of their characteristics and impact on the solute movement prediction in a structured brazilian soil. earth sciences research journal, 22(4), 281-289. doi: https://doi.org/10.15446/esrj.v22n4.68100 the retardation factor (rd) is one of the main important solute transport parameters. its value can vary significantly depending on the method used for its determination. in this paper, the sodium rd is experimentally determined using undisturbed sandy soil columns to compare four methods of rd determination and assess the impact of the chosen method on the prediction of sodium movement. column experiments in undisturbed soil samples and analytical analysis were performed. the results showed that the soil has dual-porosity and preferential pathways. the breakthrough curves were in accordance with the soil’s physical characteristics. the rd values ranged from 1.7 to 7.77 depending on the initial concentration and on the method used. these differences arise from the conceptual model of each rd determination method. the experimental and analytical analysis indicated that the higher the rd, the slower the movement. the methods that best reproduced the laboratory sodium movement were ogata and banks’ (1961), langmuir and freundlich’s isotherms. the prediction models presented smaller errors with the increase of the initial concentration. in these cases, the predicted concentrations can be overestimated up to 22.5 % when using a not suitable method. these results suggest that the rd determination method can strongly affect the prediction of the sodium movement. because of that, it is of vital importance to evaluate each method and how they can be adequate for the soil under investigation when determining rd. abstract keywords: column experiment; solute transport prediction; undisturbed soil column. determination of the sodium retardation factor using different methods: analysis of their characteristics and impact on the solute movement prediction in a structured brazilian soil determinación del factor de retardo del sodio mediante métodos diferentes: análisis de sus características e impacto en la predicción del movimiento del soluto en un suelo brasileño estructurado issn 1794-6190 e-issn 2339-3459 https://doi.org/10.15446/esrj.v22n4.68100 el factor de retardo (rd) es uno de los principales parámetros para el transporte de solutos. su valor puede variar significativamente dependiendo del método utilizado para su determinación. en este artículo, el rd del sodio se determina experimentalmente utilizando columnas de suelo arenoso no perturbado, para comparar cuatro métodos de determinación de rd y evaluar el impacto del método elegido en la predicción del movimiento de sodio. se realizaron experimentos en muestras de suelo no perturbado y análisis analítico. los resultados mostraron que el suelo tiene doble porosidad y caminos preferenciales. las curvas de llegada estaban de acuerdo con las características físicas del suelo. los valores de rd variaron de 1.7 a 7.77 dependiendo de la concentración inicial y del método utilizado. estas diferencias surgen del modelo conceptual de cada método de determinación de rd. el análisis experimental y analítico indicó que cuanto mayor es la rd, más lento es el movimiento. los métodos que mejor reprodujeron el movimiento de sodio obtenido en el laboratorio fueron las isotermas de ogata y banks (1961), langmuir y freundlich. los modelos de predicción presentaron errores más pequeños con el aumento de la concentración inicial. en estos casos, las concentraciones pronosticadas pueden sobreestimarse hasta un 22,5% cuando se utiliza un método no adecuado. estos resultados sugieren que el método de determinación de rd puede afectar fuertemente la predicción del movimiento de sodio. debido a eso, al determinar rd, es de vital importancia evaluar cada método y cómo pueden ser adecuados para el suelo bajo investigación. resumen record manuscript received:03/10/2017 accepted for publication: 29/06/2018 earth sciences research journal earth sci. res. j. vol. 22, no. 4 (december, 2018): 281-289 vanessa a. godoy1,3, gian f. napa-garcía2, lázaro valentin zuquette3 1institute of water and environmental engineering, universitat politècnica de valència 2vale institute of technology 3department of geotechnical engineering, university of são paulo. godoyalmeida@gmail.com en v ir o n m en ta l g eo te c h n ic s 282 vanessa a. godoy, gian f. napa-garcía, lázaro valentin zuquette introduction maintenance of groundwater resource quality depends on the knowledge of how solutes move in the aquifer to estimate the groundwater contamination and therefore to manage the health risks. as the processes associated with the groundwater flow and solute transport are complexes, numerical modeling is a powerful tool to perform predictions and base the decision-making processes. numerical models require input parameters that must be determined reliably to guarantee the quality of the prediction. one of the main input parameters in the modeling of reactive solute transport is the retardation factor (rd) (fetter, 1999). because of the physical, biological and chemical mechanisms involved between a liquid solution and particle surfaces of the soil, the velocity of water percolation, v, differs from the velocity of the solute, vs (freeze and cherry, 1979). the relationship between these two velocities is called retardation factor. when the solutes do not interact with the solid medium (are nonreactive), rd = 1, the percolation velocity of the solute and the water are equal. solutes with rd > 1 are called reactive solutes (freeze and cherry, 1979; shackelford, 1994). according to jury, gardner and gardner (1991), retardation factors exceeding 3 indicate a high degree of solid particles-solute interaction and a very reactive solute. the determination of the rd can be performed based on the direct evaluation of the experimental breakthrough curves (btcs) from column experiments (freeze and cherry, 1979; van genuchten and parker, 1984; godoy, gómez-hernández and zuquette, 2018). the rd can also be determined using inverse modeling by fitting the advection-dispersion equation (ade) to the experimental data (ogata and banks, 1961; markhali and ehteshami, 2016). solute transport experiments can be performed using undisturbed soil cores (vanderborght, timmerman, and feyen, 2000; logsdon keller and moorman, 2002; jarvis, 2007; jellali et al., 2010), however, in a more general way, disturbed soils columns are used (dousset et al., 2007; eberemu, amadi, and edeh, 2013; delwiche, lehmann and walter, 2014; liu et al., 2014; internò, lenti and fidelibus, 2015; silva et al., 2016). batch tests, despite their limitations, are also used to determine sorption isotherms and then rd (brigham, reed and dew, 1961; fagundes and zuquette, 2011; fonseca et al., 2011; barbieri et al., 2012; liu et al., 2014; arthur et al., 2017). sorption isotherms are efficient in the rd determination but their determination must consider the soil heterogeneity. the characteristics of the rd determination methods may affect its quantification. the use of an inadequate method to determine rd values may cause under or overestimation of the fate and solute transport, impairing the design of clean-up techniques and health risk analyses. this is especially important by considering that those who work with numerical modeling usually use the provided parameters with no concerns about the method used in its determination. studies which aim to compare rd determination methods are rare and this paper intends to contribute to reduce this lack (shackelford, 1995; porfiri et al., 2015) (3. additionally, the use of undisturbed soil samples to perform solute transport test and determine sorption isotherm have not been identified in the literature, and are also presented in this work. the objective of this study is twofold: (i) to compare four methods of rd determination and (ii) to assess the impact of the chosen method in the prediction of sodium movement. in order to achieve that, first, the sodium rd is experimentally determined using undisturbed sandy columns and then the rd were determined based on the methods proposed by freeze and cherry (1979), by van genuchten and parker (1984), by the fit of the sorption isotherms models to the experimental data obtained from column experiments, and by fitting the analytical solution of the ade to the experimental data. second, the rd obtained with all these methods were used as input parameter to reproduce the experimental btc using the analytical solution of the ade to evaluate the impact of a different rd in the prediction of the sodium movement. data and methods soil sampling and characterization the study was carried out in an experimental site located at the city of cabrália paulista (22°27′37″ s, 49°20′16″ w), state of são paulo, brazil. this area is constituted by the residual soil of sandstones from the adamantina formation, bauru group (k). the area was chosen because its geological materials are spread out at about 117,000 km² in the state of são paulo (paula e silva, kiang, and caetano-chang, 2003) and incorporate the guaraní aquifer system and some of its aquifer unities such as the lower and medium bauru unities. the soil samples were taken from an area quite close to one where salty wastewater from domestic sewage has been used for irrigation purpose. because of the high sodium (na+) concentration in the wastewater, this was the solute used in this study. we cautiously collected ten undisturbed samples from hand-excavated trenches using rigid pvc cylinders (150 mm length and 97.2 mm inner diameter). undisturbed samples were used to conduct water flow and solute transport experiments in the laboratory and to analyze the soil porosity. porosity analysis by mercury intrusion (mip) was performed in three samples (washburn, 1921). mip is based on the penetration of mercury into the soil pores, which occurs only under pressure since the mercury is a nonwetting fluid. the mercury penetration pressure should be higher the smaller the pore diameter, according to p t d s= −4 cos (1) where ts is the surface tension; α is the contact angle between the liquid and the surface of the solid; p is the applied pressure; and, d is the pore diameter. during the test, as the pressure is increased, the volumes of mercury inserted into the sample are monitored. using equation 1 the dv/dd differentials were obtained, which allowed obtaining the pore distribution of the sample. the apparatus used in these tests was the “pore size 9320 micromeritics instrument corporation”, with a maximum pressure of 30 000 psi. disturbed samples were taken to characterize, in three replicates, the physical, chemical, mineralogical and physicochemical properties of the soil (godoy and zuqutte, 2013). the mineralogical analysis was performed using x-ray diffraction(xrd) (azaroff and buerger, 1953), particle size analysis was performed according to astm d422-63. the density of the solids was determined according to astm d854-10. we also determined in three replicates the ph in h2o and in kcl, reduction potential (eh), electrical conductivity (ec) (donagema and campos, 2011), delta ph (phkcl – phh2o) (mekaru and uehara, 1972), point of zero charge (pzc) (2phkcl – phh2o) (keng and uehara, 1974), and organic matter content according to astm d2974 – 00. the arithmetic mean of the three replicates was used as the value of the measured parameter. the methylene blue adsorption test using the filter paper method described by pejon (1992) was used to determine cec in one replicate from each soil sample. in order to visually analyze the soil fabric characteristics as well as its capacity to retain solute, a solution of methylene blue (adsorbing dye) was percolated through the soil, a traditional method used to visualize flow pathways (zehe and flühler, 2001; dousset et al., 2007). first, the filter paper was placed across the base of an unsaturated and undisturbed soil column, so that the preferential paths would be marked in the cross-sectional area of the column. second, a volume of 1.5e-03 m³ of a methylene blue solution at a concentration of 3 kgm-3 was percolated through the soil. the soil was set aside to allow the excess solution to drain. third, the soil sample was removed from the pvc cylinder and longitudinally cut to observe the methylene blue pathways and visually analyze the retention and the flow through the pores. column experiments the pvc cylinders used for collecting the undisturbed soil samples were used as rigid-wall permeameters, in which column experiments were 283determination of the sodium retardation factor using different methods: analysis of their characteristics and impact on the solute movement prediction in a structured brazilian soil conducted. the soil columns were first slowly saturated from the bottom with distilled water to remove entrapped air. to ensure complete saturation, the distilled water was allowed to overflow 1 cm on the soil surface for two days. after column saturation, the flow was reversed, and the tests were performed under a constant hydraulic gradient of 1.5. once the steady-state condition was verified, that is, when flux (q) variations were below 5% in a week, the following water flow parameters were obtained from each column: the saturated hydraulic conductivity, k (darcy, 1856); darcy velocity, q (darcy, 1856)and average linear velocity, v (freeze and cherry, 1979). the mean effective porosity, ne, which is used in the calculation of v, was adopted as the percentage of pores larger than 10 µm based on the mip results (ahuja et al., 1984). three undisturbed soil columns were percolated with nacl and the effluent was collected at each pore volume (godoy, zuquette and napa garcía, 2015). the concentrations of the initial solution inserted in the columns 1, 2 and 3 were, respectively, 50, 90 and 117 mg.l-1 based on previous studies (patterson, 1997; medeiros et al., 2005; silva et al., 2012). solute displacement tests were carried out under isothermal (20 °c) conditions. the ph and electrical conductivity, both of the input solution and of the percolated effluent solution, were monitored throughout the test. the na+ concentrations were determined in three replicates using a micronal b26 flame photometer at a 1:21 dilution ratio. the column experiments design was also described by godoy, zuquette and napa garcia (2015). determination of retardation factors the rd values were determined by four methods. first, rd was determined inversely by fitting the analytical solution of the ade (ogata and banks, 1961) to the experimental btc, r c t d c x v c xd h       = − ² ² (2) where c is the concentration [ml-1], dh is the longitudinal dispersion coefficient [l2t-1], v is the linear velocity of the fluid through a column [lt-1], and rd is the retardation factor. the analytical solution of the equation 2 was proposed by ogata and banks (1961) and lapidus and amudson (1952) and was used to solve the ade when the initial condition is c=0 for the entire sample, and the boundary conditions are c=c0 at the inlet and c=0 at an infinite distance from the inlet. c c r l vt d r t vl d r l vt d r d h d h d h0 1 2 2 1 2 2 = −      +     + erfc erfcexp ddt               (3) second, rd was determined using the method proposed by van genuchten and parker (1984) where the rd is the area above the btc; third, the rd determination was performed by using the method proposed by freeze and cherry (1979) where the rd is equal to the pore volume when the relative concentration (c/c0) is 0.5; and fourth, the rd was found by fitting the sorption isotherms, obtained through column test, to the linear, freundlich and langmuir isotherm models (freeze and cherry, 1979; garga and o’shaughnessy, 1994; fetter, 1999; asgari et al., 2014). the ade and the isotherms models’ fits were performed using mathematica 9.0. table 1 shows the equations that were used to determine the rd from different sorption isotherms. lastly, the rd obtained with the aforementioned methods were directly used to predict the experimental btc using the analytical solution of the ade. the coefficient of determination r2 was computed and used to assess the goodness of the rd to explain the experimental data. the simulated btcs were used to assess the impact of the rd determination method in the sodium movement prediction. results and discussion soil characterization and flow parameters texturally, the soil is clayey silty sand and consists of 79.4% sand, 10.5% clay and 10.1% silt particles. the specific weight of the solids was consistent with sandy soil values (26.4 knm-³). the main minerals are quartz, kaolinite, and gibbsite and the results of the xrd are showed in figure 1. table 1. equations used to determine rd from sorption isotherms eq. 3 eq. 4 eq. 5 r n n kd = ( ) 1 1 + − r n n bcd s e b= ( ) 1 1 1 12+ − ρ − r n n k k cd s e = ( ) 1 1 1 1 2 + − ρ + where n represents the porosity [ ]; k is the henry coefficient for linear isotherm [ ]; b1 is the freundlich coefficient [ml-3(-b2)-1]; b2 is the freundlich exponent [ ]; k1 is the langmuir sorption coefficient[]; k2 is an exponential coefficient [l3m-1]; ce is the equilibrium concentration [ml-3]. 0 200 400 600 800 1000 5 15 25 35 45 55 in te ns ity (c ps ) angle of bragg 2θ norma l condition 3.34 (quartz) 3.54 (kaolinite) 4.26 (gibbsite) 7.25 (kaolinite) 0 200 400 600 800 1000 5 15 25 35 45 55 in te ns ity (c ps ) angle of bragg 2θ ethylene glycol 3.34 (quartz) 4.26 (gibbsite) 10.00 (haloisite) 0 200 400 600 800 1000 5 15 25 35 45 55 in te ns ity (c ps ) angle of bragg 2θ 550 °c 3.34 (quartz) 4.26 (gibbsite) 7.25 (kaolinite) 3.54 (kaolinite) a c b figure 1. xrd diffractograms of the studied soil soil: a normal condition; b ethylene glycol; c 550 ֯c 284 vanessa a. godoy, gian f. napa-garcía, lázaro valentin zuquette table 2 summarizes the soil physicochemical parameters. the soil presented values of 5.23 and 4.20 for ph in h2o and in kcl, respectively, and contains a small amount of organic matter (1.18%). with these results, it is possible to note that there was an indication of the electrostatic adsorption due to the negative delta ph (-1.03) and pzc lower than the phh2o (4.18). according to the mip, the distribution of pore sizes is: 5% macropores, 50% mesopores and 45% micropores (koorevaar, menelik and dirksen, 1983). according to godoy (2015), the soil also contains macropores larger than 100 µm that cannot be identified in the porosimetry analysis due to characteristics of the test. these pores are up to 4 mm in diameter and were presumably important in determining the movement of the sodium in the soil. table 3 shows the flow parameters obtained from column experiments. the average k coefficient is in the typical range of sandy soils (zuquette and palma, 2006). the mean v can be considered high, most probably due to the presence of preferential pathways caused by macropores (mcmahon and thomas, 1974; dousset et al., 2007; delwiche, lehmann and walter, 2014; silva et al., 2016). analysis of breakthrough curves figure 3 shows the btcs of the na+ plotted using the relative concentration measured in the leachate (c/c0) vs. the number of eluted pore volumes and table 4 summarizes the results obtained from column experiments. with the increase of the initial concentration there is a displacement of the btc to the left, indicating that the higher the concentration of the initial solution, the faster the solute is identified in the effluent solution. this can be related to the limited sorption sites in the soil as well as to the soil physical heterogeneities. the btcs have the appearance of a continuous flow of initial solution. the tails show that in the beginning, the solute speedily left the soil column and after that, there was a slow transport and a significant tailing occurred. this can indicates physical nonequilibrium and the presence of heterogeneous adsorption sites in the soil, caused by smallscale heterogeneity (gerritse, 1996; vanderborght, timmerman and feyen, 2000; dousset et al., 2007; jarvis, 2007). the small-scale heterogeneity may be due to the dual-porosity characteristic of the soil. the dual-porosity affects the timing and shape of the btc as mentioned by rezanezhad et al. (2012). that occurs because part of the solutes can move fast due to advection in large pores and part of them can be trapped in immobile regions due to percolation through micropores and non-interconnected pores, behavior also stated by others (van genuchten and wierenga, 1976; jarvis, 2007; silva et al., 2016). visual analysis of the dye transport the visual analysis of the dye transport pathways indicated that, although the methylene blue solution was equally distributed in the upper part of the column, there was a greater flow in one location than in the rest of the column and it is evident in figure 4 which shows the location where preferential flow occurred (labeled in the photo). as only water was observed in the effluent, the methylene blue is believed to have been subject to physical retardation in macropores, which do not connect to the bottom of the soil sample (figure 5), and chemical retardation due to the soil and solute chemical characteristics. table 2. physico-chemical parameters of the soil used in the column experiments ph δph pcz eh c.e ctc m.o ph(h20) ph(kcl) (mv) (ms) meq/100g (%) 5.23 4.20 -1.03 4.18 264 55.70 0.152 1.18 to ta l i nt ro du ce d vo lu m en (% ) pore diameter (1e-6 m) 0.001 0.01 0.1 1 10 100 1000 100 80 60 40 20 0 sample 1 sample 2 sample 3 figure 2. porosimetry tests results total volume of mercury versus pore diameter table 3. flow parameters obtained from column experiments column1 column2 column3 ksat(m/s) a 3.060 x 10-05 3.50 x 10-05 4.43 x 10-05 ib 1.5 1.5 1.5 q (m/s)c 4.59 x 10-05 5.25 x10-05 6.65 x 10-05 v(m/s)d 2.18 x 10-04 2.50 x10-04 3.16 x 10-04 a: saturated hydraulic conductivity; b: hydraulic gradient; c: darcy velocity; d: average linear velocity figure 3. breakthrough curves of the na+ table 4. results obtained from column experiments column 1 colum 2 colum 3 pore volume c/c0 c/c0 c/c0 0 0.00 0.00 0.00 0.395 0.00 0.11 0.09 1.185 0.10 0.36 0.38 1.975 0.40 0.56 0.66 2.765 0.50 0.72 0.77 3.555 0.60 0.78 0.83 5.135 0.70 0.83 0.91 6.715 0.80 0.86 0.94 9.085 0.85 0.91 0.96 11.455 0.95 0.96 0.98 285determination of the sodium retardation factor using different methods: analysis of their characteristics and impact on the solute movement prediction in a structured brazilian soil a similar result was found by dousset et al. (2007) who verified physical retardation of a dye tracer in non-connected pores. even after introducing three pore volumes of the solution, the methylene blue remained in the surface portion of the column. the retention of methylene blue indicates that the soil has very significant cation sorption, contrary to what is expected for sandy soils since the mineralogy of these soils is not composed by reactive clay minerals. sodium sorption isotherm the sodium sorption isotherm is shown in figure 6. there is an increase in sorption over time and saturation is reached when no additional sorption of na+ occurs. different results were obtained by müller (2005) and humenick and mattox (1978). the maximum na+ sorption value (0.1 gg-1) did not exceed the maximum soil sorption (0.5 gg-1) obtained by the cec analysis. thus, the sorption values probably are primarily related to physical mechanisms such as retention in micropores which represent 45% of the total porosity. with this result, it is possible to suppose that chemical and electrical mechanisms were probably of minor importance in the na+ retardation. it is important to mention that four points may not be sufficient to define the isotherm accurately but can show the trend of the sorption behavior as obtained by önal et al. (2007). table 5 presents the fit of the experimental data to the mathematical models of the langmuir, freundlich, and linear isotherms. the langmuir and freundlich models fitted well to the experimental data (r2 = 0.99). according to önal et al. (2007), the langmuir model can be successfully applied to many real sorption processes. the linear model had the worst fit (r2 = 0.84) indicating the nonlinearity of the experimental isotherm. retardation factor and its impact on the sodium movement prediction in this section, the rd values obtained with different methods are shown and discussed. table 6 shows the retardation factors obtained using the methods proposed by freeze and cherry (1979), van genuchten and parker (1984), ogata and banks (1691) and by the fit of the sorption isotherms. figure 4. a water spot on the filter paper indicating the location where preferential flow occurred figure 5. visual analysis of the dye transport pathways: retardation in connected macropores that are not connected to the bottom of the soil sample figure 6. sorption isotherm of na+ table 5. parameters obtained by fitting mathematical sorption models to experimental data model parameters fitted langmuir r2 0.99 k1 12.42 k2 112.1 freundlich r2 0.99 b1 0.14 b2 0.14 linear r2 0.84 k 3.17 table 6. retardation factors obtained with different methods method retardation factor column 1 column 2 column 3 (50 mg.l-1) (90 mg.l-1) (117 mg.l-1) freeze and cherry (1979) 2.75 1.75 1.51 ogata and banks (1691) 7.8 4.18 3.84 van genuchten and parker (1984) 4.42 3 2.17 linear isotherm 4.2 4.2 4.2 freundlich isotherm 7.78 4.5 3.63 langmuir isotherm 7.69 4.51 3.6 286 vanessa a. godoy, gian f. napa-garcía, lázaro valentin zuquette the rd values can be considered very high, except for the one obtained with the freeze and cherry (1979) method (jury, gardner, and gardner, 1991). rd values were higher even when compared to results obtained in compacted kaolin at low flow rates, which has a high retardation factor since they are used as liners (shackelford and redmond, 1995). using the linear isotherm model, garga and o’shaughnessy (1994) found sodium retardation factor ranging from 2.22 to 3.6 in clay soil columns with high cec. the high values of rd obtained are in accordance with the results obtained with the sorption isotherm and with the dye transport. the sorption isotherm yields values of rd smaller than those obtained by the cec from the blue methylene adsorption test suggesting a physical dependence originated by e.g. preferential pathways which were also evidenced in the dye transport pathway. given that high average linear velocity values were observed during the experiments (table 3), small values of rd can be expected because of the reduction of the contact time between the solute and soil particles, as observed by hoag and price (1997). the soil sorption and physical-chemical characteristics did not indicate that rd values would be high. however, the physical characteristics played a key role in the rd quantification, as was also noticed by rezanezhad et al. (2012). there is a trend to decrease rd with increasing initial concentration. this behavior is in accordance with the limited sorption sites in the soil but could also reflects soil heterogeneities between the columns, since they are natural soil cores and do not have the same pore configuration, as observed by dousset et al. (2007). the coefficients of determination (r2) of the prediction of the btc using the analytical solution of the ade are shown in table 7. figure 7 shows the fit of the experimental btc to the analytical equation using rd obtained with the methods used in this study. in figure 7, it can be observed that the higher the rd, the slower the movement and the btc displace to the left, as mentioned by other authors (freeze and cherry, 1979; shackelford, 1991; malusis and shackelford, 2002). the rd values obtained with the method of freeze and cherry (1979) were the smallest. this method can be suitable for soils with no preferential pathways, but when there is a dual-porosity, part of the solute will appear faster in the effluent solution, and part will be retained in micropores (rezanezhad et al., 2012). thereby, half of the initial concentration (c/c0) = 0.5, can be detected too fast in the effluent solution, not representing the solute that still is retained and that is also very important in the determination of the correct rd. according to shakelford (1994), this method can be adequate when pure advection occurs, i.e., without dispersion/diffusion. it is suitable when high ( > 50) péclet numbers are achieved (only advection) but is not appropriate when there is dispersion. moreover, this method is based on the sigmoidal btc of homogeneous soils, which was not observed in the results. therefore, this method is not adequate to determine the retardation factor of the sodium in the studied soil. the method of freeze and cherry (1979) also overestimated na+ movement prediction once the fitted btc is displaced to the left in comparison to the experimental curve (figure 7). the r2 obtained with this method were the smallest but increased with increasing concentration of the initial solution. the method proposed by ogata and banks (1961) resulted in higher rd values for all columns. this method is suitable for homogeneous soils, with linear sorption, characteristics not suitable for the studied soil. however, the r2 was 0.99 for all initial concentrations, indicating that the ade model explained the experimental btc. therefore, it is possible that the ade can be adequate to reproduce btc in heterogeneous materials for small-scale studies, in which the macrodispersion has no place. the method proposed by van genuchten and parker (1984), besides being based on the ogata and banks (1691) transport equation, considers the mass balance of the solute in the soil and can so be capable of capturing the mass of solute retained in the soil regardless of whether there is equilibrium or nonequilibrium transport. the relatively good suitability of this method was verified with r2 ranging from 0.78 to 0.91. however, in figure 7 it is noticeable that the sodium movement is overestimated. it is important to mention that the correct use of this method depends on whether the solute had reached inflow concentration, which can sometimes take a long time. in this research, the inflow concentration was not reached in any column test and to use this method, an extrapolation was made, which can explain the errors in the na+ movement prediction. column 1 50 mg/l column 2 90 mg/l column 3 117 mg/l times [minutes] times [minutes] times [minutes] c /c 0 c /c 0 c /c 0 experimental btc langmuir isotherm freundlich isotherm linear isotherm van gen. and parker ogata and banks freeze and cherry experimental btc langmuir isotherm freundlich isotherm linear isotherm van gen. and parker ogata and banks freeze and cherry experimental btc langmuir isotherm freundlich isotherm linear isotherm van gen. and parker ogata and banks freeze and cherry 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 1 0.9 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360 0 60 120 180 240 300 360 figure 7. fit of the experimental btc to the analytical solution of the ade using rd obtained with all the methods used in this study work 287determination of the sodium retardation factor using different methods: analysis of their characteristics and impact on the solute movement prediction in a structured brazilian soil conclusion aiming to compare four methods of sodium rd determination and assess the impact of the chosen method in the prediction of sodium movement, we did laboratory tests and used the analytical solution of the ade to perform solute transport predictions. with the presented study, it was possible to highlight some conclusions. in columns containing natural and completely weathered soils, with preferential paths and dual-porosity, the solute transport parameters obtained using some of the most common methods found in the literature may result in non-representative rd values mainly due to the natural heterogeneity of the soil column. the suitability of the method used for determining rd depended on the concentration of the initial solution. the higher the initial concentration, the smaller the errors in the prediction of sodium movement. for the conditions studied, the method proposed by ogata and banks (1961) and by the fit of the freundlich and langmuir isotherms models resulted in the best reproduction of the experimental breakthrough curve. differently, the method proposed by freeze and cherry (1979) resulted in the worst result. the fit of the experimental btc to the ade analytical models showed that the predicted concentration can be overestimated up to 22.5% when using a not suitable method. with this study, we conclude that the rd determination method can significantly affect the prediction of the sodium movement. because of that, it is important to evaluate carefully when a method can be adequate to determine the rd for the soil under investigation and when cannot be adequate. acknowledgements the authors would like to thank the teachers and staff of the state technical school “astor de mattos carvalho” for allowing collecting soil samples on the school’s area. this research was made possible by the involvement of financial support from the coordination for the improvement of higher education personnel (capes). references ahuja, l. r. (1984). macroporosity to characterize spatial variability of hydraulic conductivity and effects of land management. soil science society of america journal, 48(4), 699. url: https:// www.soils.org/publications/sssaj/abstracts/48/4/ss0480040699. for column 1, the use of the rd obtained with the linear isotherm model is not adequate for the prediction of the na+, since the movement is overestimated and r2 was small (0.29). these results are adequate with the fact that the soil isotherm is nonlinear and that the soil has a limited sorption capacity, which is not represented by the linear model. however, for columns 2 and 3, this model seems to be suitable with r2 equal to 0.99. this method should be used with caution since its suitability depends on the concentration of the initial solution. the results obtained with the other two isotherm models, langmuir and freundlich, were quite similar and, in comparison to the experimental btc, the simulated curves using those methods reproduced very well the experimental btc with r2 equal to 0.99 for both isotherms. these results show the efficiency of the use of isotherms obtained from undisturbed soil columns to determine rd. the comparison of the simulated btc evidenced the impact of the method used to calculate de rd in the sodium movement prediction and this impact is more pronounced with the decrease in the initial concentration of the inlet solution. the r2 increased with the increase of the initial concentration for almost all conditions studied. it is possible that, for a low concentration of the initial solution, the equilibrium adsorption has not yet been reached and the rd determination could be more difficult to achieve. by increasing the concentration of the initial solution, the rd determination is more stable and almost all methods work relatively well. table 8 shows the comparison between na+ concentrations obtained experimentally and after 180 minutes using different methods of rd determination in the analytical solution of the ade. from table 8 it is noticeable that for column 1, the predicted concentration using the rd calculated with the method of ogata and banks (1961) was 3.75 % smaller compared to the experimental data; with the langmuir and freundlich isotherms, it was 2.5 % smaller. the linear isotherm, freeze and cherry’s and van genuchten and parker’s methods overestimated the concentration in 16.25%, 22.5 %, and 15%, respectively. for column 2 and 3, all concentrations were overestimated, but this effect was smaller when compared to the results of column 1. for column 2, the overestimation of the concentrations was 4.5 % when using the method of ogata and banks, 13.80% for the method proposed by freeze and cherry, 10.34% for the methods of van genuchten and parker and the langmuir and freundlich isotherms, and 6.8 % for the linear isotherm. for column 3, the increase in the prediction of the concentration ranged from 1 % to 6%, and the worst method was the one proposed by freeze and cherry. for all columns, the highest overestimated percentage was obtained using the method proposed by freeze and cherry, as expected since this method resulted in the smallest rd due to its characteristics as discussed above. table 7. the coefficient of determination (r2) of the prediction of the btc using the analytical solution of the ade method column 1 column 2 column 3 (50 mg.l-1) (90 mg.l-1) (117 mg.l-1) r2 r2 r2 freeze and cherry (1979) 0.18 0.66 0.72 ogata and banks (1691) 0.99 0.99 0.99 van genuchten and parker (1984) 0.28 0.93 0.88 linear isotherm 0.29 0.99 0.99 freundlich isotherm 0.99 0.98 0.99 langmuir isotherm 0.99 0.98 0.99 table 8. comparison between sodium concentrations obtained experimentally and after 180 minutes using a different method of rd determination in the analytical solution of the ade method concentration (mg/l) after 180 minutes column 1 column 2 column 3 (50 mg.l-1) (90 mg.l-1) (117 mg.l-1) experimental 40 78.3 108.81 freeze and cherry (1979) 49 89.1 115.83 ogata and banks (1691) 38.5 81.9 111.15 van genuchten and parker (1984) 46 86.4 112.32 linear isotherm 46.5 83.7 109.98 freundlich isotherm 39 86.4 112.32 langmuir isotherm 39 86.4 112.32 288 vanessa a. godoy, gian f. napa-garcía, lázaro valentin zuquette arthur, j. d. (2017). batch soil adsorption and column transport studies of 2,4-dinitroanisole (dnan) in soils. journal of contaminant hydrology, 199, 14–23. url: http://www.sciencedirect.com/science/ article/pii/s0169772216301310. asgari, m. (2014) designing a commercial scale pressure swing adsorber for hydrogen purification. petroleum and coal, 56(5), 552–561. azaroff, l. & buerger, m. (1953). the powder method in x-ray crystallography. new york. mcgraw-hill book co. barbieri, m. (2012) formation of diclofenac and sulfamethoxazole reversible transformation products in aquifer material under denitrifying conditions: batch experiments. science of the total, 426, 256–263. url: http://www.sciencedirect.com/science/article/pii/ s0048969712002707. brigham, w. e., reed, p. w. & dew, j. n. (1961). experiments on mixing during miscible displacement in porous media. society of petroleum engineers journal, 1(1), 1–8. url: http://linkinghub.elsevier.com/retrieve/pii/s016977229600085x. darcy, h. (1856). les fontaines publiques de la ville de dijon : exposition et application des principes à suivre et des formules à employer dans les questions de distribution d’eau. recherche, 647. delwiche, k., lehmann, j. & walter, m. (2014). atrazine leaching from biochar-amended soils, chemosphere, 95, 346–352. url: http:// www.sciencedirect.com/science/article/pii/s0045653513012836. donagema, g. & campos, d. (2011). manual de métodos de análise de solo. embrapa solos. dousset, s. (2007). evaluating equilibrium and non-equilibrium transport of bromide and isoproturon in disturbed and undisturbed soil columns. journal of contaminant, 94(3–4), 261–276. url: http:// www.sciencedirect.com/science/article/pii/s016977220700085x. eberemu, a. o., amadi, a. a., & edeh, j. e. (2013). diffusion of municipal waste contaminants in compacted lateritic soil treated with bagasse ash. environmental earth sciences, 70(2), 789–797. url: http:// link.springer.com/10.1007/s12665-012-2168-z. fagundes, j. r. t. & zuquette, l. v. (2011). sorption behavior of the sandy residual unconsolidated materials from the sandstones of the botucatu formation, the main aquifer of brazil. environmental earth sciences, 62(4), 831–845. url: http://link.springer.com/10.1007/ s12665-010-0570-y. fetter, c. (1999). contaminant hydrogeology. 2nd edn. new york: prentice hall. fonseca, b. (2011). mobility of cr, pb, cd, cu and zn in a loamy sand soil: a comparative study. geoderma, 164(3–4), 232–237. url: http:// www.sciencedirect.com/science/article/pii/s0016706111001856. freeze, r. & cherry, j. (1979). groundwater. new jersey: prenticehall inc englewood cliffs. garga, v. k. & o’shaughnessy, v. (1994). the hydrogeological and contaminant-transport properties of fractured champlain sea clay in eastern ontario. part 2. contaminant transport. canadian geotechnical journal, 31(6), 902–915. url: http://linkinghub.elsevier. com/retrieve/pii/s0304389407004025. gerritse, r. g. (1996). dispersion of cadmium in columns of saturated sandy soils. journal of environmental quality, 25(6), 1344–1349. godoy, v. a., gómez-hernández, j. j. & zuquette, l. v. (2018). scale effect on hydraulic conductivity and solute transport: small and large-scale laboratory experiments and field experiments. engineering geology. url: https://linkinghub.elsevier.com/retrieve/ pii/s0013795218301571. godoy, v. a., zuquette, l. v. & napa garcía, g. f. (2015). transport mechanisms of sodium in sandy soil from column leaching test. in: engineering geology for society and territory volume 3: river basins, reservoir sedimentation and water resources. cham: springer international publishing, 197–200. url: http://link. springer.com/10.1007/978-3-319-09054-2_39. godoy, v. a. & zuqutte, l. v. (2013) avaliação do retardamento de fosfato em colunas indeformadas de solo residual arenoso de área irrigada com efluente de tratamento de esgotos. periódico eletrônico fórum ambiental da alta paulista, 9(11). hoag, r. s. & price, j. s. (1997). the effects of matrix diffusion on solute transport and retardation in undisturbed peat in laboratory columns. journal of contaminant hydrology, 193–205. url: http:// www.sciencedirect.com/science/article/pii/s016977229600085x. humenick, m. j. & mattox, c. f. (1978) groundwater pollutants from underground coal gasification, water research, 12(7), 463–469. url: http://www.sciencedirect.com/science/article/ pii/0043135478901537. internò, g., lenti, v. & fidelibus, c. (2015). laboratory experiments on diffusion and sorption of heavy metals in a marine clay. environmental earth sciences, 73(8), 4443–4449. url: http://link.springer.com/10.1007/s12665-014-3729-0. jarvis, n. j. (2007). a review of non-equilibrium water flow and solute transport in soil macropores: principles, controlling factors and consequences for water quality. european journal of soil science, 58(5), 523–546. url: http://www.bioone.org/doi/abs/10.1139/ cjss2011-050. jellali, s. (2010). dynamic sorption of ammonium by sandy soil in fixed bed columns: evaluation of equilibrium and non-equilibrium transport processes. journal of environmental management, 91(4), 897–905. url: http://linkinghub.elsevier.com/retrieve/pii/ s0301479709003909. jury, w. a., gardner, w. r. & gardner, w. h. (1991). soil physics. john wiley & sons, inc., new york, 276, 345–348. keng, j. & uehara, g. (1974) chemistry, mineralogy and taxonomy of oxisols and ultisols. proceedings of the soil and crop science society. koorevaar, p., menelik, g. & dirksen, c. (1983). elements of soil physics. in: p. koorevaar, g. menelik, & c. dirksen (eds). developments in soil science. elsevier science. lapidus, l. & amundson, n. (1952). mathematics of adsorption in beds vi. the effect of longitudinal diffusion in ion exchange and chromatographic columns. the journal of physical chemistry, 984–988. url: http://pubs.acs.org/doi/pdf/10.1021/j150500a014. liu, j. (2014) adsorption of arsenic(v) on bone char: batch, column and modeling studies. environmental earth sciences, 72(6), 2081– 2090. url: http://link.springer.com/10.1007/s12665-014-3116-x. logsdon-keller, k. e. & moorman, t. b. (2002). measured and predicted solute leaching from multiple undisturbed soil columns. soil science society of america journal, 66(3), 686–695. url: https:// www.soils.org/publications/sssaj/abstracts/66/3/686. malusis, m. a. & shackelford, c. d. (2002). theory for reactive solute transport through clay membrane barriers. journal of contaminant hydrology, 59(3–4), 291–316. url: http://www.sciencedirect. com/science/article/pii/s0169772202000414. markhali, s. p. & ehteshami, m. (2016). environmental assessment of leachate transport in saturated homogeneous media using finite element modeling. environmental earth sciences, 75(16), 1193. url: http://link.springer.com/10.1007/s12665-016-5994-6. mcmahon, m. & thomas, g. (1974). chloride and tritiated water flow in disturbed and undisturbed soil cores. soil science society of america journal, 38(5), 727–732. medeiros, s. de s. (2005). utilização de água residuária de origem doméstica na agricultura: estudo das alterações químicas do solo. revista brasileira de engenharia agrícola e ambiental, 9(4), 603–612. url: http://www.scielo.br/scielo.php?script=sci_arttext&pid=s1415-43662005000400026&lng=pt&tlng=pt. mekaru, t. & uehara, g. (1972). anion adsorption in ferruginous tropical soils. proceedings of the soil science society of america, 36, 296–300. müller, c. j. (2005). pozzolanic activity of natural clay minerals with respect to environmental geotechnics, no. 16299, 125. url: https://www.researchgate.net/profile/christian_mueller4/publication/270277345_pozzolanic_activity_of_natural_clay_minerals_ with_respect_to_environmental_geotechnics/links/54a569020cf257a63608cdf6.pdf. ogata, a. & banks, r. b. (1961). a solution of the differential equation of longitudinal dispersion in porous media. us geological survey professional papers, 34, 411–a. 289determination of the sodium retardation factor using different methods: analysis of their characteristics and impact on the solute movement prediction in a structured brazilian soil önal, y., akmil-başar, c. & sarici-özdemir, ç. (2007). elucidation of the naproxen sodium adsorption onto activated carbon prepared from waste apricot: kinetic, equilibrium and thermodynamic characterization. journal of hazardous materials, 148(3), 727–734. url: http://www.sciencedirect.com/science/article/pii/ s0304389407004025. patterson, r. (1997). domestic wastewater and the sodium factor. in: site characterization and design of on-site septic systems. 100 barr harbor drive, po box c700, west conshohocken, pa 194282959: astm international, 23-23–13. url: http://www.astm.org/ doilink.cgi?stp13784s. paula e silva, f., kiang, c. h. & caetano-chang, m. r. (2003). perfis de referência do grupo bauru (k) no estado de são paulo. geociências unesp, 22(especial), 21–32. url: http://www.revistageociencias.com.br/geociencias-arquivos/22_especial/2.pdf. pejon, o. (1992). mapeamento geotécnico regional da folha de piracicaba (sp): estudos de aspectos metodológicos de caracterização e apresentação de atributos. universidade de são paulo, tese de doutorado (in portuguese). university of são paulo. porfiri, c. (2015). adsorption and transport of imazapyr through intact soil columns taken from two soils under two tillage systems, geoderma, 251–252, 1–9. url: http://www.sciencedirect.com/science/ article/pii/s0016706115000877. shackelford, c. d. (1991). laboratory diffusion testing for waste disposal — a review. journal of contaminant hydrology, 7(3), 177–217. url: http://linkinghub.elsevier.com/retrieve/ pii/016977229190028y. shackelford, c. d. (1994). critical concepts for column testing. journal of geotechnical engineering, 120(10), 1804–1828. url: http://ascelibrary.org/doi/abs/10.1061/(asce)07339410(1994)120:10(1804). shackelford, c. d. (1995). cumulative mass approach for column testing. journal of geotechnical engineering-asce, 121(10), 696– 703. url: http://ascelibrary.org/doi/abs/10.1061/(asce)07339410(1995)121:10(696). shackelford, c. d. & redmond, p. l. (1995). solute breakthrough curves for processed kaolin at low flow rates. journal of geotechnical engineering, 121(1), 17–32. url: http://ascelibrary.org/ doi/10.1061/%28asce%290733-9410%281995%29121% 3a1%2817%29. silva, l. p. (2016). retention and solute transport properties in disturbed and undisturbed soil samples. revista brasileira de ciência do solo, 40. silva, w. t. l. (2012). avaliação físico-química de efluente gerado em biodigestor anaeróbio para fins de avaliação de eficiência e aplicação como fertilizante agrícola. química nova, 35(1), 35–40. url: http://www.scielo.br/scielo.php?script=sci_arttext&pid=s0100-40422012000100007&lng=pt&nrm=iso&tlng=en. van genuchten, m. t. & parker, j. c. (1984). boundary conditions for displacement experiments through short laboratory soil columns 1. soil science society of america journal, 48(4), 866–872. url: www.ars.usda.gov/arsuserfiles/20360500/pdf_pubs/p0846. pdf. van genuchten, m. t. & wierenga, p. j. (1976). mass transfer studies in sorbing porous media i. analytical solutions1, soil science society of america journal, 40(4), 473. url: https://www.soils.org/ publications/sssaj/abstracts/40/4/ss0400040473. vanderborght, j., timmerman, a. & feyen, j. (2000). solute transport for steady-state and transient flow in soils with and without macropores. soil science society of american journal, 64(4), 1305–1317. url: https://www.soils.org/publications/sssaj/abstracts/64/4/1305. washburn, e. w. (1921). note on a method of determining the distribution of pore sizes in a porous material. proceedings of the national academy of sciences of the united states of america, 7(4), 115– 116. url: http://www.pnas.org/content/7/4/115.short. zehe, e. & flühler, h. (2001). preferential transport of isoproturon at a plot scale and a field scale tile-drained site. journal of hydrology, 247(1–2), 100–115. url: http://linkinghub.elsevier.com/retrieve/ pii/s0022169401003705. zuquette, l. v. & palma, j. b. (2006). avaliação da condutividade hidráulica em área de recarga do aqüífero botucatu. rem: revista escola de minas, 59(1), 81–87. url: http://www.scielo.br/scielo.php?script=sci_arttext&pid=s0370-44672006000100011&lng=pt&tlng=pt. untitled earth sciences research journal earth sci. res. j. vol. 13, no.1 (june 2009): 16-29 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti departamento de evolución de cuencas, facultad de ciencias. iguá 4225 c.p. 11400, montevideo, uruguay corresponding author: rossana muzio, rossana@fcien.edu.uy abstract this work presents new results of a detailed geological and structural investigation focusing the easternmost uruguayan mesozoic magmatic occurrences related to the south atlantic opening. lithological descriptions, their stratigraphic relationships and complimentary lithochemical characterizations carried out in the san miguel region (east uruguay) are presented. three volcanic/sub-volcanic units have been recognized. the felsic volcanic association is composed by rhyolitic dacitic flows, mainly with porphyritic textures and sub-alkaline nature and related pyroclastic rocks. the felsic sub-volcanic association is characterized by granophyres of about 25 km2 of exposed area, crosscut by mafic and felsic dykes. finally, a mafic association has been identified characterized by dykes and a small intrusion of gabbroic composition. all these units are mesozoic in age (130 – 127 ma) and according to their chemical nature they correspond to sub-alkaline to weak peralkaline magmas. key words: petrography, geochemistry, mesozoic, uruguay. resumen este trabajo presenta nuevos resultados geológico-estructurales y geoquímicos relacionados con el magmatismo mesozoico del extremo este de uruguay (sierra de san miguel, departamento de rocha), asociados a la apertura del océano atlántico sur. fueron reconocidas para la región estudiada tres unidades volcánicas/subvolcánicas. la asociación volcánica félsica está compuesta por derrames riolíticos – dacíticos, con texturas dominantemente porfiríticas y naturaleza química subalcalina. asociadas a estas riolitas ocurren 16 manuscript receiver: january 12th, 2009. accepted for publication: march 30th, 2009. niveles de rocas piroclásticas. la asociación subvolcánica félsica está compuesta por granófiros con una extensión superficial de aproximadamente 25 km2, cortados por diques máficos y félsicos. fue identificada también una asociación básica compuesta por diques y un cuerpo de gabro. todas estas unidades son de edad mesozoico (130 – 127 ma) y de acuerdo a su carácter químico corresponden a magmas de naturaleza subalcalina a débilmente peralcalina. palabras clave: petrografía, geoquímica, mesozoico, uruguay. introduction the uruguayan geology is underlain by precambrian terrains comprising a wide range of metamorphic rocks and granitic intrusions. five sedimentary basins rest on the cratonic basement, one palaeozoic and the other four mesozoic in age. mesozoic magmatism tectonically related to the breakup of gondwana is well represented in the uruguayan extension of the paraná basin and in the rift-type basins of santa lucía and laguna merin (fig.1). recently, in the offshore punta del este basin have been also recognized relicts of this magmatic event (ucha et al., 2003). the paraná magmatic province (pmp) is one of the largest continental magmatic provinces in the world (peate, 1997). although the main outcrop areas are exposed in brazil, some extensions are well developed in argentina, paraguay and uruguay. furthermore, uruguay has a privileged position because of the presence of the southernmost extension of these magmatic exposures. during the early cretaceous, a strong volcanic activity made up of basaltic and rhyolitic rocks occurred in south eastern uruguay along the border of the paraná basin. particularly, these exposures are well represented in rift type basins located in the eastern portion of uruguay (laguna merín basin) and comprise from basic to acidic units on which this paper is concerned. several petrologic works have been carried out in the south eastern portion of uruguay, with particular emphasis in the felsic volcanic/sub-volcanic units (muzio, 2000; kirstein et al., 2000; kirstein et al., 2001; muzio et al., 2002; lustrino et al. 2003; lustrino et al. 2005). as main results can be pointed out that the petrologic and isotopic features exhibited by these units differ completely from their temporal correlatives of the paraná province (palmas and chapecó rhyolites, according to bellieni et al. 1986). recent studies performed by muzio et al. (2004) and muzio et al. (2005) in the san miguel region, allowed the identification of a gabbroic body which 17 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region cuenca de paraná b r a z i l a r g e n t i n a sedimentary basins a paraná basin b santa lucía basin c laguna merín basin crystalline basement 54º00´ 31º 00´ a b c exposed areas of mesozoic magmatism jaguarão n 0 100 km salam (santa lucía aiguá merín) lineament montevideo sierra san miguel -35º00’ figure 1. structural framework and location of the area, modified after ucha et al. (2003). has been mapped at semi-detailed scale (fig. 2). this is the first basic intrusive body that has been recognized out-cropping in the southern portion of the laguna merin basin, closed to an important gravimetric anomaly (+90 mga; reytmair, 2001). according to rossello et al. (2000), this gravimetric anomaly results from the development or reactivation of deep parallel fractures, trending n70° to e-w, which affected the crust during gondwana fragmentation. these fractures, recognized through strong structural, magnetic and gravimetric lineaments, became mantle feeding structures that allowed the emplacement of basic/mafic rocks which caused the excess of gravity in the region (rossello et al, 2000; veroslavsky et al., 2002). the aim of this paper is to present the petrologic and structural characterization of the units exposed in the sierra san miguel region, corresponding to the southernmost border of the laguna merín basin (east uruguay), with particular emphasis in the gabbroic intrusion. geologic setting and petrography the sierra san miguel complex is located in the department of rocha (east uruguay), 340 km eastwards montevideo, the capital of the country. it represents an important geomorphologic feature, of about 25 km length and regional trend ew within the wetlands and quaternary sediments of the southern extreme of the laguna merin basin (fig. 2). the structural framework controlling the development and evolution of the santa lucía and laguna merin basins is underlain by the salam (santa lucia – aiguá – merin lineament, after rossello et al.; 2000). this structural feature constitutes a transtensive tectonic corridor of the strike-slip faults (ca. 450 km), cross-cutting the río de la plata craton. as it has been described by rossello et al. (2001), either two pull-apart depocentres (santa lucia and laguna merín basins) at its extremes and the emplacement of mesozoic volcanic/sub-volcanic rocks have been controlled by this structure. according to these authors, the salam compounds two main tectonic phases: (i) a jurassic to early cretaceous extensional stage, represented by intraplate magmatism synchronous with the pmp, and (ii) an aptian – albian dextral transtensive stage. in this context, the tectonic development of the sierra san miguel complex is controlled by the aiguá india muerta – chuy fault, one of the components of the salam lineament that conditioned the opening and extensional development of the laguna merín basin (rossello et al., 2000). the pioneer geological works in the sierra san miguel correspond to walther (1927), caorsi & goñi (1958) and bossi & fernández (1963) and it has been described as a volcanic/sub-volcanic association composed by granophyres and, subordinated rhyolites. these lithologies were later integrated to the arequita formation by bossi (1966). the only available radiometric data by k/ar systematic (whole rock analyses) yielded ages between 125 – 130 ma (bossi, 1969). after recent geological studies developed by muzio et al. (2004), three petrographic associations have been identified, each-one related to different tectomagmatic stages (fig.2). unit 1 – felsic sub-volcanic association (fsa) the fsa is located mainly in the east and central portions of the sierra san miguel which occurs as an irregular body along an ew to ne axis. it is composed by strongly fractured granophyric rocks, cross-cut by sub-vertical basic and acidic/felsic dykes with structural trend n65º and ew respectively. the granophyres are progressively covered by rhyolitic lava flows. the fracture system (sub-horizontal and vertical unfilled fractures) outlining a roughly polyhedral pattern can also be observed. these granophyres are mineralogical and petrographycally homogeneous. the mineral association of these rocks is essentially anhydrous. they are medium to fine grain-sized, with equigranular and occasionally porphyritic hypidiomorphic textures (up to 10% of plagioclase phenocrysts). it can be observed the following mineralogical assemblage: quartz alkali feldspar intergrowths (micrographic/granophyric texture) and 18 rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti 19 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region e st e ro d e lap u n ta n e g ra d el tig re c a ñ a d a r in có n br av o s an m ig ue l w et la nd s v ig ia h il l r od eo v ie jo cr ee k c a r b o n e r a h il l s a nm ig u e l s an m ig ue lf or t s ar an dí cr ee k 18 de ju lio s a ra n d í b r a z il 70 0 70 2 70 4 70 6 70 8 5 3 º4 1 ’ 3 3 º3 2 ’ 71 0 71 2 70 2 70 4 70 6 70 8 71 0 71 2 70 0 69 8 69 6 69 4 69 2 69 0 68 8 69 8 69 6 69 4 68 8 69 0 69 2 27 2 27 0 26 8 26 6 26 4 27 2 27 0 26 8 26 6 27 0 lo s d if u n to s h il l o m b ú h il l 1 9 4 k m . 1 k m . 2 k m . 3 k m . 0 1 n lo ca l r ou te r oa d / l an e c re ek s in te rn at io na l b or de r v ill ag e h ill s r hy ol ite s g ab br o g ra no ph yr es f au lts / f ra ct ur es c en oz oi c de po si ts c re e k c re ek cr ee k u r u g u ay br az il argentin a s tu di ed ar ea m on te vi de o d e la pl at a r iv er at lan tic oc ea n f ig u re 2 . g eo lo g ic al sk et ch o f th e s ie rr a s an m ig u el re g io n , af te r m u zi o et al . (2 0 0 4 ). partially altered clinopyroxene (augite) often exhibiting uralitization processes. apatite and opaque minerals are frequent as accessory minerals. in the central-north part of the area, the granophyres are hardly affected by faulting with trends n70º. also, a synmagmatic foliation n250º 260º can be observed. this structural trend is coincident with the emplacement of mafic and felsic dykes. unit 2 – mafic association (ma) two different types of basic rocks have been recognized, according to their geological setting. the first type corresponds to basic dykes which vertically cross-cut the granophyres. as it was mentioned previously, they are porphyritic basic dykes with structural trend n65-70, 75º. they have porphyritic texture with plagioclase phenocrystals (up to 10 %) in a subophitic to intersertal groundmass. the mineral assemblage is composed by plagioclase, orthopyroxene and clinopyroxene (augite), opaque minerals and apatite. the second type of basic rocks is represented by an intrusive body – named as san miguel gabbro (smg). the main outcrop area is located around the central-north region of the sierra san miguel (s33º43’; w53º38’), and it can be described as a sub-horizontal intrusion cross-cutting the granophyric rocks. it is strongly fractured with sub-horizontal and sub-vertical joints. typical spherical structures due to alteration processes, ranging in diameter from 40 to 80 cm, can be observed. according to the iugs classification criteria (le maitre, 1989) this intrusive body can be described as a leucocratic gabbro/dolerite with massive structure, with equigranular and subophitic texture, medium to coarse grain sized, composed by plagioclase (an55-45), altered olivine, orthopyroxene, clinopyroxene (augite), opaque minerals and apatite. occasionally amphibole as deuteric alteration – is present. other small gabbroic intrusions appear as wedges or secondary dykes; all injected through the granophyres. euhedral to subhedral plagioclase crystals surrounded by micrographic/ granophyric intergrowths are present. this petrographic feature is also observed as an interstitial granophyric intergrowth, and occurs plagioclase and clinopyroxene crystals with intergranular arrangement, in association with k-feldspar and quartz. according to shelley (1992) this kind of texture is consistent with residual crystallization under eutectic conditions and corresponds either with the emplacement of shallow intrusions or with fast degassing processes. unit 3 – felsic volcanic association (fva) they correspond mainly to pinkish grey colored rhyolites and compound at least three lava flows softly dipping to the north. they are distributed along the north, northwestern and western region of the sierra san miguel. these lithologies are grouped in the arequita formation (bossi, 1966) and can be separated, according to their petrographic and structural features in: a) rhyolitic lavas, occasionally with porphyritic textures (approximately 10 to 15% of alkali feldspar and/or quartz phenocrysts), with vesicular levels filled by calcite, zeolite and quartz. most of these vesicles are stretched according to flow directions such as n170º to ew. microscopically, spherulitic texture and flow structures formed during the cooling of the rhyolitic lava can be observed. some of the porphyritic varieties present abundant glass phenocrysts, round shaped and brown to dark black colors. some volcaniclastic facies in association with the rhyolites have been found and according to mcphie et al. (1993) descriptive criteria they correspond to volcanic lithic breccias and autoclastic breccias (fig. 3). b) rhyolitic dykes with porphyritic texture trending n60-70º and cross-cutting all the lithologies are present in the study area. they present quartz and k-feldspar phenocrysts in an aphyric matrix composed by 20 rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti quartz, k-feldspar (orthoclase?) and clinopyroxene. their stratigraphic relationship with the mafic dykes was not observed. lithogeochemistry chemical composition of ten samples (one granophyre, three gabbros, one mafic dyke, three felsic dykes and two rhyolites) of the sierra de san miguel complex was analyzed. major and trace elements were obtained by icp – aes techniques whereas ree by using icp – ms, performed at acme laboratories inc., vancouver, canada. the analytical data are listed in table i and illustrated in fig 3-7. they were plotted in different classification and variation diagrams and they were also compared with similar rocks performed by kirstein et al. (2001) and lustrino et al. (2005) in order to evaluate the data for the region. the sio2 variation diagrams (fig.4) show no extremely high contents in the felsic volcanic samples. however, they present high contents of l.o.i. (>3%). the analyzed samples have sio2 ranges from 71.1 to 48.49 wt.%; tio2 from 1.48 to 0.58 wt.%; al2o3 from 16.22 to 11.6 wt.%; fe2o3 from 14.2 to 5.01 wt.%; mgo from 9.11 to 0.23 wt.%; cao from 12.04 to 0.82 wt%; na2o from 3.33 to 2.19 wt.% and k2o from 4.89 to 0.2 wt.%. when plotted against sio2, smooth trends are showed. tio2, al2o3, fe2o3, mgo and cao decrease with increasing sio2 content. na2o does not show a trend at all, while k2o increases towards the rhyolitic terms. furthermore, na2o and k2o present an inverse behavior when compared with lustrino et al. (2005) data. trace element contents present a short range when they are compared with published data (lustrino et al., 2005; kirstein et al., 2000). rb varies from 184.4 to 5.4; zr from 387.1 to 58; y from 76.6 to 22.2; nb from 24 to 3.7; ba from 721.1 to 111 and sr from 207.8 to 85.4 (fig. 5, table i). the ratios between incompatible elements (fig. 6) vary within narrow ranges (zr/nb= 15.03±1.05; zr/y=4.11±1.54; y/nb= 4.39 ± 1.72). these values are very similar to the treinta y tres magma type pointed out by turner et al. (1999); kirstein et al. (2000) and lustrino et al. (2005). the rock samples have been classified using the r1-r2 diagram (de la roche et al., 1980) and tas diagrams (le maitre et al., 1989) (fig. 7a and 7b). two samples of the plutonic rocks plot in the gabbro norite field of the r1-r2 diagram. the volcanic rocks plot mainly in the rhyolitic and dacitic fields and one sample in the basaltic-andesitic field. all samples show a sub-alkaline affinity according to irvine and baragar (1971) diagram (fig. 7c). however, the analyzed data are not enough to establish any petrogenetic link among the samples. the agpaitic index calculated for the samples (a.i. = molar (na+k)/al) ranges from 0.24 to 0.74. the rhyolites range from 0.71 to 0.74; dacites present values of 0.66 and andesites around 0.58. the lower values correspond to the gabbro – norite samples. none of the samples show a peralkaline character as described for the arequita formation by muzio et al. (2002). hence, they have been classified on the basis 21 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region figure 3. lithic breccias and autoclastic breccias associated with the felsic volcanics in sierra san miguel region. 22 rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti 40 50 60 70 80 0.0 1.6 3.2 4.8 6.4 8.0 sio2 k 2o 40 50 60 70 80 2 3 4 sio2 n a o 2 40 50 60 70 80 0 1 2 sio2 t io 2 40 50 60 70 80 0 1 2 3 4 5 6 7 8 9 10 sio2 m g o 40 50 60 70 80 0 10 20 sio2 c a o 40 50 60 70 80 10 11 12 13 14 15 16 17 18 19 20 sio2 a l o 2 3 40 50 60 70 80 0 10 20 sio2 f e 2 o 3 t figure 4. major and minor element diagrams for samples from the san miguel region. symbols: � felsic dykes; � felsic lavas; � gabbros; � basic dyke; � granophyre. 23 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region table 1. geochemical data of selected samples from the sierra san miguel region. samples sm50-1 sm50-2 sm50-3 sm008-7 sm008-8 sm008-9 sm51 sm13 sm27 sm29-1 sio2 (wt.%) 65,01 64,24 65,1 58,71 48,98 48,49 56,41 69,72 71,1 69,58 tio2 0,88 0,92 0,87 0,89 0,58 0,59 1,48 0,76 0,63 0,7 al2o3 12,58 12,38 12,38 13,99 16,22 16,13 12,67 12,74 11,6 12,81 fe2o3 5,87 6,43 5,66 8,89 9,53 9,8 14,2 6,15 5,01 5,66 mno 0,14 0,14 0,13 0,17 0,15 0,15 0,2 0,11 0,16 0,13 mgo 1,3 1,44 1,28 3,28 9,11 8,95 2,85 0,3 0,67 0,23 cao 3,95 4,18 3,99 6,44 11,78 12,04 6,28 1,71 1,75 0,82 na2o 3,33 3,13 2,97 3,18 2,25 2,19 2,79 3,24 2,42 2,28 k2o 2,77 2,76 3,12 2,42 0,2 0,25 2,54 3,81 4,22 4,89 p2o5 0,26 0,23 0,24 0,28 0,09 0,11 0,19 0,2 0,17 0,16 cr2o3 0,004 0,005 0,015 0,013 0,053 0,068 0,019 0,015 0,041 0,008 loi % 3,8 3,9 4 1,5 1 1,2 0,3 1,1 2,1 2,6 tot/c % 0,03 0,02 0,03 0,03 0,03 0,05 0,02 0,05 0,05 0,04 tot/s % 0,01 <.01 <.01 <.01 0,01 0,06 0,01 0,01 0,01 0,01 sum % 99,89 99,76 99,76 99,77 99,98 100 99,95 99,86 99,89 99,88 rb (ppm) 135,8 127,3 142,5 89,9 7,3 5,4 88,8 150,7 158,8 184,5 ba 625 599,2 583,9 458,1 111 112,6 365,4 721,1 654,6 712,6 sr 136,8 146,3 146,3 207,8 133,3 139,1 123,1 125,9 92,8 85,4 zr 301,1 276,4 289,6 237,1 72,2 58 189,4 387,1 293,5 327,9 y 57,6 56 56,9 56,4 22,2 22,6 50,2 76,6 64,4 58,1 ga 15,9 16,1 15,8 19,4 14 14,5 19,7 18 15,9 17 hf 8,6 8,1 8,5 7,2 2,2 1,8 5,7 11 8,7 9,3 nb 21,6 19,5 20,8 16,6 4,6 3,7 13,4 24 18,7 20,4 ta 1,4 1,4 1,5 1,2 0,4 0,3 1 1,7 1,4 1,5 th 16 14,2 15,4 7,4 1,8 1,2 8,1 14,2 14,4 14,7 u 2,5 2,3 2,6 1,4 0,4 0,3 1,3 1,7 2,5 2,7 ni 29,7 18,8 67,5 31,5 110,5 118,1 65,2 43,9 152,1 39 la 38,8 37,5 40,1 28,7 7,3 7,3 23,1 46,6 42,4 40,7 ce 96,4 91,8 99,4 68,7 16,9 16,1 54,2 107,2 99,6 102,2 pr 10,71 10,06 10,77 8,41 2,24 2,18 6,47 13,1 11,38 10,17 nd 44,3 41,3 43,6 36,8 10,2 9,9 27,3 54,8 45 41 of the alumina saturation index (a.s.i. = molar al/ (na+k+ca)) yielding values from 0.63 to 1.24. most of the samples present a metaluminous character (a.s.i. <1), and only three of them present a weak peraluminous character (a.s.i.>1). the samples present moderate light to heavy ree fractionation in chondrite normalized plot [(la/yb) í = 5.02 to 2.25]. in one hand, the rhyolites and the acidic to basic dykes have very strong negative eu anomalies (eu/eu* = 0.23 to 0.19), and a flat to slightly concave pattern of hree. on the other hand, the gabbro – norite samples show a flat pattern when ree normalized against chondrite c1 (la/yb) n = 5) and the highest values for eu/eu* (from 0.26 to 0.28) yielding less marked negative eu anomalies (fig. 8a). the rocks analyzed do not show typical patterns in the primitive mantle normalized spidergram when compared with the early cretaceous santa lucía type and aiguá type magmas (turner et al., 1999; kirstein et al., 2000). negative peaks of ba, nb, sr and pb are present. only the rhyolitic samples do not show pb troughs (fig. 8b). the plutonic rocks show almost a flat pattern in primitive mantle normalized diagram. enrichment in ba, th and u is observed. discussion and conclusions field relationships, petrography and geochemistry indicate a bimodal character for san miguel complex and reflect the general character of extensional magmatism. this complex is associated with a regional fault system responsible for the opening and extensional development of the laguna merín basin. 24 rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti samples sm50-1 sm50-2 sm50-3 sm008-7 sm008-8 sm008-9 sm51 sm13 sm27 sm29-1 sm 9,1 8,9 9 8,5 2,7 2,5 6,1 11,7 10,3 8,6 eu 1,87 1,86 1,79 2,26 0,77 0,79 1,55 2,35 2,02 1,84 gd 9,35 9,08 9,41 9,03 3,08 3,24 7,4 11,97 10,31 9,06 tb 1,8 1,7 1,75 1,78 0,64 0,66 1,41 2,36 2,01 1,81 dy 9,85 9,57 9,84 9,68 3,58 3,92 8,07 12,87 10,92 10,08 ho 1,94 1,95 2,05 1,95 0,75 0,81 1,76 2,53 2,16 1,99 er 6,08 5,95 6,18 5,91 2,32 2,37 5,25 7,59 6,55 6,06 tm 0,98 0,91 0,92 0,87 0,36 0,36 0,82 1,16 0,96 0,94 yb 5,86 5,72 5,83 5,69 2,33 2,19 5 7,09 6,17 5,82 lu 0,87 0,85 0,88 0,85 0,35 0,33 0,75 1,06 0,94 0,84 zr/nb 13,94 14,17 13,92 14,28 15,70 15,68 14,13 16,13 15,70 16,07 zr/y 5,23 4,94 5,09 4,20 3,25 2,57 3,77 5,05 4,56 5,64 y/nb 2,67 0,46 0,49 0,38 0,10 0,09 0,47 0,39 0,54 0,56 (la/yb)n 4,75 4,70 4,93 3,62 2,25 2,39 3,31 4,71 4,93 5,02 eu/eu* 0,20 0,21 0,19 0,26 0,27 0,28 0,23 0,20 0,20 0,21 rb/sr 0,99 0,87 0,97 0,43 0,05 0,04 0,72 1,20 1,71 2,16 agpaitic index 0,673 0,657 0,667 0,561 0,241 0,24 0,579 0,742 0,736 0,705 a.s.index 0,82 0,801 0,813 0,728 0,643 0,629 0,684 1,034 1,007 1,239 25 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region thus, the sierra san miguel morphological feature shows a clear correspondence with the main faults of the basin and the aiguá – india muerta – chuy lineament, trending n60º which is also discordant with other regional structures of the punta del este basement. the studied area shows structural concordance between the conspicuous foliation (n60º – 70º) exposed in the granophyres and the brittle deformation trends (e-w and n70º) affecting all the mesozoic rocks, including the regional trend of the sierra san miguel the stratigraphic array observed for the sub-volcanic/volcanic lithologies (gabbro, granophyres and rhyolites) shows that the fault system had important vertical components synchronous with the magmatic events in order to formulate a close spatial-temporal relation between these rocks. the san miguel gabbro should be linked with the tectomagmatic event responsible of the opening of the laguna merín basin 0 100 200 300 400 20 30 40 50 60 70 80 zr y 0 100 200 300 400 0 10 20 30 zr n b 0 100 200 300 100 200 300 400 500 600 700 800 sr b a 45 50 55 60 65 70 75 80 0 20 40 60 80 100 120 140 160 180 200 sio2 n i 0 5 10 15 20 0.00 0.20 0.40 0.60 0.80 1.00 1.20 1.40 th t h /n b figure 5. trace element diagrams for samples from the san miguel region. symbols: � felsic dykes; � felsic lavas; � gabbros; � basic dyke; � granophyre. 26 rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti 0 100 200 300 400 0 10 20 30 zr n b (a) 0 100 200 300 400 20 30 40 50 60 70 80 zr y (b) 0.001 0.01 0.1 1 10 40 50 60 70 80 rhyolite rhyodacite-dacite andesite tran sub-ab bas-trach-neph com/pan trachyte phonolite zr/tio2*0.0001 s io 2 ( c) figure 6. trace element diagrams and trace element ratio diagram for samples from the san miguel region. symbols: � felsic dykes;� felsic lavas; � gabbros; � basic dyke; � granophyre. 16 14 12 10 8 6 4 2 0 40 50 60 70 80 sio2 n a o + k o 2 2 (b) (c ) 35 40 45 50 55 60 65 70 75 80 85 0 2 4 6 8 10 12 14 16 18 20 alkaline subalkaline sio2 n a 2 o + k 2 o gabbronorite olivine gabbro alkali gabbro syenogabbro syenodiorite essexite nepheline syenite syenite quartz syenite alkali granite syenogranite monzogranite granodiorite diorite gabbro tonalite quartz monzonite monzonite monzonitediorite m on zo ga bb ro 400 400 800 800 1200 1200 1600 1600 2000 2000 2200 2400 2800 0 0 200 600 1000 1400 1800 r1 r 2 (a) figure 7. classification diagrams. (a) multicationic classification diagram (de la roche et al. 1980) ; (b) tas classification diagram (le maitre et al. 1989); (c) tas classification diagram (irvine and baragar, 1971). symbols: � felsic dykes; � felsic lavas; � gabbros; � basic dyke; � granophyre. that also allowed the generation of a great depocentre filled by basic lavas named puerto gómez formation (bossi, 1966), based on the geological and structural framework. these evidences allow us to point out a continuous reactivation of the basement which conditioned the faulting and jointing style that affected all the mesozoic rocks. the structural trends presented by the dykes and the spatial arrange exposed by faults and fractures, most of them concordant with ancient planes of weakness along the basement, remark the presence of dilatants situations in agreement with the dextral sense of displacement proposed for the fault system cebollatí – merín and aiguá – india muerta – chuy. this structural framework, according to rossello et al. (2000) determines the development of the salam lineament and was the main mechanism responsible for the genesis and tectomagmatic evolution of the laguna merín basin. the spatial distribution, type and volume of magmatism related to the laguna merín basin, allow us to suggest that it represents a magmatic pull-apart type depocentre, sub parallel to the atlantic margin and developed in a pre-structured cratonic domain. according to this, the volcanic rocks exposed in the eastern region of uruguay would represent different magmatic pulses synchronous with the tectonic reactivation of the basement. this would explain the coexistence of intrusive/shallow and effusive terms in a relative short time of emplacement. as a result of the drill campaign performed by private founds near lascano city (www.ume.com.uy), the rocks described correspond to gabbros/dolerites and granites (granophyres), providing us valuable information close to the sierra san miguel region. the lascano geophysical anomaly has been interpreted as being created by gabbro to granitic (granophyres) intrusive rocks underlying a package of magnetic basalts and felsic volcanic rocks. whatever be the trigger responsible for the genesis of this gravimetric anomaly, its development can be related to the tectonic evolution of the salam lineament during the early stages of the south atlantic opening (rossello et al., 2007). according to the geological background and field relationships among the different units, the basic and acidic magmatism are considered mesozoic in age and related with the magmatic events of continental scale predecessors of the opening of the south atlantic ocean (lustrino et al. 2003). further chemical analyses should be carried out in order to explain petrogenetic correlations between these magmas and the main magma types recognized by other authors. acknowledgments the authors are grateful to the comisión sectorial de investigación científica for the financial support to 27 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region s a m p le /p ri m it iv e m a n tl e s a m p le /c 1 c h o n d ri te 1 10 100 500 la ce pr nd sm eu gd tb dy ho er tm yb lu (a) 0.0001 0.001 0.01 0.1 1 10 100 1000 cs rb ba th u nb ta la ce pb pr sr nd sm zr hf eu gd tb dy y ho er tm yb lu (b) figure 8. (a) ree patterns for samples from the san miguel region normalized according to taylor and mc lennan (1985). (b) spider diagrams normalized according to sun and mc donough (1986). symbols: � felsic dykes; � felsic lavas; � gabbros; � basic dyke; � granophyre. the project “magmatismo mesozoico de la región sureste de uruguay”, grant 2005 2007. special thanks to dra. l.sánchez and to the editorial board for the revision and helpful comments. references bellieni, g.; comin-chiaramonti, p.; marques, l.s.; melfi, a.j.;. nardy, a.j.r.; papatrechas, c.; piccirilo, e.m; roisemberg, a. and stolfa, d. (1986). petrogenetic aspects of acid and basaltic lavas from the paraná plateau (brazil): mineralogical and petrochemical relationships. journal of petrology 27: 915 – 944. bossi, j. (1966). geología del uruguay. departamento de publicaciones de la universidad de la república, montevideo, 460 p. bossi, j. 1969. bossi, j. and fernández, a. (1963). evidencias de diferenciación magmática hacia el final del gondwana uruguayo. boletín da universidade de paraná, geología nº 9, pp. 1 – 20. caorsi j. and goñi, j. (1958). geología uruguaya. boletín del instituto geológico del uruguay nº 37, montevideo, 79 p. de la roche; h.; leterrier,j.; grandclaude, p. and marchal, m. (1980). a classification of volcanic and plutonic rocks using r1 – r2 diagram and major element analysis. its relationship with current nomenclature. chemical geology 29: 183 – 210. irvine, t. n. and baragar, w.r. (1971). a guide to the chemical classification of the common igneous rocks. canadian journal of earth sciences 8: 523 – 546. kirstein, l.a.; hawkesworth, c.j. and garland, f.g. (2001). felsic lavas or rheomorphic ignimbrites: is there a chemical distinction? contributions to mineralogy and petrology 142: 309 – 322. kirstein, l.a.; peate, d.w.; hawkesworth, c.j.; turner, s.p.; harris, c. and mantovani, m.s.m. (2000). early cretaceous basaltic and rhyolitic magmatism in southern uruguay associated with the opening of the south atlantic. journal of petrology 41, pp. 1413 – 1438. le maitre, r.w.; bateman, p.; dudek, a. ; keller, j.; lameyre, l.; le bas, m.j.; sabine, p.a.; schmid, r.; sorensen, h.; strekeisen, a.; woolley, a.r. and zanettin, b. (1989). a classification of igneous rocks and glossary of terms. blackwell – oxford eds., 193 p. lustrino, m.; gomes, c.b.; melluso, l.; morbidelli, l.; muzio, r.; ruberti, e. and tassinari, c.c. (2003). early cretaceous magmatic activity in southeast uruguay: trace element and sr-nd isotopic constraints. proc. iv south american symposium on isotope geology, salvador, brazil, pp. 596 – 597. lustrino, m.; melluso, l.; brotzu, p.; gomes, c.b.; morbidelli, l.; muzio, r.; ruberti, e. and tassinari, c.c. (2005). petrogenesis of the early cretaceous valle chico igneous complex (se uruguay): relationships with paraná – etendeka magmatism. lithos 82 (3/4): 407 – 434. mcphie, j.; doyle, m. and allen, r. (1993). volcanic textures. a guide to the interpretation of textures of volcanic rocks. centre for ore deposit and exploration studies. university of tasmania, australia, 198 p. muzio, r. (2000). evolução petrológica e geocronologia do maciço alcalino valle chico, uruguay. phd thesis, instituto de geociências e ciências exâtas – universidade estadual paulista, rio claro, brazil, 171 p. muzio, r.; artur, a.c. and wernick, e. (2002). petrological and geochemical evolution of the valle chico alkaline massif, southeastern uruguay. international geology review 44 (4): 352 – 369. muzio, r.; veroslavsky, g. and morales, e. (2004). geología de la sierra de san miguel, departamento de rocha (uruguay). actas iv congreso uruguayo de geología, montevideo, versión cd. 28 rossana muzio, elena peel, ethel morales, gerardo veroslavsky, bruno conti muzio, r.; veroslavsky, g., pascale, a.; cabrera, j. and furtado, a. (2005). the san miguel gabbro and its relation with mesozoic magmatism from southeastern uruguay. anais do iii simpósio de vulcanismo e ambientes asociados, cabo frio, brazil, pp. 267 – 269. peate, d.w. (1997). the paraná – etendeka province. in: mahoney, j.j. & coffin, m.f. (eds.) large igneous provinces, continental, oceanic and planetary flood volcanism. geophysical monograph 100, pp. 217 – 245. reitmayr, g. (2001). una espectacular peculiaridad uruguaya: la anomalía gravimétrica de laguna merín. actas xi congreso latinoamericano de geología y iii congreso uruguayo de geología, montevideo, cd. rossello, e.; santa ana, h. and veroslavsky, g. (2000). el lineamiento santa lucía aiguá merín (uruguay): un corredor extensivo y transcurrente dextral preecursor de la apertura atlántica. revista brasileira de geociëncias 30 (4): 749 – 756. rossello, e.; veroslavsky, g. and de santa ana, h. (2001). influencias del lineamiento transtensivo santa lucía – aiguá – merín sobre el emplazamiento del magmatismo alcalino cretácico del cratón del rio de la plata (uruguay): aportes a la prospección diamantífera. revista brasileira de geociencias 31 (4): 661 – 662. rossello, e.; veroslavsky, g.; masquelin, h. and de santa ana, h. (2007). el corredor tectónico juro-cretácico santa lucía – aiguá merín (uruguay): evidencias cinemáticas transcurrentes destrales y controles estructurales preexistentes. revista de la asociación geológica argentina 62 (1), 92 – 104. shelley, d. (1992). igneous and metamorphic rocks under microscope. chapman – hall (eds.). sun, s. and mc donough, w.f. (1989). chemical and isotopic systematics of oceanic basalts: implications for mantle composition and processes. in: magmatism in oceanic basins. geological society of london special publication 42, 313 – 345. taylor, s.r. and mc lennan, s.m. (1985). the continental crust: its composition and evolution. blackwell, oxford eds. ucha, n.; de santa ana, h. and veroslavsky, g. (2003). la cuenca punta del este: geología y potencial hidrocarburífero. in: cuencas sedimentarias del uruguay: geología, paleontología y recursos naturales. mesozoico. veroslavsky, g.; ubilla, m. and martínez, s. (eds.); dirac montevideo; p. 171 – 190. veroslavsky; g.; rossello, e. and de santa ana, h. (2002). la anomalía gravimétrica de la laguna merín (uruguay): origen y expectativas en la exploración mineral. revista uruguaya de geología vol. 1 (2), montevideo. walter, k. (1927). contribución al conocimiento de las rocas basálticas de la formación de gondwana en la sud-américa. boletín del instituto de geología y perforaciones nº 9, montevideo, 43 p. winchester, j.a. and floyd, p.a. (1977). geochemical discrimination of different magma series and their differentiation products using immobile elements. chemical geology 20, 325 – 343. www. ume.com.uy , uruguay mineral explorations inc. website. 29 mesozoic magmatism in east uruguay: petrological constraints related to the sierra san miguel region geociencias-vol 13-2 2009.vp earth sciences research journal earth sci. res. j. vol. 13, no. 2 (december 2009): 108-118 characterization of vermiculite by xrd and spectroscopic techniques a. campos, s. moreno, r. molina1 1 estado sólido y catálisis ambiental, departamento de química, facultad de ciencias, universidad nacional de colombia. ak 30 no. 45-03, bogotá, colombia; fax: 57-1-3165220 e-mail autor:ramolinag@unal.edu.co abstract a natural mineral from santa marta-colombia used as the starting material in the synthesis of pillared clays has been characterized by several techniques, including x-ray diffraction, x-ray fluorescence, electronic paramagnetic resonance, aluminum nuclear magnetic resonance and scanning electron microscopy. the information revealed that the mineral corresponds to trioctahedral vermiculite. the identification of this mineral is valuable in the control of reduction charge and pillaring processes on these materials to obtain more complex solids like the ones required to specific catalytic applications. key words: vermiculite, structural formula, layer charge. resumen un mineral natural de la región de santa marta en colombia, el cual usado como el material de partida en la síntesis de arcillas pilarizadas, ha sido caracterizado por diversas técnicas tales como difracción de rayos-x, fluoresencia de rayos-x, resonancia electrónica paramagnética, resonancia magnética nuclear de aluminio y microscopía electrónica de barrido. la información en conjunto indica que el mineral corresponde a vermiculita trioctaédrica. la identificación de este mineral es muy importante en la comprensión y el control de los procesos de reducción de carga y de pilarización de estos materiales para la obtención de sólidos más complejos, como los requeridos en aplicaciones catalíticas específicas. palabras clave: vermiculita, fórmula estructura, capa de carga. 1. introduction vermiculite is a clay mineral usually of secondary origin due to the alteration of mica, pyroxene, chlorite or similar minerals (brown 1961). the substitutions of si by al in tetrahedral sheet predominate in vermiculite. in this way, the negative charge generated on the tetrahedral sheets limits the expansion properties of the clay and this factor determines the high layer staking of its structure (mac ewan and wilson 1980; tunega et al. 2003). 108 manuscript received: 24/03/2009 accepted for publication: 26/05/2009 enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:17 p p p composite 133 lpi at 45 degrees from the catalytic point of view, the vermiculite is a very attractive material due to the thermal resistance (suvorov and skurikhin 2003) and the number of tetrahedral substitutions, which ensure the presence of a larger number of brønsted-type acid sites (del rey-pérez-caballero and poncelet et al. 2000); these substitutions are less numerous in other hydrous 2:1 minerals such as smectites. on the whole, these properties offer considerable interest in the production of pillared clays, which are distinguishable from an intercalated layered solid due to their micro and/or mesoporosity and high thermal stability with preservation of the layer stacking (moreno et al. 1997, schoonheydt et al. 1999; stefanis et al. 2006). the development of pillared clays from vermiculite during the past few years has offered particular interest especially as a heterogeneous catalyst in acid-catalyzed reactions (campos et al. 2005; campos et al. 2007; campos and gagea et al. 2008; campos and moreno et al. 2008; cristiano et al. 2005; del rey-pérez-caballero and poncelet et al. 2000; del rey-pérez-caballero and sánchez et al. 2000, hernández et al. 2007). however, as direct intercalation of vermiculite is impossible due to a very high potential of stabilization of interlayer cations (suvorov and skurikhin 2003) it has been necessary to develop alternative methodologies to achieve pillared vermiculite (cristiano et al. 2005; f. del rey-pérezcaballero 2000). one of these methodologies has been the hydrothermal treatment prior to a conventional process of intercalation, in order to reduce negative charge and make possible the subsequent cationic exchange (cristiano et al. 2005). the catalytic profile of al and al-zr modified vermiculites by means of this methodology has confirmed the potential expected for this acidic acid solid, since they are highly active with respect to other clays in alkane hydroisomerization (campos 2005; campos et al. 2007; campos and gagea et al. 2008; campos and moreno et al. 2008; cristiano et al. 2005; del rey-pérez-caballero and poncelet 2000; del rey-pérez-caballero and sánchez et al. 2000; hernández et al. 2007). in order to produce practical information to control the modification processes described previously, to achieve a clay mineral with potential catalytic and adsorptive properties, this paper focuses on the characterization of natural vermiculite issued from a colombian deposit. x-ray diffraction (xrd) and several spectroscopic techniques such as electronic paramagnetic resonance (epr) and nuclear magnetic resonance (nmr) were used. 2. experimental 2.1 materials the vermiculite selected for this work was labeled v, and it corresponds to a commercial mineral, which comes from a natural deposit in the santa marta region in colombia. its characterization was carried out with the fraction of a particle size smaller than 150 ìm, separated from the commercial form by sieve, without purification or fractioning treatment. 2.2 characterization methods the cationic exchange capacity was determined on the clay previously interchanged with an ammonium acetate solution, by means of the micro-kjeldahl method (chapman 1965). the x-ray diffraction study obtained from flakes, which had been gently pressed onto the plates by using a philips pw 1820 diffractometer (ká radiation of cu , ë=1.54056 å, 40 ma, 40 kv) in 2è geometry and bragg–brentano configuration, a step size of 0.05 and a step time of 2 s. diffraction patterns were taken at room temperature, 20 °c and 65% of humidity on the average. the identification of the mineral was assessed according to the position of basal reflection 001 in three patterns: i) natural sample, ii) after solvation in the presence of ethylene glycol for 24h and iii) after heating at 500 ºc for 2 h. in addition, mg2+ and k+ saturations were performed in the mineral to achieve a better characterization (thorez 1976). adsorption-desorption isotherms of nitrogen at the temperature of liquid nitrogen were analyzed using a micromeritics tristar 3000 instrument on samples previously degassed at 200 °c for 6 h. scanning electron microscopy images were recorded in a philips scanning electron microscope xl30 feg. the gravimetric and differential thermal-analysis were performed at a heating speed of 10 ºc min-1 in air atmosphere using a thermal analyzer tg and dsc rheometric instruments. the chemical analysis was performed by x-ray fluorescence in an xrf 2400 instrument. in addition, the fe2+ analysis was carried out through the analytic method described in wilson m. (wilson 1995). electron paramagnetic resonance spectra in the x band (9.8 ghz) was carried out in a brucker esp 3220 spectrometer adopting a modulation frequency of 100 khz whose 109 characterization of vermiculite by xrd and spectroscopic techniques enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:17 p p p composite 133 lpi at 45 degrees width was fixed at 0.4 mt for wide lines and reduced to 0.05 mt for acute bands, at a temperature of 100 k. the nuclear magnetic resonance of 27al was performed in a brucker advance 400 spectrometer; the sample was placed in 2.5mm rotors, while 1200 scannings were accumulated with a time of 100ms of recycle. rotor speed was 5000 hz. xrf, epr and nmr measurements were done additionally in the mineral washed with nitric acid 0.01 m (10 ml g-1 clay) by 1h at 30 °c. 3. results and discussion 3.1 x-ray fluorescence (xrf) and cation exchange capacity (cec) table 1 records the chemical composition of the mineral obtained by x-ray fluorescence. the structural formula of the colombian vermiculite estimated from its chemical composition and following the methodology described in wilson m. (wilson 1995) is: [(si3.04al0.92ti0.04)(al0.11fe 3+ 0.35 fe 2+ 0.07mg2.41mn0.003)o10 (oh)2] ca0.21k0.05na0.10 on the other hand, the cationic exchange carried out with ammonium-vermiculite is 1.10 meq nh g 4 1� . 3.2 x-ray diffraction analysis table 2 shows the comparison between the reflections in xrd for mg-vermiculite (brown 1961), those corresponding to a diffraction pattern calculated using difk software (wiewióra et al. 2003) and the reflections observed in the pattern for natural vermiculite v. this parallel makes it possible to observe an analogue pattern between the clays; thus the clay mineral under study corresponds to vermiculite (wiewióra et al. 2003). likewise, a high layer stacking is highlighted, which is verified by the presence of reflections in series 02l and 11l (wiewióra et al. 2003). reflection 060 located in 0.154 nm indicates that the vermiculite structure is of the trioctahedral type (dixon and weed 1989). in general, vermiculite can be dioctahedral or trioctahedral, but trioctahedral type is common in soils with a similar morphology to that of mica (moore and reynolds 1997), as observed in vermiculite (figure 3). on the other hand, 6th, 8th, and 10th order reflections located at 0.480, 0.360 and 0.288 nm respectively show a growing intensities serie typical for vermiculites (wiewióra et al. 2003). interstratified material is absent; otherwise the sequence should be abnormal (wiewióra et al. 2003). trioctahedral vermiculites usually formed by the weathering of biotite (gordeeva et al. 2002) may be found in the company of this mineral and also talc is possible. in the case of clay v we can suggest that talc is found as a non significant impurity, indicated by its characteristic diffractions in 0.93 nm, 0.48 nm and 0.31 nm (fagel et al. 2001) (figure 1). in parallel, starting from the positions of reflections 001 and 004 it can be estimated that the vermiculite is found in a proportion larger than 90 % (thorez 1976). in order to verify the purity of the mineral, table 3 records the follow up on reflection 00l of raw natural clay powder, when the sample is solvated in the presence of ethylene glycol and, when the latter is calcined at 500 ºc. this table shows that the v material exhibits the sequence typical of vermiculites (thorez 1976; dixon and weed 1989). however different authors (thorez 1976, dixon and weed 1989, malla and douglas 1987) suggest complementary analysis xrd in mg2+ and k+-vermiculite to distinguish between soil vermiculites and oh-interlayer vermiculites. as well they suggest a relation between the d001 signal and the layer charge. after saturation with magnesium followed by glycerol the basal reflection should be located in 1.4 nm compared with 1.8 nm if the clay 2:1 were a smectite (thorez 1976; fagel et al. 2001). in this sense, the basal reflection of v located at 1.4 nm was not modified. the impossibility for expansion in mg-v is then a consequence of the strong retention of magnesium between layers with a high negative 110 a. campos, s. moreno, r. molina table 1. chemical composition of colombian vermiculite. oxide al2o3 sio2 fe2o3 feo mgo mno cao k2o na2o tio2 % 13.06 45.40 6.91 1.24 24.10 0.11 2.88 0.60 0.26 0.73 enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:18 p p p composite 133 lpi at 45 degrees density like vermiculites (slade et al. 1976, tunega et al. 2003) (table 4). malla and douglas (malla and douglas 1987) established that the saturation with this potassium originates interlayer space reduction in the region between 1.12 -1.20 nm if the charge is lower than 0.57 and, between 1.00 and 1.06 nm when it is at least 0.6. in this sense, when clay v was saturated in the presence of k+ the basal spacing of 1.4 nm was reduced to 1.0 nm after dried at room temperature and calcinations of the sample at 500 °c (table 4). the reduction of interlayer space after potassium saturation, qualitatively indicate that the layer charge must be higher than 0.57 (malla and douglas 1987). in addition, the previous characteristic is a tool for the clear distinction between vermiculite and chlorite, due to the fact that the latter keeps the 001 reflection at 1.4 nm when k-chlorite, while the basal spacing of k-vermiculite is reduced to 1.0 nm (dixon and weed 1989). for this reason, the absence of chlorite is verified as a possible impurity in the mineral. 3.3 thermal analysis the thermal-gravimetric (tg) and differential (dta) analysis of vermiculite is presented in the figure 2 where a total weight loss of 21.1 % is distributed in several regions. in the first one, the dehydration of the clay takes place in two stages: i) between 20 and 110 ºc which corresponds to the loss of water physically absorbed on the surface and ii) from 110 ºc to 500 ºc, with a weight loss of 14.3 % attributed to the exit of water molecules, which are in contact with the cations in the interlayer region (pérez et al. 2003). 111 characterization of vermiculite by xrd and spectroscopic techniques table 2. xrd patterns of mg-vermiculites, for simulated pattern and the pattern of colombian vermiculite. hkl v reference (nm) (nm) 1a 2b 002 1.430 1.44 1.43 004 0.719 0.718 006 0.480 0.479 0.477 020 0.468 0.460 0.460 008 0.360 0.360 0.358 115 0.339 0.341 0010 0.288 0.287 0.287 200 0.265 0.265 0.264 132 0.259 0.260 0.258 202 0.254 0.255 0.254 204, 0012 0.240 0.238 0.239 136 0.221 0.227 138, 206 0.206 0.209 208 0.201 20î2, 1310 0.187 2010 0.185 0.184 0016, 20î4 0.179 0.174 2012 0.168 0.167 1314, 2016 1.580 0.158 060 0.154 0.154 0.154 062, 330 0.153 2018, 0020 0.144 0.144 338 0.136 0.136 a mg-vermiculite macroscopic of west chester (brown 1961). b simulated pattern (wieióra et al. 2003). table 3. identification of vermiculite from the basal reflection 001 (nm). clay mineral n eg 500 vermiculite 1.4 1.4 0.96-1.0 smectite 1.2-1.5 1.7 1.0 chlorite 1.4 1.4 1.4 v 1.4 1.4 1.0 natural clay (n), after solvation treatment with ethylene glycol (eg) and the latter heated at 500ºc (500). enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:18 p p p composite 133 lpi at 45 degrees 112 a. campos, s. moreno, r. molina a b 0 .9 3 3 1 .4 3 0 (0 0 2 ) 0 .7 1 9 (0 0 4 ) 0 .4 8 0 (0 0 6 ) 0 .4 6 8 0 .3 6 0 (0 0 8 ) 0 .3 3 9 (1 1 5 ) 0 .3 1 2 0 .2 8 8 (0 0 1 0 ) 12000 10800 9600 8400 7200 6000 4800 3600 2400 1200 o in te n s it y (a rb it ra ry u n it s ) 5 10 15 20 25 30 35 2theta (degrees) 0 .2 4 0 4 0 .2 0 6 (1 3 8 ,2 0 6 ) 1500 1350 1200 1050 900 750 600 450 300 150 o in te n s it y (a rb it ra ry u n it s ) 35 40 45 50 55 60 65 70 0 .1 8 7 (1 3 1 0 .2 0 2 ) 1 0 .1 8 5 (2 0 1 0 ) 1 .7 9 9 (0 0 1 6 .2 0 4 ) 1 1 .6 8 0 4 (2 0 1 2 ) 0 .1 5 8 (1 3 1 4 .2 0 6 ) 1 0 .1 5 4 (0 6 0 , 3 3 2 ) 0 .1 4 4 (2 0 1 8 ) 0 .1 3 6 (3 3 8 ) 2theta (degrees) figure 1. xrd patterns of colombian vermiculite, in the regions of 5-35 º2 (a) and 35-70 2 (b). table 4. comparison of the first order reflection (nm) for high and low charge vermiculites and the colombian mineral (v). vermiculite composition n eg 500 mgn mgeg kn k500 low charge 1.4 1.6 1.0 1.4 1.6 1.1 1.0 high charge 1.4 1.4 1.0 1.4 1.4 1.0 1.0 v 1.4 1.4 1.0 1.4 1.4 1.0 9.3 natural vermiculite after mg2+ saturation (mgn). mgn after solvation with ethylene glycol (mgeg). natural vermiculite after k + saturation (kn). kn heated at 500 ºc (k500). enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:18 p p p composite 133 lpi at 45 degrees the region between 500 and 850 ºc records a loss of 6.8 % in weight with a linear tendency with respect to temperature, which is attributed to dehydroxilation (pérezrodríguez et al. 2004). in fact, the unusual endothermic peak at 800 °c is associated with the thermal stability on vermiculite which is characteristic of this mineral as well this signal could be assigned to the formation of a new enstatite crystalline phase, which has been reported previously (pérez et al. 2003). 3.4 texture and morphology the selected fraction for vermiculite v corresponds to the mineral observed in figure 3, where large vermiculite layer crystals are detected, which display soft surfaces, with small protuberances. polygonal sheets with flaked borders are observed in vermiculite formed by the alteration of parental materials (kishk and barshad 1969). the specific surface area (7.41 m2 g-1) corresponding to external surface area without any microporosity. the low surface area is related to the particle but as well with the surface charge, since micas with similar particle size and charge higher than vermiculites have less surface area (del rey-pérez-caballero and sánchez et al. 2000). the strong hydrogen bonds between the tetrahedral sheet and the water of interlayer cations in vermiculite can keep the interlayer zone obstructed (tunega d. et al. 2003), which could reduce superficial nitrogen adsorption considerably. by other hand, the relation observed between the interchange cations and the penetration in the interlayer space has been indicated by different authors (pérez et al. 2003). in fact, when monovalent cations in the interlayer region are replaced by those of divalent type, a significant reduction in the superficial area for this mineral clay is observed (pérez et al. 2003). 3.5 electronic paramagnetic resonance (epr) and nuclear magnetic resonance of aluminum ( 27 al-nmr) natural vermiculite can be accompanied by iron oxides and hydroxides on its surface (iron that is not substituting aluminum in the octahedral sheet), which will cause an overestimation of iron in vermiculite if it is evaluated by chemical analysis. in this way, epr could be used to determine the presence of structural fe3+ in vermiculite. additionally, to evaluate the structural environment of paramagnetic ions found in the clay. as can be observed in figure 4, the epr of v is dominated by signals associated with fe3+ g = 4.3 and 2.1. in addition, a sextet of lines in g = 2 can be appreciated, which is typical of mn2+ species with distorted octahedral symmetry (tus¡ar et al. 2005). the signal at g = 4.3 is related with fe3+ located at an orthorhombic environment within the clay structure (bensimon et al. 2000). in general, smectites exhibit signals for fe3+ between g = 4.3 and g = 2.0, which can be attributed to the combination of two different octahedral and two tetrahedral environments. their formation depends on the possible organizations of the oh groups in the feo4(oh)2 unit (wilson 1995). mc bride et al. (mc bride et al. 1975) suggest that the difference between the two octahedral sites occupied by 113 characterization of vermiculite by xrd and spectroscopic techniques temperature ºc 30.0 20.0 10.0 0.0 -10.0 -20.0 -30.0 15.0 14.5 14.0 13.5 13.0 12.5 12.0 11.5 0.0 100.0 200.0 300.0 400.0 500.0 600.0 700.0 800.0 900.0 1000.0 w e ig h t ( ) [m g ] h e a tf lo w ( ) [m w ] figure 2. differential and gravimetric thermal curves of colombian vermiculite. figure 3. sem image of vermiculite v. enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:19 p p p composite 133 lpi at 45 degrees fe3+ ions depends on the nature of the adjacent octahedral cation (divalent or trivalent). according to timofeeva et al. (timofeeva et al. 2005), the iron found in the form of aggregates in the surface is related with signals in the region comprised between g = 2.3-2.6. alternatively, different contributions to the wide signal located around g = 2.1 are related with fexoy or feooh species (chung et al. 2004). in the case of the smectite minerals the signal at g = 4.3 is assigned to the presence of isolated fe3+ in tetrahedral or octahedral coordination, which corresponds to the iron located in the interior of the clay sheets (iron substituting aluminum in the octahedral layers). on the other hand, the signal at g = 2.0 is associated to the presence of clusters of iron (iron extra-red species) (carriazo et al. 2005). 114 a. campos, s. moreno, r. molina a b a figure 4. epr spectra of colombian vermiculite at 100 k. (a) natural mineral, (b) natural mineral after mild acid treatment. enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:20 p p p composite 133 lpi at 45 degrees as observed in figure 4a, several signals point to the presence of iron extra-red species on the surface of the clay v. for this reason, the clay was submitted to a soft acid wash to remove impurities from its surface. in figure 4b, the absence of signals between g = 2.3-2.6 indicates that these species had been removed. likewise, signals of mn2+ ions located in the region at g = 2, and g = 1.9 (figure 4a), and the multiple signal recorded is interpreted in terms of the interaction between structural ions mn2+ and fe3+ and other environments (kessissoglou 1999). after acid wash mn2+ signals remain invariable which suggest structural positions for this cation. respect to aluminum, in order to evaluate the structural environment as well as the al(iv)/al(vi) ratio to compare with the ratio estimated in the structural formula, 27al mas-nmr spectra was made. the spectrum is shown in figure 5, in which an intense signal at 60 ppm is observed (with two sidebands at -60 and 190 ppm). this signal is associated with al(iv) substituting the silica in the tetrahedral sheets (klinowski 1999). a second signal corresponding to al(iv) at 0 ppm (klinowski 1999) was observed. the intensity ratio al (iv)/ al (vi) was 10.4. 3.6 correlation of results weaver (weaver 1958) suggested that vermiculite derived from mica weathering has a high laminar charge and its 001 reflection decrease to 1.0 nm after potassium saturation, as happens in clay v (table 4). in contrast, vermiculites generated by the weathering of materials like amphiboles or volcanic material exhibit a limited layer contraction with the same treatment. in fact, the laminar charge is a fundamental element in identification of 2:1 silicates. in fact, the aipea (international association for the study of clays) established that vermiculite is a phylosilicate 2:1 with a charge of 0.6 to 0.9 per unit cell (guggenheim et al. 2006). in this sense, to determine the layer charge of v, it was performed with the structural formula, which was determined through xrf elemental analysis. in it formula can be deduced that the tetrahedral sheet originate an excess negative charge of -0.9 e/ o10(oh)2, which is compensated by ca 2+, na+ and k+ cations. as regards to the octahedral sheet, formed specially by fe3+ and mg2+ it carries a positive charge of 0.35, resulting in a total negative charge of 0.6 e/ o10(oh)2. in order to confirm the approach to the elemental organization and to evaluate the environment of the elements found in the structure, epr and nmr provide very valuable information. thus, by means of epr spectroscopy, iron species on the natural clay surface were detected, which can be easily removed through a soft acid wash (figure 4b). in consequence, the chemical analysis by xrf to evaluate the structural formula was done in the mineral washed with soft acid because unwashed sample could overestimate the iron content. regarding the nmr the al (iv)/al (vi) ratio (10.4) compared with the ratio estimated with the structural formula (8.9) is a satisfactory approach of vermiculite structure take into account that paramagnetic species such as iron can modify the tetrahedral/octahedral ratio (gates et al. 1996). the total negative charge (malla and douglas 1987; guggenheim et al. 2006) observed in the structural formula and the slight difference with the cec as well as the xrd signals and the follow up of the basal reflection 001 after submitting the sample to different treatments (thorez 1976; 115 characterization of vermiculite by xrd and spectroscopic techniques 2060140 al iv * -140-100-60-20100180220 (ppm) al vi * figure 5. 27al mas-nmr spectra of the colombian vermiculite. (*) spinning sidebands. enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:20 p p p composite 133 lpi at 45 degrees wilson 1995; moore and reynolds 1997) indicate that mineral v is trioctahedral vermiculite. the knowledge derived from the present work is very useful in further understanding, controlling and predicting the modifications that can be performed on this attractive natural mineral in order to obtain materials with technological impact such as catalyst and adsorbents (campos et al. 2005; cristiano et al. 2005; campos et al. 2007; campos and gagea et al. 2008; campos and moreno et al. 2008, hernández et al. 2007). 5. conclusion mineral clay from a deposit in the region of santa marta, colombia has been characterized through different techniques. the correlation of results indicates that the mineral corresponds to trioctaheral vermiculite as result of the mica-biotite weathering process. the layer stacking, its high thermal stability, low superficial area and the magnitude of its negative charge are characteristic of vermiculite. the details about the structure of this mineral are highly useful in the latter pillaring process of these mineral to enhance their catalytic applications. acknowledgements this research has been supported by projects code hermes 9887 vri-dib-universidad nacional de colombia. the authors wish to thank professor pierre jacobs, director of the centrum voor oppervlaktechemie en katalyse at katholieke university of leuven for carrying out the nmr and epr analysis. references bensimon y., deroide b., dijoux f., martineau m. (2000). nature and thermal stability of paramagnetic defects in natural clay: a study by electron spin resonance. journal of physics and chemistry of solids 61, 1623-1632. brown b. (1961) the x-ray identification and crystal structures of clay minerals. mineralogical society; london. campos a., moreno s., molina r. (2005). acidez e hidroisomerización de heptano en una vermiculita colombiana modificada con aluminio. revista colombiana de química. 34, 79-92. campos a., gagea b., moreno s., jacobs p., molina r. (2007). hydroisomerization of decane on pt/al,ce-pillared vermiculites. studies in surface science and catalysis. 170, 1405-1409. campos a., gagea b., moreno s., jacobs p., molina r. (2008). decane hydroconversion with al-zr, al-hf, al-ce-pillared vermiculites. applied catalysis a. 345, 112-118. campos a., moreno s., molina r. (2008). relationship between hydrothermal treatment parameters as a strategy to reduce layer charge in vermiculite, and its catalytic behavior. catalysis today. 133-135, 351-356. carriazo j., guélou e., barrault j., tatibouët j., molina r., moreno s., 2005. synthesis of pillared clays containing al, al-fe or al-ce-fe from a bentonite: characterization and catalytic activity. catalysis today 107–108, 126–132. chapman h. (1965). cation exchange capacity. in: black c, evans d, white j, ensminger l, clark e, editors. methods of soil analysis (agronomy 9), american society of agronomy. chung h., sang-won c., yong-sik o., jinho j. (2004). epr characterization of the catalytic activity of clays for pce removal by gamma-radiation induced by acid and thermal treatments. chemosphere 57, 1383-1387. cristiano d., campos a., molina r. (2005). charge reduction in a vermiculite by acid and hydrothermal methods: a comparative study. journal of physical chemistry b 109, 19026-19033. del rey-pérez-caballero f., poncelet g. (2000). microporous 18 å al-pillared vermiculites: preparation and characterization. microporous and mesoporous materials. 37, 313-327. del rey-pérez-caballero f., sánchez m., poncelet g. (2000). hydroisomerization of octane on pt/al-pillared vermiculite, and comparison with zeolites. studies in surface science and catalysis. 130, 2417-2422. dixon j., weed s. (1989). minerals in soil environments. second edition. united states. fagel n., robert c., preda m., thorez j. (2001). smectite composition as a tracer of deep circulation: the case of the northern north atlantic. marine geology. 172, 309-330. gates w., stucki j., kirkpatrick r. (1996). structural properties of reduced upton montmorillonite. physics and chemistry of minerals 2, 535-541. 116 a. campos, s. moreno, r. molina enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:20 p p p composite 133 lpi at 45 degrees gordeeva l., moroz e., rudina n., aristov y. (2002). formation of porous vermiculite structure in the course of swelling. russian journal of applied chemistry. 75, 357-361. guggenheim s., adams j., bain b., bergaya f., brigatti m., drits v., formoso m., galán e., kogure t., stanjek h. (2006). summary of recommendations of nomenclature committees relevant to clay mineralogy: report of the association international pour l’etude des argiles (aipea) nomenclature committee for 2006. clays and clay minerals. 54, 761-772. hernández w., centeno m., odriozola j., moreno s., molina r. (2007). acidity characterization of a titanium and sulfate modified vermiculite. materials reserach bulletin. 43, 1630-1640. kessissoglou d. (1999). homoand mixed-valence epr -active trinuclear manganese complexes. coordination chemistry review. 185, 837-858. kishk f., barshad i. (1969). the morphology of vermiculite clay particles as affected by their genesis. american mineralogist. 54, 849-857. klinowski j. (1999). solid-state nmr studies of molecular sieve catalysts. chemical reviews. 91, 1459-1479. laird d. (1999). layer charge influences on the hydration of expandable 2:1 phyllosilicates. clays and clay minerals. 47, 630-636. malla p., douglas l. (1987). layer charge properties of smectites and vermiculites: tetrahedral vs. octahedral. soil science society of america journal. 51, 1362-1366. macewan, d. and wilson, m. (1980). interlayer and intercalation complexes of clay minerals. pp. 197 248 in: crystal structures of clay minerals and their x-ray identification (g.w. brindley and g. brown, editors). mineralogical society, london. mc bride m., pinnavaia t., mortland m. (1975). exchange ion positions in smectite: effects on electron spin resonance of structural iron. clays and clay minerals. 23,162-164. moreno s., gutiérrez e., alvarez a., papayannakos n., poncelet g. (1997). al-pillared clays: from lab syntheses to pilot scale production characterization and catalytic properties. applied catalysis a 165, 103-114. moore d., reynolds j. (1997) . x-ray diffraction and the identification and analysis of clay minerals. second edition. oxford university press. pérez l., balek v., potayo j., pérez j., subrt j., bountsewa i., beckman i., málek z. (2003). study of natural and ion exchanged vermiculite by emanation thermal analysis, tg, dta and xrd. journal of thermal analysis and calorimetry. 71, 715-726. pérez-rodriguez j., potayo j., jiménez de haro m., pérez-maqueda l., lerf a. (2004). thermal decomposition of nh 4 � -vermiculite from santa olalla (huelva, spain) and its relation to the metal ion distribution in the octahedral sheet physics and chemistry of minerals 31, 415-420. schoonheydt r., pinnavaia t., lagaly g., gangas n. (1999). pillared clays and pillared layered solids. pure applied chemistry. 71: 2367-2371. slade p. (1976). telleria m., radoslovic w. the estructures of ornithine-vermiculite and 6-aminohexanoic acid-vermiculite. clays and clay minerals 24, 134-141. suvorov d., skurikhin v. (2003). vermiculite: a promising material for high-temperature heat insulators. refractories and industrial ceramics. 44, 186-193. stefanis a., tomlinson a. (2006). towards designing pillared clays for catalysis. catalysis today 114, 126-141. thorez j. (1976). practical identification of clay minerals. g. lelotte (ed). dison, belgique. timofeeva m., khankhasaeva s., badmaeva s., chuvilin a., burgina e., ayupov a., panchenko v., kulikova a. (2005). synthesis, characterization and catalytic application for wet oxidation of phenol of iron-containing clays. applied catalysis b. 59: 243-248. tunega d., lischka h. (2003). effect of the si/al ordering on structural parameters ad the energetic stabilization of vermiculitesa theoretical study. physics and chemistry of minerals 30, 517-522. tus¡ar n., zabukovec n.,vlaic g., arc¡on d., daneu n., kauc¡ic¡ v. (2005). local environment of manganese incorporated in mesoporous mcm-41. microporous and mesoporous materials. 82, 129-136. weaver c. (1958). effects of geologic significance of potassium “fixation” by expandable clay minerals derived from muscovite, biotite, chlorite and volcanic material. american mineralogist. 43, 839-861. 117 characterization of vermiculite by xrd and spectroscopic techniques enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:20 p p p composite 133 lpi at 45 degrees wiewióra a., perez-rodriguez l., perez-maqueda l., drapala j. (2003). particle size distribution in sonicated highad low charge vermiculites. applied clay science. 24, 51-58. wilson m. (1969). mineralogy: spectroscopy and chemical determinative methods. champan & hall ed; london. 118 a. campos, s. moreno, r. molina enero 30-geociencias-vol 13-2 2009.prn d:\geociencias 13-2 dic 2009\geociencias-vol 13-2 2009.vp sæbado, 30 de enero de 2010 18:16:21 p p p composite 133 lpi at 45 degrees untitled earth sciences research journal earth sci. res. j. vol. 13, no. 1 (june 2009): 54-63 the maximum effect of deep lakes on temperature profiles – determination of the geothermal gradient v. t. balobaev1, i. m. kutasov2 and l. v. eppelbaum3 1permafrost institute, siberian branch of the russian academy of sciences, yakutsk 677018, russia 2pajarito enterprises, 640 alta vista, suite 124, santa fe, new mexico 87505, usa 3dept. of geophysics and planetary sciences, raymond and beverly sackler faculty of exact sciences, tel aviv university, ramat aviv 69978, tel aviv, israel corresponding author, e-mail: levap@post.tau.ac.il; fax: +972 3 6409282 abstract understanding the climate change processes on the basis of geothermal observations in boreholes is an important and at the same time high-intricate problem. many non-climatic effects could cause changes in ground surface temperatures. in this study we investigate the effects of deep lakes on the borehole temperature profiles observed within or in the vicinity of the lakes. we propose a method based on utilization of laplace equation with nonuniform boundary conditions. the proposed method makes possible to estimate the maximum effect of deep lakes (here the term “deep lake” means that long term mean annual temperature of bottom sediments can be considered as a constant value) on the borehole temperature profiles. this method also allows one to estimate an accuracy of the determination of the geothermal gradient. key words: geothermal gradient, reduced temperature, laplace equation, lake resumen el entendimiento de los procesos de cambio climático basado en las observaciones geotérmicas en pozos es importante y a la vez un problema intrincadamente complejo. muchos efectos no climáticos podrían causar cambios en las temperaturas de la superficie terrestre. en este estudio investigamos el efecto de los lagos profundos sobre los perfiles de temperatura registrados en pozos, al interior o en las inmediaciones de los lagos. proponemos un método basado en el uso de la ecuación de laplace con condiciones de frontera no uniforme. el método hace posible estimar el máximo efecto de los lagos profundos (el término “lago profundo” significa que 54 manuscript received: january 01th, 2009. accepted for publication: february 27th, 2009. la temperatura media anual de los sedimentos del fondo, evaluada sobre un largo período, puede ser considerada como un valor constante) sobre los perfiles de temperatura en los pozos. este método también permite estimar la precisión en la determinación del gradiente geotérmico. palabras clave: gradiente geotérmico, temperatura reducida, ecuación laplace, lago. i. introduction at present many efforts are made to determine the trends in ground surface temperature history (gsth) from geothermal surveys. in this case accurate subsurface temperature measurements are needed to solve this inverse problem � estimation of the unknown time dependent ground surface temperature (gst). the variations of the gst during the long term climate changes resulted in disturbance (anomalies) of the temperature field of geological formations. thus, the gsth data could be evaluated by analyzing the present precise temperature-depth profiles. the effect of surface temperature variations in the past on the temperature field of formations is widely discussed in the literature (e.g., lachenbruch and marshall, 1986; beltrami et al., 1992; shen and beck, 1992; bodri and cermak, 1995; harris and chapman, 1995; huang et al., 1996; guillou-frottier et al., 1998; huang and pollack, 1998; huang et al., 2000; pollack and huang, 2000; majorowicz and safanda, 2005; eppelbaum et al., 2006; hamza et al., 2007; hopcroft et al., 2007; rath and mottaghy, 2007; gonz´alez-rouco et al., 2008; kooi, 2008). ii. previous investigations: some research background earlier the forward calculation approach (fca) was used for the analysis and interpretation of borehole temperatures in terms of the gsth (eppelbaum et al., 2006). three groups based on the geographical proximity were formed. fifteen borehole temperature profiles from europe (5), asia (4) and north america (6) were selected (huang and pollack, 1998; www.geo.lsa.umich. edu/~climate). the objective of this study was the estimation of the warming rates in the 20th century by the fca method and comparing with those obtained by the few parameter estimation (fpe) technique (huang et al., 1996; huang and pollack, 1998). it was reasonable to assume that for close spaced boreholes, the values of the warming rates obtained by the two inversion methods, should vary in narrow limits. the results of inversions (fca) have shown that for boreholes in north america the current warming rates vary in the 0.412.45 k/100a range. the wide range for the warming rate of 0.33-2.48 k/100a was also determined for boreholes in europe. interesting results were obtained for four boreholes in asia (china) (eppelbaum et al., 2006). in this case the warming rate varies in relatively narrow limits (1.16-1.59 k/100a.). the warming rate estimated by the fpe technique (huang and pollack, 1998) varied in wide ranges: 0.38-2.49 k/100a (north america); 0.21-3.75 k/100a (europe), and 0.30-2.53 k/100a (asia). thus, we can conclude that for boreholes in north america and europe both approaches provide practically the same ranges of warming rates. for asian boreholes the fca approach gives a more consistent (narrow) range of warming rates (1.16-1.59 k/100a). the results of temperature inversion by both techniques indicate that probably some of non-climatic effects (vertical and horizontal water flows, steep topography, lakes, vertical variation in heat flow, lateral thermal conductivity contrasts, thermal conductivity anisotropy, deforestation, forest fires, mining, wetland drainage, agricultural development, urbanization, etc.) may have perturbed the borehole temperature profiles. influence of these factors has been studying by many authors (e.g., carslaw and jaeger, 1959; lachenbruch, 1965; kappelmeyer and haenel, 1974; blackwell et al., 1980; majorowicz and skinner, 1997; guillou-frottier et 55 the maximum effect of deep lakes on temperature profiles – determination of the geothermal gradient al., 1998; lewis and wang, 1998; kohl, 1999; safanda, 1999; pollack and huang, 2000; cermak and bodri, 2001; gosselin and mareschal, 2003; gruber et al., 2004; bodri and cermak, 2005; mottaghy et al., 2005; nitoiu and beltrami, 2005; allen et al., 2006; taniguchi, 2006; chouinard and mareschal, 2007; hamza et al., 2007; safanda et al., 2007). at the same time exact calculation of all these factors is a complex physical-mathematical problem, which obviously will be completely solved in a future by the method of successive approximations. the temperature regime of sedimentary formations is influenced by many environmental and geological factors (local relief, sedimentation, erosion, lateral conductivity contrasts, underground water movement), past climate, and by the heat flow from the earth’s interior – terrestrial heat flow. most of temperature surveys are conducted in boreholes. in many cases the drilling sites of boreholes are located within or outside of deep lakes (we employ the term “deep lake” to designate that long term mean annual temperature of bottom sediments could be considered as a constant value). the objective of this study is to evaluate to what extent the proximity of deep lakes can affect the temperature profiles of wellbores. in 1974 balobayev and shastkevich published results of their analytical study which can be used to determine the configuration of the steady temperature field of formations beneath the lakes of an arbitrary contour (balobayev and shastkevich, 1974). taking into account that this publication is not easily accessible to researchers, we present below a brief summation of the results of this study (balobayev and shastkevich, 1974). authors assumed that the lake existed for an infinitely long period of time. in this case the solution of laplace equation with nonuniform boundary conditions can be used to describe the steady temperature field of formations beneath the lakes and estimate the maximum effect (due to assumption that the lake existed for an infinitely long period of time) of lakes on borehole temperature profiles. iii. climate reconstruction methods: some typical disturbances and restrictions we should note that all climate reconstruction methods are based on one-dimensional heat conduction equation. it is assumed that a uniform boundary condition is applied on a plane surface, the formation is a laterally homogeneous medium, and the thermal properties can depend only on a depth. for this reasons any subsurface temperature variations arising from conditions that depart from that theoretical model have the potential to be incorrectly interpreted as a climate change signature (pollack and huang, 2000). to demonstrate the well selection procedures we briefly present two examples. in the study conducted by guillou-frottier et al. (1998), only 10 from 57 temperature profiles were selected for inversion of past ground surface temperatures. as was mentioned by nitoiu and beltrami (2005) from over 10,000 borehole temperature logs worldwide (the international heat flow commission global geothermal data set), only about 10% of these data are currently used for climate studies because a number of known non-climatic energy perturbations are superimposed on the climatic signal. therefore, an extreme caution should be used in selection of temperature-depth profiles for inferring the ground surface temperature histories. the following criteria were considered in rejecting boreholes from the study: steep topography, proximity of lakes, water circulation, instrumental problems, other identifiable terrain effects (such as heat refraction, permafrost effects), and recent changes in surface conditions (clearing of trees). for most of the boreholes that were discarded, the shallowest part of the temperature profile is perturbed. as was mentioned by co-authors these perturbations are often similar to the perturbations due to changes in surface temperature. if the terrain conditions had not been considered, warming would have been inferred for 25 boreholes. ten boreholes show apparent cooling, and only one shows no difference. to screen out borehole temperature data from eastern brazil with 56 v. t. balobaev, i. m. kutasov and l. v. eppelbaum indications of possible perturbations arising from non-climatic effects, the following quality assurance conditions (we basically agree with these criteria) were imposed (hamza et al., 2007): 1. the borehole is sufficiently deep that the lower section of the temperature-depth profile allows a reliable determination of the geothermal gradient, presumably free of the effects of recent climate changes. order of magnitude calculations indicate that surface temperature changes of the last centuries would penetrate to depths of nearly 150 m, 2. the time elapsed between cessation of drilling and the temperature log is at least an order of magnitude large compared to the duration of drilling, 3. the temperature-depth profile is free from the presence of any significant non-linear features in the bottom parts of the borehole, usually indicative of advection heat transfer by fluid movements, either in the surrounding formation or in the borehole itself, 4. the elevation changes at the site and in the vicinity of the borehole are relatively small so that the topographic perturbation of the subsurface temperature field at shallow depths is not significant, and 5. the lithologic sequences encountered in the borehole, have relatively uniform thermal properties, and are of sufficiently large thickness that the gradient changes related to variations in thermal properties do not lead to systematic errors in the procedure employed for extracting the climate related signal. out of a total of 129 temperature logs only 17 were found to satisfy the above set of quality assurance conditions (hamza et al., 2007). corrections can be applied, for example, to correct borehole temperature profiles for the effect of topography (lachenbruch, 1965; blackwell et al., 1980; safanda, 1994, 1999). however, this is rarely done because the amplitude of the climatic signals is often smaller than the uncertainty on these corrections (chouinard and mareschal, 2007). safanda et al. (2007) presented interesting results of repeated temperature logs from czech, slovenian and portuguese borehole climate observatories within a time span of 8-20 years. the repeated logs revealed subsurface warming in all the boreholes amounting to 0.2-0.6 oc below 20 m depth. the warming rate of 0.05 oc/yr. at the czech observatory (located in a park within the campus of the geophysical institute in prague) was estimated. this warming rate is two times more than the simulated value (using the surface air temperature as a forcing function). it was assumed that subsurface temperature at the station is influenced by new structure built within the campus of the geophysical institute within the last 10-20 years and/or by other components of infrastructure built 40-50 years ago. the authors (safanda et al., 2007) conducted a quantitative analysis of these effects by solving numerically the heat conduction equation in a 3d geothermal model of the borehole site. it was found out that the mentioned anthropogenic structures influence the temperature in the borehole quite strongly. nitoiu and beltrami (2005) attempted to correct borehole temperature data for the effects of deforestation. the authors simulated the ground surface temperature changes following deforestation by using a combined power exponential function describing the organic matter decay and recovery of the forest floor after a clear-cut (covington, 1981). the presented examples demonstrate that application of this correction could allow incorporate many borehole data into the borehole climatology database (nitoiu and beltrami, 2005). iv. working equations let’s assume that the well site is located within or outside of a deep lake. as it was mentioned above, we consider the long-term mean annual temperature of bottom sediments as a constant value (for the first time this problem was shortly outlined in balobaev et al. (2008)). we will assume that z = 0 is the vertical 57 the maximum effect of deep lakes on temperature profiles – determination of the geothermal gradient coordinate of the lake’s bottom. the temperature regime of geological formations in this area (within and outside of the lake) is subjected to the thermal influence of the lake. the extent of this influence depends mainly on the lake’s dimensions, on the current depth, the distance from the lake, and on the difference between the long term mean annual temperature of bottom sediments and the long term mean annual temperature of surrounding lake formations (at z = 0). we will assume that the lake existed for an infinitely long period of time. the following designations will be used below: ñ,ö, z are cylindrical coordinates (ñ is the distance from the z axis, ö describes the angle from the positive xz-plane to the point, and z is depth); tis is the long term mean annual temperature of bottom sediments; and tot is the long term mean annual temperature of surrounding lake formations at z = 0. firstly, let consider a lake of an arbitrary contour (figure 1). the laplace equation for the semi-infinite solid area is � � � � � � � � 2 2 2 2 2 2 2 1 1 0 t r t t t z� � � � � � � � (1) the boundary conditions are t (ñ, ö, z = 0) = tis within the lake area. t (ñ, ö, z = 0) = tot outside the lake area t (ñ = �, ö, z) = tot + �z where � is the regional (outside the lake area) geothermal gradient. the solution of laplace equation is possible by division of an arbitrary contour lake into sectors. however, the solution is expressed through a complex poisson integral and fairly elaborate and time-consuming computations are needed (balobayev and shastkevich, 1974). let’s ñmax be the maximum value of the set ñ1, ñ2, …., ñn. by introducing a safety factor (the maximum thermal effect of the lake on temperature profiles) we can assume that the lake has a circular shape with a radius ri = ñmax. now the laplace equation and boundary conditions are � � � � � � 2 2 2 2 1 0 t r t t z� � � � � (2) t (�, z = 0) = tis � � ri t (�, öz = 0) = tot � £ ri t (ñ = �, z) = tot + �z the solution eq. (2) is (balobayev and shastkevich, 1974): t (ñ, z) = tot + �z + m(tis – tot (3) m z a a k a k( , ) [ ( , ) ( , )]p � � � � �1 1 2 1 2 3 2 2 (4) a z z r a z r zi i 1 2 2 2 2 2 2 2 � � � � � � � � ( ) ; � � � � (5) a z r z r r i 3 2 2 2 2 1 2 2 22 2� � � � � � � � � � � � � � � � � ; ; (6) r z k r z r i i 2 2 2 2 2 2 4 � � � � � � � � ; ( ) (7) where � �( , ) ( , ) 1 2 2 2k kand are the complete elliptical integrals of the third order (abramowitz and stegun, 1965). for the center of the island (� = 0) m z z z r ( ) � � � 1 2 1 2 (8) the temperature gradient for the well drilled at the center of the lake (ñ = 0) can be determined from eqs. (3) and (8). 58 v. t. balobaev, i. m. kutasov and l. v. eppelbaum figure 1. division of an arbitrary contour lake into sectors (after balobayev and shastkevich, 1974). � �* ( , ) ( ) ( ) / � � � � � � � � � � � � t z z t t r z r is ot i i 0 2 2 2 3 2 (9) �� � � �� � � � � � � � � � �* ( ) ( ) / is ot i i t r z r 2 2 2 3 2 (10) for distances ñ > 0 from eq. (3) we obtain � � �* ( , ) ( )� � � � dt z dz t dm dz is ot � (11) or �� � �� � � �* ( )t t dm dz is ot (12) introducing the reduced temperatures, tr(�, z) from eq. (3), we can write t z t z t z m t tr ot is ot( , ) ( , ) ( )� �� � � � �� (13) v. example of calculations let’s consider a 30 m deep lake with a radius of ri = 100 m and tis = 10 oc. the regional geothermal gradient is � = 0.0300 oc/m and tot = 20 oc. the drilling site of a 3000m wellbore is located at a distance of 150m from the center of the lake (table 1). 59 the maximum effect of deep lakes on temperature profiles – determination of the geothermal gradient table 1. the functions dm(z,0)/dz and m(z, ñ) z, m -dm/dz, 10-61/m, ñ = 0 the function m(z, ñ) distance from the center of the lake (ñ), m 0 50 100 150 200 300 400 20 50 100 150 200 250 300 400 500 600 700 800 900 1000 1100 1200 1400 1600 1800 2000 2200 2400 2600 2800 3000 9429 7155 3536 1707 894.4 512.3 316.2 142.7 75.43 44.43 28.28 19.08 13.47 9.852 7.421 5.727 3.617 2.427 1.707 1.245 .9362 .7215 .5677 .4547 .3698 .8039 .5528 .2929 .1679 .1056 .0715 .0513 .0299 .0194 .0136 .0100 .0077 .0061 .0050 .0041 .0034 .0025 .0019 .0015 .0012 .0010 .0009 .0007 .0006 .0006 .7612 .4937 .2606 .1538 .0991 .0683 .0496 .0292 .0119 .0135 .0100 .0077 .0061 .0049 .0041 .0034 .0025 .0019 .0015 .0012 .0010 .0009 .0007 .0006 .0006 .3828 .2815 .1787 .1189 .0827 .0598 .0449 .0275 .0184 .0131 .0098 .0076 .0060 .0049 .0041 .0034 .0025 .0019 .0015 .0012 .0010 .0009 .0007 .0006 .0006 .0525 .0950 .0985 .0805 .0628 .0488 .0384 .0249 .0172 .0125 .0094 .0073 .0059 .0048 .0040 .0034 .0025 .0019 .0015 .0012 .0010 .0009 .0007 .0006 .0006 .0168 .0370 .0518 .0512 .0450 .0379 .0315 .0219 .0157 .0117 .0090 .0071 .0057 .0047 .0039 .0033 .0025 .0019 .0015 .0012 .0010 .0009 .0007 .0006 .0006 .0042 .0100 .0174 .0212 .0222 .0215 .0198 .0159 .0125 .0099 .0079 .0064 .0052 .0044 .0037 .0032 .0024 .0018 .0015 .0012 .0010 .0008 .0007 .0006 .0006 .0017 .0091 .0076 .0101 .0116 .0122 .0122 .0111 .0095 .0080 .0066 .0056 .0047 .0040 .0034 .0030 .0023 .0018 .0014 .0012 .0010 .0008 .0007 .0006 .0005 what are the magnitudes of the formation temperature perturbations (expressed through the reduced temperatures) caused by the lake? the results of calculations after eqs. (4) and (8) are presented in table 1 and figure 2. we have to note that bottom of the lake has a coordinate z = 0 and because of this the actual depth is z* = z + 30 m. in our case tis – tot = –10 oc and the lake has a cooling effect on the temperature profiles. the values of tr(ñ, z) are decreasing with depth and practically can be neglected for radial distances of 550-600 m from the center of the lake (figure 2). let’s now assume that we have to determine the maximum values of tr(ñ, z) and ä��ñ, z) for the 300-500 m section of the wellbore. the values of tr(ñ, z) can be estimated directly from eq. (13) and table 1: tr(150,300)= –10 oc·0.0384 = –0.384 oc; tr(150,500)= –10 oc·0.0172 = –0.172 oc to determine the values of �� (ñ, z) for this case we suggest to approximate the function m by a quadratic polynomial (table 2, eq. (14)). m a a z a z* � � �0 1 2 2 (14) dm dz a a z * � �1 22 (15) a a a0 1 3 2 60 1126 0 3341 10 0 2872 10� � � � � �� �. . . then the values of ���ñ, z) can be determined from eqs. (12) and (15): ��(150,300) = 1,6174•10-3 oc/m; ��(150,500) = 0.4684•10-3 oc/m for the 300-500m section of the well the values of tr(ñ, z) and ��(ñ, z) are presented in figure 3. in our example the value of the regional geothermal gradient (�) is 0.03 °c/m. thus, the accuracy of the determined geothermal gradient is somewhere between 60 v. t. balobaev, i. m. kutasov and l. v. eppelbaum figure 3. the reduced temperature and the value of �� versus depth for ñ = 150 m. figure 2. the reduced temperatures versus radial distance for three depths. table 2. the approximation of the function m by a quadratic polynomial (eq. (14)) z, m m m* (mm*)/m ·100, % 300 320 340 360 380 400 420 440 460 480 500 0.03842 0.03505 0.03207 0.02941 0.02705 0.02494 0.02304 0.02135 0.01982 0.01844 0.01720 0.03822 0.03510 0.03221 0.02955 0.02712 0.02492 0.02295 0.02121 0.01970 0.01842 0.01737 0.51 -0.15 -0.45 -0.49 -0.27 0.07 0.38 0.65 0.60 0.12 -0.97 the average squared deviation = 0.50% (0.4684·10-3/0.03)·100% = 1.56% and (1.6174·10-3/0.03) )·100% = 5.39% at the same time for a wellbore located at the center of the lake the values tr and �� are maximal (table 1, eqs. (12) and (13)): tr(0,300) = –10 oc·0.0513 = –0.513 °c; tr(0,500)= –10 oc·0.01942 = –0.194 °c, ��(0,300) = 3.16•10-3 °c/m ��(0,500) = 0.754•10-3 °c/m a commercially available software, maple 7 (waterloo maple, 2001), was utilized to compute the function m(ñ, z). vi. conclusions it is shown that borehole paleoclimate investigations are complicated by many disturbing factors, exact calculation of which is a complex physical-mathematical problem. proposed method allows to estimate the maximum effect of deep lakes on the borehole temperature profiles observed within and outside of the lakes. authors assumed that the lake existed for an infinitely long period of time. in this case the solution of laplace equation with nonuniform boundary conditions can be used to describe the steady temperature field of geological formations beneath the lakes and estimate the maximum effect (due to assumption that the lake existed for an infinitely long period of time) of lakes on the borehole temperature profiles. a numerical example to estimate the effect was explained in detail. presented example of calculations testifies to what extent the proximity of a deep lake affects the borehole temperature profiles. an accuracy of the determination of the geothermal gradient is also estimated. acknowledgement the authors thank two anonymous reviewers and prof. luis a. montes v, editor-in-chief of the earth sciences research journal, for their useful comments and suggestions. references abramowitz, m., and i. stegun (1965). handbook of mathematical functions, dover publications inc., n. y. allen, m. r., n. p. gillett, j. a. kettleborough, g. hegerl, r. schnur, p.a. stott, g. boer, c. covey, t. l. delworth, g. s. jones, j. f. b. mitchell, and t. p. barnett (2006). quantifying anthropogenic influence on recent near-surface temperature change, survey of geophysics, 27, 491-544. balobaev, v. t., i. m. kutasov, and l. v. eppelbaum (2008). borehole paleoclimatology – the effect of deep lakes and “heat islands” on temperature profiles, climate of the past, no.4, 1-18. balobayev, v. t., and yu. g. shastkevich (1974). the estimation of the talik zones configuration and the steady temperature field of rocks beneath the lakes of arbitrary contour, in: lakes of the siberia cryolithozone (in russian), nauka publ., novosibirsk, 116-127. beltrami, h., a. m. jessop, and j.c. mareschal (1992). ground temperature histories in eastern and central canada from geothermal measurements: evidence of climate change, palaeogeogr. palaeoclimatol. palaeoecol. (global planet. change sect.), 98, 167-183. blackwell, d. d., j. l. steele, and c.a. brott (1980). the terrain effect on terrestrial heat flow, journal of geophysical research, 85(b9), 4757-4772. bodri, l., and v. cermak (1995). climate change of the last millennium from borehole temperatures: results from the czech republic – part i, global planetary change, 11, 111-125. bodri, l., and v. cermak, (2005). borehole temperatures, climate change and the pre-observational surface air temperature mean: allowance for hydraulic conditions, global planetary change, 45, 265-276. carslaw, h. s., and j. c. jaeger (1959). conduction of heat in solids, oxford university press, n. y. 61 the maximum effect of deep lakes on temperature profiles – determination of the geothermal gradient cermak, v., and l. bodri (2001). climate reconstruction from subsurface temperatures demonstrated on example of cuba, physics of the earth and planetary interiors, 126, 295-310. chouinard, c., and j.c. mareschal (2007). selection of borehole temperature depth profiles for regional climate reconstructions, climate of the past, 3, 297-313. covington, w. w. (1981). changes in the forest floor organic matter and nutrient content following clear cutting in northern hardwoods, ecology, 62, 41-48. eppelbaum, l. v., i. m. kutasov, and g. barak (2006). ground surface temperatures histories inferred from 15 boreholes temperature profiles: comparison of two approaches, earth sciences research journal, 10, no. 1, 25-34. gonz´alez-rouco, j. f., h. beltrami, e. zorita, and m. b. stevens (2008). borehole climatology: a discussion based on contributions from climate modeling, climate of the past discussions, 4, 1-80. gruber, s., l. king, t. kohl, t. herz, w. haeberli, and m. hoelzle (2004). interpretation of geothermal profiles perturbed by topography: the alpine permafrost boreholes at stockohorn plateau, switzerland, permafrost and periglacial processes, 15, 349-357. gosselin, c., and j.c. mareschal, (2003). variations in ground surface temperature histories in the thompson belt, manitoba, canada: environment and climate changes, global and planetary change, 39, 271-284. guillou-frottier, l., j.c. mareschal, and j. musset (1998). ground surface temperature history in central canada inferred from 10 selected borehole temperature profiles, journal of geophysical research, 103, no. b4, 7385-7397. hamza, v.m., a. s. b. cavalcanti, and l. c. c. benyosef (2007). surface thermal perturbations of the recent past at low latitudes – inferences based on borehole temperature data from eastern brazil, climate of the past, 3, 513-526. harris, r. n., and s. d. chapman (1995). climate change on the colorado plateau of eastern utah inferred from borehole temperatures, journal of geophysical research, 100, 6367-6381. hopcroft, p.o., k. gallagher, and c.c. pain (2007). inference of past climate from borehole temperature data using bayesian reversible jump markov chain monte carlo, geophysical journal intern., 171, 1430-1439. huang, s., p. y. shen, and h. n. pollack (1996). deriving century – long trends of surface temperature change from borehole temperatures, geophysical research letters, 23, 257-260. huang, s., h. n. pollack, and p. y. shen (2000). temperature trends over past five centuries reconstructed from borehole temperatures, nature, 403, 756-758. huang, s., and h. n. pollack (1998). global borehole temperature database for climate reconstruction, igbp pages/world data center-a for paleoclimatology data contribution series #1998-044. noaa/ngdc paleoclimatology program, boulder co, usa (see also http:// www.geo.lsa.umich.edu/~climate). kappelmeyer, o., and r. haenel (1974). geothermic with special reference to application, gebruder borntrager, berlin. kohl, t. (1999). transient thermal effects below complex topographies, tectonophysics, 306(3-4), 311-324. kooi, h. (2008). spatial variability in subsurface warming over the last three decades; insight from repeated borehole temperature measurements in the netherlands, earth and planetary science letters, 270, 86-94. lachenbruch, a.h. (1965). rapid estimation of the topographic disturbance to superficial thermal gradients, review of geophysics, 6, 365-400. 62 v. t. balobaev, i. m. kutasov and l. v. eppelbaum lachenbruch, a. h., and b. v. marshall (1986). changing climate: geothermal evidence from permafrost in the alaskan arctic, science, 234, 689-696. lewis, t. j., and k. wang (1998). geothermal evidence for deforestation induced warming: implications for the climatic impact of land development, geophysical research letters, 25, 535-538. majorowicz, j. a., and w. p. skinner (1997). potential causes of differences between ground and surface air temperature warming across different ecozones in alberta, canada, global planetary change, 15, 79-91. majorowicz, j., and j. safanda (2005) measured versus simulated transients of temperature logs – a test of borehole climatology, journal of geophysics and engineering, 2, 291-298. mottaghy, d., r. schellschmidt, y.a. popov, c. clauser, i. t. kukkonen, g. nover, s. milanovsky, and r. a. romushkevich (2005). new heat flow data from immediate vicinity of the kola super-deep borehole: vertical variation in heat flow confirmed and attributed to advection, tectonophysics, 401, 119-142. nitoiu, d., and h. beltrami (2005). subsurface thermal effects of land use changes, journal of geophysical research, 110, f01005. pollack, h. n., and s. huang (2000). climate reconstruction from subsurface temperatures, annual review of earth planetary sciences, 28, 339-365. rath, v., and d. mottaghy (2007). smooth inversion for ground surface temperature histories: estimating the optimum regularization parameter by generalized cross-validation, geophysical journal intern., 171, 1440–1448. safanda, j. (1994). effects of topography and climatic changes on the temperature in borehole gfu-1, prague, tectonophysics, 239, 187-197. safanda, j. (1999). ground surface temperature as a function of slope angle and slope orientation and its effect on the subsurface temperature field, tectonophysics, 306(3-4), 367-376. safanda, j., d. rajver, a. correia, and p. dedecek (2007). repeated temperature logs from czech, slovenian and portuguese borehole climate observatories, climate of the past, 3, 453-462. shen, p. y., k. wang, h. beltrami, and j.c. mareschal (1992). a comparative study of inverse methods for estimating climatic history from borehole temperature data, global planetary change, 98, 113-127. taniguchi, m. (2006). anthropogenic effects on subsurface temperature in bangkok, climate of the past discussions, 2, 832-846. waterloo maple (2001). maple 7 learning guide, waterloo maple inc., waterloo, canada. 63 the maximum effect of deep lakes on temperature profiles – determination of the geothermal gradient geociencias-vol 12-1 sep.vp earth sciences research journal earth sci. res. j. vol. 12, no. 1 (june 2008): 7-30 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia carlos alberto ríos reyes1, oscar mauricio castellanos alarcón2 and akira takasu3 1 escuela de geología, universidad industrial de santander, a.a 678, bucaramanga, colombia. 2 programa de geología, universidad de pamplona, colombia. 3 geosciences department, shimane university, japan. corresponding author: carlos alberto ríos reyes, school of applied sciences, university of wolverhampton, wulfruma street, wolverhampton wv1 1sb, uk tel. +44 (0)1902 322679; fax. +44 (0)1902 322714; e-mail: c.a.riosreyes@wlv.ac.uk abstract a barrovian sequence of the silgará formation at the southwestern santander massif, colombian andes, contains zoned garnets in which major and trace element zoning correlates with distribution of mineral inclusions, which may indicate that garnet growth rate varied through time and affected both composition and texture of garnets, although different garnet producing reactions have also played an important role in the chemical zoning of garent. however, a local metasomatism process associated to the action of late magmatic fluids associated to the emplacement of the pescadero pluton (external forcing mechanism) would be also considered. in particular, ca, mn and y zoning patterns in some garnets correspond with inclusion-rich vs. inclusion-free zones, although the distribution of inclusions does not correlate with chemical zoning (i.e., the same inclusions are found in ca-rich and ca-poor zones of the garnet). there is a similar lack of correlation with accessory phases (apatite, monazite, xenotime, ilmenite or rutile). in a garnet from the garnet-staurolite zone, a high mn core contains abundant and randomly oriented apatite, monazite and ilmenite inclusions, while a euhedral low ca mantle zone is inclusion-free and the high ca / low mn rim zone contains apatite, monazite and ilmenite aligned parallel to the margins of the garnet. inclusions in garnet can also represent mineral phases were not completely consumed during garnet growth. association of garnet zoning trends and patterns with inclusion distribution may help differentiate between processes that identically affect major-element zoning but that produced variable textures in the garnet. 7 manuscript received september 7, 2007. accepted for publication january 30, 2008. key words: colombia; garnet; zoning; santander massif; silgará formation. resumen una secuencia barroviana de la formación de silgará en la región suroccidental del macizo de santander, andes colombianos, contiene granates zonados en los cuales la zonación de elementos mayores y trazas se correlaciona con la distribución de las inclusiones minerales, lo cual puede indicar que la tasa de crecimiento del granate varió con el tiempo y afectó la composición y la textura de los granates, aunque diferentes reacciones que producen granate han jugado también un papel importante en la zonación química del granate. sin embargo, un proceso local de metasomatismo generado por la acción tardia de fluidos magmáticos asociados al emplazamiento del plutón de pescadero (mecanismo de fuerza externa) es también aquí considerado. en particular, los patrones de zonación de ca, mn e y en algunos granates corresponden con zonas ricas en inclusiones vs. zonas sin inclusiones, aunque la distribución de inclusiones no se correlaciona con la zonación química (es decir, las mismas inclusiones se encuentran en las zonas ricas y pobres en ca del granate). hay una carencia similar de correlación con las fases accesorias (apatito, monacita, xenotima, ilmenita o rutilo). en un granate de la zona del granate-estaurolita, un núcleo alto en mn contiene abundantes inclusiones aleatoriamente orientadas de apatito, monacita e ilmenita, mientras que una zona euhedral baja en ca carece de inclusiones y la zona de borde rica de alto ca / bajo mn contiene inclusiones de apatito, monacita, e ilmenita orientadas paralelo a los bordes del granate. inclusiones en granate pueden también representar fases minerales que no fueron completamente consumidas durante el crecimiento del granate. la asociación de los patrones de zonación del granate con la distribución de las inclusiones puede ayudar a distinguir entre los procesos que afectan idénticamente la zonación de elementos mayores pero produjo texturas variables en el granate. palabras claves: colombia; granate; zonación; macizo de santander; formación silgará. introduction garnet is one of the most studied minerals in relation with chemical variations in metamorphic rocks since it preserves a record of its growth history expressed by the chemical zoning and inclusions of other minerals. the broad range in mineral composition promotes the growth of garnet in rocks of very different chemical compositions and over a wide spectrum of metamorphic conditions. furthermore, its refractory character allows the preservation of chemical and textural zoning that is important when making interpretations about the metamorphic history of the host rock. of particular interest in recent years have been the trace element zoning in garnet as an important source of information for petrologic processes (hickmott & shimizu, 1990; schwandt et al., 1996; chernoff & carlson, 1999; yang & rivers, 2001; skelton et al., 2002) and for relating the growth of geochronologically important accessory phases (e.g., monazite) to metamorphic reaction histories (pyle & spear, 1999; gibson et al., 2004). major and trace element zoning in garnet have been typically explained by (1) elemental fractionation during mineral growth (e.g., hollister 1966; cygan and lasaga 1982), (2) by intracrystalline diffusion (e.g., anderson and buckley 1973), (3) limitations at the mineral-matrix interface (e.g., carlson 1989), (4) interaction with a metasomatic fluid (e.g., hickmott et al. 1987; young and rumble 1993; chamberlain and conrad 1993; erambert and austrheim 1993; jamtveit et al. 1993; jamtveit and hervig 1994), (5) the breakdown or growth of trace element-rich minerals (hickmott & shimizu, 1990; hickmott and spear 1992) or (6) changes in the garnet-mineral matrix partition coefficients because of changes in pressure and temperature conditions, garnet composition or the number of mineral phases in the assemblage or their proportions 8 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu (yang and rivers, 2002). the majority of these processes occur during crystal growth. diffusional reequilibration is the only primary postgrowth process that leads to zoning in minerals. in an extreme case, compositional zoning can arise in garnet after its crystallization with a homogeneous composition that is in desequilibrium with the matrix. because a chemical gradient exists between the garnet and the surrounding matrix, often biotite-rich, volume diffusion acts to reequilibrate the garnet composition with the matrix (e.g., tracy et al., 1976). diffusion occurs as long as the temperature remains sufficiently high (e.g., lasaga 1983). trace elements are extremely sensitive to changes in accessory mineral assemblage and/or fluid composition and many trace-element diffusivities in garnet must be very much slower than diffusivities for major elements (mg, mn, fe) in garnet, but probably on the same order as the diffusivity of ca in garnet (pyle & spear, 1999). trace element distributions in garnet must, however, be interpreted with caution. previous studies have documented trends in ca, mn and y zoning in garnet (e.g., chernoff & carlson, 1997; pyle & spear, 1999; yang & rivers, 2002) and discussed the possibility that lowor highannuli within garnet are related to local disequilibrium in some elements, but not all. for example, patterns that deviate from normal growth zoning in garnet (e.g., euhedral bands concentric about the garnet core, patches or spiral to curving patterns) might form by garnet overgrowth of phases enriched or depleted in particular major or trace elements (e.g., mn; yang & rivers, 2001; hirsch et al., 2003), by resorption of garnet during garnet-consuming reactions that produce staurolite (menard & spear, 1993) or by growth of garnet after a matrix phase has been depleted (e.g., epidote, chloritoid; whitney & ghent, 1993). the resulting garnet zoning patterns will be a function of the local chemical heterogeneities, the temperature of metamorphism during and after garnet growth and the growth rate of garnet, factors that will affect both the zoning and the distribution of mineral inclusions in garnet. chernoff & carlson (1997, 1999) considered the possibility of independent, local scales of equilibrium for different elements. they demonstrated that many trace elements, as well as ca, reflect disequilibrium at thin-section scale and therefore thermobarometric estimates that involve grossular contents may be in error. garnet zoning may also be affected by fluid flow (e.g., stowell et al. 1996; skelton et al., 2002) and deformation and these processes may be coupled, as deformation will change grain size and adjust grain boundaries, affecting rates and pathways for diffusion. garnets are frequently zoned in the major elements fe, mg, mn and ca. at high temperature, major element growth zoning may be significantly modified by intracrystalline diffusion while trace element zoning may be less susceptible to diffusion (e.g., hickmott & spear, 1992; lanzirotti, 1995; chernoff & carlson, 1999). in this paper, we report zoning patterns for garnets of the metapelitic silgará formation of the southwestern santander massif, colombian andes. we document major and trace element zoning in garnets and evaluate the processes that control and influence zoning during prograde metamorphism. geological setting the santander massif lies within the eastern cordillera of the colombian andes, where it divides into the northeast-trending perijá in colombia and the east-northeast-trending mérida andes in venezuela (figure 1). the metamorphic history of this massif is important for interpretation of the geologic and tectonic evolution of the northwestern continental margin of south america. the santander massif is underlain by deformed and metamorphosed rocks that have been tectonostratigraphically divided into three metamorphic units (bucaramanga gneiss complex, silgará formation and orthogneiss), which are intruded by several igneous bodies, most of them of triassic-jurassic age and some of paleozoic age (e.g., goldsmith et al., 1971; boinet et al., 1985; dörr et al., 1995; ordoñez, 2003). the intrusives form part of the santander plutonic complex and are interpreted as calc-alkaline crustal bodies emplaced after peak-metamorphism. sedimentary rocks ranging in age from devonian to tertiary flank the core rocks. the garnet-bearing samples investigated in the 9 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia present study belong to the metapelitic sequence of the caledonian silgará formation that represents what was originally a very thick pile of volcano-sedimentary rocks. well-exposed sections of this metamorphic unit crop out at the southwestern region of the santander massif (figure 1), which is long established as classic area for the study of rock metamorphism and deformation caused by continental collision during the caledonian orogeny. according to ríos et al. (2003), this region experienced crustal thickenning during the initial stages of collision, followed by a slow uplift and erosion period, resulting in a decompression accompanied by heating. the silgará formation structures and metamorphic features here are complex and early workers had great difficulty in making sense of the geology of the area. the metamorphic rocks of the silgará formation generally strike nw-se and dip to the southwest. the nw-trending bucaramanga fault represents a very major break in the crust in the northeastern part of the 10 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu figure 1. lower left side, map of the colombian andes, showing the location of the santander massif in the eastern cordillera. upper left side, simplified geological map of the santander massif, showing the distribution of the silgará formation metamorphic rocks in grey (modified after goldsmith et al., 1971) and the location of the study area in black. right side, generalized geological map of the study area (modified after ward et al., 1973), showing locations of garnet samples investigated (black stars). metamorphic zones modified after ríos et al. (2003). bf: bucaramanga fault; lsaf. los santos aratoca fault. study area and separates the precambrian bucaramanga gneiss complex from the triassic-jurassic pescadero and mogotes batholiths. this fault is probably strike slip with a protracted history of displacement (goldsmith et al., 1971). in the southwest, near aratoca, the silgará formation is cut by the nw-se-trending los santos aratoca fault and is unconformably overlain by mesozoic sedimentary rocks (figure 1). it is also unconformably overlain by a mesozoic sequence on the northwest and the southwest. the triassic-jurassic pescadero and mogotes batholiths intruded the silgará formation in the northeastern and eastern part of the area after peak-metamorphism (goldsmith et al., 1971). the silgará formation is dominated by metapelitic rocks with minor intercalations of metabasites. the lithology of the metapelitic sequence changes in composition northeastward, from quartz-rich pelitic schists in the southwest to feldspar-rich semipelitic schists in the northeast. a millimeter-scale compositional banding consists of alternating granoblastic quartzplagioclase-rich domains with minor k-feldspar and muscovite-biotite-rich domains. subordinated rock types of the silgará formation are well-foliated amphibole-bearing schists, orthoamphibolites and calc-silicate rocks. microstructural evidence shows that the silgará formation underwent prograde metamorphism during at least three deformation phases, including extensive retrograde metamorphism during the last stage. the dominant schistosity or slaty cleavage in the metapelitic rocks may have developed at the same time as early folds and thrusts. the metapelites display a well-developed schistosity, although well-preserved sedimentary features are locally preserved in low-grade rocks. in general, later structures such as flat-lying crenulations, small chevron folds and kink bands overprint the main foliation. retrograde textures include partial replacement of garnet by chlorite and/or muscovite or chlorite and biotite along cracks; heavily corroded staurolite crystals surrounded by muscovite; biotite and calcic-amphibole replaced by chlorite (although in some cases amphibole is replaced by biotite); and feldspar partly replaced by sericite. a progressive sequence of metamorphic zones has developed in pelitic rocks of the silgará formation. the regional metamorphic grade increases with structural depth from the biotite zone to the staurolite-kyanite zone. garnet-bearing pelites are widespread throughout the region. barrovian regional metamorphism occurred under lowto high-temperature and medium-pressure conditions. the distinction between the staurolite-kyanite and lower sillimanite (fibrolite) zones is not well defined because fibrolitic sillimanite is present in trace amounts in staurolite-kyanite bearing samples. the occurrence of fibrolite in sample pcm-473 lacking other al2sio5 phases near the contact with the pescadero pluton is interpreted as formed in the waning stages of a thermal event. therefore, the fibrolitization process should not be always considered as a polymorphic reaction. ríos et al. (2003) estimated the temperatures and pressures of equilibration of mineral assemblages from metamorphic rocks of the southwestern santander massif. p-t conditions were 500-520°c and 4.4-5.5 kbar in the garnet zone and 590-700°c and 5.5-7.5 kbar in the garnet-staurolite zone. analytical methods x-ray maps were collected and analyses were carried out using the jeol 8800 electron probe microanalyzer at the research center for coastal lagoon environments at the shimane university (japan) and the jeol 8900 electron probe microanalyzer at the department of geology and geophysics at the university of minnesota (usa). accessory mineral phases were identified using back-scattered electron imaging (bei) and energy-dispersive spectroscopy (eds). the analytical conditions were as follows: accelerating voltage of 15 kv and beam current of 20-25 na (for quantitative analyses) and 25-75 na (with dwell times of 40-55 msec/pixel, for major element maps) or higher beam current (100 na) and longer dwell times (up to 90 msec/pixel) for trace element maps. qualitative x-ray maps were obtained first to delineate geochemically significant zones and to guide quantitative spot analysis (in radial traverses and spacing decreasing near the rim). data adquisition and reduction were carried out us11 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia ing the zaf correction procedures. a combination of natural and synthetic minerals were used as standards. chemical zoning and textural relations in garnet a detailed petrographic study of six thin sections was carried out on selected specimens which are the same as those studied by ríos (1999) and corresponding to a series of pelitic rocks of the silgará formation composed of quartz + plagioclase + muscovite + biotite + garnet ± staurolite and accessory fibrolitic sillimanite and rare crystals of andalusite and kyanite (one sample contained the three polymorphs). in the garnet zone, garnets are commonly idioblastic, whereas in the garnet-staurolite and staurolite-kyanite zones they tend to be rounded and embayed, displaying reaction rims. as follows we describe petrographic features of the major and accessory minerals in the analyzed samples as illustrated in figure 2. the mineral assemblages of the analyzed samples are given in table 1. mineral abbreviations are after kretz (1983). all the examined samples contain garnet porphyroblasts that preserve growth zoning to varying degrees. garnet is almandine-rich and exhibits zoning in which mn decreases from core to rim as mg and fe increase. the zoning of fe, mg and mn zoning is strongly correlated with respect to the degree of local equilibrium in contrast with ca zoning that does not mimic fe, mg and mn zoning exactly (e.g., chernoff & carlson, 1997; spear & daniel, 1998). according to hirsch et al. (2003), it reflects a combination of changing availability of these elements to the surface of the growing garnet and changing partitioning of elements among the mineral phases in the rock. garnet shows abrupt variations in ca distribution from core to rim, developing in some cases a complex cyclic zoning with euhedral lowor high-ca annuli. trace element zoning in garnet will depend on the presence or absence of saturing phases, such as xenotime for y, xenotime, apatite or monazite for p and ilmenite, titatine and rutile for ti. analyzed garnets do not show zoning in y, sc, p, cr or ti, except garnet zone sample pcm-441, which exhibits y zoning. yttrium is potentially useful for monitoring reactions involving ca-rich minerals such as epidote and plagioclase because this element substitutes for ca in mineral structures and y discontinuities in garnet zoning may correlate with inflections in ca zoning (hickmott & spear, 1992). in the following sections, we describe the main features of chemical zoning in garnet from the silgará formation pelites. table 2 shows representative chemical compositions of analyzed garnets. analytical points are indicated by white dots and numbers in the x-ray maps, with numbers keyed to the table. pcm-441 garnet. pelitic schist from the garnet-zone, which typically contain a mineral assemblage of muscovite + quartz + plagioclase + garnet ± biotite, with minor k-feldspar, tourmaline, apatite, zircon, epidote, calcite and fe-ti oxides. numerous, very fine-grained (0.03-0.42 mm in diameter), euhedral grains of garnet, with a hexagonal or pentagonal outline, occur in this sample (figure 2a). garnet grew after an early foliation and contain inclusion trails of ilmenite and graphite concordant to the main foliation (crenulation cleavage), although it may have inclusion-free rims. it is replaced along their margins and fractures by chlorite, biotite and fe-oxides. the most striking characteristic of this garnet is the high mn concentration, which from core to rim varies from 54 to 25 mol% spessartine (figure 3a). fe and mg increase from core to rim with a small decrease in fe/fe+mg. garnet has a low-ca core (7 mol%) with an inflection midway (14 mol%) between core and rim, decreasing towards the rim (8 mol%), as shown in figure 4a. trace element maps for garnet in sample pcm-441 show that yttrium is the only one that displays zoning. with the possible exception of yb, which shows a slight enrichment in the garnet core, the other trace elements (sc, p, cr, ti and yb) are in low abundance and homogeneous across the garnet. garnet is characterized by a euhedral outline, high-y core (figure 4b), which is larger than the faint yb-rich region. the y-rich core region corresponds exactly to the low ca core region. relatively y-enriched regions of the garnet also occur discontinuously along the rim and in the ca-poor core of 12 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu 13 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia figure 2. plane-polarized light photomicrographs of garnet-bearing metamorphic rocks of the silgará formation. (a) numerous very fine-grained garnet crystals; sample pcm-441 (garnet zone). (b) garnet with inclusion-rich (quartz and ilmenite) core and inclusion-free rim regions; sample pcm-420 (garnet-staurolite zone). (c) garnet porphyroblast displaying two different inclusion patterns involving ilmenite; sample pcm-361 (garnet-staurolite zone). (d) inclusion trails of quartz and ilmenite preserved in garnet porphyroblasts; sample pcm-516 (garnet-staurolite zone). (e) fibrolitic sillimanite occurring at core of muscovite adjacent to garnet; sample pcm-473 (staurolite-kyanite zone). (f) garnet-bearing calc-silicate rock occurring as numerous fineto medium-grained crystals associated with calcic amphibole; sample pcm-514 (garnet-staurolite zone). 14 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu t a b le 1 . m in er al as se m b la g es o f th e an al y ze d sa m p le s o f th e s il g ar á f o rm at io n p el it ic ro ck s fr o m th e so u th w es te rn s an ta n d er m as si f m e t. z o n e s a m p le c h l b t m s q z p l k fs g rt s t a n d k y s il c a -a z r a p m g n il m r u t e p m n z x e n t u r c a l g a rn e t p c m -4 4 1 r x x x x m x m m m m m r g a rn e ts t a u ro li te p c m -3 6 1 r x x ,r x ,i x x x i i i i i i g a rn e ts t a u ro li te p c m -5 1 6 r x x x ,i x x x i i g a rn e ts t a u ro li te p c m -5 1 4 r x ,i x ,i x ,i x ,i x x ,i i i i i r g a rn e ts t a u ro li te p c m -4 2 0 r x x ,i x ,i x x i i s ta u ro li te k y a n it e p c m -4 7 3 r x ,i x ,r ,i x ,i x ,i x x m ,i i m m q tz , b t, p l, an d re tr o g ra d e c h l ar e co m m o n in al l sa m p le s. x -p re se n t, b la n k -a b se n t, m -m in o r p h as e, iin cl u si o n s in g ar n et , rre tr o g ra d e m in er al s. m in er al ab b re v ia ti o n s af te r k re tz (1 9 8 3 ) 15 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia figure 3. zoning profiles rim-core-rim of almandine, pyrope, spessartine and grossular components and fe/(fe+mg) ratio in garnet from samples (a) pcm-441 (garnet zone), (b) pcm-514 (garnet-staurolite zone), (c) pcm-361 (garnet-staurolite zone), (d) pcm-516 (garnet-staurolite zone), (e) pcm-420 (garnet-staurolite zone) and (b) pcm-473 (staurolite-kyanite zone). note the different scales for xsps, xprp and xgrs on the left and for xalm and fe/(fe+mg) on the right. an adjacent small garnet. x-ray maps for ca distribution in garnet and its vicinity indicate that there are no ca-rich accessory phases included in garnet and only one small apatite crystal in the matrix just beyond the field of view in figure 4. pcm-361 garnet. pelitic schist from the garnet-staurolite-zone characterized by the peak metamorphic assemblage of quartz + plagioclase + muscovite + biotite + garnet + staurolite. garnet porphyroblasts (1.5-2.6 mm in diameter) are anhedral to subhedral with an elliptical outline and contain numerous inclusions of ilmenite that define different pattern of distribution (figure 2c). at the core region, the inclusion patterns can be concordant or discordant to the main foliation of the rock or ramdonly oriented, whereas at the rim region, they follow the rim of the garnet, defining an approximately circular pattern. garnet porphyroblasts area wrapped by a penetrative foliation (crenulation cleavege). garnet is strongly zoned in mn (figures 3c, 5b), which decreases from 21 mol% in the core to 1-2 mol% in the rim. fe and mg increase from core to rim, with a small decrease in fe/fe+mg. ca decreases from core (5 mol%) towards the rim, with a slight discontinuity in the zoning midway between core and rim, reaching a minimum (1-2 mol%) represented by an euhedral low-ca annulus (figures 5a, 6a); then increases towards the rim (6 mol%). the low-ca annulus coincides with the euhedral zone of high-mn (6-9 mol%) and is truncated in the bottom part due to partial resorption of garnet. analyzed garnets all exhibit complex ca zoning, but there is no apparent zoning in analyzed trace elements (y, yb, p, ti, cr). garnet is texturally zoned with respect to the distribution and shape orientation of ilmenite inclusions and textural zoning can be related to chemical zoning (figures 2c, 5, 6a). in the mn-rich garnet core region, ilmenite inclusions are randomly oriented. inclusions are absent in the low-ca annulus. in the high-ca / low-mn near-rim zone, ilmenite inclusions are abundant and are aligned parallel to the garnet rim, creating a circular pattern, which is also apparent in matrix ilmenite adjacent to garnet (figure 2c). ilmenite occurs as inclusions in garnet that have been partially pseudomorphed by rutile (figure 7) or as a matrix accessory mineral. other inclusions in garnet are quartz, apatite, monazite, zircon and rutile, which 16 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu figure 4. x-ray compositional maps in garnet from sample pcm-441 (garnet zone). (a) ca map, showing a high-ca band at mantle region and the relationship between the low-ca at core and the high-y annulus. numbered spots are keyed to analyses in table 2. (b) y map, displaying a high-y annulus at core characterized by sharp euhedral and straight outlines. in the x-ray maps, the elemental distributions are indicated by white color for the highest and black color for the lowest concentration. a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia t a b le 2 . r ep re se n ta ti v e co m p o si ti o n s o f m et am o rp h ic g ar n et fr o m th e s il g ar á f o rm at io n , so u th w es te rn s an ta n d er m as si f. m a jo r e le m e n ts p c m 4 4 1 p c m -3 1 p c m 5 1 6 p c m 5 1 4 p c m 4 2 0 p c m -4 7 3 ri m c o re ri m ri m c o re ri m ri m ri m c o re ri m c o re ri m ri m c o re ri m w t% 1 2 3 4 5 1 2 3 4 5 6 7 1 1 2 3 4 5 6 7 8 1 2 3 4 1 2 3 4 s io 2 3 6 ,3 9 3 6 ,7 6 3 6 ,1 7 3 6 ,9 6 3 6 ,8 2 3 7 ,7 3 3 7 ,4 5 3 6 ,7 6 3 6 ,7 9 3 6 ,6 7 3 6 ,6 9 3 7 ,1 0 3 7 ,6 1 3 6 ,8 8 3 7 ,3 1 3 6 ,6 2 3 7 ,4 4 3 6 ,7 6 3 7 ,3 3 3 6 ,8 5 3 7 ,6 6 3 6 ,6 3 3 6 ,5 3 7 ,2 6 3 6 ,7 6 3 7 ,3 8 3 7 ,4 5 3 7 ,3 0 3 7 ,6 7 t io 2 0 ,0 2 0 ,0 6 0 ,0 8 0 ,0 5 0 ,0 5 0 ,0 0 0 ,0 0 0 ,0 9 0 ,1 2 0 ,0 0 0 ,0 1 0 ,0 5 0 ,0 2 0 ,0 4 0 ,0 2 0 ,1 2 0 ,1 6 0 ,4 6 0 ,0 5 0 ,0 3 0 ,0 3 0 ,1 3 0 ,0 2 0 ,0 0 0 ,0 0 0 ,0 8 0 ,0 0 0 ,0 0 0 ,0 4 a l 2 o 3 2 0 ,6 4 2 0 ,3 2 2 0 ,0 3 2 0 ,5 9 2 0 ,6 9 2 0 ,6 1 2 0 ,4 9 2 0 ,5 3 2 0 ,5 5 2 0 ,5 2 2 0 ,6 3 2 0 ,4 1 1 9 ,9 1 2 0 ,2 8 2 0 ,1 9 2 0 ,4 6 1 9 ,8 1 1 9 ,1 7 2 0 ,4 6 2 0 ,3 1 2 0 ,7 0 1 9 ,7 9 2 0 ,2 4 2 0 ,6 3 2 0 ,2 5 2 0 ,1 0 2 0 ,4 6 2 0 ,5 0 2 0 ,2 3 f e o * 2 5 ,4 2 1 7 ,9 5 1 6 ,6 1 2 2 ,1 3 2 6 ,2 6 3 8 ,8 6 3 7 ,0 7 3 3 ,8 4 3 2 ,1 0 3 4 ,6 5 3 4 ,9 4 3 7 ,7 5 2 7 ,9 2 2 3 ,3 9 2 3 ,4 6 2 2 ,6 4 2 3 ,8 0 2 2 ,5 5 2 2 ,9 2 2 1 ,4 5 2 3 ,8 5 2 4 ,6 6 2 5 ,7 2 3 2 ,0 9 3 1 ,2 2 3 4 ,8 3 3 4 ,1 7 3 4 ,9 1 3 4 ,1 5 m n o 1 2 ,2 8 2 0 ,3 4 2 2 ,9 2 1 4 ,5 2 1 1 ,2 0 1 ,9 1 2 ,7 8 6 ,5 1 9 ,2 1 5 ,5 8 4 ,9 9 0 ,7 5 9 ,2 7 8 ,7 2 8 ,8 5 8 ,8 5 9 ,1 4 9 ,2 4 9 ,7 9 1 0 ,1 4 8 ,4 2 1 3 ,8 9 1 1 ,6 3 ,4 3 4 ,6 8 3 ,3 9 2 ,5 7 3 ,7 3 3 ,8 m g o 2 ,2 6 1 ,1 2 0 ,9 8 1 ,7 2 2 ,4 7 2 ,2 9 2 ,7 6 2 ,0 4 1 ,6 8 2 ,2 1 2 ,2 2 2 ,7 7 3 ,5 1 2 ,8 6 2 ,7 6 2 ,4 0 2 ,6 5 2 ,3 6 2 ,5 9 2 ,4 3 2 ,8 1 1 ,5 7 1 ,7 2 2 ,7 7 2 ,4 7 1 ,7 2 2 ,8 5 1 ,7 8 2 ,0 3 c a o 2 ,9 5 4 ,2 2 2 ,5 2 5 ,0 2 2 ,7 6 1 ,2 1 0 ,6 3 1 ,6 0 1 ,6 1 1 ,2 7 0 ,9 7 1 ,9 3 2 ,0 9 6 ,3 9 5 ,9 8 7 ,9 1 6 ,5 6 7 ,5 2 6 ,2 5 7 ,3 1 5 ,9 8 3 ,5 2 3 ,6 0 2 ,9 7 2 ,7 2 1 ,8 8 2 ,1 6 1 ,8 5 2 ,0 0 n a 2 o 0 ,0 0 0 ,0 1 0 ,0 7 0 ,0 3 0 ,0 1 0 ,0 4 0 ,0 6 0 ,0 4 0 ,0 6 0 ,0 5 0 ,0 3 0 ,0 4 0 ,0 4 0 ,0 3 0 ,0 6 0 ,0 3 0 ,0 3 0 ,0 4 0 ,0 3 0 ,0 0 0 ,0 2 0 ,0 1 0 ,0 4 0 ,0 0 0 ,0 1 0 ,0 2 0 ,0 1 0 ,0 5 0 ,0 5 k 2 o 0 ,0 6 0 ,0 4 0 ,0 5 0 ,0 4 0 ,1 7 0 ,0 7 0 ,0 5 0 ,0 5 0 ,0 6 0 ,0 3 0 ,0 5 0 ,0 4 0 ,0 8 0 ,0 8 0 ,0 4 0 ,0 6 0 ,0 5 0 ,0 4 0 ,0 5 0 ,0 3 0 ,0 4 0 ,0 4 0 ,0 5 0 ,0 5 0 ,0 5 0 ,0 4 0 ,0 4 0 ,0 1 0 ,0 8 c r 2 o 3 0 ,0 0 0 ,0 0 0 ,0 0 0 ,0 0 0 ,0 0 0 ,0 4 0 ,0 1 0 ,0 0 0 ,0 3 0 ,0 0 0 ,0 0 0 ,0 7 0 ,0 0 0 ,0 1 0 ,0 0 0 ,0 0 0 ,0 1 0 ,0 2 0 ,0 0 0 ,0 2 0 ,0 0 0 ,0 1 0 ,0 0 0 ,0 1 0 ,0 2 0 ,0 0 0 ,0 1 0 ,0 2 0 t o ta l 1 0 0 ,0 1 1 0 0 ,8 2 9 9 ,4 3 1 0 1 ,0 7 1 0 0 ,4 3 1 0 2 ,2 5 1 0 1 ,2 8 1 0 1 ,4 6 1 0 2 ,2 0 1 0 0 ,9 8 1 0 0 ,8 2 1 0 0 ,9 0 1 0 0 ,4 5 9 8 ,6 8 9 8 ,6 7 1 0 0 ,0 9 9 9 ,6 5 9 8 ,1 6 9 9 ,4 7 9 8 ,5 7 9 9 ,5 1 1 0 0 ,2 5 9 9 ,4 9 9 9 ,2 1 9 8 ,1 8 9 9 ,4 2 9 9 ,7 3 1 0 0 ,1 9 1 0 0 ,0 4 c a ti o n s b a se d o n 1 2 a to m s o f o x y g e n s i 2 ,9 6 1 2 ,9 8 2 2 ,9 9 2 ,9 7 4 2 ,9 7 8 3 ,0 1 2 3 ,0 1 1 2 ,9 7 0 2 ,9 6 3 2 ,9 7 4 2 ,9 9 4 2 ,9 9 2 3 ,0 2 7 2 ,9 9 6 3 ,0 3 3 ,0 0 8 3 ,0 2 0 3 ,0 1 3 ,0 1 3 3 ,0 2 0 2 ,9 9 1 2 ,9 8 9 3 ,0 2 0 3 ,0 2 1 3 ,0 5 2 3 ,0 2 8 3 ,0 2 6 3 ,0 5 2 t i 0 ,0 0 1 0 ,0 0 4 0 ,0 1 0 ,0 0 3 0 ,0 0 3 0 ,0 0 0 0 ,0 0 0 0 ,0 0 5 0 ,0 0 7 0 ,0 0 0 0 ,0 0 1 0 ,0 0 3 0 ,0 0 1 0 ,0 0 2 0 0 ,0 0 7 0 ,0 1 0 0 ,0 3 0 0 0 ,0 0 2 0 ,0 0 8 0 ,0 0 1 0 ,0 0 0 0 ,0 0 0 0 ,0 0 5 0 ,0 0 0 0 ,0 0 0 0 ,0 0 3 a l 1 ,9 7 9 1 ,9 4 2 1 ,9 5 1 ,9 5 2 1 ,9 7 2 1 ,9 3 8 1 ,9 4 1 1 ,9 5 5 1 ,9 5 1 ,9 6 1 1 ,9 6 7 1 ,9 3 9 1 ,8 8 8 1 ,9 4 1 1 ,9 3 0 1 ,9 2 8 1 ,8 8 1 ,8 5 1 ,9 4 1 ,9 5 1 ,9 5 6 1 ,9 0 5 1 ,9 5 3 1 ,9 7 1 1 ,9 6 1 1 ,9 3 4 1 ,9 5 0 1 ,9 6 5 1 ,9 3 2 f e 3 + 0 ,0 9 6 0 ,0 8 7 0 ,0 7 0 ,0 9 3 0 ,0 6 6 0 ,0 3 8 0 ,0 3 8 0 ,0 9 4 0 ,1 0 9 0 ,0 9 1 0 ,0 4 2 0 ,0 7 2 0 ,0 5 6 0 ,0 6 2 0 ,0 1 7 0 ,0 4 1 0 ,0 5 8 0 ,0 6 0 ,0 4 0 ,0 6 0 ,0 0 0 0 ,0 9 7 0 ,0 6 7 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 f e 2 + 1 ,6 3 3 1 ,1 3 1 ,0 8 1 ,3 9 6 1 ,7 1 0 2 ,5 2 2 2 ,4 5 4 2 ,1 9 3 2 ,0 5 3 2 ,2 5 9 2 ,3 2 2 2 ,4 7 4 1 ,8 2 3 1 ,5 2 6 1 ,5 7 1 ,4 7 3 1 ,5 5 1 ,4 8 2 1 ,5 1 0 1 ,4 1 ,5 9 9 1 ,5 8 7 1 ,6 9 4 2 ,1 7 5 2 ,1 4 6 2 ,3 7 8 2 ,3 1 0 2 ,3 6 8 2 ,3 1 4 m n 0 ,8 4 6 1 ,3 9 7 1 ,6 0 ,0 9 0 0 ,7 6 7 0 ,1 2 9 0 ,1 8 9 0 ,4 4 5 0 ,6 2 8 0 ,3 8 3 0 ,3 4 2 0 ,0 5 1 0 ,6 3 2 0 ,6 0 0 0 ,6 0 8 0 ,5 9 9 0 ,6 2 0 ,6 4 0 ,6 7 0 ,7 0 ,5 7 2 0 ,9 6 1 0 ,8 0 4 0 ,2 3 5 0 ,3 2 6 0 ,2 3 4 0 ,1 7 6 0 ,2 5 6 0 ,2 6 1 m g 0 ,2 7 5 0 ,1 3 5 0 ,1 2 0 ,2 0 6 0 ,2 9 8 0 ,2 7 3 0 ,3 3 1 0 ,2 4 6 0 ,2 0 2 0 ,2 6 7 0 ,2 6 8 0 ,3 3 3 0 ,4 2 1 0 ,3 4 6 0 ,3 3 0 ,2 8 6 0 ,3 2 0 ,2 9 0 ,3 1 0 ,2 9 0 ,3 3 6 0 ,1 9 1 0 ,2 1 0 0 ,3 3 5 0 ,3 0 3 0 ,2 0 9 0 ,3 4 4 0 ,2 1 5 0 ,2 4 6 c a 0 ,2 5 7 0 ,3 6 7 0 ,2 2 0 ,4 3 2 0 ,2 3 9 0 ,1 0 3 0 ,0 5 4 0 ,1 3 8 0 ,1 3 9 0 ,1 1 0 0 ,0 8 4 0 ,1 6 7 0 ,1 8 0 0 ,5 5 6 0 ,5 2 0 0 ,6 7 8 0 ,5 7 0 ,6 6 0 0 ,5 4 0 0 ,6 4 0 ,5 1 4 0 ,3 0 8 0 ,3 1 6 0 ,2 5 8 0 ,2 3 9 0 ,1 6 4 0 ,1 8 7 0 ,1 6 1 0 ,1 7 3 n a 0 ,0 0 1 0 ,0 0 3 0 ,0 3 0 ,0 1 1 0 ,0 0 2 0 ,0 1 1 0 ,0 1 8 0 ,0 1 2 0 ,0 1 8 0 ,0 1 6 0 ,0 0 9 0 ,0 1 3 0 ,0 1 3 0 ,0 1 0 0 ,0 2 0 0 ,0 1 0 ,0 1 0 0 ,0 1 3 0 ,0 1 0 0 ,0 0 0 0 ,0 0 6 0 ,0 0 3 0 ,0 1 3 0 ,0 0 0 0 ,0 0 3 0 ,0 0 6 0 ,0 0 2 0 ,0 1 5 0 ,0 1 5 k 0 ,0 1 2 0 ,0 0 9 0 ,0 1 0 0 ,0 0 9 0 ,0 3 6 0 ,0 1 4 0 ,0 1 0 0 ,0 1 1 0 ,0 1 2 0 ,0 0 6 0 ,0 1 1 0 ,0 0 7 0 ,0 1 7 0 ,0 1 7 0 ,0 1 0 ,0 1 3 0 ,0 1 0 ,0 1 0 ,0 1 0 ,0 1 0 ,0 0 9 0 ,0 0 9 0 ,0 1 1 0 ,0 1 1 0 ,0 1 1 0 ,0 0 8 0 ,0 0 9 0 ,0 0 3 0 ,0 1 8 c r 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 0 ,0 0 0 0 ,0 0 2 0 ,0 0 0 0 ,0 0 0 0 ,0 0 2 0 ,0 0 0 0 ,0 0 0 0 ,0 0 4 0 ,0 0 0 0 ,0 0 1 0 ,0 0 0 0 ,0 0 0 0 0 0 ,0 0 0 0 0 ,0 0 0 0 ,0 0 1 0 ,0 0 0 0 ,0 0 1 0 ,0 0 1 0 ,0 0 0 0 ,0 0 1 0 ,0 0 1 0 ,0 0 0 t o ta l 8 ,0 6 1 8 ,0 5 6 8 ,0 7 8 ,0 6 7 8 ,0 7 1 8 ,0 4 3 8 ,0 4 6 8 ,0 7 0 8 ,0 8 4 8 ,0 6 7 8 ,0 4 1 8 ,0 5 4 8 ,0 5 8 8 ,0 5 8 8 ,0 4 8 ,0 4 3 8 ,0 5 8 ,0 5 3 8 ,0 3 8 8 ,0 4 8 ,0 1 5 8 ,0 6 0 8 ,0 5 8 8 ,0 0 5 8 ,0 1 2 7 ,9 9 1 8 ,0 0 7 8 ,0 1 0 8 ,0 1 2 x m g 0 ,1 4 0 ,1 1 0 ,1 0 0 ,1 3 0 ,1 5 0 ,1 0 0 ,1 2 0 ,1 0 0 ,0 9 0 ,1 1 0 ,1 0 0 ,1 2 0 ,1 9 0 ,1 8 0 ,1 7 0 ,1 6 0 ,1 7 0 ,1 6 0 ,1 7 0 ,1 7 0 ,1 7 0 ,1 1 0 ,1 1 0 ,1 3 0 ,1 2 0 ,0 8 0 ,1 3 0 ,0 8 0 ,1 x a lm 0 ,5 6 0 ,3 9 0 ,3 7 0 ,4 8 0 ,5 8 0 ,8 4 0 ,8 1 0 ,7 3 0 ,6 9 0 ,7 6 0 ,7 7 0 ,8 2 0 ,6 0 0 ,5 1 0 ,5 2 0 ,4 9 0 ,5 2 0 ,4 9 0 ,5 0 ,4 7 0 ,5 3 0 ,5 4 0 ,5 7 0 ,7 2 0 ,7 1 0 ,8 0 0 ,7 7 0 ,7 9 0 ,7 7 x sp s 0 ,2 7 0 ,4 5 0 ,5 2 0 ,3 2 0 ,2 5 0 ,0 4 0 ,0 6 0 ,1 4 0 ,2 0 0 ,1 2 0 ,1 1 0 ,0 2 0 ,2 0 0 ,1 9 0 ,2 0 ,1 9 0 ,2 0 0 ,2 0 0 ,2 2 0 ,2 3 0 ,1 9 0 ,3 1 0 ,2 6 0 ,0 8 0 ,1 1 0 ,0 8 0 ,0 6 0 ,0 9 0 ,0 9 x p rp 0 ,0 9 0 ,0 4 0 ,0 4 0 ,0 7 0 ,1 0 0 ,0 9 0 ,1 1 0 ,0 8 0 ,0 6 0 ,0 9 0 ,0 9 0 ,1 1 0 ,1 4 0 ,1 1 0 ,1 1 0 ,0 9 0 ,1 0 0 ,0 9 0 ,1 0 0 ,1 0 ,1 1 0 ,0 6 0 ,0 7 0 ,1 1 0 ,1 0 0 ,0 7 0 ,1 1 0 ,0 7 0 ,0 8 x g rs 0 ,0 8 0 ,1 2 0 ,0 7 0 ,1 4 0 ,0 8 0 ,0 3 0 ,0 2 0 ,0 4 0 ,0 4 0 ,0 4 0 ,0 3 0 ,0 5 0 ,0 6 0 ,1 8 0 ,1 7 0 ,2 2 0 ,1 8 0 ,2 1 0 ,1 8 0 ,2 1 0 ,1 7 0 ,1 0 0 ,1 0 0 ,0 9 0 ,0 8 0 ,0 5 0 ,0 6 0 ,0 5 0 ,0 6 x m g = m g/ (f e+ m g) * t o ta l f e as f eo + f e 2 o 3 also occur as matrix phases. apatite is most abundant in the mn-rich core region (also outlined by the low-ca annulus) and is rare in the outer parts of the garnet except very close to the rim, where a few large grains are located (figure 5, below). monazite increases in abundance from garnet core to rim to matrix, but is entirely lacking from the low-ca zone. xenotime occurs only as a matrix phase. pcm-516 garnet. pelitic schist from the garnetstaurolite-zone, which contains a mineral assemblage of quartz + plagioclase + k-feldspar + muscovite + biotite + garnet. garnet porphyroblasts are subhedral, 0.3-2.5 mm in diameter, pseudopentagonal and poiquiloblastic (figure 2d). they contain inclusion trails of quartz and platy ilmenite that preserve evidence of an earlier fabric occurring at a high angle to the main fabric, which corresponds to a crenulation cleavage. an inclusion-free rim separates the core from the external fabric. garnet is associated with symmetrical and asymmetrical pressure shadows. the fabric outside of garnet porphyroblasts generally displays evidence of recrystallization and grain-size 18 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu figure 5. above, x-ray compositional maps in garnet from sample pcm-361 (garnet-staurolite zone). (a) occurrence of a low-ca annulus characterized by sharp pentagonal and straight outlines. (b) relationship between the euhedral zone of high-mn region and the low-ca annulus. in the x-ray maps, the elemental distributions are indicated by white color for the highest and black color for the lowest concentration. (c) sketch of same garnet showing that the low-ca annulus lacks of inclusions. accessory mineral phases are indicated by black dots (apatite), x marks (xenotime), stars (monazite) and dashes (ilmenite). courtesy of donna whitney. coarsening, respect to the inclusions within garnet. a second stage of garnet growth is indicated by the inclusion-free rim that separates the core from the discordant external fabric. garnet shows decreasing mn from the core (34 mol%) to the rim (16 mol%). fe and mg increase from core to rim, with a small decrease in fe/fe+mg. this garnet is characterized by a small reversal in zoning near the rim. fe, mg and mn distributions are strongly correlated with each other, whereas ca distribution is not (figures 3d, 6b). ca decreases outwards and reaches a minimum at mid-region (4 mole %), then increases towards the rim. inclusions in garnet are ilmenite and quartz in the core region. irregular patches of calcite occur within garnet and in the matrix and may be texturally late, but we note that the distribution of calcite in garnet is confined within the core region, which is bordered by the low-ca ring. calcite does not occur in the grossular-rich outer core/rim of the garnet, although the outer core/rim region is highly fractured. trace elements were not analyzed. pcm-514 garnet. calc-silicate rock from the garnet-staurolite-zone characterized by a mineral assemblage of quartz + plagioclase + k-feldspar + garnet + calcic amphibole. a very interesting texture is observed in this sample, with clusters of numerous fineto medium-grained crystals of garnet, which are 0.1-3.3 mm in diameter, anhedral to subhedral and pseudohexagonal and rounded shape, containing abundant quartz inclusions throughout its core (figure 2f). other inclusions are epidote, plagioclase, biotite and ilmenite. each crystal of garnet has major element zoning (figure 3c), but no detectable trace element zoning in the analyzed elements. the grossular component displays a complex oscillatory zoning, fluctuating between 17 and 24 mol%, opposite to the trend in zoning in almandine (figure 6c). from core to rim, spessartine decreases and fe/(fe+mg) increases. analyzed garnets have at least two high-ca annuli. the variable grossular content may be related to reactions involving other calcic phases in this rock calcic amphibole, plagioclase, epidote). the garnet core, defined as the region within the low-ca ring, contains abundant mineral inclusions. quartz is most common, but epidote, plagioclase, k-feldspar, calcic amphibole, biotite, magnetite and ilmenite also occur. patchy calcite is present in the garnet and appears to be related to fractures. calcite occurs both within and beyond the low-grossular ring in the outer core. figure 8a shows that na is antipathetic with ca, which suggest that the anorthite content in plagioclase varies with the grossular content in garnet and therefore it is necessary a source of ca in addition to the garnet and plagioclase. on the other hand, ti show a positive correlation with ca (figure 8b), which could reflect the consumption of a ti-rich mineral phase such as ilmenite or rutile during garnet growth. similar oscillatory zoning has been reported by other authors (e.g., jamtveit et al., 1995; holten et al., 1997; ivanova et al., 1998; jamtveit, 1999; pollok et al. 2001), probablye due to fluctuations in supersaturation, which might have originated from external (boiling, fluid mixing, temperature, and pressure fluctuations) or internal factors, resulting from self-organization in the interplay of fluid convection and crystal growth in the near vicinity of the growing crystals (jamtveit et al. 1995; holten et al. 1997; pollok et al. 2001). pcm-420 garnet. quartz-feldespatic pelitic rock from the garnet-staurolite-zone that contains a mineral assemblage of quartz + plagioclase + k-feldspar + biotite + garnet. garnet occurs as large (1-2 mm in diameter) subhedral and subrounded porphyroblasts with cores densely packed with very fine inclusions of quartz and minor ilmenite with no preferred orientation that give the host crystal a spongy appearance and inclusion-free rims or as small (0.5-1 mm in diameter) subhedral and subrounded grains with inclusion-free cores (figure 2b). the fabric outside of garnet porphyroblasts shows recrystallization and grain-size coarsening, respect to the inclusions within garnet. small crystals of magnetite occurs adjacent to garnet. garnet exhibits growth zoning: from core to near rim, there is an increase in mg (from 3 to 12 mol%) and a decrease in mn (from 43 to 6 mol%; figure 6d) and in ca (from 18 to 8 mol%), with a small decrease in the fe/(fe+mg) ratio. concentrations of mn, mg and fe in the cores of the small garnets correspond with near-rim compositions in the 19 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia 20 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu larger crystals, but ca composition is not similarly systematic. the concentration of ca in the cores of small garnets does not correspond to near-rim compositions in the larger crystals. no discontinuity in the compositional zoning of mn occurs at the boundary between the inclusion-rich core and the inclusion-free rim. within the high-mn region, which also corresponds to an inclusion-rich region in the garnet, zoning is not concentric about the core (figure 6d). pcm-473 garnet. pelitic schist from the staurolite-kyanite-zone. it is characterized a mineral assemblage of quartz + plagioclase + muscovite + biotite + garnet + staurolite. garnet in this sample ocfigure 6. x-ray compositional maps in pelitic garnets from the garnet-staurolite zone. (a) and (b) illustrate low-ca annuli with sharp pentagonal outlines in garnet from samples pcm-361 and pcm-516, respectively. in (a), at the bottom of the garnet, the low-ca annulus is truncated against biotite; white areas correspond to calcite (garnet core region and matrix) and ca-rich plagioclase (matrix). (c) cyclic ca zoning in garnet from sample pcm-514 with lowand high-ca annuli with sharp hexagonal outlines. (d) mn distribution in garnet from sample pcm-420, showing a small reversal zoning at rim. numbered spots are keyed to analyses in table 2. in the x-ray maps, the elemental distributions are indicated by white color for the highest and black color for the lowest concentration. curs as large porphyroblasts, which are up to 5 mm in diameter, anhedral (generally elongated) and contain a small number of inclusions of quartz, biotite, muscovite, plagioclase, ilmenite and magnetite. it has been highly cracked and partially replaced by quartz, muscovite and sillimanite (figure 2e). fibrolite commonly occurs within muscovite that is associated to garnet or as minute crystals sometimes penetrating quartz grains, which develop embayment in garnet, within which the polymorph of al2sio5 also occurs. the staurolite usually occurs as relicts in muscovite. rutile occurs only as a matrix phase. garnet exhibits reverse zoning, with a decrease in mg (from 11 to 6 mol%), figure 9a, and increase in mn (from 7 to 10 mol%), figure 9b, from core to rim and a small increase in the fe/(fe+mg) ratio. composition is more homogenous in the interior of the crystal and ca content increases slightly within the outer core (5-7 mol%), where it reaches a maximum, then decreases to 5 mol% at rim, developing a low-ca annulus (figure 9c). reaction history a detailed discussion of the reaction history among major mineral phases in metapelitic rocks of the silgará formation at the southwestern santander massif has been presented by ríos (1999) and ríos et al. (2003). therefore, only metamorphic reactions in garnet-bearing pelites will be considered here. the typical barrovian progression of metamorphic reactions in these rocks involve low-grade garnet growth, intermediate-grade garnet consumption during staurolite producing and high-grade resumption of garnet growth during kyanite producing at expenses of staurolite, which is similar to what reported kohn & malloy (2004). the first appearance of garnet was the result of a fe-mg-mn continuous reaction. the characteristic garnet of the garnet zone is almandine-rich and probably grew by two continuous reactions: (1) chl + ms = grt + bt + qtz + h2o (2) chl + ms + qtz = grt + bt + h2o which can explain the formation of almandine and pyrope in garnet, but the high ca amount observed in the garnet core or rim suggests that a ca-rich phase also participated in the garnet-forming reaction. the source of mn is inferred to be mn-rich epidote, probably enclosed in clinozoisite or other matrix minerals. the garnet isograd reaction is 21 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia figure 7. back-scattered electron images of garnet in sample pcm-361 (garnet-staurolite zone), which contains ilmenite inclusions. (a)-(b) unreacted crystal of ilmenite and adjacent pseudomorph of rutile after ilmenite. strongly dependent on bulk rock composition and specially the mno and cao contents. in rocks of high amount of those components, garnet may appear at temperatures below 450°c whereas in rocks of low amount of them, garnet may not appear to well above 500°c (spear, 1993). respect to the cyclic zoning observed in calc-silicate rocks (e.g., sample pcm-514, figure 6c), we consider that the highand low-ca annuli in garnet may have grown, respectively, by the following reactions (3) chl + ms + ep = grt + bt + pl + ca-amp + + bt + h2o, (4) chl + pl + ms + qtz = grt + bt + ca-amp + + h2o however, the behavior of ca-rich fluids should be evaluated further by considering the trace element concentrations in garnet. according to pollok et al. (2001), oscillatory zoning in garnet can be explained through two different sets of processes: local growth dynamics vs external forcing mechanisms. at the staurolite isograd, the first appearance of staurolite can be related to two reactions. textural evidence shows that staurolite may be produced at the expense of garnet by the discontinuous reaction (5) grt + chl + ms = st + bt + qtz + h2o in garnet-free assemblages, staurolite may be produced by the continuous reaction (6) chl + ms = st + bt + qtz + h2o which can be deduced from metamorphic textures, considering that staurolite grew mainly in phyllosilicate-rich layers (graebner and schenk, 1999). when chlorite is completely consumed, as staurolite becomes less abundant it may be preserved as inclusions or armoured relics in muscovite and garnet growth can be explained by the reaction (7) st + bt + qtz = grt + ms + h2o staurolite may dissapears with further growth of garnet at the kyanite isograd as a result of the discontinuous reaction (8) st + ms + qtz = ky + grt + bt + h2o if garnet is not involved, the staurolite breakdown is produced by the continuous reaction (9) st + ms + qtz = al2sio5 + bt + h2o which reflects lower temperatures than the discontinuous reaction (6) and p-t conditions nearly independient of the mno and cao contents of the rock. however, if garnet breaks down as sillimanite 22 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu figure 8. zoning profiles rim-core-rim in garnet from sample pcm-514 (garnet-staurolite zone), showing the correlation between the weight percent of cao and the (a) na2o and (b) tio2 contents. lower axis indicates distance (mm). left axis indicates concentration (wt%). is produced, so the initial growth of sillimanite may be considered to result from two separate continuous reactions operating simultaneously (10) st + ms + qtz = sil + bt + h2o (11) grt + ms = bt + sil + qtz chemical zoning and sillimanite inclusions in muscovite and quartz may indicate decrease in pressure after the formation of sillimanite by reaction (9) and during staurolite-consuming reactions (increasing temperature). there is textural evidence to support the first appearance of kyanite in pelitic rocks of the silgará formation by the polymorphic reaction andalusite � kyanite. on the other hand, a net reaction such as andalusite � sillimanite must not have occurred at some point during metamorphism, because we don’t really know if the andalusite reacted directly to produce the sillimanite or if the andalusite was first dissolved in a fluid phase, with the fluid phase carrying the dissolved components to a new location where the fluid then precipitated the sillimanite. however, the presence of sillimanite, kyanite and andalusite in the area or in some samples seems to suggest a more complex p-t metamorphic evolution. discussion in this study we reveal a number of important observations regarding the major and trace element zoning and its correlation with accesory mineral phases in garnet-bearing pelites and associated lithologies of the silgará formation at the southwestern santander massif. fe, mg and mn zoning is strongly correlated with respect to the degree of local equilibrium in contrast with ca zoning that does not mimic zoning of those elements exactly (e.g., chernoff & carlson, 1997; spear & daniel, 1998). this strong correlation supports the hypothesis that these elements achieved a close approach to local equilibrium during garnet growth. if variations of these elements were produced only by local variations in bulk composition, then mass balance and phase equilibrium considerations would predict antithetic zoning of fe, mg and xfe with respect to mn. the chemical zoning in the analyzed garnets may be originated by growth zoning, as a result of changing external conditions (pressure, temperature or bulk rock chemistry) and by diffusion zoning also due to changing external conditions, but with no growth or consumption of garnet. according to spear (1993), a change in chemical zoning character from growth zoning to diffusion zoning by progressive homogenization is attributed to diffusion with increasing metamorphic grade. however, a local metasomatic 23 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia figure 9. x-ray compositional maps showing (a) mg, (b) mn and (c) ca distribution in garnet from sample pcm-473 (staurolite-kyanite zone). white arrows indicate embayments. numbered spots are keyed to analyses in table 2. in the x-ray maps, the elemental distributions are indicated by white color for the highest and black color for the lowest concentration. event has also played a very important role in controlling metamorphic reactions. major element zoning. the analyzed garnets usually show a normal zoning with mn content decreasing from core to rim, suggesting a prograde metamorphism. a high mn content correlates with a high fe/(fe+mg), which suggests that areas of high mn content formed at lower temperature. therefore, mn zoning can be used as a time line in these rocks (spear and daniel, 1998). crystal size distribution (csd) analysis of garnets is consistent with continuous nucleation and growth of garnet throughout the garnet crystallization episode (ríos et al., 2003), which indicates that large garnets nucleated early and the smallest garnets nucleated last. in contrast to the observations of chernoff & carlson (1997) for garnets in the picuris range, new mexico, high-mn, low-mg+fe cores or low-mn, high-mg+fe cores do not correlate with garnet size in the silgará formation. however, the strong correlation of mn and fe, regardless of garnet crystal size and spatial distribution, suggests a homogenous distribution of these divalent cations in the intergranular medium during garnet growth. ca zoning can exhibit different trends than the fe-mn-mg and is generally unrelated to zoning patterns in fe, mn and mg, although in some cases ca and mn zoning patterns are spatially related (e.g., sample pcm-361, figure 5a, 5b). it is well known that changes in the reactant mineral phases and product assemblages occur at different times during the reaction history of garnet crystals of different size and, therefore, they can not be the result of any event affecting the entire rock, such as a change in pressure, temperature or fluid composition but they reflect kinetic factors that cause ca to fail achieving chemical equilibrium during garnet growth. although garnet outlines are sometimes slightly rounded (e.g., pcm361, figure 6a; pcm-516, figure 6b), it is possible to observe that their rims were basically parallel to the euhedral low-ca annuli and that the majority of garnet consumption occurred at the corners or intersections of growth crystal faces of garnet. a likely explanation for the difference in appearance of the annuli is that a particular thin section plane did not necessarily pass through the centre of the garnet and in some cases a high-annuli at core could be missed. garnet shows abrupt variations in ca distribution from core to rim, probably due to the consumption of ca-enriched mineral phases in garnet-producing reactions, developing zoning patterns with euhedral low-ca annuli (figures 6a, 6b) parallel to the garnet outlines, which is typical of growth zoning and suggest that very little diffusion took place after growth (chernoff & carlson, 1999). cal-silicate rocks contain a variety of ca-rich phases (e.g., epidote, plagioclase, ca amphibole and apatite) in addition to garnet. if the consumption or production of these phases and the changes in how ca and other elements are locally partitioned between garnet and other minerals, accounts for garnet zoning, there should be trends in zoning in other ca-rich phases. matrix plagioclase typically displays normal zoning, with an increase in xan spatially associated with a decrease in xgrs of garnet rims (ríos et al., 2003). low-ca annuli within garnets are unlikely to have been caused by resorption of garnet because the bands are euhedral. a garnet resorption would be expected to produce an irregular, embayed garnet margin and an overgrowth on such irregular boundary should produce a subhedral to anhedral annulus, which has not been observed in this study. a sharp decrease of grossular content from core to the mid-region has been interpreted by menard & spear (1993) as produced by resorption of garnet during productions of staurolite or loss of epidote from the assemblage, which is unlikely because the garnet core is euhedral. a change in mineral assemblage may account for some of the zoning trends observed, but to explain the complex oscillatory zoning observed in calc-silicate rocks (sample pcm-514, figure 6c), this would require appearance and disappearance of ca-rich phases or some other process that control fluctuations in the availability of elements. the ca-rich bands in garnet may correlate with the breakdown of epidote as well as with the presence of a ca-rich fluid. garnet growth continued with decreasing grossular content after each of these events, showing that ca was fractionated into garnet, plagioclase and/or calcic amphibole. ti and ca contents in garnet from sample pcm-514 are positively correlated, which may reflect consumption of ilmenite or rutile to liberate ti during growth of carich garnet 24 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu or metasomatic addition of ti. there is not doubt that a complex oscillatory zoning as described here has a controversial origin and we agree with stowell et al. (1996) in the sense that it is difficult to explain it by a cyclic addition and loss of mineral phase(s) from the chemical system. instead, it could reflect episodic metasomatism by influx of ca-rich fluids during garnet growth near of adjacent to hydrothermal veins of quartz-epidote. therefore, the results obtained by ríos & takasu (1999) for ca concentrations would not be interpreted as event markers or as recording complex barometric histories, when in fact they reflect only local effects that can not have occurred simultaneously throughout the rock. trace element zoning. analysis of the trace element zoning in garnet was done as a useful supplement to major-element zoning to improve the petrological interpretation. according to pyle & spear (1999), trace elements zoning in garnet growing in a close system depends on the presence or absence of “saturing” mineral phases such as xenotime for y, xenotime, apatite or monazite for p and ilmenite, titanite or rutile for ti. however, the accessory phases that likely reflect interactions with major phases could be xenotime and monazite, as a consequence of reaction coupling with garnet and the high compatibility of y in all three phases. in our case, we have documented the presence of these accessory mineral phases, but unfortunately they did not show trace element zoning, except in sample pcm-441 (figure 4b), where y zoning shows a remarkably euhedral core that is not really parallel to crystal faces of garnet, representing a growth zone that probably was not modified by diffusion. high-y annuli in garnet have been reported by pyle & spear (1999) in staurolite-bearing samples, which may form by garnet overgrowth of proximal matrix enrichment in y due to garnet consumption during discontinuous staurolite-forming reactions. therefore, the high-y annulus observed in the garnet studied from the garnet zone can not be explained by a resorption-regrowth process, which also will produce a chemical zoning characterized by an asymmetric annulus with steeper slopes and embayed and irregular shapes. however, the growth history of this garnet can be represented by a period of garnet growth at an approximately constant activity of y (xenotime stable), followed by garnet at much lower y activity (xenotime absent). lanzirotti (1995) describes different mechanisms for annular rings of high-y in garnet, such as open system-fluid infiltration, garnet resorption and renewed growth (unlikely argument for y zoning described above), changes in garnet growth rates and comsumption of y-enriched mineral phases. therefore, the production of an annulus without garnet resorption requieres that the accessory-phase-out reaction be essentially discontinuous or the result would be a gradual, rather than abrupt, increase in garnet major or trace element content (pyle & spear, 1999). hickmott & spear (1992) concluded that concentrations of y, zr and ree in garnet should vary with ca concentrations because these elements susbtitute into epidote-group minerals. therefore, reactions involving consumption of these phases should supply both ca and these trace elements for garnet growth. unfortunately, garnet from sample pcm-514 does not show trace-element zoning in y and therefore is difficult to correlate it with ca in garnet. the addition of calcium would control chemical reactions producing garnet (xgrs increases), plagioclase (xan increases), calcic amphibole and epidote, whereas reactions in absence of metasomatism would return xgrs and xan to previous levels. xenotime is present only as a matrix accessory phase in sample pcm-361 (figure 5c), which can be interpreted as a stable mineral phase throughout the entire growth history of garnet. at this respect, pyle & spear (1999) consider that the scarcity of matrix xenotime at grades above the garnet-zone in garnet-bearing pelites implies that the growth of garnet is accompanied by the consumption of xenotime. monazite is a common accessory mineral in samples from the higher metamorphic zones (garnet-staurolite and staurolite-kyanite), being virtually absent at garnet-grade. monazite inclusions occur in garnet (e.g., pcm-361, figure 5c), suggesting that they grew before or during growing of garnet. garnet growth should be previous to or synchronous with matrix monazite growth. its growth may be related to the prograde consumption of allanite as been reported in different studies, assuming p is derived from apatite (e.g., catlos et al., 2001). although it is probable that a 25 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia precursor lree mineral was present, garnet, plagioclase and phyllosilicates host sufficient p and lrees and the reaction between them could stabilize monazite (kohn & malloy, 2004). where analyses are not disrupted by the presence of inclusions, ti concentration decreases from core to rim. the increasing concentration towards the rim in garnet may be a result of changes in the ti-bearing minor phase assemblage from ilmenite-rutile in the garnet core to rutile in the garnet rim. accesory ilmenite, titanite and rutile coexist with the silicates. inclusions of ilmenite (by far the most prominent ti-bearing phase) in garnet have been partly replaced by rutile as observed in figure 7 and show from core to rim an increase in mno and a decrease in feo. apatite is abundant in all rocks throughout the garnet outboard and in the matrix. accessory phases such as zircon, monazite, titanite, epidote and allanite may contain significant y, but these should be of limited modal extent in the samples examined with respect to garnet to explain why there is not apparent trace element zoning in garnet (except for y in sample pcm-441). diffusion and resorption. garnet usually shows a normal zoning with mn content decreasing from core to rim, although a minimum mn content near rim (e.g., sample pcm-420, figure 6d) sometimes is observed, and this type of zoning is characterized as reversal zoning, which in many cases reflects post-peak resorption and reequilibration during cooling by elemental diffusion during retrograde metamorphism. whitney and ghent (1993) consider that no significant increase in mn at garnet rims suggests no major late metamorphic resorption of garnet. however, a resorption process cannot explain a reversal zoning in euhedral garnet. possible explanations involve growth or post-crystallization modification. in the case of the former, asymmetric zoning can be explained by overgrowth of mn-rich phase(s) (top half of crystal in figure 6d) as well as a lower mn phase by garnet at low temperature (so that diffusion was limited and mn was confined to the vicinity of the overgrown phase(s); e.g., hirsch et al., 2003). the core, represented by a low mn concentration, is not situated in the geometrical center of the grain, implying either asymmetrical growth or that a significant amount of resorption has taken place, which is compatible with the corroded appearance of the rim. mn is not concentric about individual parts of the garnet, but rather is zoned in irregular, amoeba-like shapes, a pattern that reflects fast growth along grain boundary surfaces and slower dissolution and replacement of quartz inclusions (spear & daniel, 1999). however, we consider that this garnet where is in contact with progressive shear zones displays a tectonic dissolution, as revealed by the chemical zoning, which is abruptly truncated against the main metamorphic foliation of the rock. chemical zoning in garnet from sample pcm-473 can be affected by a post-growth thermal history. garnet zoning is reversal in this sample and is best explained by diffusion during partial resorption of first stage garnet, with a second stage of growth near the outer rim of garnet, marking the onset of a new prograde garnet producing reaction. resorption is compatible with the corroded appearance of the rim and also with the traditional interpretation of the low-mg (figure 9a) and high-mn (figure 9b) rim. figure 9c illustrates partial preservation of a low-ca annulus at the outermost rim, which is truncated by dissolution and resorption of garnet, developing embayments at rim, the chemical zoning are patchy and the mn or mg (samples pcm-361, pcm-420, pcm-514 or pcm-516) and y (sample pcm-441) distributions are characterized by a small reversal zoning, indicating some garnet resorption and back diffusion after maximum temperature was achieved as a consequence of reaction zones of chlorite or biotite around garnet that could contain inclusions of accessory mineral phases such as xenotime, monazite, epidote, zircon or titanite that can incorporate significant amounts of y as proposed in differet studies (e.g., wopenka et al., 1996; heinrich et al., 1997; finger et al., 1998). conclusions we propose that our zoning and textural observations are most consistent with a continuous garnet growth process, involving various ca-rich phases, and that garnet growth may have varied in terms of different 26 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu garnet producing reactions during prograde metamorphism. the occurrence of inclusion-rich and inclusion-free zones in garnet has been explained by a change in the rate of garnet growth (yang and rivers, 2001), i.e., garnet growth in the inclusion-rich zones must have been sufficiently rapid to allow entrapment of abundant inclusions. therefore, we can interpret inclusion-poor rims and inclusion-rich cores as representing different growth rates, which is supported here by the occurrence of garnet showing low-ca annuli lack of inclusions (e.g., pcm-361 and pcm-516) and high-mn inclusion-rich core (e.g., pcm-420). in sample pcm-361, for example, the high ca / low mn rim zone contains apatite, monazite and ilmenite aligned parallel to the margins of the garnet, whereas the euhedral low-ca annulus within the garnet corresponds to a change in mineral inclusion abundance, but does not correspond to a change in the mineral inclusion assemblage itself (figures 5, 6a). the lowand/or high-ca annuli may be used as time markers during garnet growth history and truncation of annuli is not only an evidence of tectonic dissolution in progressive shear zones as proposed by ríos et al. (2003) but also an excellent indication of subsequent garnet resorption. accessory-phase assemblages may consider an early garnet growth at the expense of chlorite or biotite, with resultant xenotime consumption and production of monazite. most importantly, this study shows that monazite crystallization is not an isolated event, but occurs throughout the metamorphic history of the rock. we have shown that valuable petrogenetic information can be obtained from major and trace-element distribution in pelitic and associated rocks of the silgará formation and that a complete understanding of reaction history may be achieved if accessory phases are considered. however, our interpretation could be tested by running gibbs method simulations, utilising internally consistent thermodynamic datasets and relevant activity-composition models. acknowledgements this research forms part of the m.sc. program carried out by c. ríos at shimane university. we are very grateful to the instituto colombiano para el desarrollo de la ciencia y la tecnología “francisco josé de caldas” (colciencias), as well as the universidad industrial de santander, entities that through grant no. 1102-05-083-95 financially supported fieldwork (research project: “metamorphism and associated metallogeny of the santander massif, eastern cordillera, colombian andes”). we are indebted to the research center for coastal lagoon environments at shimane university for allowing us the use of the electron microprobe analyzer for acquisition and assistance of major element data. we also thank dr. d. whitney from the department of geological and geophysics at university of minnesota for her assistance with the adquisition of major and trace element data, substantial contributions and help in manuscript preparation. references anderson, d.e., and buckley, g.r (1973). zoning in garnets: diffusion models. contributions to mineralogy and petrology, vol. 40, pp. 87-104. boinet, t., bourgois, j., bellon, h., et toussaint, j. (1985). age et repartition du magmatism premesozoique des andes de colombie. comptes rendus hebdomadaires des séaces de l’académie des sciences. serie d: sciences naturalles, vol. 300, pp. 445-450. bucher, k., and frey, m. (1994). petrogenesis of metamorphic rocks. 6th edition complete revision of winkler’s textbook. springer-verlag, berlin heidelberg, germany. carlson, w.d. (1989). the significance of intergranular diffusion to the mechanisms and kinetics of porphyroblast crystallization. contributions to mineralogy and petrology, vol. 103, pp. 1-24. catlos, e., harrison, t., kohn, m., grove, m., lovera, o., ryerson, f., and upreti, b. (2001). geochronologic and thermobarometric constraints on the evolution of the main central thrust, central nepal himalaya. journal of geophysics research, vol. 106, pp. 16177–16203. 27 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia chamberlain, c.p., and conrad, m.e. (1993). oxygen-isotope zoning in garnet: a record of volatile transport. geochimica et cosmochimica acta, vol. 57, pp. 2613-2629. chernoff, c., and carlson, w. (1997). disequilibrium for ca during growth of pelitic garnet. journal of metamorphic geology, vol. 15, pp. 421-438. chernoff, c. and carlson, w. (1999). trace element zoning as a record of chemical disequilibrium during garnet growth. geology, vol. 27, pp. 555-558. cygan, r.t., and lasaga, a. (1982). crystal growth and the formation of chemical zoning in garnets. contributions to mineralogy and petrology, vol. 79, pp. 187-200. dörr, w., grösser, j., rodríguez, g., and kramm, u. (1995). zircon u-pb age of the paramo rico tonalite-granodiorite, santander massif (cordillera oriental, colombia) and its geotectonic significance. journal of south american earth sciences, vol. 8, pp. 187-194. erambert, m., and austrheim, h. (1993). the effect of fluid and deformation on zoning and inclusion patterns in poly-metamorphic garnets. contributions to mineralogy and petrology, vol. 115, pp. 204-214. finger, f., broska, i., roberts, m.p., and schermaier, a. (1998). replacement of primary monazite by apatite-allanite-epidote coronas in an amphibolite facies granite gneiss from the eastern alps. american mineralogist, vol. 83, pp. 259-272. gibson, h., carr, s., brown, r., and hamilton, m. (2004). correlations between chemical and age domains in monazite and metamorphic reactions involving major pelitic phases: an integration of id-tims and shrimp geochronology with y-th-u x-ray mapping. chemical geology, vol. 211, pp. 237-260. goldsmith, r., marvin, r., and mehnert, h. (1971). radiometric ages in the santander massif, eastern cordillera, colombian andes. u.s. geological survey professional paper, vol. 750, pp. 41-49. heinrich, w., andrehs, g., and franz, g. (1997). monazite-xenotime miscibility gap thermometry. i. an empirical calibration. journal of metamorphic geology, vol. 15, pp. 3-16. hickmott, d., and shimizu, n. (1990). trace element zoning in garnet from the kwoiek area, british columbia: disequilibrium partitioning during garnet growth? contributions to mineralogy and petrology, vol. 104, pp. 619-630. hickmott, d., and spear, f. (1992). majorand trace-element zoning in garnets from calcareous pelites in the nw shelburne falls quadrangle, massachusetts: garnet growth histories in retrograded rocks. journal of petrology, vol. 33, pp. 965-1005. hickmott, d.d., shimizu, n., spear, f.s., and selverstone, j. (1987). trace-element zoning in a metamorphic garnet. geology, vol. 15(6), pp. 573-576. hirsch, d., prior, d., and carlson, w. (2003). an overgrowth model to explain multiple, dispersed high-mn regions in the cores of garnet porphyroblasts. american mineralogist, vol. 88, pp. 131-141. hollister, l.s. (1966). gamet zoning: an interpretation based on the rayleigh fractionation model. science, vol. 154, pp. 1647-1651. holten, t., jamtveit, b., meakin, p., cortini, m., blundy, j., and austrheim, h. (1997). statistical characteristics and origin of oscillatory zoning in crystals. american mineralogist, vol. 82, pp. 596–606. ivanova, t.i., shtukenberg, a.g., punin, y.o., frank-kamenetskaya, o.v., and sokolov, p.b. (1998). on the complex zonality in grandite garnets and implications. mineralogical magazine, vol. 62, pp. 857–868. jamtveit, b., wogelius, r.a., and fraser, d.g. (1993). zonation patterns of skarn garnets: re28 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu cords of hydrothermal system evolution. geology, vol. 21, pp. 113-116. jamtveit, b., and hervig, r.l. (1994). constraints on transport and kinetics in hydrothermal systems from zoned garnet crystals. science, vol. 263, pp. 505-508. jamtveit, b., ragnarsdottir, k.v., and wood, b.j. (1995). on the origin of zoned grossular-andradite garnets in hydrothermal systems. european journal of mineralogy, vol. 7, pp. 1399–1410. jamtveit, b. (1999). crystal growth and intracrystalline zonation patterns in hydrothermal environments. in growth, dissolution and pattern formation in geosystems, b. jamtveit and p. meakin eds, p.65–84, kluwer academic publishers, dordrecht, the netherlands. kohn, m., and malloy, m. (2004). formation of monazite via prograde metamorphic reactions among common silicates: implications for age determinations. geochimica et cosmochimica acta, vol. 68, pp. 101–113. kretz, r. (1983). symbols for rock-forming minerals. american mineralogist, vol. 68, pp. 277-279. lanzirotti, a. (1995). yttrium zoning in metamorphic garnets. geochimica et cosmochimica acta, vol. 59, pp. 4105-4110. lasaga, a. (1983). geospeedometry: an extension of geothermometry. in s.k. saxena, ed., kinetics and equilibrium in mineral reactions, springer-verlag, new york. p. 81-114. menard, t., and spear, f. (1993). metamorphism of calcic pelitic schists, strafford dome, vermont: compositional zoning and reaction history. journal of petrology, vol. 34, pp. 977-1005. ordoñez, j. (2003). petrology and geochemistry of the granitoids at the santander massif, eastern cordillera, colombian andes. unpublished master thesis, shimane university, matsue (japan), 150p. pollok, k., jamtveit, b., and putnis, a. (2001). analytical transmission electron microscopy of oscillatory zoned grandite garnets. contributions to mineralogy and petrology, vol. 141, pp. 358–366. pyle, j. and spear, f. (1999). yttrium zoning in garnet: coupling of major and accessory phases during metamorphic reactions. geological materials research, vol. 1, pp. 1-49. ríos, c. (1999). chemical compositions of the constituent minerals and p-t conditions of the low-grade silgará formation metamorphic rocks in the santander massif, eastern cordillera, colombian andes. unpublished master thesis, shimane university, matsue (japan), 207p. ríos, c., y garcía, c. (2001). primera ocurrencia de los tres polimorfos de al2sio5 en las rocas metapelíticas de la formación silgará, región suroccidental del macizo de santander, cordillera oriental, andes colombianos. boletín de geología uis, vol. 23, pp. 51-59. ríos, c., garcía, c., andyle, j., & takasu, a. (2003). tectono-metamorphic evolution of the silgará formation metamorphic rocks in the southwestern santander massif, colombian andes. journal of south american earth sciences, vol. 16, pp. 133-154. ríos, c., and takasu, a. (1999). chemical zoning of garnet from the low-grade metamorphic rocks of the silgará formation, santander massif, eastern cordillera (colombian andes). geosciences reports of shimane university, vol. 18, pp. 97-107. schwandt, c., papike, j., and shearer, c. (1996). trace element zoning in pelitic garnet of the black hills, south dakota. american mineralogist, vol. 81, pp. 1195-1207. skelton, a., annersten, h., and valley, j. (2002). �o18 and yttrium zoning in garnet: time markers for fluid flow? journal of metamorphic geology, vol. 20, pp. 457-466. 29 a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombia spear, f. (1993). metamorphic phase equilibria and pressure-temperature-time paths. mineralogical society of america, washington, dc. spear, f., and daniel, c. (1998). three-dimensional imaging of garnet porphyroblast sizes and chemical zoning: nucleation and growth history in the garnet zone. geological materials research, vol. 1, pp. 1-42. stowell, h., menard, t., and ridgway, ch. (1996). ca-metasomatism and chemical zoning of garnet in contact-metamorphic aureoles, juneau gold belt, southeastern alaska. the canadian mineralogist, vol. 34, pp. 1195-1209. tracy, r.j., robinson, p., and thompson, a.b. (1976). garnet composition and zoning in the determination of temperature and pressure of metamorphism: central massachusetts. american mineralogist, vol. 61, pp. 762-775. ward, d., goldsmith, r., cruz, b., jaramillo, c., y restrepo, h. (1973). geología de los cuadrángulos h-12, bucaramanga y h-13, pamplona, departamento de santander. u.s. geological survey e ingeominas. boletín geológico, vol. xxi, 132p. whitney, d., and ghent, e. (1993). prograde reactions and garnet zoning reversals in staurolite schist. british columbia: significance for thermobarometric interpretations. journal of metamorphic geology, vol. 11, pp. 779-788. wopenka, b., jolliff, b.l., zinner, e., and kremser, d.t. (1996). trace element zoning and incipient metamictization in a lunar zircon: application of three microprobe techniques. american mineralogist, vol. 81, pp. 902-912. yang, p., and rivers, t. (2001). chromium and manganese zoning in pelitic garnets and kyanite: spiral, overprint and oscillatory (?) zoning patterns and the role of growth rates. journal of metamorphic geology, vol. 19, pp. 455-474. yang, p., and rivers, t. (2002). the origin of mn and y annuli in garnet and the thermal dependence of p in garnet and y in apatite in calc-pelite and pelite, gagnon terrane, western labrador. geological materials research, vol. 4 (http://gmr.minsocam/papers/v4/v4n1). young, e.d., and rumble, d., iii (1993) the origin of correlated variations in in-situ 180/160 and elemental concentrations in metamorphic garnet from southeastern vermont, usa. geochimica et cosmochimica acta, vol. 57, pp. 2585-2597. 30 carlos alberto ríos reyes, óscar mauricio castellanos alarcón and akira takasu letter from editor a new interpretation for the garnet zoning in metapelitic rocks of the silgará formation, southwestern santander massif, colombiacarlos alberto ríos reyes, oscar mauricio castellanos alarcón and akira takasu 7 material assimilation in a shallow diapiric forceful intrusion: evidence from microstructures and csd analysis in a porphyritic intrusive body, “la línea” tunnel, central cordillera, colombialorena rayo and carlos a. zuluaga 31 mineralization controls and petrogenesis of the rare metal pegmatites of nasarawa area, central nigeriaakintola, o.f. and adekeye, j.i.d. 44 denoising resistivity phosphate “disturbances” using haar mother wavelettransform (sidi chennane, morocco)saad bakkali and mahacine amrani 62 petrogenesis of the metacarbonate and related rocks of the silgará formation, central santander massif, colombian andes: an overview of a “reaction calcic exoscarn”ríos, c.a., castellanos, o.m., gómez s.i. and ávila, g.a. 72 seismic velocity determination in gravel and sands using piezocrystalsarmando luis imhof and juan carlos santamarina 107 untitled earth sciences research journal earth sci. res. j. vol. 13, no. 1 (june 2009): 82-89 comparative evaluation of the influencing effects of geomagnetic solar storms on earthquakes in anatolian peninsula sadik cengiz yesugey dokuz eylul university, faculty of architecture, campus of tinaztepe, dogus str. no.209 kurucesme, buca, ýzmir turkey. phone: +902324535716 fax: +902324532986 corresponding author email: cengiz.yesugey@deu.edu.tr abstract earthquakes are tectonic events that take place within the fractures of the earth’s crust, namely faults. above certain scale, earthquakes can result in widespread fatalities and substantial financial loss. in addition to the movement of tectonic plates relative to each other, it is widely discussed that there are other external influences originate outside earth that can trigger earthquakes. these influences are called “triggering effects”. the purpose of this article is to present a statistical view to elaborate if the solar geomagnetic storms trigger earthquakes. as a model, the research focuses on the anatolian peninsula, presenting 41 years of historical data on magnetic storms and earthquakes collated from national and international resources. as a result of the comparative assessment of the data, it is concluded that the geomagnetic storms do not trigger earthquakes. keywords: earthquake, earthquake triggering, geomagnetic storms. resumen los terremotos son eventos tectónicos que tienen lugar dentro de las fracturas de la corteza terrestre, nominalmente las fallas. por encima de cierta escala los terremotos pueden resultar considerables pérdidas humanas y financieras. adicionalmente de los movimientos de las placas tectónicas, es ampliamente discutido que hay otras influencias originadas en el exterior de la tierra que pueden desencadenar terremotos. estas influencias son llamadas “efectos disparados”. el propósito de este artículo es presentar una observación estadística sobre si las tormentas solares geomagnéticas desencadenan terremotos. como un modelo, la investigación se centra sobre la península anatolia, presentando 41 años de datos históricos sobre tormentas 82 manuscript receiver: december 12th, 2008. accepted for publication: july 28th, 2009. julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:18 p p p composite 133 lpi at 45 degrees magnéticas y terremotos cotejados con datos nacionales e internacionales. como resultado de la evaluación comparativa de los datos, esto esta concluyendo que las tormentas geomagnéticas no desencadenan terremotos. palabras claves: terremoto, desencadenar terremoto, tormentas geomagnéticas 1. introduction earthquakes are tectonic events that take place within the fractures of the earth’s crust, namely faults. in addition to the movement of tectonic plates relative to each other, it is widely discussed that there are other external influences originate outside earth that can trigger earthquakes. these influences are called “triggering effects”. examples of such external events are the lunar and solar eclipses, planetary alignment within the solar system and the influences of geomagnetic storms taking place in the sun. the most influential of these events is considered to be the extraordinary solar activities influencing the earth’s geomagnetic field. anatolian peninsula is a historically active earthquake region within the borders of turkish republic. the peninsula, surrounded by black sea in the north, aegean sea in the west, mediterranean sea in the south, lies at the intersection of asia, europe and africa continents. hundreds of earthquakes in various scales take place every year in this active region. the most recent major earthquake, marmara earthquake, took place on 17th august 1999 and measured 7.3 on richter scale, causing extensive damage and widespread fatalities. the purpose of this article is to present a statistical view to elaborate if the solar geomagnetic storms trigger earthquakes. as a model, the research focuses on the anatolian peninsula, presenting 41 years of historical data on magnetic storms and earthquakes collated from national and international resources. 2. the causes of earthquakes and geomagnetic storm and measurement methods this section outlines the principles behind the occurrence of earthquakes and geomagnetic storms and the methods used to measure them. 2.1 the causes of earthquakes and measurements methods in 1911, professor reid established the elastic-rebound theory, supported with [reid,h.f, 1911]. a mathematical scale was required in order to obtain the data about the effects of these vibrations on the structures and charles f. richter and beno gutenberg applied “magnitude” concept on this field in 1930 [gutenberg, b.,et al., 1954]. among these; mb is calculated by taking as basis the magnitude of p and s waves (body-wave magnitude), md is calculated by using the durations of very small and close earthquakes, ms is calculated by taking as basis the magnitude of surface waves (surface-wave magnitude) and mw takes as basis the seismic moment of the released energy (moment magnitude). these are the most used magnitude measures. different magnitude methods can give different values around 0.2m. this difference can reach up to 0.5m after saturation. the magnitude of an earthquake does not change wherever on earth it is measured. giving different magnitude values of earthquakes must be associated with the method used and the fact that the geology of the place in which the recorders are located is not adequately taken into account [gürbüz, c., 2003]. 2.2 the reasons for the occurrence of geomagnetic storms and measurement systems the solar material, which accelerates as a result of the energy discharges occurring in the sun, “solar flares”, “prominences” and “coronal mass ejections”, reaches the earth from the gravitational field of the sun as different radiation forms and particles, and it interacts with the magnetic field of the earth. in cases where these effects are dense, sudden changes occur in the magnetic field and 83 comparative evaluation of the influencing effects of geomagnetic solar storms on earthquakes in anatolian peninsula julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:19 p p p composite 133 lpi at 45 degrees they create the phenomenon which is called “geomagnetic storm.” although different systems are used, one of the most important scales used in measuring this is the dst “disturbance storm time” indexes which have been recorded since 1957 [sugiura, m, 1964]. dst index’s showing high negative values means that there is a big magnetic storm [sugiura, m, 1991]. for the magnetic fields, changes in the dst index are evaluated in five major groups. weak storms -30 nt�dst >-50 nt, moderate storms -50 nt>dst >-100 nt, strongstorms -100 nt>dst>-200 nt, intense storms -200 nt>dst >-350 nt, heavy storms dst >-350 nt [loewe, c.a., at al., 1997], [natural environment research council british antarctic survey, 2001 ]. dst values’ falling to these values in their changes within 24 hours which are measured hourly is regarded as a geomagnetic storm. within the internet, data centers which provide service to the entire world on this subject are also concentrated in the usa. one of these institutions is the national environmental satellite, data, and information service which operates under the usa national oceanic and atmospheric administration, [national oceanic and atmospheric administration (noaa) space weather prediction center (swpc) 2003]. another institution is the national geophysics data center where all observations on all solar activity conducted in the entire earth, the effects of the activity on the near perimeter of the earth and its results on the atmosphere are collected, and this center provides international data coordination, [nooa national geophysics data center, 2005]. world data center system which is conducted by the usa national science foundation provides the communication between scientific research institutions all over the world. another one of these institutions is the world data center for geomagnetism, kyoto, data analysis center for geomagnetism and space magnetism graduate school of science, kyoto university, [world data center for geomagnetism, kyoto, 2005]. 3. the method used for the detection of the triggering effects of the earthquakes in accordance with the earthquake magnitudes and the changes in the dst values the occurrence of the earthquakes is dependent on very different factors. especially, the structures of the rocks are the most important features. due to the fact that some rocks show more resistant features, they may not be broken or may be broken later than usual even though they have the same energy accumulations and they are in close regions. however, it is doubtless that they all have limit values. this can be resembled to the discharge of a condenser, which is loaded with current and which reached the limit value for the discharge, with a very low current; or overflow of a glass, which is filled with water up to its utmost capacity, with a drop of water. in this study, the breakage probabilities of the rocks, which have energy accumulation that reached up to the limit value, with geomagnetic effects apart from the other factors have been examined. the below-mentioned methods have been used in order to obtain these statistics. 3.1 sources for the earthquake data and shortcomings in calculating the earthquake triggering effects, the study area is limited to the anatolian peninsula and surrounding seas, between coordinates 35°, 00’n-43°, 00’n and 25°,00’e-45°,00’e. the earthquakes which fall outside these areas are excluded from this study. the dataset used in this study is based on 41 years of data collected between 1965 and 2005. this data has been collected from the archive of the following institutions, local to the study area; republic of turkey, ministry of public works and settlement, general directorate of disaster affairs, earthquake research department, [general directorate of disaster affairs, earthquake research department, 2005], bogazici university kandilli observatory and earthquake research institute, national earthquake monitoring center –nemc, 84 sadik cengiz yesugey julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:19 p p p composite 133 lpi at 45 degrees [bogazici university kandilli observatory, 2005]. the scientific and technological research council of turkey, marmara research center, earth and marine sciences institute, [the scientific and technological research council of turkey, marmara research center, 2005]. in order to ensure that the dataset is complete, the data from the above listed sources have been validated by the archives of; u.s. geological survey, national earthquake information center usgs-neic, earthquake center, [usgs-neic, 2005], u.s. geological survey, earthquake hazard program, advanced national seismic system-anss, [anss, 2005], incorporated research institutions for seismology iris, [seismology iris, 2005]. the data from various sources was sorted by year, month, day and coordinates, and compared with each other to identify discrepancies. in the event of discrepancies, the data from the local sources was used. during this assessment, it was identified that slight differences in coordinates could result in the same earthquake event being recorded more than once. if there are a number of earthquakes reported by various sources for the same time frame, the coordinates were compared and if the coordinates measure within ± 30’ (30 minutes) of each other, it was assumed that all sources record the same event, therefore, a single earthquake was included in the study. as it has been stated in section 2.1, the measurements conducted by different institutions in different dates are archived as magnitude from time to time in accordance with mb, ms, md and mw scales, and there is no possibility to reduce them to a common unit. however, as it has been stated before, these scales take values which are very close to each other. in this condition, the values and scales provided by the turkish institutions at the same time and date have been primarily accepted. no scale difference was inquired in the different and incomplete data that were obtained by scanning various national and international institutions, and those with the greatest magnitude values were added to the list as a data. the dataset contains all the earthquakes within the coordinate limits set for the anatolian peninsula for the last 41 years. this data is sorted by year, month, day and magnitude. the total number of earthquake measurements within the study period is 122838. the earthquakes with magnitudes below 3.0m was excluded in order to prevent misleading results. the limiting upper bound was set to 7.9m, the highest earthquake recorded between 1965 and 2005. as widely known, there are many relatively minor preshocks leading up to the earthquake and aftershocks following the earthquake, which sometimes continue to occur for months. however, it is not practical to differentiate between the major event and its pre and after shocks. it can also be argued that geomagnetic storms could amplify the aftershocks. for this reasons, the study includes all pre shocks and aftershocks, as long as they coincide with a geomagnetic storm. 3.2 sources for the geomagnetic storm data and shortcomings in this study, data from world data center for geomagnetism, kyoto, data analysis center for geomagnetism and space magnetism graduate school of science, kyoto university which are constantly and regularly broadcasted on the internet and globally acknowledged have been taken as basis [world data center for geomagnetism, kyoto, 2005]. the dst -30 nt value specified in section 2.2 for the geomagnetic storm dst values has been accepted as the minimum storm value. with a computer program that has been prepared, 41 year old data on a 24 hour basis have been scanned, the days on which the dst values were inside the storm limits have been listed, and the values apart from those have been marked as “0” value. thus, “yearly and daily value” lists spanning over 41 years have been obtained. 4. methods used in the assessment of data the data assessment has been conducted with three different methods. 85 comparative evaluation of the influencing effects of geomagnetic solar storms on earthquakes in anatolian peninsula julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:19 p p p composite 133 lpi at 45 degrees in the first method; all the earthquakes occurring in a day within the range of 3.0m-7.9m remaining within the given coordinates between the years 1965-2005 and in different regions have been collectively and on a daily basis compared with the dst values within the same range in terms of their date, magnitude and dst values and presented in the form of tables (for instance; each of the values with such magnitudes as 3.1m, 3.2m, 4.5m, 5.2m that occurred on the same day and within the range of 3.0-7.9 has been taken as data). again within the same range; the earthquakes with magnitudes of 3.0m–3.9m, 4.0m–4.9m, 5.0m–5.9m, 6.0m–6.9m and 7.0m–7.9m have been compared in the form of tables and on a daily basis with the dst values corresponding to their date and magnitude values within their own range. in this comparison, all the earthquake magnitude values within the mentioned range and occurring on the same day have been included in the comparison. (for instance; each of the values with such magnitudes as 3.1m, 3.2m, 3.5m occurring within 3.0-3.9 range and on the same day has been taken as data). in the second method; the greatest rates of earthquakes detected within a day within the range of 3.0m-7.9m and with the magnitudes of 3.0m–3.9m, 4.0m–4.9m, 5.0m– 5.9m, 6.0m–6.9m and 7.0m–7.9m remaining within the given coordinates between the years 1965-2005 and in different regions have been within their own range collectively and daily compared with the dst values according to their date and magnitude values and in the form of tables. however, in this comparison, among the earthquakes occurring within mentioned range and on the same day, the earthquake with the greatest magnitude detected within the day has been included in the comparison. (for instance; as far as such values as 3.1m, 4.2m, 5.5m occurring on the same day were concerned, 5.5m being the greatest value has been taken as data. similarly, when such values as 3.1m, 3.2m, 3.5m within the range of 3.0m-3.9m occurring on the same day are concerned, only 3.5m being the greatest value has been taken as data). in the study conducted; it has been considered that in the past years, especially the records between the 1960s and 1970s might not have been as sound as the records of our present time and as a third method, the comparisons have been repeated for the next 20 years (1986-2005). in these comparisons made with a prepared computer program, the data and statistical values below have been obtained. the evaluation of the data was carried out using a bespoke computer model and the below listed data and statistical values have been obtained. input; number of earthquakes corresponding to geomagnetic storm within the magnitude interval (m), number of earthquakes not corresponding to geomagnetic storm within the magnitude interval (n), total number of earthquakes within the magnitude interval (w), number of days without geomagnetic storms (-d-days), number of days with geomagnetic storms (+d-days), total number of days between 1965-2005 (days), total number of days between 1986-2005 (days1). output; ratio of number of earthquakes corresponding to geomagnetic storms to total number of earthquakes within the magnitude interval (m/w %), ratio of number of earthquakes not corresponding to geomagnetic storms to total number of earthquakes within the magnitude interval (n/w %). 5. evaluation tables based on the approach outlined in section 4, the following evaluation tables were produced. 6. results the ratio of all the earthquakes within the range of 3.0m-6.9m magnitude between the years 1965-2005 which corresponded to the days with dst=-30nt and occurred during the day, to the number of earthquakes (w) which are within the same range differs between 19.56% and 27.50%. this ratio is 45.46% in the 7.0m-7.9m magnitude range. (table 1). the ratio of the greatest earthquake which were detected during the day among the earthquakes be86 sadik cengiz yesugey julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:19 p p p composite 133 lpi at 45 degrees tween the years 1965 and 2005, and which corresponded to the days with dst -30nt, to the number of earthquakes (w) occurring within the same range differs between 19.8% and 20.8%. this ratio is 28.57% in the 7.0m-7.9m magnitude range (table 2). the ratio of the earthquakes within the range of 3.0m-5.9m magnitude between the years 1986-2005 corresponding to the days with dst -30nt and occurring during the day, to the number of earthquakes (w) which occurred within the same range differs between 18.90% and 22.96%. this ratio is 31.15% in the 6.0m-6.9m magnitude range and 50% in the 7.0m-7.9m magnitude range (table 3). the ratio of the greatest earthquake which was detected during the day among the earthquakes within the 3.0m-5.9m range and corresponding to the days with dst -30nt between the years 1986 and 2005, to the number of earthquakes (w) that occurred within the same range differs between 20.00% and 22.67%. this ratio is 23.08% in the 6.0m-6.9m magnitude range and 25.00% in the 7.0m-7.9m magnitude range (table 4). 87 comparative evaluation of the influencing effects of geomagnetic solar storms on earthquakes in anatolian peninsula table 1. evaluation using all earthquakes within the same calendar day between 1965 and 2005. magnitude interval m (num.) n (num.) w (num.) m/w (%) n/w (%) m /(+d-day) (num./num.) n /(-d-day) (num./num.) 3.0-7.9(total) 11060 40294 51354 21.50 78.46 3.743 3.352 3.0-3.9 8966 32979 41945 21.38 78.62 3.034 2.744 4.0-4.9 1897 6552 8449 22.45 77.55 0.642 0.545 5.0-5.9 170 699 869 19.56 71.44 0.0575 0.0582 6.0-6.9 22 58 80 27.50 72.50 0.0075 0.0048 7.0-7.9 5 6 11 45.46 54.55 0.0017 0.0005 +d-day = 2955(num.) -d-day = 12020(num.) day= 14975(num.) table 2. evaluation using the earthquakes with the highest magnitudes within the same calendar day between 1965 and 2005. magnitude interval m (num.) n (num.) w (num.) m/w (%) n/w (%) m /(+d-day) (num./num.) n /(-d-day) (num./num.) 3.0-7.9(total) 2411 9174 11585 20.81 79.19 0.816 0.763 3.0-3.9 1415 5304 6719 21.06 78.94 0.479 0.441 4.0-4.9 882 3413 4295 20.54 79.46 0.298 0.284 5.0-5.9 104 421 525 19.81 80.19 0.0352 0.0350 6.0-6.9 8 31 39 20.51 79.49 0.0027 0.0026 7.0-7.9 2 5 7 28.57 71.43 0.00068 0.00042 +d-day= 2955(num.) -d-day= 12020(num.) day= 14975(num.) julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:19 p p p composite 133 lpi at 45 degrees in addition, there is not a significant quantitative increase of the empirical values m/(+d-day) ratio (table 1-4). 7. conclusion when the ratios above are examined and the highness of the ratio of dst�-30nt values to total days are taken into account, both between the years 1965-2005 and 1986-2005, an extraordinary quantitative increase can not be seen in the earthquakes which have magnitudes of 3.0m-5.9m and which correspond to days with dst�-30nt. when it is considered that the aftershocks are also included in the number of earthquakes which have been taken into the comparison, in the area within the given coordinates, there is no finding that shows that the earthquakes occur as a result of a triggering under the effect of geomagnetic storm. in order to mention such a finding, more number of earthquakes should occur with percentages that correspond to geomagnetic storms. this is not verified by the results. 88 sadik cengiz yesugey table 3. evaluation using all earthquakes within the same calendar day between 1986 and 2005. magnitude interval m (num.) n (num.) w (num.) m/w (%) n/w (%) m /(+d-day) (num./num.) n /(-d-day) (num./num.) 3.0-7.9(total) 8318 30248 38566 21.57 78.43 5.020 5.356 3.0-3.9 7096 26089 33185 21.38 78.62 4.282 4.619 4.0-4.9 1103 3701 4804 22.96 77.04 0.666 0.655 5.0-5.9 96 412 508 18.90 81.10 0.0579 0.0729 6.0-6.9 19 42 61 31.15 68.85 0.0115 0.0074 7.0-7.9 4 4 8 50.00 50.00 0.00241 0.0007 +dst-day = 1657(num.) -dst-day = 5648(num.) day1 = 7305(num.) table 4. evaluation using the earthquakes with the highest magnitudes within the same calendar day between 1986 and 2005. magnitude interval m (num.) n (num.) w (num.) m/w (%) n/w (%) m /(+d-day) (num./num.) n /(-d-day) (num./num.) 3.0-7.9(total) 1527 5238 6765 22.57 77.43 0.922 0.927 3.0-3.9 972 3317 4289 22.66 77.34 0.587 0.587 4.0-4.9 500 1706 2206 22.67 77.33 0.302 0.302 5.0-5.9 48 192 240 20.00 80.00 0.0290 0.0339 6.0-6.9 6 20 26 23.08 76.92 0.0036 0.0035 7.0-7.9 1 3 4 25.00 75.00 0.00060 0.00053 +dst-day = 1657(num.) -dst-day = 5648(num.) day1 = 7305(num.) julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:20 p p p composite 133 lpi at 45 degrees earthquakes with magnitudes of 6.9m-7.9m are very few in number. the fact that these partially have greater percentages gives the impression that they are coincidental values rather than findings that form correlation with geomagnetic storms. as a result of all these data, a hypothesis can not be put forward which suggests that geomagnetic storms trigger earthquakes in the anatolian peninsula. however, these results should not hinder the conduction of further research. a global study on this subject can potentially provide new approaches. references gutenberg b. and richter c. f. (1954), seismicity of the earth and associated phenomena, 2nd edition, princeton univ. pres, new jersey. loewe c. a., prölss, g. w. (1997), classification and mean behaviour of magnetic storms, journal of geophysical research, a 102, 14209-14213. reid, h. f., (1911), the elastic-rebound theory of earthquakes, university of california department of geology bulletin, 6(19), 413–444, berkeley. sugiura, m., (1964), hourly values of equatorial dst, annals of the international geophysical year, vol. 35, pp. 945-948, pergamon press, oxford. sugiura, m.,(1991), on dst index, http:// swdcwww.kugi.kyoto-u.ac.jp/dstdir/dst2/ondsti ndex.html, paper presented at iaga bulletin no. 40. bogazici university kandilli observatory and earthquake research institute, (2005), internet site for national earthquake monitoring center –nemc, http://www.koeri.boun.edu.tr/ (last accessed 2008). gürbüz,c. (2003). internet site for earthquake magnitude, bogazici university kandilli observatory and earthquake research institute, national earthquake monitoring center –nemc, http:// www.koeri.boun.edu.tr/jeofizik/html/info_magnitude.htm (last accessed october 2008). incorporated research institutions for seismology, (2005), internet site for iris, http:// www. iris.edu/quakes/quakes.htm (last accessed 2008). natural environment research council british antarctic survey (2005), internet site for satellite risk index, http://www.antarctica.ac.uk/ satellite risks/dstdetails.html (last accessed 2008). national oceanic and atmospheric administration (noaa) space weather prediction center (swpc) (2005), internet site for ace real time solar wind news and announcement, http://www.swpc.noaa.gov/ace/ (last accessed march 31, 2003). noaa, national geophysics data center (2005), internet site for space weather and solar events, http://www.ngdc.noaa.gov/ (last accessed july 30, 2008). world data center for geomagnetism (2005), internet site for data analysis center for geomagnetism and space magnetism graduate school of science, kyoto university, http:// swdcwww.kugi.kyoto-u.ac.jp/ (last accessed 2008). republic of turkey, ministry of public works and settlement, general directorate of disaster affairs, (2005), internet site for earthquake research department, http://www.deprem.gov.tr/ indexen.html (last accessed 2008). the scientific and technological research council of turkey, (2005), internet site for marmara research center, earth and marine sciences institute http://www.mam.gov.tr/eng/institutes/ ydbe/(last accessed october 07, 2008). u.s. geological survey, (2005), internet site for national earthquake information center usgsneic, http://earthquake.usgs.gov/ eqcenter/(last accessed july 16, 2008). u.s. geological survey, earthquake hazard program, advanced national seismic system, (2005), internet site for anss http://earthquake.usgs.gov/research/monitoring/anss/ (last accessed september 10, 2008). 89 comparative evaluation of the influencing effects of geomagnetic solar storms on earthquakes in anatolian peninsula julio 6-geociencias-vol 13-1 2009.prn d:\geociencias 13-1 jun 2009\geociencias-vol 13-1 2009.vp martes, 07 de julio de 2009 18:11:20 p p p composite 133 lpi at 45 degrees enero%2031-geociencias-vol%2012-2%20dic-en%20baja[1].pdf earth sciences research journal earth sci. res. j. vol. 12, no. 2 (december 2008): 213-234 mineralization potentials ofpegmatites in the nasarawa area of central nigeria akintola, o. f.1 and *adekeye, j. i. d.2 1 raw materials research and development council, p. m. b.232, garki, abuja, nigeria fax: 234 9 4136034, e-mail: akintolaolatunde@yahoo.com, phone: 2348055239304 2 geology and mineral sciences department, university of ilorin, p. m. b. 1515, ilorin, nigeria. email: adekeye2001@yahoo.com, phone: 2348033795444. * corresponding author abstract pegmatites in nasarawa area of central nigeria lie within a fracture-controlled east-north-east trending rare metal pegmatitebeltcloselyassociatedwith latepanafricanperaluminousgranites innigeria.traceelementsoffeldspars andmuscovitesofpegmatites innasarawaareaofcentralnigeriawereanalyzedtodeterminetheraremetalsmineralization potentials of the pegmatites. the minerals show fractionation even within units of complex mineralized pegmatites as indicated by the wide ranges of k/rb in na-feldspars (albites), muscovites and k-feldspars and the wide rangeofk/tl in the latter twominerals.cs,rbandtland the rareearthelements (rees)ce,la,pr andyare enriched in thek-bearingmuscovitesandk-feldspars;k-feldsparshave thehighestaveragevaluesoftl,ce,la,pr andywhile themuscoviteshave thehighest averagevaluesofsn,nb,ga,fandzn.averagevaluesofta,sr,ba, wandzrarehighest inthealbites,whichalsohavethelowestaveragenb/taratio.thisisanindicationthat theseelementsareenrichedalonglatena-richraremetalmineralizingfluids.therarealkalisrbandcsandtlarepositively correlatedwith the rare lithophile elements (ta, rees,pb,bi andy), inboth thepegmatitek-feldspars andmuscovites, stronglypositively correlated with sn in themuscovites but negatively correlated with nb in theminerals. in thepegmatiticalbites,nbhasverystrongpositivecorrelationswithta(0.868)andzr(0.847),which is indicativeof the incorporations/substitutionof these elements in the crystal lattice of theoreminerals. there is a general enrichmentoftherareelementstowardsthemiddleandinnerzonesofthecomplexmineralizedpegmatites.fluorine,phosphorusandboron-rich fluidsplayedsignificant roles in themagmatic complexation/fractionationandconcentration ofsn-ta-li-cs-beinthealbitizedzonesofthehighlyevolvedpagmatites.k/rbversuscsinthek-feldsparsclassify thepegmatites intobarren,be-,li-be, andli-cs-be-ta types.comparably lownb/ta ratios in the late albites and amblygonites indicate theparagenesisof thesemineralswith the tantalumores in themiddle/innerzonesof thecomplex mineralized pegmatites. while the be-, and li-be-types are highly prospective for gem tourmaline, beryl and 213 manuscript receiver: october 22th, 2008. accepted for publication: november 4 th , 2008. columbite; themostevolvedpegmatites in thisareaaregoodcandidates for tantalite, amblygonite-montebrasite,and probablypollucite prospecting/mining. key words: rare metals, pegmatites, muscovites, feldspars, mineralization, correlations. resumen pegmatitaseneláreadenasarawa,nigeriacentral, seencuentrandentrodeuncinturóndepegmatitasconmetales de tierras raras controlado por fracturas, con tendencia este-noreste, y estrechamente asociado con granitos peralumínicosdelpanafricanotardíoennigeria.elementos trazasenfeldespatosymuscovitasde las pegmatitas delareadenasarawa,nigeriacentral, fueronanalizadosparadeterminarelpotencialdemineralizacióndemetales de tierras raras en las pegmatitas. los minerales muestran fraccionamiento aun dentro de unidades en las pegmatitas complejamente mineralizadas, como lo indica el amplio rago k/rb en feldespatos sódicos (albitas), muscovitas y feldespatos potásicos y el amplio rango de k/ti en estos dos últimos minerales. las muscovitas potásicasy los feldespatospotásicos estánenriquecidosencs,rbytiyen los elementosde tierras raras (rees) ce,la,prey;losfeldespatospotásicosposeenelmayorvalorpromediodeti,ce,la,pryyeymientrasquelas muscovitas tienenlosmayoresvalorespromediosdesn,nb,ga, lafyzn.losvalorespromediosdeta,sr,ba, la w y zr son más altos en las albitas, las cuales poseen también el promedio más bajo de nb/ta. esto indica que estos elementos son enriquecidos por fluidos mineralizantes tardíos ricos en sodio y metales de tierras raras. los elementosalcalinosrarosrbandcsytisecorrelacionanpositivamenteconloselementoslitofilos(ta,rees,pb, biyy),en los feldespatospotásicosy lasmuscovitasde laspegmatitas,y tieneunafuertecorrelacionpositivacon snen lamuscovitapero lacorrelacionesnegativaconnben losminerales.en lasalbiticaspegmatiticas,nb tiene una correlacion positiva muy fuerte con ta (0.868) y zr (0.847), lo que indica la incorporación o sustitución de estos elementos en la estructura cristalina de los minerales de mena. hay un enriquecimiento general de los elementos de tierras raras hacia la mitad y las partes mas profundas del complejo pegmatitico mineralizado. fluidos ricos en flúor, fósforo y boro juegaron un papel importante en la complejización y fraccionamiento magmatico y en la concentración de snta-li-cs-be en las zonas albitizadas de las pegmatitas altamente evolucionadas.k/rbvscsen feldespatospotásicosclasifican laspegmatitasen tiposestériles, debe,deli-be,y de li-cs-be-ta. relaciones de nb/ta comparativamente bajas en las albitas y ambligonitas tardias indican la paragenesis deestosminerales con lasmenasde tantalio en las zonasmedias e internasdel complejopegmatitico mineralizado.mientras los tiposbeyli-besonaltamenteprospectivospara tourmalina,beriloycolumbitacomo gemas; las pegmatitas mas evolucionadas en esta area son buenas candidatas para la prospección minera de tantalita, ambligonitamontebrasita, y probablemente polucita. palabras clave: metales raros, pegmatitas, moscovitas, feldospastos, mineralización correlaciones. introduction nigeria possesses a very large pegmatite environment. the pegmatites are widely distributed with a marked concentrationofmineralizedpegmatites inabroadbelt, which extends from ago-iwoye in the southwest to bauchi in the northeast, an air distance of more than 400kilometers. thousands of pegmatites occur in this belt, most of which have never been mapped or sampled ina systematic, scientificmanner.potassiumfeldspar, albiticplagioclase, quartz, and subordinate, if any muscovite and/or biotite constitute the major minerals while awide spread spectrumofminerals including lepidolite, tourmaline(green,pink,blackandblue),beryl, chrysoberyl, apatite, amblygonite, monazite, lithiophyllitetriphyllite,ghanite,cassiterite,wodginite, nigerite, columbo-tantalite, tapiolite, microlite, bismuthinite, bismuthite, scheelite, cholite, andalusite and 214 akintola, o. f. and adekeye, j. i. d. sillimanite occur as accessory minerals in the pegmatites,bowden and kinnaird (1984). the ages, mineralogy, and composition of these pegmatite units appear to be analogous to those of the pegmatites environment in brazil, canada and australia. the pegmatitic belt and the orientation of the units within it appear to be related to rotational stresses created by the benue trough.fromamoreglobal perspective, this trend is probably the northern extension of the brazilian pegmatite belt, which runs from rio grande del sul toriograndedelnorte.thepegmatite fieldof this study area is part of late pan african, (jacobson and webb, 1949; wright, 1970), rare (specialty) metals granitic pegmatites. the primary mineralizationof tantalum,niobium, tin, beryllium and lithium is hosted in quartz-feldspar-muscovite pegmatites (kinnaird, 1984). the nasarawa area lies between latitudes 8o 18’ n and 8o 30’ n and longitudes 7o 35’ e and 7o 50’ e, west of the southern boundary of afu complexthe southern most unit of the younger granites complexes (figure1).miningof tantalite from both pegmatites and the eluvials started in the area some years ago. the mining continues for the major periods of the year except during the very dry monthsoffebruary toearlyaprilwhen lackofwater makes it difficult to mine and concentrate the minerals. columbite and cassiterite are recovered as by-products of the tantalite mining, while other pegmatite minerals like quartz, feldspar, amblygonite-montebrasite, cookeiteandmicaare still being discarded in waste dumps. area description, methods and materials studied regional geology centralnigeria ispartofanupperproterozoicmobile belt extending from algiers across the southern sahara into nigeria, benin, and the cameroun. this pan-african belt continues into northeast brazil where analogous rare-metal mineralized pegmatites are also known (schuiling, 1967). bordered to the west by the west african craton (stabilized around 2 ga) the pan-african belt itself is made up of gneiss-migmatites,metasediments, andmetavolcanics thathavebeensubjected topolycyclicmetamorphism, and emplacement of igneous rocks. these rocks all constitute theprecambarian tolowerpaleozoicbasement complex rocks. the gneiss-migmatites bears imprints of the liberian (ca. 2500ma), eburnean (ca 2000ma)andpanafrican(ca.600ma) tectonicevents (oversby, 1975; turner, 1983). within the sequence are domains of metasediments and metavolcanics intruded by igneous rocks, which constitute the north/south trending schist belts. the schist belt lithologies which consist of fine grained clastics, pelitic schists, phyllites, banded iron-formations,marbleandamphibolitesareconsidered to be upper proterozoic assemblages (turner, 1983). they host most of the economic minerals in thebasementcomplex.during thepan-african episode, the proterozoic gneiss-migmatite-schist complex were intruded by various granitoids resulting from oceanic closure, subduction, oblique collision between the west african craton and the hoggar – nigeria shields (black, 1984) andcrustal thickening. the pan-african granitoids of nigeria, which is collectively termed older granites, comprise gabbros, charnockites, diorites, granites, and syenites. geochronological data frompreviousworks (rb – sr whole-rock and u-pb zircon) of pan-african granitoids intruding the reactivated archean to lowerproterozoiccrustofcentralandsouth-western nigeria show that intrusive magmatite activity in these areas lasted from at least 630 to 530 ma (van breemen et al., 1977; rahaman et al., 1983; dada et al., 1987; matheis and caen-vachette, 1983; umeji and caen-vachette, 1984; akande and reynolds, 1990). results of the rock ages also show that pegmatities’ emplacement in the southwestern nigeriaoccurredmainlyafter thepeakof thepan-african orogenic event in this area. the end of the pan-african tectonic event is marked by a conjugate fracture system of the strike-slip faults (ball, 1980). fault directions have 215 mineralization potentials ofpegmatites in the nasarawa area of central nigeria consistent trend and sense of displacement; i.e. a ne-sw (nne-ssw) trending system having a dextral sense of movement and a nw-se trending system a sinistral sense (mccurry, 1971;wright, 1976;holt et al., 1978; ball, 1980). both sets crosscut all the main pan-african structures, including older n-s trending shear zones (mylonites) and late orogenic granites (ball, 1980;ajibade andwright, 1989;kuster, 1990; garba, 1992). gold and pegmatities’ rare mineralization are closely associated with the fractures in the pan-africanbelt (kuster, 1990;ekuemeandmatheis, 1995; garba, 2002, 2003). about 100km north-east of the area of study at wamba, rare metal pegmatites have also been geochemically linked to peraluminous late pan-african tectonic granitoids, the emplacement of which have largely been controlled by the regional fractures (kuster, 1990). chemical data on granites, and graniticandpegmatiticmuscovites showthatrb,cs,sn, nb, and ta are enriched during both magmatic and postmagmatic evolution, with the highest contents of these elements occurring in early muscovites of the albitized and mineralized pegmatities (kuster, 1990). albite, k-feldspar, and quartz are the main pegmatitie-forming mineral; white mica is a typical but minor component. from a geochemical point of view, quartz is of no particular interest since it diadochially (substitutionally) hosts trace elements to a negligible extent. feldspars and white micas are the most informative minerals because their element distributions reflect the traceelementscontents in either early pegmatite forming fluids or in the later metasomatizing solutions (moller and morteani, 1987). feldspars and white micas of pegmatites in the nasarawa area were sampled and analysed for traceelements todetermine themineralizationpoten216 akintola, o. f. and adekeye, j. i. d. ilorin ilesha sokoto gusau kano zaria kaduna abuja study area enugulagos maiduguri port-harcourt pan african nigeria belt 10°0° 10° 20° atlantic ocean congo craton0 5001000 50 100 150 kilometres kilometres legend faults barren pegmatites newly metal pegmatites discovered rareknown rarepegmatites metal major cities / towns cretaceoussedimentary cover recent mesozoic ring complexes [younger granites] supracustal (schist belts) basement, gneisses migmatites and pan african (older) granites n atlantic ocean 6° 4° 8° 10° 12°n 6° 10° 16°e12°8°2° west african craton 0° figure 1: geological map of nigeria showing the regional fractures and location of areas of rare-metal and barren pegmatites (after garba, 2003) tials of the pegmatites. data from this area is also compared with published data on well-studied pegmatities for comparison of the potentials. analytical procedures are as stated in akintola and adekeye (2006). the chemical data on the pegmatitemicaand feldsparswere subjected tobivariate correlations after a lognormal transformation of the data. bivariate correlation coefficient of the elements, r, were interpreted to determine elemental geochemical associations and evolutionary trends in the pegmatites. local geology the geological setting of nasarawa area is shown in figure2. fromfield evidence, theoldest rocks in the studyareaaremicaceouspeliticschists.variations in the schist composition for instance, the mica versus thequartz–feldspathiccontentsareobservable in the field.thisrockunithasundergoneapolyphasemetamorphismandductiledeformationsevidencedby local variations in the strike and dips of the foliation. like the other schist belt rocks in the country, the schistshaveageneralstrikeofnorth-south, andin the area, dips at gentle angle (10o – 30o) to the east. the schists were intruded by older granites, which outcrop as hills especially to the northwest, and west of the area. these older granites have a range of granodiorite to granite compositions. to theeastof theareaare rocksof theafucomplex of the younger granites. the afu complex is made up of high-level anorogenic granites mainly bioticgranites andminorquartzporphyryemplaced within precambrian paleozoic basement gneisses and older granites but exposed beneath the cretaceous-recent sedimentarycoverof the lowerbenue valley to the south. the geochemistry of the granites and the mica schists hosting the pegmatites in this area are discussed in akintola and adekeye (2006). two groups of pegmatites are noticeable in the area with minor muscovite in the area: (1) simple, usually barren massive quartz – microcline pegmatites with minor muscovite and accessory tourmalines and (2) complex, albitized muscovite-quartz-microline pegmatites, bearing the rare-metals ta, nb, sn, li and be mineralization. the simple, barren quartz-feldspar (with minor mica) pegmatites occur at the north western and western parts of the area and are spatially closelyassociatedwith themainphaseoldergranites.onanoutcrop scale, thepegmatites grade into patches of aplite. the complex rare-metal pegmatites are found far away from the granite plutons usually hosted by schists (exterior pegmatites). wall-rock alteration in the simple pegmatites is negligible, but more pronounced in the complex pegmatites especially at contacts with the hanging walls (jacobson and webb, 1946). although tourmalinization isby far themost common type of contact alteration, it is generally accompanied by silicification, albitization, greisenization and sometimes formation of apatites/fluorapatites which give rise to graded contacts. k-feldspars table 1 shows the trace element contents of the microcline,microperthites (k-feldspars) takenfrom different pegmatites in the study area. for example, one sample of k-feldspar has the highest cs (3489ppm), tl (73ppm), bi (21ppm), rees ce (69ppm), la (166ppm), pr (29ppm) and very high rb (9474ppm). it has the lowest total fe and mn (fe2o3, 0.13%; mno, 0.002%), and k/tl (1558), rb/tl (130), k/cs (33) and k/rb (12) ratios. these characteristic indicate that the feldspar crystallized from a highly fractionated melt and although this particular sample has low concentration of the ore elements, sn-ta-nb, an eluvial concentrate from the mine had earlier yielded high values of ta (1749ppm) with high ta/nb ratio of 5.45, nimamop (1998). the average content of ta in the k-feldspars is very low (2ppm). the k-feldspars have k/rb range of12-35andameanof19.k/rbhasaveryhighnegative correlation (-0.938) with cs in the k-feldspars, figure3. infigure3, thecscontent is plottedversus the k/rb ratio for k-feldspar together with the boundaries discriminating, according to trueman 217 mineralization potentials ofpegmatites in the nasarawa area of central nigeria and cerny (1982), between barren, be-, li-be-, and li-cs-be-ta mineralized pegmatites. from the figure, the following pegmatites in the study area of which the k-feldspars were analyzed belong to the corresponding classes (table 2). it is important tonotefromthisclassificationthat somepegmatites from the same location have differentdegreesofevolution–somearemoreevolvedbelonging to the lithium-beryllium class and some belong to the beryllium class. the pegmatites in the beryllium and lithium-beryllium classes have high prospects for gem tourmaline and beryl mineralization, especially in the albitized zones of the pegmatites (preinfalk et al., 2000). the liberia and onyelow wazobia pegmatites mainly belong to the lithium-cesium-beryllium-tantalum class. it is also important to note here that a sample of the liberia pegmatitesplots in the lithium-berylliumclass showing that the pegmatites must at least have evolved from the lithium-beryllium class to the lithium-cesium-beryllium-tantalum class. the liberia pegmatite therefore has potentials for bearing ore mineral characteristicsofboth the lithium-berylliumand lithium-cesium-beryllium-tantalumclasses in thedifferent zones of the pegmatite. active mining of sn-nb-ta ores and analysis of mica and other mineral samples such as amblygonite and fluorapatite, confirmtheenrichmentof thepegmatitein theoresof li-be-b-sn-nb-ta. na – feldspars the framework silicate samples (feldspars and feldspathoids) with low silica (sio2 < 55.95%) and na/k greater than 1 (except cookeite) were grouped together for convenience as na-feldspars (table 2). the true na-feldspars according to deer et al (1966) have the following average contents of ore and lithophile elements: p2o5 (1.17%), f (bdl), ga (27ppm),nb (145ppm),sn (221ppm),ta (195ppm), rb (175ppm) and cs (31ppm). the ore elements are finely disseminated in the albites. the aplitic footwall albite is probably secondaryorat least lateprimaryincrystallization.most of the albite in the pegmatites is found in the albite-rich aplitic zoneoftenoccurring in the footwalls of pegmatites (jahn and tuttle, 1963; jahn and burnham, 1969). thesecondaryalbitehasvery lownb/taratioof 0.26, which is comparable to the low nb/ta ratio of 218 akintola, o. f. and adekeye, j. i. d. 25 \ akpoku 30 kama tudunwada 30 30 endo nasarawa 7°40’e7°35’e 8°30’e 3000 meters 0 3 6 kilometers3 n older granites granodiorite tonalite gneiss younger granites (afu complex) mica schist tourmalinite basic dyke pegmatite strike and dip of foliation geological boundary 30 legend roads 8°25’n 8°21’n 8°18’n 8°30’e 8°25’n 8°21’n 8°18’n 7°45’e 7°50’e gidankwma figure 2: geological map of nasarawa area. 219 mineralization potentials ofpegmatites in the nasarawa area of central nigeria table 1: trace elements of the microline, microperthies (k-feldspars) of the pegmatites. sample l3 l6 l7 l10a luz lu ls s2 k1 k3 w2a p (ppm) 2400 2461 2662 2579 2130 4774 5398 6642 2854 2138 3024 f 0 0 0 0 0 0 326 0 0 0 0 ba 44 35 34 67 10 30 53 36 69 91 39 bi 20 10 17 15 17 15 10 13 12 10 21 cd 13 7 9 5 6 bdl bdl bdl bdl bdl 16 ce 40 32 32 12 52 21 0 19 35 0 69 co 16 33 18 28 20 28 31 25 23 31 18 cr 10 13 9 14 2 15 35 11 6 1 6 cs 1722 1487 1540 1226 1482 602 111 160 844 692 3489 cu 7 7 14 9 0 13 13 8 4 19 16 ga 18 15 17 15 17 16 44 19 18 14 17 la 95 73 74 58 84 27 2 16 38 31 166 nb 8 5 9 7 9 22 39 8 4 8 8 ni 20 0 31 28 0 0 0 0 0 0 27 pb 70 57 81 79 69 46 0 35 125 140 93 pr 19 15 16 13 18 8 0 6 9 8 29 rb 8546 6536 9534 8303 8420 5537 2593 3069 5440 4089 9474 sn 21 13 187 15 14 28 24 28 9 8 24 sr 38 144 58 44 51 22 30 183 64 70 32 ta bdl 1 2 bdl bdl 9 6 bdl bdl bdl bdl tl 49 38 57 49 49 30 12 16 39 30 73 w 155 241 137 187 176 203 195 154 143 204 160 v 10 0 0 14 18 0 0 0 0 0 20 y 30 18 20 0 19 24 12 11 0 14 24 zn 44 0 0 0 0 0 40 0 0 0 0 k/ba 2521 2906 3492 1694 11058 3628 1696 2841 1571 1165 2916 k/rb 13 16 12 14 13 20 35 33 20 26 12 k/cs 64 68 77 93 75 181 811 639 128 153 33 k/tl 2264 2676 2083 2316 2257 3628 7493 6393 2780 3534 1558 rb/tl 174 172 167 169 172 185 216 192 139 136 130 0.26 average in the amblygonites. obviously, both the secondary albite/mica and the amblygonites are in paragenesis with the tantalum minerals. this observation is in agreementwith the findings of moller and morteani (1987). a sample with a very high (8.8%) l. o. i. was also identified (by xrd and icp-oes) to be cookeite – a hydrated lithium aluminum silicate. it has na/k ratio slightly less than 1. cookeite crystallizes from late stage hydrothermal fluids evolved fromresidualalbiticmeltsonstabilizationof tourmaline(andthelossofother fluxingcomponentssuchas p, b, li, f and h2o) from the pegmatite melts (cerny, 1991b). cookeite is known to occur with other low temperature minerals like adularia and zeolites in miarolitic cavities of the highly evolved complex rare metal pegmatites. white micas (muscovites) in the pegmatites theaveragechemicalcompositionof themuscovites is typicalof rareelementpegmatiteswithhigh,f,cs, rbandli (deer et al., 1966;gordiyenko,1971) (table 3). the xrd diffractogram of selected samples of the micas show that the crystal structures of the muscovitesareclose to the idealdioctahedralandr2+ free composition typical of muscovites from the rare-element pegmatites class (cerny and burt, 1984).themuscoviteshaveawider rangeofrbvalues (2659-10182ppm) and lower k/rb ratios (8-31) than the k-feldspars’ (2593-9534ppm) and (12-35) respectively (table 4). a late generation mica ð taken from an inner zone of liberia pegmatite has very high cs (2353ppm), rb (9910ppm), sn (647ppm)andta(103ppm)andcorrespondingly low k/rb (8), nb/ta (0.66), and k/tl (1600) ratios. these are comparable to average values of cs (2294ppm), rb (8978ppm), sn (665ppm), and ta (464ppm) and the correspondingly low average k/rb (9), nb/ta (0.14), and k/tl (1361) ratios in muscovites from highly mineralized pegmatites in the northern part of the area. nb/ta ratios for the pegmatitic muscovites range from 0.13 to 8.07 with a mean of 2.965. the k/rbratios for themicas range from8and31witha mean of 16.20. ta has a very high positive correlation with cs (0.756) but a very high negative correlation with nb (-0.856). linnen (1998) and morteani and gaupp (1989) have observed that the differentbehaviouroftaandnbduringcrystallization is due to differences in solubility of nb and ta in li-rich pegmatitic melts. the cs vs. k/rb plots of pegmatite muscovites from thearea (figure 4) concentrate in the field typical of rare-element class pegmatites as defined by cerny and burt (1984). only one sample which was collected about 100meters south of the liberia pegmatite plots in the muscovite class. the k/rb ratios of the muscovites range between 7.58 and 30.94. these lowratiosofk/rbaswell as thehighnegative correlation of k/rb and cs (-0.830) are typical of muscovites of the rare-metals class of pegmatites (preinfalk et al., 2000) and have been used successfully as exploration tools for the pegmatites. correlation of the trace elements and fractionation trends k/tl,k/rb,k/cs,al/ga,zr/hfandnb/taratiosgive fractionation indices in the granite-pegmatites suites. some of these ratios are compared with the average valuesof the traceelements in thepegmatiticminerals (table4).the table also shows that the rare earth elements (rees) ce, la, pr, and y are enriched in the 220 akintola, o. f. and adekeye, j. i. d. barren 10p be li, be li, cs, be, ta cs 100001000100101 1 10 100 1000 k/rb figure 3. classification of the pegmatites using the plots of k/rb versus cs of their k-feldspars. note: the lines discriminating between barren pegmatites and pegmatites with significant be; li-be-, li-beta-mineralization are given according to trueman and cerny (1982). 221 mineralization potentials ofpegmatites in the nasarawa area of central nigeria table 2: trace element content of the albitic pegmatite phases phosphates albites cookeite fluorapatite amblygonites sample lb6a lb8 sj3 rna lc19 kj2fsp lb9 lb10 ch1 p2o5 (%) 0.481 0.487 2.169 1.763 0.243 0.017 24.97 49.973 43.25 f(ppm) 1240 0 0 0 0 294 11637 13897 14203 as 0 0 7 8 6 1 0 0 0 ba 19 51 37 26 102 56 31 33 61 bi 14 11 11 16 12 12 38 1 8 ce 14 12 22 14 0 0 443 0 7 cd 7 bdl bdl bdl bdl bdl 27 bdl bdl co 115 47 25 31 37 5 16 8 6 cr 1 0 0 11 0 15 1 2 1 cs 851 23 19 74 9 49 1 28 6 cu 0 0 24 16 4 0 0 16 535 ga 76 24 21 22 39 58 14 21 18 hf 2 3 0 3 3 2 0 3 2 la 48 9 0 3 3 0 204 0 0 mo 0 5 1 1 0 4 0 1 0 nb 88 75 62 326 115 33 190 26 25 nd 3 11 9 8 2 0 144 5 7 ni 14 5 3 4 0 3 9 4 3 pb 25 10 7 0 0 0 74 9 0 pr 10 2 1 2 1 0 65 0 1 rb 4829 69 347 252 31 444 46 290 62 sc 0 0 9 0 4 0 0 0 5 sm 1 3 2 2 1 1 66 1 1 sn 565 659 14 174 35 54 28 231 67 sr 263 52 1037 305 51 8 64 11 188 ta 345 109 67 305 297 37 21 107 86 th 0 0 0 0 0 3 2 1 4 tl 23 bdl bdl bdl bdl 5 bdl bdl bdl u 0 3 0 10 0 1 216 0 7 k-bearingminerals,k-feldsparsandmuscovites,with the highest concentrations of the element in the k-feldspars.becauseof their similargeochemicalmigration patterns, k, rb, cs and tl are concentrated, andhaveveryhighpositivecorrelations in thek-feldspars andmuscovites (table6).themuscoviteshowever have higher fractionation indices compared with thek-feldsparas indicatedbytheir lowaveragek/rb, k/tl, and k/cs ratios. the muscovites have high values of sn (383ppm average), nb (155ppm average), ta (96ppm average), ga (196ppm average), f (2128ppm average), and zn (379ppm average) and have higher rb/tl ratio than the k-feldspars. the muscoviteshavethehighestvaluesofga(withamean valueof161ppm)andlowestal/garatioof theminerals. this corroborates the observations of cerny et al (1985) that tourmaline and muscovites are the main concentrators of ga in pegmatites. ba, sr, w, zr and ta are highest in the na-feldspars with the lowest nb/ta ratios, which is an indication that theseelementsare enrichedalongwithta in the late mineralizing fluids. the following groups of elements (table 5) have high positive correlation indices in thena-feldspars:rbversuscs (0.922), cs versus zn (0.866), cs versus co (0.802), co versus w (0.974), ga versus co (0.890), ga versus w (0.851), nb versus zr (0.941), nb versus bi (0.850), taversusbi(0.769),taversusga(0.638),taversus nb (0.629) and ta versus w (0.538).the high positive correlations of these elements reflect their associations in the formation of rare metal ores from the na-rich late fluids. the pegmatitic k-feldspars have very high negative correlations of k/tl and k/rb with the rare lithophile elements such as cs, bi, y, rb, la, pr, pb (table 7). similarly, the muscovites have very high negative correlations of k/tl and k/rb with the rare lithophile elements as well as the ore elements sn and ta. nb is negatively correlated with rb, cs, tl, ta, la, pr and sn but positively correlated with k and ga in both the k-feldspars andmuscovites (table7).nbandgahavepositive correlations in both muscovites and k-feldspars (0.704 and 0.472) respectively. obviously, the 222 akintola, o. f. and adekeye, j. i. d. sample lb6a lb8 sj3 rna lc19 kj2fsp lb9 lb10 ch1 v 5 1 4 15 6 5 16 4 8 w 502 346 201 247 298 64 46 89 124 y 6 1 0 0 1 0 1391 1 0 zn 229 63 26 187 12 39 62 26 119 zr 17 12 17 68 18 5 17 54 7 h2o 2.42 0.43 1.1 0.82 0.36 8.8 0.51 5.27 7.54 sum 98.71 98.24 97.67 99.4 99.79 98.01 98.12 102.94 86.53 k 33457 1577 11540 3321 1079 9132 1494 3487 664 k/rb 7 23 33 13 35 21 32 12 11 mg(hx) 13 139 7 23 li(hx) 225 685 16400 20750 li(fusion) 227 2900 13366 17882 na/k 1.17 43.74 5.81 19.80 72.80 <1 21.80 4.23 n.d nb/ta 0.26 0.69 0.93 1.07 0.39 0.89 9.05 0.24 0.29 223 mineralization potentials ofpegmatites in the nasarawa area of central nigeria t a b le 3 . t ra c e e le m e n ts in th e p e g m a ti te m ic a s s a m p le (p p m ) le 7 le 8 le 1 4 le 1 3 le 1 8 l4 l1 0 b l8 a l9 a lz lu a ls 7 l c 1 9 lc 2 0 lc 2 0 a s1 k k a lc 2 3 lc 2 8 w w 2 r lc 3 0 lc 3 1 lc 3 2 lc 3 3 f 2 9 3 3 3 1 9 7 3 2 1 5 3 2 2 0 2 9 7 2 7 4 3 2 3 3 1 3 1 2 8 8 4 4 3 5 2 4 6 5 2 1 0 6 6 1 7 1 0 0 4 3 9 0 3 9 3 3 2 9 6 6 7 4 6 1 2 8 4 1 2 2 5 4 2 6 1 6 2 9 6 5 4 7 5 1 9 4 0 1 0 8 8 8 8 1 3 7 4 b a 1 5 6 0 3 1 2 7 2 4 1 2 3 0 3 5 2 0 3 1 2 3 5 3 7 6 9 3 5 1 9 1 3 2 8 1 5 4 2 2 5 1 0 0 7 4 3 3 1 6 2 3 4 4 1 8 b i 1 7 2 1 2 0 1 6 2 1 2 5 1 6 1 9 1 2 1 9 1 3 1 4 1 2 1 5 2 8 3 4 1 4 2 1 1 7 1 8 2 0 1 1 1 8 1 5 1 3 1 2 1 3 1 7 c e 1 2 2 9 5 2 0 2 6 1 3 0 1 1 0 0 0 0 9 3 2 3 9 3 7 1 8 2 6 1 4 1 2 9 0 2 7 0 2 2 0 2 3 c s 1 7 1 8 7 4 2 9 0 3 0 0 2 6 0 8 7 4 1 0 0 1 0 0 1 2 3 2 8 9 6 2 0 6 3 2 1 1 8 3 4 2 2 1 2 0 2 4 6 7 4 2 4 5 1 5 6 9 4 5 1 6 9 4 2 1 5 1 7 4 9 2 9 1 4 9 9 6 3 1 1 6 3 4 6 g a 1 6 7 1 6 3 1 6 3 1 5 3 1 6 7 1 6 8 1 5 8 1 6 4 1 6 2 1 7 4 1 5 5 2 0 3 2 0 0 1 6 6 1 5 6 1 6 4 1 5 9 1 4 1 1 7 6 1 7 4 1 9 7 1 5 4 1 4 5 1 6 2 9 7 1 4 7 1 6 1 1 3 8 l a 0 5 7 1 5 2 0 2 3 5 4 4 4 4 5 1 0 3 6 9 0 0 4 7 9 8 1 0 8 1 7 1 4 3 6 0 4 0 3 2 0 4 9 7 1 0 2 1 5 n b 1 8 1 1 1 6 1 7 8 1 6 5 1 8 2 1 3 1 1 1 9 1 1 3 1 8 5 1 4 4 1 7 8 2 1 8 2 5 6 1 8 7 7 5 5 5 1 4 6 1 5 1 1 6 0 2 1 1 1 4 3 1 9 3 2 0 0 1 2 7 6 4 2 2 3 1 9 0 1 3 5 n d 9 1 7 4 1 4 1 0 6 0 7 2 0 2 0 1 9 1 2 7 7 1 3 2 1 7 0 1 1 0 9 4 5 7 n i 7 2 7 1 7 1 8 2 1 2 3 2 4 2 4 2 1 2 3 1 3 8 9 1 8 2 7 3 1 1 8 2 1 1 6 1 1 2 0 1 0 6 1 7 1 1 1 6 1 8 2 4 p b 1 8 4 7 2 3 2 1 3 0 4 6 4 5 4 5 1 9 4 4 1 7 6 4 2 5 4 7 6 4 1 8 2 6 2 2 7 3 3 1 0 9 3 3 2 7 9 1 8 1 9 p r 3 1 5 6 8 8 1 4 1 2 1 2 6 1 1 6 1 3 1 1 1 9 2 2 6 6 9 2 1 1 7 3 1 1 1 3 2 4 5 r b 4 8 0 3 9 4 1 0 5 0 9 3 5 5 7 8 5 8 3 9 1 3 9 1 5 8 8 0 3 5 2 8 9 8 8 7 4 7 5 1 3 1 3 3 2 6 5 9 5 6 3 8 7 7 7 4 1 0 1 8 4 5 0 4 4 9 4 1 5 5 2 7 3 4 7 1 7 7 4 9 3 2 8 4 2 8 7 0 6 8 0 3 4 5 8 1 3 6 4 5 4 3 2 4 4 2 4 8 s n 2 1 7 4 7 1 2 7 5 3 3 6 3 5 7 5 2 5 5 3 3 5 9 7 2 7 4 6 8 1 2 3 9 6 1 2 6 6 3 9 7 6 8 1 6 4 9 2 9 5 3 6 4 3 9 4 8 7 4 3 7 1 3 9 1 1 8 9 0 2 5 3 9 1 4 7 2 7 1 2 1 9 s r 1 5 5 0 4 1 1 8 1 9 2 4 2 7 2 5 1 7 2 4 1 6 1 1 1 3 1 8 2 5 3 2 1 6 1 8 1 9 1 1 2 3 1 5 1 2 2 1 1 5 1 7 1 7 1 6 t a 4 4 5 8 4 5 5 0 6 3 5 3 5 9 7 1 7 1 7 1 4 1 2 7 4 6 7 2 5 0 2 4 2 5 7 5 8 5 1 1 5 3 1 5 1 6 2 6 4 1 2 0 1 8 3 3 9 3 1 1 4 0 t l 2 2 4 3 2 4 2 7 2 6 4 0 3 9 3 9 2 5 4 0 2 1 1 7 1 5 3 1 5 2 6 4 2 4 2 9 2 8 1 6 3 5 1 8 1 7 3 6 3 3 1 8 2 1 2 5 w 7 1 5 1 6 5 7 1 2 3 7 3 9 3 3 2 8 3 8 7 7 2 5 8 4 6 5 4 0 8 6 8 9 3 4 3 6 4 8 5 6 5 2 4 0 4 9 9 1 5 6 1 1 4 3 2 6 1 z n 4 2 1 1 0 0 0 4 5 2 4 3 5 4 7 2 9 6 1 3 9 1 1 4 1 6 8 0 0 3 7 9 2 3 1 1 3 1 2 1 7 1 1 1 1 2 3 1 4 2 2 4 2 1 4 2 1 8 7 3 4 1 2 4 9 1 6 3 4 5 3 4 7 1 6 1 1 7 7 1 1 2 k /r b 1 7 9 1 6 1 5 1 4 9 9 9 1 5 9 1 7 2 7 3 1 1 4 1 0 8 1 8 1 6 1 5 2 3 1 0 2 6 2 8 1 2 1 7 2 3 1 8 1 9 m g (h x ) 6 1 2 0 6 9 6 9 7 6 2 2 2 8 2 2 5 2 2 1 7 4 8 1 1 2 9 7 5 7 0 6 6 2 0 5 0 3 1 8 9 6 6 9 1 1 0 7 8 5 6 6 8 3 3 6 4 4 n b /t a 4 .1 1 2 3 .9 6 3 .3 2 .8 9 2 .4 7 2 .0 2 1 .5 9 2 .6 1 2 .0 3 4 .3 4 8 .0 7 5 .5 7 2 .6 0 .1 5 0 .1 3 1 .9 5 1 .7 8 1 .3 9 6 .8 1 2 .8 3 .1 1 3 .1 3 1 .0 6 0 .3 5 5 .7 2 6 .1 3 0 .9 6 k /r b (h x ) 1 0 .9 5 6 .0 1 1 0 .6 6 9 .4 5 9 .6 1 6 .2 4 6 .3 0 6 .4 3 9 .6 0 6 .8 6 1 0 .5 8 1 5 .5 4 1 8 .3 5 9 .2 6 6 .9 5 5 .5 4 1 1 .2 9 1 0 1 0 .0 1 1 3 .4 8 7 .2 5 1 4 .2 4 1 5 .8 9 8 .2 1 9 .9 1 1 4 .0 5 1 1 .6 9 1 1 .1 4 m g /l i( h x ) 1 .3 0 .3 4 1 .3 5 1 .1 7 0 .8 4 0 .2 6 0 .1 7 0 .1 9 1 .2 4 0 .2 4 1 .5 7 1 6 .2 3 2 .2 5 2 .2 1 5 7 .3 3 1 0 1 .6 1 1 5 .5 1 1 .1 3 1 1 4 .5 5 4 .4 6 8 .5 2 2 5 .7 2 2 .2 5 1 1 l ia a s (h x ) 4 7 5 8 5 1 5 9 9 0 8 6 1 6 8 1 1 5 4 2 8 8 4 7 5 4 3 4 1 4 9 2 3 1 2 8 6 2 0 2 4 1 0 3 1 4 .5 1 6 4 l ia a s (f u s. ) 1 5 4 2 8 3 2 0 1 1 9 8 2 7 2 3 9 5 4 6 8 3 0 3 1 7 4 3 2 4 2 3 1 3 0 4 0 2 0 4 7 3 6 1 1 2 1 5 8 1 0 4 3 3 1 7 6 1 1 1 6 0 2 9 5 4 7 2 3 0 k -a a s (h x ) 5 8 0 5 4 1 8 0 5 9 1 8 4 6 8 0 0 0 7 2 2 5 4 2 2 5 4 8 1 5 5 0 5 0 2 8 4 5 4 0 7 5 4 3 1 0 3 5 1 5 5 3 0 5 4 2 5 0 7 4 5 5 5 0 5 5 4 9 3 0 4 1 3 0 5 1 6 5 4 5 5 5 3 8 1 5 5 9 7 0 6 0 8 0 4 2 9 0 r b -a a s (h x ) 5 3 0 6 9 5 5 5 5 4 9 5 9 0 0 8 7 5 8 9 5 1 1 2 4 4 4 0 4 4 5 5 3 2 5 1 5 5 4 4 0 6 2 0 6 3 5 4 7 0 4 2 5 7 4 5 3 7 5 6 8 0 2 9 0 3 2 5 5 5 5 3 8 5 4 2 5 5 2 0 3 8 5 -a a s (f u s. ) 2 8 1 9 5 2 9 9 3 7 0 5 3 6 0 4 6 8 6 1 0 6 3 4 3 6 0 3 8 0 8 2 5 1 2 2 3 0 0 3 9 3 4 4 2 3 1 6 8 8 6 3 2 5 3 3 0 8 3 4 1 5 0 2 7 8 3 4 2 9 1 1 4 4 4 2 1 8 3 4 4 3 5 3 0 7 7 2 2 0 7 3 2 5 8 3 5 8 0 224 akintola, o. f. and adekeye, j. i. d. table 4: average contents of trace elements (range given in brackets) in the k-feldspars, na-feldspars and micas of the pegmatites k-fedlspar(n=11) muscovite(n=29) albitesn=4(true albites) ba 46ppm(10-91) 31ppm (3-100) 54ppm (26-102) cs 1214ppm (111-3489) 664ppm (32-2467) 31ppm (9-74) ga 19ppm (14-44) 161ppm (97-203) 27ppm (21-39) la 60ppm (2-166) 33ppm (1-109) bdl nb 12ppm (4-39) 155ppm (55-256) 145ppm (62-326) pb 73ppm (10-140) 27ppm (4-64) bdl pr 13ppm (1-29) 9ppm (1-22) bdl rb 6504ppm (2593-9534) 5930ppm (2659-10182) 175ppm (31-347) sn 34ppm (8-187) 383ppm (61-902) 221ppm (14-659) sr 67ppm (22-183) 22ppm (11-64) 361ppm (51-1037) f bdl 2128ppm (329-4751) bdl ni 9ppm (0-31) 18ppm (6-31) bdl ta 2ppm (0-9) 96ppm (27-502) 195ppm (67-305) ce 28ppm (0-69) 15ppm (0-47) 12ppm (0-22) zn 3.64 379ppm (47-1023) 72ppm (12-187) tl 40ppm (12-73) 30ppm (15-64) bdl w 178ppm (137-241) 62ppm (25-237) 273ppm (201-346) y 16ppm (0-30) 15ppm (0-39) 0.5ppm (0-1) k/tl 3362 (1558-7493) 3096 (1206-5485) rb/tl 168 (130-216) 198 (139-235) k/rb 19 (12-35) 16 (8-31) 26 (13-35) k/cs 211 (33-811) 121 140 k/ba 3226 (1165-11058) 2575 81 al/ga 5359 1197 3680 nb/ta 7.04 2.89 (0.15-8.07) 0.77 nb-dominant minerals crystallize earlier than ta and the positive correlations of nb, k and ga indicate thatnbenters thesilicate structuremoreeasily than ta. this agrees with the observations of cernyandburt (1989) thatnb/ta ratios in silicates are rather high. linnen (1998) has also observed that the different behaviour of ta and nb during crystallization is due to differences in solubility of nbandta inli-richpegmatiticmelts.suchbehaviour has also been observed in the bruno and xuxa pegmatites of the aracuai pegmatite district of brazil (preinfalk et al., 2000). in themuscovites (table7),tahas averyhigh positive correlation (0.756) with cs but lower correlations with rb in both muscovites and the albites (0.428 and 0.213 respectively). ta and k/rb however have better correlations in the muscovites and albites (-0.441 and -0.592 respectively)whencomparedwithtaandrb.the reason for this relationship between the elements may be explained by the fact that ideal positions for tao 4 3and tao3are lacking in the silicates. therefore, ta partitioning into mica will be small and tao 4 3which grow epitaxially at the mica faces account for much of the ta contents in the micas (moller and morteani, 1987). mg/li has a high negative (-0.856) correlation withzinc; thehighpositive correlationsofzn,li and f(³0.759) in themicasare traceable to thefact that the threeelementsoccupy theoctahedral layer in themica crystal structure. li has negative correlation with ta (-0.244),whichshowsthatlicontentof theearlypegmatite micas are higher than the later ones; obviously increased concentration of li through fractionation in the late fluidscrystallized to formdiscreteli-minerals such as amblygonite-montebrasite rather than enter micaphyllosilicatestructure.this is inagreementwith the observation of cerny et al (1985) that the buildup of li concentration in a pegmatite melt that precedes the precipitation of li minerals sensu stricto leads to increasedcontentsofli inearly rock-formingmineral phases. sn average value is highest in the muscovites; it also has much higher positive correlations with cs (0.902),rb(0.829),tl (0.868),la (0.799) compared with the very low correlations it has with these elements in the feldspars, which reflects the enrichment of tin with greisenization. usingk/rb,k/tlandcs, it is thereforepossible to recognize different generations of feldspars and micas within a complex-zoned mineralized pegmatite (table8).e.g. early and late,whichhave significant differences in the range, and mean contents of the rare elements. generally, the rare alkali (rb and cs) and tl concentrations in the muscovites and k-feldspars give a good reflection of the concentration of the rare elements in the pegmatites. the liberia (lb) pegmatite deposit in this section, the trace element contents of the k-feldspars and muscovites of pegmatites in the nasarawaareaarecomparedwith thoseofwell-studied pegmatites both within and outside nigeria. the liberia pegmatite in the study area is the most exposed of the studied pegmatites, because someblasting and drilling had been carried out on it. twelve (12) muscovite samples of the pegmatite were analyzed. extensive fractionation within the pegmatite is indicated by the wide range of k/rb 225 mineralization potentials ofpegmatites in the nasarawa area of central nigeria rare-elements pegmatites’class muscovite class cs 10000 k/rb 10001000 10 100 1000 figure 4: plots of k/rb versus cs for the pegmatites’ muscovites. note: the dash line represents the boundry between the muscovite and rare-metal classes (after cerny and burt, 1984).thesolid lines represent thebest fit lines for thelct type pegmatites. 226 akintola, o. f. and adekeye, j. i. d. t a b le 5 . c o rr e la ti o n m a tr ix o f so m e tr a c e e le m e n ts th a t h a v e si g n if ic a n t b iv a ri a te c o rr e la ti o n s in th e n a -f e ld sp a rs (n = 5 ) b a c s g a n b r b s n t a w z n z r k /r b n b /t a b a 1 .0 0 0 c s -0 .8 6 8 1 .0 0 0 g a -0 .2 3 7 0 .6 6 9 1 .0 0 0 n b -0 .1 8 3 0 .1 2 2 -0 .1 6 8 1 .0 0 0 r b -0 .9 0 0 0 .9 2 2 0 .5 6 5 -0 .0 8 7 1 .0 0 0 s n -0 .4 2 7 0 .6 1 3 0 .3 7 1 0 .1 3 2 0 .3 2 5 1 .0 0 0 t a -0 .1 4 9 0 .4 8 6 0 .6 3 8 0 .6 2 9 0 .2 1 3 0 .5 5 5 1 .0 0 0 w -0 .2 6 5 0 .6 7 7 0 .8 5 1 -0 .1 6 6 0 .4 6 5 0 .4 8 3 0 .5 3 8 1 .0 0 0 z n -0 .8 9 1 0 .8 6 6 0 .2 7 4 0 .4 0 2 0 .7 1 8 0 .7 4 0 0 .3 9 3 0 .4 6 7 1 .0 0 0 z r -0 .3 5 1 0 .1 6 0 -0 .2 7 5 0 .9 4 1 0 .0 6 9 0 .1 5 2 0 .4 5 1 -0 .3 5 5 0 .4 2 7 1 .0 0 0 k /r b 0 .8 3 4 -0 .9 7 3 -0 .6 0 8 -0 .3 0 1 -0 .4 7 6 -0 .8 1 8 -0 .5 9 2 -0 .6 8 5 -0 .9 2 8 -0 .2 9 0 1 .0 0 0 n b /t a -0 .0 1 6 -0 .4 6 1 -0 .9 6 1 -0 .3 1 5 -0 .1 3 1 -0 .3 5 4 -0 .5 4 1 -0 .8 3 6 -0 .0 4 5 -0 .3 9 6 0 .3 9 6 1 .0 0 0 227 mineralization potentials ofpegmatites in the nasarawa area of central nigeria t a b le 6 . c o rr e la ti o n m a tr ix o f tr a c e e le m e n ts th a t h a v e si g n if ic a n t b iv a ri a te c o rr e la ti o n s in th e k -f e ld sp a rs (n = 1 1 ) c s g a l a n b p b p r r b s n t l w k /t l r b /t l k /r b k /c s c s 1 .0 0 0 g a -0 .6 6 7 1 .0 0 0 l a 0 .9 4 7 -0 .8 0 1 1 .0 0 0 n b -0 .5 6 6 0 .7 0 4 -0 .6 9 4 1 .0 0 0 p b 0 .7 3 8 -0 .8 5 2 0 .7 9 2 -0 .7 8 3 1 .0 0 0 p r 0 .9 2 3 -0 .8 1 2 0 .9 9 7 -0 .6 8 5 0 .7 7 7 1 .0 0 0 r b 0 .9 4 3 -0 .5 8 5 0 .9 0 3 -0 .4 5 4 0 .6 1 0 0 .8 8 8 1 .0 0 0 s n 0 .2 6 0 .1 4 7 0 .0 3 3 0 .2 9 2 -0 .2 1 6 0 .0 5 7 0 .2 3 6 1 .0 0 0 t l 0 .9 8 0 -0 .6 6 2 0 .9 4 0 -0 .5 6 4 0 .7 6 1 0 .9 2 1 0 .9 6 7 0 .1 1 4 1 .0 0 0 w -0 .1 3 3 -0 .0 3 3 -0 .2 1 9 0 .2 4 5 -0 .2 4 7 -0 .2 3 6 -0 .2 5 4 -0 .4 6 3 -0 .2 7 5 1 0 0 0 k /t l -0 .9 8 9 0 .6 3 6 -0 .9 3 4 0 .5 6 1 -0 .7 4 1 -0 .9 1 1 -0 .9 6 3 -0 .0 8 1 -0 .9 9 7 0 .2 3 8 1 .0 0 0 r b /t l -0 .6 5 0 0 .5 9 3 -0 .6 2 8 0 .6 3 7 -0 .8 6 7 -0 .6 0 2 -0 .4 3 2 0 .3 0 1 -0 .6 4 8 0 .2 1 5 0 .6 5 0 1 .0 0 0 k /r b -0 .9 3 8 0 .5 3 2 -0 .8 8 5 0 .4 2 8 -0 .5 6 1 -0 .8 6 8 -0 .9 9 6 -0 .2 4 3 -0 .9 5 4 0 .2 3 6 0 .9 5 5 0 .3 9 7 1 .0 0 0 k /c s -0 .9 9 9 0 .6 5 1 -0 .9 3 9 0 .5 6 3 -0 .7 2 3 -0 .9 1 3 -0 .9 3 3 -0 .0 0 3 -0 .9 7 1 0 .1 0 4 0 .9 8 3 0 .6 4 6 0 .9 3 1 1 .0 0 0 228 akintola, o. f. and adekeye, j. i. d. t a b le 7 . c o rr e la ti o n m a tr ix o f tr a c e e le m e n ts th a t h a v e si g n if ic a n t b iv a ri a te c o rr e la ti o n s in th e p e g m a ti te s’ m u sc o v it e s f c s g a l a n b r b s n t a t l z n l i( x ) l i( fu s) r b (x ) r b (f u s ) k /r b n b /t a k /t l r b /t l m g /l i f 1 .0 0 0 c s -0 .1 1 6 1 .0 0 0 g a -0 .1 5 5 -0 .3 2 9 1 .0 0 0 l a -0 .0 0 2 0 .9 1 3 -0 .3 2 7 1 .0 0 0 n b 0 .2 1 7 -0 .8 7 5 0 .4 7 2 -0 .7 5 3 1 .0 0 0 r b 0 .1 8 0 0 .8 2 1 0 .0 1 2 0 .7 7 1 -0 .6 9 6 1 .0 0 0 s n -0 .0 3 6 0 .9 0 2 -0 .1 8 5 0 .7 9 9 -0 .7 0 0 0 .8 2 9 1 .0 0 0 t a -0 .5 0 7 0 .7 5 6 -0 .3 9 4 0 .6 7 5 -0 .7 9 6 0 .4 2 8 0 .5 9 5 1 .0 0 0 t l -0 .0 4 8 0 .9 2 2 -0 .1 5 6 0 .8 5 1 -0 .8 5 8 0 .9 3 9 0 .8 6 8 0 .6 8 2 1 .0 0 0 z n 0 .7 5 9 0 .1 6 0 0 .3 4 8 0 .1 9 3 0 .0 7 5 0 .5 5 9 0 .2 8 1 -0 .3 7 5 0 .2 8 7 1 .0 0 0 l i( x ) 0 .6 6 6 0 .1 2 7 0 .0 4 4 0 .1 8 3 0 .0 5 6 0 .4 5 2 0 .2 3 2 -0 .2 4 4 0 .2 5 0 0 .8 1 4 1 .0 0 0 l i( fu s) 0 .6 3 0 0 .1 6 7 0 .0 0 5 0 .2 1 4 0 .0 0 5 0 .4 5 5 0 .2 7 5 -0 .1 9 4 0 .2 8 4 0 .7 7 3 0 .9 8 0 1 .0 0 0 r b (x ) 0 .2 7 9 0 .5 8 3 0 .0 7 2 0 .5 5 8 -0 .4 3 1 0 .8 5 9 0 .6 1 9 0 .1 8 2 0 .7 0 9 0 .6 0 8 0 .5 1 8 0 .4 4 5 1 .0 0 0 r b (f u s) 0 .1 1 6 0 .6 9 6 0 .1 2 9 0 .6 2 3 -0 .5 6 8 0 .9 1 9 0 .7 6 2 0 .3 3 3 0 .8 3 3 0 .5 1 8 0 .4 0 8 0 .4 3 1 0 .8 3 9 1 .0 0 0 k /r b -0 .1 6 4 -0 .8 3 0 0 .0 0 6 -0 .7 7 2 0 .7 0 8 -0 .9 9 8 -0 .8 4 0 -0 .4 4 1 -0 .9 4 3 -0 .5 3 4 -0 .4 4 1 -0 .4 4 6 -0 .8 5 8 -0 .9 2 1 1 .0 0 0 (8.0-17), k/cs (66-406), k/tl (1600-3764), and nb/ta (1.59-4.11) ratios in the micas (table 9). rb andcscontents andk/rb,k/csaswell asnb/ta ratios in themicas show that there are at least two generations of primary muscovites in the pegmatite, the early and the late. table 8. some trace elements distribution in the different muscovite generations in the liberia pegmatite deposit element early generation late generation rb(%): range 0.48-0.58 0.88-0.99 mean 0.52 0.92 cs(ppm):range 171-300 874-2353 mean 243 1168 tl(ppm):range 21-27 39-47 mean 24.16 41.33 ta(ppm):range 41-71 53-103 mean 52.33 69.16 nb(ppm):range 165-185 68-144 mean 178 115 sn(ppm):range 217-357 471-681 mean 283 575.67 k/rb:range 13.8-17 8.0-9.3 mean 15.58 8.77 k/cs:range 242-406 66-81 mean 301 74.6 k/tl:range 3007-3764 1600-2058 mean 3375 1941 nb/ta:range 2.61-4.11 1.59-2.47 mean 3.54 2.02 number of samples 6 6 the rb, cs and tl contents show enrichment in late muscovites by factors of 1.77, 4.81 and 1.71 respectively with a corresponding enrichment in the oreelementssnandtabyfactorsof2.03and1.32respectively but a noticeable depletion of nb by a factor of 1.55 in the late muscovite. thus, the enrichment of ta and sn in the late muscovites is accompanied by decrease in the k/rb, k/cs, k/tl and nb/ta ratios. while there is also enrichment in zn in the late muscovite, they are depleted in mgo and tio2. similar enrichments of the rare elements havebeenobserved in the lateprimarymuscovitesof well-studied pegmatites like tanco, noumas, tip-top and harding. such late generation primary muscovitesare found in themiddleandcentral zones of the pegmatites where they occur in paragenesis with ores of the rare elements. intable8,acomparisonismadeof therb,cs,ta and ga contents of muscovites of selected pegmatites in the study area, wamba (100km north-east of the area), and some well studied pegmatites from other partsof theworld.themicashavehighrb(0.8978%) and cs (2294ppm) which are comparable to those of the higly mineralized tanco pegmatite of manitoba, canada as shown in table 9. pegmatites in the areahave awide rangeofmineralization potential from low ta – pegmatites through medium ta-pegmatites to high ta-pegmatite. from table 9, it is evident that the rb and cs contents of the micas reflect ta mineralization in pegmatites. extremely low k/rb (8.86 or less), and k/cs (34 or less) characterize such highly mineralized pegmatites like tanco and liberia. the high ta-mineralization potentials of the complex pegmatites in thisarea iscorroboratedbythehighaverageta2o5content (406ppm)andta/nbratio(3)of eighteen concentrate samples taken from different mines in the area (okunlola, 1998). muscovites of the mineralized pegmatites from wambaareahave lowta-potentialwith correspondingly low cs (116ppm), rb (0.3150%) and high k/rb (25) and k/cs (800) nb/ta (4.50) ratios. the wamba pegmatites vary from barren, muscovite class to complex albitized, and sn-mineralized pegmatites (kuster, 1990). on the whole, it appears thepegmatities innasarawa, thestudyarea, aremore fractionated and therefore have higher ta-mineralization potential than those of wamba and the other 229 mineralization potentials ofpegmatites in the nasarawa area of central nigeria studied pegmatites in nigeria (matheis and kuster, 2001, personal communication). discussion and conclusion the very low ta contents in the k-feldspars may be attributed to the fact thattao3 cannot enter the fully polymerized (alsi)o4 network of the k-feldspar (moller and morteani, 1987). thus, the rare alkalis rbandcsarebetter indicatorsof theraremetalsmineralization potentials in the pegmatites’ k-feldspars than the ore elements ta. the higher contents of the rare earth elements (rees) ce, la, pr, and y in the k-feldspar when compared with those of the other majorpegmatiteminerals shows that therees in the fluids from which the pegmatites crystallized were partitioned more into the k-feldspar than the other rock-forming minerals. this corroborates the observations of simmons and heinrich (1980) that ree-bearing minerals occur in k-feldspar-rich parts ofpegmatites.thehighnegativecorrelationofk/rb versus cs is characteristic of k-feldspars of rare metalpegmatitesandhasbeenusedalongwithk/rb versuscs in themicasas reliableprospectionaids for rare metal pegmatites (preinfalk et al., 2000). high positive correlations between cs, rb and the rees: ce, la, and pr are notable in the k-feldspars, compared with the muscovites. this may be due to selective complexing of the rees in the residual melt by the framework silicate feldspar. the same process of selective complexing by p and f must have enriched the fluorapatite with high concentrations of the rees. the low average nb/ta ratio for the muscovites (2.965) is still lower than the (4±0.7) upper limit diagnostic of ta-pegmatites (cerny, 1989). higher nb/ta ratios (9.5±1.5) and high nb-concentrations (³200ppm) are diagnostic of the less specialized nb-rich columbite pegmatites. beus (1966) determined that ³20ppm ta concentrations are characteristic of columbotantalitepegmatites andgordiyenko (1971)alsodetermined that65-75ppmtaconcentrationsarecharacteristic of the ta-enriched pegmatite (figure 5). simultaneousenrichmentofrb,csalongwithta in the lct granite pegmatite suites makes the alkalis reliable indicators of rare-metals mineralization in pegmatites. thus k/rb versus cs, and ta versus cs plotsofprimarymuscoviteshavebeen reliablyused to determine the mineralization potentials of pegmatites. 230 akintola, o. f. and adekeye, j. i. d. table 9. geochemical characteristics of some well-studied pegmatites compared with nasarawa pegmatites k(%) rb(%) cs(ppm) ta(ppm) ga(ppm) k/rb k/cs no of samples liberia 8.0831 0.6972 556 57 163 12.6 198 11 loc 20/20a 7.7997 0.8978 2294 464 160 8.86 30 2 k/ka 8.2231 0.5234 605 100 159 15.75 119 2 w/w2 8.3145 0.3077 196 63 150 27.12 374 2 tip top, south dakota (low ta) 8.52 0.355 222 56.0 175 24 384 13 tanco, manitoba (high ta) 8.33 2.450 2420 240.9 433 3.4 34 19 noumas, namaqualand 8.21 0.357 566 74.2 92 23 145 2 harding, new mexico (low ta) 9.47 0.631 1917 64.0 123 15.0 49 21 wamba, central nigeria n.d 0.3150 116 53 n.d 25 800 51 matheis (1979) has earlier shown that mineralized pegmatites in ijero and egbe area of southwesternnigeriaarestronglydepleted inba,sr,andzrbut are enriched inrb,li,y,be,sn,nb, andta.kuster (1990) has alsoobserved similargeochemical evolution in latepanafrican tectonicgranitesandmineralized pegmatites in wamba area. the relative enrichment of ba, sr, w, zr and ta in the na-feldspars indicate that pegmatites innasarawaarea crystallize from a more fluid and rare metals-enriched melt than those of ijero and egbe, as well as wamba areas.these resultsagreewithobservationsofcerny et al. (1985) who noted that these elements are enriched in late hydrothermal stages of pegmatite formation. high activity of p during the primary pegmatite crystallization in this area resulted in the formation of the amblygonite subtype (burt and london, 1982; london and burt, 1982b; cerny, 1991b) of the complex pegmatites. the exhaustion ofli andfby thephosphates frompegmatiticmelt may account for the li-, and f-poor micas with the highesttacontentswhichprobablycrystallizedafter the crystallization of the amblygonites in the area. tl, rb, cs and the rees la, ce, pr fractionation in k-feldspars and white mica demonstrates congruent/similar trends. k/rb, k/tl and k/cs ratios are lower in the white mica than the k-feldspars. ta and cs have the highest positive correlation (0.756) in the micas, therefore low k/rb (16),k/tl (3096) andk/cs (121) as found in the nasarawa area indicate high rare metal ta-nb-sn-li-be mineralization potentials as observed in the nasarawa pegmatites. acknowledgement financialassistance for the fieldworkby thenigerian government through the raw materials research and development council, rmrdc is gratefully acknowledged. the german government through the german academic exchange programme (daad) provided financial assistance for the rocks/minerals chemical analyses in the technicaluniversityofberlin (tub)which is also gratefully acknowledged. the authors are grateful to dr. g. matheis of thetechnicaluniversityofberlin (tub) forhis assistance on the research. we are also grateful for the assistance of professor peter moller and dr. peter dulski of the geoforshungs zentrum (gfz), potsdam for their assistance in analyzing the rare earth and trace elements of some of the samples. references ajibade, a. c. and wright, j. b. (1989) the togo-benin-nigeria shield: evidence of crustal aggregation in the pan african belt. tectonophysics 165, 125-129. akande, s. o. and reynolds, p. h. (1990) 40 ar/ 39 ar spectrum ages of micas from the sn-nb-ta bearing pegmatites in nigeria. in: proceeding, volume 15, colloquium of african geology. centre internationalpour laformation et les echanges geologiques (cifeg) occassional publication 1990/20 p243. ball, e. (1980) an exampleof very consistent brittle deformation over a wide intra-continental area: the late pan-african fracture system of the 231 mineralization potentials ofpegmatites in the nasarawa area of central nigeria gordiyenko (1971) bcus (1966) 10000100010010 1 10 100 1000 cs ta figure 5. plot of ta versus cs for the muscovites of nasarawa pegmatites. note: the limits of 20ppm ta according to beus (1966), and of 65-75ppm set by gordiyenko (1971) for a ta mineralization are also given. tuareg and nigerian shield. tectonophys 61, 363-379. beus, a. a. (1966). geochemical exploration for endogenic deposits of rare elements on the example of tantalum. nedra, moscow, engl. transl. gse libr, ottawa. black, r. (1984) the pan african event in the geological framework of africa, pangea 2, 6-16. bowden, p. and kinnaird, j. a. (1984) geology and mineralization of the nigerian anorogenic ring complexes. geol. jb. b56, 3-65. burt,d.m.andlondon,d. (1982)subsolidusequilibria. in: cerny, p. (ed). granitic pegmatites in science and industry: mineralogical association of canada, short course handbook 22, 221-232. cerny, p. (1989). exploration strategy and methods for pegmatite deposits of tantalum. in: moller, p., cerny, p. and saupe, f. (eds). lanthanides, tantalum and niobium: society for geology applied to mineral deposits, special publication 7, springer-verlag, 271-299 pp. cerny, p. (1991b) rare-element pegmatites, part 1: anatomyand internal evolutionofpegmatitedeposits: geoscience canada, 18(2), 49-67. cerny, p. and burt, d. m. (1984). paragenesis, crystallochemical characteristics and geochemical evolutionofmicas ingranitespegmatites: in: bailey,s.w. (ed.),mica.minerologicalsociety of america, reviews in mineralogy 13, 257-297. cerny,p.,meintzer,r.e.andanderson,a.j. (1985) extreme-fractionation in rare-element granitic pegmatites: selectedexamplesofdataandmechanisms: canadian mineralogist 22, 381-421. dada, s. s., lancelot, j. r. and briqueu, i. (1987) age and origin of a pan-african charnockitic complex: u-pb and rb-sr evidence from the charnockitic complex at toro, nothern nigeria. abtr. vol. 14 coll. afri. geol. berlin, 72-73. deer, w. a., howie, r. a. and zussman, j. (1966) an introduction to the rock-forming minerals. longman, london. 528pp ekwueme,b.n.andmatheis,g. (1995)geochemistry and economic value of pegmatites in the pre-cambrianbasementofsoutheastnigeria. in: magmatism in relation to diverse tectonic settings (srivastava, r. k. and chandra, r. eds) ibh publishing co., new delhi, india, 375-392 pp. garba, i. (1992)geology, geochemistryandorigin of gold mineralization at bin yauri, nigeria. phd thesis, university of london, uk. 264 pp. garba, i. (2002) late pan-african tectonics and origin of gold mineralization and rare-metal pegmatites in the kushaka schist belt, north-western nigeria. j. min. & geol., 38(1), 1-12. garba, i. (2003) geochemical discrimination of newly discovered rare-metal bearing and barren pegmatites in the pan-african (600±150ma) basement of northern nigeria. applied earth science (trans. inst. min. metall.) 112; 287-292. gordiyenko, v. v. (1971) concentrations of li, rb and cs in potash feldspar and muscovites as criteria for pegmatites. int. geol. reviews 13, 134-142. holt, r. w., egbenuwe, i. g., fitches, w. r. and wright, j. b. (1978) the relationships between low-grade metasedimentary belts, calc-alkaline volcanismand thepan-africanorogeny inn.w. nigeria. geol. rund. 67, 631-646. jacobson, r. and webb, j. s. (1949) the pegmatites of central nigeria. geol. surv. nig. bull. 17, 61p. jahns,r.h. andburnham,c.w. (1969)experimental studiesofpegmatitesgenesis.amodel for the derivation and crystallization of granitic pegmatites. econ. geol. 64, 843-864. 232 akintola, o. f. and adekeye, j. i. d. jahns,r.h.andtuttle,o.f. (1983)layeredpegmatite-aplite intrusives. mineralogical soc. amer. spec. paper 1, 78-92. kinnaird, j. a. (1984) contrasting styles of sn-nb-ta-zn mineralization in nigeria. j. afr. ear. sci., 2(2), 81-90. kuster, d. (1990) rare-metal pegmatites of wamba, centralnigeria-their formationinrelationship to late pan-african granites. mineral deposita 25, 25-33. linnen, r. l. (1998) the solubility of nb-ta-zr-hf-w in granitic meltswith li and li + f: constraints for mineralization in rare-metal granites and pegmatites. econ. geol. 93, 1013-1025. london, d. and burt, d. m. (1982b) lithiumminerals in pegmatites. in: cerny, p (ed). granitic pegmatites in science and industry: mineralogical association of canada, short course handbook 8, 99-133. matheis, g. (1979) geochemical exploration around the pegmatitic sn-nb-ta mineralization of sw-nigeria. geol. soc. malaysia bull. 11, 33-351. matheis, g. and caen-vachette, m. (1983). rb-sr isotopic studyof rare-metalbearingandbarren pegmatites in the pan-african reactivation zone of nigeria. j. afr. ear. sci. 1, 35-40. mccurry, p. (1971). pan-african orogeny in northern nigeria. geol. soc. amer. bull. 82: 3251-3262. moller, p. and morteani,g. (1987) geochemical exploration guide for tantalum pegmatites. econ. geol. 42, 1888-1897. morteani, g. and gaupp, r. (1989) geochemical evaluation of the tantalum potential of pegmatites. in:lanthanides,tantalumandniobium. moller, p., cerny, p. and saupe, f. (eds). 303-310. nigerian minerals appraisal and monetization programme nimamop stage i report (1998). publ. ministry of solid minerals development, abuja. okunlola, o. a. (1998). nimamop stages ii/iii batch ifinalreport: specialtymetals exploration programme in the udegi (ogapa oto) areaofnasarawastate.publishedby theministry of solid minerals development, abuja, nigeria. oversby,v.m. (1975) lead isotopic studyof aplites from the pre-cambrian basement rocks near ibadan, southwestern nigeria.earthplanet. sci. lett. 27, 177-180. preinfalk, c., morteani, g. and huber, g. (2000) geochemistry of the granites and pegmatites of the aracuai, minas gerais (brazil). chem. erde 60, 305-326. rahaman,m.a.,emofurieta,w.d.,caen-vachette, m. (1983) the potassic granites of the igbeti area: further evidence of the polycyclic evolution of the pan-african belt in south-western nigeria. precamb. res. 22, 27-92. schuiling, r. d. (1967) tin belts on the continents around the atlantic ocean. econ. geol. 62, 540-550. simmons, w. b. and heinrich, e. w. (1980) rare-earth pegmatites of the south pit district colorado. col. geol. surv., denver, resources ser. 11, 131 pp. trueman,d.l. andcerny,p. (1982)exploration for rare-element granitic pegmatites. in: cerny, p (ed). granitic pegmatites in science and industry. minerological association of canada, short course handbook 8, 463-494. turner,d.c. (1983)upperproterozoic schistbelts in thenigerian sector of thepan-africanprovince of west africa. precamb. res. 21, 55-79. umeji, a. c. and caen-vachette, m. (1984) geochronology of pan-africa nasarawa eggon and mkar gboko granites, south east nigeria. precambr. res. 23, 317-324. 233 mineralization potentials ofpegmatites in the nasarawa area of central nigeria van breemen, o., pidgeon, r. t. and bowden, p. (1977) age and isotopic studies of pan-african granites from north-central nigeria precambr. res. 4, 307-319. wright, j.b. (1970)controlsofmineralizationin the older and younger tin fields of nigeria. econ. geol. 65, 945-951. wright, j. b. (1976) fracture systems in nigeria and initiationof fracture zones in thesouth atlantic. tectonophys. 34, 43-47. 234 akintola, o. f. and adekeye, j. i. d.